DARST-WEBBE TENANT ASSOCIATION BOARD, a Missouri non-profit corporation; Housing Comes First, a Missouri non-profit corporation; Peabody Tenant Association, a Missouri non-profit corporation, Plaintiffs-Appellants, v. ST. LOUIS HOUSING AUTHORITY, a Municipal corporation; Cheryl A. Lovell, in her official capacity as Director of the St. Louis Housing Authority; United States Department of Housing and Urban Development; Mel Martinez, in his official capacity as Secretary of United States Department of Housing and Urban Development, Defendants-Appellees, 417 F.3d 898

MELLOY, Circuit Judge. On remand following partial

summary-judgment, a bench trial, and a prior appeal, the distinct court found in favor of the defendants on six claims that relate to alleged discrimination in housing. We affirm. I. Background Plaintiffs-Appellants Darsb-Webbe Tenant Association Board, Peabody Tenant Association, and Housing Comes First, Inc., brought this action against Defen-danb-Appellees the United States Department of Housing and Urban Development (“HUD”) and the St. Louis Housing Authority (the “Housing Authority”).

The plaintiffs’ grievances arose from the defendants’ funding, approval, and implementation of revitalization plans for the Darsb-Webbe and Clinton Peabody public housing complexes in the Near South Side of St. Louis, Missouri. The revitalization plans employed federal HOPE VI program funds and called for the demolition of older housing projects that contained a high density of low-income, public housing rental units. The plans also called for the construction of mixed-income housing.

The proposed mix of housing for the revitalization area included: low-income, public housing rental units in which rent was a percentage of the occupants’ income; low-income tax credit units in which rent was a percentage of the area median income; market rate rental units; and units built for sale at market and below-market rates. The revitalization plans called for a reduction in the total number of low-income, public housing rental units and also a reduction in the number of large, multi-bedroom units capable of comfortably housing large families. To aid in the redevelopment, the City of St.

Louis applied for and received

Section 108 loan guarantees for improvements to the streets and other infrastructure around the reconstructed area. The infrastructure improvements were designed to eliminate the “superblock” model of the old projects, allow improved access for law enforcement and emergency services, reconnect the neighborhood to the street grid and the surrounding historic districts, and restore normal pedestrian and vehicle flows. We set forth the history of this matter in Darst-Webbe Tenant Ass’n Bd. v. St. Louis Housing Auth., 339 F.3d 702 (8th Cir.2003) (“Darst-Webbe I”).

There we explained that the Darst-Webbe buildings, built in the 1950s, originally contained 758 low-income, public housing rental units for families and 242 units designated for use by elderly residents. By 1995, when the Housing Authority formulated its initial revitalization plan, many of the units were dilapidated and only 220 of the family public housing rental units remained occupied. Those units were occupied almost entirely by African-American residents, resulting in a segregated development. Households with children and female-headed households were disproportionately represented among the residents.

These disproportions also existed to a substantial extent among households on the waiting list for public housing and households eligible for public housing. The Housing Authority did not proceed with demolition and reconstruction under the initial, 1995 plan. As a result, HUD placed the Housing Authority in default under the terms of the HOPE VI program. HUD then assisted the Housing Authority in the development of a new revitalization plan. The product of this collaboration was a modified, 1998 revitalization plan.

Under the 1998 plan, which is currently being implemented, the existing 758 total units/220 occupied units were to be replaced with a mix of housing that included eighty low-income, public housing rental units. Of the eighty replacement low-income, public housing rental units, twelve contained four bedrooms and none contained more than four bedrooms.

The plaintiffs brought the present action against the Housing Authority requesting that an additional 120 low-income, public housing rental units be built on-or off-site and that the unit mix include a greater number of units with four or more bedrooms to accommodate large families. The plaintiffs also brought claims against HUD, alleging various infirmities with HUD’s process of approving, and HUD’s decision to approve and fund, the Housing Authority’s plan. The district court granted

summary judgment in favor of the defendants on one count and held a bench trial to address the other counts. Following the bench trial, the district court returned a mixed verdict, ruling in favor of the plaintiffs on counts that involved the separate, Clinton Peabody housing complex, but ruling in favor of the defendants on counts related to the Darst-Webbe complex. The defendants did not appeal the adverse rulings regarding the Clinton Peabody housing complex. In the prior appeal, Darst-Webbe I, the plaintiffs appealed the

summary judgment ruling and eight of the counts related to Darst-Webbe. We affirmed the district court’s grant of

summary judgment in favor of the defendants. We also affirmed the district court’s trial judgment in favor of the defendants on two of the eight appealed counts. On the remaining six counts, we remanded for further findings of fact and law. The first three remanded counts were original Counts I — III against the Housing Authority. In these counts, the plaintiffs alleged disparate impact discrimination under the Fair Housing Act, 42 U.S.C. § 3604(a) & (b) (“FHA”) based on race, sex, and familial status, respectively.

