Full opinion text
JOHN R. GIBSON, Circuit Judge. This case arises from claims of unconstitutional racial segregation of school children in the Kansas City metropolitan area. After trial, the district court imposed an intradistrict remedy against the Kansas City, Missouri School District (KCMSD) and the State of Missouri with the State bearing approximately three-fourths of the cost. The district court found that the KCMSD and the State had not eradicated vestiges of the racially segregated dual school system once required under state law in violation of the Constitution. The claim for interdistrict relief in the form of consolidation or realignment of the suburban school districts (SSDs) was rejected as the SSDs were found to have eliminated all vestiges of their segregated school systems and there was no finding of interdistrict violation or interdistrict effect. The SSDs were dismissed from the action. The United States Department of Housing and Urban Development (HUD) also was held not liable. In this appeal, both the KCMSD and Kalima Jenkins, the named plaintiff in a certified class of present and future KCMSD students (Jenkins class), challenge the denial of interdistrict relief. The Jenkins class also appeals the judgment in favor of HUD. In addition, both the KCMSD and the State challenge the scope and the allocation of costs of the intradis-trict remedy ordered against each. The district court’s findings and conclusion that the suburban school districts are not liable for interdistrict violation and may not be ordered to participate in interdistrict relief, on those issues not related to housing, are affirmed by the vote of five judges (Judges Ross, Fagg and Wollman join in this opinion; Judge Arnold concurs in the result; Chief Judge Lay, joined by Judges Heaney and McMillian, dissents). The interdistrict findings and conclusion on the housing issues are affirmed by an equally divided court (Judges Ross, Fagg and Wollman join in this opinion; Judge Arnold files a concurring and dissenting opinion in which he concludes that the case should be remanded to the district court to consider whether there are interdistrict housing violations which would require relief; Chief Judge Lay, joined by Judges Heaney and McMillian, dissents). The dismissal of the SSDs and HUD at the close of the plaintiff’s evidence is therefore affirmed. The intradistrict remedy ordered against the State of Missouri and KCMSD is modified in some respects with five judges concurring (Judges Ross, Fagg and Wollman join in the opinion and Judge Arnold concurs in the result; Chief Judge Lay, joined by Judges Heaney and McMillian, dissents). This suit was filed in 1977 by the KCMSD, the School Board, and four children of two School Board members. The complaint alleged that the State, surrounding school districts in Missouri and Kansas, and several federal agencies including HUD had helped cause or had been part of a system of racial segregation among Kansas City metropolitan area school districts. In October 1978, the district court dismissed the Kansas defendants for want of jurisdiction. It also concluded that the KCMSD lacked standing to bring an action against the State as party plaintiff and realigned the KCMSD as a defendant. School District of Kansas City, Missouri v. Missouri, 460 F.Supp. 421 (W.D.Mo. 1978). In May 1979, an amended complaint was filed against KCMSD, and the federal and Missouri defendants named in the original complaint. The amended complaint made two distinct allegations: first, of an inter-district violation, committed by the SSDs, and the State and federal defendants; and second, of an intradistrict violation within the KCMSD, committed by the KCMSD and the State defendants. KCMSD pursued its claim by filing, in July 1979, a cross-claim against the State, reiterating the allegation of interdistrict violation made by the Jenkins class, and seeking indemnification against any intradistrict liability. Trial to the court commenced October 31, 1983. Over the course of 64 trial days, plaintiff called over 140 witnesses, offered 2,100 exhibits, and designated approximately 10,000 pages of depositions. Before hearing evidence in response, the district court, based on standards set out by the Supreme Court in Milliken v. Bradley, 418 U.S. 717, 94 S.Ct. 3112, 41 L.Ed.2d 1069 (1974), dismissed the eleven SSDs from the case under Fed.R.Civ.P. 41(b). The court found that school districts in Missouri are autonomous and that none of the districts had committed any acts with intent to discriminate on the basis of race, to contain blacks in the KCMSD, or bar blacks from the SSDs. The court further found that within four years after Brown v. Board of Education, 347 U.S. 483, 74 S.Ct. 686, 98 L.Ed. 873 (1954) (Brown I), and in most cases within a shorter time, all the SSDs had eradicated the vestiges of their dual school systems and were operating unitary systems. The court further found that none of the alleged discriminatory actions committed by the State or the federal defendants had caused any significant current interdistrict segregation. The district court therefore denied interdistrict relief. See generally Order of June 5, 1984. The district court then heard additional evidence and, on September 17, 1984, issued its order holding the State and the KCMSD liable for racial segregation of students within the KCMSD. Jenkins v. Missouri, 593 F.Supp. 1485 (W.D.Mo.1984). The court observed that before 1954, the State and the KCMSD had maintained a racially segregated dual school system. The court found, in the existence of twenty-four KCMSD schools with a black enrollment in excess of 90%, and in other areas, vestiges of the now unlawful dual school system. It therefore held that the State, and the KCMSD, under principles announced by the Supreme Court in Brown v. Board of Education, 349 U.S. 294, 301, 75 5. Ct. 753, 756-57, 99 L.Ed. 1083 (1955) (Brown II), had failed in their affirmative duty to dismantle the unconstitutional pre1954 system. The court also held that HUD was not liable, finding that HUD had followed a balanced approach to sponsoring subsidized housing projects in both innercity and suburban areas. On June 14, 1985, after a two-week hearing on the scope of appropriate relief, the court, 639 F.Supp. 19 (1985), issued a remedial order requiring the State and the KCMSD to fund compensatory and remedial educational programs and necessary capital improvements in KCMSD schools. The plan is projected to cost $87,000,000 over the next three years, with the State bearing approximately $67,000,000 and the KCMSD approximately $20,000,000. See Memorandum Opinion of June 14, 1985 at 41-42. In this appeal the Jenkins class challenges the district court’s dismissal of HUD, and joined by KCMSD, its dismissal of the SSDs, and its denial of interdistrict relief. The KCMSD also challenges the district court’s realignment of the KCMSD as a party defendant. The State challenges the scope of the district court’s remedy and the allocation of costs. INTERDISTRICT LIABILITY FINDINGS In Milliken, 418 U.S. 717, 94 S.Ct. 3112, the Supreme Court held that an interdis-trict remedy may not be imposed absent a finding of a constitutional violation within one district producing a significant segre-gative effect in another school district. Id. at 744-75, 94 S.Ct. at 3127. Without an interdistrict violation and interdistrict effect, there is no constitutional wrong requiring an interdistrict