426 F.3d 1162 (2005)

PARENTS INVOLVED IN COMMUNITY SCHOOLS, a Washington non-profit corporation, Plaintiff-counter-defendant-Appellant,
v.
SEATTLE SCHOOL DISTRICT, NO. 1, a political subdivision of the State of Washington; Joseph Olchefske, in his official capacity as superintendent; Barbara Schaad-Lamphere, in her official capacity as President of the Board of Directors of Seattle Public Schools; Donald Neilson, in his official capacity as Vice President of the Board of Directors of Seattle Public Schools; Steven Brown; Jan Kumasaka; Michael Preston; Nancy Waldman, in their official capacities as members of the board of Directors, Defendants-counter-claimants-Appellees.

1163*1163 No. 01-35450.United States Court of Appeals, Ninth Circuit.

Argued and Submitted June 21, 2005.Filed October 20, 2005.1164*1164 1165*1165 Harry J.F. Korrell (argued) and Daniel B. Ritter, Davis Wright Tremaine LLP, Seattle, WA, for the plaintiff-counter-defendant/appellant.

Michael Madden (argued) and Carol Sue Janes, Bennett Bigelow & Leedom, P.S., Seattle, WA, and Mark S. Green, Office of the General Counsel, Seattle School District No. 1, Seattle, WA, for the defendants-counter-claimants/appellees.

Sharon L. Browne, Pacific Legal Foundation, Sacramento, CA, and Russell C. Brooks, Pacific Legal Foundation, Bellevue, WA, for the amici curiae Pacific Legal Foundation, American Civil Rights Institute, American Civil Rights Union and Center for Equal Opportunity in support of plaintiff-counter-defendant/appellant.

Paul J. Lawrence, Preston Gates & Ellis LLP, Seattle, WA, for the amicus curiae American Civil Liberties Union in support of defendants-counter-claimants/appellees.

Before: SCHROEDER, Chief Judge, PREGERSON, KOZINSKI, KLEINFELD, HAWKINS, W. FLETCHER, FISHER, TALLMAN, RAWLINSON, CALLAHAN and BEA, Circuit Judges.

Argued and Submitted En Banc June 21, 2005.

1166*1166 Opinion by Judge FISHER; Concurrence by Judges KOZINSKI; Dissent by Judge BEA.

FISHER, Circuit Judge, with whom Chief Judge SCHROEDER and Judges PREGERSON, HAWKINS, W. FLETCHER and RAWLINSON join concurring; Judge KOZINSKI, concurring in the result.

This appeal requires us to consider whether the use of an integration tiebreaker in the open choice, noncompetitive, public high school assignment plan crafted by Seattle School District Number 1 (the “District”) violates the federal Constitution’s Equal Protection Clause. Our review is guided by the principles articulated in the Supreme Court’s recent decisions regarding affirmative action in higher education, Grutter v. Bollinger, 539 U.S. 306, 123 S.Ct. 2325, 156 L.Ed.2d 304 (2003), and Gratz v. Bollinger, 539 U.S. 244, 123 S.Ct. 2411, 156 L.Ed.2d 257 (2003), and the Court’s directive that “[c]ontext matters when reviewing race-based governmental action under the Equal Protection Clause.” Grutter, 539 U.S. at 327, 123 S.Ct. 2325. We conclude that the District has a compelling interest in securing the educational and social benefits of racial (and ethnic) diversity, and in ameliorating racial isolation or concentration in its high schools by ensuring that its assignments do not simply replicate Seattle’s segregated housing patterns.[1] We also conclude that the District’s Plan is narrowly tailored to meet the District’s compelling interests.

I. Background[2]

A. Seattle Public Schools: A Historical Perspective

Seattle’s historical struggle with the problem of racial isolation in its public school system provides the context for the District’s implementation of the current challenged assignment plan. Seattle is a diverse community. Approximately 70 percent of its residents are white, and 30 percent are nonwhite. Seattle public school enrollment breaks down nearly inversely, with approximately 40 percent white and 60 percent nonwhite students. A majority of the District’s white students live in neighborhoods north of downtown, the historically more affluent part of the city. A majority of the city’s nonwhite students, including approximately 84 percent of all African-American students, 74 percent of all Asian-American students, 65 percent of all Latino students and 51 percent of all Native-American students, live south of downtown.

The District operates 10 four-year public high schools. Four are located north of downtown—Ballard, Ingraham, Nathan Hale and Roosevelt; five are located south of downtown—Chief Sealth, Cleveland, Franklin, Garfield and Rainier Beach; one is located west of downtown—West Seattle. For over 40 years, the District has made efforts to attain and maintain desegregated schools and avoid the racial isolation or concentration that would ensue if school assignments replicated Seattle’s segregated housing patterns. Since the 1960s, while courts around the country 1167*1167 ordered intransigent school districts to desegregate, Seattle’s School Board voluntarily explored measures designed to end de facto segregation in the schools and provide all of the District’s students with access to diverse and equal educational opportunities.

In the late 1950s and early 1960s, school assignments were made strictly on the basis of neighborhood.[3] In 1962, Garfield High School reported 64 percent minority enrollment and it accommodated 75 percent of all African-American students. Meanwhile, the eight high schools serving other major areas of the city remained more than 95 percent white.

The District responded to this imbalance, and racial tensions in the de facto segregated schools, in various ways. In the early 1960s, the District first experimented with small-scale exchange programs in which handfulls of students switched high schools for five-week periods. In 1963, expanding on this concept, 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. In the 1970s, the District increased its efforts again, this time adopting a desegregation plan in the middle schools that requested volunteers to transfer between minority- and majority-dominated neighborhood schools and called for mandatory transfers when the number of volunteers was insufficient, though this portion of the plan was never implemented. The District also took steps to desegregate Garfield High School by changing its educational program, improving its facilities and eliminating “special transfers” that had previously allowed white students to leave Garfield. Finally, for the 1977-78 school year, the District instituted a magnet-school program. According to the District’s history:

While 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.

History of Desegregation at 32.

By the 1977-78 school year, segregation had increased: Franklin was 78 percent minority, Rainier Beach 58 percent, Cleveland 76 percent and Garfield 65 percent. Other high schools ranged from 9 percent to 23 percent minority enrollment.

In the spring of 1977, the Seattle branch of the National Association for the Advancement of Colored People (“NAACP”) filed a complaint with the United States Department of Education’s 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 American Civil Liberties Union (“ACLU”), formally threatened to file additional actions if the District failed to adopt a mandatory desegregation plan. When the District agreed to develop such a plan, the Office of Civil Rights concomitantly agreed to delay its investigation, and the ACLU agreed to delay filing a lawsuit.

During the summer of 1977, the District and community representatives reviewed five model plans. Ultimately, the District 1168*1168 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 desegregation. 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. Opponents of the Seattle Plan immediately passed a state initiative to block its implementation, but the Supreme Court ultimately declared the initiative unconstitutional. Washington v. Seattle Sch. Dist. No. 1, 458 U.S. 457, 470, 102 S.Ct. 3187, 73 L.Ed.2d 896 (1982).

The Seattle Plan furthered the District’s school desegregation goals, but its operation was unsatisfactory in other ways.[4] In 1988, a decade after its implementation, 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 students’ homes, but others were in “racially and culturally different neighborhoods.” Findings and Conclusions at 30-31. Although roughly 70 percent of students received their first choices, the controlled choice plan still resulted in mandatory busing for 16 percent of the District’s students.

In 1994, the Board directed District staff 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 a 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 racial diversity. Neighborhood and regional plans were viewed as unduly limiting student choice, on which the District placed high value because student choice 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 racially integrated education by establishing diversity goals while moving away from the rigid desegregation guidelines and mandatory assignments prevalent in the 1970s and 1980s.

The Board adopted the current open choice plan (the “Plan”) for the 1998-99 1169*1169 school year. Under the Plan, students entering the ninth grade may select any high school in the District. They are assigned, where possible, to the school they list as their first choice. If too many students choose the same school as their first choice, resulting in “oversubscription,” the District assigns students to each oversubscribed school based on a series of tiebreakers. If a student is not admitted to his or her first choice school as a result of the tiebreakers, the District tries to assign the student to his or her second choice school, and so on. Students not assigned to one of their chosen schools are assigned to the closest school with space available; students who list more choices are less likely to receive one of these “mandatory” assignments. The most recent version of the Plan, which the School Board reviews annually, is for the 2001-02 school year and is the subject of this litigation.

B. The Plan

The District has sought to make each of its 10 high schools unique, with programs that respond to the continually changing needs of students and their parents. Indeed, the District implemented the Plan as part of a comprehensive effort to improve and equalize the attractiveness of all the high schools, including adoption of a weighted funding formula, a facilities plan and a new teacher contract that would make teacher transfers easier. Nevertheless, the high schools vary widely in desirability. Three of the northern schools— Ballard, Nathan Hale and Roosevelt—and two of the southern schools—Garfield and Franklin—are highly desirable and oversubscribed, meaning that more students wish to attend those schools than capacity allows.[5] The magnitude of the oversubscription is noteworthy: For the academic year 2000-01, approximately 82 percent of students selected one of the oversubscribed schools as their first choice, while only about 18 percent picked one of the undersubscribed high schools as their first choice. Only when oversubscription occurs does the District become involved in the assignment process.

If a high school is oversubscribed, all students applying for ninth grade are admitted according to a series of four tiebreakers, applied in the following order: First, students who have a sibling attending that school are admitted. In any given oversubscribed school, the sibling tiebreaker accounts for somewhere between 15 to 20 percent of the admissions to the ninth grade class.

Second, if an oversubscribed high school is racially imbalanced—meaning that the racial make up of its student body differs by more than 15 percent from the racial make up of the students of the Seattle public schools as a whole— and if the sibling preference does not bring the oversubscribed high school within plus or minus 15 percent of the District’s demographics, the race-based tiebreaker 1170*1170 is “triggered” and the race of the applying student is considered. (For the purposes of the race-based tiebreaker, a student is deemed to be of the race specified in his or her registration materials.) Thus, if a school has more than 75 percent nonwhite students (i.e., more than 15 percent above the overall 60 percent nonwhite student population) and less than 25 percent white students, or when it has less than 45 percent nonwhite students (i.e., more than 15 percent below the overall 60 percent nonwhite student population) and more than 55 percent white students, the school is considered racially imbalanced.

Originally, schools that deviated by more than 10 percent were deemed racially imbalanced. For the 2001-02 school year, however, the triggering number was increased to 15 percent, softening the effect of the tiebreaker.[6] For that year, the race-based tiebreaker was used in assigning entering ninth grade students only to three oversubscribed schools—Ballard, Franklin and Nathan Hale. Accordingly, in seven of the 10 public high schools in 2001-02, race was not relevant in making admissions decisions.

The race-based tiebreaker is applied to both white and nonwhite students. For example, in the 2000-01 school year—when the trigger point was still plus or minus 10 percent—89 more white students were assigned to Franklin than would have been assigned absent the tiebreaker, 107 more nonwhite students were assigned to Ballard than would have been assigned absent the tiebreaker, 82 more nonwhite students were assigned to Roosevelt than would have been assigned absent the tiebreaker and 27 more nonwhite students were assigned to Nathan Hale than would have been assigned absent the tiebreaker.[7] These assignments accounted for about 10 percent of admissions to Seattle’s high schools as whole. That is, of the approximately 3,000 incoming students entering Seattle high schools in the 2000-01 school year, approximately 300 were assigned to an oversubscribed high school based on the race-based tiebreaker.

In addition to changing the trigger point for the 2001-02 school year to plus or minus 15 percent, the District also developed a “thermostat,” whereby the tiebreaker is applied to the entering ninth grade student population only until it comes within the 15 percent plus or minus variance. Once that point is reached, the District “turns-off” the race-based tiebreaker, and there is no further consideration of a student’s race in the assignment process. The tiebreaker does not apply, and race is not considered, for students entering a high school after the ninth grade (e.g., by transfer).

As demonstrated in the chart below, the District estimates that without the race-based tiebreaker, the nonwhite populations of the 2000-01 ninth grade class at Franklin would have been 79.2 percent, at Hale 30.5 percent, at Ballard 33 percent and at Roosevelt 41.1 percent. Using the race-based 1171*1171 tiebreaker, the actual nonwhite populations of the ninth grade classes at the same schools respectively were 59.5 percent, 40.6 percent, 54.2 percent and 55.3 percent.

2000-01 DIFFERENCE IN PERCENTAGES OF NONWHITE STUDENTS IN NINTH GRADE WITH AND WITHOUT TIEBREAKER

               WITHOUT       WITH        PERCENT
SCHOOL       TIEBREAKER   TIEBREAKER    Difference

  FRANKLIN       79.2        59.5        -19.7
NATHAN HALE      30.5        40.6        +10.1
  BALLARD        33.0        54.2        +21.2
 ROOSEVELT       41.1        55.3        +14.2

In the third tiebreaker, students are admitted according to distance from the student’s home to the high school. Distance between home and school is calculated within 1/100 of a mile, with the closest students being admitted first. In any given oversubscribed school, the distance-based tiebreaker accounts for between 70 to 75 percent of admissions to the ninth grade.

In the fourth tiebreaker, a lottery is used to allocate the remaining seats. Because the distance tiebreaker serves to assign nearly all the students in the District, a lottery is virtually never used.

C. Procedural History

Parents Involved in Community Schools (“Parents”), a group of parents whose children were not, or might not be, assigned to the high schools of their choice under the Plan, claimed that the District’s use of the race-based tiebreaker for high school admissions is illegal under the Washington Civil Rights Act (“Initiative 200”),[8] the Equal Protection Clause of the Fourteenth Amendment[9] and Title VI of the Civil Rights Act of 1964.[10]

Both Parents and the District moved for summary judgment on all claims. In a published opinion dated April 6, 2001, the district court upheld the use of the racial tiebreaker under both state and federal law, granting the District’s motion. Parents Involved in Cmty. Schs. v. Seattle Sch. Dist. No. 1, 137 F.Supp.2d 1224, 1240 (W.D.Wash.2001) (“Parents I“). Parents timely appealed, and on April 16, 2002, a three-judge panel of this court issued an opinion reversing the district court’s decision, holding that the Plan violated Washington state law and discussing federal law only as an aid to construing state law. Parents Involved in Cmty. Schs. v. Seattle Sch. Dist. No. 1, 285 F.3d 1236 (9th Cir. 2002) (“Parents II“). The panel subsequently withdrew its opinion and certified the state law question to the Washington Supreme Court. Parents Involved in Cmty. Schs. v. Seattle Sch. Dist. No. 1, 294 F.3d 1084, 1085 (9th Cir.2002) (“Parents III“). The Washington Supreme Court disagreed with the panel’s decision, holding that the open choice plan did not violate Washington law. Parents Involved in 1172*1172 Cmty. Schs. v. Seattle Sch. Dist. No. 1, 149 Wash.2d 660, 72 P.3d 151, 166 (2003) (“Parents IV“) (holding that Washington law “does not prohibit the Seattle School District’s open choice plan tie breaker based upon race so long as it remains neutral on race and ethnicity and does not promote a less qualified minority applicant over a more qualified applicant”). Thereafter, a majority of the three-judge panel of this court held that although the District demonstrated a compelling interest in achieving the benefits of racial diversity, the Plan violated the Equal Protection Clause because it was not narrowly tailored. Parents Involved in Comty. Schs. v. Seattle Sch. Dist., No. 1, 377 F.3d 949 (9th Cir.2004) (“Parents V“). We granted en banc rehearing and now affirm the district court.[11]

II. Discussion

A. Strict Scrutiny

We review racial classifications under the strict scrutiny standard, which requires that the policy in question be narrowly tailored to achieve a compelling state interest. See Johnson v. California, ___ U.S. ___, 125 S.Ct. 1141, 1146, 160 L.Ed.2d 949 (2005); Grutter, 539 U.S. at 326, 123 S.Ct. 2325; Adarand Constructors, Inc. v. Pena, 515 U.S. 200, 226-27, 115 S.Ct. 2097, 132 L.Ed.2d 158 (1995).[12] The strict scrutiny standard is not “strict in theory, but fatal in fact.” Adarand, 515 1173*1173 U.S. at 237, 115 S.Ct. 2097 (internal quotation marks omitted). “Although all governmental uses of race are subject to strict scrutiny, not all are invalidated by it.” Grutter, 539 U.S. at 326-27, 123 S.Ct. 2325. We employ strict scrutiny to “smoke out” impermissible uses of race by ensuring that the government is pursuing a goal important enough to warrant use of a highly suspect tool. Id. at 327, 123 S.Ct. 2325 (internal quotation marks omitted). This heightened standard of review provides 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. Smith v. Univ. of Washington, 392 F.3d 367, 372 (9th Cir.2004). In evaluating the District’s Plan under strict scrutiny, we also bear in mind the Court’s directive that “[c]ontext matters when reviewing race-based governmental action under the Equal Protection Clause.” Grutter, 539 U.S. at 326, 123 S.Ct. 2325.

B. Compelling State Interest

Under strict scrutiny, a government action will not survive unless motivated by a “compelling state interest.” See id. at 325, 327, 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 fit. See United States v. Paradise, 480 U.S. 149, 171, 107 S.Ct. 1053, 94 L.Ed.2d 203 (1987) (stating that, in order to determine whether an order was narrowly tailored, “we must examine the purposes the order was intended to serve”).

Although the Supreme Court has never decided a case involving the consideration of race in a voluntarily imposed school assignment plan intended to promote racially and ethnically diverse secondary schools, its decisions regarding selective admissions to institutions of higher learning demonstrate that one compelling reason for considering race is to achieve the educational benefits of diversity. The compelling interest that the Court recognized in Grutter was the promotion of the specific educational and societal benefits that flow from diversity. See Grutter, 539 U.S. at 330, 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 advantages 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 group of national leaders who are representative of our country’s diversity, id. at 332-33, 123 S.Ct. 2325. The Court also mentioned the role of diversity in challenging stereotypes. Id. at 330, 333, 123 S.Ct. 2325. The Court largely deferred to the law school’s educational judgment not only in determining that diversity would produce these benefits, but also in determining that these benefits were critical to the school’s educational mission. Id. at 328-33, 123 S.Ct. 2325.[13]

1174*1174 Against this background, we consider the specific interests that the District’s Plan seeks to advance. These interests are articulated in the “Board Statement Reaffirming Diversity Rationale” as:

Diversity in the classroom increases the likelihood that students will discuss racial or ethnic issues and be more likely to socialize with people of different races. Diversity is thus a valuable resource for teaching students to become citizens in a multi-racial/multi-ethnic world.

Providing students the opportunity to attend schools with diverse student enrollment also has inherent educational value from the standpoint of education’s role in a democratic society. . . . Diversity brings different viewpoints and experiences to classroom discussions and thereby enhances the educational process. It also fosters racial and cultural understanding, which is particularly important in a racially and culturally diverse society such as ours.

The District’s commitment to the diversity of its schools and to the ability to voluntarily avoid racially concentrating enrollment patterns also helps ensure that all students have access to those schools, faculties, course offerings, and resources that will enable them to reach their full potential.

Based on the foregoing rationale, the Seattle School District’s commitment is that no student should be required to attend a racially concentrated school. The District is also committed to providing students with the opportunity to voluntarily choose to attend a school to promote integration. The District provides these opportunities for students to attend a racially and ethnically diverse school, and to assist in the voluntary integration of a school, because it believes that providing a diverse learning environment is educationally beneficial for all students.

The District’s interests fit into two broad categories: (1) the District seeks the affirmative educational and social benefits that flow from racial diversity; and (2) the District seeks to avoid the harms resulting from racially concentrated or isolated schools.

1. Educational and Social Benefits that Flow from Diversity

The District has established that racial diversity produces a number of compelling educational and social benefits in secondary education. First, the District presented expert testimony 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.”

Second, the District demonstrated the socialization and citizenship advantages of racially diverse schools. School officials, relying on their experience as teachers and administrators, and the District’s expert all explained these benefits on the record. According to the District’s expert, the social science 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 . . . 1175*1175 inclusive experience for all citizens. . . . 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.) Even Parents’ expert conceded that “[t]here is general agreement by both experts and the general public that integration is a desirable policy goal mainly for the social benefit of increased information and understanding about the cultural and social differences among various racial and ethnic groups.”[14] That is, diversity encourages students not only to think critically but also democratically.

Third, the District’s expert noted that “research shows that a[ ] desegregated educational experience opens opportunity networks in areas of higher education and employment . . . [and] strongly shows that graduates of desegregated high schools are more likely to live in integrated communities than those who do not, and are more likely to have cross-race friendships later in life.”[15]

The District’s interests in the educational and social benefits of diversity are similar to those of a law school as articulated in Grutter. The contextual differences between public high schools and universities, however, make the District’s interests compelling in a similar but also significantly different manner. See Grutter, 539 U.S. at 330, 123 S.Ct. 2325 (noting that the compelling state interest in diversity is judged in relation to the educational benefits that it seeks to produce).

