Full opinion text
MEMORANDUM OPINION LIMBAUGH, District Judge. It was the best of times for the St. Louis city school system. The coveted Triple AAA rating bestowed by the Missouri Department of Education has again been received. A quality education is generally provided for all the city children attending public schools. A sound faculty of teachers and principals are dedicated to use their utmost talent in developing the skills of their students. Parents and friends are caring more about the children and are beginning to work with the educators in the business of youth education. The breadth of educational opportunity is unlimited. An impressive black principal of an all black primary school and his remarkable first assistant stated “our goal is to make certain these children learn to communicate and when they leave our school for middle school, they can.” A white principal of a mixed middle school remarked that “these students are now being exposed to sophisticated educational experiences and they will be equipped to take their place in this competitive society.” A high school principal observed that “our teachers are back to the basics with their students, but a huge variety of other educational opportunities exists.” Throughout the system, remedial programs are administered to those students with learning disabilities and devoted teacher after teacher provide tender care for these youngsters giving them “something to hang on to.” Enrichment programs are evident and are receiving increased attention. It is amazing to enter a room in which a dozen or more fifth graders are each performing an assignment on their Apple-type computers. An air of friendly but strict discipline pervades the system. Respect is being instilled. When the principal enters the room in some delightful primary schools, the children rise and acknowledge in animated voices, “Good morning Ms._” There is great pride among the middle and high school students. Their excellent art work is everywhere. They are striving to keep their schools clean and presentable. Custodial staffs are attentive to their duties. School spirit abounds. In many difficult situations, racial problems are alleviated and the number and intensity of racial incidents remains low. Parents and relatives are coming to school meetings and are working with the teachers in seeking the best educational interests of the children. On Saturdays and after hours, some parents can be seen with the children painting the walls of school buildings with paint and equipment they provide themselves. Vocational education is virtually unlimited. Students can receive special training in every field imaginable. A young black female has just finished a difficult welding assignment done with a highly sophisticated computer system. Others are in aviation mechanics or are using all trades to build a house (within the school confines), or are developing beautician skills or a host of other things. A middle school has its own radio broadcasting network. Teachers, principals and administrators, generally, have worked diligently to promote a sound integrated school system and much progress has been made. It is a good school system. But, it was also the worst of times for the St. Louis city school system. The schools have not received major repairs in the memory of most of the staffs. Roofs leak in over half the schools. The leaks receive only temporary attention. In classroom after classroom, in gymnasiums, in libraries and study halls and in cafeterias, water is everywhere. It drips from the ceiling, down the walls and even from light fixtures. Cans, buckets and other receptacles are all over. A sixth grader in a reading class leans over in her chair to avoid the steady drip of water going into a bucket at her feet. Some of the plumbing is intolerable. On one occasion in a school when the water was flushed from a urinal, portions came down a wall in the room below, while a devoted teacher was attempting to teach her students in that room. Ceiling tile in many rooms no longer exists or is so permeated with water that it hangs perilously. Plaster falls to the floor sometimes placing the student or teacher in some danger. Paint peels from many walls and exposes the plaster or wall board and sometimes the studs. Many buildings are old and dilapidated and were designed for education seventy-five years ago. Some have no gymnasiums. Others have gymnasiums, but the ceilings are only seven or eight feet high. In many schools, a student must bend or duck going from some rooms to others to avoid hitting exposed pipes or mechanical supports. In the face of this deplorable and impossible situation, those in charge use ban-daids to cover gaping wounds and, with some hypocrisy lament the lack of funds to conduct a necessary structural overhaul of the city school buildings as being the sole basis of their inaction. And, too, without addressing them, the school system has its other problems and deficiencies as do all metropolitan schools. This opinion, therefore, addresses the capital improvement problems of the city integrated and non-integrated schools as well as those of the existing magnet schools and those which may reasonably be needed in the foreseeable future. Although this case was filed in 1972, the physical condition of the schools was not considered until the so-called Settlement Plan was approved and later accepted by the Court. No attempt will be made here to recount the history of the events lead-ings to the enactment of the Settlement Plan as it has been adequately reviewed in Liddell v. Board of Education of City of St. Louis, 567 F.Supp. 1037 (E.D.Mo.1983), and Liddell v. State of Missouri, 731 F.2d 1294 (8th Cir.1984). The Settlement Agreement basically was an attempt to address previous determinations of this Court and of the United States Court of Appeals for the Eighth Circuit, that the defendant Board of Education of the City of St. Louis (City Board), and State of Missouri defendants were liable because of their establishment and maintenance of a racially segregated public school system within the City of St. Louis. The remedy for the liability was ordered in 1980 by this Court, (Meredith, J.), Liddell v. Board of Education, 491 F.Supp. 351 (E.D.Mo.1980). The order directed the implementation of a mandatory, intradistrict desegregation plan within the City of St. Louis public school system, and the development and submission of plans involving voluntary interdis-trict transfers between suburban school districts and the St. Louis city school district. Following the 1980 order, the Court approved two interdistrict transfer plans (the 12(a) voluntary plan and the 12(b) vocational education plan). The 12(c) Settlement Plan was approved in 1983 in Liddell v. Board of Education of City of St. Louis, 567 F.Supp. 1037 (E.D.Mo.1983). This was done after a fairness hearing in which all parties had complete opportunity to present evidence and the point of view of each. The physical condition of existing schools following the Settlement Agreement (Settlement Agreement and Settlement Plan are used interchangeably), was first addressed by the appellate court in Liddell VII, 731 F.2d at 1318 (decided February 8, 1984). As noted in Liddell VII, p. 1318, the Settlement Agreement suggested the age and condition of the city schools. The Court in Liddell VII, at p. 1318, also observed that over the years, the school board has attempted unsuccessfully to raise funds