No parties sought to augment the record on remand, and the district court considered no new evidence. The district court applied the burden shifting analysis for FHA disparate impact claims, found that the plaintiffs had failed to demonstrate a disparate impact upon a protected class, and found in the alternative that even if the plaintiffs had demonstrated a disparate impact, the Housing Authority had met its burden to prove that any disparate impact was justifiable as necessary to achieve legitimate policy objectives. See, e.g., Oti Kaga, Inc. v. South Dakota Housing Dev.

Auth., 342 F.3d 871, 883 (8th Cir.2003) (applying a burden-shifting analysis to an FHA disparate impact claim and finding a public housing fund allocation decision “justifiable on the ground it is necessary to [the defendant’s] exercise of its funding responsibilities.”). The district court found in the further alternative that the plaintiffs had failed to offer an alternative policy that could meet the many and varied policy goals set out for the HOPE VI program and revitalization of the Near South Side without discriminatory effects.

See, e.g., id. at 883 (noting that a plaintiff “may nonetheless prevail by showing another policy would accomplish [the proffered policy] objectives without the discriminatory effects”); Huntington Branch, N.A.A.C.P. v. Town of Huntington, 844 F.2d 926, 936 (2d Cir.1988) (stating that if defendants set out a legitimate, non-discriminatory justification, the burden shifts back , to plaintiffs to show alternative means to achieve the legitimate goals with a less discriminatory effect). The second three remanded counts were original Counts XIII, XVII, and XVIII against HUD.

In these counts, the plaintiffs alleged that HUD did not adequately consider the impact of the Housing Authority’s revitalization plan on protected classes. The plaintiffs also alleged that the revitalization plan discriminated against African-Americans, women, and children and consigned displaced Darst-Webbe family households to live in racially segregated public housing developments.

As a result, the plaintiffs argued, HUD violated the Administrative Procedures Act, 5 U.S.C. § 701 et seq. (“APA”) and the Fair Housing Act, 42 U.S.C. § 3608(e), by failing to consider impact on protected classes and failing to satisfy HUD’s general statutory mandate to affirmatively further fair housing. Count XIII focused specifically on HUD’s approval of the

Section 108 loan guarantee to the City of St. Louis as an objectionable action that failed to affirmatively further fair housing. In Count XVII, the plaintiffs alleged that HUD violated the APA by approving the

Section 108 loan guarantee. Finally, in Count XVIII, the plaintiffs alleged that HUD violated the APA by approving the DarsN Webbe HOPE VI plan. As to these latter three Counts, the district court found against the plaintiffs. The district court discussed the history of the specific HOPE VI plan for the Near South Side and the relationship between the HOPE VT plan and the

Section 108 loan program to help the city improve infrastructure around the project. The district court also discussed the HOPE VI program in general, identified many of the numerous goals that HUD is required by statute to satisfy and noted that the plan in this case was supported by detailed studies regarding the unique needs on the Near South Side. Ultimately, the district court found that HUD considered the impact of the proposed actions on protected classes and acted in an effort to affirmatively further fair housing.

Accordingly, the district court found that HUD did not abuse, its broad discretion by acting arbitrarily or capriciously when it decided to approve and fund the projects. II. FHA Claims Against the Housing Authority To establish a prima facie FHA claim under Oti Kaga, the plaintiffs must demonstrate that the objected-to action results in, or can be predicted to result in, a disparate impact upon protected classes compared to a relevant population. We assume for the sake of our analysis that the plaintiffs’ statistical proof established a prima facie case.

Accordingly, we turn our attention to the subsequent steps of the burden shifting analysis. Under the second step, the Housing Authority must demonstrate that the proposed action has a “manifest relationship” to the legitimate, non-discriminatory policy objectives and “is justifiable on the ground it is necessary to” the attainment of these objectives. Oti Kaga, 342 F.3d at 883. If the Housing Authority makes this showing, the burden shifts back to the plaintiffs to show that a viable alternative means is available to achieve those legitimate policy objectives without discriminatory effects. Id.