The Supreme Court in Grutter noted the importance of higher education in “preparing students for work and citizenship.” 539 U.S. at 331, 123 S.Ct. 2325. For a number of reasons, public secondary schools have an equal if not more important role in this preparation. First, underlying the history of desegregation in this country is a legal regime that recognizes the principle that public secondary education serves a unique and vital socialization function in our democratic society. As the Court explained in Plyler v. Doe, “[w]e have recognized the public schools as a most vital civic institution for the preservation of a democratic system of government, and as the primary vehicle for transmitting the values on which our society rests.” 457 U.S. 202, 221, 102 S.Ct. 2382, 72 L.Ed.2d 786 (1982) (internal quotation marks and citations omitted); 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, 457 U.S. at 221-23, 102 S.Ct. 2382 (noting that public 1176*1176 education perpetuates the political system and the economic and social advancement of citizens and that “education has a fundamental role in maintaining the fabric of our society”); Ambach v. Norwick, 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 v. Bd. of Educ., 347 U.S. 483, 493, 74 S.Ct. 686, 98 L.Ed. 873 (1954) (“[Education] is required in the performance of our most basic public responsibilities. . . . It is the very foundation of good citizenship. Today it is a principal instrument in awakening the child to cultural values, in preparing him for later professional training, and in helping him to adjust normally to his environment.”). Under Washington law, such civic training is mandated by the state constitution: “Our constitution is unique in placing paramount value on education for citizenship.” Parents IV, 72 P.3d at 158.

Second, although one hopes that all students who graduate from Seattle’s public schools would have the opportunity to attend institutions of higher learning if they so desire, a substantial number of Seattle’s public high school graduates do not attend college.[16] For these students, their public high school educational experience will be their sole opportunity to reap the benefits of a diverse learning environment. We reject the notion that only those students who leave high school and enter the elite world of higher education should garner the benefits that flow from learning in a diverse classroom. Indeed, it would be a perverse reading of the Equal Protection Clause that would allow a university, educating a relatively small percentage of the population, to use race when choosing its student body but not allow a public school district, educating all children attending its schools, to consider a student’s race in order to ensure that the high schools within the district attain and maintain diverse student bodies.

Third, the public school context involves students who, because they are younger and more impressionable, are more amenable to the benefits of diversity. See Comfort, 418 F.3d at 15-16 (“In fact, there is significant evidence in the record that the benefits of a racially diverse school are more compelling at younger ages.”); Comfort v. Lynn School Committee, 283 F.Supp.2d 328, 356 (D.Mass.2003) (noting expert testimony describing racial stereotyping as a “`habit of mind’ that is difficult to break once it forms” and explaining that “[i]t is more difficult to teach racial tolerance to college-age students; the time to do it is when the students are still young, before they are locked into racialized thinking”); see also Goodwin Liu, Brown, Bollinger, and Beyond, 47 How. L.J. 705, 755 (2004) (“[I]f `diminishing the force of [racial] stereotypes’ is a compelling pedagogical interest in elite higher education, it can only be more so in elementary and secondary schools — for the very premise of Grutter’s diversity rationale is that students enter higher education having had too few opportunities in early grades to study and learn alongside peers from other racial groups.”) (citing Grutter, 539 U.S. at 333, 123 S.Ct. 2325) (emphasis added)).

The dissent insists that racial diversity in a public high school is not a compelling interest, arguing that Grutter endorsed a law school’s compelling interest in diversity 1177*1177 only in some broader or more holistic sense. Bea, J., dissenting, infra. at 1202. To attain this broader interest, the dissent contends, the District may only consider race along with other attributes such as socioeconomic status, ability to speak multiple languages or extracurricular talents. We read Grutter, however, to recognize that racial diversity, not some proxy for it, is valuable in and of itself. 539 U.S. at 330, 123 S.Ct. 2325 (discussing the “substantial” benefits that flow from a racially diverse student body and citing several sources that detail the impact of racial diversity in the educational environment).

In short, the District has demonstrated that it has a compelling interest in the educational and social benefits of racial diversity similar to those articulated by the Supreme Court in Grutter as well as the additional compelling educational and social benefits of such diversity unique to the public secondary school context.

2. Avoiding the Harms Resulting from Racially Concentrated or Isolated Schools

The District’s interest in achieving the affirmative benefits of a racially diverse educational environment has a flip side: avoiding racially concentrated or isolated schools. In particular, the District is concerned with making the educational benefits of a diverse learning environment available to all its students and ensuring that “no student should be required to attend a racially concentrated school.” See “Board Statement Reaffirming Diversity Rationale,” quoted supra p. 1174. Research regarding desegregation has found that racially concentrated or isolated schools are characterized by much higher levels of poverty, lower average test scores, lower levels of student achievement, with less-qualified teachers and fewer advanced courses — “[w]ith few exceptions, separate schools are still unequal schools.” See Erica Frankenberg et al., A Multiracial Society with Segregated Schools: Are We Losing the Dream? 11 (The Civil Rights Project, Harvard Univ. Jan. 2003), at http://www.civilright sproject.harvard.edu /research/ reseg03/AreWeLosing theDream.pdf) (hereinafter “Civil Rights Project“) (last visited October 11, 2005) (cited in Grutter, 539 U.S. at 345, 123 S.Ct. 2325 (Ginsburg, J., concurring)).

In Seattle, the threat of having to attend a racially concentrated or isolated school is not a theoretical or imagined problem.[17] As the district court found, the District “established that housing patterns in Seattle continue to be racially concentrated,” and would result in racially concentrated or isolated schools if school assignments were based solely on a student’s neighborhood or proximity to a particular high school. Parents I, 137 F.Supp.2d at 1235. Accordingly, the District’s Plan strives to ensure that patterns of residential segregation are not replicated in the District’s school assignments. Cf. Comfort, 418 F.3d at 29 (“The problem is that in Lynn, as in many other cities, minorities and whites often live in different neighborhoods. Lynn’s aim is to preserve local schools as an option without having the housing pattern of de facto segregation projected into the school system.”) (Boudin, C.J., concurring). Although Parents make much of the fact that “Seattle has never operated a 1178*1178 segregated school system,” and allege that “this is not a school desegregation case,” each court to review the matter has concluded that because of Seattle’s housing patterns, high schools in Seattle would be highly segregated absent race conscious measures. See Parents I, 137 F.Supp.2d at 1237; Parents II, 285 F.3d at 1239-40; Parents III, 294 F.3d at 1088; Parents IV, 72 P.3d at 153.

The district court found that, “[t]he circumstances that gave rise to the court-approved school assignment policies of the 1970s [e.g., Seattle’s segregated housing patterns] continue to be as compelling today as they were in the days of the district’s mandatory busing programs. . . . [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 1970s did not.” Parents I, 137 F.Supp.2d at 1235. Thus, it concluded that “[p]reventing the re-segregation of Seattle’s schools is . . . a compelling interest.” Id. at 1237; see id. at 1233-35. Several other courts have also conceived of a school district’s voluntary reduction or prevention of de facto segregation as a compelling interest. See Comfort, 418 F.3d at 14 (holding that the “negative consequences of racial isolation that Lynn seeks to avoid and the benefits of diversity that it hopes to achieve” constituted compelling interests); 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)); McFarland v. Jefferson County Pub. Sch., 330 F.Supp.2d 834, 851 (W.D.Ky.2004) (concluding that voluntary maintenance of the desegregated school system was a compelling state interest and the district could consider race in assigning students to comparable schools), aff’d 416 F.3d 513 (6th Cir.2005).[18] We join these courts in recognizing that school districts have a compelling interest in ameliorating real, identifiable de facto racial segregation.

The dissent, however, contends first that the District is not “desegregating” but rather is engaged in racial balancing. Bea, J., dissenting, infra. at 1197-1198. Further, for the dissent, segregation requires a state actor intentionally to separate the races; and in the absence of such offensive state conduct, the Supreme Court cases detailing the remedies for Fourteenth Amendment violations are of no relevance. Bea, J., dissenting, infra. at 1208, n. 17. Thus, without a court finding of de jure segregation the elected school board members of the District may not take voluntary, affirmative steps towards 1179*1179 creating a racially diverse student body. We disagree. The fact that de jure segregation is particularly offensive to our Constitution does not diminish the real harms of separation of the races by other means. “Segregation of white and colored children in public schools has a detrimental effect upon the colored children. The impact is greater when it has the sanction of law. . . .” Brown v. Bd. of Educ., 347 U.S. 483, 494, 74 S.Ct. 686, 98 L.Ed. 873 (1954) (emphasis added). The benefits that flow from integration (or desegregation) exist whether or not a state actor was responsible for the earlier racial isolation. Brown’s statement that “in the field of public education . . . [s]eparate educational facilities are inherently unequal” retains its validity today. Id. at 495, 74 S.Ct. 686. The District is entitled to seek the benefits of racial integration and avoid the harms of segregation even in the absence of a court order deeming it a violator of the U.S. Constitution.

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.[19] 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 Los Angeles, 439 U.S. 1380, 1383, 99 S.Ct. 40, 58 L.Ed.2d 88 (1978) (denying a request to stay implementation of a voluntary desegregation 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.”); Washington v. Seattle Sch. Dist. No. 1, 458 U.S. at 480, 487, 102 S.Ct. 3187 (holding unconstitutional the state initiative that blocked the Seattle School District’s use of mandatory busing to remedy de facto segregation).

In sum, we hold that the District’s interests in obtaining the educational and social benefits of racial diversity in secondary education and in avoiding racially concentrated or isolated schools resulting from Seattle’s segregated housing pattern are clearly compelling.

C. Narrow Tailoring

We must next determine whether the District’s use of the race-based tiebreaker is narrowly tailored to achieve its compelling interests. See Grutter, 539 1180*1180 U.S. at 333, 123 S.Ct. 2325. 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. Richmond v. J.A. Croson Co., 488 U.S. 469, 493, 109 S.Ct. 706, 102 L.Ed.2d 854 (1989). Here, our analysis is framed by the Court’s narrow tailoring analysis in Grutter and Gratz, which, though informed by considerations specific to the higher education context, substantially guides our inquiry. See Grutter, 539 U.S. at 334, 123 S.Ct. 2325 (stating that the narrow tailoring inquiry is context-specific and must be “calibrated to fit the distinct issues raised” in a given case, taking “relevant differences into account”) (internal quotation marks omitted).

In Gratz, the Court held unconstitutional the University of Michigan’s undergraduate admissions program, which automatically assigned 20 points on the admissions scale to an applicant from an underrepresented racial or ethnic minority group. 539 U.S. at 255, 272, 123 S.Ct. 2411. In Grutter, by contrast, the Court upheld the University of Michigan Law School’s admissions policy, which took race into account as one of several variables in an individual’s application. 539 U.S. at 315-16, 340, 123 S.Ct. 2325. The law school’s policy also attempted to ensure that a “critical mass” of underrepresented minority students would be admitted in order to realize the benefits of a diverse student body.[20] Id. at 316, 123 S.Ct. 2325.

In its analysis, the Court identified five hallmarks of a narrowly tailored affirmative action plan: (1) individualized consideration of applicants; (2) the absence of quotas; (3) serious, good-faith consideration of race-neutral alternatives to the affirmative action program; (4) that no member of any racial group was unduly harmed; and (5) that the program had a sunset provision or some other end point. Smith v. Univ. of Washington, 392 F.3d 367, 373 (9th Cir.2004); Comfort, 418 F.3d at 17 (characterizing Grutter as outlining a “four-part narrow tailoring inquiry”).

Hallmarks two through five are applicable here despite significant differences between the competitive admissions plans at issue in Gratz and Grutter and the District’s high school assignment Plan. The first hallmark, however, is less relevant to our analysis because of the contextual differences between institutions of higher learning and public high schools.

1. Individualized, Holistic Consideration of Applicants

a. An applicant’s qualifications

In the context of university admissions, where applicants compete for a limited number of spaces in a class, the Court in Grutter and Gratz focused its inquiry on the role race may play in judging an applicant’s qualifications. The Court’s underlying concern was that the “admissions policy is flexible enough to consider all pertinent elements of diversity in light of the particular qualifications of each applicant, and to place them on the same footing for consideration, although not necessarily according them the same weight.” Grutter, 539 U.S. at 337, 123 S.Ct. 2325 (emphasis added) (internal quotation marks omitted); see Adarand, 515 U.S. at 211, 115 S.Ct. 2097 (“The injury in cases of this kind is that a discriminatory classification prevent[s] the plaintiff from competing on an equal footing.“) (emphasis added) (internal quotation marks omitted). 1181*1181 The focus on fair competition is due, in part, to the stigma that may attach if some individuals are viewed as unable to achieve success without special protection. 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) (“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.”).

In Grutter and Gratz, in order to prevent race from being used as a mechanical proxy for an applicant’s qualifications, the Court required individualized, holistic consideration of each applicant across a broad range of factors (of which race may be but one). Grutter, 539 U.S. at 336-37, 123 S.Ct. 2325; see Gratz, 539 U.S. at 272, 123 S.Ct. 2411 (holding that the undergraduate admissions policy was not narrowly tailored because the “automatic distribution of 20 points has the effect of making `the factor of race . . . decisive‘ for virtually every minimally qualified underrepresented minority applicant”) (emphasis added). This focus on an applicant’s qualifications — whether these qualifications are such things as an applicant’s test scores, grades, artistic or athletic ability, musical talent or life experience — is not applicable when there is no competition or consideration of qualifications at issue.

All of Seattle’s high school students must and will be placed in a Seattle public school.[21] Students’ relative qualifications are irrelevant because regardless of their academic achievement, sports or artistic ability, musical talent or life experience, any student who wants to attend Seattle’s public high schools is entitled to an assignment; no assignment to any of the District’s high schools is tethered to a student’s qualifications. Thus, no stigma results from any particular school assignment.[22] Accordingly, the dangers that are present in the university context — of substituting racial preference for qualification-based competition — are absent here. See Comfort, 418 F.3d at 18 (“Because transfers under the Lynn Plan are not tied to merit, the Plan’s use of race does not risk imposing stigmatic harm by fueling the stereotype that `certain groups are unable to achieve success without special protection.'”) (quoting Bakke, 438 U.S. at 298, 98 S.Ct. 2733).

b. Differences in compelling interests

The Court’s requirement of individualized, holistic review in Grutter is also more 1182*1182 relevant to the compelling interest advanced by the law school (“the robust exchange of ideas” fostered by viewpoint diversity) than it is to the District’s (racial diversity and avoiding racially concentrated or isolated schools). See Grutter, 539 U.S. at 337, 123 S.Ct. 2325. The Court noted that the law school did not “limit in any way . . . the broad range of qualities and experiences that may be considered valuable contributions to student body diversity.” Id. at 338, 123 S.Ct. 2325. To this end, the law school’s policy made clear that “[t]here are many possible bases for diversity admissions, and provide[d] examples of admittees who have lived or traveled widely abroad, are fluent in several languages, have overcome personal adversity and family hardship, have exceptional records of extensive community service, and had successful careers in other fields.” Id. (internal quotation marks and citations omitted). These multiple bases for diversity ensure the “classroom discussion is livelier, more spirited, and simply more enlightening and interesting when the students have the greatest possible variety of backgrounds.” Id. at 330, 123 S.Ct. 2325 (internal citations omitted).

Although the District’s Plan, like the plan in Grutter, is designed to achieve the educational and social benefits of diversity, including bringing “different viewpoints and experiences to classroom discussions,” see “Statement Reaffirming Diversity Rationale,” viewpoint diversity in the law school and high school contexts serves different albeit overlapping ends. In the law school setting, viewpoint diversity fosters the “robust exchange of ideas.” Grutter, 539 U.S. at 324, 123 S.Ct. 2325; see Comfort, 418 F.3d at 16 (“[L]ively classroom discussion is a more central form of learning in law schools (which prefer the Socratic method) than in a K-12 setting.”). In the high school context, viewpoint diversity fosters racial and civic understanding.[23] For example, Eric Benson, the principal of Nathan Hale High School, one of the District’s most popular schools, testified that as a result of racial diversity in the classroom, “students of different races and backgrounds tend to have significant interactions both in class, and outside of class. When I came to Nathan Hale, there were racial tensions in the school, reflected in fighting and disciplinary problems. These kind of problems have, to a large extent, disappeared.”

In addition, the law school takes other diversity factors, besides race and ethnicity, into consideration in order to achieve its other compelling interest — cultivating a group of national leaders. For example, extensive travel, fluency in foreign languages, extensive community service and successful careers in other fields demonstrate that a candidate is somehow exceptional or out of the ordinary. cf. Gratz, 539 U.S. at 273, 123 S.Ct. 2411 (disapproving of the undergraduate admissions plan, in part, because of its failure to consider whether an applicant was extraordinary and noting that “[e]ven if [a] student[‘s] `extraordinary artistic talent’ rivaled that of Monet or Picasso, the applicant would receive, at most, five points” as opposed to 1183*1183 the automatic 20 points given to an applicant from an underrepresented minority). In contrast, the District is required to educate all high school age children, both the average and the extraordinary, regardless of individual leadership potential.

The District also has a second compelling interest that is absent from the university context — ensuring that its school assignments do not replicate Seattle’s segregated housing patterns. The holistic review necessary to achieve viewpoint diversity in the university context, across a broad range of factors (of which race may be but one), is not germane to the District’s compelling interest in preventing racial concentration or racial isolation. Because race itself is the relevant consideration when attempting to ameliorate de facto segregation, the District’s tiebreaker must necessarily focus on the race of its students. See Comfort, 418 F.3d at 18 (holding that when racial diversity is the compelling interest — “[t]he only relevant criterion, then, is a student’s race; individualized consideration beyond that is irrelevant to the compelling interest”); Brewer v. W. Irondequoit Cent. Sch. Dist., 212 F.3d at 752 (“If reducing racial isolation is — standing alone — a constitutionally permissible goal, . . . then there is no more effective means of achieving that goal than to base decisions on race.”). We therefore conclude that if a noncompetitive, voluntary student assignment plan is otherwise narrowly tailored, a district need not consider each student in a individualized, holistic manner.[24]

The dissent insists that absent such individualized consideration, the District’s plan cannot serve a compelling interest and is not narrowly tailored to protect individuals from group classifications by race. Bea, J., dissenting, infra. at 1209. This is a flawed reading of the Fourteenth Amendment.[25] The District’s compelling interest is to avoid the harms of racial isolation for all students in the Seattle school district. As we have explained, to accomplish that objective the District may look to the racial consequences of honoring the preferred choices of individual students (and their parents). It is true that for some students their first choice of school, based on geographical proximity, will be denied because other students’ choices are granted in order to advance the overall interest in maintaining racially diverse school enrollments. The Fourteenth Amendment in this context does not preclude the District from honoring racial diversity at the expense of geographical proximity. We must not forget that “race unfortunately still matters,” Grutter, 539 U.S. at 333, 123 S.Ct. 2325, and it is race that is the relevant consideration here.

1184*1184 In sum, the contextual differences between public high schools and selective institutions of higher learning make the first of the Grutter hallmarks ill-suited for our narrow tailoring inquiry.[26] The remaining hallmarks, however, are relevant and control our analysis.

2. Absence of Quotas

In Grutter, the Court approved the law school’s plan, in part, because it did not institute a quota, whereby a fixed number of slots are reserved exclusively for minority groups, thereby insulating members of those groups from competition with other candidates.[27] 539 U.S. at 335, 123 S.Ct. 2325. Although the law school’s plan did not seek to admit a set number or percentage of minority students, during the height of the admission’s season, the law school would consult “daily reports” that kept track of the racial composition of the incoming class. Id. at 318, 123 S.Ct. 2325. The Court held that this attention to numbers did not transform the law school plan into a quota, but instead demonstrated that the law school sought to enroll a critical mass of minority students in order “to realize the educational benefits of a diverse student body.” Id. Similarly, we conclude that the District’s 15 percent plus or minus variance is not a quota because it does not reserve a fixed number of slots for students based on their race, but instead it seeks to enroll a critical mass of white and nonwhite students in its oversubscribed schools in order to realize its compelling interests.[28]

a. No fixed number of slots

The District’s race-based tiebreaker does not set aside a fixed number of slots for nonwhite or white students in any of the District’s schools. The tiebreaker is used only so long as there are members of the underrepresented race in the applicant pool for a particular oversubscribed school. If the number of students of that race who have applied to that school is exhausted, no further action is taken, even if the 15 1185*1185 percent variance has not been satisfied. That is, if the applicant pool has been exhausted, no students are required or recruited to attend a particular high school in order to bring it within the 15 percent plus or minus range for that year.

