Full opinion text
OPINION RE MOTIONS FOR UNITARY STATUS HILLMAN, Senior District Judge. After operating nearly 22 years under a remedial order, the Benton Harbor Area School District (“BHASD”) and the State of Michigan defendants move to terminate this 35-year-old school desegregation case by filing motions for unitary status. Defendants’ motions have been vigorously opposed by the plaintiff class and intervenor Michigan Education Association (“MEA”). The court heard sixteen days of testimony, received hundreds of exhibits, and reviewed extensive briefs and proposed findings of fact and conclusions of law. The court has carefully considered the facts presented as well as the governing law. For the reasons that follow, the motions of the BHASD and the State of Michigan defendants for unitary status are GRANTED. I. BACKGROUND This case began on November 16, 1967, with the filing of a complaint by plaintiffs Barbara Jean Berry, et al., as parents of African American children then attending the public schools of Benton Harbor, Michigan, against the School District of the City of Benton Harbor, the members of its Board of Education and its Superintendent. In the complaint, plaintiffs sought preliminary and permanent injunctive relief as to various acts and practices by the defendants, which plaintiffs deemed to be discriminatory or segregative. In July 1971, the district court found several practices carried out by the defendants to be constitutionally discriminatory. On November 1, 1974, the Sixth Circuit Court of Appeals affirmed the district court’s findings that the practices were discriminatory and that plaintiffs had made out a prima facie case of de jure segregation. On August 21, 1974 and September 25, 1975, plaintiffs added the following defendants to the case: the State of Michigan, the Attorney General of the State of Michigan, the Michigan State Board of Education, the Superintendent of Public Instruction (collectively, “the State of Michigan defendants” or “State defendants”), the Boards of Education of the Eau Claire Public Schools and the Coloma Community Schools, and the Berrien County Intermediate School District and its Superintendent (“BCISD”). On August 22, 1977, following a trial on the liability of Benton Harbor Area School District (Phase I trial), the district court (then Chief Judge Noel P. Fox) ordered that the case be certified as a class action under Rule 23 of the Federal Rules of Civil Procedure. The class was identified as “all present and future students within the Benton Harbor Area School District.” The court also found defendant Benton Harbor Area School District (“BHASD”) guilty of acts of segregation in violation of the United States Constitution. See Berry v. School Dish of City of Benton Harbor, 442 F.Supp. 1280 (W.D.Mich.1977). On August 7, 1978, following a second liability trial (Phase II trial), the district court ruled against the State of Michigan defendants, the Berrien County Intermediate School District and its Superintendent, and the Coloma and Eau Claire School Districts and their Superintendents, finding that by their policies, practices, actions and inactions, defendants had helped to create, perpetuate or contribute to the unlawfully segregated conditions in the Benton Harbor Area School District. The district court issued an amended order requiring the defendants found liable in Phases I and II to formulate a plan to remedy the constitutional violations. See Berry v. School Dish of City of Benton Harbor, 467 F.Supp. 630 (W.D.Mich.1978). In February 1980, the case was reassigned to the undersigned for remedial proceedings. Following a remedy trial, the court entered its Opinion and Order on May 1, 1981, adopting and ordering the implementation of a desegregation plan. See Berry v. School Dist. of City of Benton Harbor, 515 F.Supp. 344 (W.D.Mich.1981). In broad outlines, this plan: (1) ordered the Eaman residential area be returned to the Benton Harbor Area School District; (2) enjoined the transfer of the Sodus II residential area from the Benton Harbor Area School District to the Eau Claire Public School District; (3) ordered the Benton Harbor Area School District to eliminate racially identifiable schools; (4) ordered the creation of magnet programs in the Benton Harbor Area School District; (5) ordered a voluntary program for interdistrict transfers of students between the Benton Harbor, Colo-nia and Eau Claire School Districts; (6) ordered further remedies relating to curriculum, faculty and staff reassignment and affirmative action goals, in-service training, student discipline, community involvement, monitoring and reporting, and for financing of the court’s remedial plan. In addition, in its 1977 liability findings, the court found that individual student achievement of black students had been detrimentally affected by defendants’ past segregative actions and inactions. The court therefore included an extensive “achievement and social skills component” in its remedial plan. 515 F.Supp. at 369-373. On January 24, 1983, the Sixth Circuit Court of Appeals affirmed the May 1, 1981, remedial order and the United States Supreme Court denied certiorari on October 11, 1983. Berry v. School Dist. of City of Benton Harbor, 698 F.2d 813 (6th Cir.), cert. denied, 464 U.S. 892, 104 S.Ct. 235, 236, 78 L.Ed.2d 227 (1983). In September 1991, following ten years of implementation of the remedial plan, defendants Coloma, Eau Claire, the BCISD and the State defendants filed separate motions requesting the court to terminate court supervision and control and to declare the districts unitary. Plaintiffs and the BHASD opposed the motions. Thereafter, with the court’s encouragement, the parties undertook settlement negotiations. On December 3, 1993, the court dismissed the motions, for administrative purposes only, while the parties continued settlement discussions. (Dkt. # 974.) In 1996, this court considered a proposed partial settlement between plaintiffs and defendants Coloma, Eau Claire and the State. Following a preliminary approval hearing, notice and a fairness hearing, the court concluded that the 1996 proposed partial settlement was neither fair, adequate nor reasonable, and rejected the proposed agreement. Thereafter, further settlement discussions ensued, resulting in new settlement agreements between plaintiffs and Coloma, Eau Claire and the BCISD. These agreements were signed and filed on June 23, 1998, and approved and adopted as consent judgments by this court on August 20, 1998, following notice to the class and a fairness hearing. On November 13, 2001, pursuant to the terms of the settlement agreements, the court terminated jurisdiction over Coloma, Eau Claire and the BCISD. However, under the terms of the consent judgments, Coloma and Eau Claire must continue to allow BHASD resident students who were enrolled in those districts in the 1998-99 school year under the interdistrict transfer program to complete their educations in those districts. Settlement discussions continued between the remaining parties, but ultimately were unsuccessful. On April 17, 2000, the BHASD filed a motion for unitary status and the State defendants renewed their earlier motion for unitary status. Meanwhile, plaintiffs filed a motion to modify the remedial order to include supplemental remedial measures. The parties proceeded with extensive discovery. On February 2, 2001, this court bifurcated the hearing on the motions “horizontally,” determining