dissenting:
Fifty years after Brown v. Board of Education, 347 U.S. 483, 74 S.Ct. 686, 98 L.Ed. 873 (1954), required the integration of public schools “with all deliberate speed,” 349 U.S. 294, 301, 75 S.Ct. 753, 99 L.Ed. 1083 (1955), segregated schools remain in many communities, often as a result of segregated housing patterns.1 Seattle has been no exception to the struggle to achieve and maintain integrated schools. After decades of more coercive efforts to counteract the effects of segregated housing patterns, Seattle School District No. 1 (“the District”) in 1998 adopted a high school assignment plan (“the Plan”) to maximize school choices for students and their families while continuing to ensure that integrated public schools are available to all. I respectfully dissent from my colleagues’ conclusion that the Plan is unconstitutional.2 When understood in context, the Plan is narrowly tailored to serve a compelling governmental interest in ensuring that all students in Seattle’s public high schools receive the educational benefits of an integrated learning environment.
The Supreme Court of the United States has never decided a case involving the consideration of race in a voluntarily imposed school assignment program that is intended to promote integrated secondary schools. The Court’s recent decisions regarding the consideration of race in selective admissions to institutions of higher learning do not control in the secondary-school context, but they provide several guiding principles. First, the Court in Grutter v. Bollinger, 539 U.S. 306, 328, 123 S.Ct. 2325, 156 L.Ed.2d 304 (2003), clarified that remediation of past official or de jure discrimination is not the only permissible reason for a government to use racial classifications; one permissible reason for considering race is to achieve the educational benefits of diversity. Second, as the Court reminded us in Grutter:
Not every decision influenced by race is equally objectionable and strict scrutiny is designed to provide a framework for carefully examining the importance and the sincerity of the reasons advanced by the governmental decisionmaker for the use of race in that particular context.
Id. at 327, 123 S.Ct. 2325.3 Finally, the Court emphasized that “narrow tailoring” is a fact-based analysis, noting that the inquiry in Grutter had to be “calibrated to fit the distinct issues raised by the use of race to achieve student body diversity in public higher education.” Id. at 334, 123 S.Ct. 2325.
In order to calibrate our inquiry to fit the distinct issues raised by the use of race as a factor in a school assignment program in a public school district, I believe it is *990necessary, first, to understand the governmental interests that the District is trying to further and, second, to employ a narrow-tailoring analysis that is appropriate to the secondary-school setting and to the process of assigning every student to a high school. Doing so leads me to conclude that the Plan is narrowly tailored to serve compelling governmental interests.4
I. The District’s interests in employing a race-conscious classification are compelling, but are not identical to those asserted in Grutter and Gratz v. Bollinger, 539 U.S. 244, 123 S.Ct. 2411, 156 L.Ed.2d 257 (2003).
As the majority rightly notes, “diversity” can be an amorphous concept. Maj. op. at 962. Indeed, the compelling interest that the Court recognized in (Pmtter is not “diversity” per se but, rather, promotion of the specific educational and societal benefits that flow from diversity. See Grutter, 539 U.S. at 329-30, 123 S.Ct. 2325 (noting that the law school’s concept of critical mass must be “defined by reference to the educational benefits that diversity is designed to produce”). In evaluating the relevance of diversity to higher education, the Court focused principally on two benefits that a diverse student body provides: (1) the learning advantage of having diverse viewpoints represented in the “robust exchange of ideas” that is critical to the mission of higher education, id. at 329-30, 123 S.Ct. 2325; and (2) the greater societal legitimacy that institutions of higher learning enjoy by cultivating a cadre of national leaders who are representative of our country’s diversity, id. at 331-33, 123 S.Ct. 2325. The Court also mentioned the role of diversity in challenging stereotypes. Id. at 330, 333, 123 S.Ct. 2325; see also Lani Guinier, Admissions Rituals as Political Acts: Guardians at the Gates of Our Democratic Ideals, 117 Harv. L.Rev. 113, 175-76 (2003) (observing that the diversity interest recognized in Grutter has “three important elements ...: diversity is pedagogical and dialogic; it helps challenge stereotypes; and it helps legitimate the democratic mission of higher education” (footnotes omitted)). The Court explicitly deferred to the law school’s “educational judgment that such diversity is essential to its educational mission.” Grutter, 539 U.S. at 328, 123 S.Ct. 2325.
Because strict scrutiny requires us to evaluate the “fit” between the government’s means and its ends, Wygant v. Jackson Bd. of Educ., 476 U.S. 267, 280 n. 6, 106 S.Ct. 1842, 90 L.Ed.2d 260 (1986), it is critical to identify precisely the governmental interests — the ends — to which the government’s use of race must be fitted. See United States v. Paradise, 480 U.S. 149, 171, 107 S.Ct. 1053, 94 L.Ed.2d 203 (1987) (noting that, in order to determine whether an order was narrowly tailored, “we must examine the purposes the order was intended to serve”). In other words, before we assume that the District shares the interest identified in Grutter, we must consider carefully the interests the District asserts and then determine whether the District’s interests are compelling.
The District’s interests are connected. First, the District seeks the affirmative educational benefits that flow from racial diversity (which, as I will discuss below, are different in K-12 education than in higher education). Second, and related, is the District’s interest in preventing its *991school assignment system from replicating Seattle’s segregated housing pattern;5 that is, the District has an interest in ensuring that each one of its students has access to the educational benefits of an integrated school environment.
A. The District has a compelling interest in the educational benefits of racial diversity in secondary education.
The District has established that racial diversity produces compelling educational benefits in secondary education.6 Because the educational benefits that the District seeks are materially different from those sought by the university in Grutter, however, the type of diversity required to produce those benefits is also different.
The university sought to further the academic and professional development of its students through the “livelier, more spirited, and simply more enlightening and interesting” classroom discussions that result when students have “the greatest possible variety of backgrounds.” Grutter, 539 U.S. at 330, 123 S.Ct. 2325 (internal quotation marks omitted). Aggrieved applicants accused the university of using race as an impermissible proxy for particular viewpoints and perspectives, but the Court disagreed, holding that racial diversity added to the mix of diversity factors by representing the “unique experience of being a racial minority in a society, like our own, in which race unfortunately still matters.” Id. at 333, 123 S.Ct. 2325. encompasses the K-12 context as a whole.
Although secondary-school educators share the university’s academic goals to some extent,7 achieving diversity of viewpoint and background is not the sole — or even the primary — reason for promoting an integrated secondary-school environment. Cf. Comfort ex rel. Neumyer v. Lynn Sch. Comm., 283 F.Supp.2d 328, 381 n. 90 (D.Mass.2003) (“The value of a diverse classroom setting at these ages does not inhere in the range of perspectives and experience that students can offer in discussions; rather, diversity is valuable because it enables students to learn racial tolerance by building cross-racial relationships.”).8
*992The District begins its own explanation of its interest in classroom diversity by noting the socialization and citizenship advantages of racially diverse schools. See maj. op. at 960-61 (quoting the School Board’s “Statement Reaffirming [the] Diversity Rationale”). Indeed, courts have recognized that the fundamental goal of K-12 education is to prepare children to be good citizens — to socialize children and to inculcate civic values.9 See Bethel Sch. Dist. No. 403 v. Fraser, 478 U.S. 675, 683, 106 S.Ct. 3159, 92 L.Ed.2d 549 (1986) (stating that the inculcation of civic values is “truly the work of the schools” (internal quotation marks omitted)); Plyler v. Doe, 457 U.S. 202, 221-23, 102 S.Ct. 2382, 72 L.Ed.2d 786 (1982) (noting that public education perpetuates the political system and the economic and social advancement of citizens); Ambach v. Norwich, 441 U.S. 68, 76-77, 99 S.Ct. 1589, 60 L.Ed.2d 49 (1979) (observing that public schools transmit to children “the values on which our society rests,” including “fundamental values necessary to the maintenance of a democratic political system”); Brown, 347 U.S. at 493, 74 S.Ct. 686 (“[Education] is required in the performance of our most basic public responsibilities.... It is the very foundation of good citizenship.”); see also Comfort, 283 F.Supp.2d at 381 n. 90 (“[A]t the elementary, middle, and high school level, the goal of teaching socialization is at least as important as the subject matter of instruction.”).10 In Washington, such civic training is mandated by the state constitution: “Our constitution is unique in placing paramount value on education for citizenship.” Parents Involved in Cmty. Schs. v. Seattle Sch. Dist., No. 1, 149 Wash.2d 660, 72 P.3d 151, 158 (Wash.2003).
In our society, in which “race unfortunately still matters,” Grutter, 539 U.S. at 333, 123 S.Ct. 2325, the “goals of teaching tolerance and cooperation among the races[ ][and] of molding values free of racial prejudice ... are integral to the mission of public schools,” Parents Involved, 72 P.3d at 162.11 Achieving those teaching goals requires the presence of a racially diverse student body. See Comfort, 283 F.Supp.2d at 376-77 (“If the compelling goal of the Plan is to train citizens to *993function in a multiracial world, actual in-tergroup racial contact is' essential.”). The District has emphasized the importance of interaction with peers of other races in educating students for citizenship; school officials, relying on their experience as teachers and administrators, and the District’s expert all explained these benefits on the record. Even Plaintiffs expert admitted that students are widely perceived to benefit from the information that they gain from increased contact with children of other races.12 See also Boston’s Children First v. City of Boston, 62 F.Supp.2d 247, 259 (D.Mass.1999) (“Diversity may well be more important at this stage than at any other — [because elementary school] is when first friendships are formed and important attitudes shaped....”). As the United States Supreme Court has noted, this educational goal is relevant for the entire community:
Attending an ethnically diverse school may help accomplish this goal by preparing minority children for citizenship in our pluralistic society while, we may hope, teaching members of the racial majority to live in harmony and mutual respect with children of minority heritage.
