Brown v. Board of Education of City of Chicago

MEMORANDUM OPINION AND ORDER McLAREN, District Judge. I. Introduction This is a class action civil rights claim against the Board of Education of the City of Chicago (Board) and its general superintendent, James F. Redmond (Redmond) contending that the Board arbitrarily allocates the city’s educational funds in a manner which systematically discriminates against non-Caucasian and poor children. Plaintiffs sue individually and on behalf of all students attending those Chicago public elementary schools for which the defendants allocate and spend a less than equal per-pupil share of state and local funds; plaintiffs seek to represent two subclasses: (1) the sub-class of all non-Caucasian students and (2) the sub-class of all students from families of low or moderate income or dependent economic status. This opinion shall constitute the Court’s findings of fact and conclusions of law as provided by F.R.Civ.P. 52(a). The case presented to the Court by all parties is largely composed of the Board’s own statistical data which plaintiffs contend establishes a prima facie case of racial and economic discrimination. Defendants argue that the differentials in per-pupil expenditures revealed in this data are not significantly correlated to the racial or economic composition of individual schools. Instead, defendants argue that spending differentials result from natural causes such as school size, participation in experimental programs, and demographic make-up of an individual school. Other evidence in this case includes the depositions of Dr. Redmond and other agents of the Board, and the depositions of plaintiffs’ and defendants’ independent educational experts. Plaintiffs attempt to draw from this testimonial evidence proof that the level of per-pupil expenditures has a substantial impact on the quality of the education offered students and on the level of student achievement. None of the parties has chosen to present in-court testimonial evidence. II. Statements of Facts A. Background of Racial and Economic Demographics in the Chicago Public Elementary Schools The City of Chicago, like most large American urban centers, has racially segregated residential housing patterns. These housing patterns are significant in the instant action because the defendants here have adopted the neighborhood school concept. Thus an overwhelming majority of the public elementary schools in the City of Chicago have student body memberships comprised of 90% or more of one racial group: (i) In the 1969-70 school year, of the 416 elementary schools, 174 had student populations over 90% non-Caucasian, 142 had student populations over 90% Caucasian, and 100 had student populations that were “mixed,” that is, no race accounted for more than 90%. (ii) In 1970-71, of the 427 elementary schools, 198 had student populations over 90% non-Caucasian, 101 had student populations over 90% *116 Caucasian, and 128 had student populations that were “mixed.” (iii) In 1971-72, of the 434 elementary schools, 208 had student populations over 90% non-Caucasian, 94 had student populations over 90% Caucasian, and 132 had student populations that were “mixed”. (iv) In 1972-73, of the 446 elementary schools, 221 had student populations over 90% non-Caucasian, 92 had student populations over 90% Caucasian, and 133 had student populations that were “mixed”. (v) The percentage of schools having student populations over 90% non-Caucasian or over 90% Caucasian changed in only a minor degree over the four-year 1969-73 period: Defendants correctly argue that no evidence is present in this record of any act by the Board which directly caused segregated housing patterns. Moreover, plaintiffs do not argue here that the neighborhood school concept, where a de jure policy of segregation has not existed, is constitutionally infirm. Therefore, evidence that one racial group comprises more than 90% of the student membership in a given school does not, standing alone, establish discrimination on the basis of race. However, this evidence of the pattern of racial composition of Chicago elementary schools is relevant to the instant action because the Board’s funding allocation policies do not operate in a vacuum; in establishing funding policies the Board must recognize that a racially neutral policy may result in an invidiously discriminatory result because of the established racial attendance patterns in the school system. Moreover, many neighborhoods in the City of Chicago are almost exclusively comprised of low or moderate income families or families of dependent economic status; similarly, many city neighborhoods are almost exclusively comprised of middle or upper income families. Public elementary school students, with few exceptions, are required by defendants’ districting and pupil assignment policies to attend schools in the neighborhoods where they reside. Plaintiffs maintain that as a consequence, children from lower income families generally are required by defendants’ policies to attend schools located in poorer neighborhoods, while children from higher income families attend schools in wealthier neighborhoods. Defendants implicitly accept this conclusion. They admit that eligibility for funds under Title I of the Elementary and Secondary Education Act of 1965 (ESEA) 1 can serve as an indicator of *117 low economic status. Title I schools appear to be clustered in neighborhoods with relatively low socio-economic status [SES]. Since most elementary students attend school in the