that it was proper for this court to decide the unitary status motion first, but concluding that it was most efficient to hear all evidence regarding the proposed school improvement plan, whether or not that evidence was strictly relevant to the issue of unitary status or to the proposed modification. Hearing on other aspects of the proposed modification of the remedial order was reserved until the court had decided the unitary status motions. From July 24, 2001 to August 16, 2001, and again on October 24, 2001, the court heard lay and expert testimony regarding the motions for unitary status and the proposed school improvement plan. The principle issues at the hearing were (1) whether racial disparities in the areas of school operations known as the Green factors (i.e., student assignment, faculty and staff assignment, transportation, facilities, and extracurricular activities) continued to exist in the BHASD as vestiges of defendants’ past segregative conduct or whether they had been eliminated to the extent practicable; (2) whether vestiges of segregation remain in areas outside the Green factors, particularly student achievement disparities, or whether such vestiges have been eliminated to the extent practicable; (8) whether the BHASD and the State defendants have complied in good faith with the court’s remedial orders. Following the hearing and the filing of the transcript of the proceedings, the parties submitted proposed findings of fact and conclusions of law, together with post-hearing briefs and responsive briefs. The matter presently is ripe for decision. II. FINDINGS OF FACT A. 1981 Remedial Order On August 22, 1977, in his initial liability findings, Judge Noel P. Fox found the BHASD guilty of unlawful segregation. During the course of his opinion, Judge Fox recognized that both white and black students had been affected by the de jure segregation. 442 F.Supp. at 1289, 1302, and 1307. The court found in particular that black students experienced significant achievement deficits and detriments to social and psychological development as a result of segregation. 442 F.Supp. at 1289, 1307. White students, in contrast, developed irrational attitudes of inherent superiority, an unrealistic concept of homogeneous society and lack of preparation for a pluralistic society. Id. In the May 1, 1981 remedial order, the undersigned reiterated Judge Fox’s earlier liability determination that the “system-wide segregation, such as the Benton Harbor district has created and the State Board of Education condoned and perpetuated, results in measurably reduced achievement.” 516 F.Supp. at 369. The remedial order recognized that past segregation and the resultant low achievement “set the stage for the exodus of white families from the district and attempted property transfers out of the consolidated district by white property owners.” Id. at 349. The remedial order, however, acknowledged the complexity of devising an effective desegregation remedy in circumstances in which population trends and the economic deterioration of Benton Harbor also operated to increase the racial isolation of the BHASD. Id. at 347. At the time it issued its remedial order, this court recognized that its authority to impose a remedy was limited exclusively to remedying constitutional violations and that the “ ‘scope of the remedy is determined by the nature and extent of the constitutional violation.’ ” Id. at 848 (quoting Swann v. Charlotte-Mecklenburg Bd. of Ed., 402 U.S. 1, 16, 91 S.Ct. 1267, 28 L.Ed.2d 554 (1971)); see also id. at 353 (quoting Milliken v. Bradley, 418 U.S. 717, 738, 94 S.Ct. 3112, 41 L.Ed.2d 1069 (1974) (“Milliken I ”) (“A federal remedial power may be exercised only on the basis of a constitutional violation and, as with any equity case, the nature of the violation determines the scope of the remedy.”) (quotations omitted)). The court determined that the remedy appropriate to address defendants’ constitutional violations included components designed both to enhance desegregation within the three districts and to address problems of student achievement caused by segregation and the denial of educational opportunities. Id. at 348-49, 355, 358-65. First, the remedy enjoined future property transfers from the BHASD to other districts and ordered the return of the Eamon area previously transferred to Eau Claire. It also prohibited Eau Claire from accepting tuition students from the BHASD. Id. at 358. Second, the remedial order required the BHASD to adopt a student reassignment plan that would eliminate the racial identi-fíability of schools and achieve desegregation within all schools in the BHASD. Id. Third, the remedy created an interdis-trict transfer program between the BHASD and Coloma and the BHASD and Eau Claire, permitting students to voluntarily transfer to another district whenever such transfer would result in decreasing segregation in each school system. Id. at 367. Fourth, to encourage voluntary transfers, the court ordered the development of magnet programs that would be attractive to students from all three districts at all levels in such a way that all three districts would be further desegregated. Id. at 365-66. Fifth, the remedial order noted that low achievement was highly correlated with racially-identifiable black schools, id. at 357 n. 15, and reiterated its liability finding that segregated black students experience reduced self-esteem, diminished motivation to succeed, and a concomitant measurable reduction in achievement, id. at 369 & n. 20. The remedy therefore included a component designed to address chronic deficits in achievement and social skills created by past segregated and “inherently unequal” schools. Id. at 373 & n. 24. The achievement and social skills component included the implementation of a school improvement program to be developed by Dr. James Comer at the Child Study Center of Yale University. Id. at 371. The BHASD was directed to continue to implement the achievement component “until achievement within the district reaches the average attained by Michigan students in state-wide achievement tests. Once that goal has been achieved, the program will be phased out over two additional school years.” Id. at 370. Sixth, the remedial order required review of and additions to the curricula of the three districts. Id. at 373-74. Seventh, the court ordered changes in the hiring and assignment of faculty, staff and administration in the BHASD, Coloma and Eau Claire in order to diversify the districts as a whole and to eliminate racial identification of particular schools. Id. at 375-76. Eighth, the court ordered the districts to adopt an in-service workshop component of the plan, which would provide mandatory training on a range of diversity issues for all members of the school districts, and would provide opportunities for parental participation. Ninth, the court directed the districts to adopt a uniform discipline code and an inter-district committee on student conduct and discipline. Id. at 380. The court also adopted a series of measures necessary to implement the plan. The court designated Dr. Michael Stolee as its representative for implementation. Id. at 381. The court also established the Community Education Council, a community-wide representative board with a full-time director (the “CEC”). The CEC was established to provide accurate information to the communities about the desegregation process, to improve citizen involvement in the process, to serve as liaison between communities and the school districts, to assist districts in dispute resolution, and to serve as the primary body monitoring implementation of the remedy for the court. Id. at 382-83. In addition, the court ordered the transportation of transfer students, established timetables for implementation of the plan, and required the districts to submit annual progress reports. Id. at 383-85. Further, the court established the mechanisms for funding all aspects of the plan. Id. at 385-88. In particular, to prevent a student’s transfer election from imposing a financial penalty on the student’s district of residence, the State was required to “continue to pay to the school district of residence, the amount of State financial aid that the student would have generated had he or she not elected to transfer,” as well as to “pay the receiving school district 100% of the costs involved in educating that pupil,” at the same rate as annual costs for regularly enrolled students. 