remedy. Id. Plaintiffs advanced to the district court three theories to support their claim of interdis-trict violation and effect, requiring the imposition of an interdistrict remedy in the form of consolidation of the KCMSD and SSDs: [Fjirst, that the SSDs, as agents of the state, were guilty of operating a regional system of segregated schools that centered on and impacted Kansas City with blacks and made the suburbs whiter before 1954; second, that the SSDs failed in their affirmative duty to eliminate the vestiges of the dual school system; and third, that the SSDs are liable for the effects of actions by the KCMSD and other defendants and can therefore be included in an interdistrict remedy. Order of June 5, 1984 at 5. The district court found, based on the Milliken holding, that there was lack of proof of discriminatory intent in the establishment or change of any school district boundary and thus distinguished the interdistrict cases upon which the plaintiffs relied. It rejected the arguments that the SSDs were not separate and autonomous but were instead agents of the State. It further rejected the argument that, as agents and subdivisions of the State, the SSDs must be included in an interdistrict remedy absent any finding that they had committed specific violations by their own initiative. The court examined the plaintiffs’ argument that the pre-1954 dual school systems had caused racial segregation in the Kansas City metropolitan area. The court acknowledged the substantial increase in the KCMSD black population from 1910 to 1960, but found that the increase resulted principally from the unusual economic and employment ramifications of the World Wars and intervening Depression. Order of June 5, 1984 at 17. The absence of black schools in the SSDs, it found, had not discouraged black families outside or within Missouri from moving to and living in those districts. The district court also rejected the argument that within the SSDs any vestiges or significant effects of the pre-1954 dual school system remained. The acts thirty years past, the court found, had negligible current effects. While some of the SSDs had dual school systems in the pre-1954 era and some did not, after Brown I each had disestablished its pre-1954 school system with deliberate speed and all vestiges of those systems have long since disappeared. The district court entered separate findings with respect to each SSD. It found that all were unitary, most by 1955, in immediate compliance with the mandate of Brown I, two more by the 1957-58 school year, and the last, Park Hill, was fully integrated by the 1959-60 school year. Id. at 43-95. The court then considered the plaintiffs’ second theory, that the SSDs had failed in their duty, pursuant to the mandate of Brown II, to eliminate vestiges. The district court found that the establishment of an association of regional school districts did not negate the autonomy of the individual SSDs, and that the sharing of sales tax revenues by certain of the SSDs whose area may have been partly within the Kansas City, Missouri, city limits did not amount to a constitutional violation. Id. at 19-21. Further, a juvenile home, local special and vocational education programs, and particularly the creation of certain vocational education districts were examined and found not to have been constitutional violations on the part of the SSDs. Id. at 21-26. The district court also examined a statute enacted by the Missouri General Assembly in 1957 enlarging the size of a city that should constitute a single school district. The court found that the enactment was not driven by an intent to concentrate black students within KCMSD, and therefore, was not an intentional interdis-trict violation. Id. at 28-29. The court also examined the SSDs’ response to recommendations of the Spainhower School District Commission of 1979, regarding realignment of boundaries so as to create some twenty school districts in the state, and with respect to financing and local control, and found no evidence that any of the SSDs rejected or declined to consider these proposals for reasons at all related to race. It further rejected arguments based upon employment practices in the SSDs. The district court finally analyzed in detail plaintiffs’ third theory, that the SSDs may be included in an interdistrict remedy because the effects of others’ discriminatory acts were felt within the SSDs. It found no evidence that white families had been lured to or black families discouraged from living in the SSDs. Nor did any of the factual evidence concerning housing implicate any of the SSD defendants. Id. at 34-35. The steady increase in black enrollment in the SSDs since 1968, particularly in Raytown, Hickman Mills, Center, and Grandview districts, and the fact that in 1980, 23,434 blacks resided outside the principal contiguous area was found to refute the testimony of plaintiffs’ experts. The court rejected white-flight theory as a constitutional violation by the SSDs. Id. at 36-39. The district court also addressed plaintiffs’ claims regarding housing violations. While the court found that before 1948, in several areas within the SSDs restrictive covenants were enforced, the court further found that such enforcement did not have a current significant segregative effect. With respect to public housing, it found that such housing is available in significant amounts within every SSD and that the HUD section 8 program is available in all parts of the metropolitan area. No testimony linked any SSD to the transition of neighborhoods within the KCMSD from white to black and the accompanying white flight which had taken place in Kansas City since 1940. Id. at 41. The court therefore concluded that there was no interdistrict violation and interdis-trict effect. Thus, as required by Milliken, interdistrict relief was denied. Id. at 95. The court stressed that awarding relief “ ‘would impose on the outlying districts, not shown to have committed any constitutional violation, a wholly impermissible remedy based on a standard not hinted at in Brown I and II or any holding of this Court.’ ” Id. at 96 (quoting Milliken, 418 U.S. at 746, 94 S.Ct. at 3128). It rejected a cumulative effect and found plaintiffs’ evidence de minimis at most and legally insufficient to justify the relief sought against the SSDs. Id. at 97-98. The court thus dismissed the case against the SSDs. INTRADISTRICT LIABILITY FINDINGS The district court then tried the intradis-trict issues and found that the inferior education indigenous to the State-compelled dual school system has lingering effects in the KCMSD. Jenkins v. Missouri, 593 F.Supp. at 1492. The district court recognized the stipulation that before 1977 KCMSD was not unitary and did not comply with federal regulations. Id. at 1489. KCMSD was majority white in enrollment until 1970 and could have achieved mathematical racial balance in its schools. Id. at 1492. Instead, it chose to operate some completely segregated schools and other integrated ones. In 1974, twenty years after Brown I, 39 schools were more than 90% black; another 38 had 10% to 90% black enrollment. Eighty percent of all blacks in the district attended schools that were 90% black. Only 19% of the blacks attended a school that was 10% to 90% black. Id. at 1492-93. As of 1977, 25 one-race schools under the pre-1954 system remained 90% or more of the same race. Id. In addition, four schools that were black under