Summary
explaining that Ninth Circuit precedent "does not require Plaintiffs to identify those people who were actually affected by the rezoning denial. ... [P]roof of a predictably discriminatory effect is sufficient"
Summary of this case from Nat'l Fair Hous. Alliance v. Travelers Indem. Co.Opinion
2:09–cv–00297 JWS
04-30-2017
AVENUE 6E INVESTMENTS, LLC, et al., Plaintiffs, v. CITY OF YUMA, ARIZONA, a municipal corporation, Defendant.
Christopher Brancart, Elizabeth Brancart, Brancart & Brancart, Pescadero, CA, John A. Weil, Halls General Contractor, Yuma, AZ, for Plaintiffs. Richard Gray Erickson, Vaughn A. Crawford, Benjamin Michael Mitsuda, Snell & Wilmer LLP, Phoenix, AZ, Andrew Martin Jacobs, Snell & Wilmer LLP, Tucson, AZ, Richard Wade Files, Yuma City Attorneys Office, Yuma, AZ, for Defendant.
Christopher Brancart, Elizabeth Brancart, Brancart & Brancart, Pescadero, CA, John A. Weil, Halls General Contractor, Yuma, AZ, for Plaintiffs.
Richard Gray Erickson, Vaughn A. Crawford, Benjamin Michael Mitsuda, Snell & Wilmer LLP, Phoenix, AZ, Andrew Martin Jacobs, Snell & Wilmer LLP, Tucson, AZ, Richard Wade Files, Yuma City Attorneys Office, Yuma, AZ, for Defendant.
ORDER AND OPINION
[Re: Motion at Docket 218]
JOHN W. SEDWICK, SENIOR JUDGE
I. MOTION PRESENTED
At docket 218, the City of Yuma, Arizona, ("the City") has renewed its motion for summary judgment as to plaintiffs' Fair Housing Act disparate impact claim. The City argues that plaintiffs Avenue 6E Investments, LLC and Saguaro Desert Land, Inc. (jointly "Plaintiffs" or "the Hall Company") failed to present appropriate statistics to establish a prima facie case of disparate impact and, alternatively, that it had a legitimate and nondiscriminatory basis for denying the Hall Company's rezoning request. The City originally filed this motion at docket 148, but the motion was denied as moot after the court dismissed Plaintiffs' disparate impact claim on other grounds. On appeal, the Ninth Circuit reversed the court's dismissal and directed it to consider the arguments raised regarding the sufficiency of Plaintiffs' statistical showing of disparate impact. The City's statement of facts and supporting materials are located at docket 149. Plaintiffs' response to the motion is at docket 227. Plaintiffs filed a response to the City's statement of facts at docket 228 and filed their own statement of facts at docket 229. Their supporting materials are at docket 165. The City's reply is at docket 232, and its response to Plaintiffs' statement of facts is at docket 231. Oral argument was requested but would not assist the court.
The members and stockholders of Avenue 6E Investments, LLC and Saguaro Desert Land, Inc. are brothers Brian L. Hall, Fred T. Hall, and Michael T. Hall.
II. BACKGROUND
This action arises from the City's denial of the Hall Company's rezoning application for a 42–acre parcel of undeveloped land in Yuma, Arizona ("the Property"). The Property is located in the southeast portion of Yuma; specifically, on the west side of Avenue 6E and about one-half mile south of 32nd Street. The south end of the Property abuts a low-density R–1–8 subdivision, Belleza Phase 1. The north end of the Property is bordered by a recreational vehicle village ("RV Village"). The City owns the parcel of land to the east of the Property, which is designated for use as a wastewater facility and municipal park. To the west of the Property is the Terra Bella development. At the time of the events in this case, Terra Bella was zoned R–1–6 (minimum 6,000–square-foot lots) and had two phases of development. The first phase abutted the Property on the Property's southwest corner and had been platted for lots larger than the minimum lot size of 6,000 square-feet, ranging from 8,000 to 20,000 square feet. Only a couple of lots within the first phase of development had been sold and developed during the relevant time period. The second phase abutted the Property to the west and northwest. It was vacant land that had not been platted.
Prior to 2006 the Property was part of a larger 80–acre parcel of land owned by KDC of Yuma, LLC ("KDC"). KDC applied to rezone the 80–acre parcel from agricultural to R–1–8 (minimum 8,000–square-foot lots). The City Council conditionally granted the rezoning application, after which KDC obtained approval of a preliminary plat on the entire parcel that set out single family lots that were at least 9,000 square feet. KDC developed the southern 38 acres, known as Belleza Phase 1, and then sold the remaining 42 acres, the Property, to the Hall Company. The Hall Company purchased the Property from KDC for $5.8 million, or around $135,000 an acre. At the time of the sale, the Property was still conditionally zoned R–1–8 and still had KDC's preliminary plat providing for 129 lots of least 8,000 square feet.
In 2008, the Hall Company determined that development of the Property with R–1–8 zoning was not financially viable because there was no demand for large-lot, higher-priced homes in Yuma due to existing inventory and the housing market decline. Consequently, the Hall Company designed a development consisting of smaller lots: approximately 198 lots, each about 6,000 square feet. The Hall Company intended to construct affordable and moderately priced homes using a housing product—the Sunrise model home—it had built in another one of its subdivisions, Ocotillo 5, located approximately 1.25 miles south of the Property. Unlike more expensive large-lot homes, it believed that these affordable or moderately priced homes were still in demand. The Hall Company deemed "affordable" to mean entry-level houses priced between $120,000 and $150,000 and moderately priced to mean mid-level houses priced between $150,000 and $175,000. Specifically, the proposed price range for the houses to be constructed on the Property was between $125,200 and $159,800. They were not seeking to develop low-income housing as defined by the Department of Housing and Urban Development.