neighborhoods in which they reside, it is a logical inference that schools located in poor neighborhoods will have student populations largely composed of children from poorer families. B. Variations in Per-Pupil Staffing Expenditures With this background data established, we turn now to plaintiffs’ central factual contention that the Board’s funding policies cause invidious racial and economic discrimination in per-pupil staffing expenditures. Plaintiffs’ cause on this point is based entirely on an analysis of the Board’s expenditures for the school years 1969-70 and 1970-71. Plaintiffs’ expert witness, Richard A. Berk, analyzed the 1969-70 per-pupil cost for the staffing of the 416 elementary schools in the Chicago school system. His analysis showed that the 174 schools with more than a 90 % black student body received an average of $391 per pupil, while the 142 schools with more than 90% white membership received an average of $423 per pupil. The remaining 100 “mixed” schools received an average of $406 per pupil. Plaintiffs’ expert also found that the same sort of systematic allocation of resources was apparent when schools were measured on the basis of socio-economic status. Children in the 184 Title I schools (generally poorer children) received an average of $387 per pupil, while children in the 232 non-Title I schools (relatively richer children) received $420 per pupil. Plaintiffs’ analysis of the data for 1970-71 showed a like distribution of funds. Total instructional staffing expenditures per pupil for the 198 nonwhite schools equaled $451 while the expenditures for the 101 predominantly white schools equaled $488. Expenditures per-pupil in the mixed schools equaled $458. Defendants object to certain aspects of the methodology used to obtain these statistics, but defendants’ own reports to the citizens of Chicago, Selected School Characteristics, 1969-1970 and Selected School Characteristics, 1970-1971 use similar statistical methods. Moreover, this Court’s examination of the July 1972 Board Report entitled Per Pupil Staffing Costs — Elementary Schools — First Semester 1971-72 and other reports reveals that the Board found that mean racial differentials in 1970-71 in staff expenditures were of a similar magnitude — about 8.3%. Likewise, the Berk analysis shows for 1969-70 and 1970-71, 8.2% (Defendants’ exhibit D, page V-6; Berk direct testimony, p. 8). This Court therefore finds that for the school years 1969-70 and 1970-71 the mean per-pupil expenditure for staffing costs for white schools exceeded the expenditure in predominantly non-Caucasian schools by more than 8%. 2 The relationship between racial composition of a school and total per pupil instructional expenditures for 1970-71 can also be demonstrated by ranking the various schools by dollars allotted. When rank ordered, plaintiffs’ data shows that while 22.7 % of the non-white schools are in the lowest 20% expenditure category, only 4% of the white schools fall in the bottom 20%. Similarly, 35.8% of the predominantly non *118 white schools are in the bottom 30% while only 6% of the predominantly white schools fall in the bottom 30% category. In contrast, 34.6% of the white schools score in the top 20% group, while only 14.7% of the predominantly black schools are in the top 20% expenditure category. Defendants’ data contained in defendants’ Exhibit B, table 3, comports substantially with plaintiffs’ rank order evidence. Thus, even though the parties disagree as to the proper methodological approach to establishing rank orders by deciles, since the parties’ data is substantially similar, the Court finds that when all the elementary schools are rank ordered by staffing expenditure for 1970-71, non-white schools are significantly more likely to appear in the bottom deciles than white schools. Plaintiffs’ have introduced no evidence analyzing the per-pupil staffing expenditures for the school years since 1970-71. Defendants’ data, however, shows that significant changes in the pattern of staffing expenditures have occurred in the school years since 1970-71. Defendants’ data for the school year 1971-72 shows that the expenditure difference between predominantly white and predominantly black schools had dropped to 4.0%. Data for the 1972-73 school year reveals that the difference between the spending for predominantly white and predominantly black schools had decreased to 1.0%. The rank ordering data for these two years also reflects increased equality of expenditure between white and black schools. As noted, plaintiffs put in no evidence rebutting this information and therefore the Court finds that in the school years 1971-72 and 1972-73 the Board made substantial progress in equalizing per-pupil expenditures in schools with different racial compositions. C. Explanations of Expenditure Patterns In attempting to explain the variations in per pupil staffing expenditures in any school year, the parties agree that the major determinants of costs are the result of two basic factors: “the educational preparation and experience of the teachers, as indicated by their lane and step on the salary schedule (and) “The level of staffing (number of teachers) in the class and in the auxiliary areas in relation to the number of pupils.” (Plaintiffs’ Exhibit No. 4, page A3, entitled “Report, Per Pupil Staffing Costs — Elementary Schools, October 30, 1970, Projected 1970-71” prepared by the Department of Operations Analysis, Chicago School Board) These factors cause a variation in expenditures per pupil because the more