Posted on 28 March 2011.
CECIL BARRETT, JR., et al.
H&R BLOCK, INC., OPTION ONE MORTGAGE CORPORATION and H&R BLOCK MORTGAGE CORPORATION n/k/a OPTION ONE MORTGAGE SERVICES, INC.
Civil Action No. 08-10157-RWZ.
United States District Court, D. Massachusetts.
March 21, 2011.
MEMORANDUM OF DECISION
RYA W. ZOBEL, District Judge.
Now pending before the court is Plaintiffs’ motion for class certification. Plaintiffs are African-American homeowners who bring suit on behalf of themselves and similarly situated homeowners against H&R Block, Inc. (“H&R Block”), and its wholly-owned subsidiaries, San Canyon Corp., f/k/a Option One Mortgage Corporation (“Option One”) and Ada Services Corporation, f/k/a H&R Block Mortgage Corporation (“H&R Block Mortgage”) (collectively, “Option One” or “Defendants”).
The gravamen of Plaintiffs’ complaint is that H&R Block and Option One violated the Equal Credit Opportunity Act, 15 U.S.C. §§ 1691-1691f (“ECOA”), and the Fair Housing Act, 42 U.S.C. §§ 3601-3619 (“FHA”), by giving its authorized brokers discretion to impose additional charges to the borrower’s wholesale mortgage loans unrelated to a borrower’s creditworthiness, a policy that had a disparate impact on African-American borrowers in that it resulted in their being charged higher rates than similarly situated whites.
Plaintiffs now move to certify a class of “[a]ll African-American borrowers who obtained a mortgage loan from one of the Defendants between January 1, 2001 and the [d]ate of [j]udgment” under Federal Rule of Civil Procedure 23(b)(3). See Docket # 74, Pl.’s Mem. in Support of Mot. for Class Certification at 17.
Plaintiffs Cecil and Cynthia Barrett (“the Barretts”) purchased their home in 2004 for approximately $277,000. See Second Am. Compl. ¶ 76. They refinanced the mortgage on their home in Mattapan, Massachusetts, in 2005, taking out a $416,000 loan with a 30-year term and a disclosed Annual Percentage Rate, or “APR,” of 8.653 percent. Id. at ¶¶ 77-78. The Barretts were assisted by Money-Wise Solutions, a mortgage broker authorized to originate loans with Option One. Id. at ¶ 79. On April 6, 2006, they again refinanced. Id. at ¶ 81. That loan, also with Option One, was for $500,000, and had an adjustable rate with a balloon feature, providing for a final payment of $344,113.90. Id. at ¶ 82. The APR on the second loan was 10.536%. Id. The remaining plaintiffs similarly used brokers to obtain wholesale mortgage loans from Option One and allege that they were charged a higher APR than similarly-situated whites.
H&R Block made home mortgage loans to consumers through its subsidiaries, H&R Block Mortgage and Option One. See Second Am. Compl. ¶¶ 23, 49. Option One was primarily a wholesale mortgage lender and offered its services through its branches and a national network of mortgage brokers. Id. at ¶ 22.
In the wholesale mortgage lender market, independent mortgage brokers act as intermediaries between borrowers and lenders like Option One. A broker identifies prospective borrowers, facilitates the loan origination process, and transmits prospective borrowers’ respective applications to lenders for a determination of whether or not to grant the loan. This reliance on brokers enabled Option One to fund mortgages in areas where it had not established any retail presence of its own. Option One worked with numerous authorized brokers when it was in the wholesale mortgage business, which it abandoned in late 2007. Between 2001 and 2007, H&R Block Mortgage’s subprime retail originations represented approximately 10% of Option One’s overall loan origination volume.
The pricing of Option One’s mortgage loans was comprised of an objective and a subjective component. According to Plaintiffs, when a proposed borrower applied for a loan, Option One first computed a risk-based financing rate (the “Par Rate”) based on objective criteria of creditworthiness, such as FICO score, property value, and loan-to-value ratio to determine credit parameters, and set prices for its loan products. This information was communicated to brokers on a rate sheet listing Option One’s “par” interest rate, which did not result in any broker compensation. That objective component of loan pricing is not at issue here.