Washington v. Seattle Sch. Dist. No. 1, 458 U.S. 457, 473, 102 S.Ct. 3187, 73 L.Ed.2d 896 (1982) (citations and internal quotation marks omitted).
The majority misperceives the distinction between the interest recognized in Grutter and the one that I would recognize here. See maj. op. at 982-83. The District’s socialization and citizenship training is no more tangential or external to the educational experience of a secondary school than is academic training to the educational experience offered by a law school. The University of Michigan wanted to promote a stimulating academic environment so that its graduates would become accomplished and well-rounded members of the legal profession; the District wants to encourage integrated schools so that its graduates will become tolerant, productive, and well-adapted members of this racially diverse society. In both cases, although the benefits are external and long-term, the teaching occurs — and can be observed and evaluated — within thfe school environment.
In short, the District has a compelling interest in educating all students in a racially diverse learning environment, to educate them effectively to take their places in a racially diverse society.
B. The District has a compelling interest in reducing racial isolation and ameliorating de facto segregation.
The District’s interest in achieving the affirmative benefits of a racially diverse educational environment has a flip side: avoiding racially concentrated, or racially isolated, schools. In particular,' the District is concerned with making the educational benefits of an integrated school environment available to all its students. Thus, in addition to striving for better academic and social outcomes across the board, the District has been motivated by its belief that “[n]o student should be required to attend a racially concentrated school.”13 In other words, the Plan *994strives to ensure that patterns of residential segregation are not repeated as patterns of educational segregation that would be “determinative of a child’s opportunity.” Comfort, 283 F.Supp.2d at 384.
This “flip side” makes the District’s interest different from any that could have been posited in Grutter. Universities (like most other entities that select a few from among a pool of competitive applicants) are not, in any direct sense, responsible for the welfare of the entire universe of their applicants. That is, so long as all applicants are treated fairly in the competition for access to the limited government benefit, a university may design a class of students that satisfies its academic objectives without worrying about the effect that its admissions decisions have on rejected applicants.14 Public school districts, on the other hand, must consider not only the affirmative effect that a student’s assignment to a particular school will have on the level of diversity in that school, but also the concomitant effect of that assignment on the entire school system.
As the district court did in this case,15 several courts have conceived of a school district’s voluntary reduction or prevention of de facto segregation as a compelling interest. See Brewer v. W. Irondequoit Cent. Sch. Dist., 212 F.3d 738, 752 (2d Cir.2000) (holding that “a compelling interest can be found in a program that has as its object the reduction of racial isolation and what appears to be de facto segregation”), superseded on other grounds as stated in Zervos v. Verizon N.Y., Inc., 252 F.3d 163, 171 n. 7 (2d Cir.2001); Parent Ass’n of Andrew Jackson High Sch. v. Ambach, 738 F.2d 574, 579 (2d Cir.1984) (“[W]e held that the Board’s goal of ensuring the continuation of relatively integrated schools for the maximum number of students, even at the cost of limiting freedom of choice for some minority students, survived strict scrutiny as a matter of law.”) (citing Parent Ass’n of Andrew Jackson High Sch. v. Ambach, 598 F.2d 705, 717-20 (2d Cir.1979)); Comfort, 283 F.Supp.2d at 384-86 (holding that a school district had a compelling interest in ameliorating the effects of de facto residential segregation); Hampton v. Jefferson County Bd. of Educ., 102 F.Supp.2d 358, 379 (W.D.Ky.2000) (noting that “voluntary maintenance of the desegregated school system should be considered a compelling state interest,” such that a district may consider race in assigning students to comparable schools).
None of the school districts in the above-cited cases was subject to a court-ordered *995desegregation decree nor, with one exception, did the schools face an imminent threat of litigation to compel desegregation.16 Like the Seattle School District, they may have been vulnerable to litigation in decades past,17 but the districts’ voluntary desegregation measures today would make it difficult to make the required showing that the districts intended to create segregated schools. See, e.g., Comfort, 283 F.Supp.2d at 390 (explaining that the district’s vulnerability to litigation had been “headed off by the very Plan in contention here”). It is well established that school districts have no obligation to remedy de facto (as distinct from de jure) segregation. Freeman v. Pitts, 503 U.S. 467, 495, 112 S.Ct. 1430, 118 L.Ed.2d 108 (1992). Nevertheless, several courts have pointed out the irony of a conclusion that a measure that could be required to remedy segregation could not be adopted voluntarily to prevent segregation. See, e.g., Comfort, 283 F.Supp.2d at 384-85 (“It would make no sense if [school] officials were obliged to take responsibility for addressing these adverse consequences [of segregated schools] but at the same time were constitutionally barred from taking voluntary action aimed at nipping some of these effects in the bud.”); Parents Involved in Cmty. Schs. v. Seattle Sch. Dist. No. 1, 137 F.Supp.2d 1224, 1235 (W.D.Wash.2001) (“[I]t would defy logic for this court to find that the less intrusive programs of today violate the Equal Protection Clause while the more coercive programs of the 1970’s did not.”); Hampton, 102 F.Supp.2d at 379 (“It is incongruous that a federal court could at one moment require a school board to use race to prevent resegregation of the system, and at the very next moment prohibit that same policy.”).
In essence, what these courts have recognized is that school districts have a prospective, even if not a remedial, interest in avoiding and ameliorating real, identifiable de facto racial segregation. Support for this conclusion comes from statements in the Supreme Court’s school desegregation cases, which repeatedly refer to the voluntary integration of schools as sound educational policy within the discretion of local school officials. See Swann v. Charlotte-Mecklenburg Bd. of Educ., 402 U.S. 1, 16, 91 S.Ct. 1267, 28 L.Ed.2d 554 (1971) (stating that school authorities “are traditionally charged with broad power to formulate and implement educational policy and might well conclude ... that in order to prepare students to live in a pluralistic society each school should have a prescribed ratio of Negro to white students reflecting the proportion for the district as a whole”); N.C. State Bd. of Educ. v. Swann, 402 U.S. 43, 45, 91 S.Ct. 1284, 28 L.Ed.2d 586 (1971) (“[A]s a matter of educational policy school authorities may well conclude that some kind of racial balance in the schools is desirable quite apart from any constitutional requirements.”); Bustop, Inc. v. Bd. of Educ. of L.A., 439 U.S. 1380, 1383, 99 S.Ct. 40, 58 L.Ed.2d 88 (Rehnquist, Circuit Justice 1978) (denying a request to stay implementation of a de-*996segregation plan and noting that there was “very little doubt” that the Constitution at least permitted its implementation); Keyes v. Sch. Dist. No. 1, 413 U.S. 189, 242, 93 S.Ct. 2686, 37 L.Ed.2d 548 (1973) (Powell, J., concurring in part and dissenting in part) (“School boards would, of course, be free to develop and initiate further plans to promote school desegregation.... Nothing in this opinion is meant to discourage school boards from exceeding minimal constitutional standards in promoting the values of an integrated school experience.”); Seattle Sch. Dist., 458 U.S. at 480, 487, 102 S.Ct. 3187 (reinstating the Seattle School District’s authority to use mandatory busing to correct de facto segregation).
Of course, these statements must be considered in the light of the Court’s later decisions in City of Richmond v. J.A. Cro-son Co., 488 U.S. 469, 507, 109 S.Ct. 706, 102 L.Ed.2d 854 (1989) (holding that “outright racial balancing” did not constitute a permissible reason to establish a quota for awarding construction contracts); Freeman, 503 U.S. at 494, 112 S.Ct. 1430 (holding that, in the absence of a constitutional violation, the district court had no power to order “[rjacial balance ... for its own sake”); and Grutter, 539 U.S. at 329-30, 123 S.Ct. 2325 (stating that the law school’s concept of “critical mass” was permissible only because it was not “outright racial balancing,” but rather was “defined by reference to ... educational benefits”). Those decisions establish that the government may not act in furtherance of racial balance without a compelling nonraeial reason. Unless and until the Supreme Court says otherwise, however, I would heed its repeated statements that the voluntary integration of public schools, in response to specific conditions of de facto segregation and in furtherance of legitimate educational policies, can be a constitutionally permissible interest.
In sum, I would hold that the District has a compelling interest in structuring its assignment policies to prevent a return to the era in which Seattle’s undisputedly segregated housing pattern was the exclusive determinant of school assignments to neighborhood schools.