experienced the teachers a school had and the more educated they were in terms of advanced degrees and courses, the higher were their salaries and the higher were the per pupil staffing costs in the school in which they were teaching. The causes of variations in staffing levels are less easily explained. The Board asserts that staffing level variations, while difficult to explain, may result from the fact that some schools have classrooms which physically accommodate fewer than the usual number of students and thus require a larger number of teachers for the same number of students. Additionally, a school with a very small enrollment may need to operate at a low pupil-teacher ratio, with possibly a resulting higher expenditure per pupil, in order to avoid having too wide a range in age or grade groups in one classroom. The Board explains variations in auxiliary staffing costs as being caused by such factors as existence of experimental programs in a school or the need for bilingual programs. Auxiliary costs also vary with school size. Small schools, generally speaking, have higher auxiliary costs than larger schools because the Board provides some standard educational services regardless of the number of pupils using the service. However, no direct *119 evidence exists in the record why schools with a lower staffing level tend to cluster in non-white neighborhoods. The Board explains clustering of experienced teachers in certain schools as a result of many factors, including: the desire of teachers, particularly elementary school teachers, to teach as close to home as possible, thus resulting in experienced teachers, except during the moratorium on teacher transfers, transferring to schools near their homes. Thus a school in a community of a socio-economic level which produces teachers is likely to have more experienced teachers than one in a community in which teachers are not likely to reside. As a school increases in membership and requires additional numbers of teachers, these teachers, especially in schools in low socio-economic communities which do not produce teachers, are likely to come from the ranks of those with the least experience and hence the lowest salaries. As a school decreases in membership and requires fewer teachers, it is the teachers with the least experience and hence the lowest salaries who will be required to leave. Thus, in general, schools which are increasing in membership are likely to have increasing numbers of inexperienced teachers and lower costs while schools decreasing in membership are likely to have an increasing percentage of experienced teachers and higher costs. Some communities appear to provide greater support to the school program and to teachers in the schools than do others and thus are able to retain teachers on the staff. These explanations of experience clustering demonstrate how a differential racial impact arose creating the expenditure variations shown in the 1969-70 and 1970-71 school years. Experienced teachers in the Chicago school system have historically used a transfer policy which allowed teachers with seniority to transfer out of certain schools to other schools which the teachers thought more desirable. As the Board recognized, these more desirable schools tended to be in neighborhoods with higher socio-economic status (SES). Census data reveals that these high SES areas also tend to be predominantly Caucasian neighborhoods. As a consequence, those teachers with less experience and education did cluster in non-Caucasian neighborhoods. D. Corrective Measures Undertaken by the Board This experienced teacher clustering phenomenon has been somewhat alleviated by the Board's imposition of a moratorium on voluntary teacher transfers. Under the transfer moratorium, no regular voluntary transfers of teachers from school to school were effected in September 1971 or in September 1972, the usual time for teacher transfers. Instead, the Board mandated a policy that the vacancies to which the more experienced teachers would have been transferred if there were no moratorium were largely filled by teachers new to the school system who were at the lower range of the salary schedule. Concomitantly, the more experienced teachers were retained in schools from which they might normally transfer. As the Board reports: “[t]hus there was a tendency toward lower per pupil staffing costs in schools which usually receive transfers in vacancies created by' attrition or membership growth and, on the other hand, a tendency toward higher per pupil staffing costs in schools from which, were it not for the moratorium, more experienced teachers would be transferring to be replaced by teachers lower on the salary schedule.” Defendants’ Exhibit D, page III-5. In the school years 1971-72 and 1972-73, the Board implemented two other policies which tended to. alleviate the disparities in per pupil expenditures between schools. These programs were denominated by the Board as the “maximum class size program” and the “equalization program.” The maximum *120 class size program actually began in 1970 as part of the Board’s contractual agreement with the Chicago Teachers Union (CTU). This agreement selected certain elementary schools where the pupil-teacher ratio was above the norm for additional staffing to lower class size; prime consideration was to be given “to schools with the lowest per pupil staffing costs and these were selected to receive the additional teachers if space was available to make it possible to actually lower class size and if the schools