Ronald M. Whyte United States District Judge

ORDER ON DEFENDANT'S MOTION FOR SUMMARY JUDGMENT; PLAINTIFFS' MOTION FOR SUMMARY JUDGMENT ON AFFIRMATIVE DEFENSES; MOTION TO BIFURCATE; MOTION TO DECERTIFY; AND MOTION TO STRIKE

[Re Docket Nos. 386, 383, 392, 387, 388]

In this certified class action, plaintiffs seek to show that defendant First American eAppraiseIT ("EA") violated Section 8(a) of the Real Estate Settlement Procedures Act ("RESPA"), 12 U.S.C. § 2607(a), by agreeing with Washington Mutual Bank ("WMB") to provide inflated home appraisals in exchange for business referrals. The parties bring the following motions addressed in this order:

• Defendant's Motion for Summary Judgment (Dkt. No. 386);



• Plaintiffs' Motion for Summary Judgment on Defendant's Affirmative Defenses (Dkt. No. 383);
• Plaintiffs' Motion for Bifurcation (Dkt. No. 392);



• Defendant's Motion to Decertify (Dkt. No. 387); and



• Defendant's Motion to Strike Supplemental Expert Reports (Dkt. No. 388).

For the reasons explained below, the court allows this case to proceed as a class action, and bifurcates the question of whether EA appraisals were inflated in the aggregate as a result of an agreement between WMB and EA from the question, if it is reached, of what damages were suffered by plaintiffs.

At present, the court tentatively envisions the case proceeding as follows: the issue of whether there was inflation of appraisals on an aggregate basis as a result of an agreement between WMB and EA will be bifurcated and tried in accordance with the current trial schedule. The amount of class damages, either in total or on an individual basis, will not be determined as part of the bifurcated issue. If plaintiffs prevail on the bifurcated issue, potential class members will submit verified claims confirming that they paid the appraisal fee, the amount of the appraisal fee, and the RESPA status of their loan. Concurrently, plaintiffs may obtain the HUD-1 forms from Chase. Then, the parties will review the claim forms and any newly submitted HUD-1 forms, and determine which, if any, claims of the putative class members require further investigation to determine their eligibility for recovery. Although the defendant believes that this process will swallow the entire litigation, the court is not convinced that an overwhelming number of putative class members' claims will require further investigation beyond the claim form. For those class members submitting claims that defendant wishes to challenge or further investigate, the court will hold proceedings, if requested, to determine their eligibility. Thus, questions of individual damages and eligibility will be reviewed on a claim-by-claim basis, preserving EA's rights.

I. Background

A. Summary of Plaintiffs' RESPA § 8(a) Claim

Plaintiffs Felton Spears and Sidney Scholl bring this action on behalf of themselves and the class of "[a]ll consumers in California and throughout the United States who, on or after June 1, 2006, received home loans from Washington Mutual Bank, FA in connection with appraisals that were obtained through eAppraiseIT." Dkt. No. 249 (Cert. Order).

RESPA § 8(a) states:

(a) Business referrals



No person shall give and no person shall accept any fee, kickback, or thing of value pursuant to any agreement or understanding, oral or otherwise, that business incident to or a part of a real estate settlement service involving a federally related mortgage loan shall be referred to any person.

As detailed in the court's prior orders, plaintiffs must provide enough evidence for a fact finder to (1) conclude that WMB received appraisal values that were inflated in the aggregate; (2) conclude that WMB and EA agreed to exchange inflated appraisals for referrals; (3) conclude that class members' loans are covered under RESPA; (4) determine the cost of the appraisal for each of the class members' loans; (5) determine that WMB funded the loan; and (6) determine that the class member paid the appraisal fee. If plaintiffs are successful, pursuant to 12 U.S.C. § 2607(d)(2), damages are set at "an amount equal to three times the amount of any charge paid for such [appraisal]."