to provide for school improvements. In approving the 12(c) Settlement Plan, Liddell, supra, 567 F.Supp. 1037, the District Court (Hungate, J.) ordered that: (b) The City Board shall submit to its voters, on or before February 1, 1984, a proposed bond issue of an amount determined by the City Board as sufficient to meet those of its capital improvement needs as are deemed necessary to meet its constitutional obligation to desegregate the city’s public schools; (and) (c) Should that bond issue fail to obtain the two-thirds majority vote required by state law, the court will consider an appropriate order to obtain the funds deemed sufficient to meet the capital improvement needs of City Board in complying with its constitutional obligation to desegregate the city’s public schools. Liddell supra, 567 F.Supp. at 1056 and Liddell VII, supra at 1318. Following the District Court’s order, the City Board formulated a building program with a total cost of $127,000,000.00 with one-half of the total to be financed by the issuance of $63,500,000.00 in City Board bonds. Ostensibly, the other one-half was to be paid by the State of Missouri. The bond issue was presented to the voters on November 8, 1983 and 55% of the voters approved the issue. Eighty-four percent of the voters in the predominantly black wards voted for the issue, but 65% of the voters in the predominantly white wards voted against it. The bond issue was defeated because it failed to receive a two-thirds majority. Liddell VII, supra at p. 1319. Following the February 8, 1984 Liddell VII decision a bond issue identical to the one defeated on November 8, 1983 was again submitted to the voters on June 5, 1984. It, too, was defeated with a voting pattern similar to that in the 1983 election. The Court of Appeals for the Eighth Circuit specifically held in Liddell VII, supra, at p. 1319 “that the State had an obligation to pay one-half of the costs of the capital improvements program necessary to restore the city facilities to a constitutionally acceptable level ...” The Court of Appeals in Liddell VII, supra, at p. 1319 also directed, On remand, therefore, the City Board should promptly identify the projects to be undertaken, estimate the cost of each project, and set a reasonably detailed schedule for the completion of each project. The projects having the highest priority must be scheduled for completion at the earliest possible date. To that end, the City Board should consider the desirability of a referendum on a bond issue which can be initiated at a very early date, and subsequent bond issue for those projects to be built in later years. The State will pay one-half of the cost of preparing the detailed plans and schedules. As soon as the City Board has prepared the new plans, estimates and schedules, it shall submit them to the Budget Review Committee, discussed infra Section VI, and then to the district court. When the District Court has approved them, a new bond issue shall be submitted to the voters. If it is defeated again, the District Court shall determine how the improvements shall be funded. See infra Section V. Following the decision in Liddell VII of February 8, 1984, the District Court directed the United States Magistrate to conduct an evidentiary hearing and recommend the legal standard to be applied in ascertaining the desegregation-related costs of a capital improvements program for the schools operated by the Board of Education in the City of St. Louis. After extensive hearings, United States Magistrate Noce filed his report and recommendation on October 17, 1984, H(3422)84. Although a standard was not set out with specificity, Magistrate Noce noted that there ought to be no distinction in the basic quality and educational usefulness of the capital facilities of the non-integrated schools, the integrated schools, the magnet schools, or any capital asset necessary to spport any remedial program. Districtwide, the capital facilities must be brought to a uniform minimum level of educational utility. No specific order of the District Court was ever entered approving or disaffirming the findings of the Magistrate. During the same period of time, the City Board determined that professional assistance would be needed in the capital improvement area to follow the mandate of Liddell VII, so as to enable it to “identify the projects to be undertaken, estimate the cost of each project and set a reasonably detailed schedule for the completion of each project.” Liddell VII, at 1319. The professional assistance needed would include engineers and architects who could survey the approximate 150 schools in the system and make recommendations to the City Board and to the Court. While discussions were being carried out between the State and the City Board concerning the procedure to be followed to employ professional assistance, Judge Hungate was relieved of the responsibility of handling this case and on February 1, 1985, Judge Limbaugh was appointed as the successor to Judge Hungate. On March 20, 1985, the State and the City Board reached an accord and agreed to prepare jointly an invitation to submit bids to employ architects and engineers to make the necessary evaluation of the capital improvement problem. The parties agreed to share the cost equally for the preparation of detailed plans and schedules. L(122)85. On April 15, 1985, the Court approved the agreement of the State and the City Board and ordered the parties to file a joint report on the status and findings of the facilities study no later than August 15, 1985. L(160)85. During this period, the United States Court of Appeals for the Eighth Circuit rendered Liddell VIII on March 26, 1985. Liddell v. Board of Education of City of St. Louis, 758 F.2d 290 (8th Cir.1985). That Court in addressing the capital improvement problem determined that “the district court shall on remand direct the City Board to prepare promptly a building program, not to exceed $40,000,000.00 to meet the most urgent capital needs of the district, with emphasis on the all-black schools and the magnet schools.” The appellate court then directed the district court to conduct a hearing in order to determine that all projects in the program are necessary to cure constitutional deficiencies in the City Board’s facilities. If the district court determined that question in the affirmative, it was directed to order the City Board to submit a bond issue of up to $20,000,000.00 to the voters which was to be matched by the State. If the bond issue failed, then the appellate court stated that “the district court shall enter a judgment against the City Board in the amount determined and in an equal amount against the State.” The appellate court determined further that “we recognize the need for long-term planning, and that should proceed, but the needs of the all-black schools appear to be so obvious that they should be met at the earliest possible time.” The Court then went on to address further suggestions as to a capital improvement plan as to new magnet schools authorized under the Settlement Agreement. The appellate court did not explain the basis on which it had arrived at a total of $40,000,000.00 but apparently, it was calculated on the basis of a recommendation of the financial advisor that a bond issue of $20,000,000.00 would have continued the debt service levy for the City Board at a rate which was required to retire a 1964 bond issue in 1984. Liddell VIII, at 302. The