also had large numbers of underachievers and high mobility, or were in changing communities.” (Defendants’ Exhibit B, page III — 1). The 1970 plan as the Board admits was not entirely successful because space to actually lower class size was often not available. Thus in some instances schools with relatively high per pupil staffing costs with extra space were designated to receive additional staff. In 1971 the maximum class size program was modified to correct this problem; schools with the lowest per pupil staffing costs, based on the Board’s 1970-71 ranking, received first priority for additional staff under the maximum class size program, even though they did not have space to establish classes of smaller size. In addition to the maximum class size program, the equalization program also added staff to low expenditure schools. This program was directly focused on the disparate funding problem; additional teachers were assigned to schools where the per pupil staffing costs continued to be low in spite of the staff added because of designation of the school as a maximum class size school. During the 1972-73 school year, a third program was instituted which had the effect of equalizing per pupil expenditures. This program, termed the “ESEA comparability program,” was established to effectuate ESEA guidelines which require that Title I participating schools have equal or lower pupil-teacher ratios and equal or higher per pupil instructional costs than the average for non-Title I schools, within 5%. During the 1972-73 school year, 635.0 teaching positions were assigned to low expenditure schools under the maximum class size program, 212.5 positions under the equalization program, and the ESEA comparability program provided 45.0 positions for a total of 892.5 positions. The 1973 budget of the Board allocated $667,000 to fund the equalization program and approximately $300,000 to fund the cost equalization program. There is no evidence that the measures taken by the Board to equalize expenditures were adopted in response to this litigation. Moreover, it is the stated intention of the Board to continue these corrective policies. The Board’s contractual agreement with the CTU requiring maintenance of the maximum class size program remains in force, federal regulations require continuation of the ESEA comparability program, and the Board has stated that it will continue the equalization program of its own volition. E. The Effect of Expenditures on Educational Quality As a final step in proof, plaintiffs attempt to show that the expenditure level affects the quality of education. Plaintiffs’ expert, Arthur E. Wise (Wise) stated that variations in the quality of a school’s program do result from differences in expenditure levels; Wise also asserted that the most important factors which have been found to be associated with student achievement are the characteristics of the teacher, even when the experimenter controls for the SES of the students studied. Teacher experience, student/teacher contact, and the academic preparation of the teacher all appear to have some effect on student performance. Defendants’ expert, William P. McLure, disagrees. He states that studies examining the relationships between teacher characteristics, expenditures and educational outcomes are tenuous at best. Each party’s expert cites numerous studies which *121 support one position or the other; in the instant action, however, this Court does not have to choose which scholarly camp is correct; the Board’s behavior here serves as an admission that it believes expenditures have some effect on educational outcomes. The corrective measures undertaken by the Board were specifically directed at the factors Wise found significant — class size, teacher experience and teacher educational qualifications. The Board by its actions therefore recognizes that plaintiffs’ theory has some relevance in planning educational policy for the Chicago school system. The Board’s action in spending over $1,000,000 per year on programs which more equitably distribute the resources which plaintiffs claim affect student achievement serves as persuasive, objective evidence that expenditures within the Chicago Public School system do play a role in student achievement. 3 III. Conclusions of Law A. Jurisdiction This Court has jurisdiction over the parties and subject matter of the instant controversy. Jurisdiction over Dr. Redmond can be maintained under 28 U.S.C. § 1343(3) and 42 U.S.C. § 1983. As chief administrative officer of the Chicago Public School System, it goes without saying that Dr. Redmond “acts under the color of state law” and that if his actions cause invidious racial discrimination, a claim has been stated under the Civil Rights Act of 1871. Under 28 U.S.C. § 1343 no jurisdictional amount need be met. Maintaining jurisdiction over the Board, as a municipal corporation, presents more complex problems. The Supreme Court’s decision in City of Kenosha v. Bruno, 412 U.S. 507, 93 S.Ct. 2222, 37 L.Ed.2d 109 (1973) holds that state political subdivisions are not “persons” within the meaning of the Civil Rights Act of 1871 and thus are not amenable to suit under § 1983. Therefore, the Board is not a proper party-defendant under 42 U.S.C. § 1983 or 28 U.S.C. § 1343. Taliaferro v. State Council of Higher Education, 372 F.Supp. 1378 (E.D.Va.1974); United Farmworkers of Florida v. City of Delray Beach, 493 F.2d 799 (5th Cir. 1974). 