B. Factual Allegations

Home purchases in the United States have traditionally been financed through a third-party lender who retains a security interest in the property until the loan is repaid. Dkt. No. 149, Second Amended Complaint ("SAC") ¶ 2. In order to ensure that the secured lender will recoup the value of the loan if the borrower defaults before the loan is repaid, the lender generally requires that the property be professionally appraised. Id. In recent years, however, rather than hold mortgage loans until repaid, banks have frequently sold the loans to other financial institutions. Id. ¶ 23. This shift created an incentive for banks to seek higher appraisals for the properties underlying their mortgage loans, thus justifying larger loans. Id. ¶ 24.

Plaintiffs allege that beginning in June of 2006, WMB conspired with EA and Lender's Service, Inc. ("LSI") to inflate the appraised value of property underlying their mortgage loans so that WMB could sell the aggregated security interests in these properties at inflated prices. Id. ¶ 6. Around June 2006, WMB retained EA and LSI to administer its appraisal program. Id. ¶ 36. EA and LSI have since performed almost all of WMB's appraisals, and WMB's borrowers became EA and LSI's largest source of business. Id.

Plaintiffs allege that defendants engaged in the following conduct as part of the conspiracy to inflate appraisals: (1) EA and LSI complied with WMB's demand that all of its appraisals be performed by appraisers on its "Proven Appraiser List," which contained appraisers selected by WMB's loan origination staff; (2) WMB maintained the contractual right to challenge appraisals by requesting a reconsideration of value ("ROV") and used ROV requests to get EA and LSI to increase appraisal values; (3) WMB requested that EA and LSI hire former WMB employees as appraisal business managers ("ABMs"), who had the authority to override the values determined by third-party appraisers; and (4) EA and LSI altered appraisal reports to reflect higher property values, remove negative references, and make other changes so that the final appraisal reports complied with WMB's wishes. Id. ¶¶ 6, 37-40, 43-45.

C. Procedural Background

On February 8, 2008, plaintiffs filed suit against WMB, EA, and LSI, alleging breach of contract, unjust enrichment, and violations RESPA, the Unfair Competition Law, and the Consumers Legal Remedies Act. The only claim remaining is plaintiffs Felton A. Spears and Sidney Scholl's claim against EA for violation of Section 8(a) of RESPA, 12 U.S.C. § 2607(a). The court dismissed all other claims against EA in orders issued on March 9, 2009 (Dkt. No. 147) and August 30, 2009 (Dkt. No. 169). In its August 30, 2009 order, the court also denied a motion to permit Juan and Carmen Bencosme to intervene and dismissed LSI from this case.

Spears and Scholl moved for class certification on May 25, 2010. On July 2, 2010, the court denied class certification and held that, while the class was ascertainable and the prerequisites of Federal Rule of Civil Procedure 23(a) had been satisfied, plaintiffs had not satisfied the requirements of Rule 23(b)(3). Dkt. No. 209. To give the parties an opportunity to address the question of whether the inflation of appraisal values in the aggregate could be established by common proof, the court denied the motion for class certification without prejudice.

Plaintiffs then renewed their motion for class certification, Dkt. No. 217, and the court granted the motion, Dkt. No. 249. The court certified the class of "[a]ll consumers in California and throughout the United States who, on or after June 1, 2006, received home loans from Washington Mutual Bank, FA in connection with appraisals that were obtained through eAppraiseIT." Id.

In February 2013 the parties filed cross-motions for judgment on the pleadings, and the court denied defendant's motion and granted in part plaintiffs' motion. Dkt. No. 309 (JOP Order). The case then proceeded with discovery, giving rise to several discovery disputes. As a result of discovery and production issues with nonparty J.P. Morgan Chase Bank ("Chase"), described in detail below, the court extended the discovery schedule. Dkt. No. 350 (Sch. Order). The case is currently set for a pretrial conference on October 9, 2014 and for trial on November 24, 2014. Id.

D. Preliminary Issues

The court begins by providing background for two issues that are raised in multiple motions.