the dual system were predominantly black when closed in 1968. Id. at 1492. In the 1983-84 school year no KCMSD school had less than 30% black enrollment, but 24 schools were racially isolated at more than 90% black. Id. at 1493. The court found that the KCMSD still has not entirely dismantled the dual school system. The court then made particular findings on the liberal transfer policy, the neighborhood school policy, and attendance zones, and found that these, rather than facilitating integration, fostered the maintenance of segregated attendance patterns. Intact busing was found to have been seg-regative in intent and effect, but, as it was stopped in the 1960’s, no continuing violation existed. The court found that plans had not been adopted for district-wide desegregation until 1977-78. Id. at 1493-94. Similarly, the court imposed liability on the State based on its intentional creation of the dual school system and the obligation that existed to disestablish such a system. As vestiges of the State’s dual school system lingered in KCMSD, the obligations of KCMSD and the State had not been met. The court rejected the State’s argument that constraints imposed by the state constitution or statutes prevented its affirmative disestablishment of the ■ dual school system. It held the issues in favor of the Jenkins class and against KCMSD and the State and held the issues in favor of KCMSD and against the State on the cross-claim. Id. at 1505. LIABILITY OF HUD With respect to the claim against HUD, the court examined Federal Housing Administration appraisal practices before 1949, cooperative agreements with the City of Kansas City under the auspices of the Housing Authority of Kansas City (HAKC) and the Land Clearance for Redevelopment Authority (LCRA), and section 8 certificates. It found that HUD followed a balanced housing policy and attempted to insure that assisted housing was located in both innercity and suburban areas. There was no evidence that HUD’s site selection and approval practices for federally subsidized multi-family housing substantially affected the racial composition of schools within KCMSD. Id. at 1499. The court pointed to HUD’s affirmative marketing efforts to apprise participating section 8 certificate-holders that affordable housing was available throughout the entire community. Id. at 1500. There was no evidence that HUD routinely denied blacks FHA mortgages, that it discriminatorily foreclosed on mortgages held by blacks, or that it avoided selling the homes on which its insured mortgages had been foreclosed to blacks. Further, there was no evidence that neglect of HUD-held homes was racially discriminatory in intent or purpose. The issues were held in favor of HUD. I. We have had earlier occasion to deal with the legal standards governing interdistrict school desegregation cases. See, e.g., Little Rock School District v. Pulaski County Special School District No. 1, 778 F.2d 404 (8th Cir.1985); Liddell v. Missouri, 731 F.2d 1294 (8th Cir.), cert. denied, 469 U.S. 816, 105 S.Ct. 82, 83 L.Ed.2d 30 (1984). We draw our basic guidance from the Supreme Court’s decision in Milliken v. Bradley, 418 U.S. 717, 94 S.Ct. 3112, 41 L.Ed.2d 1069 (1974): Before the boundaries of separate and autonomous school districts may be set aside by consolidating the separate units for remedial purposes or by imposing a cross-district remedy, it must first be shown that there has been a constitutional violation within one district that produces a significant segregative effect in another district. Specifically, it must be shown that racially discriminatory acts of the state or local school districts, or of a single school district have been a substantial cause of interdistrict segregation. Thus an interdistrict remedy might be in order where the racially discriminatory acts of one or more school districts caused racial segregation in an adjacent district, or where district lines have been deliberately drawn on the basis of race. In such circumstances an interdistrict remedy would be appropriate to eliminate the interdistrict segregation directly caused by the constitutional violation. Conversely, without an interdistrict violation and interdistrict effect, there is no constitutional wrong calling for an inter-district remedy. 418 U.S. 744-75, 94 S.Ct. at 3127 (citations omitted). Two courts of appeals have read this language to require clear proof of cause and effect of a constitutional violation, and a careful delineation of the extent of the effect, before an interdistrict remedy may be invoked. Goldsboro City Board of Education v. Wayne County Board of Education, 745 F.2d 324, 332 (4th Cir.1984); Lee v. Lee County Board of Education, 639 F.2d 1243, 1256 (5th Cir.1981). As the Fifth Circuit stated in Lee: We believe the Court’s deliberate choice of phrases such as “substantial” or “direct cause” and “significant segregative effect” also expresses an insistence that in cases where an interdistrict remedy is requested, there must be clear proof of cause and effect and a careful delineation of the extent of the effect. In the absence of such a showing, school district lines are to be carefully observed and desegregation remedies confined to orders affecting the school district in which the condition of segregation is manifest. Id. at 1256. In addition to “clear proof” of the interdistrict violation and its interdistrict effect and “a careful delineation” of the extent of the interdistrict effects, it must be shown that the interdistrict segregative effects are current. Lee, 639 F.2d at 1260. Federal courts may not invoke their equitable power to fashion a remedy to correct a condition unless it currently offends the Constitution. As the Court stated in Milliken: “A federal remedial power may be exercised ‘only on the basis of a constitutional violation' and, ‘[a]s with any equity case, the nature of the violation determines the scope of the remedy.’ ” Id. at 738, 94 S.Ct. at 3124 (quoting Swann v. Charlotte-Mecklenburg Board of Education, 402 U.S. 1, 16, 91 S.Ct. 1267, 1276, 28 L.Ed.2d 554 (1971)); see also General Building Contractors v. Pennsylvania, 458 U.S. 375, 399, 102 S.Ct. 3141, 3154-55, 73 L.Ed.2d 835 (1982) (a remedial decree should “extend no farther than required by the nature and the extent of that violation”). II. We must also recognize at the outset the importance of the district court’s factual findings in school desegregation cases. As we stated in Little Rock: We will not reverse the district court’s factual findings with respect to liability unless we conclude that they are clearly erroneous. Fed.R.Civ.P. 52(a); Anderson v. City of Bessemer City [470 U.S. 564], 105 S.Ct. 1504 [84 L.Ed.2d 518] (1985); Pullman-Standard v. Swint, 456 U.S. 273, 287-90 [102 S.Ct. 1781,1789-91, 72 L.Ed.2d 66] (1982); Dayton II [v. Brinkman], 443 U.S. [526] at 534 n. 8 [99 S.Ct. 2971 at 2977 n. 8, 61 L.Ed.2d 720 (1979)]; Columbus Board of Education v. Penick, 443 U.S. [449] at 468-71 [99 S.Ct. 2941 at 2952, 2983, 61 L.Ed.2d 666 (1979) ] (concurring opinions of Burger, C.J., and Stewart, J.); United States v. United States Gypsum Co., 333 U.S. 364, 395 [68 S.Ct. 525, 542, 92 L.Ed. 746] (1978). Nor will we reverse such findings when they are based on inferences from other facts unless the rigorous standards of the same rule are met. Anderson, 105 S.Ct. at 1511. The Supreme Court has emphasized the importance of the clearly erroneous rule in civil rights cases, see, e.g., Pullman-Standard