4 Jurisdiction over the Board can be maintained however, pursuant to the Fourteenth Amendment and 28 U.S. C. § 1331. Federal courts have the authority to directly fashion remedies for violations of the Fourteenth Amendment without specific statutory authority. See, e. g., Bell v. Hood, 327 U.S. 678, 66 S.Ct. 773, 90 L.Ed. 939 (1946). This power has been invoked in innumerable civil rights cases vindicating rights granted under the Equal Protection clause. See, e. g., Baker v. Carr, 369 U. S. 186, 82 S.Ct. 691, 7 L.Ed.2d 663 (1962). The Court has jurisdiction over the Board in the instant action to fashion such remedies if the $10,000 jurisdictional requirement contained in 28 U.S.C. § 1331 is met. In injunctive actions, such as this, the amount in contro *122 versy is the value of the right to be protected or the extent of the injury to be prevented. Wright, Law of Federal Courts, § 34 at page 116. Each member of the plaintiff class is attempting to protect his right to be free from racial or economic discrimination. That right in the context of a claim of a deprivation of equal educational opportunity is easily worth more than $10,000' for each member of the plaintiff class. Wattenberg, The Real America, page 76 (1974). Cf. Zahn v. International Paper Co., 414 U.S. 291, 94 S.Ct. 505, 38 L.Ed.2d 511 (1974). See also Stanton v. Bond, 504 F.2d 1246 at 1251, n. 25 (7th Cir. 1974). At the very least, the Court is not persuaded that the value of the right to every member of the class is “to a legal certainty” worth less than $10,000; since the allegation of the amount in controversy was made in good faith, dismissal of the Board for lack of the jurisdictional amount is inappropriate. St. Paul Mercury Indemnity Co. v. Red Cab Co., 303 U.S. 283, 58 S.Ct. 586, 82 L.Ed. 845 (1938). 5 B. Equal Protection Claims and Racial and Economic Classifications In analyzing state action under the equal protection clause of the Fourteenth Amendment, two divergent tests have developed — the “classical” test and the “compelling state interest” test. In order to establish a violation of the classical equal protection test, a litigant must show the existence of some discrimination and that the classification involved is not rationally related to some legitimate governmental purpose. Weber v. Aetna Cas. & Sur. Co., 406 U.S. 164, 92 S.Ct. 1400, 31 L.Ed.2d 768 (1972). The classical test, however, has proved insufficiently stringent to check certain types of discrimination; the Supreme Court requires in cases where the classification system concerns “fundamental rights” or creates “suspect classifications” a more demanding standard of review. In these cases, the distinctions created may stand only if they fulfill some compelling state interest rather than a merely legitimate governmental purpose. See, e. g., Shapiro v. Thompson, 394 U.S. 618, 89 S.Ct. 1322, 22 L.Ed.2d 600 (1969). 6 1. Economic Discrimination Plaintiffs’ claim of economic discrimination must be measured under the “classical” rational-reasonable test. In San Antonio Independent School District v. Rodriguez, 411 U.S. 1, 93 S.Ct. 1278, 36 L.Ed.2d 16 (1973) the Supreme Court held that education was not a fundamental right in the constitutional sense; therefore, a relative deprivation of educational opportunity in and of itself was not entitled to strict judicial scrutiny under the equal protection clause. Moreover, the Rodriguez court held that the only type of wealth discrimination arguably entitled to strict judicial scrutiny was discrimination against poor persons whose incomes fell below some identifiable level of poverty or who might be characterized as functionally indigent. Plaintiffs here fail to show that the economic class they seek to represent can be identified in such traditional terms. *123 As the previous discussion of the distribution of Title I schools indicates, these schools, largely located in neighborhoods with a low SES, are attended by relatively poorer students. Plaintiffs fail to prove, however, that all the students in these schools fall below the poverty line in family income or are functionally indigent. In fact plaintiffs concede that students of all economic levels attend each Chicago public school. Since even Title I schools have some diversity in the economic status of their student bodies, the system of alleged discrimination against Title I schools does not meet the heavy burden of proof Rodriguez required to establish a suspect economic class: “The class is not saddled with such disabilities or subjected to such a history of purposeful unequal treatment, or relegated to such a position of political powerlessness as to command extraordinary protection from the majoritarian political process.” 411 U.S. at 28, 93 S.Ct. at 1294. Additionally, the Rodriguez decision requires an absolute deprivation of the desired benefit in order for wealth discrimination to be measured by the compelling state interest test. In the instant case plaintiffs do not allege nor have they proved that such an absolute deprivation occurred. Plaintiffs only claim that members of the plaintiff economic class received a relatively poorer education; to invoke the compelling state interest test they must show more; “where wealth is involved the Equal Protection Clause does not require absolute equality or precisely equal advantages.” 