Moreover, the number of white and nonwhite students in the high schools is flexible and varies from school to school and from year to year.[29] This variance in the number of nonwhite and white students throughout the District’s high schools is because, under the Plan, assignments are based on students’ and parents’ preferences.[30] The tiebreakers come into play in the assignment process only when a school is oversubscribed. As Morgan Lewis, the Manager of Enrollment Planning, Technical Support and Demographics, testified, “If all the parents . . . don’t pick [a] school in a massive number, then everyone gets in. And so it’s . . . a case where the choice patterns, the oversubscription . . . [is] the reason the [tiebreaker] kicks in. . . . Everything happens when more people want the seats. And why they want the seats sometimes we don’t know.”

b. Critical mass

Within this flexible system, where parental and student choices drive the assignments to particular schools, the District seeks to enroll and maintain a relatively stable critical mass of white and nonwhite students in each of its oversubscribed high schools in order to achieve its compelling interest in racial diversity and to prevent the assignments from replicating Seattle’s segregated housing patterns. Faced with the question of what constituted a critical mass of students in this particular context, the District determined that a critical mass was best achieved by adopting the 15 percent plus or minus variance tied to demographics of students in the Seattle public schools. Thus, when an oversubscribed high school has more than 75 percent nonwhite students (i.e., more than 15 percent above the overall 60 percent nonwhite student population) and less than 25 percent white students, or when it has less than 45 percent nonwhite students (i.e., more than 15 percent below the overall 60 percent nonwhite student population) and more than 55 percent white students, the school is considered racially concentrated or isolated, meaning that it lacks a critical mass of students needed “to realize the educational benefits of a diverse student body.”

Parents attack the District’s use of the 15 percent plus or minus variance tied to the District’s school population demographics because they believe that the District cannot use race at all in its assignment process. We have rejected this argument, however, applying Grutter and Gratz. See supra Part II.B. Alternatively, Parents contend that the District’s goal of enrolling between 75 and 45 percent nonwhite 1186*1186 students and between 25 and 55 percent white students in its oversubscribed schools establishes a quota, not a critical mass. They note that the critical mass sought by the law school in Grutter was smaller, consisting of between 12 and 20 percent of underrepresented minority students in each law school class.

Parents’ argument, however, ignores Grutter’s admonition that the narrow tailoring inquiry be context-specific. First, like the District’s enrollment goals, which are tied to the demographics of the Seattle schools’ total student population, the law school’s goal of enrolling between 12 to 20 percent of underrepresented minorities in a given year was tied to the demographics of its applicant pool.[31] Second, in tying the use of the tiebreaker to the District’s demographics with a 15 percent plus or minus trigger point, the District adopted a common benchmark in the context of voluntary and court-ordered school desegregation plans. As the District’s expert testified,

Most of the cases I’ve participated in . . . generally worked with numbers that reflect the racial composition of the school district but, at the same time, tr[ied] to allow the district sufficient flexibility so that it would not have to regularly and repeatedly move students on a short-term basis simply to maintain some specific number. That’s why we see ranges of plus or minus 15 percent in most cases of school desegregation.

Even Parents’ expert testified that school districts throughout the country determine whether a district is sufficiently desegregated by looking to the “population of the district” in question. See also Comfort, 418 F.3d at 21 (holding that a “transfer policy conditioned on district demographics (+/-10-15%)” was not a quota because it “reflects the defendants’ efforts to obtain the benefits of diversity in a stable learning environment”); Belk v. Charlotte-Mecklenburg Bd. of Educ., 233 F.3d 232, 287-88 (4th Cir.2000) (Traxler, J., dissenting) (citing to a book written by David J. Armor, Parents’ expert, Forced Justice: School Desegregation and the Law 160 (1995), which observed that over 70 percent of the school districts with desegregation plans use a variance of plus or minus 15 percent or greater); cf. 34 C.F.R. § 280.4(b) (defining “minority group isolation” as a “condition in which minority group children constitute more than 50 percent of the enrollment of [a] school”). Given this empirically and time-tested notion of critical mass in the public high school desegregation context, it would make little sense to force the District to utilize the same percentages that constituted a critical mass in the elite law school context to determine what constitutes a critical mass for Seattle public high schools. See Grutter, 539 U.S. at 336, 123 S.Ct. 2325 (“[S]ome attention to numbers, without more, does not transform a flexible admissions system into a rigid quota.”) (internal quotation marks and citations omitted).

Accordingly, we conclude that the District’s 15 percent plus or minus trigger point tied to the demographics of the Seattle school population is not a quota. It is a context-specific, flexible measurement of racial diversity designed to attain and maintain a critical mass of white and nonwhite students in Seattle’s public high schools.

1187*1187 3. Necessity of the Plan and Race-Neutral Alternatives

Narrow tailoring also requires us to consider the necessity of the race-based plan or policy in question and whether there are equally effective, race-neutral alternatives.

a. Necessity of the Plan

The District argues that the compelling interests that it seeks are directly served by the race-based tiebreaker. The tiebreaker allows the District to balance students’ and parents’ choices among high schools with its broader compelling interests — achieving the educational and social benefits of diversity and the benefits specific to the secondary school context, and discouraging a return to enrollment patterns based on Seattle’s racially segregated housing pattern.

i. Need for race-based tiebreaker

When the District moved from its controlled choice plan to the current Plan, see supra Part I.A, 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 — created the risk that Seattle’s high school enrollment would again do no more than reflect its segregated housing patterns. See supra Part II.C.2.

It is this de facto residential segregation across a white/nonwhite axis that the District has battled historically and that it seeks to ameliorate by making the integration tiebreaker a part of its open choice Plan.[32] The District, mindful of both Seattle’s history and future, appropriately places its focus here. In the 2001-02 school year, the integration tiebreaker operated in three high schools (that is, three high schools were oversubscribed and deviated by more than 15 percent from the ratio of white to nonwhite students district-wide). The integration tiebreaker served to alter the imbalance in the schools in which it operated in a minimally intrusive manner. The tiebreaker, therefore, successfully achieved the District’s compelling interests.

ii. White/Nonwhite distinction

Parents argue that the District paints with too broad a brush by distinguishing only between white and nonwhite students, without taking into account the diversity within the “nonwhite” group. However, the District’s choice to increase diversity along the white/nonwhite axis is rooted in Seattle’s history and current reality of de facto segregation resulting from Seattle’s segregated housing patterns. The white/nonwhite distinction is narrowly tailored to prioritize movement of students from the north of the city to the south of city and vice versa. This white/nonwhite focus is also consistent with the history of public school desegregation measures throughout the 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 “minority group.” 34 C.F.R. § 280.4(b); see Grutter, 539 U.S. at 316, 1188*1188 123 S.Ct. 2325 (noting that the law school sought to enroll a critical mass of “minority students,” a category that included African Americans, Hispanics and Native Americans); Comfort, 418 F.3d at 22 (“By increasing diversity along the white/nonwhite axis, the Plan reduced racial tensions and produced positive educational benefits. Narrow tailoring does not require that Lynn ensure diversity among every racial and ethnic subgroup as well.“) (emphasis added).

b. 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, “[n]arrow tailoring does not require exhaustion of every conceivable race-neutral alternative.” Id. Furthermore, 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.[33] See id. at 328, 340, 123 S.Ct. 2325. Here, the record reflects that the District reasonably concluded that a race-neutral alternative would not meet its goals.

i. Using poverty as an alternative measure of diversity

The record demonstrates that the School Board considered using a poverty tiebreaker in place of the race-based tiebreaker. It concluded, however, that this proxy device would not achieve its compelling interest in achieving racial diversity, and had other adverse effects. Although there was no formal study of the proposal by District staff, Board members’ testimony revealed two legitimate reasons why the Board rejected the use of poverty to reach its goal of racial diversity. First, the Board concluded that it is insulting to minorities and often inaccurate to assume that poverty correlates with minority status. Second, for the group of students for whom poverty would correlate with minority 1189*1189 status, the implementation would have been thwarted by high school students’ understandable reluctance to reveal their socioeconomic status to their peers.

Because racial diversity is a compelling interest, the District may permissibly seek it if it does so in a narrowly tailored manner. We do not require the District to conceal its compelling interest of achieving racial diversity and avoiding racial concentration or isolation through the use of “some clumsier proxy device” such as poverty. See Comfort, 418 F.3d at 29 (Boudin, C.J., concurring).

ii. The Urban League plan

Parents also assert that the District should have more formally considered an Urban League proposal, which did not eliminate the integration tiebreaker but merely considered it after other factors. The Urban League plan was a comprehensive plan seeking to enhance the quality of education in Seattle’s schools by focusing on educational organization, teacher quality, parent-teacher interaction, raising curricular standards, substantially broadening the availability of specialized and magnet programs (which could attract a broader cross-section of students to undersubscribed schools) and supporting extra-curricular development. The plan proposed decreasing the School District’s reliance on race in the assignment process by pairing neighborhoods with particular schools and creating a type of neighborhood/regional school model. Under the Urban League plan, preference initially would be given to students choosing a school in their paired region, and the existing racial tiebreaker would be demoted from second to third in the process of resolving any remaining oversubscription. The plan also suggested adding an eleventh high school.

Board members testified that they rejected the plan because of the high value the District places on parental and student choice. Moreover, given Seattle’s segregated housing patterns, by prioritizing a neighborhood/regional school model where students are assigned to schools close to their homes, the Urban League plan did not sufficiently ensure the achievement of the District’s compelling interests in racial diversity and avoidance of racial concentration or isolation. As one member of the School Board testified, “[it] would become Controlled Choice all over again. That’s basically what Controlled Choice was, [ ] a regional plan; it controlled your options by using regions or geography.” It was therefore permissible for the District to reject a plan that neither comported with its priorities nor achieved its compelling interests.

iii. Lottery

Parents additionally contend in this court that the District should have considered using a lottery to assign students to the oversubscribed high schools. As an initial matter, we note that Parents did not argue before the district court that a lottery was a workable race-neutral alternative that would achieve the Districts’ compelling interests. Parents now argue on appeal, however, that a lottery would achieve the District’s compelling interests without having to resort to the race-based tiebreaker. They ask us to assume that because approximately 82 percent of all students want to attend one of Seattle’s oversubscribed schools, the makeup of this 82 percent, as well as that of the applicant pool for each school, mirrors the demographics of the District (60 percent white and 40 percent nonwhite). Employing this assumption, Parents also ask us to assume that a random lottery drawing from this pool would produce a student body in each of the oversubscribed schools that falls within the District’s 15 percent plus or minus variance. These assumptions, however, are not supported — indeed, are undercut 1190*1190 —by the factual record. For example, Superintendent Olchefske explained that District patterns indicate that more people choose schools close to home. That would mean that the pool of applicants would be skewed in favor of the demographic of the surrounding residential area. That is, the applicant pool for the north area oversubscribed high schools would have a higher concentration of white students and the applicant pool for the south area oversubscribed high school would have a higher concentration of nonwhite students. Thus, random sampling from such a racially skewed pool would produce a racially skewed student body. As one Board member testified, a lottery was not a viable alternative because “[i]f applicants are overwhelmingly majority and you have a lottery, then your lottery — the pool of your lottery kids are going to be overwhelmingly majority. We have a diversity goal.”

Although the District has the burden of demonstrating that its Plan is narrowly tailored, see Gratz, 539 U.S. at 270, 123 S.Ct. 2411, it need not “exhaust[ ] every conceivable race-neutral alternative.” Grutter, 539 U.S. at 339, 123 S.Ct. 2325. Parents’ belated and bald assertion that a lottery could achieve the District’s compelling interests, without any evidence to support their claim, fails to demonstrate that a lottery is a viable race-neutral alternative. See id. at 340, 123 S.Ct. 2325 (dismissing the race-neutral alternative of “percentage plans,” advocated by the United States in an amicus brief, because the “United States [did] not . . . explain how such plans could work for graduate and professional schools”); Comfort, 418 F.3d at 23 (noting that Lynn rejected the use of a lottery in place of the race-based tiebreaker and holding that “Lynn must keep abreast of possible alternatives as they develop . . . but it need not prove the impracticability of every conceivable model for racial integration”) (internal citation omitted).

c. The District’s use of race

The dissent posits variables the District could use instead of race, for example, embracing the San Francisco school district’s approach as a possible model for integration that would meet the dissent’s criteria. Bea, J., dissenting, infra. at 1218, n. 26. Perhaps San Francisco has experienced success (however that school district defines it) in its multi-variable plan — the details and evaluations of which are not in the record. The District is free to consider the San Francisco model when it engages in the annual review of its own Plan. However, even assuming that San Francisco’s plan is working, that does not mean that it must be used by other cities in other states. Much can be gained from the various states employing locally appropriate means to achieve desirable ends. In our system, where states are considered laboratories to be used to experiment with myriad approaches to resolving social problems, we certainly should not punish one school district for not adopting the approach of another. Justice Brandeis said it well,

There must be power in the States and the Nation to remould, through experimentation, our economic practices and institutions to meet changing social and economic needs. . . . To stay experimentation in things social and economic is a grave responsibility. Denial of the right to experiment may be fraught with serious consequences to the Nation. It is one of the happy incidents of the federal system that a single courageous State may, if its citizens choose, serve as a laboratory; and try novel social and economic experiments without risk to the rest of the country. 1191*1191 New State Ice Co. v. Liebmann, 285 U.S. 262, 311, 52 S.Ct. 371, 76 L.Ed. 747 (1932) (Brandeis, J., dissenting).

In sum, the District made a good faith effort to consider feasible race-neutral alternatives and permissibly rejected them in favor of a system involving a sibling preference, a race-based tiebreaker and a proximity preference. Over the long history of the District’s efforts to achieve desegregated schools, it has experimented with many alternatives, including magnet and other special-interest programs, which it continues to employ, and race-conscious districting. But when a racially diverse school system is the goal (or racial concentration or isolation is the problem), there is no more effective means than a consideration of race to achieve the solution. Even Parents’ 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 is self-evident: When racial diversity is a principal element of the school district’s compelling interest, then a narrowly tailored plan may explicitly take race into account.[34] Cf. Hunter v. Regents of Univ. of Cal., 190 F.3d 1061, 1067 (9th Cir.1999) (upholding as narrowly tailored the admissions policy of an elementary school — operated as a research laboratory — that explicitly considered race in pursuit of a racially balanced research sample).

4. Undue Harm

A narrowly tailored plan ensures that no member of any racial group is unduly harmed. Grutter, 539 U.S. at 341, 123 S.Ct. 2325. Parents argue that every student who is denied his or her choice of schools because of the integration tiebreaker suffers a constitutionally significant burden. We agree with the Supreme Court of Washington, however, in its assessment that the District’s Plan imposes a minimal burden that is shared equally by all of the District’s students. Parents IV, 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). As that court noted, it is well established that “there [is] no right under Washington law to attend a local school or the school of the student’s choice.” Id. at 159.[35] Indeed, public schools, unlike universities, have a tradition of compulsory assignment. See Bazemore v. Friday, 478 U.S. 385, 408, 106 S.Ct. 3000, 92 L.Ed.2d 315 (1986) (White, J., concurring) (noting that “school boards customarily have the power to create school attendance areas and otherwise designate the school that particular students may attend”). When an applicant’s qualifications are not under consideration at all, there is no notion that one student is entitled to a place at any particular school. See Comfort, 418 F.3d at 20 (“The denial of a transfer under the [District’s] Plan is … markedly different from the denial of a spot at a unique or selective educational institution.”).

1192*1192 Moreover, it is undisputed that the race-based tiebreaker does not uniformly benefit one race or group to the detriment of another. At some schools, white students are given preference over nonwhite students, and, at other schools, nonwhite students are given preference over white students. For example, in the 2000-01 school year, 89 more white students were assigned to Franklin, one of Seattle’s most popular schools, than would have been assigned absent the tiebreaker; 107 more nonwhite students were assigned to Ballard, another of Seattle’s most popular schools, than would have been assigned absent the tiebreaker; 27 more nonwhite students were assigned to Nathan Hale than would have been assigned absent the tiebreaker; and 82 more nonwhite students were assigned to Roosevelt than would have been absent the tiebreaker.[36]

In sum, because (1) the District is entitled to assign all students to any of its schools, (2) no student is entitled to attend any specific school and (3) the tiebreaker does not uniformly benefit any race or group of individuals to the detriment of another, the tiebreaker does not unduly harm any students in the District.

5. Sunset Provision

A narrowly tailored plan must be limited not only in scope, but also in time. Grutter, 539 U.S. at 342, 123 S.Ct. 2325. The Court held in Grutter that this durational requirement can be met by “periodic reviews to determine whether racial preferences are still necessary to achieve student body diversity.” Id. The District’s Plan includes such reviews. It revisits the Plan annually and has demonstrated its ability to be responsive to parents’ and students’ choice patterns and to the concerns of its constituents. For example, in 2000, when a higher than normal number of students selected the same schools, the Board responded by increasing the race-based trigger from 10 percent to a 15 percent deviation from the school population, adopting the thermostat that turns off the tiebreaker as soon as the school has come within the 15 percent plus or minus trigger point and by using the tiebreaker solely for the incoming ninth grade class.

With respect to the dissent’s concern for a “logical end point,” Bea, J., dissenting, infra. at 1217, like Justice O’Connor this court shares in the hope that “25 years from now, the use of racial preferences will no longer be necessary to further the interest approved today.” Grutter, 539 U.S. at 343, 123 S.Ct. 2325. We expect that the District will continue to review its Plan, and we presume, as did the Court in Grutter, that school officials will demonstrate a good faith commitment to monitoring the continued need for the race-based tiebreaker and terminating its use when that need ends.[37] See 539 U.S. at 343, 123 S.Ct. 2325.

III. Conclusion

For the foregoing reasons, we hold that the Plan adopted by the Seattle School 1193*1193 District for high school assignments is constitutional and the use of the race-based tiebreaker is narrowly tailored to achieve the District’s compelling interests. Accordingly, we AFFIRM the district court’s judgment.

AFFIRMED.

KOZINSKI, Circuit Judge, concurring:

My colleagues in the majority and the dissent have written extensively and well. Given the exacting standard they are attempting to apply, I cannot say that either is clearly wrong. But there is something unreal about their efforts to apply the teachings of prior Supreme Court cases, all decided in very different contexts, to the plan at issue here. I hear the thud of square pegs being pounded into round holes. Ultimately, neither analysis seems entirely persuasive.

I start as did our eminent colleague Chief Judge Boudin of the First Circuit, in commenting on a highly-analogous plan adopted by the city of Lynn, Massachusetts:

[The] plan at issue in this case is fundamentally different from almost anything that the Supreme Court has previously addressed. It is not, like old-fashioned racial discrimination laws, aimed at oppressing blacks, e.g., Brown v. Bd. of Educ., 347 U.S. 483, 74 S.Ct. 686, 98 L.Ed. 873 (1954); Strauder v. West Virginia, 100 U.S. 303, 25 L.Ed. 664 (1880); nor, like modern affirmative action, does it seek to give one racial group an edge over another (either to remedy past discrimination or for other purposes). E.g., Adarand Constructors, Inc. v. Pena, 515 U.S. 200, 115 S.Ct. 2097, 132 L.Ed.2d 158 (1995). By contrast to Johnson v. California, ___ U.S. ___, 125 S.Ct. 1141, 160 L.Ed.2d 949 (2005), the plan does not segregate persons by race. See also Loving v. Virginia, 388 U.S. 1, 87 S.Ct. 1817, 18 L.Ed.2d 1010 (1967). Nor does it involve racial quotas. E.g., Regents of the Univ. of Cal. v. Bakke, 438 U.S. 265, 98 S.Ct. 2733, 57 L.Ed.2d 750 (1978).

Comfort v. Lynn Sch. Comm., 418 F.3d 1, 27 (1st Cir.2005) (Boudin, C.J., concurring).