515 F.Supp. at 386. The court also directed that “[f]ederal funding of programs outlined in th[e] plan should be sought to the maximum feasible levels” by the three local school districts, the intermediate school district and the state acting cooperatively. Id. at 385. The court required that additional funds “generated above and beyond regular district operating budgets under the financial plan set forth in the paragraph below shall be used by the districts acting cooperatively to finance enriched educational opportunities and magnet programs, in each district, for the benefit of children in all three districts.” Id. B. The Green Factors The court finds that the BHASD has eliminated the vestiges of past segregation to the extent practicable in each of the areas of school operations identified by the Supreme Court in Green, 391 U.S. 430, 88 S.Ct. 1689, 20 L.Ed.2d 716, i.e., student assignment, faculty and staff, transportation, extracurricular activities, and facilities. Plaintiffs admit that the BHASD is unitary with respect to each of the Green factors. (Ex. S-255: Pi’s admissions); (Tr. 489-90 Rossell; Tr. 260-62 Williams; Tr. 63-64 Peeples Burks.) In addition, the court finds that the uncontroverted evidence presented at hearing fully supports the determination that the BHASD has fully desegregated its schools under each of the Green factors. (See generally, Tr. 476-577 and specifically, 489-91 Rossell.) C. Student Achievement Vestiges As previously mentioned, in its 1977 liability findings the court found that defendants’ past segregative conduct had resulted in measurably reduced achievement in the BHASD. The court’s 1981 remedial order provided for an achievement and social skills component to address chronic low achievement. The goal of the achievement component was to “raise the level of student achievement until achievement within the district reaches the average attained by Michigan students in statewide achievement tests.” 515 F.Supp. at 370. On May 11, 2001, this court ruled that defendants have the burden of proving that current achievement levels are not causally related to past segregation. (Dkt. # 1610.) After careful analysis of the evidence presented at the hearing, and despite certain inadequacies in the implementation of the Comer education model, the court finds that defendants have met their burden. As stipulated by the parties, “no children in the Benton Harbor schools today were enrolled in the BHASD when the remedy began in 1981.” (Stip. No. 2a.) Moreover, current levels of educational attainment in the district are not attributable to deficiencies in school funding in the district. Under the desegregation order, the BHASD has enjoyed resource advantages compared to most other school districts within Michigan generally and within Berrien County in particular. The ratio of pupils to teacher is generally lower and the teachers are generally better educated. The average teacher salary is not on the low end of salaries in the region. (S-3 to S — 7; Tr. 91-104 Armor; 932-33 Williamson.) The needs of the district are greater than many other districts because of the financial condition of the community. (Tr. 287-88, 291, 334 Williams (citing mandatory breakfast and lunch programs all days; high teen pregnancy; large numbers of entering children with impoverished communications skills).) Poverty in the BHASD has risen steadily both before and after the implementation of the remedial order. (S-39.) In 1990, 60% of black households in the City of Benton Harbor were in poverty, compared with only 42% in 1980. (S — 89; Tr. 452 Clark.) Benton Harbor also has suffered very significant population losses overall. (S-30.) White and middle class families have left the BHASD in substantial numbers, while black households have increased. (S-30 to S-33.) The white population in the City of Benton Harbor declined from 14,-000 in 1960 to approximately 600 in the year 2000, while the black population increased from about 4,800 to over 10,000 over the same period. (S — 31; Tr. 437 Clark.) On a percentage basis, the change in black population from 1960 to 2000 was from 27% to 95%. (Tr. 438 Clark.) In the BHASD as a whole, the white population declined from about 35,000 in 1960 to about 12,000 in 1990, while the black population increased from 10,000 to almost 22,000 over the same period. These population changes have been affected both by patterns of migration (out-migration of white families and in-migration of black families) and by higher black fertility rates. (S — 38; Tr. 448-50 Clark.) As a result, the enrollment of the BHASD has changed markedly from the time of the remedial order to the present. Between 1980-81 and 1999-2000, white enrollment in the BHASD declined from about 2,000 students to fewer than 500, but black enrollment only declined from about 7,000 to 5,500. (S-30 to S-33; S-45.) The current racial composition of the BHASD enrollment is not causally related to the earlier Constitutional violations found by this court, but rather is the result of demographic and economic forces that are not unique to Benton Harbor and that both predated and continued through this litigation. (Tr. 461-62 Clark.) In an effort to assess how the BHASD schools were performing, defendants’ achievement expert, Dr. David Armor, performed a regression analysis comparing the BHASD math and reading MEAP scores to the state average for 4th and 7th grades for the years 1999 and 2000. Dr. Armor controlled for the effects of poverty in his regression analysis in order to eliminate the extant poverty in Benton Harbor as a factor affecting student achievement. Dr. Armor found that BHASD students in the 4th grade exceeded the state average for both years in reading. (Indeed, even without controlling for poverty, on the most recent MEAP tests, 4th grade reading exceeded the state average. J — 1; Tr. 942 Williamson.) In the 7th grade results, BHASD students trailed the state average by a statistically significant margin for both years, although the difference was equal to approximately one test question. In math, BHASD students exceeded the state average in 2000 and slightly trailed it in 1999, but again by only about one test question. (S — 9 to S-12; Tr. 134-41 Armor.) Dr. Armor concluded that BHASD’s poverty accounts for virtually all of the differences in educational outcomes observed between the BHASD and the rest of Michigan. (Tr. 143-44 Armor.) Dr. Armor also studied