The district court’s findings under the second and third steps of the disparate impact burden-shifting analysis are factual findings. See Chambers v. Omaha Girls Club, Inc., 834 F.2d 697, 702 (8th Cir.1987) (stating in the context of a disparate impact employment discrimination claim that determinations under the burden shifting analysis “are reviewed under the clearly erroneous standard of review applied to factual findings”). Following a bench trial, “we review the trial court’s findings of fact for clear error.” Cooper Tire & Rubber Co. v. St. Paul Fire & Marine Insur.

Co., 48 F.3d 365, 369 (8th Cir.1995); see also Fed. R. Civ. Pro. 52(a). The Housing Authority identified numerous legitimate policy objectives to support its revitalization plan, some of which are expressly set forth in federal legislation. These include the overarching statutory goals of reducing the concentration of low-income housing and developing sustainable, mixed income communities.

See, e.g., 42 U.S.C. § 1437p(d) (implementing the goal of reducing the concentration of low-income public housing by providing that replacement housing is permitted on-site following the demolition of obsolete housing “if the number of replacement public housing units is significantly fewer than the number of units demolished”) (emphasis added); Id. at § 1437v(a)(3) (calling for “housing that will avoid or decrease the concentration of very low-income families”).

The Housing Authority also identified that the concentration of low-income housing was not merely an abstract issue, but at Darst-Webbe, an acute problem in need of a remedy.

Above and beyond the general statutory mandate of de-concentration, the Housing Authority set forth more specific objectives as follows: (1) providing for the marketability of a new mixed income community to families from a range of incomes by providing a balanced mix of housing types; (2) providing a strong home ownership component to ensure the presence of residents with vested, ownership-related incentives to improve and maintain the neighborhood; (3) providing for partnership with the City of St.

Louis to provide social services to existing and new residents; (4) providing for partnership with the Missouri Department of Family Services to provide on-site services to eligible residents of the community; and (5) providing for partnership with the St. Louis Agency on Training and Employment to provide assessment and services to eligible area residents. Finally, as already noted, the Housing Authority developed its revitalization plan to be implemented in concert with infrastructure investments by the City of St. Louis.

In general, the overall plan was an effort, consistent with the statutory goal of deconcentration, to shift the model of public housing in the Near South Side from that of densely clustered low income housing to a model of economic integration, with social services available to residents, with attendant public services restored, and without isolation of the development from the rest of the city. These general and specific objectives are legitimate and facially neutral. The plaintiffs’ contentions that these objectives are illegitimate are without merit.

The plaintiffs focus their primary efforts on proving that the proffered objectives are pretextual and/or insufficient to justify the action taken by the Housing Authority. We, however, find no evidence of pretext. Also, the district court did not err when it concluded as a factual determination that the Housing Authority’s comprehensive plan for redevelopment of the Darst-Webbe site and the surrounding area bears the requisite manifest relationship to the legitimate policy objectives. Oti Kaga, 342 F.3d at 883.

To achieve the objectives of creating a sustainable, mixed-income community, it was necessary to employ a marketable housing mix so that the post-redevelopment population would include actively employed residents to serve as role models and homeowners with a vested interest in the upkeep of the neighborhood. Determination of an appropriate housing mix necessarily relied upon imprecise predictive judgments regarding the likely marketability of various housing mixes.

Accordingly, there was no one piece of evidence that stated with certainty how many low-income rental units should have been included or what the optimal housing mix may have been. The plaintiffs make much of this fact and urge us to find that the lack of conclusive evidence prevents any successful showing that the specific housing mix was “necessary” to achieve the policy objectives. As a result, the plaintiffs argue, the plan may be amended to accommodate their request for more replacement, low-income public housing rental units.

We, however, do not sit as a secondary legislative body to amend or rework funding and planning decisions based on our own, alternative predictive judgments about the likely success of alternative proposed actions. As a result, our review under the burden shifting analysis is not an exercise in balancing the need for low-income housing units against the need for less-heavily subsidized units or market rate units. In other words, our review is not to determine if one, ten, or one hundred additional, replacement, low-income rental units should have been included in the amended plan.

Rather, our review is to ensure that “ a demonstrated disparate impact in housing be justified by a legitimate and substantial goal of the measure in question.” Langlois v. Abington Housing Auth., 207 F.3d 43, 51 (1st Cir.2000). The plaintiffs ask for more, seeking to have the court hold that the Housing Authority’s policy decisions and predictive judgments regarding marketability can be incrementally improved. “But to have federal judges make such policy choices is essentially to impose on them the job of making decisions that are properly made by Congress or its executive branch delegates.” Id.