decision in Liddell VIII caused some concern as well as confusion to the new Judge who had just been assigned the 13-year old complex litigation for a period less than two months and to the parties as well. At that point, the parties had agreed on the mechanics for obtaining professional help to determine the proper method for meeting the criterion in the capital improvements area of all of the city schools within the guidelines of Liddell VII. It finally became apparent that the mechanics and the cost for preparing a building program for $40,000,000.00 to meet urgent capital needs of the district with emphasis on all-black schools and the submission of the program to the Budget Review Committee, the holding of a hearing for the purpose of determining that the projects in the plan are necessary to cure constitutional deficiencies and the ultimate submission of a bond issue for one-half of that cost, would require not much less effort than if the planning was directed to the entire school system. After consultation with counsel for the parties, it appeared more feasible to attack the problem as a whole for it would be a duplication of effort and cost to plan for urgent needs utilizing an uncertain sum of $40,000,000.00, all the while duplicating this effort with a long-range plan. Accordingly, this Court read all of the opinions of the Court of Appeals in pari materia and determined that the proper function of the district court would be to urge the planning to continue to apply to the school system as a whole as this plan would produce costs obviously in excess of $40,000,000.00 and as the framework for the payment of these costs had been established by the spirit and intent of Liddell VII and the other decisions of the appellate court. The engineering firm of Zurheide-Herr-mann in St. Louis, Missouri was employed and it began the facility study of all of the City Board schools. This study was necessary for long-range planning with respect to all schools and would have been equally necessary in addressing the specific $40,-000,000.00 program set out in Liddell VIII. Obviously, a study of all the schools must be made before it could be determined what “the most urgent capital needs of the district, with emphasis on the all-black schools” were. As the Zurheide-Herrmann study progressed, the lack of cooperation of the City Board and the State became evident, as was noted in an order of the District Court on August 30, 1985. L(446)85. In that order, the District Court directed that after the Zurheide-Herrmann facilities study is filed, the Court would hear oral arguments and accept briefs regarding the issue of the applicable standard to be utilized in an ultimate determination as to which capital improvements are desegregation-related. After the Court determined the applicable standard, the parties were then directed to propose a detailed capital improvement program pursuant to the standard and those needs addressed by the Zurheide-Herrmann study. The Budget Review Committee was to consider the proposal and the arguments of the parties, and then report its findings to the Court and all parties were entitled to provide input for the benefit of the Court in making an ultimate decision. This Court announced its ambitious plan to have a proposal to submit to the voters with respect to a bond issue in early 1986. On September 13, 1985, in response to requests by both the City Board and the State, the Court stayed the time requirements relating to capital improvements because of the requests by the City Board to propose a bond issue of $155,300,000.00 on the November 5, 1985 ballot. At that point, it was represented that the city administration and other influential entities were prepared to support this issue which the City Board felt would be adequate in and of itself to address the overall capital improvement needs of the city school system. As the Zurheide-Herrmann study had still not been completed, and as nothing could proceed further without the benefit of that study, the District Court stayed the original schedule. The November 5, 1985 bond issue of $155,300,000.00 was defeated with 56.2% approving the issue when 66% was required. Only 33% of the registered voters voted and in the predominantly white wards 31% voted in favor of the issue and in the predominantly black wards 88% voted in favor of the issue and 74% voted in favor of the issue in the mixed wards. Only 26% of the registered voters voted in the predominantly white and black wards, and 36% of the registered voters voted in the mixed wards. On November 14, 1985, shortly after the election, the Zurheide-Herrmann study was filed. It contained over 50 pages of recommendations with well over 6,000 pages of supporting data. After the study was filed, the Court addressed the schedule which had been on hold, pending the election. On January 10, 1986, a standard was established for determining what capital improvements are desegregation related. L(680)86. In setting out that standard, the Court considered the findings of Magistrate Noce on October 17, 1984, H(3422)84, the briefs and arguments of all of the participating parties, all other relevant information in the case file and the opinions of the Court of Appeals involving this case. This Court noted that “capital improvements are part of the remedial effort to eliminate the effects of a dual-system. The purpose of the total desegregation plan is to remedy the vestiges of segregation and to attract and maintain white students. The goal is to provide ‘quality, integrated educational opportunities to black and white students alike.’ ” Liddell VIII, at 304. This Court continued by finding that “everyone agrees that the physical condition of many city schools is in a deplorable state. Thirty years of neglect by the City Board cannot be condoned; however piecemeal repairs will not promote the desegregation effort. Equal educational opportunity mandates that district-wide, capital facilities must be improved. There are limitations in achieving a uniform minimum level of educational utility.. The city schools must meet the basic requirements of safety and health standards in order to enhance the appeal of the city schools, especially to suburban white students. Additional renovation work must be necessary for the implementation of the approved 12(c) programs.” The Court then set the standard, perhaps in an simplistic way, by holding that “sum-rnarily, desegregation-related capital improvements are those expenditures necessary for the city schools to fulfill the bare minimum requirements of the St. Louis city safety and health codes, and to provide the facilities necessary to implement the 12(c) programs.” This Court then directed the City Board to propose a desegregation-related capital improvements plan without regard to a fixed monetary amount. The City Board was further directed to give top priority to the non-integrated schools and the intra-district magnet schools, to have a school-by-school breakdown similar to the Zurheide-Herrmann report format, to consider demographic trends and all other areas which would conform to the standard announced. L(680)86. Again, the enormity of this directive was obvious. It was equally obvious that a same or similar type of directive would have been necessary to carry out the specific $40,000,000.00 mandate set out in Lid-dell VIII. In addition, it became apparent that substantial time would be needed to meet the District Court’s