C. Deference to administrators’ expertise in education policy is warranted.
In addition to signaling specifically its approval of voluntary measures to promote integrated schools, the Supreme Court repeatedly has shown deference to school officials at the intersection between constitutional protections and educational policy. See generally Wendy Parker, Connecting the Dots: Grutter, School Desegregation, and Federalism, 45 Wm. & Mary L.Rev. 1691 (2004). Local control over public education has animated Supreme Court jurisprudence from the dawn to the apparent twilight of federal-court involvement in the desegregation of public schools. See, e.g., Brown, 349 U.S. at 299, 75 S.Ct. 753 (directing local school officials, with court oversight, to devise remedies for segregation in the light of “varied local school problems”); Milliken v. Bradley, 418 U.S. 717, 741-42, 94 S.Ct. 3112, 41 L.Ed.2d 1069 (1974) (“No single tradition in public education is more deeply rooted than local control over the operation of schools; local autonomy has long been thought essential both to the maintenance of community concern and support for public schools and to quality of the educational process.”); Freeman, 503 U.S. at 490, 112 S.Ct. 1430 (“As we have long observed, ‘local autonomy of school districts is a vital national tradition.’ ”) (quoting Dayton Bd. of Educ. v. Brinkman, 433 U.S. 406, 410, 97 S.Ct. 2766, 53 L.Ed.2d 851 (1977)). In the context of a challenge to a school-funding system, the Court was motivated, in part, by concerns about the judiciary’s lack of *997competency in the area of educational policy. See San Antonio Indep. Sch. Dist. v. Rodriguez, 411 U.S. 1, 42, 93 S.Ct. 1278, 36 L.Ed.2d 16 (1973) (stating, in its rational-basis review of a school-funding system, that “this case ... involves the most persistent and difficult questions of educational policy, another area in which this Court’s lack of specialized knowledge and experience counsels against premature interference with the informed judgments made at the state and local levels”).18 Thus, although I agree that public secondary schools do not have “a constitutional right to select their student body,” maj. op. at 981-82,19 they have been given considerable discretion to devise school assignment policies, even in the face of adjudicated constitutional violations.
The Supreme Court also has shown solicitude toward the educational objectives of public school administrators in balancing those educational objectives with students’ First Amendment rights. See, e.g., Hazelwood Sch. Dist. v. Kuhlmeier, 484 U.S. 260, 273, 108 S.Ct. 562, 98 L.Ed.2d 592 (1988) (holding that educators may censor student speech in school-sponsored forums for valid educational reasons and noting that “[tjhis standard is consistent with our oft-expressed view that the education of the Nation’s youth is primarily the responsibility of parents, teachers, and state and local school officials, and not of federal judges”); see also Comfort, 283 F.Supp.2d at 374 & n. 83 (citing Supreme Court cases involving the balancing of schools’ curricular needs against students’ rights under the First, Fourth, and Eighth Amendments, as well as the Due Process Clause of the Fourteenth Amendment). This deference recognizes not merely a school’s need to preserve order so as to promote pure “academic” learning, as the majority suggests, maj. op. at 982, but also to teach students that certain kinds of discourse are “wholly inconsistent with the ‘fundamental values’ of public school education,” Bethel Sch. Dist., 478 U.S. at 685-86, 106 S.Ct. 3159.
These Supreme Court decisions suggest that secondary schools, like universities, occupy a “special niche” in our constitutional tradition, albeit one that owes more to the values of federalism and to the public schools’ broad educational mission than to a desire to safeguard academic freedom. For this reason, I would afford some deference to the District’s judgment that integrated schools are essential to its educational mission and would extend to the District’s identification of its core values a deference similar to that which the Grutter Court afforded the university. See Grutter, 539 U.S. at 328, 123 S.Ct. 2325; cf. Petit v. City of Chicago, 352 F.3d 1111, 1114 (7th Cir.2003) (extending deference, pursuant to Grutter, to the “views of experts and Chicago police executives that affirmative action was warranted to enhance the operations” of the Chicago Police Department), cert. denied, — U.S. —, 124 S.Ct. 2426, 158 L.Ed.2d 984 (2004).
In sum, I am convinced by the record, as well as by deference to the District’s expertise in educational policy, that the District’s interests in obtaining the educational benefits of diversity in secondary education and in ameliorating the de facto *998segregation caused by Seattle’s segregated housing pattern are compelling as a matter of law.
II. The District’s Plan is narrowly tailored to achieve its compelling governmental interests.
The narrow-tailoring inquiry is intended to “ ‘smoke out’ illegitimate uses of race” by ensuring that the government’s classification is closely fitted to the compelling goals that it seeks to achieve. Croson, 488 U.S. at 493, 109 S.Ct. 706. As discussed above, the analysis must fit the context of the challenged governmental action. Grutter, 539 U.S. at 327, 333-34, 123 S.Ct. 2325. For example, the factors that the Court uses to assess narrow tailoring in the employment context, Paradise, 480 U.S. at 171, 107 S.Ct. 1053,20 must be modified for use in the context of higher education. See Johnson v. Bd. of Regents of Univ. of Ga., 263 F.3d 1234, 1252 (11th Cir.2001) (“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.”). Likewise, some of the factors used to assess programs related to employment or higher education are of doubtful relevance in the context of K-12 school assignment plans. See Hampton, 102 F.Supp.2d at 380 (“The workplace, marketplace, and higher education cases are poor models for most elementary and secondary public school education.... ”). We must consider which narrow-tailoring factors are appropriate to this context; our inquiry
pivots not merely on the fact that race is used in a school plan, but how it is used, in what settings, for what purposes, whether it is race conscious or race preferential, whether it involves an examination school (or a college or law school) for which there are significant qualifications, or an elementary school, for which there are not, whether the use of race excludes or simply affects the distribution of a benefit....
Boston’s Children, 62 F.Supp.2d at 259.
Because of the differences in setting, several of the narrow-tailoring factors employed by the Supreme Court in Grutter and Gratz — and by the majority in this case — have no logical relevance to the evaluation of secondary-school assignment plans like the District’s. After fashioning an appropriately contextualized narrow-tailoring analysis, I will consider the Plan in its broader historical and factual context and conclude that the Plan satisfies strict scrutiny.
A. Narrow-tailoring factors involving “holistic review” and “quotas” have no relevance in the context of assigning students to secondary schools.
For two reasons, cases involving selective admissions to institutions of higher learning do not provide a proper “narrow tailoring” model for this case.21 First, *999they involve situations in which a school grants or denies access to a limited government benefit based on the school’s evaluation of a particular applicant’s merit; using race as a proxy, or as a substitute, for merit in awarding this benefit raises problems of stereotyping and stigma that are absent from the District’s Plan. Second, the institutions involved in those cases seek the “trae diversity” befitting their advanced academic orientations; as I have discussed, the diversity interest in the K-12 context involves different educational benefits and, like the District’s related interest in ameliorating de facto segregation, is more appropriately achieved through an explicit consideration of racial diversity.
1. Where competition for a limited government resource is absent, rules about how competition may be conducted are irrelevant.
In Grutter, the Supreme Court “define[d] the contours of the narrow-tailoring inquiry with respect to race-conscious university admissions programs.” See Grutter, 539 U.S. at 333, 123 S.Ct. 2325. In the context of university admissions, where applicants compete for a limited number of spaces in a class, the Court focused its inquiry on what role race may play in judging an applicant’s qualifications. The Court’s underlying concern is for fair competition — to prevent race from being used as an outright substitute for merit in the competition for access to a limited government resource, in part because of the stigma that may attach. See Regents of Univ. of Cal. v. Bakke, 438 U.S. 265, 298, 98 S.Ct. 2733, 57 L.Ed.2d 750 (1978) (Powell, J., concurring) (stating that “preferential programs may only reinforce common stereotypes holding that certain groups are unable to achieve success without special protection based on a factor having no relationship to individual worth”); Croson, 488 U.S. at 493, 109 S.Ct. 706 (“Classifications based on race carry a danger of stigmatic harm. Unless they are strictly reserved for remedial settings, they may in fact promote notions of racial inferiority and lead to a politics of racial hostility.”); see also Gratz, 539 U.S. at 272-73, 123 S.Ct. 2411 (applying the narrow-tailoring inquiry to ensure that applicants of all races have the opportunity to prove their merit based on a broad range of criteria). In Grutter, the Court discussed two specific rules to ensure fair competition. 539 U.S. at 334, 123 S.Ct. 2325. The first prohibits “quotas,” which insulate applicants from certain groups from competition with applicants from other groups for some portion of the available slots.22 Id. *1000at 335, 123 S.Ct. 2325. The second rule prevents race from being used as a mechanical proxy for merit by requiring individualized consideration of the merit of each applicant, across a broad range of factors (of which race may be but one). Id. at 336-37, 123 S.Ct. 2325.
Neither of those requirements, which concern how universities are permitted to evaluate merit, is relevant in a situation where there is absolutely no competition or consideration of merit at issue. All high school students must and will be placed in a Seattle public school. The students’ relative merit is irrelevant. There are no special qualifications for assignment to any school, so no stigma results from any particular school assignment.23 The dangers of substituting racial preference for fair, merit- or worth-based competition are absent here. Justice Powell recognized this very fact in his landmark opinion in Bakke, which reasoned that the use of racial classifications to desegregate schools was fundamentally different from the selective admissions context because, in the school assignment context, “white students [were not] deprived of an equal opportunity for education.” Bakke, 438 U.S. at 301 n. 39, 98 S.Ct. 2733.
Justice Powell’s comment suggests an even more fundamental reason why a careful, holistic, individualized consideration of an applicant’s worth is not necessary here: No student is being excluded from the government resource at issue — a free public education.24 In the assignment process, all students are accommodated with the same baseline high school education.25 As *1001I will discuss in more depth below, differences among the high schools may be relevant to our consideration of the “burdens” that the Plan imposes, but perceived or actual differences in academic quality do not transform the District’s assignment process into a competition for access to a limited government resource. This is especially true where it is clear that every student can enroll in at least one of Seattle’s oversubscribed, “quality” high schools.