Here, the Housing Authority relied upon an extensive housing market study performed by a consulting firm, Zimmerman Volk and Associates, to predict a marketable housing mix. The Housing Authority did not rely exclusively on this study, as acknowledged by the plaintiffs, but rather, considered other sources and incorporated a greater number of tax-credit units than recommended under the study.

The Housing Authority determined that, consistent with the general statutory mandate to achieve a deconcentration of low-income public housing, eighty low-income rental units mixed with a balance of market and below market for-sale and rental units would be marketable to residents from a variety of income levels. Marketability of the unsubsidized units and tax credit units was a critical component of the planned neighborhood.

The plaintiffs argue that the project could have a strong home ownership component, enjoy the desired partnerships with city and state social service providers, and remain marketable even if the number of replacement, low-income rental units were increased to 200. The plaintiffs also argue that the requested, additional 120 low-income public housing units could be built off-site with HOPE VI funds by diverting funding from certain aspects of the current Darst-Webbe revitalization plan but leaving the actual, on-site, proposed housing mix unchanged.

The plaintiffs point to expert testimony in which various professionals opined that marketability and other goals could be met even if HOPE VI funds were used to provide more low-income public housing on or off-site. These experts, in turn, relied on representative housing mixes in another St. Louis area redevelopment site as well as redevelopment sites in other communities. One expert the plaintiffs relied upon was Dr.

Paul Fischer, a political science professor, who in turn relied on a literature review, a March 2000 appraisal report by a company named Design Solutions, and the original plan submitted by the Housing Authority to HUD in 1995. Plaintiffs claim that these items of evidence demonstrate that a housing mix with 120 additional low income units would meet the Housing Authority’s stated objectives as effectively as the Housing Authority’s plan while reducing the discriminatory effect. Dr. Fischer does opine that such a housing mix would be marketable.

His opinion, however, relies in part on the appraisal report, and the appraisal report does not support his view. To the contrary, the appraisal assesses the marketability of the housing mix contained in Phase I of the redevelopment (a phase that included 36 of the 80 replacement public housing units) against the backdrop of the overall plan to build eighty replacement units. Dr. Fischer draws from the appraisal that the housing mix contained wholly within Phase I would be marketable in the Near South Side if applied to the revitalization area as a whole.

This conclusion, however, relied on the Phase I mix when taken out of context and viewed in isolation as a model housing mix. Taken out of context, as done by Dr. Fischer, this evidence falls far short of demonstrating that the Housing Authority’s plan is unjustified in light of the legitimate policy objectives. Also, we believe the district court properly declined to rely upon the original, 1995 redevelopment proposal as evidence that the Housing Authority could have included a greater number of low-income rental units and maintained marketability.

Redevelopment stalled under that proposal, and HUD declared the Housing Authority to be in default. With HUD’s assistance, the Housing Authority revised the plan and arrived at the housing mix that the plaintiffs now challenge. While the 1995 redevelopment proposal demonstrates that, at an earlier date, the Housing Authority believed that a different housing mix would be marketable, we do not find it appropriate to declare the district court’s findings infirm on the basis of an abandoned plan that was never implemented.

Also, we do not believe that the plaintiffs arguments related to the viability of housing mixes at other revitalization locations demonstrate error in the Housing Authority’s plan. The plaintiffs argued that the Murphy Park neighborhood in St. Louis and revitalization housing developments in other cities contained a substantially higher percentage of low-income public housing rental units and achieved marketability. As noted by the district court, these comparison neighborhoods each carried unique circumstances given their settings, which differed considerably from the Near South Side.

A different planning body may have elected to give these comparison sites more weight when developing a housing mix for the Near South Side. However, the role of the court is not to balance interests or second guess the minutia of a comprehensive redevelopment plan. Simply put, we do not believe that these other sites’ alternative housing mixes demonstrate that the Housing Authority in this case erred when it reached different conclusions about a marketable housing mixes.

Overall, our review reveals that the plaintiffs’ evidence fails to show that the district court committed error when it found that any discriminatory impact of the Housing Authority’s plan was justified by the need to achieve the stated objectives Importantly, we note that even if the plaintiffs had provided sufficiently strong and clear evidence to prove error in the district court’s determination as to necessity (i.e., marketability), it is not sufficient for the plaintiffs merely to prove the viability of an alternative housing plan or housing mix.