In order to implement the new plan, the Hall Company submitted an application to the City to rezone the Property from R–1–8 to R–1–6. Zoning designations R–1–8 and R–1–6 are both considered low-density zoning in Yuma, only one density gradient apart. The City's planning staff recommended that the City Council approve the rezoning request, finding it consistent with the City's General Plan for the area, which designated the area for low-density residential development.
Neighborhood opposition to the rezoning was subsequently communicated to City Council members through letters and at public meetings. People voicing opposition primarily based their objection to the rezoning on their belief that higher-density development and lower-priced homes would increase crime and reduce property values. Based on the substance of these comments, their expectation of increased crime and lower property values was based on the "demographics" that they associated with the Hall Company's other developments; the Hall Company was known for developing low and moderately priced homes, and its representative estimates that at least half of the purchasers of its homes were Hispanic. An example of the comments made by opposing neighbors includes a letter which stated that the rezoning will cater to a "group of people" with incomes of less than $75,000 that "statistically account for 91% of rape, murder, assault, armed robbery, etc." Another neighboring homeowner sent a letter imploring the City Council to consider the homes and neighborhoods that the Hall Company had built in the past and to notice the "higher rate of unattended juveniles roaming the streets." The homeowner stated that "many of [the Hall Company's] homes were not single family dwellings like they were designated to be and instead turned into multifamily dwellings which in turn led to more unattended juveniles and crime." One neighbor questioned at a City Council public hearing why the City would try to "mix these groups," referring to the property owners who would live in the Hall Company's proposed development and those that live in the neighboring R–1–8 subdivisions. Another homeowner voiced concerns about the "ownership demographic[s]" and indicated that while the nearby RV Village to the north of the Property had smaller lots, it nonetheless had "very different ownership demographics" than the "prospective owners of the houses that would be built [on the Property]." In September of 2008, the City Council denied the rezoning application. It had been three years since the Council had denied a rezoning application, approving over 60 requests.
Doc. 229 at pp. 26–27 (Plaintiffs' Fact 80).
Id. at p. 29 (Plaintiffs' Fact 82).
Id.
Id. at p. 30 (Plaintiffs' Fact 83).
Id. at p. 31 (Plaintiffs' Fact 84).
Doc. 228 at p. 14 (Plaintiffs' Response to City Fact 15); Doc. 165–2 at pp. 60–62 (Plaintiffs' Exhibit 33).
The Hall Company filed a complaint against the City, alleging violations of their equal protection and substantive due process rights under 42 U.S.C. § 1983 ; claims of discriminatory intent and disparate impact under the federal Fair Housing Act, 42 U.S.C. § 33601 et seq. ("FHA"); and violations of Arizona constitutional and statutory law. The City moved to dismiss all claims under Rule 12(b)(6) of the Federal Rules of Civil Procedure. The court originally granted that motion as to all of the claims except for the disparate impact claim under the FHA.
The City later filed two motions for summary judgment on that claim. In the motion at docket 146, the City argued that Plaintiffs could not meet their prima facie burden of establishing disparate impact claim under the FHA because of the glut of housing opportunities in the southeast portion of Yuma that were similar to Plaintiffs' proposed development on the Property. The City presented an alternative basis for summary judgment at docket 148, arguing that Plaintiffs failed to present the appropriate statistics to meet the prima facie test for disparate impact and also that it had a legitimate and nondiscriminatory basis for denying the rezoning request. The court granted the motion at docket 146, finding that Plaintiffs could not meet their prima facie burden because of other housing opportunities in the area, relying on an Eleventh Circuit case, Hallmark Developers, Inc. v. Fulton County . It then denied the motion at docket 148 as moot and therefore did not address the City's challenge to Plaintiffs' statistical showing of disparate impact.
466 F.3d 1276 (11th Cir. 2006). The court in Hallmark held that a developer failed to establish disparate impact on a protected group as a result of the county's denial of the developer's application to rezone land for the purpose of building a development with affordable housing because there was an oversupply of homes in the developer's projected price range in the southern part of the county where the property at issue was located.
On appeal, the Ninth Circuit reversed the court's initial dismissal of Plaintiffs' discriminatory intent claim under the FHA. It held that Plaintiffs' claim—that the City Council's denial of their rezoning request constituted disparate treatment under the FHA and Equal Protection—was plausible given allegations that 1) the public comments consisted of language that a reasonable jury could interpret to be racially charged code words; 2) the planning commission had unanimously voted in favor of the rezoning request; and 3) the City Council had not denied a rezoning request in three years. The Ninth Circuit also reversed this court's finding of summary judgment in favor of the City as to Plaintiffs' disparate impact claim under the FHA. It rejected the Eleventh Circuit's holding in Hallmark that similar housing opportunities within the general area of the property denied for rezoning negated the possibility of any disparate impact. It reasoned that adopting such a holding "would prematurely cut short the carefully constructed mode of analysis that the Supreme Court just recently established" in Texas Department of Housing & Community Affairs v. Inclusive Communities Project, Inc. , which was issued after this court's order granting summary judgment. On remand, the City renewed its motion for summary judgment that it had originally filed at docket 148.
––– U.S. ––––, 135 S.Ct. 2507, 2516, 192 L.Ed.2d 514 (2015).
III. STANDARD OF REVIEW
Summary judgment is appropriate where "there is no genuine dispute as to any material fact and the movant is entitled to judgment as a matter of law." The materiality requirement ensures that "only disputes over facts that might affect the outcome of the suit under the governing law will properly preclude the entry of summary judgment." There can be no genuine issue as to any material fact if a party "fails to make a showing sufficient to establish the existence of an element essential to that party's case, and on which that party will bear the burden of proof at trial." Ultimately, "summary judgment will not lie if the ... evidence is such that a reasonable jury could return a verdict for the nonmoving party." In resolving a motion for summary judgment, a court must view the evidence in the light most favorable to the non-moving party. The reviewing court may not weigh evidence or assess the credibility of witnesses.