411 U.S. at 24, 93 S.Ct. at 1291. Thus, with respect to the claims of economic discrimination here, the proper legal analysis requires application of the “classical” rational-reasonable test. The variance in expenditures between Title I and non-Title I schools passes constitutional muster under the rational-reasonable test. Using this limited standard of judicial review, the Board has demonstrated that its teacher assignment policies, the major cause of the expenditure differentials, serve legitimate governmental purposes. Allowing experienced teachers to transfer to more “desirable” schools presumably helps keep these teachers in the Chicago school system, which upgrades the skill level of the system as a whole. Moreover, variations as between schools in the student-teacher ratio is not constitutionally infirm under the classical test. Here, administrative convenience, sudden unexpected demographic changes, and response to the special educational need of certain neighborhoods serve as sufficient justifications for the Board’s policies. Thus, plaintiffs have failed to prove that the equal protection clause was violated with respect to expenditure levels between Title I and non-Title I schools. 2. Racial Discrimination The Rodriguez decision does not change the standard of review of state action in cases where the aggrieved party can prove racial discrimination. 7 Since race is a suspect classification, such state action must be measured under the compelling state interest test. See, e.g., Loving v. Virginia, 388 U.S. 1, 87 S.Ct. 1817, 18 L.Ed.2d 1010 (1967); cf. Shapiro v. Thompson, 394 U.S. 618, 658-659, 89 S.Ct. 1322, 22 L.Ed.2d 600 (1969) (Harlan, J., dissenting). It must be emphasized, however, that this stricter standard of judicial scrutiny is imposed only after an aggrieved party has proved that racial discrimination exists. In the instant ease, as noted, plaintiffs’ proof of racial discrimination consists almost entirely of statistical data showing an 8% expenditure differential *124 between non-Caucasian and Caucasian schools in 1969-70 and 1970-71. No evidence was introduced to show that the Board’s behavior involved anything but unintended de facto racial discrimination. In Jefferson v. Hackney, 406 U.S. 535, 548, 92 S.Ct. 1724, 32 L.Ed.2d 285 (1972) the Court stated that a “naked statistical argument” showing differential racial impact where the differential could be explained as an infrequent coincidence would not present an appropriate circumstance for imposing the compelling state interest test. Where only a naked statistical argument was made showing differential racial impact, the Court suggests that racial motivation or an irrational classification must be shown before the stricter standard of review is invoked. Statistics, however, remain an important tool in measuring the discriminatory impact of governmental action. Ultimate effect can be as probative of an intent to discriminate as proof of subjective motivation. Gautreaux v. CHA, 436 F.2d 306 (7th Cir. 1970). Moreover, in eases where racial discrimination is alleged, the historical context of the governmental conduct lends color to the statistical argument presented. Kennedy Park Homes Ass’n v. City of Lackawanna, 318 F.Supp. 669 (W.D.N.Y.), aff’d, 436 F.2d 108 (2d Cir. 1970), cert. den., 401 U.S. 1010, 91 S.Ct. 1256, 28 L.Ed.2d 546 (1971); Sisters of Prov. of St. Mary of Woods v. City of Evanston, 335 F.Supp. 396 (N.D.Ill.1971). If there is a history of racial separation involved in the governmental action challenged, then statistical proof alone may be enough evidence to prove racial discrimination. 8 See also Hawkins v. Town of Shaw, 437 F.2d 1286 (5th Cir. 1971), aff’d en banc, 461 F.2d 1171 (1971); Hobson v. Hansen, 269 F.Supp. 401, 496 (D.D.C.1967), aff’d sub nom. Smuck v. Hobson, 132 U.S.App.D.C. 372, 408 F.2d 175 (1969). Thus it is not necessary in cases where the plaintiff attempts to prove structurally imposed racial discrimination to prove intent, motive or purpose. Consequently, as the court noted in Norwalk CORE v. Norwalk Redevelopment Agency, 395 F.2d 920, 931 (2d Cir. 1970): “ ‘equal protection of the laws’ means more than merely the absence of governmental action designed to discriminate; . . . ‘we now firmly recognize that the arbitrary quality of thoughtlessness can be as disastrous and unfair to private rights and to the public interest as the perversity of a willful scheme.’ ” *125 In the instant action, plaintiffs’ statistical proof must be read in light of historical circumstances creating segregated housing patterns in Chicago. The Board’s staffing policies, particularly allowing the more educated and more experienced teachers to transfer to the schools of their choice, resulted in the expenditure level in non-Caucasian schools being lower than that of white schools. This policy, considering the historical circumstances, does contain that arbitrary thoughtlessness which may result in unconstitutional racial discrimination. The 8% variation in expenditures in 1969-70 and 1970-71, while small in dollar amount, establishes a prima facie case of racial discrimination; the expenditure differential is especially significant considering the fact that it was caused by the clustering of beneficial educational benefits such as smaller classes and more educated, experienced teachers in white schools. The Board’s own corrective measures, as indicated, demonstrate the importance of these factors. Measured against the compelling state interest test, the staffing expenditure differentials in 1969-70 and 1970-71 between Caucasian and non-Caucasian schools are constitutionally unacceptable. Administrative convenience or employee desires do not provide sufficient justifications where governmental action results in differential racial impact. 