2. Where “true diversity” is not the goal, consideration of a broad range of diversity factors is unnecessary.
Another rationale for the Court’s requirement of holistic review is that it provides a closer fit with a university’s interest in viewpoint diversity. Grutter, 539 U.S. at 337, 123 S.Ct. 2325. The Court held that it was impermissible to presume that race would correlate with viewpoint, perspective, or background; rather, the university must evaluate each applicant’s viewpoint, perspective, and background. Id. at 338-39, 123 S.Ct. 2325.
The danger that race would be used to “fill[ ] stereotyped ‘viewpoint’ niches,” Comfort, 283 F.Supp.2d at 379, is not present here. The diversity interest in K-12 education is much simpler: that children learn to interact with peers of different races. That interest requires that there be children of different races in the classroom. Rather than relying on stereotypes, intergroup contact has the opposite effect; it inhibits the formation of stereotypes by teaching children that “all people are different no matter what their color or ethnic background.” Id. In other words, the District’s focus on racial diversity is not a presumption that students of the same race will share common viewpoints; the presence of students of different races is meant to break down, rather than to further, racial stereotypes by giving students an opportunity to learn that race does not signal an individual’s viewpoint, perspective, or background.
Moreover, the holistic review necessary to achieve “true diversity” is of even less relevance to the District’s interest in preventing and ameliorating de facto racial segregation. As the Second Circuit has said:
If reducing racial isolation is- — standing alone — a constitutionally permissible goal, as we have held it is ..., then there is no more effective means of achieving that goal than to base decisions on race. ... [T]he cases cited by the District Court in support of its decision that the use of race alone in the Program was not narrowly tailored only address the efficacy of employing strictly racial classifications to achieve “true diversity.” Those decisions are, therefore, inapplicable to the present situation where the Program’s aim ... is precisely to ameliorate racial isolation in the participating districts.
Brewer, 212 F.3d at 752-53 (citations omitted). In other words, when the District’s compelling interest is in racial diversity, it makes little sense to ask it instead to evaluate a student’s musical talent, athletic prowess, or eligibility for a free lunch.26
*1002B. Viewed in its historical and factual context, the District’s Plan satisfies an appropriate narrow-tailoring test.
Except for rejecting the narrow-tailoring factors peculiar to situations of competition and “true diversity,” I have no disagreement with the majority’s identification of the remaining narrow-tailoring factors. As I will discuss, I conclude that the Plan satisfies the narrow-tailoring test. But first, in order to facilitate an accurate narrow-tailoring inquiry, I believe it is necessary to supplement the majority’s statement of facts by placing the District’s adoption of the Plan in a broader factual context.
1. The broader context of desegregation efforts and the Board’s decision to adopt the Plan.
The increase in Seattle’s minority population after World War II was concentrated first in the central, then in the southeast, area of the city.27 Because school assignments were made strictly on the basis of neighborhood, schools in the central and southeast areas reflected that population concentration. In 1962, the central area’s Garfield High School reported 64 percent minority enrollment (it accommodated 75 percent of all black students), and six of the central area elementary schools had at least 75 percent minority enrollment.28 Meanwhile, the eight high schools serving other major areas of the city remained more than 95 percent white.
In the 1960’s, the District responded to this imbalance in various ways. It granted the central area principals’ request for special financial assistance. It responded to racial tensions by experimenting with exchange programs, in which a handful of students switched high schools for several weeks. And in 1963, the District implemented a voluntary racial transfer program, through which a student could transfer to any school with available space, if the transfer would improve the racial balance at the receiving school. Although this program had some positive results, it did not reduce the racial imbalance significantly.
In the 1970’s, the District stepped up its efforts. It adopted a plan to desegregate central-area middle schools by requesting volunteers to transfer between minority- and majority-dominated neighborhood schools and ordering mandatory transfers when the numbers of volunteers were insufficient. The District also took steps to desegregate Garfield High School by changing its educational program, improving its facilities, and eliminating the “special transfers” that had allowed white students to leave the Garfield area. In addition, for the 1977-78 school year, the District instituted a magnet-school program to promote desegregation. According to the District’s history:
*1003While it appeared evident that the addition of magnet programs would not in itself desegregate the Seattle schools, there was supportive evidence that voluntary strategies, magnet and non-magnet, could be significant components of a more comprehensive desegregation plan.
By the 1977-78 school year, Franklin was 78 percent minority, Rainier Beach 58 percent, Cleveland 75 percent, and Garfield 64 percent. Other high schools ranged from 9 percent to 23 percent minority enrollment, with one school (Lincoln) at 37 percent. See Seattle Sch. Dist., 458 U.S. at 461, 102 S.Ct. 3187 (noting that the racial imbalance in Seattle’s schools had increased between the 1970-71 and 1977-78 school years).
In the spring of 1977, the NAACP filed a complaint with the Office of Civil Rights, alleging that Seattle’s School Board had acted to further racial segregation in the city’s schools. Several other organizations, principally the ACLU, threatened to file an action in court if the District failed to adopt a mandatory desegregation plan. When the District agreed to develop a desegregation plan, the Office of Civil Rights concomitantly agreed to delay its investigation, and the ACLU agreed to delay filing a lawsuit. See Seattle Sch. Dist., 458 U.S. at 460 n. 2, 102 S.Ct. 3187 (describing this threat of litigation).
During the summer of 1977, the District and community representatives reviewed five model plans. Ultimately, the District incorporated elements of each model into its final desegregation plan, adopted in December 1977 and known as the “Seattle Plan.” The Seattle Plan divided the district into zones, within which majority-dominated elementary schools were paired with minority-dominated elementary schools to achieve racial balance. Mandatory high school assignments were linked to elementary-school assignments, although various voluntary transfer options were available. With the Seattle Plan,
Seattle became the first major city to adopt a comprehensive desegregation program voluntarily without a court order. By doing so the District maintained local control over its desegregation plan and was able to adopt and implement a plan which in the eyes of the District best met the needs of Seattle students and the Seattle School District.
“History of Desegregation” at 36-37. An initiative was passed immediately to block implementation of the Seattle Plan, but the initiative ultimately was declared unconstitutional by the United States Supreme Court. Seattle Sch. Dist., 458 U.S. at 470, 102 S.Ct. 3187 (holding that the initiative violated equal protection).
The Seattle Plan apparently furthered the District’s school desegregation goals, but its operation was unsatisfactory in other ways.29 In 1988, the District abandoned the Seattle Plan and adopted a new plan that it referred to as “Controlled Choice.” Under the Controlled Choice plan, schools were grouped into clusters that met state and district desegregation guidelines, and families were permitted to rank schools within the relevant cluster, increasing the predictability of assignments. Because of Seattle’s housing patterns, the District’s planners explained that “it was impossible to fashion clusters in a geographically contiguous manner”; some cluster schools were near the student’s home, but others were in “racially and culturally different neighborhoods.” Although roughly 70 per*1004cent of students received their first choices, the Controlled Choice plan still resulted in mandatory busing for 16 percent of the District’s students.
In the mid-1990’s, District staff were directed to devise a new plan for all grade levels to simplify assignments, reduce costs, and increase community satisfaction, among other things; the guiding factors were to be choice, diversity, and predictability. Staff developed four basic options, including the then-existing Controlled Choice plan, a regional choice plan, a neighborhood assignment plan with provision for voluntary, integration-positive transfers, and an open choice plan.
Board members testified that they considered all the options, as they related to the District’s educational goals — with special emphasis, at the secondary-school level, on the goals of choice and diversity. Neighborhood and regional plans were viewed as unduly limiting student choice, on which the District placed high value because it was seen to increase parental involvement in the schools and promote improvements in quality through a marketplace model. The District sought to maintain its commitment to integrated education by establishing diversity goals, but moving away from the rigid desegregation guidelines and mandatory assignments prevalent in the 1970’s and 1980’s. The Board adopted the current, open choice Plan for the 1998-99 school year.
The Plan now under review permits students to rank their choices among the District’s 10 high schools. The District has sought to make each of the 10 schools unique, with programs that attempt to respond to the continually changing needs of students and their parents.30 Only when oversubscription results from families’ choices — as, of course, it has — does the District become involved in the assignment process. Assignments to oversubscribed schools proceed by way of a series of tiebreakers: first, students with a sibling attending the selected school are assigned; second, if but only if the school deviates from the District’s proportion of white and minority students by more than a specified percentage,31 students who bring that school closer to the ratio are assigned; third, students are assigned in order of the distance from their homes to the school.32 The first and third tiebreakers seek to further the District’s goal of parental involvement, and the second is directed toward the District’s diversity goal. Students not assigned to one of their chosen schools are assigned to the closest school with space available; naturally, students who list more choices are less likely to receive one of these “mandatory” assignments.
Having examined the District’s interests and the specifics of the Plan in its historical context, I will turn next to a consideration of whether the Plan is narrowly tailored to serve the compelling governmental interests that I have identified.