v. Swint, 456 U.S. at 287-90 [102 S.Ct. at 1789-91], and, more particularly, in school desegregation cases: The elimination of the more conspicuous forms of governmentally ordained racial segregation * * * counsels undiminished deference to the factual adjudications of the federal trial judges in cases such as these, uniquely situated as those judges are to appraise the societal forces at work in the communities where they sit. Columbus Board of Education v. Pen-ick, 443 U.S. at 449, 470 [99 S.Ct. at 2941, 2983] (1979) (Stewart, J., concurring, with whom Burger, C.J., joins, concurring). Little Rock, 778 F.2d at 410-11. See Morrilton School District No. 32 v. United States, 606 F.2d 222, 230 (8th Cir.1979), cert. denied, 444 U.S. 1071, 100 S.Ct. 1015, 62 L.Ed.2d 753 (1980); see also Riddick v. School Board of the City of Norfolk, 784 F.2d 521, 533 (4th Cir.1986) (factual findings by a district court in school desegregation cases are entitled to great deference on review “especially where the presiding judicial officer has lived with a case for many years”). The Supreme Court in Anderson v. City of Bessemer City, 470 U.S. 564, 105 S.Ct. 1504, 84 L.Ed.2d 518 (1985), has explained that the trial judge’s role is to determine fact, and duplication of these efforts in the courts of appeals “would very likely contribute only negligibly to the accuracy of fact determination.” Id., 105 S.Ct. at 1512. The Court stated: The reviewing court oversteps the bounds of its duty under Rule 52 if it undertakes to duplicate the role of the lower court. “In applying the clearly erroneous standard to the findings of a district court sitting without a jury, appellate courts must constantly have in mind that their function is not to decide factual issues de novo.” Zenith Radio Corp. v. Hazeltine Research, Inc., 395 U.S. 100, 123 [89 S.Ct 1562, 1576, 23 L.Ed.2d 129] (1969). If the district court’s account of the evidence is plausible in light of the record viewed in its entirety, the court of appeals may not reverse it even though convinced that had it been sitting as the trier of fact, it would have weighed the evidence differently. Where there are two permissible views of the evidence, the factfinder’s choice between them cannot be clearly erroneous. Id. at 1511-12. Chief Judge Lay’s dissent, in violation of the teachings of Anderson, duplicates the role of the district court and is an exercise in appellate factfinding. A reading of the dissent is sufficient to demonstrate the rejection of the district court’s carefully weighed findings of fact and substitution of those more desirable, frequently based on evidence considered and rejected by the district court. While on some occasions placing the mantle of the clearly erroneous rule about the discussion, it is evident that Chief Judge Lay was engaging in an original evaluation of the voluminous and ofttimes contradictory record seeking an ultimate conclusion more satisfactory than that reached by the district court. This is directly contrary to the mandate of Anderson. III. The Jenkins class does not challenge the findings of the district court. Rather, it argues that the findings establish continuing interlocking, interdistrict violations whose cross-district nature and metropolitan-wide scope require relief encompassing the SSDs. It argues particularly that there is an interdistrict violation or effect under Milliken and Hills v. Gautreaux, 425 U.S. 284, 96 S.Ct. 1538, 47 L.Ed.2d 792 (1976), and points to several independent bases for interdistrict relief. These are: the pre-1954 interdistrict system of locating dual schools; the State’s and the SSDs continuing interdistrict violations; KCMSD’s officially sanctioned suburban flight violation; the State’s market-wide dual housing systems; and LCRA and HAKC’s racial steering and siting violations. The Jenkins class further argues that the district court denied interdistrict relief based on a “concatenation of legal error” as to interdistrict liability. It argues that in six respects the district court abandoned controlling legal principles in concluding that: the existence of dual schools before 1954 does not now amount to a constitutional violation; the SSDs were autonomous and independent; the SSDs were absolved of liability for their pre-1954 violations and erroneously ignoring the post-1954 segregative acts and omissions; without a finding of fault, the SSDs may not be required to take part in eliminating the effects on children in their districts of the State's metropolitan-wide dual school and housing violations; the SSDs did not invidiously entice white families to move into their district, and that unconstitutional actions within one district making it blacker which reciprocally affects adjacent districts by making them whiter may not provide an independent basis for interdistrict relief; and absent intentional segregative acts by each school board sought to be included, a finding of liability against actors in the housing market is irrelevant to the SSDs. The housing argument will be discussed separately in IV and V. Finally, the Jenkins class argues that the district court applied an improper burden and standard of proof of significant effects in an improperly piecemeal fashion. It further argues that the district court improperly fragmented the evidence of unconstitutional acts rather than evaluating the evidence as a whole. It argues that the district court’s finding of six distinct metropolitan-wide constitutional violations satisfies Milliken’s significant segregative effect standard. It urges under Swann that proof of a system with a history of segregation warrants a presumption against schools that remain disproportionate in their racial composition. It argues that the proof acknowledged by the court, of prior de jure segregation on an area-wide basis, coupled with a patently continuing racial imbalance among the SSDs’ student and faculty compositions, established the liability of the SSDs. It further argues that the district court improperly rejected much of the evidence, and improperly barred the presentation of quantitative evidence, as irrelevant. The broad scope of these arguments is considerably narrowed when viewed in light of significant factual findings of the district court. These findings are not challenged as clearly erroneous. The district court found that the SSDs were autonomous and locally controlled, and that the State was powerless to require a merger or consolidation. The district court recognized the stipulation that before 1954 the State mandated dual school systems and mandated SSDs’ compliance in maintaining dual systems. It found that all SSDs met their constitutional obligation to operate a unitary school system within a maximum of four years after Brown I, most of them doing so within one year. The dismantling of the dual system was accomplished with all deliberate speed. Order of June 5, 1984 at 99. It specifically found that the evidence established that there was no barrier to movement of blacks into the SSDs. Id. at 39. It found the pre-Brown acts thirty years past to have negligible present effects. It further found that the faculty and staff composition of the SSDs do not affect the racial composition of the student bodies. The court found the faculty were hired and promoted on a racially neutral basis. The court could not conclude that there was an interdistrict violation from any possible mtradistrict factors. The