These are meaningful differences. When the government seeks to use racial classifications to oppress blacks or other minorities, no conceivable justification will be sufficiently compelling. See, e.g., Yick Wo v. Hopkins, 118 U.S. 356, 374, 6 S.Ct. 1064, 30 L.Ed. 220 (1886). Similarly, when lawyers use peremptory challenges to exclude jurors of a particular race, thereby denying them the right to participate in government service, they must justify their challenges based on objective, non-racial considerations; justifications based on race will be rejected out of hand, no matter how compelling they might seem. See Batson v. Kentucky, 476 U.S. 79, 85-88, 106 S.Ct. 1712, 90 L.Ed.2d 69 (1986). When government seeks to segregate the races, as in Johnson, the courts will look with great skepticism at the justifications offered in support of such programs, and will reject them when they reflect assumptions about the conduct of individuals based on their race or skin color. See Johnson, 125 S.Ct. at 1154 (Stevens, J., dissenting) (concluding that California’s policy of racially segregating inmates “supports the suspicion that the policy is based on racial stereotypes and outmoded fears about the dangers of racial integration”). When the government engages in racial gerrymandering, it not only keeps the races apart, but exacerbates racial tensions by making race a proxy for political power. See Shaw v. Reno, 509 U.S. 630, 648, 113 S.Ct. 2816, 125 L.Ed.2d 511 (1993) (“When a district obviously is created solely to effectuate the perceived common interests of one racial group, elected officials are more likely to believe that their primary 1194*1194 obligation is to represent only the members of that group, rather than their constituency as a whole.”). Programs seeking to help minorities by giving them preferences in contracting, see, e.g., Adarand, and education, see, e.g., Bakke, benign though they may be in their motivations, pit the races against each other, and cast doubts on the ability of minorities to compete with the majority on an equal footing.

The Seattle plan suffers none of these defects. It certainly is not meant to oppress minorities, nor does it have that effect. No race is turned away from government service or services. The plan does not segregate the races; to the contrary, it seeks to promote integration. There is no attempt to give members of particular races political power based on skin color. There is no competition between the races, and no race is given a preference over another. That a student is denied the school of his choice may be disappointing, but it carries no racial stigma and says nothing at all about that individual’s aptitude or ability. The program does use race as a criterion, but only to ensure that the population of each public school roughly reflects the city’s racial composition.

Because the Seattle plan carries none of the baggage the Supreme Court has found objectionable in cases where it has applied strict scrutiny and narrow tailoring, I would consider the plan under a rational basis standard of review. By rational basis, I don’t mean the standard applied to economic regulations, where courts shut their eyes to reality or even invent justifications for upholding government programs, see, e.g., Williamson v. Lee Optical of Okla., Inc., 348 U.S. 483, 75 S.Ct. 461, 99 L.Ed. 563 (1955), but robust and realistic rational basis review, see, e.g., City of Cleburne v. Cleburne Living Ctr., Inc., 473 U.S. 432, 105 S.Ct. 3249, 87 L.Ed.2d 313 (1985), where courts consider the actual reasons for the plan in light of the real-world circumstances that gave rise to it.

Under this standard, I have no trouble finding the Seattle plan constitutional. Through their elected officials, the people of Seattle have adopted a plan that emphasizes school choice, yet tempers such choice somewhat in order to ensure that the schools reflect the city’s population. Such stirring of the melting pot strikes me as eminently sensible.

The record shows, and common experience tells us, that students tend to select the schools closest to their homes, which means that schools will reflect the composition of the neighborhood where they are located. Neighborhoods, however, do not reflect the racial composition of the city as a whole. In Seattle, “as in many other cities, minorities and whites often live in different neighborhoods.” Comfort, 418 F.3d at 29 (Boudin, C.J., concurring). To the extent that students gravitate to the schools near their homes, the schools will have the same racial composition as the neighborhood. This means that student patterns of interacting primarily with members of their own race that are first developed by living in racially isolated neighborhoods will be continued and exacerbated by the school experience.

It is difficult to deny the importance of teaching children, during their formative years, how to deal respectfully and collegially with peers of different races. Whether one would call this a compelling interest or merely a highly rational one strikes me as little more than semantics. The reality is that attitudes and patterns of interaction are developed early in life and, in a multicultural and diverse society such as ours, there is great value in developing the ability to interact successfully with individuals who are very different from oneself. It is important for the individual student, to be sure, but it is also vitally important for us as a society.

1195*1195 It may be true, as the dissent suggests, that students are influenced far more by their experiences in the home, church and social clubs they attend outside of school. But this does not negate the fact that time spent in school and on school-related activities, which may take up as much as half of a student’s waking hours, nevertheless has a significant impact on that student’s development. The school environment forces students both to compete and cooperate in the classroom, as well as during extracurricular activities ranging from football to forensics. Schoolmates often become friends, rivals and romantic partners; learning to deal with individuals of different races in these various capacities cannot help but foster the live-and-let-live spirit that is the essence of the American experience. I believe this is a rational objective for an educational system—every bit as rational as teaching the three Rs, advanced chemistry or driver’s education. Schools, after all, don’t simply prepare students for further education, though they certainly can and should do that; good schools prepare students for life, by instilling skills and attitudes that will serve them long after their first year of college.

To borrow Judge Boudin’s words once again, the plan here is “far from the original evils at which the Fourteenth Amendment was addressed…. This is not a case in which, against the background of core principles, all doubts should be resolved against constitutionality.” Comfort, 418 F.3d at 29 (Boudin, C.J., concurring). I am acutely mindful of the Supreme Court’s strong admonition only last Term that any and all racial classifications must be adjudged under the strict scrutiny standard of review. See Johnson, 125 S.Ct. at 1146 (citing cases). But the Supreme Court’s opinions are necessarily forged by the cases presented to it; where the case at hand differs in material respects from those the Supreme Court has previously decided, I would hope that those seemingly categorical pronouncements will not be applied without consideration of whether they make sense beyond the circumstances that occasioned them.

When the Supreme Court does review the Seattle plan, or one like it, I hope the justices will give serious thought to bypassing strict—and almost always deadly—scrutiny, and adopt something more akin to rational basis review. Not only does a plan that promotes the mixing of races deserve support rather than suspicion and hostility from the judiciary, but there is much to be said for returning primacy on matters of educational policy to local officials. Long past is the day when losing an election or a legislative vote on a hotly contested issue was considered the end of the matter—at least until the next election when the voters might “throw the rascals out.” Too often nowadays, an election or a vote is a mere precursor to litigation, with the outcome of the dispute not known until judges decide the case many years later.

Whatever else the strict scrutiny standard of review may do, it most certainly encourages resort to the courts and often delays implementation of a program for years. The more complex and exacting the standard of review, the more uncertain the outcome, and the greater are the incentives for the parties to bloat the record with depositions, expert reports, exhibits, documents and various other materials they hope will catch the eye of the judges who ultimately decide the issue. This is a perfectly fine example, the litigation having taken over five years so far, generating 11 published opinions from the 24 judges who have considered the matter in the federal and state courts. In the meantime, the plan was put on hold, and at least one class has entered and will have completed its entire high school career without ever being affected by it.

1196*1196 While it’s tempting to adopt rules of law that give us the ultimate say on hotly contested political questions, we should keep in mind that we are not infallible, nor are we the repository of ultimate wisdom. Elected officials, who are much closer to ground zero than we are—and whose political power ebbs and flows with the approval of the voters—understand the realities of the situation far better than we can, no matter how many depositions and expert reports we may read in the quiet of our chambers. It therefore behooves us to approach issues such as those presented here with a healthy dose of modesty about our ability to understand the past or predict the future. It should make us chary about use of the strict scrutiny standard of review, which proclaims us the ultimate arbiters of the issue and gives those who oppose the policy in question every incentive to turn litigation, to paraphrase Clausewitz, into a continuation of politics by other means.

To resort to Chief Judge Boudin’s words one last time, “we are faced with a local experiment, pursuing plausible goals by novel means that are not squarely condemned by past Supreme Court precedent. The problems that the … plan addresses are real, and time is more likely than court hearings to tell us whether the solution is a good one….” Comfort, 418 F.3d at 29 (Boudin, C.J., concurring). I share Judge Boudin’s preference for resolving such difficult issues by trial and error in the real world, rather than by experts jousting in the courtroom. When it comes to a plan such as this—a plan that gives the American melting pot a healthy stir without benefitting or burdening any particular group—I would leave the decision to those much closer to the affected community, who have the power to reverse or modify the policy should it prove unworkable. It is on this basis that I would affirm the judgment of the district court.

BEA, Circuit Judge, with whom Circuit Judges KLEINFELD, TALLMAN and CALLAHAN join dissenting:

I respectfully dissent.

At the outset, it is important to note what this case is not about. The idea that children will gain social, civic, and perhaps educational skills by attending schools with a proportion of students of other ethnicities and races, which proportion reflects the world in which they will move, is a notion grounded in common sense. It may be generally, if not universally, accepted.[1] But that is not the issue here. The issue here is whether this idea may be imposed by government coercion, rather than societal conviction; whether students and their parents may choose, or whether their government may choose for them.[2]

In the Seattle School District (“District”), some schools are oversubscribed 1197*1197 and in higher demand than others, so the District uses a tiebreaker to assign some ninth-grade students, and not others, to those schools. The tiebreaker operates solely on the basis of the student’s race. In fact, rather than differentiating between African-American, Asian-American, Latino, Native American, or Caucasian students, the tiebreaker classifies students only as “white” or “nonwhite.”[3] The District seeks a racially balanced student body of 40% white, 60% nonwhite children; the tiebreaker excludes white or nonwhite students from an oversubscribed school if their admission will not further that preferred ratio.

Notwithstanding the majority’s fervent defense of that plan, the District is engaged in simple racial balancing, which the Equal Protection Clause forbids. The majority can arrive at the opposite conclusion only by applying a watered-down standard of review — improperly labeled “strict scrutiny” — which contains none of the attributes common to our most stringent standard of review. I respectfully disagree with the majority’s gentle endorsement of the racial tiebreaker and would instead hold the District violates the Equal Protection Clause whenever it excludes a student from a school solely on the basis of race.

I.

As an introductory note, I call attention to the majority’s frequent misuse of the terms “segregation,” “segregated schools,” and “segregated housing patterns.” See, e.g., Majority op. at 1166, 1167. As a perfectly understandable rhetorical ploy, the majority continually uses those charged terms when there has been no such segregation in the Seattle schools in any textual or legal sense.[4] Throughout the desegregation cases, the U.S. Supreme Court stated that only the remediation of de jure segregation justified the use of racial classifications. Freeman v. Pitts, 503 U.S. 467, 494, 112 S.Ct. 1430, 118 L.Ed.2d 108 (1992). “[T]he differentiating factor between de jure segregation and so-called de facto segregation … is purpose or intent to segregate.” Keyes v. School Dist. No. 1, 413 U.S. 189, 208, 93 S.Ct. 2686, 37 L.Ed.2d 548 (1973) (emphasis in original); see Swann v. Charlotte-Mecklenburg Bd. of Educ., 402 U.S. 1, 17, 91 S.Ct. 1267, 28 L.Ed.2d 554 (1971) (“`Desegregation’ means the assignment of students to public schools and within such schools without regard to their race, color, religion, or national origin, but `desegregation’ shall not mean the assignment of students to public schools in order to overcome racial imbalance.“) (emphasis added).

“Segregate” is a transitive verb. It requires an actor to do an act which effects segregation. See OXFORD ENGLISH DICTIONARY (2d ed.1989) (“segregate, v. 1. a. trans.: To separate (a person, a body or class of persons) from the general body, or from some particular class; to set apart, isolate, seclude”).[5] Instead of de jure segregation, what the majority describes is racial imbalance in the District’s schools and Seattle’s residential makeup.

1198*1198 Of course, it is much easier to argue for measures to end “segregation” than for measures to avoid “racial imbalance.” Especially is this so in view of the U.S. Supreme Court’s frequent pronouncements that “racial balancing” violates the Equal Protection Clause. See Grutter v. Bollinger, 539 U.S. 306, 330, 123 S.Ct. 2325, 156 L.Ed.2d 304 (2003) (“[O]utright racial balancing … is patently unconstitutional.”); Freeman, 503 U.S. at 494, 112 S.Ct. 1430 (“Racial balance is not to be achieved for its own sake.”); Regents of the Univ. of Calif. v. Bakke, 438 U.S. 265, 307, 98 S.Ct. 2733, 57 L.Ed.2d 750 (1978) (Powell, J.) (“If petitioner’s purpose is to assure within its student body some specified percentage of a particular group merely because of its race or ethnic origin, such a preferential purpose must be rejected not as insubstantial but as facially invalid. Preferring members of any one group for no reason other than race or ethnic origin is discrimination for its own sake. This the Constitution forbids.”).

It should be remembered by the reader of the majority opinion that one can no more “segregate” without a person actively doing the segregation than one can separate an egg without a cook.

Like Judge Boudin,[6] in his concurring opinion Judge Kozinski tries to distinguish past Supreme Court cases involving racial discrimination by focusing on the effects of the discrimination, rather than the fact of the discrimination.

This creates for them two categories different from the effects of the Seattle plan: (1) the effects of other race discrimination plans were much worse than Seattle’s and (2) the effects were visited on certain races.

But the difference reflected in these two categories are irrelevant. “[T]here is no de minimis exception to the Equal Protection Clause. Race discrimination is never a `trifle.'” Monterey Mechanical Co. v. Wilson, 125 F.3d 702, 712 (9th Cir.1997). Second, the Fourteenth Amendment protects individual rights, not the rights of certain races or groups.

Further, that a “plan does not segregate persons by race”[7] does not justify it in refusing school admission to a qualified scholar because he does not belong to a particular race. There was no segregation by race at Cal Davis medical school, when Bakke was improperly refused admission. See Bakke, 438 U.S. 265, 98 S.Ct. 2733, 57 L.Ed.2d 750.

Also, it is quite accurate to say the Seattle plan does not “involve racial quotas.”[8] The numerical quota is the percentage by which the school in question’s racial composition differs from the school district’s target.[9] Not calling it a quota, does not make it something other. “A rose by any other name … etc.”

Perhaps the Supreme Court will adopt a “rational relation” basis for review of race-based discrimination by government, based on the concurrence’s view of what is “realistic” or what are “real-world circumstances.”[10] As indicated above, however, it certainly has given no such indication.[11] 1199*1199 But if it does, one doubts that it will do so based on a “melting pot” metaphor.

Up to now, the American “melting pot” has been made up of people voluntarily coming to this country from different lands, putting aside their differences and embracing our common values. To date it has not meant people who are told whether they are white or non-white, and where to go to school based on their race.

The suggestion that local political forces should decide when to employ racial discrimination in the allocation of governmental resources is certainly nothing new in American history. Such “local option” discrimination was adopted in the Missouri Compromise of 1820, which established the Mason-Dixon line, and the Compromise of 1850. But since then, the Civil War, the post-war Amendments to the Constitution and Brown v. Bd. of Ed. of Topeka, Shawnee County, Kan., 347 U.S. 483, 74 S.Ct. 686, 98 L.Ed. 873 (1954) have made racial discrimination a matter of national concern and national governance.

As noted in the opening lines of this dissenting opinion, it certainly is rational to believe that racial balancing in schools achieves better racial socialization and, as a result, better citizens. The issue is not that, but whether what is rational can be achieved by compulsory racial discrimination by the State.

II.

I agree with the majority that the District’s use of the racial tiebreaker is a racial classification, and all racial classifications are subject to “strict scrutiny” review under the Equal Protection Clause. See Majority op. at 1173. Yet the majority conceives of strict scrutiny as some type of relaxed, deferential standard of review. I view it differently.

“No State shall … deny to any person within its jurisdiction the equal protection of the laws.” U.S. Const., amend. XIV, § 1. The right to equal protection is an individual one, and so where federal or state governments classify a person according to race—”a group classification long recognized as in most circumstances irrelevant and therefore prohibited”—we review such state action under the most “detailed judicial inquiry”—that is, under strict scrutiny. Grutter, 539 U.S. at 326, 123 S.Ct. 2325; see Miller v. Johnson, 515 U.S. 900, 911, 115 S.Ct. 2475, 132 L.Ed.2d 762 (1995) (“At the heart of the Constitution’s guarantee of equal protection lies the simple command that the Government must treat citizens as individuals, not as simply components of a racial, religious, sexual or national class.)” (internal quotation marks omitted).

The right to equal protection is held equally among all individuals. “[A]ll racial classifications reviewable under the Equal Protection Clause must be strictly scrutinized.” Adarand, 515 U.S. at 224, 115 S.Ct. 2097, 132 L.Ed.2d 158 (1995) (emphasis added). Strict scrutiny applies regardless whether the racial classifications are invidious or benign and “is not dependent on the race of those burdened or benefited by a particular classification.” Gratz, 539 U.S. at 270, 123 S.Ct. 2411; see Johnson, 125 S.Ct. at 1146 (“We have insisted on strict scrutiny in every context, even for so-called `benign’ racial classifications, such as race-conscious university admissions policies, race-based preferences in government contracts, and race-based districting intended to improve minority representation.”) (internal citations omitted). We require such a demanding inquiry “to `smoke out’ illegitimate uses of race by assuring 1200*1200 that the legislative body is pursuing a goal important enough to warrant use of a highly suspect tool.” Adarand, 515 U.S. at 226, 115 S.Ct. 2097.

The right to equal protection provides a liberty; it represents freedom from government coercion based upon racial classifications. See Miller, 515 U.S. at 904, 115 S.Ct. 2475 (the Equal Protection Clause’s “central mandate is racial neutrality in governmental decisionmaking”). Thus, under strict scrutiny, all racial classifications by the government, regardless of purported motivation, are “inherently suspect,” Adarand, 515 U.S. at 223, 115 S.Ct. 2097, and “presumptively invalid,” Shaw v. Reno, 509 U.S. 630, 643-44, 113 S.Ct. 2816, 125 L.Ed.2d 511 (1993). They are permissible only where the government proves their use is “narrowly tailored to further compelling governmental interests.” Grutter, 539 U.S. at 326, 123 S.Ct. 2325.

It follows, then, that the government carries the burden of proving that its use of racial classifications satisfies strict scrutiny. Johnson, 125 S.Ct. at 1146 n. 1 (“We put the burden on state actors to demonstrate that their race-based policies are justified.”); Gratz, 539 U.S. at 270, 123 S.Ct. 2411; W. States Paving Co., Inc. v. Wash. State Dep’t of Transp., 407 F.3d 983, 990 (9th Cir.2005) (“The burden of justifying different treatment by ethnicity … is always on the government.”) (quoting Monterey Mech. Co. v. Wilson, 125 F.3d 702, 713 (9th Cir.1997)).

Despite this formidable standard of review, the majority does not hesitate to endorse the District’s use of the racial tiebreaker. Rather than recognizing the protections of the individual against governmental racial classifications, the majority instead endorses a rigid racial governmental grouping of high school students for the purpose of attaining racial balance in the schools. For the reasons expressed below, I do not share in the majority’s confidence that such a plan is constitutionally permissible.

III.

I consider first whether the District has asserted a “compelling governmental interest,” the first element of the strict scrutiny test. The District contends it has a valid compelling governmental interest in using racial balancing to achieve “the educational and social benefits of racial … diversity” within its high schools and avoid “racially concentrated” schools. The District argues its interest will enhance student discussion of racial issues in high school and will foster cross-racial socialization and understanding, both in school and later in the students’ lives.

The U.S. Supreme Court has “declined to define compelling interest or to tell [the lower courts] how to apply that term.” Hunter v. Regents of the Univ. of Calif., 190 F.3d 1061, 1070 n. 9 (9th Cir.1999) (Beezer, J., dissenting); Mark R. Killenbeck, Pushing Things Up to Their First Principles: Reflections on the Values of Affirmative Action, 87 Calif. L.Rev. 1299, 1349 (1999) (the definition of a compelling interest “is admittedly imprecise. The Supreme Court has never offered a workable definition of the term … and is unlikely ever to do so, preferring to approach matters on a case-by-case basis”).

The majority is correct in noting the U.S. Supreme Court has never endorsed “racial balancing” as a “compelling interest.” Indeed, throughout the history of strict scrutiny, the Supreme Court has rejected as invalid all such asserted compelling interests, save for two exceptions. With respect, the majority errs in creating a third.

A.

The Court has endorsed two race-based compelling governmental interests in the public education context. First, the Court 1201*1201 has allowed racial classifications to remedy past racial imbalances in schools resulting from past de jure segregation. Freeman, 503 U.S. at 494, 112 S.Ct. 1430. Second, the Court has allowed undergraduate and graduate universities to consider race as part of an overall, flexible assessment of an individual’s characteristics to attain student body diversity. Grutter, 539 U.S. at 328, 123 S.Ct. 2325; Gratz, 539 U.S. at 268-69, 123 S.Ct. 2411.