2. The District’s Plan is narrowly tailored.
A narrow-tailoring analysis requires consideration of three traditional groups of
*1005factors: (1) the necessity for the action and the efficacy of alternative, race-neutral remedies; (2) the extent to which the action is proportional to the District’s interests (particularly, whether it is of limited duration and is flexible, in relation to its objective); and (3) the relative weight of any burden on third parties. See Paradise, 480 U.S. at 171, 107 S.Ct. 1053; see also Comfort, 283 F.Supp.2d at 371-73. As I stated above, the purpose of this inquiry is to “ ‘smoke out’ illegitimate uses of race” by ensuring that the government’s means are closely fitted to its ends. Croson, 488 U.S. at 493, 109 S.Ct. 706.
(a) The Plan achieves the District’s diversity goals more effectively than any workable race-neutral alternative.
(i) Need for the integration tiebreaker
The integration tiebreaker allows the District to make students’ and parents’ choices among high schools the primary feature of its educational plan,33 while discouraging a return to enrollment patterns based on Seattle’s racially segregated housing pattern. When the District moved from its Controlled Choice plan to the current, open choice Plan, it predicted that families would tend to choose schools close to their homes. Indeed, this feature was seen as a positive way to increase parental involvement. However, unfettered choice — especially with tiebreakers based on neighborhood or distance from a school' — -raised the risk that Seattle’s high school enrollment would begin to reflect its segregated housing patterns. The District’s 2000-01 enrollment data showed that, of the students living in the southern half of Seattle, only 23 percent are white (6,247 out of 27,377 students), as compared with 64 percent of the students living in the northern half of the city (12,571 out of 19,555 students).
It is de facto residential segregation across this white/ nonwhite axis that the District has battled historically and that it sought to prevent by making the integration tiebreaker a part of its open choice Plan. Although I have no doubt that other forms of race-based tension exist, see maj. op. at 983-84 n. 47, the District reasonably placed its focus here. The District has consistently faced a pattern in which its white students live predominately in the northern half of the city (in 2000-2001, 66.8 percent of the District’s white students lived in the northern half of the city) and its students of color — in each of the three largest categories that the District tracks34 — live predominately in the southern half of the city. In 2000-2001, 74.2 percent of the District’s Asian students, 83.6 percent of its Black students, and 65 percent of its Hispanic students lived in the southern half of the city:
[[Image here]]
*1006Moreover, Seattle’s peculiar geography-makes its northern neighborhoods and schools distant from its southern neighborhoods and schools. In these circumstances, the District permissibly could prioritize white/nonwhite, primarily north/ south, movement. This white/nonwhite focus also is consistent with the history of public school integration measures in this country, as reflected in a current federal regulation defining “[m]inority group isolation” as “a condition in which minority group children constitute more than 50 percent of the enrollment of the school,” without distinguishing among the various categories included within the definition of “[mjinority group.” 34 C.F.R. § 280.4(b).
To discourage choices that would perpetuate this north-south division between the district’s white and nonwhite populations, the District gave priority to choices that would counter it and create north-south movement within the District. In the 2000-01 school year, the integration tiebreaker operated in four high schools (that is, four high schools were oversubscribed and deviated by more than 15 percent from the ratio of white to nonwhite students District-wide). Although the integration tiebreaker was a limited measure, in contrast to the District’s previous efforts, it did serve to alter the imbalance in the schools in which it operated.
The majority’s contrary view is based on a skewed presentation of the enrollment statistics. Figures reflecting the tiebreaker’s total effect on a school’s enrollment, such as those cited in the majority opinion, see maj. op. at 984 (“Table 2”), 985 (“Table 3”), 985 (citing a maximum shift of 6.1 percent in the white/nonwhite ratio), artificially minimize the tiebreaker’s effect by failing to recognize that students enter the ninth grade in much greater numbers than they transfer to other schools after the ninth grade. The following statistics submitted by the District, which portray directly the effect of the tiebreaker on the make-up of the ninth grade classes at the four affected schools, illustrate the function of the tiebreaker far more accurately:
[[Image here]]
In other words, the majority’s references to a 6.1 percent maximum shift do not tell the whole story.
Still, without the integration tiebreaker, the freshman classes at some of the affected north-end schools may well have been sufficiently diverse to promote interaction across the white/nonwhite axis and to prevent the tokenization of nonwhite students. See maj. op. at 984 n. 49. However, without the integration tiebreaker, the Plan would have operated to prevent students of color who lived in the south end of Seattle from attending those schools because of the schools’ distance from south-end neighborhoods. The tiebreaker furthered the District’s goal of giving south-end students of color the opportunity to *1007opt out of attending the more racially concentrated schools in their neighborhoods, if they so desired.
Certainly, the integration tiebreaker does not attempt to achieve perfect adherence to the District-wide ratios in each of the District’s high schools. Except by encouraging an optout, the tiebreaker does not directly alter the racial make-up of schools that are not oversubscribed, even that of the south-end schools 'that diverge widely from District-wide proportions. See supra note 13. I do not, however, view the Plan’s underinclusiveness as a fatal flaw.
Indeed, I find it peculiar that the majority’s rebuttal makes so much of the failure of the District’s Plan to achieve perfect racial balance. To be sure, in strict scrutiny review, especially in the First Amendment context, a law’s underinclusiveness can be a sign that the enacting authority was not in fact motivated by its stated objectives. See Republican Party of Minn. v. White, 536 U.S. 765, 780, 122 S.Ct. 2528, 153 L.Ed.2d 694 (2002) (noting that underinclusiveness impairs the credibility of the government’s rationale for restricting speech) (citing City of Ladue v. Gilleo, 512 U.S. 43, 52-53, 114 S.Ct. 2038, 129 L.Ed.2d 36 (1994)). Apparently, in order for the majority to find that a race-conscious means is necessary to achieve the District’s stated goals, the means would have to produce perfect adherence to the District’s racial make-up — which, in Seattle’s circumstances, only aggressive districting, forced busing, and more intrusive racial classifications could do. However, requiring that the chosen means achieve perfect balance would make strict scrutiny impossible to satisfy, because the burden caused by the Plan — the third element of the narrow-tailoring inquiry— would be enormous. That is why, in contexts more relevant to this case, such as employment, courts have found that the modesty — i.e., the underinclusiveness — of government action is a point in favor of a conclusion that the action was narrowly tailored. See, e.g., Cotter v. City of Boston, 323 F.3d 160, 171 (1st Cir.) (“The necessity for relief was great, but the means chosen by the Department were modest — only three African-American officers were promoted out of rank — indicating narrow tailoring.”), cert. denied, — U.S. —, 124 S.Ct. 179, 157 L.Ed.2d 47 (2003).
Because strict scrutiny is not meant to be “fatal in fact,” Grutter, 539 U.S. at 326, 123 S.Ct. 2325, I would hold that the District’s modest measures, which were enacted to decrease the intrusiveness and burden of its assignment policy, do not cause its Plan to become unnecessary or the District’s motivations to become suspect. And, the Plan furthers the District’s goals better than any workable race-neutral alternative.
(ii) Race-neutral alternatives
In Grutter, the Court explained that narrow tailoring “require[s] serious, good faith consideration of workable race-neutral alternatives that will achieve the diversity the university seeks.” 539 U.S. at 339, 123 S.Ct. 2325 (emphasis added). On the other hand, “[njarrow tailoring does not require exhaustion of every conceivable race-neutral alternative.”35 Id. Fur*1008thermore, the Court made clear that the university was not required to adopt race-neutral measures that would have forced it to sacrifice other educational values central to- its mission. Id. at 340, 123 S.Ct; 2325. Implicit in the Court’s analysis was a measure of deference toward the university’s identification of those values. See id. at 328, 340, 123 S.Ct. 2325 (affording deference to the university’s judgment that diversity and “academic selectivity” were important to its educational mission). By affording deference to the university’s identification of its core educational values in this context, the Court simply recognized that whether a race-neutral measure is truly an alternative in the first place depends on whether that measure is consistent with an institution’s core values.
The majority faults the District for failing to consider seriously three specific race-neutral measures. Maj. op. at 970-975. One of the majority’s suggestions— that the District measure and assign students according to their “true diversity,” maj. op. at 971-72 — cannot properly be considered an “alternative,” because it is not directed toward achieving the District’s interest in a racially integrated learning environment. See supra pp. 1001. Furthermore, it is clear from the record that the school board has discussed the use of other diversity measures, including poverty, as a tiebreaker. Although the majority correctly points out that there has been no formal study of that proposal by District staff, Board members’ testimony reveals two legitimate reasons why the majority of the Board rejected the use of poverty measures to reach its goal of racial diversity: one, it is insulting to minorities and often inaccurate to assume that the two populations are coextensive; and, two, implementation would be thwarted by high school students’ reluctance to reveal their socioeconomic status to their peers.