high school dropout rates as another measure of BHASD performance. Again controlling for the BHASD’s extreme poverty, Dr. Armor found that in the year 2000, the BHASD dropout rate was statistically the same for the BHASD as for the rest of the state. (S — 13; Tr. 141-43 Armor.) Dr. Armor did not, however, evaluate achievement in the BHASD at the high school level using standardized assessments. Instead, he opined that such individual studies of high school students were less reliable for two reasons: (1) by eleventh grade, the time of MEAP achievement assessment, a large number of students drop out of school, resulting in a comparison of a smaller subset; (2) in high school, fewer students are willing to use the free and reduced lunch program, rendering it a less accurate surrogate for socioeconomic status (“SES”) of the individual students. (Tr. 125 Armor.) Although I partially accept Dr. Armor’s explanation for failing to conduct a study of high school performance — ie., I accept that individual achievement studies of high school students might be less accurate— Dr. Armor failed to explain why a statewide comparison, comparing school district test averages and school district free lunch rates, is unreliable. Where, as here, Dr. Armor concludes that the drop-out rate in Benton Harbor as a whole is comparable to other Michigan districts when controlled for poverty (Tr. 141-43 Armor), and where Dr. Armor testified that the district-wide rate of free lunch was highly correlated with all other SES factors in state-wide comparisons (Tr. 232-33 Armor), I question why district-wide data could not be used to control for poverty since the smaller cohort should be comparable to that of similar districts. Despite these reservations, I nevertheless accept Dr. Armor’s testimony that student achievement in Benton Harbor is statistically comparable to state-wide performance, when controlled for poverty. Dr. Armor also studied whether there existed any gaps in the achievement outcomes of black and white children in the BHASD. The goal of his study was to determine the existence of any measurable differences in the performance of white and minority students in the BHASD that could be attributable to past segregation in the district. The black-white achievement gap exists in every school district in the United States, including those never found to have operated a segregated system. Further, the gap is found in children at the beginning of their academic careers. (Tr. 144-45 Armor.) I accept Dr. Armor’s conclusion that after adjusting for SES factors and first grade test scores, no statistically significant gap exists between the performance of white and minority students in the BHASD. (S — 14; Tr. 146-49 Armor.) In other words, no “achievement gap” exists in the BHASD that is attributable to the race of students. Indeed, even without controlling for SES, the achievement gap between black and white students in the BHASD is one of the smallest Dr. Armor has ever studied. (Tr. 155-56 Armor.) Using unadjusted figures, the achievement gap in the BHASD at the fourth grade level is .23 standard deviations in reading and .38 standard deviations in math. At the seventh grade level, the achievement gap is greater, but still low: .46 standard deviations in math and .59 standard deviations in reading. All of these numbers are significantly lower than for districts such as Grand Rapids, which have never been found to have engaged in segregation. (Id.) Plaintiffs’ expert, Dr. Robert Crain, concluded that achievement deficits remain in the BHASD since the time of the remedial order. Dr. Crain evaluated individual transcripts for 1,419 students entering ninth grade for the first time and residing in the poorest census tracts in the BHASD. Dr. Crain divided the students into six year-categories: 1975-77 (graduating class of 1980-81); 1978-83 (graduating class of 1985-86); 1983-86 (graduating class of 1990-91); 1987-1991 (graduating class of 1995-96); 1992-1996 (graduating class of 1999-00); 1997-2000 (not yet graduated). Dr. Crain also examined MEAP scores in each of the years from 1980-81 to 1999-2000. Dr. Crain found a substantial increase in the number of students repeating ninth grade, but no increase or decrease in the rate of graduation. (P 861) However, I note that another of plaintiffs’ exhibits (P 5) shows a decline in the dropout rate between 1997 and 1999 (16.22% to 8.35%), and a significant increase in the graduation rate (51.87% to 69.92%). (P 5.) Going back to 1995, the rate of change is generally consistent, but the trend is interrupted by an outlying peak in 1996, where graduation rates were high and dropout rates low. Dr. Crain found that the percentage of students repeating grades ten through twelve was consistent within and across years. (P-86j.) In addition, he found that grade point averages of graduating students had risen since the first year category. (P-86k.) Dr. Crain also found significant improvement in fourth grade reading and math performance on the MEAP, but a loss of ground by seventh grade. He concluded that the BHASD has failed to close the gap between its seventh grade students’ performance and that of the state average. Dr. Crain’s study, however, faked to control for poverty in assessing student performance on the MEAP against the state average. He also failed to acknowledge that seventh grade scores have begun to improve in recent years. Dr. Crain also noted a decline in the difficulty of courses taken within the math and science curricula by the 1992-96 cohort. He observed that most students had ceased taking a second semester of basic math and science, compared to earlier years. (P — 862.) The court notes, however, that Dr. Crain’s data is unreliable on this point because it fails to account for the intervening adoption of block programming in the BHASD, in which students take a full year of courses in a single semester. (Tr. 678 Garrison.) Dr. Crain also noted that the percent of students having transcripts sent out to colleges was consistent across all year-categories (30%), with only one year-category (year category 3) slightly higher. While Dr. Crain assumed these numbers were lower than other districts, he examined no evidence to support that assumption. No significant change can be identified across the years in the BHASD. (P-86r.) The court finds that Dr. Crain’s findings are partially persuasive. That is, the court finds that, in raw terms, student achievement deficits (as measured by the MEAP) between the BHASD and the state averages have not been eliminated. The court also accepts that no significant change has occurred in graduation rates or college application rates, but the court finds that real growth in GPA has been achieved. The court also recognizes that neither plaintiffs nor defendants have provided the court with a systematic analysis of individual student performance, controlled for poverty, across the years of the remedial order. This deficit results from the failure of the BHASD to retain records of free and reduced lunch eligibility beyond five years. The court accepts that the data collected and relied upon by Dr. Crain and MCIC provides some useful information comparing individual achievement over time within the BHASD. While Dr. Crain does not control for poverty in his analy-ses, the use of specific census tracts for individual data collection provides some control for poverty in comparisons between performance of students