9 Thus, the plaintiffs’ statistical proof remains unrebutted and if it represented current conditions, the Court would be forced to intervene. C. Appropriateness of Injunctive Relief Even though plaintiffs have proved that racial discrimination occurred in 1970-71 and prior years, the Court still must consider whether injunctive relief is appropriate. As the court stated in Walling v. Clinchfield, 159 F.2d 395, 399 (4th Cir. 1946): “The rule is well settled that the extraordinary writ of injunction will not issue for the purpose of punishing past offenses, but will issue only in those cases where the court is convinced that such relief is necessary to prevent future violations.” Civil rights cases provide no exception to this general rule. See, e.g., James v. Carnegie Public Library, 235 F.Supp. 911 (E.D.S.Car.1964). Thus, a federal court of equity when dealing with state action, statutes or ordinances generally should not issue an injunction when the alleged illegal activity has stopped and there is a bona fide intention not to resume it. Magtab Publ. Corp. v. Howard, 169 F.Supp. 65 (D.La.1951); see also, Wright & Miller, Federal Practice and Procedure § 2942 at Vol. 11, pp. 371-73. Additionally, as Chief Justice Burger stated in Lemon v. Kurtzman, 411 U.S. *126 192, 200, 93 S.Ct. 1463, 1469, 36 L.Ed.2d 151 (1973): “[I]n constitutional adjudication as elsewhere, equitable remedies are a special blend oí what is necessary, what is fair, and what is workable. In equity, as nowhere else, courts eschew rigid absolutes and look to the practical realities and necessities inescapably involved in reconciling competing interests, notwithstanding that those interests may have constitutional roots.” Therefore, the question of whether an injunction should issue here is in the sound discretion of this Court. See, e. g., Conner v. Brierley, 57 F.R.D. 144 (E.D.Pa.1972). Under the facts of the instant case this Court has determined that injunctive relief is inappropriate. As the statement of facts demonstrates, the Board has undertaken a series of comprehensive programs to correct the funding disparities between Caucasian schools and non-Caucasian schools. 10 By the 1972-73 school year these programs had reduced the average expenditure disparity to about 1%. The Board has spent substantial sums on these programs. Moreover, the Board has pledged to continue the effort to equalize expenditures. In a case where the racial discrimination which occurred was de facto instead of de jure, and unintentional, these genuinely effective corrective measures are enough; further court action is unnecessary. Both the Rodriguez and Milliken v. Bradley, 418 U.S. 717, 94 S.Ct. 3112, 41 L.Ed.2d 1069 (1974) decisions demonstrate a judicial policy that local control over public education should be maintained wherever possible; since the Board has adequately demonstrated its sincere desire to comply with the law, this Court will not interfere. IV. CONCLUSION In sum, (a) plaintiffs failed to prove that any unconstitutional wealth discrimination occurred here against an identifiable economic class; (b) unjustified expenditure differentials caused unconstitutional racial discrimination between Caucasian and non-Caucasian schools in the 1969-70 and 1970-71 school years; but (e) the Board’s corrective measures make injunctive relief unnecessary. It is therefore ordered that the complaint be dismissed with costs against the defendants. It is so ordered. Footnotes [1] Title I of the Elementary and Secondary Education Act of 1965 declared a Congressional policy of providing federal funds to concentrations of students from poorer families “to expand and improve their educational programs.” 20 U.S.C. § 241a. Eligibility of a school district for Title I funds depends on the district-wide percentage of children from families of low income, families receiving Aid to Families with Dependent Children, and children living in foster homes. 20 U.S.C. § 241c; 45 C.F.R. §§ 116.2 and 116.3. Title I funds are to be used to supplement and not supplant state and local funds, and are not to be included in the determination of average per-pupil expenditures. 20 U.S.C. § 241e; 45 C.F.R. § 116.53. [2] The expenditure differentials between Caucasian and non-Caucasian schools in 1969-70 and 1970-71 seem to be representative of a long-term historical pattern. Harrold M. Barron’s study of the Board’s expenditures for 1961, 1963 and 1966 shows that the racial composition of a school was significantly correlated witli expenditure levels. The Barron study also indicated that expenditure variations within the Chicago school system were caused by clustering of more educated, experienced teachers in white schools. See H. Barron, “Race and Status in School Spending: Chicago, 1961-1966.” 