The majority also asserts that the District should have considered more formally a proposal developed by the Urban League (which, incidentally, did not eliminate the integration tiebreaker, but merely demoted it). Maj. op. at 973-75. The majority quotes at length from the colorful and often emotional testimony of one Board member, who clearly was not impressed by the proposal. But there was other testimony from other Board members suggesting that the Board was aware of, and informally considered, the Urban League’s proposal. Board member Schaad-Lam-phere testified that she remembers reading the Urban League’s plan and considered it to be similar to other regional assignment plans being proposed at the same time; she testified that she weighed it as she did other community input. Furthermore, the testimony of Board member Nielsen is consistent with Superintendent Olchefske’s understanding that regional plans, including the Urban League’s, had been disfavored by the Board because of the high value the District placed on choice among the different academic offerings at the various high 'schools. In other words, when all the testimony in the record is examined, it is clear that the Urban League’s plan was in fact considered and that it was rejected for legitimate educational reasons.36
*1009The majority concludes that such informal consideration is inadequate and that we should evaluate the District’s consideration of race-neutral alternatives using the same rigorous evidentiary standard that we use to evaluate whether a local government has proved that past discrimination justified its enactment of a remedial minority set-aside program. Maj. op. at 972 (citing cases that elucidate the latter standard). To the contrary, our cases on set-aside programs plainly employ different standards to these different analyses— with a more permissive view toward the analysis of race-neutral alternatives. Compare Coral Constr. Co. v. King County, 941 F.2d 910, 916-22 (9th Cir.1991) (discussing, within its “compelling government interest” analysis, the high burden of demonstrating actual discrimination by the county), with id. at 923 (noting, under the race-neutral alternatives prong of the narrow-tailoring analysis, that “some degree of practicality is subsumed in the exhaustion requirement” and that exhaustion of every possible alternative is not required); see also Associated Gen. Contractors of Cal., Inc. v. Coalition for Econ. Equity, 950 F.2d 1401, 1416-17 (9th Cir.1991) (citing Coral). Under the relevant standard, the District adequately considered the alternatives.
Finally, the majority would require the District to “earnestly appraise[]” a random, citywide lottery for high school assignments. Maj. op. at 970 (emphasis omitted). In view of the District’s clear commitment to educational choice among high schools, and in view of its desire to provide students with an opportunity to attend school closer to home, the majority’s suggestion flatly contradicts the Grut-ter Court’s approach to narrow tailoring:
The District Court took the Law School to task for failing to consider race-neutral alternatives such as “using a lottery system”.... But [this] alternative! ] would require a dramatic sacrifice of diversity, the academic quality of all admitted students, or both.
... We are satisfied that the Law School adequately considered race-neutral alternatives currently capable of producing a critical mass without forcing the Law School to abandon the academic selectivity that is the cornerstone of its educational mission.
539 U.S. at 340, 123 S.Ct. 2325. The university, in other words, was not required to “earnestly appraise!],” maj. op. at 970 (emphasis omitted), a lottery system. The majority is incorrect in its conclusion that the District must do so, despite the harm that a lottery would do to the goals of choice and parental involvement that lie at the heart of its educational mission.37
It is a closer question whether the District’s goals could be met by using a pure lottery tiebreaker — that is, a lottery to determine which of the students who had chosen a particular school would be enrolled there. This format would allow the District to retain its emphasis on choice and would cause unhappiness more ran*1010domly. However, as Superintendent 01-chefske explained, District patterns suggested that more people would choose schools close to home, thus raising the justifiable concern that the pool of students choosing a particular school would be skewed in favor of the demographic of the surrounding residential area. Indeed, when the District adopted the Plan, it could not predict exactly how much harm open choice would do to the racial diversity of its schools. A lottery, in the face of this uncertainty, would have left the District without a safety net for its diversity goals and, moreover, would have prevented the District from furthering the policy goals reflected in its sibling and proximity tiebreakers.
Over the long history of its efforts to achieve integrated schools, the District has experimented with many alternatives, including magnet and other special-interest programs, which it continues to employ, and race-conscious districting.38 But when a racially integrated school system is the goal (or racial isolation is the problem), there is no more effective means than a consideration of race to achieve a solution. Even Plaintiffs expert conceded that, “if you don’t consider race, it may not be possible to offer an integrated option to students.... [I]f you want to guarantee it you have to consider race.” As Superintendent Olchefske stated, “when diversity, meaning racial diversity, is part of the educational environment we wanted to create, I think our view was you took that issue head on and used — you used race as part of the structures you developed.” The logic of this point is sound: When race is a principal element of the government’s compelling interest, then race-neutral alternatives seldom will be equally efficient. Cf. Hunter v. Regents of Univ. of Cal., 190 F.3d 1061, 1066 (9th Cir.1999) (upholding, as narrowly tailored, a research elementary school’s admissions policy that explicitly considered race in pursuit of a racially balanced research sample). Of course, race-conscious remedies still must be proportional to the government’s interest.
(b) The Plan satisfies requirements of proportionality, flexibility, and limited duration.
The District’s plan is proportional to its interests and is sufficiently flexible and time-limited to meet the requirements of narrow tailoring.
(i) Proportionality
To determine whether the means adopted are proportional to the government’s interest, courts have considered the “relationship between the numerical relief ordered and the percentage of nonwhites in the relevant [school population].” Paradise, 480 U.S. at 179, 107 S.Ct. 1053; see Tuttle v. Arlington County Sch. Bd., 195 F.3d 698, 706 (4th Cir.1999) (per curiam) (applying Paradise’s proportionality analysis to K-12 student assignment plans); Comfort, 283 F.Supp.2d at 372 (same); see also Brewer, 212 F.3d at 756-57 (Miner, C.J., dissenting) (same).
The principal question here is whether linking the integration tiebreaker to the racial demographic of the District’s population — rather than, for instance, that of the city of Seattle — overshoots the District’s goals. Specifically, Plaintiff suggests that, even when they are considered *1011“out of balance” by the District (i.e., when they deviate from the 60/40 ratio by more than 15 percent and thus enroll less than 45 percent or more than 75 percent nonwhite students), Seattle’s oversubscribed schools are sufficiently racially diverse to achieve the District’s goals.
I disagree that the District’s means significantly overshoot its goals. First, the District is trying to teach its students to be effective participants in the racially diverse environment in which they exist. Superintendent Olchefske noted that Seattle’s school-age demographic is significantly more racially diverse than the demographic for its population as a whole. (“There [are] a lot of elderly white people in this town,” he noted.) And he stated that the District has no regular access to data on the racial make-up of Seattle’s private school students.
Second, even if the racial mix at some of the oversubscribed high schools would be sufficiently diverse for the District to achieve its goals39 in those schools without the integration tiebreaker, this fact would not account for the effect of the integration tiebreaker on the overall school system. A clear objective of the School Board was that “no child should be required to attend a racially concentrated school.” Removing the integration tiebreaker would mean that non-white students living in the southern area of the city, where neighborhoods and schools are more racially concentrated, would not have an opportunity for access to the more diverse schools in the northern part of the city, simply because of where they live. Giving them this access furthers the District’s diversity goals.
Furthermore, the fact that a particular oversubscribed high school would draw a sufficiently diverse population in a given year without the integration tiebreaker does not guarantee that it would continue to do so. As I discussed above, open choice puts school assignment in the hands of the students; a tiebreaker tied to the District’s racial demographic is a natural way for the District to retain a safety net.
(ii) Flexibility
The District also has shown that its Plan is flexible in the short term and that its approach has been flexible over the long term. The District no longer forces white students south, nor nonwhite students north. For this reason, racial concentration has increased in some schools. But the District’s response has been measured. Responsive to community concerns and its own educational goals, the District has abandoned its complicated and mandatory systems for integrating its schools. Instead, it has developed a system that gives south-end non-white students an opportunity to leave racially concentrated schools (if they wish to) and promotes integrated schools across the district, while preserving the choice that it considers so critical to parental involvement. Mandatory assignments are kept to a minimum,40 and waivers are permitted for various reasons. The District’s consistent movement from coercive to voluntary integration measures *1012lends credence to its argument that it is working, through improvements to its programs, to reduce or eliminate oversub-scription and therefore to reduce its reliance on the integration tiebreaker.
Furthermore, the Plan is not inflexible in the manner of a quota; the integration tiebreaker operates only when patterns of individual choice result in oversubscription, and only until the school approximates the characteristics of the district as a whole. Choice, not a prescribed ratio of white to nonwhite students, controls the overwhelming majority of assignments. And choice patterns have been shown to change over time, as new facilities and programs are offered at different schools.
The District has demonstrated its ability to be responsive to these choice patterns and to the concerns of its constituents. It revisits the plan annually.41 In 2000, when a higher than normal number of students selected the same schools, the Board responded by increasing the integration trigger from a 10 percent to a 15 percent deviation from the school population and adopting a “thermostat” that turns off the integration tiebreaker as soon as the school has come into balance. The majority considers it constitutionally significant that the Board rejected a staff suggestion that the trigger instead be increased to 20 percent. Board members testified that they rejected a 20 percent trigger, in effect, because it would fail to assist students in moving from racially concentrated south-end schools. In other words, the proposed 20 percent trigger would no longer promote the Board’s goals; it therefore could not be considered narrowly tailored to achieve the District’s compelling interest because it would not achieve that interest at all. The Board’s decision thus is not a sign that the Board has failed, as the majority suggests, to “minimize [the] adverse impact on third parties,” maj. op. at 975.
(c) Relative burden on third parties.
The majority assumes that every student who is denied his or her choice of schools because of the integration tiebreaker suffers a constitutionally significant burden. As I foreshadowed .above, however, I consider the District’s Plan to impose a minimal burden that is shared equally by all of the District’s students. See Parents Involved, 72 P.3d at 159-60 (noting that the burden of not being allowed to attend one’s preferred school is shared by all students equally).
When we view the Plan on the large scale, without attempting to anticipate students’ subjective and shifting preferences for different schools, all the District’s students are equally subject to the possible burden of being denied their first choices. Only when we conceive of the Plan narrowly, by imagining two students — one white, one nonwhite — who are next door neighbors and have identical preferences for Ballard or for Franklin high school, will one student bear a “burden” and the other gain a “benefit.”