1. The Chase Discovery: HUD-1 Forms and Chase Spreadsheets

The background of the Chase discovery is complicated and not always clear from the parties' papers. This discussion is based primarily on the parties' statements in their October 10, 2013 Joint Case Management Statement ("Joint CMS"), Dkt. No. 342, and the parties' briefing on the Motion to Strike, Dkt. Nos. 359-9, 409-5, and 416-6. The court first discusses the procedural history of the Chase discovery and then how the parties propose to use the Chase documents in this case. The Chase production is relevant to Defendant's Motion for Summary Judgment, Motion to Decertify, Motion to Bifurcate, and Motion to Strike.

a. History of the Chase Discovery

Chase, as the entity that purchased all of the class members loans from the FDIC after WMB went into receivership, has WMB's loan files for most or all of the class members. The loan files typically include a "HUD-1 Settlement Statement" ("HUD-1"), a form required by law for all RESPA loans, which itemizes all charges related to a loan.

In October 2012, plaintiffs first subpoenaed from Chase all HUD-1 forms for mortgage loans "made by [WMB]" for which EA performed the appraisal. Dkt. No. 409-5 at 4; Dkt. No. 371. Plaintiffs estimate that this would be about 115,000 HUD-1 forms. Dkt. No. 371 at 1. Plaintiffs also sought production of "loan data" from all of the loan files related to an EA appraisal. Plaintiffs also requested 450 "funded loan files." The court refers to these as the "450 sample loans." The 450 sample loans were to be used to evaluate whether WMB received appraisals that were inflated in the aggregate, in a process described by plaintiff's expert Dr. French. See Dkt. No. 222. The 450 sample loans needed to be representative of the class, and therefore plaintiffs wanted to ensure that the sample only included loans that were funded by WMB.

During the course of discovery, both parties agreed that the Chase information was necessary to show whether a loan was funded by WMB, who paid for the appraisal, the amount of the appraisal fee, and to determine whether the loan is covered by RESPA. Plaintiffs also needed the Chase "loan data" to confirm that the 450 appraisals they selected to prove inflation were in fact related to funded WMB loans. Plaintiffs therefore needed the "loan data" to complete their expert reports on the key liability issue: whether EA appraisals were inflated in the aggregate as a result of an agreement between WMB and EA.

Through a lengthy meet and confer process, Chase and plaintiffs narrowed the discovery to "more specifically identif[y] the particular loan-level data" plaintiffs needed. Dkt. No. 342 at 3. "Chase then located and provided some of the loan data responsive to the Parties' subpoenas, which it produced on July 31, 2013 in an Excel spreadsheet." Id. (emphasis in original). The July 2013 Spreadsheet is referred to by the parties as "JPMC000000."

In August 2013, Chase informed the plaintiffs that it could not complete the requested discovery by the discovery cut-off, but would produce additional reports responsive to the parties' subpoenas. These additional reports were produced September 3, 2013. Id. at 4. The reports did not have the information plaintiffs required. Chase then offered to send more information in early October 2013.

Following the deposition of a Chase corporate representative, plaintiffs concluded that they could not rely on the Chase data produced up to that point to determine the funding status, RESPA status, or appraisal fee payment of the WMB loans. On December 6, 2013, plaintiffs finally obtained "a definitive answer" that the Chase spreadsheets could not be used to reliably determine whether the borrower paid for the appraisal. Dkt. No. 409-5 at 5; Dkt. No. 371.

Plaintiffs also had concerns with the 450 sample loans Chase produced. Plaintiffs initially relied on Chase's representation that "a loan number field in the spreadsheet would indicate whether or not the loans associated with the loan files were funded," but later determined they could not rely on Chase's representation. Dkt. 342 at 7. Accordingly, plaintiffs had to select some new loans for their sample, which took more time to produce. By October 2013, it appears that plaintiffs still did not have a complete sample set of relevant loans.