6 The Journal of Human Resources 1 1969. [3] While there may be debate as to the precise effect of money on the educational process, no one denies the fact that money does have some impact. The President’s Commission on School Finance concluded that “there must be fruitful ways to spend money to improve schools, to equalize educational opportunity.” Schools, People, and Money, Final Report of the President’s Commission on School Finance, xi. Yet even if money had no clear impact on the quality of education, the equal protection clause would still require with respect to race that all students be afforded the same chance to discover that fact for themselves. [4] The status of school boards as public corporations in Illinois is unique. The 1970 Illinois constitution creates three types of governmental corporations: municipalities, other units of local government, and school districts. See, e. g., Ill.Const. Art. VIII, §§ 1, 2 and 4. The intent of these three categories is to define different types of home rule powers. Within the meaning of Bruno, however, none of the three types of state agencies appear to be “persons” under § 1983. [5] Two post -Bruno cases explicitly hold that state agencies can be directly sued under the constitution for violation of Fourteenth Amendment rights. See Amen v. City of Dearborn, 363 F.Supp. 1267 (E.D.Mich. 1973); Noe v. County of Lake, 73 H 157 (N.D.Ind., Nov. 1, 1973). Moreover, many cases suggest that constitutional rights, although incapable of exact valuation, are of sufficient value to confer jurisdiction under § 1331. See, e. g., Cortright v. Resor, 325 F.Supp. 797 (E.D.N.Y.1971), rev’d on other grounds, 447 F.2d 245 (2d Cir. 1971); Marquez v. Hardin, 339 F.Supp. 1364 (N.D.Cal. 1969); Martinez v. Richardson, 472 F.2d 1121 (10th Cir. 1973); Bass v. Rockefeller, 331 F.Supp. 945 (S.D.N.Y.1971). See generally, Bodensteiner, “Federal Court Jurisdiction of Suits Against Non-Persons for Deprivation of Constitutional Rights,” 8 Valparaiso U.L.R. 215 (1974). [6] See Bennett, “Liberty, Equality and Welfare Reform,” 68 Nw.U.L.R. 74 (1973) for a summary of the development of the two separate equal protection tests. [7] The Supreme Court would have held the Texas system of school financing unconstitutional if it had involved a suspect classification. 411 U.S. at 16-17, 93 S.Ct. 1278. Rodriguez thus supports the position here of the class of non-Caucasian students who claim to have been discriminated against on the basis of race. [8] The Supreme Court’s treatment of statistical data on racial imbalances can be found through cases involving discrimination in jury selection. The leading case in this area is Norris v. Alabama, 294 U.S. 587, 55 S.Ct. 579, 79 L.Ed. 1074 (1934). There the Court took note of the proportion of blacks on grand and petit juries in relation to the proportion of blacks in the general population and held that the petitioner had thereby established a prima facie case of racial discrimination in the selection of juries. The Norris approach has been followed in a long line of jury selection cases, Whitus v. Georgia, 385 U.S. 545, 87 S.Ct. 643, 17 L.Ed.2d 599 (1966); Sims v. Georgia, 389 U.S. 404, 88 S.Ct. 523, 19 L.Ed.2d 634 (1967); Turner v. Fouche, 396 U.S. 346, 90 S.Ct. 532, 24 L.Ed.2d 567 (1969), and has been applied in related areas such as school desegregation cases. Green v. County School Board, 391 U.S. 430, 88 S.Ct. 1689, 20 L.Ed.2d 716 (1967). For a similar legal analysis, see Griggs v. Duke Power Co., 401 U.S. 424, 91 S.Ct. 849, 28 L.Ed.2d 158 (1971). There, the Court held that when a statistical proof was made that employment practices, including intelligence tests, discriminated against blacks, the burden of proof shifts to the defendant company to show that the questioned practice or test is related to job performance. If the defendant company is unable to make such a proof, the challenged practice must be abandoned because of its discriminatory effect, even though intent is not proved. See also Cypress v. Newport News Gen. Hospital, 375 F.2d 648 (4th Cir. 1967). Thus where statistical proof and historical pattern are coupled together, racial discrimination is proved. In sum, as one court stated: “[I]n the problem of racial discrimination, statistics often tell much, and Courts [will] listen.” Alabama v. United States, 304 F.2d 583 at 586 (5th Cir. 1962), aff’d, 371 U.S. 37, 83 S.Ct. 145, 9 L.Ed.2d 112 (1962). [9] Administrative convenience involved in defendants’ procedures for school staffing and expenditure does not constitute adequate justification. “[T]he constitutional imperatives of the Equal Protection Clause must have priority over the comfortable convenience of the status quo.” Williams v. Illinois, 399 U.S. 235, 245, 90 S.Ct. 2018, 2024, 26 L.Ed.2d 586 (1970). Inequality is not a necessary element of defendants’ functions, for “normal administrative practice should produce schools of like quality, facilities, and staffs.” Swann v. Charlotte-Mecklenburg Bd. of Ed., 402 U.S. 1, 18-19, 91 S.Ct. 1267, 1277, 28 L.Ed.2d 554 (1971). Similarly, the failure to keep pace with substantial growth in the non-Caucasian student population is insufficient to provide the necessary rational basis for defendants’ policies and actions. “[J]ustification must be in terms not of excusing reasons of this stripe but of positive social interests protected or advanced.” Hobson v. Hansen, supra, 269 F.Supp. 401 at 498 (D.D.C.1967). Additionally, defendants cannot justify the differences in school staffing and expenditure on the ground that these inequalities result from defendants’ contracts with the Chicago Teachers Union (Ans. Para. 10 and Third Defense). Responsibility for the operation of the Chicago public schools by law is vested in defendants, not the Union. Ill. Rev.Stat. ch. 122, Art. 34. See Hobson v. Hansen, supra, 269 F.Supp. at 502. Indeed, the collective bargaining agreements upon which defendants attempt to rely explicitly state that such agreements shall not apply where terms would be “inconsistent with constitutional, statutory, or other legal provisions.” [10] These programs have also corrected funding disparities between Title I and non-Title I schools. Thus even if plaintiffs had proved that unconstitutional wealth discrimination occurred, injunctive relief for this portion of the case would also be inappropriate.