over time within the BHASD. It does not, however, provide a control for poverty vis-a-vis state averages, and Dr. Crain attempted no regression analysis to control for poverty in his comparisons between the BHASD and the state average. On balance, the court is persuaded that the information produced by Dr. Crain is insufficient to undermine Dr. Armor’s narrower studies of student MEAP performance, controlled for poverty, for the most recent school years. The court also credits the analysis of defendants’ expert, Dr. John Murphy, who found no lingering effect of past segregation on achievement. Dr. Murphy analyzed student achievement in the BHASD based on the racial composition of each school and found that achievement was somewhat higher at schools with a higher percentage of minority students. (S-77; S — 78; Tr. 854-57.) Dr. Murphy also found that student performance at the schools that the court previously found were “racially identifiable” was higher than performance at other BHASD schools. (S — 79; S — 80; S — 81; Tr. 857-59 Murphy.) Weighing all these findings together, including the passage of more than twenty years, the absence of any segregative conduct during the educational fives of present BHASD students, the overall current performance of students when controlled for poverty, and the extremely low performance difference based on race, I find that no achievement vestige exists that is attributable to the past segregative conduct of defendants. As a result, the question of whether it is practicable to further eliminate student achievement deficits is not reached. While farther efforts to improve student achievement may be practicable, such efforts would be addressed to achievement deficits not caused by past segregative conduct of these defendants. D. Good Faith Compliance with Court Orders Property Transfers. Defendants undis-putedly complied with the order of this court requiring the return of the Eamon residential area from Eau Claire to the BHASD. Defendants also observed the injunction from transferring the Sodus II residential area or any other property area out of the BHASD. (Stip. No.2d, p. 4; J-2.) Elimination of Racially Identifiable Schools. The court also required the BHASD to adopt a student reassignment plan to eliminate racially identifiable schools and to have it in place for the 1981-82 school year. The district complied with this requirement. According to Dr. Rossell’s testimony, in 1981-82, the BHASD implemented a plan achieving racial balance within the court’s goal of 10% of the district average for 82% of its schools. In the 2000-01 school year, 94% of BHASD schools met the 10% goal. (Tr. 502-03 Rossell.) The district further complied with the court’s suggestion of using a uniform grade structure to keep children in the same school building in grades one through four until 1990-91, when it returned to a neighborhood school system with the court’s permission. Magnet Schools and Interdistrict Transfers. The district also substantially complied with the court’s order directing the development of magnet programs to provide greater educational opportunities for BHASD students and to encourage voluntary transfer of students between the BHASD and Coloma and Eau Claire. The district structured several highly successful magnet programs, including the Boyn-ton Montessori Program, the Gifted and Talented Program, the Creative Arts Academy and the McCord Renaissance Program, each of which remained available to students in the three districts. (P-18.) While not drawing as large numbers of suburban students as the court had hoped, the magnet schools in the BHASD achieved a transfer-in rate from Coloma and Eau Claire of 1%, a rate equivalent to the transfer-in rate of the largest interdis-trict transfer programs in the United States. (S — 96; Tr. 517-19 Rossell.) I note that plaintiffs introduced a 1992 study of magnet schools and student transfers conducted by representatives from Eau Claire, Coloma, Benton Harbor, the NAACP and the State of Michigan. (P-108.) The study demonstrated that the implicit goal in the remedial order of 25% out-of-district participation was never met for any Benton Harbor magnet. The report identified ways in which participation might be increased, specifically, providing additional Montessori magnet opportunities and altering the Gifted and Talented magnet. The study reported that Coloma and Eau Claire students were placed on a waiting list despite the 25% goal not having been reached. (P-108 at 6.) The study then continued that a total of 124 students had applied for 40 slots. The study did not, however, provide any numbers in any years of how many Coloma and Eau Claire students were placed on such a list. As a result, while the report suggests that in one year Coloma and Eau Claire students may .have been turned away, it provides little evidence of a significant failure to comply with the court order. In contrast, testimonial evidence of BHASD officials strenuously testified that Coloma and Eau Claire students were not placed on waiting lists and that they were unaware of any such occurrence. (Tr. 731-32 Garrison; Tr. 355 Williams.) In response to this testimony, plaintiffs relied solely on the limited evidence contained in the magnet study, which, as I have observed, did not clearly state how frequently and how many students from Eau Claire and Coloma were placed on waiting lists. The remedial order also required that “[f]ederal funding of programs outlined in this plan should be sought to the maximum feasible levels by the three local school districts, the intermediate school district, and the state acting cooperatively.” 515 F.Supp. at 385. Plaintiffs introduced evidence that the BHASD sought funding of its magnet programs only three times, in 1985, 1986 and 1989. (Tr. 357-58 Williams.) In 1989, the district received a substantial grant to fund magnets. Thereafter, the district was eligible to apply for federal grants but elected - not to do so. (Tr. 359 Williams.) Although Superintendent Williams suggested that grants possibly were not applied for because funding was reduced or the likelihood of getting the grants was so small as to make development of the proposal a wasted effort (Tr. 360 Williams), Williams did not testify as to the precise reason for the failure. In addition, the minutes of the Superintendents’ Council meeting of Friday, December 1, 1995, reflects that the magnet study committee had been resumed but discontinued due to a lack of participation from Benton Harbor. (BH-2.) However, at that same meeting, the district superintendents, including Ms. Williams, recommitted to support this committee. At the subsequent meeting of the council, January 26, 1996, Dr. Staten reported that two meetings of the magnet committee had been scheduled. Study of the magnet programs continued through 1996. (BH-2.) I find that while defendants did not fully comply with the court’s order, either by seeking maximum funding for the magnet programs or by making aggressive efforts to evaluate and improve those programs, defendants nevertheless substantially complied with the court’s orders regarding magnet schools. I therefore conclude that defendants substantially complied in good faith with this aspect of the transfer program. With respect to transfers of black students out of Benton Harbor to Eau Claire and Coloma, defendants implemented and financially supported a substantial inter-district transfer program during