Besides those two valid compelling interests, the Court has struck down every other asserted race-based compelling interest that has come before it. See Shaw v. Hunt, 517 U.S. 899, 909-12, 116 S.Ct. 1894, 135 L.Ed.2d 207 (1996) (rejecting racial classifications to “alleviate the effects of societal discrimination” in the absence of findings of past discrimination, and to promote minority representation in Congress); Richmond v. J.A. Croson Co., 488 U.S. 469, 511, 109 S.Ct. 706, 102 L.Ed.2d 854 (1989) (plurality) (rejecting racial classifications in the awarding of public construction contracts in the absence of findings of past discrimination); Wygant v. Jackson Bd. of Educ., 476 U.S. 267, 274-76, 106 S.Ct. 1842, 90 L.Ed.2d 260 (1986) (rejecting racial classifications in a school district’s teacher layoff policy when offered as a means of providing minority role models for its minority students and as a means of alleviating past societal discrimination); Bakke, 438 U.S. at 310-11, 98 S.Ct. 2733 (Powell, J.) (rejecting the application of race-conscious measures to improve “the delivery of health-care services to communities currently underserved”). A crucial guiding point here — and one elided entirely by the majority—is the Court’s consistent reiteration that “outright racial balancing … is patently unconstitutional.” See, e.g., Grutter, 539 U.S. at 330, 123 S.Ct. 2325.

Thus, we face a landscape littered with rejected asserted “compelling interests” requiring race-based determinations, but with two exceptions still standing. The first exception is inapplicable here because the Seattle schools have never been de jure segregated. See Freeman, 503 U.S. at 494, 112 S.Ct. 1430.

The second exception is also inapplicable, albeit not so directly acknowledged. At oral argument, the District conceded that it is not asserting the Grutter “diversity” interest; the majority recognizes this in stating the District’s asserted interest is “significantly different” in some ways from the interest asserted in Grutter. Majority op. at 1176. Nonetheless, the majority concludes those differences are inconsequential because of the different “context”[12] between high schools and universities, 1202*1202 and the District’s asserted interest is a compelling governmental interest in its own right.

Not so. The very differences between the Grutter “diversity” interest and the District’s asserted interest illustrate why the latter violates the Equal Protection Clause as opposed to the former. The Grutter “diversity” interest focuses upon the individual, of which race plays a part, but not the whole. The District’s asserted interest, however, focuses only upon race, running afoul of equal protection’s focus upon the individual.

B.

In Grutter and Gratz, the Court made clear that the valid compelling interest in “diversity” does not translate into a valid compelling interest in “racial diversity.” The “diversity” interest

is not an interest in simple ethnic diversity, in which a specified percentage of the student body is in effect guaranteed to be members of selected ethnic groups…. Rather, the diversity that furthers a compelling state interest encompasses a far broader array of qualifications and characteristics of which racial or ethnic origin is but a single though important element.

Grutter, 539 U.S. at 324-25, 123 S.Ct. 2325 (emphasis added); see Gratz, 539 U.S. at 272-73, 123 S.Ct. 2411 (“[T]he critical criteria [in a permissible race-conscious admissions program] are often individual qualities or experiences not dependent upon race but sometimes associated with it.”).

The Grutter “diversity” interest focuses upon the individual, which can include the applicant’s race, but also includes other factors, such as the applicant’s family background, her parent’s educational history, whether she is fluent in other languages, whether she has overcome adversity or hardship, or whether she has unique athletic or artistic talents. See 539 U.S. at 338, 123 S.Ct. 2325. Such a focus is consistent with the Equal Protection Clause, which protects the individual, not groups.

But here, the District’s operation of the racial tiebreaker does not consider the applicant as an individual. To the contrary, the racial tiebreaker considers only whether the student is white or nonwhite. While the Grutter “diversity” interest pursues genuine diversity in the student body (of which race is only a single “plus” factor), the District pursues an interest which considers only racial diversity, i.e., a predefined grouping of races in the District’s schools.[13] Such an interest is not a valid 1203*1203 compelling interest; it is simple racial balancing, forbidden by the Equal Protection Clause. See id. at 330, 123 S.Ct. 2325 (stating a government institution’s interest “to assure within its student body some specified percentage of a particular group merely because of its race … would amount to outright racial balancing, which is patently unconstitutional”).

Grutter emphasized the dangers resulting from lack of an individualized consideration of each applicant. Observing that the Michigan Law School sought an unquantified “critical mass” of minority students to avoid only token representation, rather than some defined balance, id. at 330, 123 S.Ct. 2325, the Court reasoned the law school’s individualized focus on students forming that “critical mass” would avoid perpetuating the stereotype that all “minority students always … express some characteristic minority viewpoint on any issue,” id. at 333, 123 S.Ct. 2325.

But here, the District’s concept of racial diversity is a predetermined, defined ratio of white and nonwhite children. The racial tiebreaker works to exclude white students from schools that have a 50-55% white student body (depending on the tiebreaker trigger used in a particular year), and works to exclude nonwhite students from schools with a 70-75% nonwhite student body (depending on the tiebreaker trigger used). Thus, the District’s concept of racial diversity does not permit a school with a student body that is too white, or a school with a student body that is too nonwhite.

The District argues its concept of racial diversity is necessary to foster classroom discussion and cross-racial socialization. That argument, however, is based on the stereotype that all white children express traditional white viewpoints and exhibit traditional white mannerisms; all nonwhite children express opposite nonwhite viewpoints and exhibit nonwhite mannerisms, and thereby white and nonwhite children will better understand each other. Yet there is nothing in the racial tiebreaker to ensure such viewpoints and mannerisms are represented within the preferred student body ratio. As noted in Grutter, the only way to achieve diverse viewpoints and mannerisms is to look at the individual student. White children have different viewpoints and backgrounds than other white children; the same goes for nonwhite children; and some white children have the same viewpoints and backgrounds as some nonwhite children. The assumption that there is a difference between individuals just because there is a difference in their skin color is a stereotype in itself, nothing more.[14]

The District also claims it must use the racial tiebreaker to avoid racially imbalanced schools, which may result in schools with large white or nonwhite student bodies and in which the supposed benefits from the District’s concept of racial diversity will not occur. This theory, however, presents another racial stereotype, which assumes there is something wrong with a school that has a heavy nonwhite student body population, or something better 1204*1204 about a school that has a heavy white student body population. See Missouri v. Jenkins, 515 U.S. 70, 122, 115 S.Ct. 2038, 132 L.Ed.2d 63 (1995) (Thomas, J., concurring) (“After all, if separation itself is a harm, and if integration therefore is the only way that blacks can receive a proper education, then there must be something inferior about blacks. Under this theory, segregation injures blacks because blacks, when left on their own, cannot achieve. To my way of thinking, that conclusion is the result of a jurisprudence based upon a theory of black inferiority.”).

Besides the District’s reliance on racial stereotypes, there is good reason categorically to forbid racial balancing. The process of classifying children in groups of color, rather than viewing them as individuals, encourages “notions of racial inferiority” in both white and nonwhite children and incites racial hostility. See Grutter, 539 U.S. at 328, 123 S.Ct. 2325. Indeed, those risks are particularly great here because of the blunt nature of the racial tiebreaker. The District’s racial grouping of students, either as white or nonwhite, assumes that each minority student is the same, regardless whether he is African-American, Asian-American, Latino, or Native American; the only difference noted by the District is that the minority student is not white.[15] The District thus “conceives of racial diversity in simplistic terms as a dichotomy between white and nonwhite, as if to say all nonwhites are interchangeable.” Parents Involved in Cmty. Schs. v. Seattle Sch. Dist., No. 1, 149 Wash.2d 660, 72 P.3d 151, 169 n. 5 (2003) (Sanders, J., dissenting). I join my colleague on the Washington Supreme Court in observing that “[a]s a theory of racial politics, this view is patently offensive and as a policy to promote racially diverse schools, wholly inadequate.” Id.

Unlike a voluntary decision by parents to expose their children to individuals of different races or background, the District classifies each student by skin color and excludes certain students from particular schools—solely on the basis of race—to ensure those schools remain racially balanced. Even if well-intentioned, the District’s use of racial classifications in such a stark and compulsory fashion risks perpetuating the same racial divisions which have plagued this country since its founding:

Race is perhaps the worst imaginable category around which to organize group competition and social relations more generally. At the risk of belaboring the obvious, racial categories in law have played an utterly pernicious and destructive role throughout human history. This incontrovertible fact should arouse wonder … at the hubris of those who imagine that we can distinguish clearly enough between invidious and benign race discrimination to engrave this distinction into our constitutional order. Vast human experience mocks this comforting illusion, as does the fact that most Americans, including many minorities, think racial preferences are invidious, not benign. Whether benignly intended or not, using the category of race—which affirmative action proponents oddly depict as socially constructed and primordial and immutable—to 1205*1205 distribute advantage and disadvantage tends to ossify the fluid, forward-looking political identities that a robust democratic spirit inspires and requires.

Peter H. Schuck, Affirmative Action: Past, Present, and Future, 20 Yale L. & Pol’y Rev. 1, 92-93 (2002).

We should not minimize these shadows that are cast over the supposed benefits of the District’s asserted interest. The District’s stark racial classifications not only offend intrinsic notions of individuality, they even suggest principles opposite to what the District claims to seek. Although the District contends it uses the racial tiebreaker for good, i.e., to foster cross-racial socialization and understanding, the District’s concept of racial diversity also suggests other principles which many may find objectionable, especially when taught to children:

While a public law preference does express a certain kind of compassion for and commitment to the preferred groups, other signals dominate its message — among them, that American society thinks it just to group people by race and ethnicity, to treat those groups monolithically, and to allocate precious resources and opportunities accordingly; that it holds equal treatment and individual merit as secondary, dispensable ideals; that the preferred groups cannot succeed without special public favors; that such favors do not stigmatize them in the minds of fair-minded others; that those who oppose preferences thereby oppose the aspirations of the preferred groups; and that society can assuage old injustices by creating new ones. When public law says such things, it speaks falsely, holds out vain promises, and brings itself into disrepute.

Id. at 87-88.

The District’s asserted interest may be supported by noble goals. But the stereotypes on which it is based, and the risks that it presents, make that interest far from compelling.

C.

The sociological evidence presented by the District, relied upon strongly by the majority, does not change my view. The majority discusses much of the evidence that supports the District’s position that racially balanced schools foster cross-racial socialization and understanding in school and later in the students’ lives. Majority op. at 1174-1175. Yet the majority puts aside the other evidence suggesting there is no definitive agreement as to the beneficial effects of racial balance in K-12 schools, that the benefits attributed to racially balanced schools are often weak, and that any benefits do not always have a direct correlation to racial balance. Yet again, a private citizen is free to accept one body of opinion and reject another in deciding to send his child to a particular school. Is the state similarly privileged when required to determine that its claimed goal is a “compelling interest”? One would think that to be “compelling” there would be no room for doubt of the need for the measure. That is certainly not the case here.

For example, a source provided by the District states that “family background has a significantly stronger effect on student achievement than any other single school factor or constellation of school factors, including school racial and ethnic composition.” [SER 182.] Another source presented by the District states that court-ordered desegregation (i.e., a court-ordered breakup of a de jure segregated student body) resulted in only minimal benefits:

[R]esearch suggests that desegregation has had some positive effect on the reading skills of African American youngsters. The effect is not large, nor does it occur in all situations, but a modest measurable effect does seem apparent. 1206*1206 Such is not the case with mathematical skills, which seem generally unaffected by desegregation. Second, there is some evidence that desegregation may help to break what can be thought of as a generational cycle of segregation and racial isolation. Although research on this topic is scant and often marred by unavoidable flaws, evidence has begun to accumulate that desegregation may favorably influence such adult outcomes as college graduation, income, and employment patterns. The measured effects are often weak ….

[SER 205, 207-208.]

That source concludes that “[t]he evidence regarding the impact of desegregation on intergroup relations is generally held to be inconclusive and inconsistent.” [SER 208.]. See Grutter, 539 U.S. at 364-65, 123 S.Ct. 2325 (Thomas, J., dissenting) (collecting studies suggesting black students perform at higher levels of achievement at historically black colleges); David I. Levine, Public School Assignment Methods after Grutter and Gratz: The View from San Francisco, 30 Hastings Const. L.Q. 511, 536 (2003) (noting that a high school’s focus on racial balance misses the “key element” in the context of education, i.e., that “the life chances of students are improved only with economic integration”).[16]

The serious risks presented by racial classifications counteract the marginal benefits provided by racial balancing. Courts have long recognized racial classifications promote “notions of racial inferiority and lead to a politics of racial hostility.” See Grutter, 539 U.S. at 328, 123 S.Ct. 2325; Michael Perry, Modern Equal Protection, 79 Colum. L.Rev. 1023, 1048 (1979) (“Affirmative action “`inevitably foments racial resentment and thereby strains the effort to gain wider acceptance for the principle of moral equality of the races.'” “). Other studies suggest that where racial classifications are a means of achieving racial balance, academic achievement by minorities is hindered, and racial tensions are riled:

In a culture that ardently affirms the principles of individual freedom, merit, and equality of opportunity, [the] demoralization and anger [precipitated by being victim to government-imposed racial classifications] must be counted as a very large social cost. It is no less a 1207*1207 cost because it is borne by whites, and often less privileged whites at that. If these principles make it unfair to impose this cost, the fact that the unfairness is spread across a large group of people may not make it any more palatable. In fact, diffusing the unfairness in this way will simply increase the number of people who feel themselves aggrieved.

Schuck, supra, at 69.

But despite the inconsistencies in the sociological evidence and the vivid risks of the District’s asserted interest, the majority implicitly defers to the District’s position. Grutter took a similar approach, emphasizing that its endorsement of the “diversity” interest relied in large part upon deference to the educational judgment of the Michigan Law School. 539 U.S. at 330, 123 S.Ct. 2325.

Yet perhaps to steal a line from the majority, the “context” here is different. We are not faced with a university’s “academic freedom,” which arises from “a constitutional dimension, grounded in the First Amendment, of educational autonomy,” and which includes the freedom to select its student body. Id. We instead consider a public high school’s admissions plan which admits or excludes students from particular schools solely on the basis of their race. For several reasons, we should not defer to such a plan.

First, other than for race-conscious university admissions based on holistic diversity, deference to a government actor is inconsistent with strict scrutiny. See Johnson, 125 S.Ct. at 1146 n. 1 (stating generally that “deference [by the courts in applying strict scrutiny] is fundamentally at odds with our equal protection jurisprudence”); id. at 1150 (stating the Supreme Court has “refused to defer to state officials’ judgments on race … where those officials traditionally exercise substantial discretion.”). In Grutter, the Court deferred to the Michigan Law School’s “diversity” interest because of the law school’s “academic freedom” — grounded in the First Amendment and including the law school’s freedom to select its own student body — and the law school’s asserted need for diversity to achieve a “robust exchange of ideas” within its classrooms, a vital part of the law school’s mission. 539 U.S. at 330, 123 S.Ct. 2325.

None of those same issues are implicated here. The “academic freedom” of a university allows it “to determine for itself on academic grounds who may teach, what may be taught, how it shall be taught, and who may be admitted to study.” Bakke, 438 U.S. at 312, 98 S.Ct. 2733 (Powell, J.). High schools do not have such similar freedoms. They cannot determine who may teach, at least when that determination is based upon racial grounds. See Wygant, 476 U.S. at 274-76, 106 S.Ct. 1842. They also cannot determine who may be admitted to study; when the government chooses to provide public education in secondary schools, it “must be made available to all on equal terms.” See Plyler v. Doe, 457 U.S. 202, 221-23, 102 S.Ct. 2382, 72 L.Ed.2d 786 (1982). Further, there is no comparable line of U.S. Supreme Court cases affording high schools the special “[A]cademic freedom[s]” granted to universities by the First Amendment. See United States v. Fordice, 505 U.S. 717, 728-29, 112 S.Ct. 2727, 120 L.Ed.2d 575 (1992) (“a state university system is quite different in very relevant respects from primary and secondary schools.”); Jay P. Lechner, Learning From Experience: Why Racial Diversity Cannot Be a Legally Compelling Interest in Elementary and Secondary Education, 32 SW. U.L.Rev. 201, 215 (2003) (stating the Supreme Court “has been less deferential to the discretion of elementary or secondary school officials in Equal Protection cases, in part because the Court has viewed school desegregation as serving social rather than educational goals. The Court has acknowledged that 1208*1208 even the most important, delicate, and highly discretionary functions of state educators are subject to the limits of the Bill of Rights and subordinate to the Constitutional freedoms of the individual. Moreover, the educational benefits from diversity, if any, are much greater at the higher educational level because such benefits are greatly magnified by the learning that takes place outside the classroom — in dormitories, social settings, and extracurricular activities — as students must learn to live and work with persons of other races and ethnic backgrounds.”) (internal quotation marks omitted).

Moreover, there is a crucial difference between the “robust exchange of ideas” theory referenced in Grutter and the District’s claim that its interest “brings different viewpoints and experiences to classroom discussions and thereby enhances the educational process.” [ER 237.] The District applies the racial tiebreaker only to entering ninth-grade students. [ER 253, 308.] It is self-evident that classroom discussion plays a significantly more vital role in universities with their typical dialectic or Socratic teaching method, than in ninth-grade high school courses with their typical didactic or rote teaching method.

Last, the District’s claim that its asserted interest helps to foster cross-racial socialization and understanding later in the students’ lives is a sociological judgment outside the expertise of the District’s educators. Those external benefits are diffuse, manifest long after students leave the classroom, and cannot be measured with skills possessed uniquely by educators. Unlike Grutter, which deferred to the Law School on the basis that diversity in the classroom was vital to its educational mission during the three-year law school curriculum, here, the District’s asserted interest depends upon benefits only loosely linked to the District’s educational mission and to take effect years after its schooling of the children, or entirely outside the expertise of its educators. Here, high school administrators and teachers are predicting what sociologists will find years later.

Strict scrutiny cannot remain strict if we defer to judgments not even within the particular expertise or observation of the party being scrutinized. Hence, deference is not due to the District regarding the benefits the District contends are attributable to its claimed interest.[17]

1209*1209 In the absence of deference to the District’s sociological evidence, the faults of the District’s asserted interest come into sharper focus. It has none of the saving graces present in the Grutter holistic diversity interest. It perpetuates racial stereotypes and risks fomenting racial hostility. Last, the District enforces the interest through government compulsion in the starkest black and white terms, espousing the principle that race trumps the individual.

The sociological evidence presented by the District suggests that some benefits will accrue from racial balancing. To me, evidence of some benefits does not satisfy the District’s burden of proving a compelling governmental interest, especially in light of the Supreme Court’s frequent pronouncements that racial balancing itself is unconstitutional. Thus, viewed under the lens of strict scrutiny, and without the deference invoked in Grutter, the District’s interest is simply not a compelling governmental interest. Hence, I would hold that the District’s operation of the racial tiebreaker is an impermissible racial classification and violates the Equal Protection Clause.

IV.

Even if the District’s asserted interest were a compelling governmental interest, the means used by the District must still be narrowly tailored to serve that interest. See Grutter, 539 U.S. at 333, 123 S.Ct. 2325. For argument’s sake, I here assume, without conceding, the District has asserted a valid compelling governmental interest in using racial balancing to achieve “the educational and social benefits of racial … diversity” within its high schools and to avoid “racially concentrated” schools. Yet even under that assumption, the District’s use of the racial tiebreaker is not narrowly tailored to serve that interest.

The majority notes that Grutter set forth “five hallmarks of a narrowly tailored affirmative action plan: (1) individualized consideration of applicants; (2) the absence of quotas; (3) serious, good-faith consideration of race-neutral alternatives to the affirmative action program; (4) that no member of any racial group was unduly harmed; and (5) that the program had a sunset provision or some other end point.” Majority op. at 1180. I agree with that general formulation. Yet the majority’s application of those factors again evinces an improper deference to the District; such deference is ill suited for the searching inquiry needed under the narrow-tailoring prong of strict scrutiny. See Johnson, 125 S.Ct. at 1146 n. 1. I consider below whether the District’s use of the racial tiebreaker is narrowly tailored to its asserted interest, and conclude that racial tiebreaker is not narrowly tailored.

1210*1210 A.

The first narrow-tailoring factor requires the District to engage in an individualized consideration of each applicant’s characteristics and qualifications. See Grutter, 539 U.S. at 337, 123 S.Ct. 2325. The importance of this factor is self-evident; individualized consideration serves the primary purpose of the Equal Protection Clause, which protects the individual from group classifications, especially those by race. See id. at 326, 123 S.Ct. 2325.

Yet the majority concludes that individualized consideration of each applicant is irrelevant here “because of the contextual differences between institutions of higher learning and public high schools.” Majority op. at 1180. I could not disagree more.[18] By removing consideration of the individual from the narrow tailoring analysis, the majority threatens to read the Equal Protection Clause out of the Constitution. It is the very nature of equal protection to require individualized consideration when the government uses racial classifications: “the Fourteenth Amendment “protects persons, not groups.“” Grutter, 539 U.S. at 326, 123 S.Ct. 2325 (quoting Adarand, 515 U.S. at 227, 115 S.Ct. 2097) (emphasis in original). Grutter emphasized the importance of the individualized consideration of each applicant: in the context of a race-conscious university admissions program, such consideration

must remain flexible enough to ensure that each applicant is evaluated as an individual and not in a way that makes an applicant’s race or ethnicity the defining feature of his or her application. The importance of this individualized consideration in the context of a race-conscious admissions program is paramount.

