Using the measuring stick provided by Arlington Heights II, we believe that the plaintiffs in this case could have demonstrated that the Village's TIF-related activities produced a discriminatory effect on its Hispanic residents. Over 49% of the TIF district residents are Hispanic, while their Village-wide population is just 13.4%. The districts also contain the two largest Hispanic neighborhoods -- in fact the only two majority Hispanic areas -- in the Village. Moreover, almost 44% of the Village's entire Hispanic population resides in these two districts. Given these facts, the way in which the Village fashioned the TIF districts had a greater impact on Hispanics than on whites because they were included disproportionately -- at a rate almost four times their population in the Village as a whole. The statistical picture here is even more compelling than in Arlington Heights, both because it is more disproportionate (49% affected/13.4% population versus 40% affected/18% population), and more concrete: the Arlington Heights court had to assume that the demographics of the proposed housing project would mirror the percentage of blacks and whites eligible for subsidized housing; here we know the racial composition of the affected population because the TIFs are already in place.
And all indications were that, for Hispanics, the effect of being disproportionately included in the TIF districts was adverse. The Village began and ended (constrained by the stand-still agreement) an aggressive course of TIF-related acquisition and demolition in the majority Hispanic neighborhoods of Green Oaks and Michael Lane. Green Oaks was affected most dramatically, with 44 of its families displaced, and several more buildings slated to be demolished or condemned. Although only one-half of the displaced residents was Hispanic, 50% is still far out of proportion to a 13.4% Village-wide population. Further evidence of discriminatory effect is found in the fact that the Village began TIF activities in majority Hispanic areas, and never reached the vacant, deteriorating commercial sites or predominantly non-minority sectors in the districts. Simply because the buildings in Green Oaks may have been integrated does not change that fact that the Village embarked on a selective course of action focused on the largest Hispanic neighborhoods in Addison. Exacerbating the effect on Hispanics was the Village's failure to provide Spanish-language notices alerting Michael Lane or Green Oaks residents to the consequences of living in a TIF district, although the Village knew these were majority Hispanic areas and had sent bilingual announcements there in the past. Considered together, these facts could satisfy the first type of discriminatory effect, a greater adverse impact on one race than another.
The Village attempted to dodge this evidence by claiming that plaintiffs had used the wrong "reference population" to show discriminatory effect. According to the Village, the plaintiffs should not have compared the percentage of Hispanics in the TIF districts to those in the general Village population, but rather should have pit the percentage of Hispanics displaced by TIF redevelopment activities against those living in areas needing redevelopment. Conceding that Arlington Heights II employs the first method, derived from Title VII case law, the Village insisted that the test for discriminatory effect in Title VII cases has evolved since then. In Wards Cove Packing Co. v. Atonio, 490 U.S. 642, 104 L. Ed. 2d 733, 109 S. Ct. 2115 (1989), a discriminatory failure-to-hire-and-promote case, the Supreme Court defined the proper statistical comparison in disparate impact cases as between "the racial composition of the qualified persons in the labor market and the persons holding the at-issue jobs." Id. at 651. The Court reasoned that if the shortage of minorities in certain positions was the product not of discrimination, but due instead to the "dearth of qualified nonwhite applicants," the employer should not be held liable for employment discrimination. Id. at 651-52.
By analogy, the Village asserted the only "qualified persons" in this case are those residing in areas needing aggressive redevelopment, more specifically, areas in the Village "with high crime, deteriorating housing, and poor planning." Def. Br. at 21. According to the Village, the only regions in this category are the TIF districts themselves, because "they are in the worst condition in the Village" and plaintiffs did not prove that other areas were similarly situated. Id. Because both the TIFs and the persons displaced from them are about 50% Hispanic, and equal numbers of Hispanics and whites were affected, the Village strongly maintained the plaintiffs had not established any discriminatory effect.
However, the Court believes the Village's analogy to Wards Cove is flawed in several respects. First, the Village's interpretation of the "qualified persons" test is inconsistent with reference populations used in fair housing cases, which unlike Wards Cove, are directly on point. Second, Wards Cove's precedential value has been considerably weakened by Congress' statutory overruling in the Civil Rights Act of 1991. Third, the Village's formulation is illogical because it assumes the validity of its own justifications for adopting the TIFs, fusing the first and third Arlington Heights II factors in a manner entirely contrary to the court's analysis.