the entire period of the remedial order. It should be noted that Benton Harbor achieved the second highest percentage of African American students transferring out-of-district (13%) of any of the largest interdis-trict transfer programs in the country. Only St. Louis exceeded the rate experienced in Benton Harbor. (S — 95; Tr. 513-17 Rossefl.) No evidence suggested that the transfers out of Benton Harbor could or should have been increased. No question exists of the good faith of defendants in implementing this aspect of the program. Curriculum. Defendants presented un-controverted evidence at the hearing that the BHASD has an appropriate and uniform curriculum, which is tied to the state MEAP standards. (Tr. 685-86 Garrison; S-61; Tr. 840-41 Murphy.) The BHASD curriculum includes courses on African American history and multi-cultural studies. (S-210, p. 60.) It is clear that no evidence exists that the curriculum or materials used by the BHASD includes cultural or racial bias or that practices or policies exist that racially discriminate. (Tr. 53-54 Peeples Burks; Tr. 260-62 Williams.) Faculty and Staff Reassignment and Affirmative Action Goals. It was uncontested at the hearing that faculty and staff in the BHASD have been reassigned and are well-integrated throughout the BHASD. No school is racially identifiable on the basis of faculty and staff composition. (S — 255; Tr. 521-25 Rossell.) In addition, the BHASD has met and exceeded the affirmative action goals identified in the remedial order concerning the composition of faculty and staff. (Id.; S-99 through S-101.) In-Service Training. BHASD personnel participated in extensive in-service training and workshops over the years, including workshops geared toward the improvement of student achievement, the desegregation process, diversity and multicultural issues. (S-lll through S — 128; S-150; Tr. 679-81, 686-87 Garrison; Tr. 266-70 Williams; Tr. 740^42 Staten.) The court finds that defendants have complied in good faith with the in-service training requirement of the remedial order. Discipline. In the 1981 remedial order, the court required defendants to adopt a uniform code of conduct and discipline. It is undisputed that such a code was developed and adopted. (BH-12; Tr. 306 Williams.) The record at trial also established that the code is fairly enforced as evidenced by the fact that the percentage of African American suspensions is nearly equal to the percentage of African American student enrollment. (Tr. 530-31 Ros-sell; S — 105; S-106.) The court therefore finds good faith compliance with this aspect of the remedial order. Community Involvement and Monitoring. As part of the remedial order, the court established the Community Education Council (“CEC”) as the “primary body monitoring implementation of the court’s order on behalf of the court.” The CEC was established and funded in accordance with this court’s orders at all times since adoption of the remedial order. Since monitoring implementation of the court’s order was one of the functions of the CEC, the court is aware of the existence of some tensions between the CEC and various defendants during the course of the 21-year remedial period. However, no evidence was presented at the hearing that suggested that defendants had failed to cooperate in good faith with the CEC. (Tr. 936 Williamson; Tr. 1134 Cain; Tr. 772-73 Staten.) Annual Desegregation Progress Reports. The court required the BHASD to annually submit desegregation progress reports. This was done every year through the 1999-2000 school year. (S-111 through S — 128; S-150; Tr. 306-07 Williams.) Achievement and Social Skills. Compliance with this aspect of the court’s remedial order was the subject of a very large percentage of the evidence presented at the unitary status hearing. In the 1981 order, the court determined that the remedy for past segregation in Benton Harbor required implementation of a program designed to improve “self-esteem, achievement motivation, social skills, coping, behavior, and vocational and career awareness in students at all grade levels.” 515 F.Supp. at 370. The court directed the school boards and the school district superintendent, under the direction of Dr. James Comer, to develop, implement and evaluate such a program, which the court ruled was of “critical importance ... to a complete equitable remedy of the adverse effects of defendant’s segregative conduct.” Id. The court retained jurisdiction to issue further orders on this component. The court also determined that [t]he goal of persons working within this component will be to raise the level of student achievement, until achievement within the district reaches the average attained by Michigan students in statewide achievement tests. Once that goal has been achieved, the program will be phased out over two additional school years. 515 F.Supp. at 370. The BHASD worked with Dr. Comer to adapt the Comer plan to the BHASD, sending an administrator, Ms. Erma Mitchell, to Yale University for one year to receive training from Dr. Comer and his staff. (Tr. 1414 Mitchell.) During the course of the program, the State paid for the training of numerous administrators, board members, superintendents and teacher and parent representatives at Yale University. The State also paid for the court and the court master to observe the program in New Haven. (Tr. 1413-15 Mitchell.) At Dr. Comer’s recommendation, the Comer program began to be implemented in only four elementary schools in the BHASD during the 1982-83 school year. (S-368.) The court approved stipulated modifications in the required state funding payments consistent with the staggered implementation. (P-10; P-90a; P-90j; P-90k; P-90Z; Tr. 1465 Stevens.) The BHASD did not attempt to reserve for later use funds originally authorized by the remedial order, though the program coordinator believed those funds were being placed aside for later use when implementation was at its peak. (Tr. 1334 Mitchell; Tr. 1465 Stevens; P-10; P-90j; P-90k.) In 1987, the parties entered into a stipulation, which the court approved, that provided predictable funding for school years 1986-87 through 1991-92. The stipulation provided funding at higher levels than originally mandated by the remedial order, but also provided for the funds to decrease each year after 1988 and to be terminated at the end of the 1991-92 school year. (P-90h; Tr. 1467 Stevens.) In the 1986-87 school year, in anticipation of declining financing from the State under the remedial order, the BHASD superintendent ordered that the program be implemented in all remaining elementary schools in the district. (S-368; S-223; Tr. 595 Gavin; Tr. 1338-39, 1345-47 Mitchell.) The decision was contrary to the recommendation of both the program director, Erma Mitchell, and Dr. Comer. (Tr. 1346 Mitchell.) During that same school year, while the program was being implemented for the first time in all elementary schools, Mitchell was removed from the position of program director. (Tr. 1350 Mitchell.) She was not replaced until after the end of the school year, and then by an individual holding a lower status position who had no say on the management team. (Tr. 1350, 1352 Mitchell; Tr. 606, 612 Gavin.) In addition, Mitchell was directed to have no contact with her successor. It is also undisputed that on July 14, 1992, the BHASD school board voted to terminate the Comer model due to lack of finances. (P-90f: Board Minutes of 7/14/92.) After