directive, but the end result anticipated was thought to overcome the necessary delay. Accordingly, the City Board was to comply with the District Court’s order, L(680)86, on or before May 1, 1986, and all interested parties were to respond to the proposal on or before May 16, 1986. L(734)86. In this Court’s order, L(680)86, certain guidelines for the capital improvements plan were stated. The guidelines were to be met with the data collected in the Zu-rheide-Herrmann facilities study. These guidelines suggested that top priority be given to nonintegrated and intra-district magnet schools, that consolidation of enrollments be considered with closing of small schools, that there be rehabilitation of buildings presently in operation only, that low priority be given to schools which probably should close within the next five years, and that plan standards be in compliance with minimum local, state and federal health and safety requirements and codes. In response to the order, L(680)86, the City Board filed its so-called “680 Plan” on May 1,1986. In general, the plan provided for the rehabilitation of 127 buildings for approximately $145,000,000.00. The City Board plan used as its source material the Zurheide-Herrmann report, as well as the so-called “Billingsly Report” and the “Win-sor Model”. The Billingsly Report forecasts four population scenarios for the City of St. Louis. The Winsor Model is an educational prototype used to formulate plant-classroom capacity and sets standards of educational requirements for physical improvements. The 680 Plan is based upon a projected enrollment in the City of St. Louis in 1990 of 53,399 students. This number is somewhat misleading, as it fails to account for the 10,000 approximate transfer students which could serve to reduce the resident student population to 43,399 persons. The 680 Plan of the City Board also utilizes a lower pupil-teacher ratio than that required by Liddell IX, which will be addressed later, or those set out in AAA requirements. Finally, the Winsor Model prototype allows for extremely generous space allocations. Otherwise, the 680 Plan basically adheres to the guidelines set forth in L(680)86. There were various responses to the City Board’s 680 Plan by the parties, some of which were favorable and others of which were not. The Court finds that the projected population figure of the 680 Plan is, in fact, flawed because the student population estimate did not account for the 10,000 students transferred from the City to the County, and the pupil/teacher ratios conflict with the directives of Liddell IX. The Winsor Model is impractical when applied to an old urban school district. Unfortunately, the Court finds that certain data in the Zurheide-Herrmann Report is flawed. There are some serious inconsistencies in the unit costs which will be addressed later. In addition, the recommendations of the 680 Plan, based upon Zurheide-Herrmann data can be considered extravagant. For illustration, suggestion was made that all boilers in a building which were old should be replaced whether or not they were in good working condition. In addition to the 680 Plan calling for an approximate $145,000,000.00 expenditure, the City Board also projected a long-range facilities plan whereby all schools in the system should be rehabilitated in accordance with the data set out in the Zurheide-Herrmann Report for $420,000,000.00. This figure included 10% for architect costs, as well as a 5% built-in inflation cost. At least 150 buildings were involved in this proposal. The State of Missouri chose not to analyze the Zurheide-Herrmann data or the City Board’s 680 Plan, and instead, it proposed an alternative in response to the directives of L(680)86. The State’s plan, L(1047)86, assumed that the 1985-1986 enrollment of city students would remain stable, and then projected a 1990 enrollment of 42,298 persons. This reflected the magnet school enrollment and the city to county transfer students. The State plan rejected the Winsor Model, and used AAA standards in calculating the classroom capacity required. The State proposed that only 67 schools be rehabilitated for a cost of approximately $44,000,000.00. The State’s plan did utilize Zurheide-Herrmann unit costs, however, it eliminated certain projects which it considered to be non-desegregation related. These involved energy efficient items, accoustical ceilings, fluorescent lighting, landscaping, removal of graffiti, handicap-accessibility items in the form of elevators, etc. The plan also did not provide for fees for architects, engineers or contingencies. During the late summer and early fall of 1986, the Court conferred with representatives of the State and the City Board for the purpose of attempting to reach some accord as to various aspects of the problem. Thereafter, the Court was prepared to address the issues and make a final determination. During this time, the Court of Appeals for the Eighth Circuit rendered an opinion referred to as Liddell IX, 801 F.2d 278, which was issued September 5, 1986. In that opinion, the Court noted that several bond issues had been submitted to the voters and rejected. The Court then directed “that upon receipt of this opinion, the Board of Education of the City of St. Louis shall prepare a capital improvement program for the non-integrated elementary and middle schools which is estimated to cost not more than $40,000,000.00. It shall provide for the replacement or complete rehabilitation of the buildings and grounds of those non-integrated elementary and middle schools which (1) are in the worst physical condition and (2) are so located that the best available demographic studies indicate that they will continue to serve students for the foreseeable future. This program should be presented to the district court within thirty (30) days of the receipt of this decision. (The parties have advised this Court that a comprehensive study and report of repairs to the entire school system has already been accomplished.) Thereafter, the District Court shall promptly either approve the program as submitted, or make such modifications in the program as it feels necessary to carry out the mandate of this Court. At the earliest possible date after such approval, a bond issue for the sum of $20,000,000.00 shall be submitted to the voters of the City of St. Louis. If the voters fail to approve the issue by the required majority, the District Court shall follow the course set forth in this Court’s en banc opinion in Liddell VII, 731 F.2d at 1323, as reiterated in Liddell VIII, 758 F.2d at 302.” This directive in Liddell IX obviously created a substantial dilemma to the District Court as well as to the parties involved. The parties had spent $800,000.00 for the Zurheide-Herrmann study, the District Court had established a standard for the parties to follow and various plans had been submitted in response to the standard and the data in the Zu-rheide-Herrmann Report. At the time of the issuance of Liddell IX, the District Court was in the process of preparing an order which would address the capital improvement program for the entire city district and the proper methods of financing the project. Liddell IX, on the other hand, gave specific directions for handling the matter on a piecemeal basis. On October 6, 1986, in response to Lid-dell IX, the City Board filed a $40,000,-000.00 capital improvement site selection plan. The plan envisioned renovation costs of