Rather, the plaintiffs must offer a viable alternative that satisfies the Housing Authority’s legitimate policy objectives while reducing the revitalization plan’s discriminator impact. This the plaintiffs did not do. No party offered evidence to show that a plan which diverts funds for 120 additional low-income housing units from other functions would actually result in a different post-development concentration of protected class members. During trial, defense counsel questioned Dr. Fischer and the following exchange took place: Q.

You don’t have a prediction in your report as to the number of Clinton-Peabody and Darst-Webbe residents that would be displaced from the Near South Side revitalization area as a result of the [Hope VI plan]? A. That’s correct. Q.

And, Professor Fischer, you did not make a prediction in your report as to the likely demographic characteristics as to the families that will occupy any newly constructed units developed as part of the [Hope VI plan], for instance the market rate rental units, the low-income housing tax credit units, the affordable home ownership and the market rate home ownership units, is that correct? A. Explain what you mean by population mix — If you are talking about racially or talking about in terms of income? Q. Any demographics [sic] characteristics, aside from income, any demographic characteristics? A. Correct.

Without evidence to demonstrate a reliable prediction as to the composition of the post-redevelopment population, the plaintiffs cannot prove that their proposed plan achieves a lesser impact than the plan as implemented. Further, under the clear error standard, it is not sufficient for the plaintiffs merely to offer some evidence regarding the alternative policy’s ability to satisfy the Housing Authority’s legitimate policy objectives and reduce disparate impact.

Rather, we must affirm unless the plaintiffs’ evidence in this regard is so compelling that we are “left with the definite and firm conviction that a mistake has been committed.” Anderson v. City of Bessemer City, 470 U.S. 564, 573, 105 S.Ct. 1504, 84 L.Ed.2d 518 (1985) (quoting United States v. United States Gypsum Co., 333 U.S. 364, 395, 68 S.Ct. 525, 92 L.Ed. 746 (1948)). By this measure, the plaintiffs fail.

The plaintiffs have not presented evidence to demonstrate by this standard that their proposed housing mix would allow the Housing Authority to achieve its numerous goals with a less discriminatory impact. III. Claims Against HUD Regarding the claims against HUD, Count XIII alleges that HUD failed to affirmatively further fair housing in violation of the general mandate in 42 U.S.C. § 3608(e). Count XVII and XVIII are claims under the APA that HUD failed to adequately consider impact on protected class members and affirmatively further fair housing when it approved the City’s

Section 108 loan guarantee and approved the Housing Authority’s Hope VI plan. We review HUD’s actions for compliance with the APA and a general statutory mandate under a highly deferential standard, reversing only if we find an abuse of HUD’s broad discretion. Norton v. Southern Utah Wilderness Alliance, 542 U.S. 55, -, 124 S.Ct. 2373, 2381, 159 L.Ed.2d 137 (2004); NAACP v. Secretary of Hous. & Urban Dev., 817 F.2d 149, 157 (1st. Cir.1987). Our review is deferential because it is not the role of the courts to micromanage agency actions for compliance with broad, general statutory mandates.

In Norton, the Supreme Court clearly articulated the problems attendant to such an invasive form of judicial review over agency action. There the court stated: The principal purpose of the APA limitations we have discussed — and of the traditional limitations upon mandamus from which they were derived — is to protect agencies from undue judicial interference with their lawful discretion, and to avoid judicial entanglement in abstract policy disagreements which courts lack both expertise and information to resolve.

If courts were empowered to enter general orders compelling compliance with broad statutory mandates, they would necessarily be empowered, as well, to determine whether compliance was achieved — which would mean that it would ultimately become the task of the supervising court, rather than the agency, to work out compliance with the broad statutory mandate, injecting the judge into day-to-day agency management. The prospect of pervasive oversight by federal courts over the manner and pace of agency compliance with such congressional directives is not contemplated by the APA. Id. at 2381.

This case, like Norton, involves a broad, general statutory mandate, HUD’s “duty to affirmatively further fair housing.” See 42 U.S.C. § 3608(e)(5) (providing HUD’s duty to “administer activities relating to housing and urban development in a manner affirmatively to further the policies of this subchapter”). Darst-Webbe I, 339 F.3d at 713. Norton makes clear that our review of HUD’s action is not a review to determine whether HUD has, in fact achieved tangible results in the form of furthering opportunities for fair housing.