Id. at 337, 123 S.Ct. 2325 (emphasis added). The differences between university and secondary education do not justify denial of individualized equal protection of the law to secondary school students.

Individualized consideration of an applicant does not require an admissions program to be oblivious to race; the program may consider race, but in doing so, it must remain “flexible enough to consider all pertinent elements of diversity in light of the particular qualifications of each applicant, and to place them on the same footing for consideration, although not necessarily according them the same weight.” Id. at 334, 123 S.Ct. 2325. There can be “no policy, either de jure or de facto, of automatic acceptance or rejection based on any single `soft’ variable … [such as the awarding of] mechanical, predetermined diversity `bonuses’ based on race or ethnicity.” Id. at 337, 123 S.Ct. 2325.

Here, the racial tiebreaker works to admit or exclude high school students from certain oversubscribed schools solely on the basis of their skin color. No other consideration affects the operation of the racial tiebreaker; when it operates, it operates to admit or exclude either a white or nonwhite student, depending upon how the admission will affect the preferred balance at the oversubscribed school. Such a program is precisely what Grutter warned against, and what Gratz held unconstitutional: a mechanical, predetermined policy “of automatic acceptance or rejection based on a[] single `soft’ variable,” that being the student’s skin color. See id.

The racial tiebreaker’s overbroad classification of students as “white” or “nonwhite” also runs counter to the required individualized consideration of each applicant. The District does not even consider the student’s actual race. Instead, the 1211*1211 District presumably places all Caucasian students into the “white” category, and then places all African-American, Latino, Asian-American, Pacific Islander and Native Americans into the “nonwhite” category. This puts aside the categorization of any individuals whose skin color does not correlate directly with the classifications. Although parents and students may identify their particular group on the registration materials, if they do not, the District will make the racial identification itself through visual inspection of the parent or student. Thus, a fair-skinned minority may wind up in the “white” category, or a darker-skinned Caucasian may wind up in the “nonwhite” category.

Courts have often recognized that the inclusion of all minorities within a “nonwhite” classification suggests the operation of a racial classification is not narrowly tailored. See Wygant, 476 U.S. at 284 n. 13, 106 S.Ct. 1842 (noting the “definition of minority to include blacks, Orientals, American Indians, and persons of Spanish descent further illustrates the undifferentiated nature of the plan”); Monterey Mech. Co., 125 F.3d at 714 (noting the inclusion of all minority races within a broad “minority” category serves as a “red flag[] signaling that the statute is not, as the Equal Protection Clause requires, narrowly tailored”). At the very least, a narrowly tailored program would require an individualized focus which would separate the student according to his or her correct race, rather than as a process of simple pigmental matching.

The majority concludes, however, that individualized consideration of each applicant is unnecessary because the District does not exclude any student from a public education by operation of the racial tiebreaker. The majority reasons that because all students are entitled to a public education in one of the District’s schools, there is no competition in the District for admission to any of those schools, and thus no racial stigma could attach when a student is excluded from admission to one of the schools on the basis of his race. Majority op. at 1181-1182.

Yet the majority offers no explanation why, in the 2000-01 school year, 82% of the students selected one of the oversubscribed schools (i.e., the schools subject to the racial tiebreaker) as their first choice, while only 18% picked one of the undersubscribed schools as their first choice. Majority op. at 10-11. Clearly, the students’ and their parents’ “market” appraise some of the schools as providing a better education than the others. Even the District’s superintendent confirmed that the students’ parents considered some of the schools to be of higher quality. [ER 534.]

It is common sense that some public schools are better than others. Parents often move into areas offering better school districts, and ubiquitous research guides compare the quality of public schools according to standardized test scores, program offerings, and the sort. It may be that soothing, if self-interested, bureaucratic voices sing a lullaby of equal educational quality in the District’s schools. But the facts show that parents and children have voted with their feet in choosing some schools rather than others. The verdict of that “market” makes a hash out of such assurances by the District.

Thus, the District’s operation of the racial tiebreaker in reality does limit access to a governmental benefit among certain students. The District insulates applicants belonging to certain racial groups from competition for admission to those schools perceived to be of higher quality. A narrowly tailored race-conscious admissions program “cannot insulate each category of applicants with certain desired [racial] qualifications from competition with all other applicants.” Grutter, 539 U.S. at 1212*1212 334, 123 S.Ct. 2325. The racial tiebreaker fails that test.

Yet the majority insist that because the District seeks to avoid racially concentrated schools, “the District’s tiebreaker must necessarily focus on the race of its students.” Majority op. at 1183. Again, the majority misses the crucial protection provided by the Equal Protection Clause. The District’s narrow-tailoring obligation does not prohibit it from considering race; it just cannot consider only race. The constitutional guarantee of equal protection requires the District to focus upon the individual’s whole make up, rather than just a group’s skin color; this protects each student’s right to equal protection under the law. See Grutter, 539 U.S. at 326, 123 S.Ct. 2325.

The counter-argument, of course, is that administrative inconveniences would prohibit the District from examining each student’s file for individual characteristics, of which race may be a part. To the contrary, the record shows such an effort is certainly feasible.

First, thirteen- or fourteen-year-old students[19] are not so young that they have not yet developed unique traits to set themselves apart from other students and add greater diversity to the student body. The students’s race is a factor in assessing the student as an individual, but the student may also speak English as a second language, come from a different socioeconomic stratum than other students, have overcome adversity, be a talented baseball player, musician, or have participated in community service.

Second, as noted by the majority, in the 2000-01 school year, approximately 3,000 students entered the District’s high schools as ninth graders. Ten percent of those students were subject to the racial tiebreaker. Majority op. at 1170. Thus, under an individualized approach, the District would have had to examine only three hundred applications to determine who to admit to the oversubscribed schools. Instead, the District grouped those three hundred students into white and nonwhite categories and allowed a computer to select their assignment based solely upon their race.[20]

Thus, rather than providing an individualized consideration of applicants, the District is engaged in a “de jure [policy] of automatic acceptance or rejection based on a[] single `soft’ variable.” See Grutter, 539 U.S. at 337, 123 S.Ct. 2325. Such inflexibility shows the racial tiebreaker is not “narrowly tailored to any goal, except perhaps outright racial balancing.” See Croson, 488 U.S. at 507, 109 S.Ct. 706 (plurality).

B.

The second narrow-tailoring factor prohibits the use of quotas based upon race. Grutter, 539 U.S. at 334, 123 S.Ct. 2325. A quota is defined as “a program in which a certain fixed number or proportion of opportunities are reserved exclusively for certain minority groups. Quotas impose a fixed number or percentage which must be attained, or which cannot be exceeded.” Id. at 335, 123 S.Ct. 2325 (internal quotation marks and citations omitted).

1213*1213 Here, when a District school is oversubscribed and “integration positive” — i.e., the white or nonwhite student body of the school deviates by plus or minus 10% or 15% (depending on the school year)[21] of the preferred 40% white/60% nonwhite ratio — the District uses the racial tiebreaker to admit students whose presence will move the overall student body closer to the preferred ratio. Using the 2000-2001 school year as an example, the District would employ the racial tiebreaker to exclude white students and admit nonwhite students where the white student body population exceeded 50%. The District would also employ the racial tiebreaker to exclude nonwhite students and admit white students where the nonwhite student body population in a particular school exceeded 70%.

By its nature, the tiebreaker aims for a rigid, predetermined ratio of white and nonwhite students, and thus operates to reach “a fixed number or percentage.” (emphasis supplied). Gratz specifically rejected such a plan as not narrowly tailored. See 539 U.S. at 270, 123 S.Ct. 2411 (“[T]he University’s policy, which automatically distributes [20%] … of the points needed to guarantee admission, to every single `underrepresented minority’ applicant solely because of race, is not narrowly tailored….”); id. at 271-72, 123 S.Ct. 2411 (“The only consideration that accompanies this distribution of points is a factual review of an application to determine whether an individual is a member of one of these minority groups.”).

Yet the majority argues no quota exists here because the racial tiebreaker “does not set aside a fixed number of slots for nonwhite or white students,” nor is the 10 or 15% variance always satisfied (generally because there are insufficient numbers of white or nonwhite students needed to balance the school). Majority op. at 1185.[22] With respect, the majority misses the point. A quota does not become less of a quota because there are an insufficient number of whites or nonwhites to fill the preselected spots. The District created a quota when it established the predetermined, preferred ratio of white and nonwhite students. In Bakke, the medical school argued that it did not operate a quota in its admissions system because it did not always fill the preselected seats; thus, its admissions system only had a “goal.” Justice Powell rejected that argument, stating that regardless of whether the preselected seats were a “quota” or a “goal,” such a

semantic distinction is beside the point: The special admissions program is undeniably a classification based on race and ethnic background. To the extent that there existed a pool of at least minimally qualified minority applicants to fill the 16 special admissions seats, white applicants could compete only for 84 seats in the entering class, rather than the 100 open to minority applicants. Whether this limitation is described as a quota or a goal, it is a line drawn on the basis of race and ethnic status.

1214*1214 Bakke, 438 U.S. at 289, 98 S.Ct. 2733 (Powell, J.).

The majority makes a further attempt to avoid Grutter’s admonition against quotas by attempting to classify the District’s predetermined ratio as a “critical mass.” The District’s preferred ratio could not be further from the definition of a “critical mass.” Grutter recognized that a “critical mass” had no quantified definition; instead, it was generally referred to as “meaningful numbers” or “meaningful representation” of minorities. 539 U.S. at 318, 123 S.Ct. 2325. The Court expressly stated that a “critical mass” was not a means “simply to assure within its student body some specified percentage of a particular group merely because of its race or ethnic origin.” Id. at 329, 123 S.Ct. 2325 (internal quotation marks omitted).

But unlike the unquantified “critical mass” from Grutter, the District’s preferred ratio is firmly set at 40% white, 60% nonwhite. When the 15% deviation trigger is used with the racial tiebreaker, the District seeks to enroll between 75% and 45% nonwhite students and 25% to 55% of white students. The District’s admissions plan clearly seeks to assure a specified percentage of white or nonwhite students in its schools; rather than seeking a “critical mass,” the District instead seeks racial balance. Thus, the District’s operation of the racial tiebreaker fails this factor as well.

C.

The third narrow-tailoring factor requires the District to have engaged in a “serious, good-faith consideration of workable race-neutral alternatives.” See id. at 339, 123 S.Ct. 2325. The majority concludes the District made such an effort. Majority op. at 1188. For several reasons, I disagree.

First, the District’s superintendent flatly admitted the District did not engage in a serious, good-faith consideration of race-neutral alternatives. When asked whether the District “g[a]ve any serious consideration to the adoption of a plan for the assignment of high school students that did not use racial balancing as a factor or goal,” the District’s superintendent stated: “I think the general answer to that question is no … I don’t remember a significant body of work being done. I mean it’s possible informally ideas were floated here or there, but I don’t remember any significant staff work being done.” [ER 521.]

The record supports this concession. The District never asked its demographer to conduct any analysis regarding the effect of using a race-neutral lottery. [ER 483.] The District also never asked its demographer to conduct any analysis regarding a diversity program with non-racial indicia such as a student’s eligibility for free lunch or the students’s socioeconomic background.[23] [ER 481-82.]

1215*1215 Also, in 2000, the Urban League of Metropolitan Seattle presented a high school assignment plan to the District. The plan proposed that each neighborhood region in Seattle would have a designated high school. Students would still be able to apply to any high school in Seattle, but when oversubscription occurred, students living in the designated “reference area” would first be assigned to their regional high school ahead of those who did not. To avoid racial concentration in the schools, the plan proposed “merit-based academic, avocational and vocational magnet programs.” These programs “will help each school address racial diversity issues by encouraging students to travel outside of their communities to participate in a specific magnet program.”[24]

Despite the majority’s assertion, the record suggests the District did not seriously consider this plan. The District did not ask its demographer to conduct any analysis as to the effect or workability of the plan [ER 504]; one District board member 1216*1216 stated the District “didn’t deal with” the plan [ER 514]; another board member stated the District didn’t consider the plan [ER 643]; and last, another board member stated he refused to read the proposal because he would “rather play with my bass lunker fishing game.” [ER 573.]

Of course, “[n]arrow tailoring does not require exhaustion of every conceivable race-neutral alternative,” Grutter, 539 U.S. at 339, 123 S.Ct. 2325, but it does require an earnest, good-faith consideration of the alternatives. Here, the District made no such attempt, and thus the District’s use of the racial tiebreaker fails this narrow-tailoring factor.[25]

D.

The fourth narrow-tailoring factor requires that the District’s use of the racial tiebreaker “must not unduly burden individuals who are not members of the favored racial and ethnic groups.” See Grutter, 539 U.S. at 341, 123 S.Ct. 2325. The majority adjusts this test slightly to consider “any racial group,” rather than just members of the disfavored group. Majority op. at 1180. Because the racial tiebreaker disadvantages both white and nonwhite children, I agree that the modification is valid. But unlike the majority, I conclude the District’s operation of the racial tiebreaker fails this factor as well.

The racial tiebreaker unduly burdens thirteen- and fourteen-year-old school children by (1) depriving them of their choice of school, and (2) imposing on them tedious cross-town commutes, solely upon the basis of their race.

First, as recognized above, the “good” schools in Seattle are a limited government benefit. Thus, the racial tiebreaker burdens white or nonwhite students, and often deprives them of the opportunity to enroll at what are considered the better schools, solely on the basis of race.

Second, the children of plaintiff members Jill Kurfurst and Winnie Bachwitz were denied admission to Ballard High School based on their race and instead were forced to attend Ingraham, a school on the other side of Seattle from their home. To attend that school, the two white students faced a daily multi-bus round-trip commute of over four hours. The parents instead enrolled their children in private schools. Those children were not only deprived of the school of their choice, they were effectively denied a public education (surely at much lower cost than private tuition), based on nothing but their race.

A look at the operation of the tiebreaker provides further evidence of the injury the District inflicts on both white and nonwhite students. As noted by the majority, 1217*1217 in the 2000-01 school year, 89 more white students were assigned to Franklin than would have occurred absent the tiebreaker; 107 more nonwhite students were admitted to Ballard; 82 more nonwhite students were admitted to Roosevelt; and Twenty-seven more nonwhite students were admitted to Nathan Hale. Majority op. at 1170. To place the racial tiebreaker into proper perspective, in the 2000-01 school year, 89 nonwhite, minority students were denied admission to Franklin, and had to attend what to them was a less desirable school, solely because of their skin color. One hundred-seven white students were denied admission to Ballard, and had to attend what to them was a less desirable school, solely because of their skin color. Eighty-two white students were denied admission to Roosevelt, and had to attend what to them was a less desirable school, solely because of their skin color. Twenty-seven white students were denied admission to Nathan Hale, and had to attend what to them was a less desirable school, solely because of their skin color.

Yet the majority discounts the burdens imposed by the racial tiebreaker, concluding that (1) the “minimal burden” of the tiebreaker is shared equally among white and nonwhite students; (2) no student is entitled to attend any specific school in any event; and (3) the tiebreaker does not uniformly benefit one race over the other because the tiebreaker operates against both whites and nonwhites. Majority op. at 1191-1192. Regarding the first point, the U.S. Supreme Court has long rejected the notion that a racial classification which burdens races equally is any less objectionable under the Equal Protection Clause. In Loving v. Virginia, 388 U.S. 1, 87 S.Ct. 1817, 18 L.Ed.2d 1010 (1967), the U.S. Supreme Court held a Virginia statute criminalizing interracial marriages was unconstitutional under the Equal Protection Clause. Id. at 12, 87 S.Ct. 1817. The Court rejected the state’s argument that the miscegenation statute did not discriminate on the basis of race because it “punish[ed] equally both the white and the Negro participants in an interracial marriage.” Id. at 8, 87 S.Ct. 1817. The Court reasoned: “In the case at bar … we deal with statutes containing racial classifications, and the fact of equal application does not immunize the statute from the very heavy burden of justification which the Fourteenth Amendment has traditionally required of state statutes drawn according to race.” Id. at 9, 87 S.Ct. 1817. Hence, it is irrelevant whether the racial tiebreaker disadvantages both races equally.

Second, I think I have already disposed of the majority’s argument that no student is entitled to attend any specific District school. The students and parents clearly value some of the District’s schools above the others, and limiting access to those higher quality schools on the basis of race is just the same as any other preferential racial classification.

Third, I agree the tiebreaker does not uniformly benefit one race over the other and can exclude both white and nonwhite students from the preferred schools. Yet that does not lessen the injury of being subject to a racial classification. Equal protection is an individual right, and whenever the District tells one student, whether white or nonwhite, he or she cannot attend a particular school on the basis of race, that action works an injury of constitutional proportion. See Adarand, 515 U.S. at 230, 115 S.Ct. 2097 (“[A]ny individual suffers an injury when he or she is disadvantaged by the government because of his or her race, whatever that race may be.”); Monterey Mech. Co., 125 F.3d at 712 (“Race discrimination is never a `trifle.'”).

The District’s use of the racial tiebreaker thus unduly burdens members of the disfavored class, and the tiebreaker fails this narrow-tailoring factor as well.

1218*1218 E.

The fifth and final narrow-tailoring factor requires the District’s use of the racial tiebreaker to “be limited in time,” and “have a logical end point.” See Grutter, 539 U.S. at 342, 123 S.Ct. 2325. A workable “sunset” provision within any government-operated racial classification is vital:

[A] core purpose of the Fourteenth Amendment was to do away with all governmentally imposed discrimination based on race…. The requirement that all race-conscious admissions programs have a termination point assures all citizens that the deviation from the norm of equal treatment of all racial and ethnic groups is a temporary matter, a measure taken in the service of the goal of equality itself.

Id. at 341-42, 123 S.Ct. 2325 (internal quotation marks and alterations omitted).

Citing Grutter, the majority contends the racial tiebreaker satisfies this factor because “this durational requirement can be met by periodic reviews to determine whether racial preferences are still necessary to achieve student body diversity,” and the District engages in such periodic reviews. Majority op. at 1192. Yet citing Grutter in full shows that “the durational requirement can be met by sunset provisions in race-conscious admissions policies and periodic reviews to determine whether racial preferences are still necessary to achieve student body diversity.” 539 U.S. at 342, 123 S.Ct. 2325 (emphasis added). Periodic reviews are not enough; there must be some “durational requirement,” some “logical end point,” to the racial classifications.

The District argued the end point is in the “thermostat” to the tiebreaker, in which the District ceases to use the racial tiebreaker at any school for the year once its use had brought the school into racial balance. Yet it is undisputed that the District has never been segregated by law; the racial imbalance in its schools results from Seattle’s racially imbalanced housing patterns. If Seattle’s children were simply assigned to the high schools nearest their homes, those schools supposedly would tend to reflect such imbalance.

Because there is no reason — much less evidence — to conclude Seattle’s housing patterns will change, or that the District’s student assignment program will affect such patterns, I must respectfully disagree that such a provision satisfies the “sunset provision” requirement enunciated in Grutter. Presumably, where the District employs the racial tiebreaker, the schools will become racially balanced, that is 40% white, 60% nonwhite (plus or minus a few percentage points, depending on the particular percentage deviation triggering the tiebreaker that year). Pursuant to the “thermostat,” the District would then stop using the racial tiebreaker. But because Seattle’s residential makeup is racially imbalanced[26] (and remains so despite the use of the racial tiebreaker), assignment to the oversubscribed schools would then occur only with use of (1) the sibling tiebreaker; and (2) the distance tiebreaker. Assuming that not every student also has a sibling attending one of the District’s schools, the schools will inevitably become racially imbalanced again because of the racially imbalanced residential makeup, thus rendering the thermostat useless as a “sunset provision.”

One could argue, then, that this result supports the need for use of the racial tiebreaker. Not necessarily so. If the racial imbalance in the schools is caused 1219*1219 not by the students, but by the choices of the parents as to where to live, then why not put the onus of remedying that imbalance on the parents rather than the students? Seattle’s city council could create “incentives” for whites to move into nonwhite areas, and for nonwhites to move into white areas. And if incentives do not accomplish the task, well, why not use compulsion, as the District does to high school children? The city council could take measures to prevent new persons taking up residence in Seattle from living in areas where their presence might otherwise alter the sought-after racial balance. This would protect the racial balance within the schools and squarely put the burden of remedying the racial imbalance upon the parents, rather than the students.