the program was terminated, the program had no director or monitoring, no further training took place, the social skills component of the program was discontinued and the parent stipend program was terminated. (Tr. 651-654 Gavin; Tr. 1430-31 Mitchell.) In addition, the program was never implemented at the junior high and high school levels. (Tr. 624 Gavin; BH-38.) Although defendants assert that they continued the Comer philosophy after July 14, 1992, by compliance with Public Act 25, the evidence does not support that contention. The Comer program requires all of the elements for its effectiveness. (10/24/01 Tr. 17-28, 81 Comer.) Compliance with Public Act 25 is not equivalent to the Comer program, though a school implementing the Comer program will comply with Public Act 25. (BH-29; Tr. 594-97, 625-34, 650 Gavin.) I find, based on all of these facts, that the BHASD did not fully comply with the achievement component of the court’s order. Indeed, the school board appeared to pay little attention to the ultimate goal or method of its achievement obligation under the remedial order. Gladys Peeples Burks, director of BHASD’s desegregation advisory services in 1982-83 and district board member from 1993 to present, stated that she did not become aware of the remedial order’s goal for student achievement until two weeks before the hearing on unitary status, when she read the order for the first time. (Tr. 37-38, 59-60, 71 Peeples Burks.) She testified that during her eight-year tenure on the BHASD school board, the board has not discussed the goals of the remedial order. (Tr. 39, 70, 72 Peeples Burks.) Even at the time of her testimony, Peeples Burks did not believe the BHASD was under direction to continue the Comer program until MEAP test scores equaled state averages. (Tr. 73-74 Peeples Burks.) Yet in 1987, then-superintendent of the BHASD, Harry Stephens, received a letter from court master Michael Stolee advising that any program that might interfere with or replace the Comer model would have to be approved by the court. (P-40.) Further, no evaluations were conducted by the district of compliance with the remedial order. Despite the above, I find that the BHASD has made substantial and continuous efforts to improve student achievement notwithstanding this lack of attention to the specific method of the remedial order. (Tr. 39-50 Peeples Burks; Tr. 311-12 Williams; Tr. 658-700, 731-32 Garrison.) The evidence is clear that while the district abandoned the Comer model as a whole, board members and administrators actively sought to improve student achievement in the BHASD with the understanding that the remedial order required such efforts. (Tr. 256-57, 265-302, 312 Williams; Tr. 38, 40, 72 Peeples Burks; Tr. 731-32 Garrison.) Recently, those efforts included working extensively with plaintiffs’ expert Dr. Stephens to develop a comprehensive school improvement plan, 45% of which has been implemented despite the lack of specially dedicated funding. (Tr. 659, 662-63 Garrison.) In addition, it is apparent that the program was not terminated in willful contempt of the court’s order. While the order clearly states the requirement that the student achievement component be implemented until BHASD MEAP scores reached the state average, the evidence does not indicate that all BHASD officials fully appreciated that fact. Further, the memos produced at the time of the discontinuation of the program, as well as the board minutes, suggest that the district believed, whether reasonably or not, that it could continue to utilize the Comer philosophy without the formality of the program. (S-162.) Moreover, while it always remained the BHASD’s primary responsibility under the order to implement and assure compliance with the order, 515 F.Supp. at 383, it appears that although Dr. Stolee and James Turner of the CEC initially questioned the district’s actions (S — 163), neither Dr. Stolee, the CEC, nor plaintiffs’ counsel ever sought to follow up their early concerns until the motions for unitary status were filed years later. Finally, the BHASD appeared to consider the termination of special state funding for the component, as approved by the court, as tacit approval to not fund the program. Again, however unreasonable this assessment may have been, the court concludes that the decision made by the BHASD to terminate the Comer plan was not taken in bad faith. I further find that the State of Michigan has fully complied with its obligations under this component of the remedial order. It was undisputed that the State fully funded all aspects of the achievement component as directed in the original and subsequent orders. (Tr. 1601 Stevens; Tr. 321-22 Williams; Tr. 1407 Mitchell; S-207; P-90j; P-90k; S-227; S-368; S-402.) The State’s funding obligation for this component ended in 1992, pursuant to a stipulated order of this court issued in 1987. (S-368.) In addition, the court finds that a substantial number of state officials and department staff placed great emphasis on effective implementation of the order and promoting achievement in the BHASD, fully implementing both the letter and spirit of the order. Dr. Paslov, Dr. Donovan and later Dr. Teressa Staten, all top officials of the Department of Education, routinely chaired the Superintendents Council meetings. (Tr. 737-740 Staten.) Indeed, Dr. Staten continued in this role following her retirement. Dr. Staten traveled regularly to Benton Harbor and consulted on the telephone on issues involving the BHASD. (Tr. 747-49 Staten.) Dr. Staten testified at length about additional efforts undertaken by the State beyond funding the court-ordered remedy. (Tr. 761-64, 767-69.) In addition, Dr. Eugene Cain testified to the large number of equity initiatives the State pursued in Benton Harbor, making efforts to see that Benton Harbor, in particular, was one of the beneficiaries of various competitive State programs. (Tr. 1118-30 Cain.) Dr. Cain also testified to being influenced by a meeting that occurred shortly after he began working at the Department of Education, in which the State Superintendent of Instruction, Dr. Runkel, discussed the agency’s responsibilities in Benton Harbor and his belief that the State should go beyond the court order. Dr. Cain expressed his own abiding concern about the district arising out of his personal experiences attending segregated schools in Birmingham, Alabama. (Tr. 1116-17,1132 Cain.) Further, the State has continuously provided a variety of technical services to the BHASD particularly, and to other districts with large numbers of disadvantaged students. These services include technical assistance such as individual coach/mentor visits to the BHASD schools, consultation with individual school leaders on the creation of school improvement plans, grant awards to individual schools in the BHASD, other discretionary grants, support for early childhood activities, assistance in the area of school health programs, and cash awards for school improvement initiatives. (S-292; S — 269; S-270; S-263 to S-265; S-293; S-266; S-268; S-276; S-289; Tr. 916-31 Williamson.) III. CONCLUSIONS OF LAW “The duty and responsibility of a school district once segregated by law is to take all steps necessary to eliminate the vestiges of the unconstitutional de jure system.” Freeman v. Pitts, 503 U.S. 467, 485, 112 S.Ct. 1430, 118 L.Ed.2d 108 (1992). This obligation has been phrased as a duty to convert the prior dual system to one that is unitary, in which “ ‘racial discrimination would be eliminated root and branch.’ ” Freeman, 503 U.S. at 486, 112 S.Ct. 1430 (quoting Green, 391 U.S. at 437-38, 88 S.Ct. 1689). The term “unitary” is not a precise concept. Instead, the concept is used generally to describe a school system that has been brought into compliance with the requirements of the Constitution. Id. at 487, 112 S.Ct. 1430. The elimination of the effects of the prior violation is the goal and limit of the court’s equity power. See generally id. at 487, 112 S.Ct. 1430 (“The essence of a court’s equity power lies in its inherent capacity to adjust remedies in a feasible and practical way to eliminate the conditions or redress the injuries caused by unlawful action.”). The concept of a “unitary” system is to “guide district courts in assessing when it is appropriate to restore local control.... ” Manning v. School Bd. of Hillsborough County, 244 F.3d 927, 942 (11th Cir.2001). In imposing a remedy and determining whether a unitary district has been established, the Court has identified six important indicia of a racially segregated school system: “student assignments, ... faculty, staff, transportation, extracurricular activities and facilities.” Id. (quoting Green, 391 U.S. at 435, 88 S.Ct. 1689). Those six factors, the “Green factors,” have been identified by the Supreme Court as deficiencies considered to be per se vestiges of discrimination. Id. The Court has acknowledged, however, that Green does not set forth the only factors that may be implicated by segregative practices. See Jenkins III, 515 U.S. at 88-89, 115 S.Ct. 2038 (“[W]e have identified ‘student assignments, ... faculty, staff, transportation, extracurricular activities and facilities’ as the most important indicia of a racially segregated school system.”) (emphasis added). Indeed, the Court specifically has recognized that disparate student achievement and other ancillary effects in addition to the Green factors may properly be attributable to past de jure discrimination. Id. at 90, 115 S.Ct. 2038; see also Milliken v. Bradley (“Milliken II”), 433 U.S. 267, 97 S.Ct. 2749, 53 L.Ed.2d 745 (1977). In determining whether a school district previously found to have engaged in de jure segregation has achieved unitary status, the ultimate question before the court is “ “whether the constitutional violator ha[s] complied in good faith with the desegregation decree since it was entered, and whether the vestiges of past discrimination ha[ve] been eliminated to the extent practicable.’ ” Missouri v. Jenkins, 515 U.S. 70, 89, 115 S.Ct. 2038, 132 L.Ed.2d 63 (1995) (quoting Freeman v. Pitts, 503 U.S. 467, 492, 112 S.Ct. 1430, 118 L.Ed.2d 108 (1992), and Board of Education of Oklahoma City Public Schools v. Dowell, 498 U.S. 237, 249-50, 111 S.Ct. 630, 112 L.Ed.2d 715 (1991)). It is well established that defendants bear the burden to proof on their compliance with remedial orders and on the existence of any vestiges of the Green factors. See Freeman, 503 U.S. at 494, 112 S.Ct. 1430. In addition, this court previously has held that student achievement vestiges should be treated like other indicia of a segregated school district. See Berry v. School Dist. of City of Benton Harbor, 141 F.Supp.2d 802, 810 (W.D.Mich.2001). Therefore, as with the Green factors, defendants bear the burden of showing that any continuing student achievement deficits are not vestiges of their prior segre-gative conduct. Id. A. Good Faith Compliance As previously discussed, the court’s initial remedial order and various supplementary orders included a variety of components, including the elimination of racially identifiable schools within the BHASD on all of the Green factors, the development of magnet schools and interdistrict transfers, as well as various means of assuring the success of those components such as community involvement, monitoring and financing. In addition, the court ordered an achievement and social skills component, directing the defendants to implement the Comer Program throughout the district. The court stated that the “goal of persons working within this component will be to raise the level of student achievement, until achievement within the district reaches the average attained by Michigan students in state-wide achievement tests.” 515 F.Supp. at 370. The student achievement component was to be phased out over two additional school years if this standard had been met. Subsequently, in 1987, in accordance with a stipulation of the parties, the court modified the order with respect to the funding of the achievement and social skills component. (Dkt. #882; P 90h.) The order altered both the timeline for full implementation of the Comer program and the funding formula for the program. Under the stipulation and order, the State’s obligation to provide full funding for the program-was terminated after the 1991-92 school year, at which time the Comer model was expected to be fully internalized in the system. The order, while fully and finally resolving the State’s obligations to pay for the component, did not alter the measure set in the original order for terminating the program. As I previously concluded in my findings of fact, the BHASD did not continue the Comer program after the 1991-92 school year, when specific funds from the State were terminated, despite the fact that the district at no time met the goal that district MEAP averages would meet those of the state. In the absence of another order of the court, this failure amounts to facial noncompliance with the court order. However, I note that the remedial order did not expressly state that the BHASD would be barred from unitary status unless the Comer model was continued until the district MEAP averages met state-wide averages. The order stated that the “goal of persons working within this component will be to raise the level of student achievement, until achievement within the district reaches the average attained by Michigan students in state-wide achievement tests.” 515 F.Supp. at 370 (emphasis added). The court’s phrasing was not accidental. A “goal” is not an imperative. The phrasing followed the court’s reiteration of its liability findings to the effect that student achievement deficits were directly linked to past discrimination and achievement remedies were required to overcome the effects of the past segregation. Id. at 368. In fact, the remedial order reserved its authority to make additional orders with respect to the component “as required to fully effect a complete, equitable remedy.” Thus, the order implicitly recognized the possibility that the goal would not be met prior to termination of court jurisdiction, if achievement deficits were no longer linked to past liability. As a result, as I have held, the BHASD did not fully comply with the requirements of the order with respect to the Comer program. That failure to comply, however, does not itself demonstrate bad faith that necessarily bars a grant of unitary status. Indeed, perfect compliance with the court’s remedial orders is not required for a constitutional violator to be released from judicial oversight. When examining evidence of compliance to determine “whether a school board has acted in good faith, a court should not dwell on isolated discrepancies, but rather should ‘consider whether the school board’s policies form a consistent pattern of lawful conduct directed to eliminating earlier violations.’ ” Man