Anderson v. Liberty Lobby, Inc., 477 U.S. 242, 248, 106 S.Ct. 2505, 91 L.Ed.2d 202 (1986).
Celotex Corp. v. Catrett, 477 U.S. 317, 322, 106 S.Ct. 2548, 91 L.Ed.2d 265 (1986).
Anderson, 477 U.S. at 248, 106 S.Ct. 2505.
Lopez v. Smith, 203 F.3d 1122 (9th Cir. 2000).
Dominguez–Curry v. Nevada Transp. Dept., 424 F.3d 1027, 1036 (9th Cir. 2005).
The moving party has the burden of showing that there is no genuine dispute as to any material fact. The moving party need not present evidence; it need only point out the lack of any genuine dispute as to material fact. Once the moving party has met this burden, the non-moving party must set forth evidence of specific facts showing the existence of a genuine issue for trial. All evidence presented by the non-movant must be believed for purposes of summary judgment, and all justifiable inferences must be drawn in favor of the non-movant. However, the non-moving party may not rest upon mere allegations or denials, but must show that there is sufficient evidence supporting the claimed factual dispute to require a fact-finder to resolve the parties' differing versions of the truth at trial.
Celotex, 477 U.S. at 323, 106 S.Ct. 2548.
Id. at 323–25.
Anderson, 477 U.S. at 248–49, 106 S.Ct. 2505.
Id. at 255.
Id. at 248–49.
IV. DISCUSSION
Under the FHA it is unlawful to "make unavailable or deny" a "dwelling" to a person because of that person's race, color, religion, sex, familial status, or national origin. A dwelling includes "any vacant land which is offered for sale or lease for the construction or location thereon of any such building, structure, or portion thereof." What it means to "make unavailable or deny" a dwelling is not specifically defined in the statute, but municipal land use decisions that block or impede the provision of housing are included. A plaintiff can establish a FHA violation under a theory of disparate treatment or disparate impact. Disparate treatment is intentional discrimination; a governmental body cannot "zone land or refuse to zone land out of concern that minorities would enter a neighborhood." Disparate impact discrimination, on the other hand, includes actions taken by "governmental bodies that create a discriminatory effect upon a protected class or perpetuate housing segregation without any concomitant legitimate reason." Disparate impact " ‘permits plaintiffs to counteract unconscious prejudices and disguised animus that escape easy classification.’ " It also "targets ‘artificial, arbitrary, and unnecessary barriers' to minority housing and integration that can occur through unthinking, even if not malignant, policies of developers and governmental entities." In a recent case, Texas Department of Housing , the Supreme Court affirmed that disparate impact claims were cognizable under the FHA.
San Pedro Hotel Co., Inc. v. City of Los Angeles, 159 F.3d 470, 475 (9th Cir. 1998) ; Gallagher v. Magner, 619 F.3d 823, 831 (8th Cir. 2010).
Avenue 6E Investments, LLC v. City of Yuma, 818 F.3d 493, 502 (9th Cir. 2016).
Id. at 503.
Id. (quoting Tex. Dep't of Hous. & Cmty. Affairs v. Inclusive Cmties., ––– U.S. ––––, 135 S.Ct. 2507, 2522, 192 L.Ed.2d 514 (2015) ).
Id. (quoting Tex. Dep't of Hous., 135 S.Ct. at 2522 ).
As noted above, the claim at issue in this motion is the FHA claim based on a disparate impact theory. Such a claim is evaluated under the familiar burden-shifting framework, and therefore a plaintiff must first make a prima facie showing of disparate impact. The Ninth Circuit has identified the elements of an FHA prima facie case under a disparate impact theory: " ‘(1) the occurrence of certain outwardly neutral ... practices, and (2) a significantly adverse or disproportionate impact on persons of a particular [type] produced by the [defendant's] facially neutral acts or practices.’ " That is, a plaintiff must show that defendant's actions had a discriminatory effect. Actions have a discriminatory effect if they actually or predictably result in discrimination. A plaintiff must prove the discriminatory impact at issue; merely raising an inference of discriminatory impact is insufficient. Statistical analysis is often the method by which a plaintiff makes the required demonstration. The statistical measure used must include " ‘a comparison between two groups—those affected and those unaffected by the facially neutral policy.’ "
See Comm. Concerning Cmty. Improvement v. City of Modesto, 583 F.3d 690, 711 (9th Cir. 2009) ; see also 24 C.F.R. § 100.500(c) (setting forth burden-shifting framework for disparate-impact claims under the FHA).
Id. (quoting Palmer v. United States, 794 F.2d 534, 538 (9th Cir. 1986) ) (brackets in original).
Pfaff v. U.S. Dep't of Hous. and Urban Dev., 88 F.3d 739, 745 (9th Cir. 1996) (quoting Keith v. Volpe, 858 F.2d 467, 482 (9th Cir. 1988) ).
Id. ; see also 24 C.F.R. § 100.500(a).
Id. at 746.
Darensburg v. Metro. Transp. Comm'n, 636 F.3d 511, 519–20 (9th Cir. 2011) (quoting Tsombanidis v. W. Haven Fire Dep't, 352 F.3d 565, 575 (2d Cir. 2003) ).
If a plaintiff successfully meets its burden to show a prima facie case of disparate impact, the burden shifts to the defendant to prove that the challenged practice "is necessary to achieve one or more substantial, legitimate, nondiscriminatory interests." If the defendant satisfies that burden, the plaintiff may still prevail by showing that the defendant's interests could be served by a less discriminatory alternative.