Option One also authorized a subjective component in its credit pricing system (the “Discretionary Pricing Policy”), which governed brokers’ compensation for their services. This is the policy at issue. Under this policy, brokers were permitted to set interest rates higher than the par rate, as well as to charge loan origination and processing fees. Option One paid brokers a “yield spread premium” or “rebate” when they did so. Brokers were paid more for loans that cost the borrower more, though their total compensation was capped at 5 percent of the loan amount. As the name implies, there were no objective criteria for the imposition of these higher rates and fees, which were set by the brokers in their discretion. These discretionary charges were negotiated between the broker and borrower as part of the total finance charge (the “Contract APR”), without specific disclosure that a portion of the Contract APR was a non-risk related charge.
Option One, along with H&R Block and H&R Block Mortgage, jointly established the Discretionary Pricing Policy and participated in the decisions to grant credit to borrowers. (Id. ¶¶ 53-54.) Option One monitored the fees charged by its brokers to ensure they complied with its policies.
Plaintiffs allege that “by design,” the Discretionary Pricing Policy “caused persons with identical or similar credit scores to pay different amounts for the cost of credit.” (Id. ¶ 68.) Although facially neutral, the policy had an adverse effect on African-Americans in that they paid higher discretionary charges on their home loans than did similarly situated white borrowers. Plaintiffs bring these claims under a disparate impact theory, challenging “practices that are facially neutral in their treatment of different groups but that in fact fall more harshly on one group than another and cannot be justified by business necessity.” Int’l Bhd. of Teamsters v. United States, 431 U.S. 324, 335 n. 15 (1977).
II. Legal Standard
To obtain class certification, plaintiffs must satisfy four requirements of Federal Rule of Civil Procedure 23(a) as well as one of several requirements of Rule 23(b). Smilow v. Southwestern Bell Mobile Systems, Inc., 323 F.3d 32, 38 (1st Cir. 2003).
Rule 23(a) provides that a class may be certified only if “(1) the class is so numerous that joinder of all members is impracticable; (2) there are questions of law or fact common to the class; (3) the claims or defenses of the representative parties are typical of the claims or defenses of the class; and (4) the representative parties will fairly and adequately protect the interests of the class.” Fed. R. Civ. P. 23(a). Courts have characterized this rule to require plaintiffs to satisfy the requirements of numerosity, commonality, typicality, and adequacy. See Smilow, 323 F.3d at 38.
Rule 23(b) allows for several different types of class actions. Plaintiffs seek to certify the class under Rule 23(b)(3) which requires a showing “that the questions of law or fact common to class members predominate over any questions affecting only individual members, and that a class action is superior to other available methods for fairly and efficiently adjudicating the controversy.” Fed. R. Civ. P. 23(b).
Before certifying a class, courts are required to engage in “a rigorous analysis of the prerequisites established by Rule 23.” In re New Motor Vehicles Canadian Export Antitrust Litigation, 522 F.3d 6, 17 (1st Cir. 2008). Accordingly, when considering disputed issues for class certification, a district court may “probe behind the pleadings to formulate some prediction as to how specific issues will play out.” DeRosa v. Massachusetts Bay Commuter Rail Co., 694 F. Supp. 2d 87, 95 (D. Mass. 2010) (citations omitted). However, the court may not consider whether the party seeking class certification has stated a cause of action or is likely to prevail on the merits. See In re Initial Public Offering Securities Litigation, 471 F.3d 24, 36-37 (2d Cir. 2006). A district court must certify a class if it concludes that the moving party has met its burden of proof on each element.
A. Rule 23(a)
Under Rule 23(a)(1), the numerosity requirement is met if “the class is so numerous that joinder of all members is impracticable.”
From 2001 through 2007, Option One made at least 130,000 wholesale and retail loans to African-American borrowers located across the United States. Defendants do not dispute that the numerosity requirement has been met.
To demonstrate commonality under Rule 23(a)(2), Plaintiffs must establish “common questions of law and fact.” Fed. R. Civ. P. 23(a)(2). It is not necessary that members of the proposed class share every fact in common or present identical legal issues. See In re Transkaryotic Therapies, Inc. Securities Litig., 03-cv-10165-RWZ, 2005 WL 3178162, at *2 (D. Mass. Nov. 28, 2005) (internal quotations omitted). Rather, the rule requires “a sufficient constellation of common issues [that] binds class members together.” Waste Mgmt. Holdings, Inc. v. Mowbray, 208 F.3d 288, 296 (1st Cir. 2000). In actions based on disparate impact, commonality is satisfied if the lawsuit “tend[s] to show the existence of a class of persons affected by a company-wide policy or practice of discrimination.” Attenborough v. Const. and General Bldg. Laborers’ Local 79, 238 F.R.D. 82, 95 (S.D.N.Y. 2006). Individual factual differences among the putative class members will not preclude a finding of commonality. See Armstrong v. Davis, 275 F.3d 849, 868 (9th Cir. 2001).