Of course, less political resistance can be expected from choosing students for social engineering experiments in racial balancing, than in telling everyone — including voters — into which neighborhood they can move. Further, regulation of residence by race might run afoul of Shelley v. Kraemer, 334 U.S. 1, 68 S.Ct. 836, 92 L.Ed. 1161 (1948), although it is difficult to distinguish why the “compelling interest” of socialization among the races could not as easily be pressed in housing regulation as it is in schooling regulation.

The simple truth is that some people choose to live near members of their own ethnic or racial group.

There is no denying that American blacks often live in their own residential enclaves, especially in our big cities. But the same is true of whites and of every other racial and ethnic group — Jews, Chinese, Cambodians, Cubans, Arabs. Such racial and ethnic clustering means that a third of non-Asian minorities attend schools that are less than 10-percent white. And even though whites constitute just over 60 percent of the nation’s schoolchildren, the average white student goes to a school that is 80-percent white.

But why should we expect identical proportions of blacks and whites to live in each and every neighborhood? People like to live near others with whom they identify, and the schools mirror their choices. When asked about their residential preferences, only about 5 percent of blacks said they wished to live on an entirely or almost entirely white block. The vast majority preferred neighborhoods that were half or more than half African-American — in other words, neighborhoods in which the black concentration was “disproportionately” high. According to the 2000 census, this happens to correspond closely to the actual distribution of black city-dwellers.

In a complex, heterogenous society, it is only natural that people should sort themselves out in urban space along lines of race as well as of religion and social class. This pattern was firmly established in the U.S. by the European immigrants who landed in the cities of the North in the 19th and early 20th centuries. The sociologists who studied these settlements recognized the important social functions served by “Little Italies” and “Poletowns”.

Abigail Thernstrom[27] & Stephan Thernstrom, Have We Overcome?, Commentary, Nov. 2004, at 51-52.[28]

Of course, the continuing racial imbalance in some residential areas is in significant 1220*1220 part a byproduct of past efforts to exclude minority groups from predominately white areas. Yet as racial tolerance and enforcement of civil rights laws have increased, neighborhoods are becoming more racially balanced. Id. In 1960, 15% of African-Americans lived in suburbs. In 2004, 36% live in suburbs. Id. African-Americans account for 9% of the total suburban population, “surprisingly close to proportionality for a group that constitutes only 12 percent of the American population.” Id. Moreover, from 1960 to 2000, the proportions of African-American living in census tracts that were over 80% black fell from 47% to under 30%. Id. During that same period, the proportion residing in census tracts that were over 50% black fell from 70% to 50%. Id. Most importantly, this balancing takes place without any government coercion, except perhaps by the enforcement of fair housing laws which prevent racial discrimination such as California’s Unruh Civil Rights Act, CAL. CIV.CODE § 51 (West 2001).

No one who understands what makes America great can quarrel with ethnic pride. At home, on the weekend, in the family and the neighborhood, Jews will be Jews, Italians Italian — and there is no reason blacks should be any different. Religion and ethnicity are essential parts of our lives, and government should not curtail how we express them in the private sphere. But when it comes to public life, even the benevolent color coding of recent decades has proved a recipe for alienation and resentment. Society need not be colorblind or color-less, but the law cannot work unless it is color-neutral, and the government should not be in the business of abetting or paying for the cultivation of group identity.

Schuck, supra, at 88 (quoting Tamar Jacoby, Someone Else’s House: America’s Unfinished Struggle for Integration 541 (1998)) (internal alteration omitted).

The racial imbalance in Seattle’s schools results not from de jure segregation nor from any invidious exclusion of nonwhite minorities from the schools. Instead, it results from racially imbalanced residential housing patterns, an issue which the District does not even contend it can alter. Hence, the method chosen by the District to impose racially balanced schools is fatally flawed. Because it does not respond to the racial imbalances in Seattle’s residential makeup, and instead only attempts to fix it within the schools, there will be no sunset to the use of the racial tiebreaker. See Grutter, 539 U.S. at 343, 123 S.Ct. 2325 (“It would be a sad day indeed were America to become a quota-ridden society, with each identifiable minority assigned proportional representation in every desirable walk of life. But that is not the rationale for programs of preferential treatment; the acid test of their justification will be their efficacy in eliminating the need for any racial or ethnic preferences at all.”). Thus, the District’s operation of the racial tiebreaker fails this factor as well.

V.

As pointed out in the majority opinion, other courts have concluded that a school district’s use of a racial tiebreaker in search of racial balance in the student body passes muster under the Equal Protection Clause.[29] I respectfully disagree. 1221*1221 The District’s use of the racial tiebreaker to achieve racial balance in its high schools infringes upon each student’s right to equal protection and tramples upon the unique and valuable nature of each individual. We are not different because of our skin color; we are different because each one of us is unique. That uniqueness incorporates our opinions, our background, our religion (or lack thereof), our thought, and our color. Grutter attempted to strike a balance between the individual protections of equal protection and being conscious of race even when looking at the individual. The District’s use of the racial tiebreaker, however, attempts no such balance; it instead classifies each ninth-grade student solely by race. Because of that, I must conclude such a program violates the Equal Protection Clause.

The majority’s decision risks unfortunate repercussions. On the short-term, the specter of “white flight” (a recurring issue in the aftermath of the elimination of de jure desegregation) manifests itself here. The racial balancing of students will require busing and long-distance transportation to schools outside of some students’ neighborhoods. Parental involvement in those distant schools (such as with the PTA) will undoubtedly decrease. Parents who can afford private education (such as those in the more affluent northern part of Seattle) may very well choose to pull their children from the District schools and enroll them elsewhere, much like the Kurfurst and Bachwitz children. On the long-term, such an exodus could result in a decreased tax base and public support for the District schools and may result in the exact opposite the District hopes to achieve-a loss of white students from their school campuses.

One of the greatest stains upon the history of our country is our struggle with race discrimination. Perhaps that stain would not be so deep had we chosen a different approach to our equal protection jurisprudence, an approach often-quoted:

Our Constitution is color-blind, and neither knows nor tolerates classes among citizens. In respect of civil rights, all citizens are equal before the law. The humblest is the peer of the most powerful. The law regards man as man, and takes no account of his surroundings or of his color when his civil rights as guaranteed by the supreme law of the land are involved.

Plessy v. Ferguson, 163 U.S. 537, 559, 16 S.Ct. 1138, 41 L.Ed. 256 (1896) (Harlan, J., dissenting).

Or, as more recently said by the late Justice Stanley Mosk of the California Supreme Court:

Racism will never disappear by employing devices of classifying people and of thus measuring their rights. Rather, wrote Professor Van Alstyne, `one gets beyond racism by getting beyond it now: by a complete, resolute, and credible commitment [n]ever to tolerate in one’s own life or in the life or practices of one’s government the differential treatment of other human beings by race. Indeed, that is the great lesson for government itself to teach: in all we do in 1222-1240*1222-1240 life, whatever we do in life, to treat any person less well than another or to favor any more than another for being black or white or brown or red, is wrong. Let that be our fundamental law and we shall have a Constitution universally worth expounding.’

Price v. Civil Serv. Comm., 26 Cal.3d 257, 161 Cal.Rptr. 475, 604 P.2d 1365, 1391 (1980) (Mosk, J., dissenting) (quoting William Van Alstyne, Rites of Passage: Race, the Supreme Court, and the Constitution, 46 U. CHI. L.REV. 775, 809-10 (1979)).

The way to end racial discrimination is to stop discriminating by race.

For the reasons expressed above, I respectfully dissent and would reverse the judgment of the district court, holding the District’s use of the racial tiebreaker in its high school admissions program violates the equal protection rights of each student excluded from a particular school solely on the basis of that student’s race.

[1] The terms “racial diversity,” “racial concentration” and “racial isolation” have been used by the District to encompass racial and ethnic diversity, concentration and isolation. For the purposes of this opinion, we adopt this shorthand.

[2] We draw the following restatement of facts largely from the district court opinion, see Parents Involved in Cmty. Schs. v. Seattle Sch. Dist. No. 1, 137 F.Supp.2d 1224 (W.D.Wash. 2001) (“Parents I“), and the Washington Supreme Court Opinion, see Parents Involved in Cmty. Schs. v. Seattle Sch. Dist. No. 1, 149 Wash.2d 660, 72 P.3d 151 (2003) (“Parents IV“).

[3] The history that follows comes principally from two documents in the district court record. One is a report 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 assignment plan. (They are cited as History of Desegregation and Findings and Conclusions, respectively.)

[4] For example, the Seattle Plan was confusing, required mandatory busing of nonwhite students in disproportionate numbers, made facilities and enrollment planning difficult and contributed to “white flight” from the city schools. Findings and Conclusions at 30.

[5] The current popularity of Ballard High School is illustrative of the constantly changing dynamic of Seattle’s public high schools. In the fall of 1999, Ballard moved to a new facility under the leadership of a new principal. Prior to the move, Ballard was not oversubscribed; now it is one of the most popular high schools in Seattle.

Similarly, the popularity and demographics of Nathan Hale High School changed significantly when it acquired a new principal who instituted a number of academic innovations, including joining the “Coalition of Essential Schools” and instituting a “Ninth Grade Academy” and “Tenth Grade Integrated Studies Program.” Prior to 1998, Nathan Hale, a north area high school, was not oversubscribed, and the student body was predominantly nonwhite. Starting in 1998, the high school began to have a waitlist, and more white students, who had previously passed on Nathan Hale, wanted to go there. As a result, the number of nonwhite students declined dramatically between 1995 and 2000.

[6] Although the record reflects the general effects of the tiebreaker in 2001-02, it does not include the specific number of students affected by the tiebreaker in the three oversubscribed schools where the tiebreaker applied. The record, however, does include these numbers for the 2000-01 school year. Although the tiebreaker operated differently in 2000-01, and applied to four schools rather than three, the 2000-01 numbers illustrate the general operation of the tiebreaker.

[7] The Board’s decision to change the trigger point for use of the tiebreaker from plus or minus 10 percent to plus or minus 15 percent, however, had the effect of rendering Roosevelt High School neutral for desegregation purposes. Thus, the tiebreaker did not factor into assignments to Roosevelt High School in the 2001-02 school year.

[8] Wash. Rev.Code § 49.60.400 (“The state shall not discriminate against, or grant preferential treatment to, any individual or group on the basis of race, sex, color, ethnicity, or national origin in the operation of public employment, public education, or public contracting.”).

[9] U.S. Const. amend. XIV, § 1 (“No state shall . . . deny to any person within its jurisdiction the equal protection of the laws.”).

[10] 42 U.S.C. § 2000d (“No person in the United States shall, on the ground of race, color, or national origin, be excluded from participation in, be denied the benefits of, or be subjected to discrimination under any program or activity receiving Federal financial assistance.”). Because “discrimination that violates the Equal Protection Clause of the Fourteenth Amendment committed by an institution that accepts federal funds also constitutes a violation of Title VI,” we address the twin challenges to the racial tiebreaker simultaneously. Gratz, 539 U.S. at 276 n. 23, 123 S.Ct. 2411.

[11] We review the district court’s resolution of cross-motions for summary judgment de novo. United States v. City of Tacoma, 332 F.3d 574, 578 (9th Cir.2003).

[12] Judge Kozinski’s concurrence makes a powerful case for adopting a less stringent standard of review here because the Plan does not attempt to “benefit[] or burden[] any particular group;” therefore it “carries none of the baggage the Supreme Court has found objectionable” in earlier equal protection cases. Kozinski, J., concurring, infra at 1194 and 1196. Recognizing the importance of context in the Supreme Court’s equal protection jurisprudence, Judge Kozinski proposes “robust and realistic” rational basis rather than strict scrutiny review. Id. at ___. Cf. Coalition for Economic Equity v. Wilson, 122 F.3d 692, 708 n. 16 (9th Cir.1997) (“We have recognized . . . that `stacked deck’ programs trench on Fourteenth Amendment values in ways that `reshuffle’ programs do not. Unlike racial preference programs, school desegregation programs are not inherently invidious, do not work wholly to the benefit of certain members of one group and correspondingly to the harm of certain members of another group, and do not deprive citizens of rights.”) (internal quotation marks, alterations and citations omitted).

Nonetheless, the Supreme Court in Johnson v. California, ___ U.S. ___, 125 S.Ct. 1141, 160 L.Ed.2d 949 (2005), rejected the argument that a California Department of Corrections (“CDC”) policy in which all inmates were segregated by race should be subjected to relaxed scrutiny because the policy “neither benefits nor burdens one group or individual more than any other group or individual.” Id. at 1147 (internal quotation marks omitted); see also id. at 1146 (noting that all racial classifications “raise special fears that they are motivated by an invidious purpose” and that “[a]bsent searching judicial inquiry into the justification for such race-based measures, there is simply no way of determining . . . what classifications are in fact motivated by illegitimate notions of racial inferiority or simple racial politics” (internal quotation marks and citation omitted)). As Judge Kozinski aptly notes, Johnson is not entirely analogous to the instant case because the CDC segregated inmates on the basis of race, whereas the District’s use of race is aimed at achieving the opposite result—attaining and maintaining integrated schools. Kozinski, J., concurring, infra. at 1194. Nevertheless, like the First and Sixth Circuits—the only other circuits to rule, post-Grutter and Gratz, on the constitutionality of a voluntary plan designed to achieve the benefits of racial diversity in the public secondary school setting—we conclude that the Plan must be reviewed under strict scrutiny. See Comfort v. Lynn School Committee, 418 F.3d 1, 6 (1st Cir.2005) (en banc); McFarland v. Jefferson County Public Schools, 416 F.3d 513, 514 (6th Cir.2005) (per curiam).

[13] The Court also heeded the judgment of amici curiae—including educators, business leaders and the military—that the educational benefits that flow from diversity constitute a compelling interest. Grutter, 539 U.S. at 330, 123 S.Ct. 2325 (“The Law School’s claim of a compelling interest is further bolstered by its amici, who point to the educational benefits that flow from student body diversity.”); see also id. (“These benefits are not theoretical but real, as major American businesses have made clear that the skills needed in today’s increasingly global marketplace can only be developed through exposure to widely diverse people, cultures, ideas, and viewpoints.”); id. at 331, 123 S.Ct. 2325 (“[H]igh-ranking retired officers and civilian leaders of the United States military assert that, `[b]ased on [their] decades of experience,’ a `highly qualified, racially diverse officer corps . . . is essential to the military’s ability to fulfill its principle mission to provide national security.'”).

[14] Academic research has shown that intergroup 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).

[15] The District’s compelling interests in diversity have been endorsed by Congress. In the Magnet Schools Assistance Act, Congress found that “It is in the best interests of the United States—(A) to continue the Federal Government’s support of local educational agencies that are voluntarily seeking to foster meaningful interaction among students of different racial and ethnic backgrounds, beginning at the earliest stages of such students’ education; (B) to ensure that all students have equitable access to a high quality education that will prepare all students to function well in a technologically oriented and a highly competitive economy comprised of people from many different racial and ethnic backgrounds.” 20 U.S.C. § 7231(a)(4) (emphasis added).

[16] According to the Seattle Times’ School Guide submitted by Parents, for the year 2000, on average 34 percent of Seattle’s high school graduates attend four-year colleges after graduation and 38.2 percent attend two-year colleges, although percentages vary from high school to high school.

[17] The prospect of children across the nation being required to attend racially concentrated or isolated schools is a crisis that school boards, districts, teachers and parents confront daily. See Civil Rights Project 4 (“At the beginning of the twenty-first century, American public schools are now twelve years in the process of continuous resegregation. The desegregation of black students, which increased continuously from the 1950s to the late 1980s, has now receded to levels not seen in three decades.”).

[18] Like the District, none of the school districts in the above-cited cases was subject to a court-ordered desegregation decree nor, with the exception of Andrew Jackson, did the schools face an imminent threat of litigation to compel desegregation. Like the District, they may have been vulnerable to litigation in decades past, but the districts’ voluntary desegregation measures would make it difficult today 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”).

[19] The dissent correctly notes that these decisions were rendered in the context of de jure segregation. But their import is also significantly compelling in the context of de facto segregation, as in Seattle. Indeed, in Swann, the Court further stated, “Our objective in dealing with the issues presented by these cases is to see that school authorities exclude no pupil of a racial minority from any school, directly or indirectly, on account of race. . . .” 402 U.S. at 23, 91 S.Ct. 1267 (emphasis added).

[20] The Court explained that “critical mass” was defined by the law school as “meaningful numbers” or “meaningful representation,” or “a number that encourages underrepresented minority students to participate in the classroom and not feel isolated.” Grutter, 539 U.S. at 318, 123 S.Ct. 2325 (internal quotation marks omitted).

[21] Parents do not claim that their children have a right to attend a particular school, nor could they. See Bustop Inc., 439 U.S. at 1383, 99 S.Ct. 40 (rejecting any legally protected right to have children attend their nearest school). In any case, under the current Plan, all students can attend a school close to their home. Because there are multiple schools in the north and south of Seattle, students for whom proximity is a priority may elect as their first choice one of the schools in their residential area that is not oversubscribed and be guaranteed an assignment to that school.

[22] In Bakke, Justice Powell noted:

Respondent’s position is wholly dissimilar to that of a pupil bused from his neighborhood school to a comparable school in another 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.

438 U.S. at 301 n. 39, 98 S.Ct. 2733.

[23] The dissent believes that “the educational benefits from diversity, if any, are much greater at the higher educational level because such benefits are greatly magnified by the learning that takes place outside the classroom. . . .” Bea, J., dissenting, infra. at 1207. This belittles the substantial role of high school classroom discussions in contributing to the educational development of our young citizens. “The [high school] classroom is peculiarly the marketplace of ideas. The Nation’s future depends upon leaders trained through wide exposure to that robust exchange of ideas which discovers truth out of a multitude of tongues.” Tinker v. Des Moines Independent Community Sch. Dist., 393 U.S. 503, 512, 89 S.Ct. 733, 21 L.Ed.2d 731 (1969) (internal quotation marks omitted).

[24] The dissent calculates that individualized consideration would be administratively feasible because only 300 students would need to be considered holistically. Though it is true that 300 students were subject to the race-based tiebreaker, it does not follow that only those 300 would require individualized consideration. Under the dissent’s view of the way the District should operate, all 3,000 students would have to be subject to holistic consideration to determine their proper school assignment. Whether or not this is administratively feasible is not clear in the record, but we believe it is ultimately irrelevant because individualized consideration is not required in the context presented here.

[25] Reliance on group characteristics is not necessarily constitutionally infirm under Fourteenth Amendment jurisprudence. See, e.g., Kimel v. Florida Bd. of Regents, 528 U.S. 62, 84, 120 S.Ct. 631, 145 L.Ed.2d 522 (2000) (“Under the Fourteenth Amendment, a State may rely on age as a proxy for other qualities, abilities, or characteristics that are relevant to the State’s legitimate interests. The Constitution does not preclude reliance on such generalizations. That age proves to be an inaccurate proxy in any individual case is irrelevant.”)

[26] The dissent’s alternative proposals to achieve the District’s interests in diversity illustrate the difficulty of individualized consideration in the high school context. For example, the dissent offers socioeconomic status as a more narrowly tailored and acceptable form of diversifying the District’s schools. However, socioeconomic status does nothing more than substitute a number from a family’s tax return for race. There is no holistic, individualized consideration under such an approach.

[27] Much like the rationale underlying the Court’s requirement of individualized, holistic review, the rationale underlying the Court’s prohibition of quotas does not apply to the race-based tiebreaker. In paradigmatic affirmative action settings — employment and admissions to institutions of higher learning — the Court disapproves of quotas because they are viewed as insulating minority candidates from competition with nonminority candidates for scarce government resources usually awarded on the basis of an applicant’s qualifications — jobs, promotions or places in a law school class. See Bakke, 438 U.S. at 317, 98 S.Ct. 2733 (opinion of Powell, J.). This is objectionable because no “matter how strong their qualifications,” nonminority candidates are never afforded the chance to compete with applicants from the preferred groups for the set-aside. Id. at 319, 98 S.Ct. 2733. Because noncompetitive assignment to Seattle’s public high schools is not based on a student’s relative qualifications, the dangers that are presented by a quota — of substituting racial preference for qualification-based competition — are absent here.

[28] Although the dissent contends that the “tiebreaker aims for a rigid, predetermined ratio of white and nonwhite students,” we believe it is more appropriately viewed as a “permissible goal.” Such a goal “requires only a good faith effort . . . to come within a range demarcated by the goal itself.” Grutter, 539 U.S at 334, 123 S.Ct. 2325 (internal quotation marks and citation omitted). The tiebreaker’s broad, 30% range and the District’s willingness to turn off the use of the tiebreaker after the ninth grade are consistent with a goal as opposed to a rigid ratio.