24 C.F.R. § 100.500(c)(2) ; Darensburg, 636 F.3d at 519.
24 C.F.R. § 100.500(c)(3) ; Darensburg, 636 F.3d at 519.
Plaintiffs' Prima Facie Showing of Disparate Impact
In this case, the City's denial of Plaintiffs' rezoning application is the outwardly neutral practice at issue. In order to show that the denial had a disproportionate effect on Hispanics, Plaintiffs hired an expert who specializes in research, training, policy evaluation, and program development in the areas of housing and community development discrimination, Dr. Calvin Bradford. Dr. Bradford's analysis was designed to compare who could afford housing on the Property if the Hall Company's proposed R–1–6 development had been allowed, with who could afford the larger-lot housing that will be built as a result of the denial. His objective was to determine the percentages of protected and non-protected persons who were priced out of the market as a result of the housing price increase related to the rezoning denial.
Dr. Bradford's report is found at docket 165–7 (Plaintiffs' Exhibit 108).
In order to make this comparison, Dr. Bradford first had to establish how the rezoning denial would affect housing costs on the Property; that is, he had to determine the price range of homes that would have been built on the Property had the rezoning been granted and the price range of homes that will be built on the Property as a result of the denial. Based on the evidence provided, it is undisputed that typically density conversely affects housing prices. It is also undisputed that the Hall Company intended to have more than the 129 lots originally planned by the previous owner in order to make the development financially viable and that it intended to build its Sunrise style of home on the Property. It needed to rezone the Property in order to implement its plan. The Sunrise model homes were also located in the Hall Company's nearby development Ocotillo 5, and, therefore, for the cost of housing associated with Plaintiffs' proposed development, Dr. Bradford used the average sale price of homes sold in the Ocotillo 5 development during the relevant time period of 2007–2009 to calculate a normal expected price range for homes within Plaintiffs' proposed R–1–6 development. Dr. Bradford determined that price range to be $123,000 to $171,000 ("Plaintiff's Proposed Price Range"). For the cost of housing associated with the rezoning denial, Dr. Bradford used the actual sale price of homes between 2007–2009 in the neighboring Terra Bella I and Belleza I developments, which had lots greater than 8,000 square feet. From the sale data for these two developments he calculated a normal expected price range of homes built on larger lot developments to be from $258,000 to $386,000 ("Neighboring Properties Price Range").
Doc. 299 at pp. 18–19 (Plaintiffs SOF 52, 53).
Dr. Bradford later re-ran his analysis to include only those homes sales in Ocotillo 5 with lots that were 6,000 square feet. The change did not affect his analysis or opinion. Doc. 165–7 at pp. 97–98.
Doc. 165–7 at pp. 45–46.
Id. at pp. 46–51. Dr. Bradford later revised this price range to $259,000 to $385,000 to account for a sale that had originally been omitted from the analysis. He concluded that the change did not affect his analysis or opinion. Id. at p. 92.
In order to compare the percentages of Hispanics and whites who were qualified home buyers in the identified price ranges in the Yuma market during the relevant time period, Dr. Bradford used data from the Home Mortgage Disclosure Act (HMDA). HMDA "requires home mortgage lenders who make first or second lien home purchase or refinance loans to report specific data elements on an application by application basis each year." The data includes information about the ethnicity of the applicant, the race of the applicant, the amount of the loan, the type of housing unit, the purpose of the loan, the lien status, and the census tract location of the property. Dr. Bradford used this data because it is tied to who actually purchased owner-occupied homes in the Yuma market and can be linked to the race and ethnicity of the buyer.
Id. at p. 51.
Dr. Bradford conducted a variety of different statistical analyses based on the HMDA data. He ran the data using three different assumptions about the relationship of the loan amounts to the sale prices of the homes purchased in the Yuma market. "The data were tested for assumptions that the loan data represented 85% of the home prices, 90% of the home prices, or 95% of the home prices." He looked at the data for a period of three years, 2007–2009, in order to "create a robust set of data for the three-year period as a whole," but he also looked at the data for each individual year to account "for the unusual variations in the mortgage markets in [those three years]." He defined the Yuma market area in two different ways, "producing two variations in the geographic market area to ensure that the results of the tests were not simply the result of a particular definition of the City of Yuma home sales market."
Id. at p. 40.
Id.
Id.
Dr. Bradford concluded that his analysis "shows a consistent pattern of statistically significant disparate impacts on Hispanics for housing in the value and price ranges reasonably comparable to those proposed for the ... Property by the Plaintiffs in this case ... versus those built in the neighboring subdivisions of Belleza 1 and Terra Bella...." The pool of persons purchasing a home in the Hall Company's proposed price range was roughly 45% Hispanic and 50% white, while the pool of persons purchasing a home in the more expensive range was roughly 30% Hispanic and 65% white. Dr. Bradford concluded that there were "statistically significant decreases in the percentages of Hispanics in the Neighboring Properties Price Range to the Plaintiffs' Proposed Price Range." This was true regardless of the assumptions made about the loan amounts and homes prices, and regardless of the two ways to define the Yuma market area. He ran comparable tests on white borrowers and concluded that "whites showed a significant increase in their share of the market when comparing a change from the Plaintiffs' Proposed Price Range to the Neighboring Properties Price Range." He summarized his findings as follows:
Id. at p. 39.
Id. at p. 40.
Id. at pp. 40–41.
[F]or the City of Yuma Market Area over three years from 2007–2009, for loans representing a range of 85% of two prices ranges, the share of Hispanic borrowers decreased by 40% when comparing their market shares in the Neighboring Properties Price Range to their market share in the Plaintiffs' Proposed Price Range. On the other hand, when making the same comparison for whites, their market share increased by 39%. For the loans representing 90% of the two price ranges, the Hispanic market shares decreased by over 36% while the market share for whites increased by 30%. For the loans representing 95% of the two price ranges, the Hispanic market share decreased by more than 37% while the white market share increased by over 29%.
All of the tests were statistically significant at the 95% level or higher for every test for each individual year, and for all the Loan-to-Value ratios for both market areas. The tests for the three years as a whole were statistically significant well above the 99% level for all Loan-to-Value ratios and for both market areas.
Id. at p. 41.