seven middle schools, and eight elementary schools, which were non-integrated and the data used as to the cost was that included in the Zurheide-Herrmann study. Other parties responded to the City Board’s plan, L(1093)86, and the dilemma was then placed squarely in the District Court. This Court reasoned that even the specific mandate of Liddell IX could not be carried out with any more rapidity that the overall school plan, the solution for which was also before the District Court. Accordingly, the District Court, on October 17, 1986, directed that an evidentiary hearing be held on December 8, 1986, for the purpose of having available for testimony and questioning representatives of Zu-rheide-Herrmann and others who have been instrumental in collecting and preparing the data and the material in the various plans before the Court. This Court stated in the order, L(1113)86, that the information needed to make a final determination as mandated by Liddell IX, as well as the other decisions of the appellate court, is contained in the Zurheide-Herrmann facilities study, L(598)85, City Board’s desegregation related capital improvements plan, L(852)86, the City Board’s long-range facilities plan, L(857)86, the State’s response to these documents, L(1047)86, and the City Board’s $40,000,000.00 capital improvement site selection plan, given in response to Liddell IX, L(1093)86. The evidentiary hearing was held when scheduled and at the beginning of 1987 the District Court began the task of making a final resolution of the capital improvements problem, keeping in mind the directives of all of the decisions of the Court of Appeals. Beginning on April 22, 1986, and on sporadic occasions since that time, the Court personally visited representative schools in the system. At the time this opinion is written the Court will have been in and examined 43 of the schools involved. These include representative schools in the primary, middle and high school level and integrated, non-integrated and magnet schools. During these visitations, the Court has examined the physical facilities, observed classroom teachers and students and had personal conversations with principals and members of their staff, teachers, students and maintenance staff personnel. In addition, the Court has observed at least 20 other schools from the outside without an intimate examination of the interior. The Court has also directed its financial advisor, Warren M. Brown, Chairperson of the Budget Review Committee (BRC), and his associate Jay Moody, to gather certain material and data for the benefit of the Court from the Zurheide-Herrmann report and other documents on file in this case. In particular, Brown and Moody were directed to inspect all of the schools in the system, analyze the Zurheide-Herrmann Report, project enrollment and consider budgetary requirements for capital improvements. The Court has carefully reviewed all of the appellate decisions in this case, as well as the District Court findings concerning capital improvements, the City Board proposals as well as those of the State of Missouri and the responses of all of the parties to those proposals, the evidence adduced at the capital improvements hearing, and all other pertinent matters in the file of this case relating to capital improvements. With that review and after the Court’s personal examination of representative schools and after conferences with Warren Brown and Jay Moody concerning their assignment, the Court is now ready to enter findings of fact upon which an order can be entered. In its order, L(680)86, this Court stated its intention to approve a district-wide plan for facilities rehabilitation, based on the following standard for desegregation-related capital improvements: The order: 1. Gives top priority to nonintegrated and intradistrict magnet schools. 2. Encourages consolidation of enrollments and closing of small schools. 3. Includes only buildings which are now in operation. 4. Abandons or gives low priority to schools which probably should close in the next five years.- 5. Requires a plan of work which describes each particular facility improvement in detail (i.e., renovation or rehabilitation work on an existing building). 6. Includes projects which meet St. Louis health and safety codes. 7. Provides facilities necessary to implement approved 12(c) programs. In response to the order L(680)86, the Board filed its Desegregation Related Capital Improvements Plan, L(852)86 (Plan 680), which met in part and exceeded in part the provisions of the order L(680)86. The Board’s response to the order contains two proposals. The Plan 680, L(852)86, restates the Board’s Zurheide-Herrmann (ZH) costs for rehabilitating 127 buildings, taken from L(598)85. The Plan 680, costing $145,000,000 is presented as a subset of a comprehensive long-range facilities plan, the second part of which would cost another $145,000,000 plus escalation costs. [L(857)86]. The Board has left out capital costs for 17 current and future 12(c) magnets, pursuant to L(680)86, which are subject later to 100 percent state financing. It contends that the entire plan is essential to meet desegregation objectives. The Plan 680 conforms to the Board’s interpretation of the order L(680)86 which the Board believes is too restrictive. The State’s response to the order L(680)86 and to the Board’s plan recites each of the elements of the Court’s standard, shows how the Board allegedly exceeds it, and proposes an alternative plan to meet the State’s interpretation of the order. [L(1047)86]. The State concedes an obligation to help pay for capital improvements which are needed to meet desegregation objectives, not those which generally upgrade the educational system. The State analyzed the validity of ZH improvement recommendations by visiting a sample of 40 schools. The State’s architectural consultants reported that ZH unit costs for construction projects in the schools visited were reasonable. The State agrees with many projects proposed by the Board but rejects others as unnecessary to meet Court standards. The State agrees that 67 of the 127 buildings proposed by the Board should be rehabilitated (using State-approved projects). The State also accepts the Board’s Billingsley enrollment projections for the year 1990 as a basis for planning. Architects, engineers and educators who have inspected the Board’s facilities find them to be fundamentally sound although in a condition of deterioration. The Court agrees with this concensus opinion. As a practical matter, these soundly constructed buildings should be the base point from which facilities improvements proceed. The most productive approach is to start from today’s realities and to move forward constructively to rehabilitate existing facilities. Further, the cost of the plan to rehabilitate buildings currently in use is within the Board’s financial means, as detailed later in this opinion. Recognizing the urgent need for prompt action to rehabilitate existing school facilities in St. Louis, as noted in the Settlement Agreement, Section IY, 9-12, H(2217)83, and in numerous orders and related filings, this Court’s approved capital improvements plan embraces the following objectives: 1. Provide completely rehabilitated buildings and grounds for a sufficient number of nonintegrated, magnet, integrated, and special school facilities to house enrollments projected for the foreseeable future. 