[29] Notably, the District’s percentage of white and nonwhite enrollment is significantly more varied than the percentage of underrepresented minorities admitted to the University of Michigan’s Law School, which remained relatively consistent. From 1995 to 1998, the percentage of minority students enrolled in the law school was 13.5 percent, 13.8 percent, 13.6 percent and 13.8 percent. Grutter, 539 U.S. at 389-90, 123 S.Ct. 2325 (Kennedy, J., dissenting). In contrast, the District’s percentage of white and nonwhite enrollment encompasses a wide range. For example, for the 2000-01 school year, the percentage of nonwhite students in the ninth grade classes of the four oversubscribed public high schools after the racial tiebreaker was applied, varied from 54.2 percent at Ballard, to 59.5 percent at Franklin, to 40.6 percent at Nathan Hale to 55.3 percent at Roosevelt.

[30] Slightly more than 80 percent of all entering ninth grade students were assigned to their first choice school.

[31] For example, in 1995, 662 (approximately 16 percent) of the 4147 law school applicants were underrepresented minorities; in 1996, 559 (approximately 15 percent) of the 3677 law school applicants were underrepresented minorities; in 1997, 520 (approximately 15 percent) of the 3429 law school applicants were underrepresented minorities. See Grutter, 539 U.S. at 384, 123 S.Ct. 2325 (Rehnquist, C.J., dissenting).

[32] Although we characterize it as de facto residential segregation, we are mindful of Justice Marshall’s dissent in Board of Education v. Dowell, “The . . . conclusion that the racial identity of the northeast quadrant now subsists because of `personal preference[s]’ pays insufficient attention to the roles of the State, local officials, and the Board in creating what are now self-perpetuating patterns of residential segregation.” 498 U.S. 237, 263, 111 S.Ct. 630, 112 L.Ed.2d 715 (1991) (internal citation omitted).

[33] 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). The theme of local control over public education has animated Supreme Court jurisprudence. 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))); see also Bethel Sch. Dist. v. Fraser, 478 U.S. 675, 683, 106 S.Ct. 3159, 92 L.Ed.2d 549 (1986) (“The determination of what manner of speech in the classroom or in the school assembly is inappropriate properly rests with the school board.”); LaVine v. Blaine School District, 257 F.3d 981, 988 (9th Cir.2001) (“In the school context, we have granted educators substantial deference as to what speech is appropriate.”) (citing and quoting Hazelwood Sch. Dist. v. Kuhlmeier, 484 U.S. 260, 267, 108 S.Ct. 562, 98 L.Ed.2d 592 (1988)). These Supreme Court decisions suggest that secondary schools occupy a unique position in our constitutional tradition. For this reason, we afford deference to the District’s judgment similar to that which Grutter afforded the university. See Grutter, 539 U.S. at 328-29, 123 S.Ct. 2325.

[34] The dissent urges, “The way to end discrimination is to stop discriminating by race.” Bea, J., dissenting, infra. at 1221. More properly stated, the way to end segregation is to stop separation of the races. The Seattle school district is attempting to do precisely that.

[35] 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).

[36] As detailed earlier, the Board’s decision to change the trigger point for use of the tiebreaker from plus or minus 10 percent to plus or minus 15 percent had the effect of rendering Roosevelt High School neutral for desegregation purposes. Thus, the tiebreaker did not factor into assignments to Roosevelt High School in the 2001-02 school year.

[37] It is worth noting that plans like the District’s may actually contribute to achieving the Court’s vision in Grutter that racial preferences will no longer be necessary in 25 years — or even sooner. As Justice Ginsburg observed, “As lower school education in minority communities improves, an increase in the number of [highly qualified and competitive] students may be anticipated.” Grutter, 539 U.S. at 346, 123 S.Ct. 2325 (Ginsburg, J., concurring).

[1] For a dissenting view, see infra pp. 1204-1205.

[2] Because of our country’s struggle with racial division and the injustices of compelled government de jure segregation, we must be especially suspicious of any compulsive government program based upon race, even when such a program is supposedly beneficial. Good intentions cannot insulate the government’s use of race from the commands of the Equal Protection Clause; history is rife with examples of well-intentioned government programs which later caused grievous harm to society and individuals. See Adarand Constructors v. Pena, 515 U.S. 200, 226, 115 S.Ct. 2097, 132 L.Ed.2d 158 (1995) (“More than good motives should be required when government seeks to allocate its resources by way of an explicit racial classification system.”); Olmstead v. United States, 277 U.S. 438, 479, 48 S.Ct. 564, 72 L.Ed. 944 (1928) (Brandeis, J., dissenting) (“Experience should teach us to be most on our guard to protect liberty when the Government’s purposes are beneficent…. The greatest dangers to liberty lurk in insidious encroachment by men of zeal, well-meaning but without understanding.”).

[3] This makes all the more puzzling the majority’s assertion that “that the District has a compelling interest in securing the educational and social benefits of racial (and ethnic) diversity.” Majority op. 1166 (emphasis added). There simply is no ethnic tiebreaker.

[4] Remediation of de jure segregation is not at issue here; the parties concede the District’s schools have never been de jure segregated. No one even suggests that Seattle’s housing market has ever been affected by de jure segregation.

[5] Indeed, the term “de facto segregation” is somewhat of an oxymoron. That is perhaps why the Supreme Court preceded the term with the qualifier “so-called.” See Keyes, 413 U.S. at 208, 93 S.Ct. 2686.

[6] See Comfort v. Lynn Sch. Comm., 418 F.3d 1, 27 (1st Cir.2005) (Boudin, C.J., concurring).

[7] Id.

[8] Concurrence at ___-___ (citing Comfort, 418 F.3d at 27 (Boudin, C.J., concurring)).

[9] See infra pp. 1212-1214 (discussion of why the racial tiebreaker used by Seattle is a quota).

[10] What is “the reality” or “realistic” or “real-world” is usually a rhetorical tool for dressing up one’s own view as objective and impartial, and therefore, more presentable.

[11] See e.g. Adarand, 515 U.S. at 224, 115 S.Ct. 2097, Gratz v. Bollinger, 539 U.S. 244, 270, 123 S.Ct. 2411, 156 L.Ed.2d 257 (2003), Johnson v. California, ___ U.S. ___, 125 S.Ct. 1141, 1146, 160 L.Ed.2d 949 (2005). On this point, the majority agrees. See Majority op. pp. 1172 n. 12.

[12] The majority cites often to Grutter’s statement that “context matters” in reviewing racial classifications under the Equal Protection Clause. See 539 U.S. at 327, 123 S.Ct. 2325 (“Context matters when reviewing race-based governmental action under the Equal Protection Clause.”). There, the Court counseled that strict scrutiny was to take “relevant differences” into account. Id.

Indeed, “context” does matter; context always matters in the application of general rules of law to varied factual settings. See Gomillion v. Lightfoot, 364 U.S. 339, 343-44, 81 S.Ct. 125, 5 L.Ed.2d 110 (1960) (“Particularly in dealing with claims under broad provisions of the Constitution, which derive content by an interpretive process of inclusion and exclusion, it is imperative that generalizations, based on and qualified by the concrete situations that gave rise to them, must not be applied out of context in disregard of variant controlling facts.”). In Grutter, the “context” was a public law school’s race-conscious, individualized consideration of applicants for purposes of admissions, designed to achieve diversity. Here, the context is different; we consider a rigid racial tiebreaker, which considers only race, designed to avoid racial imbalance in the schools. And so, as we do for all cases, we look to general principles of law and apply them through the correct standard of review, cognizant of the different results reached in other cases because of different facts and the “context” in which the cases arose. But what must be remembered is that a different “context” does not change the general rules of law, nor does a different “context” change the applicable standard of review (at least for government-imposed racial classifications).

Yes, “context” matters, but the mention of “context” should not be a talisman to banish further enquiry. The “context” of the Michigan Law School is different from the District’s schools. But the difference is in the age of the students, their number and the obligation of the District to admit all students. Does that change the fact that some students are sent to certain schools solely because of their races? How does “context” change that? Let us not succumb to the use of an abstraction (“context”) to invoke “sensitivity” to “nuances,” thus to attempt to change the bald fact of selection based on race.

[13] The majority fails to recognize this distinction. For example, comparing the District’s claimed interest with those endorsed in Grutter, the majority reasons high schools “have an equal if not more important role” in preparing students for work and citizenship, and concludes “it would be a perverse reading of the Equal Protection Clause that would allow a university, educating a relatively small percentage of the population, to use race when choosing its student body but not allow a public school district, educating all children attending its schools, to consider a student’s race in order to ensure that the high schools within the district attain and maintain diverse student bodies.” Majority op. at 1175, 1176. Yet Grutter did not allow universities to consider race in admissions to achieve racial balancing. The whole point of Grutter and Gratz was that universities may consider race, but only as part of the overall individual. I see nothing perverse in recognizing the Equal Protection Clause to be the protector of the individual, whether he be among the few at an elite law school, or among the many in a public high school.

[14] Again, there is nothing illegal in freely choosing to believe in this stereotype and to act upon it as a private citizen in sending one’s child to a particular school. The case changes when such racial stereotype is accepted by the state, and is the basis for the imposition of racial discrimination.

[15] The majority notes that for purposes of the racial tiebreaker, “a student is deemed to be of the race specified in his or her registration materials.” Majority op. at 1169. That generalization declines to note a particularly overbearing facet of the racial tiebreaker. Although the District encourages the students’ parents to identify the race of their student in the registration materials, if a parent or student chooses to follow the example of Tiger Woods and refuses to identify his or her race, the District then engages in a visual inspection of the student or parent and will decide the child’s color notwithstanding the parent’s or student’s choice.

[16] See also David J. Armor & Christine H. Rossell, Desegregation and Resegregation in the Public Schools, in Beyond the Color Line: New Perspectives on Race and Ethnicity in America 251 (Abigail Thernstrom & Stephan Thernstrom eds., 2002) (“[R]acial composition by itself has little effect on raising the achievement of minority students or on reducing the minority-white achievement gap. Some studies show that there is no relationship at all between black achievement and racial composition . . ., and other studies show that there is no relationship between the black-white achievement gap and racial composition. In either case, though there is some evidence here that achievement can be affected by programmatic changes, there is no evidence that it responds to improved racial balance by itself.”); id. at 252 (“The evidence on the benefit of school desegregation for race relations is probably the weakest of all. Indeed, there are more studies showing harmful effects than studies showing positive effects.” This led to another and more recent reviewer of the race relations literature to conclude, somewhat generously: “`In general, the reviews of desegregation and intergroup relations were unable to come to any conclusion about what the probable effects of desegregation were…. Virtually all of the reviewers determined that few, if any, firm conclusions about the impact of desegregation on intergroup relations could be drawn. The reluctance of reviewers to draw conclusions about the benefits of school desegregation for race relations or self-esteem only reinforces our conclusion that the psychological harm theory of de facto segregation and the social benefit theory of desegregation are clearly wrong, at least when applied to desegregation as a racial balance policy.'”).

[17] The majority states that Swann v. Charlotte-Mecklenburg Bd. of Educ., 402 U.S. 1, 91 S.Ct. 1267, 28 L.Ed.2d 554 (1971), supports the proposition that the District has broad discretion to engage in racial balancing as an “educational policy.” In Swann, the Supreme Court stated: “School authorities are traditionally charged with broad power to formulate and implement educational policy and might well conclude, for example, 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. To do this as an educational policy is within the broad discretionary powers of school authorities; absent a finding of a constitutional violation, however, that would not be within the authority of a federal court.” Id. at 16, 91 S.Ct. 1267; see also North Carolina State Bd. of Educ. v. Swann, 402 U.S. 43, 45, 91 S.Ct. 1284, 28 L.Ed.2d 586 (1971) (same, citing Swann, 402 U.S. at 16, 91 S.Ct. 1267).

Swann’s passage seems to provide powerful language for the majority’s position, but alas, the majority takes the passage out of context. Swann considered the remedies available to a federal court to combat past de jure segregation. The Court never considered whether a school district could use racial classifications to achieve racial balance absent de jure segregation. Indeed, the Court stated: “We are concerned in these cases with the elimination of the discrimination inherent in the dual school systems, not with myriad factors of human existence which can cause discrimination in a multitude of ways on racial, religious, or ethnic grounds…. Our objective in dealing with the issues presented by these cases is to see that school authorities exclude no pupil of a racial minority from any school, directly or indirectly, on account of race; it does not and cannot embrace all the problems of racial prejudice, even when those problems contribute to disproportionate racial concentrations in some schools.” Id. at 22-23, 91 S.Ct. 1267 (emphasis added).

Swann was also decided decades before the Court resolved the issue of the level of scrutiny to apply to “benign” racial classifications, vis-a-vis “invidious” racial classifications. Thus, Swann’s dictum cannot shelter the District’s use of the racial tiebreaker from the searching inquiry required by strict scrutiny.

The majority similarly errs in relying on Washington v. Seattle Sch. Dist. No. 1, 458 U.S. 457, 102 S.Ct. 3187, 73 L.Ed.2d 896 (1982). There, the Court also specifically stated it did not reach the issue of the constitutionality of “race-conscious student assignments for the purpose of achieving integration, even absent a finding of prior de jure segregation.” Id. at 472 n. 15, 102 S.Ct. 3187.

[18] See supra pp. 1201-1202 n. 12 (explaining why the talismanic use of “context” can not alter the fact of racial discrimination).

[19] As noted, the District applies the racial tiebreaker only to entering ninth-grade students (presumably around thirteen to fourteen years old).

[20] Three hundred applications seem like only a minor administrative challenge, but the Supreme Court’s admonition bears repeating nonetheless: “[T]he fact that the implementation of a program capable of providing individualized consideration might present administrative challenges does not render constitutional an otherwise problematic system.” Gratz, 539 U.S. at 275, 123 S.Ct. 2411.

[21] In 2000-01, the District used a 10% deviation trigger, but increased the trigger to 15% for the 2001-02 school year.

[22] Although the majority concludes a quota does not exist here, it also concludes “the rationale underlying the … prohibition of quotas does not apply” here. Majority op. at 1184 n. 27. The majority reasons that because there is no competition in assignment to the District’s schools, the dangers presented by a quota — i.e., insulating applicants from competition on the basis of race — are absent here. Majority op. at 1184 n. 27. But saying it does not make it so, whether it is said by the District or by the majority. As explained above, there is clearly a “market” for higher quality schools in the District, and there is competition for the schools the parents and students view to be the better schools.

[23] The majority makes the conclusory statement that the District’s “white/nonwhite distinction is narrowly tailored to prioritize movement of students from the north of the city to the south of the city and vice versa” as an effort to combat Seattle’s racially imbalanced residential patterns. Majority op. at 1188. Yet the District’s attempt to balance students from north Seattle and south Seattle strongly suggests a less-restrictive, race-neutral approach to achieve such balancing: socioeconomic balancing. As the majority notes, the northern Seattle area contains a majority of “white” students and is “historically more affluent.” Majority op. at 1166. This would mean the southern Seattle area is less affluent. Thus, moving more affluent students south, and less affluent students north, could possibly provide a more diverse student body. At the very least, serious consideration would have been warranted into this race-neutral alternative. See Levine, supra, at 536 (noting the key element to successfully integrating students of different backgrounds and race is not racial balance, but “economic integration”).

Yet the majority accepts the District’s rejection of the use of socioeconomic factors, reasoning that “[a]lthough there was no formal study of the proposal by District staff, Board members’ testimony revealed two legitimate reasons” for rejecting the socioeconomic alternative: (1) “it is insulting to minorities and often inaccurate to assume that poverty correlates with minority status;” and (2) students would be reluctant to reveal their socioeconomic status to their peers. Majority op. at 1189. Such analysis seems far from the “serious, good-faith consideration of workable race-neutral alternatives” demanded by Grutter. See 539 U.S. at 339, 123 S.Ct. 2325. First, without formal studies (or indeed any earnest consideration of the alternatives), we have no way of knowing whether the District actually seriously considered, and rejected for valid reasons, less-restrictive race-neutral alternatives. In Croson, the Court emphasized the importance of a satisfactory record to determine whether race-neutral alternatives were considered. See Croson, 488 U.S. at 498-511, 109 S.Ct. 706 (plurality) (detailing the government actor’s failure to document the basis for its use of a racial quota and stressing the need to do so). Second, the majority’s insistence that the District’s consideration of poverty would be “insulting” ignores the demeaning — and indeed, constitutionally objectionable — effect of placing persons into groups solely by their skin color for the purpose of receiving or being denied a governmental benefit. See Loving v. Virginia, 388 U.S. 1, 11, 87 S.Ct. 1817, 18 L.Ed.2d 1010 (1967) (“[T]his Court has consistently repudiated distinctions between citizens solely because of their ancestry as being odious to a free people whose institutions are founded upon the doctrine of equality.”). Even if a sole focus on poverty might be insulting to some minorities, socioeconomic considerations need not inquire only into poverty status; eligibility for free lunch, the parents’ levels of education, or whether English is a second language for the child are also relevant determinations in evaluating diversity. Third, there is no reason students would have to reveal their socioeconomic status to their peers; the District could, of course, keep such information confidential.

[24] Similar race-neutral alternatives are common throughout the United States. For example, the San Francisco, California public school district employs a program focused on enhancing diversity in the classrooms. The program allows students to choose any school within the district. When a school is oversubscribed, the program first assigns students with siblings to the same school, and then accommodates students with specialized learning needs. After that, the “Diversity Index” handles further assignments. “Under the Diversity Index process, the school district calculates a numerical profile of all student applicants. The current Diversity Index is composed of six binary factors: socioeconomic status, academic achievement status, mother’s educational background, language status, academic performance index, and home language.” David I. Levine, Public School Assignment Methods after Grutter and Gratz: The View from San Francisco, 30 HASTINGS CONST. L.Q. 511, 528-31 (2003). Notably, the San Francisco system “does not use race as an express criterion for school assignments” and thus avoids the sharp focus of strict scrutiny. Id. at 531.

[25] In assessing whether the District seriously considered race-neutral alternatives, the majority applies deference to the District’s consideration (or lack thereof) and rejection of the various alternatives. Majority op. at 1188 n. 33. With respect, the majority errs in two respects. First, as previously noted, deference to local officials’ use of race is generally barred in the application of strict scrutiny. See Johnson, 125 S.Ct. at 1146 n. 1. Second, if the majority is attempting to apply the deference invoked in Grutter, the Court there applied deference in determining whether the Law School asserted a compelling governmental interest, not whether the means used to achieve that interest were narrowly tailored. See 539 U.S. at 328, 123 S.Ct. 2325 (“The Law School’s educational judgment that such diversity is essential to its educational mission is one to which we defer.”).

The pattern now established by the majority seems suspicious. Out of five narrow-tailoring factors, the majority has concluded two are inapplicable, and now a third is entitled to deference. I find it difficult to understand how such analysis could truly be considered strict scrutiny as to the narrowing requirement.

[26] About 70% of the residents of Seattle, Washington are white, and 30% of the residents are nonwhite. Sixty-six percent of white students live in the northern part of Seattle, while 75% of nonwhite students live in the southern part of Seattle.

[27] Mrs. Thernstrom is presently the Vice Chair of the U.S. Commission on Civil Rights.

[28] Further evidence that such self-selection results is submitted by this year’s Nobel Laureate, Thomas C. Schelling, by application of game theory in chapter four of his book Micromotives and Macrobehavior (1978). Schelling employs an exercise using coins to demonstrate how an integrated neighborhood can become largely segregated as long as each resident desires at least one third of his or her neighbors to be of his or her race. When one person moves to get a preferred set of neighbors, it causes a chain reaction which settles down only when the neighborhood is effectively segregated.

[29] Cf. Comfort v. Lynn Sch. Comm., 418 F.3d 1, 5 (1st Cir.2005) (en banc) (holding a public high school district had a compelling interest, in the absence of de jure segregation, in using race-based assignments to “secur[e] the educational benefits of racial diversity,” and the means used to serve that interest were narrowly tailored); McFarland v. Jefferson County Pub. Sch., 330 F.Supp.2d 834, 850 (W.D.Ky.2004) (holding a public high school district had a compelling interest in using race-based assignments to maintain racially integrated schools, and the means used to serve that interest were narrowly tailored), aff’d, 416 F.3d 513 (6th Cir.2005); Brewer v. W. Irondequoit Central Sch. Dist., 212 F.3d 738, 752 (2d Cir.2000) (holding a public middle school district had a compelling interest, in the absence of de jure segregation, in using race-based assignments to reduce “racial isolation” in its schools).