The court concludes that Plaintiffs' evidence is sufficient to show a prima facie case of disparate impact. Indeed, there is no rigid mathematical formula to show disparate impact and the inquiry is necessarily fact-specific. HUD recognized as much when it promulgated its disparate impact rule: "Given the numerous and varied practices and wide variety of private and governmental entities covered by the Act, it would be impossible to specify in the rule the showing that would be required to demonstrate a discriminatory effect in each of these contexts." Here, Plaintiffs did not just show that the denial of a request to lower the density increases housing costs and that Hispanics are generally less wealthy than whites in Yuma. Rather, it provided statistical evidence—based on a pool of qualified home purchasers within the relevant market area and during the relevant time frame—regarding the racial makeup of those priced out of the market as a result of the price increase associated with the City's denial of Plaintiffs' rezoning application.
Watson v. Fort Worth Bank & Trust, 487 U.S. 977, 994–95, 108 S.Ct. 2777, 101 L.Ed.2d 827 (1988).
78 Fed. Reg. 11460, 11468 (2013).
The City raises a handful of objections to Dr. Bradford's conclusions that it believes makes his reasoning faulty and his conclusions speculative enough for the court to discount. First, the City argues that Plaintiffs have improperly identified the population affected by the City's action. It argues that the affected group is comprised of only those who actually would have purchased a home in the Hall Company's proposed development and not those people who purchased homes within the relevant price ranges throughout the Yuma market area as a whole. It argues that Dr. Bradford did not know where within Yuma any particular home purchaser actually wanted to buy a house, and therefore he could not quantify whether there were lost housing opportunities.In support the City relies on Darensburg v. Metro. Transportation Commission , a Ninth Circuit case in which the plaintiffs alleged that the defendant's transit expansion plan disparately impacted racial minorities because it favored rail expansion over bus expansions. The Ninth Circuit concluded that the plaintiffs had failed to establish a prima facie case of disparate impact because the statistical showing of disparate impact was too general:
636 F.3d 511 (9th Cir. 2011).
The district court erred by relying on a faulty syllogism: (1) a greater percentage of bus riders than rail riders are minorities; (2) fewer bus expansion projects than rail expansion projects were included in the RTEP, and bus projects received a lesser percentage of requested funding than did rail projects; (3) therefore, minorities were adversely affected. The fault lies in the reliance on the overall regional ridership statistics for existing bus and rail service. What is key is that these statistics say nothing about the particular ridership of the planned expansions. Simply because minorities represent a greater majority of bus riders as opposed to rail riders, the rejection of a particular new bus expansion project in favor of a new rail expansion project will not necessarily work to the detriment of minorities. It is a real possibility that a particular bus project supported by AC Transit—Transbay service, for example—will serve a largely white ridership. On the other hand, a rail expansion project—the BART project connecting the East Bay to San Jose, or the MUNI central subway project connecting Bayview and Chinatown, for example–may benefit minority riders more than white riders by serving areas with high concentrations of minorities, and integrating them more fully into the regional rail system. In fact, although AC Transit's ridership may have a higher percentage of minorities, BART annually carries over two million more minority riders than AC Transit. It is entirely plausible that an RTEP with a heavy emphasis on rail could significantly benefit Bay Area minorities. However, a court simply could not determine from Plaintiffs' statistical evidence whether the projects in the RTEP will benefit or harm the Bay Area's minority transit riders.
Id. at 520–21.
In Darensburg, the regional ridership statistics failed to show that any particular expansion project would either benefit or harm minority riders. Here, Plaintiffs did not simply rely on statistics that show increased home prices adversely affect Hispanics generally. They did not simply look at "all persons who purchased any kind of housing in Yuma at any price range during the relevant time period." Rather, they "narrowly defined the particular type of housing and price ranges at issue, based on the facts of the case, and showed how decisions affecting the availability of housing in those price ranges, using actual Yuma homebuyer data, had a significantly disproportionate impact on qualified Hispanic buyers." That is, they showed the racial makeup of those priced out of a housing opportunity due to the denial of Plaintiffs' rezoning request and consequently Plaintiffs' proposed housing development.
Doc. 227 at p.13.
Id.
Darensburg does not require Plaintiffs to identify those people who were actually affected by the rezoning denial. As noted by Plaintiff, proof of a predictably discriminatory effect is sufficient. Indeed, because the proposed R–1–6 development was not built, identifying with specificity those who would have purchased a home within Plaintiffs' proposed development on the Property is impossible. As a result, Plaintiff used a pool of qualified persons—those who did in fact purchase homes in Yuma in the predicted price range between 2007 and 2009—as a proxy, which is acceptable in the absence of data reflecting actual applicant flow.
24 C.F.R. § 100.500(c)(1) ; see also 78 Fed. Reg. 11460 at 11468 (rejecting the assertion that proof that a practice or decision which only predictably results in discrimination is too speculative to sustain a disparate impact claim and noting that discriminatory housing practices that block the development of certain housing necessarily rest on proof of predictably discriminatory effects).
See, e.g., Hazelwood Sch. Dist. v. United States, 433 U.S. 299, 309 n.13, 97 S.Ct. 2736, 53 L.Ed.2d 768 (1977) (noting that proof of actual applicants would be "very relevant" and "firmer proof" but acknowledging that in the absence of such data statistical evidence of the pool of qualified individuals was still probative); Williams v. Owens–Illinois, Inc., 665 F.2d 918, 927–28 (9th Cir. 1982) (recognizing that if actual applicant flow data does not exist, "the district court may accept any other reasonable proxy" that would indicate the source of the defendant's potential employees).