2. Provide sufficient classroom space to accommodate projected enrollment in accordance with (1) court-mandated pupil/teacher ratios in nonintegrated schools and magnets and (2) current pupil/teacher ratios in integrated schools, which now exceed AAA accreditation requirements. 3. Make space available to accommodate court-approved 12(c) programs such as (1) preschool rooms, kindergartens, libraries, art, music and physical education facilities, and special education rooms in all schools; (2) enrichment labs and schools of emphasis in nonintegrated schools; (3) computer labs in noninte-grated middle and high schools. 4. Consolidate enrollment, housing students in larger buildings which are in sound operating condition. 5. Provide a planning and reporting system for completion of all needed capital projects and for continuing maintenance of physical school plant. 6. Set aside a stream of revenue to meet the Board’s obligation to fund capital expenditures needed to restore, repair, maintain, or enhance existing facilities. Provide matching State funding for all desegregation-related capital improvements. 7. Make funds available at the earliest possible date so that the most urgent work can begin promptly. 8. Provide for a continuing flow of revenue to be applied to desegregation-related capital projects and to the ongoing maintenance of facilities after improvements are completed. The Court’s plan is detailed under the following nine identified issues: (1) the target year for planning; (2) the projected enrollment to be used for planning purposes; (3) the 12(c) programs to be accommodated; (4) the capacity of presently operating schools; (5) consolidation of facilities; (6) the number of buildings needed to house projected enrollments; (7) the construction projects approvable as desegregation-related; (8) the budgeted cost of rehabilitation and reconfiguration projects; and (9) the method of financing the proposed plan. Target Year for Projections What is an appropriate target year for estimating Board enrollments for the purpose of determining plant capacity needed to meet constitutional requirements? This section of the plan summarizes the parties’ positions as compared with Court guidelines. Directions from the Courts The Court directed the City Board to conduct a long-range study of the City Board’s plant facilities, ‘especially in light of possible changes that may occur throughout the course of this remedial effort in student and staff populations, as well as in specialized educational programs.’ L(680)86, page 2. ... [Bjuildings closed as of this date are not to be considered for renovation. The vitality of open schools must be closely scrutinized in light of declining enrollment as pointed out in Liddell VII ... Schools with low enrollments, which in all reasonable probability should close in the next five years, are good candidates for consolidation and the capital improvement needs for some should be abandoned or prioritized accordingly. Id., page 7. Nonintegrated elementary and middle schools planned for rehabilitation should be “so located that the best available demographic studies indicate that they will continue to serve students for the foreseeable future.” Liddell IX, 801 F.2d at 284. With regard to determining the facilities needed to reach a 20:1 pupil/teacher ratio, “planning should assume that by 1990, fewer than 15,000 black students will be attending non-integrated schools in the city.” Liddell XII, 823 F.2d at 1254. The Board uses the year 2000 as its target year for enrollment projection. The Board believes that redevelopment in the city, improvements to school facilities and K-12 curriculum revisions toward the year 2000 will revitalize the educational program in the public schools. Significant enrollment increases will result. To plan for the short-term will not embrace long-run upward trends in enrollment. Therefore, the year 2000 has been selected for program, enrollment, and plant improvement planning. If enrollment gains do not materialize, proposed plans can be scaled down over the proposed four-year construction period without disadvantage to the plant improvement program. The State chooses the year 1990-91 because it believes that enrollment projections beyond five years are speculative and because the desegregation-related capital improvements program should not be concerned with growth which may occur independent of any past constitutional violation. The last victim of the dual system will graduate in 1991. The State accepts Billingsley’s 1990 enrollment number using the “most optimistic scenario.” This Court and the Eighth Circuit have repeatedly used the phrase “foreseeable future” in connection with capital improvements. A five-year limit (from 1986) was set in Order L(680)86. The Eighth Circuit utilized the school years 1989-1990 and 1990-1991 as target years with regard to enrollments in the interdistrict magnets and the nonintegrated schools. All these factors suggest a target year of 1990. The Court finds that for planning purposes, with regard to capital improvements, the year 1990 provides the most reasonable target year for projected enrollment analysis. Beyond 1990 is simply too speculative. Enrollment Projection What is a reasonable estimate of the enrollment to be served in the target year 1990-91, taking into account demographic changes and the relocation of students under the Settlement Plan? Directions from the Courts. “Planning and scheduling are important in view of the expected decline in enrollment.” Liddell VII, 731 F.2d at 1319. The Board shall plan to improve those nonintegrated elementary and middle schools so located that the best available demographic studies indicate that they will continue to serve students for the foreseeable future. Liddell IX, 801 F.2d at 284. The goal must be to have 2,000 more students in interdistrict magnets by 1987-88 and a total of 6,000 by 1989-90. Liddell IX, at 282. ... planning should assume that by 1990 fewer than 15,000 black students will be attending nonintegrated schools in the City. Liddell XII, 823 F.2d at 1254. The Board projects its target year 2000 local enrollment at 61,995 (including 3,930 county transfers). The enrollment projections used in the Plan 680 are from the 1984 Phase I report, “Population Analysis: St. Louis Public Schools,” by Billingsley Consultants. Four population scenarios, with attendance enrollment scenarios, were developed. The projections ranged from continued decline to stabilization followed by moderate growth. The 680 Plan uses the most optimistic scenario of stabilization followed by moderate growth. Billingsley employed a projection technique which extracted public school enrollment from projected total school-age population, based on an assumed “capture ratio.” For 1990, Billingsley Consultants project 51,584 in its “most optimistic” scenario, assuming stabilization plus moderate growth. [L(848)86, population analysis, pages 19-20]. Bill-ingsley clearly states that its enrollment projections do not account for interdistrict transfers. But, the Board’s Plan 680 states that interdistrict transfers have been removed. The Board then adds 1,815 county-to-city students, deriving a total of 53,-399 for 1990. The Board’s projection does not appear to deduct city-to-county transfers. [L(852)86, page 18]. This fact was later confirmed at the evidentiary hearing held in December, 1986. The State accepts the Board’s 1990-91 enrollment projection data produced by Billingsley, L(848)86, and