Yet it is well established that “there [is] no right under Washington law to attend a local school or the school of the student’s *1013choice.” Id. at 159.42 Of course, students and their parents will nonetheless prefer some schools over others; their preferences may be based on their perceptions of a school’s academic quality, on their subjective preference for a particular educational theme or program, or on the convenience of attending a particular school, among other things. These preferences result in changing choice patterns and the oversubscription of certain high schools. But oversubscribed schools do not become a limited government resource because of their popularity in a given year or their convenience for a given family. See Hampton, 102 F.Supp.2d at 380 n. 43 (“The Court understands that students and their parents might prefer one school over another. The preference may even arise from a perception that one school is better than others due to its location, its teachers and principal, or its classroom environment. However, these matters of personal preference do not distinguish those schools in a constitutionally significant sense.”); see also supra note 24. Despite any differences in academic quality (or perceptions thereof), all students who enroll in a Seattle high school will receive a high school education that meets state standards. And, as the District points out, even if Plaintiffs assertions of objectively unequal school quality were accepted,43 it is undisputed that the integration tiebreaker operates to give every student an opportunity to attend at least one of five oversubscribed “quality” high schools (because at least one is “integration positive” for both white and nonwhite students). I do not believe that students’ subjective preferences for one school over another, where the existence and educational relevance of objective differences among them is disputed, make the inconvenience of a nonpreferred assignment weightier than the District’s legitimate educational goals.
Finally, the District has minimized any burden by working to ameliorate the inconvenience and frustration for families who do not receive their preferred school assignments. When the popularity of the District’s five oversubscribed schools spiked for 2000-01 assignments, the School Board met to consider ways to soften the adverse effects. The administration immediately began to “aggressively move the waitlists” and to attempt to increase capacity at the oversubscribed schools. The Board and the administration discussed specific, long-range plans to increase the attractiveness of the undersubscribed schools. Finally, the District reached out to students receiving assignments to un-dersubscribed schools to share with them advantages of the schools of which they may not have been aware.
III. Conclusion.
For all these reasons, the Plan adopted by the Seattle School District for high school assignments is constitutional notwithstanding its inclusion of an integration tiebreaker. I would affirm the district *1014court’s judgment, and I dissent from the majority’s contrary holding.
. See Erica Frankenberg et al., A Multiracial Society with Segregated Schools: Are We Losing the Dream? 4 (The Civil Rights Project, Harvard Univ. Jan. 2003), at http://www.civil-rightsproject.harvard.edu/research/ re-seg03/AreWeLosingtheDream.pdf., cited in Gratz v. Bollinger, 539 U.S. 244, 299 n. 4, 123 S.Ct. 2411, 156 L.Ed.2d 257 (2003) (Ginsburg, J., dissenting).
. I agree with the majority’s conclusion that the case is not moot. Maj. op. at 957-60.
.In other words, strict scrutiny is a tool that we use to root out the improper prejudices ' and stereotypes that are the baseline concern of the Equal Protection Clause. See City of Richmond v. J.A. Croson Co., 488 U.S. 469, 493, 109 S.Ct. 706, 102 L.Ed.2d 854 (1989) (stating that the aim of strict scrutiny is to determine "what classifications are in fact motivated by illegitimate notions of racial inferiority or simple racial politics”).
. I agree with, the majority that the rights afforded by Title VI are coextensive with those guaranteed by the Equal Protection Clause. Maj. op. at 956-57 n. 10, 959-60 n. 14. I therefore conclude that the Plan does not violate Title VI. See Grutter, 539 U.S. at 343, 123 S.Ct. 2325 (concluding that the uni*991versity's admissions policy did not violate Title VI).
. A map created by the District's planners shows a striking pattern: Between 70 and 100 percent of the students who live in the various elementary-school "reference areas” in the south and southeast areas of the city are nonwhite, compared with 20 to 50 percent in the northern half of the city.
. The Plan under consideration here involves only high school assignments. However, because the District adopted the high school assignment Plan as part of a process of redesigning its school assignment system for all grade levels, the District understood its interest in diversity to span the K-12 system. Accordingly, when I discuss the details of "the Plan,” I refer only to high school assignments; my discussion of the District's compelling governmental interests, however,
. The Board explained that "[djiversity brings different viewpoints and experiences to classroom discussions and thereby enhances the educational process.” Maj. op. at 961. The District's expert noted that, in racially diverse schools, "both white and minority students experienced improved critical thinking skills — the ability to both understand and challenge views which are different from their own.”
. Comfort involved an elementary-school assignment plan, in which the school district allowed students to transfer from their assigned neighborhood schools only if the transfer would further the district's desegregation goals. 283 F.Supp.2d at 347-48. I agree with the court’s statement in Comfort that, "[w]hile a high school's mission is surely more academic-oriented than that of the elementary schools, citizenship training is still part and parcel of the enterprise.” Id. at 375 n. 84.
.The District's expert explained the civic benefits of diverse schools this way:
The research clearly and consistently shows that, for both white and minority students, a diverse educational experience results in improvement in race-relations, the reduction of prejudicial attitudes, and the achievement of a more democratic and inclusive experience for all citizens. More specifically, these benefits include more cross-race friendliness, reduction in prejudicial attitudes and increases in cross-race understanding of cultural differences. Recent research has identified the critical role of early school experiences in breaking down racial and cultural stereotypes. The research further shows that only a desegregated and diverse school can offer such opportunities and benefits. The research further supports the proposition that these benefits are long lasting.
(Emphasis added.)
. The Supreme Court in Grutter also recognized the importance of higher education in “preparing students for work and citizenship.” 539 U.S. at 331, 123 S.Ct. 2325. For the Court in Grutter, this point related less to the academic benefits of diversity and more to the Court's second rationale: ensuring open access to selective institutions of higher education in order to maintain their democratic legitimacy. See id. at 331-33, 123 S.Ct. 2325.
. Although it has not decided whether the state constitution requires integrated schools, the Supreme Court of Washington has written:
[I]f it is determined that in a contemporary setting de facto segregated schools cannot provide children with the educational opportunities necessary to equip them for their role as citizens, then the state constitution would most certainly mandate integrated schools.
Parents Involved, 72 P.3d at 162-63.
. Academic research has shown that inter-group contact reduces prejudice and supports the values of citizenship. See Derek Black, Comment, The Case for the New Compelling Government Interest: Improving Educational Outcomes, 80 N.C. L.Rev. 923, 951-52 (2002) (collecting academic research demonstrating that interpersonal interaction in desegregated schools reduces racial prejudice and stereotypes, improving students' citizenship values and their ability to succeed in a racially diverse society in their adult lives).
. Seattle's Cleveland and Rainier Beach High Schools, located in the minority-domi*994nated southeast area of the city, enrolled 90 and 92 percent nonwhite students, respectively, in the 2000-2001 school year. The District’s view that these schools are racially concentrated is, at the very least, reasonable, if not compelled by the evidence. See 34 C.F.R. § 280.4(b) (defining ''[m]inority group isolation, in reference to a school, [as] a condition in which minority group children constitute more than 50 percent of the enrollment of the school”); see also infra pp. 1005 (discussing the majority’s assertion that “inter-ethnic diversity” obviates the District's need for pro-diversity and pro-integration policies, maj. op. at 983-84).
. There are at least two exceptions to this general proposition. First, public university systems can be ordered to desegregate to remedy the effects of past intentional or de jure segregation, as in United States v. Fordice, 505 U.S. 717, 112 S.Ct. 2727, 120 L.Ed.2d 575 (1992). Second, the Court in Grutter recognized that universities do exist in, and affect, the society as a whole and that they have a compelling interest in taking into account the effects of their admissions policies on that society. 539 U.S. at 332-33, 123 S.Ct. 2325.
. The district court held that”[pjreventing the re-segregation of Seattle’s schools is ... a compelling interest.” Parents Involved in Cmty. Schs. v. Seattle Sch. Dist. No. 1, 137 F.Supp.2d 1224, 1237 (W.D.Wash.2001); see also id. at 1233-35.
. The Andrew Jackson cases arose out of an action by minority students seeking compulsory desegregation, but the school district was held not to have engaged in intentional or de jure segregation and therefore could not be ordered to remedy the de facto racial imbalance that existed. 598 F.2d at 715. The court then addressed the question whether the district's voluntary plan itself violated equal protection, id. at 717, holding that it did not, id. at 718-19.
. As I will discuss below, the District voluntarily adopted its first mandatory desegregation plan in 1977, in order to forestall legal action by the NAACP and the ACLU, who alleged that the District had acted to further de facto segregation.
. The logical corollary to this concern about the judiciary's lack of competency is a recognition that public school educators are, in fact, trained in and qualified to assess educational policies and their outcomes.
. Indeed, as I will argue below, a fundamental difference between the university and the public school settings is that public schools generally do not "select” their students at all. Rather, they are obliged to educate all students in the relevant district.
. In an oft-quoted sentence, the Court described the analysis as follows:
In determining whether race-conscious remedies are appropriate, we look to several factors, including the necessity for the relief and the efficacy of alternative remedies; the flexibility and duration of the relief, including the availability of waiver provisions; the relationship of the numerical goals to the relevant labor market; and the impact of the relief on the rights of third parties.
Paradise, 480 U.S. at 171, 107 S.Ct. 1053.