Defendant was unsympathetic to the failure of plaintiffs to obtain the necessary Chase documents, and opposed any extension of discovery for Chase's production. Defendant noted that plaintiffs did not file any motions to compel and that plaintiffs were still not sure that Chase could produce the requested information. Therefore, defendant believed that any discovery extension should be denied.

After holding a case management conference, the court noted both parties' concerns and granted plaintiffs an extension to December 22, 2013 to obtain documents from Chase. See Dkt. No. 350 (Scheduling Order). The court's order expressly noted "Spears will have to make his case with whatever information he can obtain by December 22, 2013." Id. at 2.

Following the discovery extension, plaintiffs further communicated with Chase about the reliability and usefulness of the spreadsheets produced. On December 6, 2013, Chase informed the plaintiffs that it could not confirm that the spreadsheets showed who paid the appraisal fee. Dkt. No. 371 at 2. On December 9, 2013, plaintiffs filed a motion before Magistrate Judge Lloyd to compel production of all the HUD-1 forms, which was granted, Dkt. No. 359, and allowed production by a date mutually agreed to by plaintiffs and Chase. The paper HUD-1 forms have still not been produced.

On December 23, 2013 (the last business day of the discovery period), Chase produced three additional spreadsheets. Dkt. No. 409-5 at 6. The parties label these spreadsheets "JPMCHASE_WAMU_000001" through "JPMCHASE_WAMU_000003." These spreadsheets included "codes" that "render[ed] it useless." Id. Plaintiffs asked Chase for an explanation of the codes "but they received no response from Chase." Id. Plaintiffs state that they could not fully understand the codes in the Chase spreadsheets until May 3, 2014. Id. Defendant does not appear to dispute this.

b. How the Parties Propose to Use the Chase Discovery in this Case

Plaintiffs' current position on the Chase data appears to be that they can reliably determine the number of RESPA-qualifying, WMB funded loans from the data Chase has already provided. Plaintiffs can also calculate an average appraisal fee. However, plaintiffs acknowledge that they must have the HUD-1 forms to determine whether the borrower, as opposed to the lender, paid the appraisal fee. Plaintiffs therefore rely on the 450 sample loans to determine the percentage of borrowers who paid the appraisal fee.

Defendant's position is that plaintiffs cannot reliably determine the RESPA status of the loan from the Chase data, and that because the class-wide HUD-1 forms were not produced by the close of discovery, they cannot be used by plaintiffs. Defendant also disputes that the HUD-1 forms will reliably show whether the borrower paid the appraisal fee, pointing specifically to named plaintiff Spears' HUD-1 form. Spears' HUD-1 form shows that he paid $361 for an appraisal fee to EA (or an EA subsidiary), but also received a $980 "credit-customer retent[ion] to Washington Mutual Bank." Dkt. No. 149-3 at 2 (lines 803 and 813). The total of all charges for "items payable in connection with loan" is $1,165. Id. Therefore, EA states the most Spears could have paid for the appraisal was $185 ($1,165 less the $980 credit). See also Dkt. No. 390-20 (Grice Rebuttal Report) ¶¶ 49-51 (analyzing Spears' HUD-1 and questioning whether HUD-1 forms reliably show the appraisal fee actually paid by the borrower). Neither party provides any explanation as to how much Spears paid beyond pointing to the HUD-1 form.

2. How Plaintiffs Will Show the RESPA Status of a Loan, and How Defendant Will Challenge Plaintiffs' Showing

The second key issue underlying the instant motions is how plaintiffs will show how many and which loans are covered by RESPA and how defendant will challenge plaintiffs' showing. This is especially relevant to the Motion to Decertify and Motion to Bifurcate.

As explained in more detail below, RESPA does not apply to all credit transactions.