Rather, our review is to assess whether HUD exercised its broad authority in a manner that demonstrates consideration of, and an effort to achieve, such results. Norton, 124 S.Ct. at 2380-81. We instructed the district court, on remand, to determine whether HUD had “ ‘assessed] negatively those aspects of [the] proposed course of action that would further limit the supply of genuinely open housing and ... assess[ed] positively those aspects of [the] proposed course of action that would increase that supply.’ ” Darst-Webbe I, 339 F.3d at 713 (quoting NAACP, 817 F.2d at 156).

We also asked the district court to explain “its conclusion that HUD appropriately considered its duty to affirmatively further fair housing.” Id. On remand, the district court found that the lengthy history of the revitalization efforts had created a large volume of documentary evidence. In reviewing HUD’s action, the district court reviewed all of this evidence, rather than merely reviewing the final documents issued by HUD regarding approval of the loans and revitalization plans.

The district court noted the unique, drawn-out nature of the agency action below starting with the initial revitalization plan, the Housing Authority’s default under that plan, and the collaborative effort to revise the plan. In this evidence, the district court found a record of HUD’s consideration of the project’s impact on protected class members. Plaintiffs argue that the district court should have ignored this larger body of evidence and focused only on HUD’s failure to explicitly discuss impact upon protected class members in the final communication of approval to the Housing Authority and the City.

We believe the district court properly found that the evidence to be considered was the evidence spanning the entire time frame for the project and not merely HUD’s final communication of approval to the Housing Authority. Review for compliance with a general statutory mandate — a deferential review only for abuse of discretion— should not focus myopically only on an agency’s end product. The plaintiffs are correct to criticize HUD’s failure to discuss the issue of impact on protected class members.

The plaintiffs also correctly note that we must not invent justifications for the agency’s actions, but must view what the agency actually considered. However, in the end, the issue is not whether HUD adequately discussed the impact on protected class members in one document. The issue is whether HUD adequately considered the impact on protected class members. We are convinced HUD did so in a manner sufficient to satisfy the general mandate. The district court cited evidence that showed how HUD assessed negatively the impact of the Housing Authority’s proposed action.

For example, HUD considered an analysis of impediments report prepared by the University of Missouri-St. Louis Public Policy Research Centers with Community Development Block Grant funds and support from the St. Louis City and County housing agencies. As described by the district court, this report included information gathered from lenders, real estate professionals, owners and renters. It also included analyses of fair housing actions and census data.

The conclusions in this report were that better information about housing opportunities needed to be provided to minority tenants, realistic opportunities for home ownership needed to be provided, and ancillary services were needed to aid minority tenants in obtaining jobs to empower home ownership. As discussed above, the revitalized plan incorporated these ancillary services to provide truly open housing and opportunities to integrate the Darst-Webbe neighborhood and its residents into the city.

The plan as adopted was an explicit rejection of the former model of public housing: attempts at integration and opportunities for residents rather than publicly funded, high density, high rise apartments. In short, as found by the district court, the record demonstrates consideration of impact and pursuit of a course of action that, on its face, demonstrates responsiveness to the perceived impact. Also, HUD’s initial approval of the 1995 plan demonstrates consideration of alternative plans — in fact, consideration of elements of the plaintiffs’ desired plan.

HUD placed the Housing Authority in default when the Housing Authority failed to move forward with that plan. It cannot reasonably be said that the Housing Authority failed to consider alternatives when, in fact, HUD initially approved the plan offered by the plaintiffs. The plaintiffs rely heavily on the testimony of one HUD employee, Jo Anne Garrison, who did not recall reviewing specific data regarding impact on African-Americans.

The plaintiffs infer that this testimony demonstrates HUD personnel were unaware that Darst-Webbe was overwhelming segregated and that its residents were predominantly African-American. The position that such an inference can be made is simply incredible as it ignores the well-documented reality of the situation that existed at Darsfl-Webbe before the Housing Authority began relocating residents.

While we would hope that future actions of this type would involve a more explicit discussion from the agency regarding impact on protected classes, we agree with the district court in this case and find the agency satisfied the requirements of the FHA and the APA. The judgment of the district court is affirmed. . The Honorable Stephen N. Limbaugh, United States District Judge for the Eastern District of Missouri. . Like the first appeal in this litigation, the present appeal does not involve the Clinton-Peabody complex or its residents. .

By the time of the prior appeal, all residents of the former Darst-Webbe project had been relocated, mostly through the use of

Section 8 vouchers.