As part of their discriminatory effect calculations, Fair Housing Act cases have often looked at area-wide population statistics and compared them to statistics for the group adversely affected. In Huntington Branch NAACP v. Town of Huntington, a case hauntingly similar to Arlington Heights II, zoning laws limited construction of multi-family housing to the town's "urban renewal zone," which was 52% minority populated. 844 F.2d at 930. When plaintiffs petitioned for a change in the zoning laws to build an integrated, federally subsidized apartment complex in a nearly all-white neighborhood, and the town refused, they sued under Title VIII. Id. at 928. In finding that the town's "failure to rezone . . . had a substantial adverse impact on minorities," the court observed that "minorities constitute a far greater percentage of those currently occupying subsidized rental projects compared to their percentage in the Town's population." Id. at 938. A similarly disproportionate percentage of blacks comprised the families holding subsidized housing certificates and on the waiting list to get certificates. Id. The court also premised its discriminatory effect determination on the fact that only 7% of all the town's families were in need of subsidized housing, while 24% of the town's black families needed such housing. Id.
Similarly, in Smith v. Town of Clarkton, the court found discriminatory effects produced by the town's decisions to withdraw from a county-wide housing authority and terminate construction of a low-income housing project. While blacks comprised under 40% of the county's population, 56% of all poverty-level families in the county were black, and 69.2% of all black families in the county were eligible for low-income housing. 682 F.2d at 1061. This "undisputed statistical picture [left] no doubt" that terminating the housing project affected blacks adversely. Id. at 1065.
The principle to be gleaned from these decisions is that using town-wide demographics for comparison to the affected population is a perfectly legitimate method of establishing discriminatory effect. Neither of these courts limited the reference population based on some notion of "qualified persons." Instead, these courts used the statistics that were most relevant given the facts before them, and did not establish hard-and-fast rules about the appropriate reference population. Likewise, as discussed above, we find the Village-wide demographics very telling under the particular facts presented here. As the Supreme Court recognized, "Statistics . . . come in infinite variety . . . . Their usefulness depends on all of the surrounding facts and circumstances." Hazelwood Sch. Dist. v. United States, 433 U.S. 299, 312, 53 L. Ed. 2d 768, 97 S. Ct. 2736 (1977) (internal quotations and citations omitted).
The Village cited only one FHA case, Gomez v. Chody, 867 F.2d 395 (7th Cir. 1989), that purportedly rejects a city-wide reference population. A closer look at the case belies this interpretation. During the course of a renovation project, the private developer in Gomez demolished portions of an apartment building complex that was 95% Hispanic, but "in an advanced state of dangerous disrepair, unsanitary, and infested with insects and rodents." Id. at 397-98. The apartment complex had also been declared a public nuisance. Id. at 397. As part of his contract to rehabilitate the apartments, the developer agreed that at least 51% of the renovated apartments would be affordable to low-and moderate-income persons. Id. at 398. Because gutting and renovation required displacing all the apartment residents, the County adopted a plan to provide those displaced with relocation assistance. Id. The displaced Hispanic residents sued, alleging disparate impact under the FHA based on the fact that Hispanics constituted only 60% of the city's population, but comprised 95% of the building residents. Id. at 400. The court found no discriminatory effect, holding that the record conclusively established that, whatever the tenants' race or the demands of the rehabilitation project, the apartments "had to be closed because they were unfit for human habitation." Id. at 402. Consequently, "'everyone living in the apartments would have to be displaced, regardless of their national origin . . . .'" Id. (quoting the district court). The Gomez court understandably determined that statistics were irrelevant given the building conditions; it did not, however, extend that holding to every fair housing case, especially those in which the need for aggressive redevelopment may be in serious question.
In conclusion, we find that, although the requisites for establishing discriminatory effect under the FHA are not crystal clear in every case, a careful analysis of the law and the evidence here reveals that the facts presented could have shown discriminatory effect under the established law in this Circuit.
(b.) Evidence of Discriminatory Intent
The second Arlington Heights II factor directs courts to ascertain whether the plaintiff has offered "some evidence" of discriminatory intent. Arlington Heights II, 558 F.2d at 1292. This is the least important of the four factors, id., and the required showing under it is modest. The plaintiffs' evidence need only give "some indication -- which might be suggestive rather than conclusive -- of discriminatory intent." Phillips v. Hunter Trails Community Ass'n, 685 F.2d 184, 190 (7th Cir. 1982).
On appeal from Metropolitan Housing Devel. Corp. v. Village of Arlington Heights (Arlington Heights I), 517 F.2d 409 (7th Cir. 1975), the Supreme Court examined the issue of intent in relation to Equal Protection Clause violations. It recognized that an intent to discriminate must often be inferred from surrounding circumstances, and instructed courts on the kinds of circumstantial evidence that might be relevant:
Sometimes a clear pattern, unexplainable on grounds other than race, emerges from the effect of the state action even when the governing legislation appears neutral on its face. . . . But such cases are rare . . . impact alone is not determinative, and the Court must look to other evidence. . . .