factual argument based on the pre-Brown interdistrict system of locating dual schools. flies in the face of the district court’s finding that the pre-Brown acts have negligible present effects. Chief Judge Lay’s dissent suffers from a similar infirmity as the district court made numerous findings interspersed through its order concerning the negligible and de minimis nature of any such effects. These findings are not clearly erroneous. These findings make unnecessary a detailed recitation of the pre-1954 evidence. The argument that the State and SSDs participated in continuing interdistrict violations is contrary to the district court’s finding that the SSDs had met their constitutional obligation to operate unitary school systems within four years after Brown I, most of them doing so within a year. Similarly, the argument that KCMSD officially sanctioned suburban flight looks first to KCMSD’s violation which the district court clearly found to be only intradistrict in nature. The argument based on flight into neighboring SSDs making the southeast area schools blacker and recipient SSD schools whiter, thereby causing racial segregation in adjacent districts, is contrary to the district court’s finding that there was no barrier of movement to blacks to the SSDs. The district court specifically found that there was considerable increase in the black enrollment of the SSDs, particularly Center, Grandview, Hickman Mills, and Raytown. Similarly, in the enumeration of six legal errors, three are based upon pre-1954 actions of the SSDs. The district court’s findings that these acts have negligible present effect fundamentally undermines the validity of these arguments. As we have seen, the argument based on the actions of KCMSD and white flight run contrary to the factual findings mentioned above. So also, the argument that the SSDs, though not at fault, would still be liable to take part in eliminating the effects on children in their districts of the State’s metropolitan-wide dual school violations, suffers from two fatal deficiencies. It ignores the lack of finding of any such effects, and is contrary to the district court’s findings that within at most four years after Brown I, any vestiges of the dual school systems that may have existed in the SSDs had been eliminated. The Jenkins class further argues that Swann raises a presumption against schools that are disproportionate in their racial composition. The district court did not err in rejecting the Swann presumption. Milliken was careful to point out that disparity in the racial composition of pupils within a single district merely signals an inquiry into the causes for pronounced racial identifiability of schools within one school system. The Milliken limitation of this presumption is the subject of comment by the Fifth Circuit in Lee v. Lee County Board of Education, where the court stated: It seems important to note also that Milliken, unlike Swann and Keyes v. School Dist. No. 1, 413 U.S. 189 [93 S.Ct. 2686, 37 L.Ed.2d 548] (1973), did not sanction the use of any presumptions on the question of the cause of interdistrict segregation. Thus, while Swann, as noted above, permits an inference that the continued existence of one-race schools in a system that formerly practiced de jure segregation is a vestige of such segregation, and Keyes permits one to infer the existence of systemwide de jure segregation from proof that school authorities have pursued an intentional policy of segregation in a substantial portion of a school district, Milliken refused to sanction a presumption that significant disparities in the racial composition of autonomous school districts resulted from impermissible action by those districts and thus justified imposing upon them the burden of remedying conditions of segregation existing in other districts. The Milliken Court noted that both Keyes and Swann merely involved “the use of a significant racial imbalance in schools within an autonomous school district as a signal which operates to shift the burden of proof [which] is a very different matter from equating racial imbalance with a constitutional violation calling for a remedy.” 418 U.S. at 741 n. 19 [94 S.Ct. at 3125 n. 19] * * *. 639 F.2d at 1254. Contrary to the argument of the Jenkins class, cases of this court do not adopt the Swann presumption in interdistrict cases. See Morrilton School District No. 32 v. United States, 606 F.2d at 230; see also United States v. Missouri, 515 F.2d 1365 (8th Cir.), cert. denied, 423 U.S. 951, 96 S.Ct. 374, 46 L.Ed.2d 288 (1975); Haney v. County Board of Education, 410 F.2d 920 (8th Cir.1969). They simply recognize that racial gerrymandering, which the district court specifically found is absent here, will make school authorities responsible for the foreseeable effects. So also, the argument concerning the presumption runs directly contrary to the teachings of Washington v. Davis, 426 U.S. 229, 240, 96 S.Ct. 2040, 2047-48, 48 L.Ed.2d 597 (1976), that there must be a showing in a school desegregation case of “a current condition of segregation” resulting from intentional state action. The district court’s finding that the SSDs were autonomous prevents a conclusion that there is a single system to which such a presumption applies. The arguments of the Jenkins class run contrary to the facts found by the district court. In essence, in the detailed recitation of evidence, it seeks to have this court substitute its judgment for that of the district court. This is contrary to the general rule of Anderson v. City of Bessemer City, supra, and to the specific dictate of this court that we give “undiminished deference to the factual adjudications of federal trial judges in cases such as these, uniquely situated as those judges are to appraise the societal forces at work in the communities where they sit.” Little Rock, 778 F.2d at 410-11. IV. We next consider arguments advanced by the Jenkins class that racially discriminatory acts by the State in housing-related areas justified interdistrict relief involving the SSDs. Before considering the district court’s findings on this matter, we must address the class’ argument that the district court misread Milliken in deciding whether the SSDs could be required to participate in an interdistrict remedy. In doing so, we must read all of Milliken and all of the district court’s order. According to the class, the district court read Milliken as holding that each SSD could not be required to participate in an interdistrict remedy absent a finding that that SSD had engaged in racially discriminatory conduct. The passage from Milliken set forth above in Part I, supra, made plain that, before a federal court may impose an inter-district remedy, “it must first be shown that there has been a constitutional violation within one district that produces a significant segregative effect in another district.” 