To make out a prima facie case of discrimination under the disparate impact theory, a plaintiff must (1) identify the specific practice being challenged; and (2) show that it effected different results in different populations. See Watson v. Ft. Worth Bank and Trust, 487 U.S. 977, 994-995 (1988). “[I]t is not enough to simply allege that there is a disparate impact on workers, or point to a generalized policy that leads to such an impact. Rather, the [plaintiff] is responsible for isolating and identifying the specific … practices that are allegedly responsible for any observed statistical disparities.” See Smith v. City of Jackson, 544 U.S. 228, 241 (2005) (internal quotations omitted). Moreover, “[p]roof of disparate impact is based not on an examination of individual claims, but on a statistical analysis of the class as a whole.” In re Wells Fargo Residential Mortg. Lending Discrimination Litigation, 08-md-01930, 2010 WL 4791687, *2 (N.D. Cal. 2010) (internal citations omitted).
Once the plaintiff has established a prima facie case of disparate impact, the burden of proof shifts to the defendant, who may either discredit the plaintiff’s statistics or proffer its own computations to demonstrate that no disparity exists. See Watson, 487 U.S. at 996-97.
First, Option One argues that the named Plaintiffs cannot satisfy the requirement of commonality because the results of an aggregated statistical regression cannot supply classwide proof of discrimination, particularly where individual Plaintiffs did receive a lower APR. It relies on its studies that the majority of putative class members paid an amount that was statistically the same as they would have paid had they been white, and that another 2.6% of class members paid an amount less than predicted. Defendants contend that such disparities as existed are explainable not by race but by factors such as geography and the individual broker. Second, Defendants contend that individualized pricing, changes in policy, and other practices preclude classwide adjudication of Plaintiffs’ claims.
Plaintiffs demonstrate common questions of fact and law through the expert report of Yale Law School Professor Dr. Ian Ayres (“Professor Ayres”), whose analysis of Option One’s mortgage data leads him to conclude that the Discretionary Pricing Policy did have a disparate impact on minority borrowers because “African Americans paid more for Option One mortgage loans than whites with similar risk-characteristics.” Docket # 89-3, Report of Professor Ayres (“Ayres Report”) at 6, ¶ 10. In his study, Professor Ayres compares the annual percentage rate, or “APR,” paid by white and minority borrowers for Option One wholesale loans originated from 2001 to 2007. He finds that the mean APR for African-Americans was 9.876%, as compared with a mean APR of 9.415% for whites, a difference of 0.461%. See Ayres Report at 7, ¶ 10.
To compare similarly situated whites and minorities, Professor Ayres also performed regression analysis, a statistical method that allows him to control for legitimate risk factors that may affect the cost of a loan. Controlling for such risk factors, he concluded that the APRs of African-Americans are 0.086% higher than those of similarly situated whites, resulting in an average payment of $134 more per year for each of the former group’s loans. Professor Ayres’ study relies entirely on evidence common to the class and does not require any individualized inquiry.
The central question of fact and law is common to the class. Plaintiffs assert that the discretionary pricing strategy they challenge was executed uniformly, and its adverse effects were felt in the same way by Plaintiffs and all class members. Therefore, common questions include whether Option One’s policy resulted in a pricing disparity between white and minority borrowers and whether those disparities are justified by legitimate risk factors.
Defendants dispute commonality through their own expert, Dr. Darius Palia (“Dr. Palia”), Professor of Finance and Economics at Rutgers Business School, who asserts that there is no evidence that there was “a commonly applied `Discretionary Pricing Policy’ that was the cause of a class-wide disparate impact on African-American borrowers.” See Docket # 89-1 (Rebuttal Report of Dr. Palia dated May 4, 2010, hereinafter “Palia Report”).