Defendants also rely on a Tenth Circuit case, Reinhart v. Lincoln County . In Reinhart , the plaintiffs alleged that new county development regulations that increased minimum lot sizes and imposed various development requirements would increase the cost of housing generally and increase the sale price of undeveloped lots and therefore would cause a disparate impact on protected groups under the FHA. The Tenth Circuit affirmed the lower court's dismissal of the complaint: "It is not enough for [the plaintiffs] to show that (1) a regulation would increase housing costs and (2) members of a protected group tend to be less wealthy than others." Rather, the plaintiffs should have "compare[d] who could afford the housing before the new regulations with who could afford it afterwards." It noted that the plaintiffs "told the district court nothing about the expected prices of homes in the development, or, more importantly, the amount by which those prices would exceed what they would have been absent the change in the regulations" and that their income data did not allow "anyone to calculate the specific impact on protected groups, as compared to the general population, from an increase in home prices from $X to $Y."
482 F.3d 1225, 1231 (10th Cir. 2007).
Id. at 1230.
Id.
Id. at 1231.
Dr. Bradford's analysis did not suffer from the same defects described in Reinhart , but rather it followed the Tenth Circuit's instruction as to what would constitute a sufficient statistical demonstration. The analysis showed that the City Council's decision will increase the cost of housing by a certain amount and then showed that the "increase disparately impacts the ability of members of the protected group to buy a dwelling" on the Property. His analysis showed that "to the extent that the higher price reduces the size of the purchaser market for the dwelling, the reduction is disproportionately high for the protected group."
Id. at 1230.
Id.
Second, the City argues that Dr. Bradford's comparison group is flawed because Dr. Bradford wrongly assumed there was a market for R–1–8 housing. That is, it argues that larger lot, R–1–8, housing was not going to be built on the Property given the glut of existing large lot homes on the market at that time and so the Property would merely be vacant as a result of the rezoning denial. Therefore, the City argues that the only actual effect of the denial of the rezoning request was to prevent the R–1–6 development from being built. On this point, the City argues that Dr. Bradford should have simply compared the proportion of Hispanics who would have lived in the Hall Company's proposed development to either (1) the proportion of whites who would have lived there or (2) the proportion of Hispanics in Yuma's total population. Based on Dr. Bradford's data, the purchasers of homes in Plaintiffs' proposed price range were roughly half Hispanic and half white. Hispanics also make up about 55% of Yuma's population. Therefore, the City argues that there is no disparate impact when "comparing the proportion of Hispanics in the affected group with the proportion of [w]hites in the affected group" nor when "comparing the percentage of Hispanics in the affected group with their proportion in the total population."
The court agrees with Plaintiffs that given the facts of this case, a comparison between the price range of housing the Hall Company intended to build and the price range of housing in the surrounding larger lot developments is sufficient to withstand summary judgment. The record shows that limiting development of the Property to lot sizes and prices similar to those in the neighboring subdivisions was the stated goal of the residents opposing the rezoning, and "[t]he reliance of those homeowners on keeping the status quo is cited by the City as a legitimate reason for its denial of [Plaintiffs'] request for rezoning." Moreover, the facts show that after denying Plaintiffs' request, the City granted the rezoning request by the developer of the neighboring Terra Bella property to rezone the Terra Bella property from R–1–6 to R–1–8 to ensure that the Terra Bella property remains larger-lot development. A reasonable jury could conclude, based on these facts, that by denying Plaintiffs' rezoning request, the City is reserving the Property for future development proposals similar to that in the neighboring larger-lot developments. Dr. Bradford's analysis shows that Hispanics are disproportionately priced out of those neighboring developments as compared to whites. On this record, construing the evidence in the light most favorable to Plaintiffs, it would be an error to conclude as a matter of law that Plaintiffs are unable to establish a prima facie case of disparate impact.
Doc. 227 at p. 15.
Id.
The City's third challenge to Plaintiffs' prima facie case is that Dr. Bradford's analysis is otherwise unreliable. It cites NAACP v. City of Kyle for the proposition that he failed to take into account supply, demand, and other affordable housing options. In Kyle , the NAACP sued the city alleging that changes to a zoning ordinance, which resulted in increased lot and home size requirements throughout its jurisdiction, disproportionately impacted minority home buyers. The Texas district court held that the NAACP's statistical showing was too general to establish a prima facie case. It found that the analysis did not consider applicant flow (those who purchased homes to occupy within the city during the relevant time period), the effect of the exemptions provided, the affordable housing alternatives provided under the ordinance, and the "supply and demand and other available affordable-housing alternatives." The City argues that, as in Kyle , Plaintiffs failed to consider supply and demand and other available affordable-housing alternatives and therefore failed to show there is an actual shortage of housing available to Hispanics. As noted by Plaintiffs, however, this court originally concluded that Plaintiffs could not show a prima facie case of disparate impact considering the adequate supply of similarly priced housing in southeast Yuma, but the Ninth Circuit rejected that conclusion. While the Ninth Circuit made clear that its ruling did not "mean that the existence of available housing in close proximity is irrelevant to determining whether a plaintiff proves a disparate impact," but it explained that a city must prove that "truly comparable housing is available in close proximity to a proposed development" and that the housing is comparable not only in price and type but also in relation to schools, infrastructure, convenience, amenities, and crime. The City did not make such a showing here. Moreover, Kyle is distinguishable. In Kyle the court rejected the plaintiff's analysis in part because it relied on general affordability pools that did not show whether any person in such a pool would have purchased a home in Kyle . Here, Plaintiffs used HMDA data that reflects actual home buyer behavior in the Yuma housing market at the relevant price ranges. As discussed above, the court finds this to be a reasonable proxy for actual applicant flow.
No. A-05-CA-979, 2009 WL 6574497 (W.D. Tex. Mar. 20, 2009).
Avenue 6E, 818 F.3d at 512.