418 U.S. at 744-45, 94 S.Ct. at 3127. Specifically, the court must find that racially discriminatory acts “have been a substantial cause of interdistrict segregation.” Id. at 745, 94 S.Ct. at 3127. Absent this showing, “there is no constitutional wrong calling for an interdistrict remedy.” Id. On the facts before it, the Court was forced to conclude in Milliken that: With no showing of significant violation by the 53 outlying school districts and no evidence of any interdistrict violation or effect, the court went beyond the original theory of the case as framed by the pleadings and mandated a metropolitan area remedy. To approve the remedy ordered by the court would impose on the outlying districts, not shown to have committed any constitutional violation, a wholly impermissible remedy based on a standard not hinted at in Brown I and II or any holding of this Court. Id. The Court’s decision in Hills v. Gau-treaux, 425 U.S. 284, 96 S.Ct. 1538, two years later, forcefully emphasized this central holding of Milliken. Justice Stewart wrote for the Court: Once a constitutional violation is found, a federal court is required to tailor “the scope of the remedy” to fit “the nature and extent of the constitutional violation.” * * * In Milliken, there was no finding of unconstitutional action on the part of the suburban school officials and no demonstration that the violations committed in the operation of the Detroit school system had had any significant segregative effects in the suburbs. * * * The desegregation order in Milliken requiring the consolidation of local school districts in the Detroit metropolitan area thus constituted direct federal judicial interference with local governmental entities without the necessary predicate of a constitutional violation by those entities or of the identification within them of any significant segregative effects resulting from the Detroit school officials’ unconstitutional conduct. Under these circumstances, the Court held that the interdistrict decree was impermissible because it was not commensurate with the constitutional violation to be repaired. Id. at 293-94, 96 S.Ct. at 1544-45 (emphasis added) (citations omitted); see also Golds-boro City Board of Education v. Wayne County Board of Education, 745 F.2d at 328 (“An independent school district which has not caused segregation in a neighboring independent district has no duty to rectify a racial imbalance in the other districts.”). Milliken and Hills make clear that we may grant interdistrict relief only to remedy a constitutional violation committed by the SSD, or to remedy an inter-district effect in the SSD caused by a constitutional violation in KCMSD. The district court first carefully considered whether the SSDs had violated the Constitution. It found they had not. Indeed, the district court concluded that to award relief in this case would, as the Court concluded in Mil-liken, “impose on the outlying districts, not shown to have committed any constitutional violation, a wholly impermissible remedy.” 418 U.S. at 745, 94 S.Ct. at 3127. Order of June 5, 1984 at 96. It further found that the Jenkins class had not established any acts or omissions by the SSDs “which have had a substantial segregative impact in any other district.” The district court was compelled to make this inquiry by Milliken and Hills, stressing as they do the significance of the commission of a constitutional violation. The court’s consideration of the appropriateness of interdistrict relief would have been incomplete without this inquiry. Milliken makes plain that the issue of whether the parties are constitutional violators must be considered as well as the issue of whether there is a constitutional violation in one district causing a significant segregative effect in a neighboring district. Moreover, contrary to the argument of the class that the court looked only to the culpability of the SSDs, the scope of the order is far broader. The court’s order admittedly emphasizes the absence of culpability of the suburban districts. However, the court explicitly recognized that under Milliken “there must be evidence of a constitutional violation in one district that produces a significant segregative effect in another district.” Order of June 5,1984 at 14, 95. The district court compared the theories before it with those in Milliken. It noted that only the schools in one district were affected and that the remedy must be limited to that system. In examining the cause and effect issue, the court noted that “not only is plaintiff’s evidence here blurred as to cause and effect, there is no ‘careful delineation of the extent of the effect.’ ” Id. at 96 (quoting Lee, 639 F.2d at 1256). The district court thus dealt not only with the issue of whether the SSDs were constitutional violators but also whether there were significant interdistrict segregative effects. See V, infra. When it did so, it made specific findings that negate current significant interdistrict effects, and concluded that the requirements of Milliken had not been met. In Bell v. Board of Education, 683 F.2d 963 (6th Cir.1982), the Sixth Circuit suggested a number of practical problems in attempting to order school districts to remedy housing violations: We do not find any case addressing the argument that a school board otherwise innocent of segregative intent is liable for the discriminatory housing practices of other governmental agencies. We decline to accept this argument. Under this argument the discriminatory conduct of the FHA in making housing loans and local housing authorities in the construction and rental of public housing is attributable to school boards. Such a proposal places too heavy a burden on the schools to remedy wrongs for which they are no more or less responsible than the plaintiffs, the courts, the churches, the Congress or other institutions. Plaintiffs do not suggest how the schools, after a finding of liability, would go about remedying this problem or what kind of order a federal court could enter that might as a practical matter have a chance of changing the fact that black and white families live in separate neighborhoods in most towns and cities. Id. at 968 (footnote omitted). The Supreme Court has given consistent admonition in Hills and Milliken that federal courts may not “restructur[e] the operation of local governmental entities that were not implicated in any constitutional violation.” 425 U.S. at 296, 96 S.Ct. at 1546. The district court’s findings amply establish that the SSDs were not “implicated” in the State’s housing violations. See Part V. We have considered that culpability is an important factor in interdistrict analysis. In Little Rock, 778 F.2d 404, we ordered that the boundaries of the North Little Rock School District remain intact “in partial recognition of the fact that the nature and extent of its interdistrict violations are less severe than those of the other defendants.” Id. at 435. Judge Arnold’s separate concurring and dissenting opinion recognizes that the remedy to be imposed against the SSDs must be “limited by the fact that the SSD is not itself a constitutional violator.” Post at 694. We recognize that Morrilton School District No. 32 v. United States, 606 F.2d 222, states that school districts may be required to participate in a remedy when “there was no evidence implicating them in a direct way with the establishment” of the segregated district. Id. at 228. Morrilton is a gerrymandering case. Its broad language must be limited to its facts, an interdistrict violation in the consolidation of school districts, with the effects felt in each. An attempt to imbue Morrilton with broader force disregards not only its salient facts, but also the Seventh Circuit precedent it quotes: “If the state has contributed to the separation of the races, it has the obligation to remedy the constitutional violations. That remedy may include school districts which are its instrumentalities and which were the product of the violation.” Morrilton, 606 F.2d at 229 (quoting United States v. Board of School Commissioners, 573 F.2d 400, 410 (7th Cir.), cert. denied sub nom. Bowen v. United States, 439 U.S. 824, 99 S.Ct. 93, 58 L.Ed.2d 116 (1978)). Here the district court clearly found that the SSDs were autonomous, not mere instrumentalities of the State, and their boundaries not the product of any constitutional violation by the State. Despite plaintiffs’ urging, the language of Morrilton is limited to its facts, and does not reach the issues we face today. y. The Jenkins class argues that a dual housing violation, fueled by State enforcement of racially restricted covenants and felt in virtually every other housing program in the three-county area touching lending, appraisal, and sales practices, effectively channelled black and white low-income families to separate communities. Such violation and effects, it argues, were interdistrict, and therefore require the in-terdistrict remedy of consolidation. The Jenkins class further argues that as part and parcel of the State’s dual housing market violations, there were constitutional violations by the LCRA and HAKC, which diverted blacks exclusively to the black areas within the KCMSD. The district court made a general finding that not only is the evidence “blurred as to the cause and effect, there was no ‘careful delineation of the extent of the effect.’ ” Order of June 5, 1984 at 96. It made specific findings that negate the existence of significant interdistrict current effects in the SSDs. As Milliken requires an interdistrict violation causing a significant segregative effect in another district, these findings pose an insuperable barrier to granting interdistrict relief. The district court found that school district boundaries had not constrained black movement in any way, id. at 39, and no acts of the SSDs contained blacks in KCMSD or kept them out of the SSDs. Id. at 3. The absence of black schools in the defendant districts before Brown did not discourage black families from outside or within Missouri from moving to and living in those districts. Id. at 18. The district court found that there was steady increase in black enrollments in the SSDs since 1968, and in 1980, 23,434 blacks resided outside the principal contiguous area. Id. at 36-37. We have observed the increase in black enrollment in Hickman Mills, Center, Grandview, and Raytown. See note 9, supra. The court found that the substantial increases in the KCMSD black population were in large part due to the unusual economic and employment ramifications of the World Wars and intervening Depression. Id. at 17. None of the SSDs were shown by the expert Dr. Tobin to have played any significant role in housing matters or to have exerted any control whatsoever over the private decisions people made about where to live. Id. at 37-38. The district court made numerous findings concerning housing choices, economics, and job opportunities. Jenkins v. Missouri, 593 F.Supp. at 1490. In rejecting the expert testimony of Dr. Kane, the court pointed to the importance of job location, ethnic clustering, and personal preference. Id. at 1491. These findings demonstrate the importance of personal choice, referred to by at least two courts as “voting with feet.” See Riddick, 784 F.2d at 537; Davis v. East Baton Rouge Parish School Board, 721 F.2d 1425, 1435 (5th Cir.1983). Regarding the existence of racially restrictive covenants, stressed by the Jenkins class and by Judge Arnold in his concurring and dissenting opinion, the district court made it plain that the restrictive covenants have no current effect in the SSDs. The district court found no evidence that the covenants were enforced by state courts following Shelley v. Kraemer, 334 U.S. 1, 68 S.Ct. 836, 92 L.Ed. 1161 (1948). It recognized the tremendous housing growth in many suburban areas after 1948, when the covenants were a nullity. That blacks presently live in locations previously covered by restrictive covenants further undermines their significance. Order of June 5, 1984 at 39. These findings of the district court that we have outlined above deal with conditions or effects that would have been expected had there been an interdistrict effect in the SSDs flowing from housing violations committed by the State or other actors in KCMSD. Judge Arnold’s concurring and dissenting opinion argues, as does the Jenkins class, that Evans v. Buchanan, 393 F.Supp. 428 (D.GDel.), aff'd 423 U.S. 963, 96 S.Ct. 381, 46 L.Ed.2d 293 (1975), and United States v. Board of School Commissioners, 637 F.2d 1101 (7th Cir.), cert. denied, 449 U.S. 838, 101 S.Ct. 114, 66 L.Ed.2d 45 (1980), both approved interdis-trict remedies based in part upon housing discrimination practiced by state actors. Post at 689. In Evans, the district court found that governmental authorities had provided public housing almost exclusively within the confines of Wilmington and restricted the availability of private and public housing to blacks in suburban New Castle County. 393 F.Supp. at 435. In Board of School Commissioners, the Seventh Circuit affirmed a finding that all public housing in the county had been located within the boundaries of the Indianapolis public school district and this was the segregative intent of state agencies. 637 F.2d at 1110-11. In this case the district court findings are far different. There was no lack of balance between federally assisted housing within the KCMSD and the suburban areas, as demonstrated by the 6,832 HUD-insured or subsidized multi-family units within KCMSD and 9,872 such units in the eleven SSDs. Jenkins v. Missouri, 593 F.Supp. at 1499. HUD efforts were designed to insure that participating section 8 certificate-holders were aware that affordable housing was available throughout the entire community and that they were informed of their rights under the fair housing laws. Id. at 1500. The court pointed to the affirmative marketing efforts in the Parvin Estates area, located in predominantly white North Kansas City, which were never able to attract more than 12% minority residents. Id. The district court further found no evidence that the practices of HUD in site selection and approval for federally subsidized multi-family housing had a substantial effect upon the racial make-up of schools in KCMSD. Id. at 1499. The findings contained in the analysis of HUD’s liability also relate to LCRA and HAKC. The facts relied on to support the housing violations in Evans and Board of School Commissioners are directly contrary to those found in this case. These cases simply give no assistance to the argument that there have been constitutional violations by housing actors having a significant segregative effect in the SSDs. Other findings on housing practices are significant. The district court specifically found that FHA appraisal practices before 1950 had at most a de minimis effect on present racial housing patterns. It examined LCRA practices and its role in administering urban renewal programs under the Housing Act of 1949. It found that HUD had investigated LCRA and issued a report in April 1972 finding discrimination in LCRA’s referral practices and in requiring that such practices be ceased and reports filed. Failure to correct this led in June 1973 to refusal to fund the program and a cooperative agreement between HUD, LCRA, and the City of Kansas City by which the city assumed responsibility for relocation under the Act. The relocation report showed that during the period from 