shown on page 19 of the Plan 680 and page 27 of the State’s filing. Billingsley projected a local enrollment of 51,584 for the 1990-91 school year. When adjusted for interdistrict transfers, the net city enrollment would be 40,574, according to the State. The State also uses current enrollment data to project enrollment, based on the 1985-86 total of 51,236. The State assumes that the 1985-86 enrollment will remain stable through 1990-91. The State then reduces total current enrollment to reflect additional city-to-county transfers and increased magnet school enrollment. After adjustment for these factors, the State predicts that 42,298 children will need to be housed in all St. Louis public schools in 1990-91. [L(1047)86, page 13]. Dr. Brown’s enrollment projection for 1990 is based upon a technique he refers to as the “cohort survival method.” He believes that this technique is much more realistic than the methods employed by the City Board and the State. Neither of those techniques takes into account the City Board’s long-term enrollment history. Neither technique considers student survival rates which may be statistically derived from the City Board’s actual experience. He believes that student survival rates will be influenced by improved learning opportunities through quality education programs, magnet schools and interdistrict transfers. He further believes that the Billingsley enrollment projections are skewed because they are based on total population estimates in the city for five-year periods 1970 to 2000. [L(848)86, pg. 14]. The trend has been steadily downward since 1950. {Id., at 10). Dr. Brown’s projection procedure, the cohort survival method, recognizes the actual retention rate of first grade enrollment as it has moved through the grades since the school year 1971-72. This historical rate is applied to first grade enrollments which will produce enrollments in all grades through the year 1990-91. He extracted his data from the following sources: 1. The Board’s annual October enrollment reports required by paragraph 14 of the district court order of May 21, 1980. 2. Kindergarten enrollment for the years 1971-80 as reported by the Board to the State Department of Elementary and Secondary Education (DESE). Enrollment for the years 1980-86 was obtained from the Board’s annual October enrollment reports. 3. First grade enrollment for the years 1971-1986 as reported by the Board to the DESE. 4. Interdistrict transfers as reported by vice. 5. Vocational interdistrict transfers as reported by MCC. The enrollment projection derived from the cohort survival method indicates that 44,737 will need to be accommodated in St. Louis schools in 1990-91. Dr. Brown recommends that facilities be provided to house 31,261 regular students (5,301 in grades 9-12 plus 24,960 in grades K-8); 12,876 magnet students (4,530 local students in grades 9-12, 7,103 local students in grades K-8, plus 1,243 students from county districts); and 600 vocational students. Since 1980, there has been an increasing interchange of students between St. Louis and suburban schools as a result of desegregation court orders. Interdistrict transfers will have a significant impact upon student assignments and housing. Liddell IX established the goal of at least 2,000 additional students in interdistrict magnets by the opening of the 1987-88 school year and a total of 6,000 students in these magnets by the 1989-90 school year. The goal is 40 percent white students from the county. Interdistrict magnet enrollment also reduces the classroom space needed in both regular integrated and nonintegrated schools. Housing provided for students attending interdistrict magnets reduces the housing needed for regular enrollment. The Court has carefully reviewed the enrollment projections of the City Board, the state and Dr. Brown and the various methods employed to calculate these projections. It appears to this Court that the most reliable estimate is derived via the cohort survival method. This procedure for analysis accurately makes allowance for students who have transferred in and out of city schools since 1980. It is also the only procedure which tracks actual enrollment prior to 1980 and adjusts the historical enrollment trend for the effects of the desegregation programs. The Court finds that this method provides the most realistic basis for a 1990 enrollment projection. Accordingly, the Court accepts a 1990 enrollment forecast of 44,737 City students. Accommodating 12(c) Programs What Settlement Plan 12(c) programs must be provided for in rehabilitated schools? The allocation of space for the 12(c) programs is an important factor in determining available physical plant capacity- Directions from the Courts. “Additional renovation work must be necessary for the implementation of the approved 12(c) Programs.” L(680)86, page 6. “(The) model standard that the (ZH) engineering firm created ... is a higher standard than this Court has pronounced or will accept.” Id., at page 8. The Board uses the Winsor Model in making its capacity projections. The model provides space for the Board’s year 2000 curriculum, which includes all 12(c) program requirements. L(852)86, pages 10, 16, 20-29. It contends that its 680 Plan does not include all programs, in the absence of building additions. But it does not specificaly show the program requirements by budget function as required in L(680)86. The State claims that the Board’s proposal contains elements which exceed 12(c) needs, such as elementary foreign language rooms, computer literacy labs in integrated schools, handicapped access, counseling offices for high schools and middle schools, expanded facilities for counselors, nurses, security guards, middle school science rooms, and excess storage space. However, the State used the Winsor capacity estimates in assigning its 1990-91 enrollment to buildings. The courts have clearly enumerated 12(c) programs which require space allocations in nonintegrated and/or integrated schools. In analyzing plant capacity space has been allowed both nonintegrated and integrated schools for art, music and physical education rooms; all-day kindergarten rooms; preschool rooms; special education rooms; and libraries. Also, space has been set aside for 12(c) programs implemented in úie nonintegrated schools only, as follows: 1. Rooms for school of emphasis labs and enrichment labs in elementary, middle and high schools. 