The City argues that Dr. Bradford's analysis is also unreliable because the HMDA data is flawed in that it does not include cash or seller-financed homes. The City relies on tables in its expert's report containing data that show cash and seller-financed transactions, which are not included in HMDA data, made up 36% of homes sales in Yuma in 2009, 44% in 2010, and 58% in 2011, thus rendering the HMDA data incomplete and any analysis stemming therefrom speculative. Plaintiffs object to the City's data, arguing that the data contained in the tables was not gathered by the expert himself but rather was provided to the expert from the City's attorney. The expert did not verify the data for accuracy and admits that he cannot verify the accuracy of the tables. Plaintiffs argue that data provided by counsel in the midst of litigation are not reasonably relied upon by experts under Rule 702 or 703 of the Federal Rules of Evidence. The court need not rule on the admissibility of the data at this time because, regardless of its admissibility, Plaintiffs present evidence raising an issue of fact concerning the negligible impact that cash-sales have on Dr. Bradford's statistical analysis. They point to evidence showing that the majority of sales involve loans and that the majority of cash sales are in the Foothills, an area separate from the Property at issue here, with seasonal manufactured homes, rather than year-round, site-built homes. The City's statistical expert who relied on that data was not able to state what percentage of cash sales involved the purchase of recreational vehicles or manufactured homes for seasonal visitors. This evidence presents a dispute as to whether the presence of cash sales renders Dr. Bradford's analysis unreliable. Construing the evidence in the light most favorable to Plaintiffs, the court cannot conclude that as a matter of law Dr. Bradford's analysis is too speculative to support a prima facie case of disparate impact.
Doc. 165–1 at pp. 454 (Plaintiffs' Exhibit 15).
Plaintiffs cite Dukes v. Wal – Mart, Inc., 222 F.R.D. 189, 196, 198 (N.D. Cal. 2004) and Munoz v. Orr, 200 F.3d 291, 301 (5th Cir. 2000).
The City also says that Dr. Bradford should not have used HMDA data at all and that the better data source to compare the racial make-up of those who will live on the Property in its current R–1–8 zoning should have been the 2010 Census data for the Belleza and Terra Bella subdivisions. The City asserts that the 2010 Census data shows that the population of the Census blocks making up Belleza I and Terra Bella is 49% Hispanic and therefore any larger lot development on the Property would be similar. As noted by Plaintiffs, however, the 2010 Census data is based on persons living in the census block and not the number of households. Plaintiffs provide evidence to show that the percent of Hispanic households in Terra Bella and Belleza is 36.9%, whereas the percent of white households is 59.1%. This is consistent with the analysis conducted by Dr. Bradford using the HMDA data. Based on the evidence provided by Plaintiffs, the court cannot conclude that as a matter of law the 2010 Census data is the only appropriate data source.
The City also says that the price ranges chosen by Dr. Bradford are unsubstantiated. It asserts that the Hall Company could have provided the Sunrise product on the Property regardless of the lot size. It argues that while generally there is a relationship between density and cost, both R–1–6 and R–1–8 are considered low-density zones and R–1–6 does not guarantee lower home prices, relying particularly on the Terra Bella development, which was zoned for a minimum of 6,000 square foot lots, but was nonetheless a larger-lot development. The court is not persuaded by this argument. Plaintiffs do not need to prove that all property zoned for R–1–8 results in more expensive homes than a property zoned for R–1–6 or vice versa. Here, the undisputed evidence shows that the Hall Company was seeking to rezone the Property to build a specific type of housing in the range of $125,200 and $159,800 and that the denial made it unable to do so. As stated by Plaintiff, "[a]lthough a zoning decision lies at the heart of the case, it is the effect of zoning on the minimum price that can be built that is at issue."
Doc. 227 at p. 18.
Additional Evidence Supports Plaintiffs' Prima Facie Case
Evidence of discriminatory intent can bolster a disparate impact case, and there is evidence of discriminatory intent here. In its order remanding the case back to this court, the Ninth Circuit concluded the language alleged to have been used by the neighbors opposing Plaintiffs' rezoning request was sufficiently racially charged to raise the inference of racial animus and to put the decision-making body on notice. Plaintiffs have now put forth evidence to support its allegations that such comments were in fact made in letters and during council meetings. Therefore, there is evidence that the neighbors opposing the rezoning request did so because of discriminatory animus against Hispanics and that they communicated such animus to the City Council. The evidence shows that the City Council, at least in part, based its denial on the opposing neighbors' concerns. Such evidence bolsters Plaintiffs' disparate impact claim.
Casa Marie, Inc. v. Superior Court, 988 F.2d 252, 270 n.20 (1st Cir. 1993).
The City's Justification for the Rezoning Denial
As the Ninth Circuit stressed in its ruling remanding the case back to this court, the Supreme Court's decision in Texas Department of Housing makes clear that a showing of a disparate impact is not the end of the analysis. "[A] developer's ability to show disparate impact does not impose a duty on a municipality to approve all zoning applications in a particular price range." Rather, "such a showing merely requires the city to demonstrate that the action that creates an adverse effect on minorities is supported by adequate justification." If this justification is shown, then the plaintiff can still prevail by showing that there were less discriminatory alternatives available that would accommodate the city's legitimate concerns and the developer's goals. This "carefully constructed" burden-shifting analysis supports the FHA's objectives by " ‘stopping municipalities from enforcing arbitrary and, in practice, discriminatory ordinances barring the construction of certain types of housing units.’ "
Avenue 6E, 818 F.3d at 512.
Id.
Id. at 510 (quotingTex. Dep't of Hous., 135 S.Ct. at 2522 ).