1971 to 1976, only 174 black families were relocated. Jenkins v. Missouri, 593 F.Supp. at 1497-98. Again, the finding of the district court concerns and is limited to intradistrict claims. This finding is not sufficient to justify remand to the district court for consideration, as Judge Arnold suggests, of what interdistrict effect may have resulted from such actions. The district court found that while the HAKC constructed seven family projects between 1952 and 1963 within the model cities or urban renewal areas, the location was in accordance with congressional acts and the assignment practices were approved by HUD. Id. at 1498-99. The court reviewed HUD policies and found expenditures to be balanced between KCMSD and the suburban areas. Id. at 1499. It further found no evidence that the practices of HUD in site selection and approval for federally subsidized multifamily housing substantially affected the racial make-up of schools within the KCMSD. Id. These findings clearly were made with reference to the claims against HUD. They do not apply to the housing practices in KCMSD. They belie the possibility of interdistrict effect occurring in the SSDs. Another weakness immediately is apparent in analyzing the Jenkins class’ arguments in the context of the district court’s findings. The activities of LCRA and HAKC to which attention is primarily pointed, the investigation and the change in practices, both occurred in the 1970’s after KCMSD had experienced the substantial increase in black school population. The exhibit relied upon by the district court in its findings demonstrates that the KCMSD black population was 18.9% in 1955-56, that it had grown to 30.7% in 1961-62, and had reached 50.2% in the 1970-71 school year. The substantial increase which resulted in the district becoming majority black thus had .occurred before the particular events to which the Jenkins class points. To argue that alleged LCRA and HAKC violations created these interdistrict effects simply ignores logic. The court specifically found that KCMSD assignments of children to particular schools had an intradistrict effect and were not influenced by any SSD. Further, these actions did not have a significant effect on the enrollment in any SSD. Order of June 5, 1984 at 38. The court considered many non-governmental housing factors and rejected the theory that liability may be imposed upon the SSDs for being the recipients of people moving for whatever reason. The court found that red-lining, steering, and blockbusting practices by private real estate agents, which may have affected the racial composition of KCMSD, were not the actions of KCMSD or of any SSD and are beyond the control of any school district. The transition of neighborhoods from white to black and accompanying white flight existed in Kansas City since the 1940’s, but no testimony linked the SSDs to the process. Id. at 41. Absent a nexus between the conduct of the SSDs and the policies or practices of independent housing actors, the court would not hold the SSDs liable for racial imbalance. Id. at 42. Judge Arnold stresses the district court’s findings that there is a dual housing market impacting blacks in the KCMSD and causing the public schools to swell in black enrollment, 593 F.Supp. at 1491, and that the State had encouraged racial discrimination by private individuals. Id. at 1503. The court went no further but commented that the question of State liability would be close if it hinged solely on this encouragement of private discrimination. It then proceeded to base its finding against the State on the intentional creation of the dual school system in KCMSD and the failure to eradicate its vestiges. The Jenkins class makes much of these statements by the district court. Nevertheless, they fall short of making a specific finding of a constitutional violation, and certainly make no finding of any interdistrict effect. That the findings are in the district court’s separate opinion concerning intradistrict liability prevents extension of such findings to arguments regarding interdistrict effect in the SSDs. Thus, even accepting the Jenkins class’ arguments that the district court misread Milliken by improperly limiting its inquiry solely to whether the SSDs were guilty of committing constitutional violations, we conclude that the factual findings of the district court effectively foreclose findings of interdistrict effect in the SSDs flowing from constitutional violations with respect to housing that may have occurred in KCMSD. It is, of course, evident from the findings that there were no constitutional violations in the SSDs. Under Mil-liken this answers the arguments made by the Jenkins class, as well as the further articulation in the arguments of KCMSD, and forecloses efforts to require the SSDs to be subject to the claim for interdistrict relief, be it consolidation or some step less drastic. To consider isolated bits of evidence, de minimis in nature, and to mix this with speculation is insufficient to support the conclusion that there are additional factual issues not passed upon by the district court. The findings made were adequate to dispel the conclusion required by Milliken that there be significant segregative effects in the SSDs caused by actions of a constitutional violator in KCMSD. The housing discussion is not complete without another word of caution. Plaintiffs expert, Dr. Gary Orfield, testified that the Kansas City metropolitan area, located in both Missouri and Kansas, comprised a single housing market. Dr. Kane testified that 7,239 black students now in Missouri schools would be in Kansas schools had there not been housing discrimination. Record at 7669-70, Pi’s Exhibit 1265-R. A number of witnesses called by plaintiff testified their children were moved from the KCMSD to Johnson County, Kansas. Record at 6722-23, 6809, 6863-64, 6953. The district court dismissed the Kansas districts from this action in its order of October 6, 1978. School District of Kansas City v. Missouri, 460 F.Supp. at 431. Accordingly, it made no findings on these issues. This evidence, however, points to the serious complications present in a claim based on housing violations where the housing market extends into another state and there is movement from an impacted district into the other state. VI. The Jenkins class relies particularly on decisions from Wilmington, Louisville, Indianapolis, and Allegheny County, in which interdistrict remedies had been ordered. These cases involved gerrymandering; the present case does not. The district court specifically found that the establishment and maintenance of school district boundaries was a local matter in Missouri, determined through local initiative, and uninfluenced by racial animus. The district court thus held the gerrymandering cases inappo-site. We agree. Evans v. Buchanan, 393 F.Supp. 428, considered the effect of Delaware legislation explicitly prohibiting the state board from altering the boundaries of the Wilmington school district, fixed as coterminous with the Wilmington city limits. The district court found no racially discriminatory purpose in the Act’s freezing of the Wilmington district’s boundaries. Id. at 439. It held, however, that the Act helped to maintain the racial identifiability of the Wilmington and the suburban New Castle County school districts and thus contributed to the segregation of the races. Id. at 445-46. The district court also found that the Wil