Using Professor Ayres’ numbers, Dr. Palia replicated Professor Ayres’ exact regression analysis to highlight alleged errors. Dr. Palia points to two major flaws in Professor Ayres’ analysis. First, he argues that the Ayres regression model was improperly applied to the aggregate, and not separately to the individual mortgage brokers that used the so-called “Discretionary Pricing Policy.” If such a policy had, in fact, been applied, “the disparate impact caused by the policy should be observed consistently across the various brokers that applied it”; if not, “that would suggest that loan pricing is the result of individualized decision-making rather than the result of a common policy.” Second, Dr. Palia contends that Professor Ayres’ failure to apply his regression model separately to local geographic markets in which borrowers applied for and obtained their mortgage loans renders his conclusions inaccurate. After completing his own analysis, Dr. Palia concluded: “(1) the statistical evidence does not show any common pattern of disparate impact against African-American borrowers either across the brokers that originated the loans or across the geographic markets in which the largest numbers of loans were originated; (2) even among the minority of brokers and geographic markets in which African-Americans experienced statistically higher APRs than similarly-situated whites, there is no common cause of such pricing differences; and (3) nine of ten loans extended to named plaintiffs had APRs that were not statistically different from the APR that would have been predicted had the borrowers been white.”
Although Defendants hotly dispute the merits of Professor Ayres’ analysis, it has long been the rule that disputes about the respective experts’ statistics are tantamount to disputes about the parties’ proof of the merits and are not grounds for denying class certification. See In re Initial Public Offerings Securities Litigation, 471 F.3d 24, 35 (2d Cir. 2006) Plaintiffs have satisfied the commonality requirement. (experts’ disagreement on whether a discriminatory impact could be shown is a disagreement as to the merits, and is not a valid basis for denying class certification). Statistical disputes in civil rights cases “encompass the basic merits inquiry and need not be proved to raise common questions and demonstrate the appropriateness of class resolution.” Id. at 594.
A plaintiff may represent a class only if his or her claims are “typical” of those of the putative class. See Fed. R. Civ. P. 23(a)(3). In general, a plaintiff’s claim is typical if it “arises from the same event or practice or course of conduct that gives rise to the claims of other class members, and if his or her claims are based on the same legal theory.” In re Pharm. Indus. Average Wholesale Price Litig., 230 F.R.D. 61, 78 (D. Mass. 2005). Where, however, “a named plaintiff may be subject to unique defenses that would divert attention from the common claims of the class, that plaintiff cannot be considered typical of the class.” In re Bank of Boston Corp. Securities Litigation, 762 F.Supp. 1525, 1532 (D. Mass. 1991). While commonality “examines the relationship of facts and legal issues common to class members,” typicality “focuses on the relationship of facts and issues between the class and its representatives.” Dukes v. Wal-Mart Stores, Inc., 603 F.3d 571, 613 n. 37 (9th Cir. 2010) (en banc).
Here, Plaintiffs contend that their claims are typical because Option One made loans to each Plaintiff under the same subjective Discretionary Pricing Policy to which the class was subjected.
Option One counters that the named Plaintiffs are not typical for two reasons: (1) some have suffered no injury in connection with their loans and therefore lack standing; and (2) individualized defenses demonstrate that there is no “typical” named plaintiff.
Defendants assert that certain Plaintiffs were not injured because they received loans that were priced more favorably than similarly situated white borrowers. Further they argue that Plaintiffs’ reliance on Dr. Ayres’ conclusions of disadvantage to African-American borrowers as a group does not support the inference that the named Plaintiffs were so disadvantaged. Absent such individualized evidence, the named Plaintiffs are not typical of the class they represent, and thus lack standing.
Plaintiffs have alleged that the disparate impact was the result of the Discretionary Pricing Policy, a common practice that governed the pricing of all class members’ mortgages. The named Plaintiffs were subject to that policy, and have advanced a viable theory showing that it produced harm. That is sufficient to satisfy the typicality requirement.
ii. Individualized Defenses
Next, Defendants contend that the individual circumstances surrounding each named Plaintiff’s loans expose each to individual defenses which defeat typicality. In particular, Defendants contend that several Plaintiffs submitted loan applications which contained false information, subjecting them to a defense of unclean hands. This argument is unavailing. The U.S. Supreme Court has held that because the purpose of the ECOA is to eradicate discrimination, the unclean hands defense is not available to question liability. See McKennon v. Nashville Banner Publishing Co., 513 U.S. 352, 356-57, 360 (1995) (holding that the unclean hands defense “has not been applied where Congress authorizes broad equitable relief to serve important national policies” including civil rights statutes such as the ADEA); see also Moore v. U.S. Department of Agriculture, 55 F.3d 991, 995-96 (5th Cir. 1995) (holding that an unclean hands defense did not defeat liability under the ECOA).