The City asserts two legally sufficient reasons for denying the rezoning. The first stated reason for the denial was the neighbors' reliance on the Property's preexisting zoning and plat. It cites to evidence where one council member explained that he voted against the rezoning because the neighbors relied on the R–1–8 zoning and preliminary plat that already existed for that Property when they bought their homes in the neighboring developments. The City cites to Budnick v. Town of Carefree for the proposition that a city's interest in achieving its zoning goals is a legitimate governmental interest. While promotion of zoning goals is a legitimate governmental interest, appeasement of neighboring property owner's private expectations and assumptions about a property is not necessarily an equivalent objective. While the evidence demonstrates that neighboring owners expected the Property to be large-lot development and the preliminary plat indicated the plan for the Property was to sell at least 8,000 square foot lots, there is evidence to show that such definitive expectations were not entirely supported by the City's zoning plans and ordinances. It is undisputed that Plaintiffs' rezoning request was consistent with and conformed to the City's General Plan. One of the purposes of the General Plan is to "protect[ ] homeowners and business property values for the long term by identifying appropriate locations for those [uses] in adjacent properties." The General Plan sets forth "long-term land use parameters such that people can rely on those land uses in making investment decisions and purchasing property and know whether there's going to be business or housing right next to them." Therefore, in passing a General Plan that allowed either R–1–6 or R–1–8 zoning on the Property, because both were considered low-density housing, the City had already determined that these two densities were compatible land uses. Moreover, the R–1–8 zoning on the Property was conditional on the developer meeting certain conditions by a specified date, which did not occur; the plat in place on the Property was only preliminary and because a final one was never filed, the plat automatically expired in 2007. Also, the fact that the Zoning and Planning Commission unanimously recommended approval of Plaintiffs' request lends support to the conclusion that achievement of zoning goals was not the purpose here. Plaintiffs also point to evidence in the record to support a conclusion that the actual concern here was the different objectives of neighboring developers. All of the evidence is sufficient to raise an issue of fact for the jury as to whether the denial was indeed necessary to achieve the City's zoning goals and plans.
518 F.3d 1109 (9th Cir. 2008).
Doc. 165–1 at p. 183 (Plaintiffs' Exhibit 7 at pp. 41–42).
Doc. 165–1 at p. 183 (Plaintiffs' Exhibit 7 at pp. 42–43).
Doc. 149 at p. 2 (City Fact 6); Doc. 228 at pp. 13–14 (Plaintiffs' response to City Fact 15); Doc. 165–2 at pp. 64–65 (Plaintiffs' Exhibit 34); Doc. 165–1 at p. 354 (Plaintiffs' Exhibit 12).
Doc. 149 at p. 2 (City Fact 6); Doc. 228 at pp. 13–14 (Plaintiffs' response to City Fact 15); Doc. 165–2 at pp. 64–65 (Plaintiffs' Exhibit 34); Doc. 165–1 at p. 354 (Plaintiffs' Exhibit 12).
Doc. 228 at pp. 15–16 (Plaintiffs' response to City Fact 15)
The City argues that its denial based on private reliance on an approved preliminary plat is legally sufficient basis because the plat is part of the character of the neighborhood that the City sought to maintain for its residents. While various Ninth Circuit cases support the City's argument that concern for a neighborhood's character is a legitimate reason for a denial of a land use or building permit request, the circumstances surrounding the land use decision at issue in those cases are distinguishable from the circumstances here. n Budnick , the town denied a developer's request for a special use permit to build a multi-story, continuing care retirement community in a residential zone. In Gamble v. City of Escondido , the city denied a developer's request for a conditional use permit to build a 10,360 square foot, eight bedroom, twelve bathroom structure for disabled elderly living and day care after it concluded that the proposed structure was too large for the lot and did not conform in size or bulk to the neighboring structures. In Affordable Housing Development Corp. v. City of Fresno , the city's denial was based on the incongruity of the proposed 324–unit apartment complex in comparison to the surrounding single-family homes. None of these cases present a situation where the members of a large-lot pocket development objected to the development of a neighboring, slightly less dense residential community with smaller, less expensive homes.
104 F.3d 300, 303–04 (9th Cir. 1997).
433 F.3d 1182, 1189, 1196 (9th Cir. 2006).
To the extent the City argues that its denial was out of concern for property values, the court agrees with Plaintiffs that the record does not support such a finding. There was no evidence presented to the City Council showing that Plaintiffs' proposed development would adversely affect neighboring property values. Indeed, at the time of the City's denial, there were no homes within the development abutting much of the Property's western boundary, and the evidence shows that Plaintiffs agreed to provided a buffer consisting of larger, 8,000 square foot lots on its southern boundary for the neighbors in the existing Belleza I development.
The City's second reason for the denial was based on Plaintiffs' failure to accept the City's proposed compromise, which consisted of providing a buffer of 8,000 square foot lots along the full extent of the Property's western border as well as its southern border. As noted by Plaintiffs, "[s]ummary judgment cannot be granted on this issue because material facts surrounding the buffer issue are in dispute." Plaintiffs provide evidence to show that the Hall Company agreed to a buffer consisting of 8,000 square foot lots on the Property's southern border and a portion of its western border in order to address the concerns of neighbors already living in the area. The Hall Company did not agree to extend the larger-lot buffer all the way up the western edge of the Property because the abutting land was vacant and not yet platted. Plaintiffs also presented evidence to support its argument that the City requested the extended buffer on the western edge of the Property because the owner of that undeveloped land agreed to officially rezone her property to R–1–8, thereby guaranteeing large-lot development, if the City would deny Plaintiffs' request. Plaintiff points to evidence to show that this situation was unusual and that any rezoning denial to benefit a neighboring developer would be improper. On this record, viewing the evidence in favor of Plaintiffs, the court cannot conclude the buffer issue was a legitimate basis for denying the rezoning request. Moreover, evidence of Plaintiffs' proposed buffer creates an issue of fact regarding whether there was an alternative to the City's outright denial of the rezoning request that would have had a less discriminatory effect.
Doc. 227 at p. 25.
See Doc. 229 at pp. 40–41 (Plaintiffs' Fact 109).
See Doc. 229 at pp. 41–42 (Plaintiffs' Facts 111–112).
Doc. 229 at p. 43, 46 (Plaintiffs' Facts 116, 123).
--------
V. CONCLUSION
For the foregoing reasons, the City's motion at docket 218 for summary judgment as to Plaintiffs' disparate impact claim under the FHA is DENIED.