2. Computer labs in high schools and middle schools. Analysis of Plant Capacity What school building capacity will be required to house projected enrollment in St. Louis schools by the target date? Is the capacity of presently operating schools adequate? Directions from the Courts. Consolidation of the city school district is the key to an effective capital improvements program. The vitality of open schools must be closely scrutinized in light of declining enrollment. Schools with low enrollments which will probably be closed in the next five years should be abandoned or given a low priority for renovation. L(680)86, page 7. Nonintegrated elementary and middle school classrooms should have P/T ratios of 20/1 by 1987-88, excluding remedial and compensatory program staff. Liddell IX, 801 F.2d 280-281. Integrated and nonintegrated high schools should have P/T ratios meeting AAA Standards (35/1). Integrated elementary and middle schools should have ratios of 30/1. Id., at 281-282. Magnets should have the same P/T ratios as nonintegrated schools. Id., at 283-284. The requirements of Liddell VII (i.e. P/T ratios of 20:1 in nonintegrated elementary and middle schools) are to be met at the earliest possible time. Liddell XII, 823 F.2d at 1254. (2) Even though the goal of twenty to one may not be fully achieved in the 1987-88 school year, it is apparent that some reduction in class size in the non-integrated elementary and middle schools can be made at the beginning of the 1987-88 school year by making more efficient use of classrooms in the elementary and middle school buildings. Liddell XII, at 1254. In determining the facilities needed to reach the twenty to one pupil/teacher ratio in the nonintegrated elementary and middle schools, the parties are to assume — and accomplish as a fact — that by the 1990-91 school year, an additional 5,500 black city students will be attending county schools, that by the same year an additional 2,750 black city students will be attending magnet schools in the city, and that there will be some space available for additional black students in the integrated comprehensive schools. Thus, unless demographic factors clearly indicte otherwise, planning should assume that by 1990, fewer than 15,000 black students will be attending nonintegrated schools in the city. Liddell XII, at 1254. (4) The record reveals that many of the nonintegrated elementary and middle schools are very small. For example, sixteen nonintegrated elementary schools have enrollments of fewer than 250 students per school, with an average enrollment of 170 students. The record also reveals that many of the nonintegrated elementary and middle schools are poorly maintained and located and would be very expensive to rehabilitate. In its long-term planning for the noninte-grated elementary and middle schools, the City Board should carefully consider closing the smaller, poorly maintained and poorly located nonintegrated elementary and middle schools and consolidating the students in either new or existing buildings that are well located and designed specifically to meet a pupil/teacher ratio of twenty to one, the goal being to achieve significant operating savings while providing the students with suitable facilities. Liddell XII, at 1254. (See Liddell XII, footnote 1 regarding consolidation). The Board states that the necessary reconfiguration required to accommodate all Settlement Plan programs will cause a drastic reduction in pupil capacity of each school. It assumes that program fitness has been minimally and essentially described in the Winsor Model. L(852)86, page 14. Building capacities, present and future, listed in the Plan 680 are based upon the Winsor Model. The Board downgrades a number of schools to a lower capacity category because the Order, L(680)86, precludes building additions which provide Winsor-model facilities. Id., page 22. The Plan 680 shows inadequate building capacity by the year 2000. The unhoused enrollment is expected to be: elementary, 9,200 (page 63); middle, 637 (page 46); high school, 530 (page 35). The State uses the Board’s own building capacity data as shown in the Plan 680 to show that future enrollments can be housed in fewer buildings. These capacities allow for libraries, music, art and other special rooms and for generous storage space. They also provide for some programs which the State claims are unrelated to desegregation. L(1047)86, page 11 and Attachment 1. The State’s capacity projections also assume that provision of classrooms — other than in nonintegrated schools — to house students in ratios lower than required by AAA standards is not required for desegregation purposes. If it is done, the State contends that the local board should pay for the extra classrooms needed to meet desired class sizes. Most of the Board’s proposed 12(c) reconfiguration is not required for desegregation. Id., pages 30-34. The Board’s class size goals (K-15/1; elementary — 20/1; middle — 20/1; high school — 25/1) in all schools greatly exceed Liddell IX mandates, AAA standards and suburban school ratios. Dr. Brown’s analysis of capacity focuses on the facilities needed to house anticipated regular enrollment in integrated and non-integrated schools, K-12, by the year 1990-91. The procedure for determining available spaces differs for each level of instruction: high school, middle school, and elementary school. The data base for determining available school building capacity is the current number of schools serving students in instructional programs in the 1986-87 school year. Closed schools are not considered, per L(680)86. The P/T ratio used by Dr. Brown in assigning classroom space in the integrated elementary and middle schools is 26/1. Although this P/T ratio is below the AAA standards as required by Liddell IX, Dr. Brown chose to keep this P/T ratio in his calculations because it is consistent with the capacity analyses used by the City Board and the State; and many of these schools are already at or near this P/T ratio. He further chose to accept the City Board’s capacity estimates for the eight regular high schools currently in operation (integrated and nonintegrated). These capacity estimates are based on a P/T ratio of 28/1, which also exceeds AAA accreditation and Liddell IX requirements. He created a new model for the nonintegrated elementary and middle schools to account for the P/T ratio of 20/1 as mandated by Liddell IX and XII. Dr. Brown’s capacity analysis concludes that a surplus of instructional space will exist at all levels. He surmises that by 1990-91, there will be a surplus capacity in the regular high schools of 8,007; equivalent to 285 classrooms. In the integrated middle schools, a surplus of 86 regular classrooms will remain in 1990. A surplus of 99 classrooms will remain in the noninte-grated middle schools in 1990. Finally, he concludes that an excess of 75 regular classrooms will remain in 1990 in the integrated elementary schools and an excess of 53 regular classrooms in the nonintegrated elementary schools. He further found present intradistrict magnet capacities to be adequate. Since the Magnet Panel is currently studying all magnets, capacities in the Settlement Plan magnets were not considered. There is abundant space available for vocational education (O’Fallon) and for the three special program schools: Gal-laudet, Gratiot and Michael. City Board’s Plan 680 assumes the same P/T ratio in all schools. This is inconsistent with the directives of Liddell IX and XII The Winsor Model prototype is an unrealistic model for facilities planning within the purview of Court requirements for desegregation. The St. Louis City school system is an old urban school system. None of its facilities could ever approximate the Winsor Model. The Winsor Model is an idealistic contemporary educational facility. The City Board’s facilities are structurally sound old buildings. Simply because they are not new does not mean they are not usable. This Court and the Eighth Circuit have repeatedly stressed the rehabilitation and utilization of existing buildings. This is the main thrust of a capital improvements plan for the St. Louis City school system. The Winsor Model impairs realistic planning, which should begin with conditions as they now exist. For example, the downgrading of schools’ capacities for minor inconsistencies with the Winsor Model(s) eliminates about 900 classroom spaces from the City Board’s original capacities. L(852)86, pgs. 21-28. Su