Finally, Defendants say that the necessity for an individualized statute of limitations defense determination defeats typicality. This, too, is without merit. First, this court has already ruled against Defendant’s statute of limitations defense with respect to the named Plaintiffs when it denied their motion to dismiss. Second, all named Plaintiffs but one filed within the requisite time. Third, at the class certification stage, a court’s analysis of unique defenses focuses on whether those defenses will “unacceptably detract from the focus of the litigation to the detriment of absent class members.” Baffa v. Donaldson, Lufkin & Jenrette Sec. Corp., 222 F.3d 52, 59 (2d Cir. 2000). Here, any statute of limitations defense will not do so.
Rule 23(a)(4) requires that the proposed class representatives “fairly and adequately protect the interests of the class.” This requirement has two parts. Plaintiffs must first demonstrate that “the interests of the representative party will not conflict with the interests of any of the class members,” and second, that “counsel chosen by the representative party is qualified, experienced and able to vigorously conduct the proposed litigation.” In re M3 Power Razor System Marketing & Sales Practice Litigation, 270 F.R.D. 45, 55 (D. Mass. 2010) (citing Andrews v. Bechtel Power Corp., 780 F.2d 124, 130 (1st Cir. 1985)).
Option One does not dispute the adequacy of these class representatives, and the court discerns no conflicts between Plaintiffs and any members of the class. Accordingly, all four requirements of Rule 23(a) have been met.
B. Rule 23(b)(3)
As they request certification under Rule 23(b)(3), Plaintiffs must present evidence showing the predominance of common issues and the superiority of a class action. The court now turns to these two requirements.
Rule 23(b)(3) requires the court to find “that the questions of law or fact common to class members predominate over any questions affecting only individual members.” This predominance requirement “tests whether proposed classes are sufficiently cohesive to warrant adjudication by representation” and is a “far more demanding” standard than Rule 23(a)’s commonality requirement. Amchem Prod., Inc. v. Windsor, 521 U.S. 591, 623-24 (1997). The Rule is intended to ensure “that common issues predominate, not that all issues be common to the class.” In re Transkaryotic Therapies, Inc. Securities Litigation, 2005 WL 3178162, *2 (D. Mass. 2005) (citations omitted).
Option One disputes predominance by reiterating its arguments against commonality. The disparity in APR is explained not by race, Option One argues, but by other legitimate variables.
The key question again is whether Option One’s Discretionary Pricing Policy had a disparate impact, that is, whether it fell “more harshly on one group than another and cannot be justified by business necessity.” Int’l Bhd. of Teamsters v. United States, 431 U.S. 324, 335 n. 15 (1977). Since the claim is disparate impact, the relevant evidence will focus on “statistical disparities, rather than specific incidents, and on competing explanations for those disparities.” Watson v. Ft. Worth Bank & Trust, 487 U.S. 977, 987 (1988).
Professor Ayres’ analysis provides evidence of the disparate impact on a class-wide basis. Competing explanations for those disparities are examined by way of regression analyses that assess the effect of competing variables. Option One can defend against Plaintiffs’ case either by demonstrating that its discretionary policy had a valid business justification, or by challenging the statistical basis for Plaintiffs’ claim. In either case, the legal contention applies across the class. Thus, Plaintiffs have carried their burden of showing the predominance of common questions.
The final requirement for class certification is “that a class action is superior to other available methods for fairly and efficiently adjudicating the controversy.” Fed. R. Civ. P. 23(b)(3). The pertinent factors in assessing superiority are “the class members’ interests in individually controlling the prosecution or defense of separate actions” and “the likely difficulties in managing a class action.” Id. Superiority exists where “there is a real question whether the putative class members could sensibly litigate on their own for these amounts of damages, especially with the prospect of expert testimony required.” Gintis v. Bouchard Transp. Co., Inc., 596 F.3d 64, 68 (1st Cir. 2010).
It would be neither economically feasible nor efficient for class members to pursue these claims against Option One individually. The amounts recoverable for individual class members are too low for class members to bring individual claims. The class action is manageable because liability will be determined based on statistical proof, and remedies can be calculated on a class-wide basis. A class is therefore superior to other methods for adjudicating these claims.