. The same is true of selective admissions to special high school programs, as in Wessmann v. Gittens, 160 F.3d 790, 799-800 (1st Cir.1998) (employing a "true diversity” analysis in the context of competitive admissions to a prestigious examination high school); and Hampton, 102 F.Supp.2d at 380-81 *999(noting that admissions to magnet schools, unlike basic school assignments, have "vertical effects"). See Brewer, 212 F.3d at 752-53 (distinguishing Wessmann because it was a selective admissions case in which "true diversity” was the compelling interest). I disagree with the majority's reasoning that the District’s plan is "necessarily selective” merely because some schools are oversubscribed or more popular. See maj. op. at 979-80. Under this logic, assignment between two first-grade classrooms in a single school — classrooms that are equivalent but for the popularity of the teacher — could be considered "selective.”
. I use the term "quota” here, as I believe the Court did in Grutter, 539 U.S. at 335, 123 S.Ct. 2325, to mean "a fixed number or percentage of minority group members who may be admitted into some activity or institution” — not to mean, more generally, "a proportional part.” Webster's Third New Int’l Dictionary 1868 (unabridged ed.1993). Quotas, as I understand them, are not at issue here because no student is preferentially admitted to, or turned away from, the Seattle public high schools. Thus, the District does not run afoul of this aspect of the Court’s narrow-tailoring analysis.
The majority seizes upon the more general sense of the word "quota” — as meaning "proportion” — and argues that "racial balancing” is per se unconstitutional. See maj. op. at 986-87. The cited cases do not support the *1000majority’s sweeping statements. First, the Court in Croson held that a minority set aside was prohibited because the government could not prove that past discrimination provided a compelling reason for the program, not because "racial balancing” as a mechanism for achieving a compelling state interest would necessarily be inconsistent with the Fourteenth Amendment. Second, states are prohibited from making race the predominant factor in drawing legislative districts because of the impermissibility of racial stereotyping (that is, using race as a proxy for political characteristics), not because any consideration of a district's racial proportions is per se improper. See Bush v. Vera, 517 U.S. 952, 968, 116 S.Ct. 1941, 135 L.Ed.2d 248 (1996) (plurality); see also id. at 958, 116 S.Ct. 1941 (noting that redistricting may be performed with "consciousness of race”). These cases do not establish that the District is per se prohibited from linking its assignment practices to the racial make-up of its student enrollment, especially where student choices, not fixed proportions, are the principal determinant of student assignments. I simply cannot agree that we have reached the majority’s promised land: Our governments, schools, and courts may yet be forced, by compelling reasons, to acknowledge that race exists in America. See Grutter, 539 U.S. at 333, 123 S.Ct. 2325 (noting that, in our society, "race unfortunately still matters”).
. Students are selected by merit into at least one District program (which carries a corresponding school assignment), but not into any District school. Those who test in the top 2 percent of their grade levels are offered admission to the Advanced Placement Program for academically talented students. Selection for that program is not at issue in this case.
. Of course, I agree with the majority that the government must offer its benefits on equal terms, regardless of race. See maj. op. at 987-89. But the governmental benefit at issue here is a high school education, not free student choice about school assignments. Indeed, the District could devise a permissible assignment system drat is devoid of student choice among high schools. Nonetheless, the District has opted to offer some choices to families — not as an abstract benefit, but rather as an educational policy that interacts with and is necessarily constrained by other District policies, including the District's diversity and integration goals.
. Justice Powell noted:
Respondent’s position is wholly dissimilar to that of a pupil bused from his neighborhood school to a comparable school in an*1001other neighborhood in compliance with a desegregation decree. Petitioner did not arrange for respondent to attend a different medical school in order to desegregate Davis Medical School; instead, it denied him admission and may have deprived him altogether of a medical education.
Bakke, 438 U.S. at 301 n. 39, 98 S.Ct. 2733.
. The majority's view is that diversity is an all-or-nothing proposition and that it is improper to try to achieve racial diversity without simultaneously trying to achieve every other conceivable type of diversity (e.g., socio*1002economic, religious, or linguistic). See maj. op. at 971-72. I disagree. The District’s interest is in the socialization benefits that come from racial diversity in particular. The narrow tailoring inquiry is not concerned with how a different compelling interest might lead to similar benefits, but rather asks whether race-neutral means are effective in achieving the race-related compelling interest at issue.
. The history that follows comes principally from two documents in the district court record. One is a paper entitled, "The History of Desegregation in Seattle Public Schools, 1954-1981,” which was prepared by the District's desegregation planners. The other is the "Findings and Conclusions” adopted by the Board in support of the current Plan.
. South and southeast area high schools Franklin and Cleveland would experience similar enrollment changes in the 1960's and the 1970's, with minority enrollment at Franklin reaching 78 percent in 1977.
. For example, the District’s History of Desegregation reports that the Seattle Plan was extremely confusing, required mandatory busing of non-white students in disproportionate numbers, made facilities and enrollment planning difficult, and contributed to "white flight" from the city schools.
.Indeed, the District implemented the school assignment Plan as part of a comprehensive plan to improve and equalize the attractiveness of all the high schools, which included a weighted funding formula, a facilities plan, and a new teacher contract that would make teacher transfers easier.
. Originally, schools that deviated by more than 10 percent were considered "imbalanced.” For the 2000-2001 school year, the trigger was increased to 15 percent, softening the effect of the tiebreaker.
. A fourth tiebreaker, a random lottery, is seldom used because distance is calculated to l/100th of a mile.
. "Today choice is a popular way to reform American education....” Wendy Parker, The Color of Choice: Race and Charter Schools, 75 Tul. L.Rev. 563, 564 (2001).
. Native American students are by far the smallest group and are the only group spread evenly between the two halves of the city.
. Indeed, later in the opinion, the Court noted that universities in states with laws against "racial preferences” were experimenting with "a wide variety of alternative approaches.” Grutter, 539 U.S. at 342, 123 S.Ct. 2325. Yet the Court did not require the university to have analyzed fully and rejected each of these alternatives; instead, it noted that universities in states without such laws should monitor and "draw on the most promising aspects of these race-neutral alternatives as they develop.” Id. (emphasis added).
. The majority also notes the Urban League's suggestion that the District improve and better market its specialty programs, especially at racially concentrated schools. In fact, the District demonstrated that it is striving to improve its programming in a manner intended to make all schools equally attractive, thereby reducing or eliminating its dependence on the integration tiebreaker; it has installed new principals, constructed new buildings, undertaken major renovations, introduced an International Baccalaureate program, and introduced an information technology program linked with community colleges, *1009among other things. Similar steps taken at Ballard and Nathan Hale High Schools led to their recent turn-arounds in popularity.
. The majority interprets Grutter to demand that every school consider a pure lottery unless, but only unless, such a lottery would sacrifice the academic quality or diversity of the student body. See maj. op. at 971. This narrow reading does not fit a context, like this one, in which the school's mission is not to fashion an academically exceptional student body. Instead, the more appropriate principle to draw from Grutter is that schools need not consider alternatives that would do violence to the values central to their particular educational missions — -here, choice and parental involvement.
. We have held that a local government's continuing efforts to combine race-neutral measures with a minority set-aside program are "one factor suggesting that [a set-aside] plan is narrowly tailored.” Coral, 941 F.2d at 923; Associated Gen. Contractors, 950 F.2d at 1417 (citing Coral). See supra pp. 1002-1003, 1004 & n. 30, 1008-1009 n. 36 (discussing the District’s many race-neutral efforts to promote integrated schools).
. To reiterate, in this context the District's relevant goals are for regular intergroup contact to occur and for students not to feel isolated or tokenized.
. For the 2000-01 school year, roughly 350 students received ''mandatory” assignments, meaning that their assigned school was not one of their choices. Roughly 100 of these students had listed only one choice and another hundred had listed only two choices. Of the roughly 300 students affected by the integration tiebreaker, only 84 were given "mandatory” assignments. Of these, 29 were ultimately assigned to the same school they would have been attending without the tiebreaker, and 55 received assignments affected by the tiebreaker.
. Like the majority, maj. op. at 976 n. 32, I believe that this annual review, combined with the fact that the tiebreaker operates only until a school comes into balance, satisfies the durational requirement of narrow tailoring. See Grutter, 539 U.S. at 342-43, 123 S.Ct. 2325; Adarand Constructors, Inc. v. Pena, 515 U.S. 200, 238, 115 S.Ct. 2097, 132 L.Ed.2d 158 (1995) (holding that narrow tailoring requires that a program be limited in scope and duration "such that it will not last longer than the discriminatory effects it is designed to eliminate” (internal quotation marks omitted)).
. Subject to federal statutory and constitutional requirements, structuring public education has long been within the control of the states, as part of their traditional police powers. See Barbier v. Connolly, 113 U.S. 27, 31-32, 5 S.Ct. 357, 28 L.Ed. 923 (1884) (describing the states' traditional police powers).
. The District has disputed Plaintiff's assertion of significant differences in objective quality among the 10 high schools. Before granting summary judgment to Plaintiff, the majority must accept the District's version of the facts. See Simo v. Union of Needletrades, Indus. & Textile Employees, 322 F.3d 602, 609 (9th Cir.2003) (stating that, on summary judgment, facts are to be viewed in the light most favorable to the nonmoving party). If this factual issue were material, summary judgment would not be proper. Id. at 610.