C. Class Period
While the result on the merits is by no means certain, the proposed class satisfies the requirements of Rule 23, and class certification is appropriate. However, several questions remain unresolved. Most notably, the proposed dates of the beginning and end of the class period are left singularly unsubstantiated. Moreover, it is unclear how and when Option One began to identify loan applicants by race.
D. Rule 23(g)
Since the court has determined that Plaintiffs have met Rule 23’s requirements for class certification, the court must appoint class counsel. See Fed. R. Civ. P. 23(g). On or before April 11, 2011, any counsel who wishes to serve as class counsel shall file the requisite motions and documentation to support his/her request.
Plaintiffs’ motion for class certification (Docket # 72) is ALLOWED, subject to limitation by time. A class of “[a]ll African-American borrowers who obtained a mortgage loan from one of the Defendants” is hereby certified.
 H&R Block Bank, a Federal Savings Bank, Member FDIC, was named as a defendant but has since been voluntarily dismissed from the action. H&R Block, Inc. was dismissed for lack of personal jurisdiction. The only defendants remaining are Option One Mortgage Corporation and Option One Mortgage Services, Inc., which became the new name of H&R Mortgage Services in July 2007.
 Plaintiffs are Cecil Barrett, Jr., Cynthia Barrett, Jean Blanco Guerrier, Angelique M. Bastien, Jacqueline Grissett, Craig Grissett, Steven Parham, Betty and Edward Hoffman, Doris Murray, Joslyn Day and Keisha Chavers (collectively “Plaintiffs”),
 The ECOA provides that it is unlawful “for any creditor to discriminate against any applicant, with respect to any aspect of a credit transaction-(1) on the basis of race, color, religion, national origin, sex or marital status, or age (provided the applicant has the capacity to contract).” 15 U.S.C. § 1691(a). Similarly, the FHA makes it unlawful “for any person or other entity whose business includes engaging in residential real estate-related transactions to discriminate against any person in making available such a transaction, or in the terms or conditions of such a transaction, because of race, color, religion, sex, handicap, familial status, or national origin.” 42 U.S.C. § 3605(a).
 This court previously concluded that disparate impact claims are cognizable under both the FHA and ECOA. See Order Denying Mot. to Dismiss (Docket # 45) at 3-5; see also Langlois v. Abington Hous. Auth., 207 F.3d, 43, 49 (1st Cir. 2000) (disparate impact claims allowable under FHA); and Miller v. Countrywide Bank, N.A., 571 F. Supp. 2d 251, 256-257 (D. Mass. 2008) (disparate impact claims allowable under ECOA).
 Citing Stastny v. Southern Bell Tel. & Tel. Co., 628 F.2d 267 (4th Cir. 1980), Defendants further contend that delegation of discretion cannot, as a matter of law, form the policy required to make out a claim of disparate impact discrimination. This argument is unavailing. It is not the delegation of discretion that constitutes the policy, but rather the existence of a commonly applied practice that satisfies the requirement. See Watson v. Fort Worth Bank, 487 U.S. 977 (1988) (policies which designate discretionary authority to individual actors are actionable if they have a verifiable discriminatory impact on a protected class); see also Dukes v. Wal-Mart Stores, Inc., 603 F.3d 571, 612 (9th Cir. 2010) (en banc) (same).
Moreover, this court previously held that Plaintiffs adequately identified the practice at issue, namely “establishing a par rate keyed to objective indicators of creditworthiness while simultaneously authorizing additional charges keyed to factors unrelated to those criteria.” Barrett v. H & R Block, 08-cv-10157-RWZ (Docket # 45) at 7.
 Defendants contend that arguments that one party’s statistics are “unreliable or based on an unaccepted method” must be resolved at the certification stage. See Dukes v. Wal-Mart Stores, Inc., 603 F.3d 571, 591-592 (9th Cir. 2010) (en banc). Here, however, Defendants do not contend that the statistical analysis was based on an unaccepted method. Rather, they contend that Dr. Ayres’ model produces results which do not prove a disparate impact caused by any policy.
[ipaper docId=51688987 access_key=key-7e4f5josfyyppxsp3ka height=600 width=600 /]
© 2010-17 FORECLOSURE FRAUD | by DinSFLA. All rights reserved.