The Regulations [applicable to RESPA] provide that RESPA applies to all federally related mortgage loans except for the exemptions contained in 24 C.F.R. § 3500.5(b). That subsection sets forth seven different exceptions, including a loan on property greater than 25 acres, a loan primarily for business purposes, temporary financing, vacant or unimproved property, subsequent assumptions without lender approval, loans converted to different terms, and secondary-market transactions. 24 C.F.R. § 3500.5(b).

a. The Ghiglieri Report

Plaintiffs rely on the expert report of Catherine Ghiglieri, "an expert with respect to the manner in which banks makes loans" to establish the RESPA-status of class members' loans. Dkt. No. 390-20, Ex. 63 (Ghiglieri Rep.); Dkt. No. 390-20, Ex. 65 (Ghiglieri Supp. Rep.). Ms. Ghiglieri opines that if certain mandatory RESPA disclosures are found in a loan file, then the loan is "presumptively RESPA." The basis for this is that "a bank examiner starts with the presumption that a loan secured by a one-to-four family property is subject to RESPA and would expect to see the RESPA documents in the loan file. Where there are no RESPA documents in the loan file, then the examiner would want to review the bank's determination and reasoning that the loan is exempt from RESPA, which the examiner would expect to be in writing and contained in the loan file." Ghiglieri Rep. at 11. Ms. Ghiglieri notes that "there is no requirement that a bank make an affirmative statement in a loan file that it has determined that a given loan qualifies as a RESPA-covered transaction." Id. Ms. Ghiglieri further concludes that the presence of only the HUD-1 form, and not the other mandatory disclosures, still signals a presumptively RESPA loan. She then concluded that Spears' and Scholl's loans, which include the RESPA documents in their files, are covered by RESPA. In her first report, Ms. Ghiglieri did not make any estimate as to the number of EA appraisals associated with RESPA-qualifying loans.

In Ms. Ghiglieri's supplemental report, she reviews (1) the December 2013 Chase spreadsheets to identify which "loan codes" were subject to RESPA, and (2) the 450 loan sample set to determine which loans were subject to RESPA. Ghiglieri Supp. Rep. at 2. Ms. Ghiglieri also provides some elaboration on her RESPA-presumption theory:

When I was performing examinations of national banks for the OCC to determine compliance with RESPA, I would request the bank to identify the loan categories subject to RESPA. I then selected a sample of loans from the bank records presented. If there was an indicia of RESPA because of the bank's codes, loan types or otherwise, and bank did not treat a loan as a RESPA loan, I would cite the bank for a violation of RESPA, unless there was supporting documentation in the bank's records to support why a loan was not subject to RESPA. In other words, the bank would need to prove to the examiners and provide supporting documentation as to why a loan that appeared to be subject to RESPA was exempt and did not require the use RESPA-compliance documents such as the HUD-1, Good Faith Estimate, etc., in order to avoid being cited for a violation of RESPA.
I used the same methodology in my review of the loans in this case. A loan with a loan code that appeared to be otherwise subject to RESPA was presumed to be a RESPA loan absent documentation in the bank records presented showing why the loan was not a RESPA loan.

Reviewing the sample 450 loans and "consistent with [her] review of the Chase spreadsheets," Ms. Ghiglieri concluded that 192 of the loans were definitely subject to RESPA, 256 were presumed to be subject to RESPA, and 2 were not subject to RESPA. Id. at 5. The two loans that were not subject to RESPA had a "bank documented business purpose," and none of the other loans in the sample did. Ms. Ghiglieri's spreadsheets containing her review of the 450 sample loans include a column for "Chase Spreadsheets" and a column for "Data in Lo_Type field of JPMChase_WaMu_000003", but entries are missing for the majority of the loans.

b. The Grice Report

Defendant submitted the Expert Report of Charles H. Grice, Dkt. No. 390-20, Ex. 64 (Grice Rep.), to rebut Ms. Ghiglieri. Mr. Grice is a "banking consultant with more than 30 years of experience working with financial institutions on regulatory and risk management matters." Id. at ¶ 1.

Mr. Grice opines that the presence of a HUD-1 form in a loan file is "insufficient to create a presumption that the underlying transaction is covered by RESPA" because "mortgage lenders during the relevant period of 2006-2007 (the 'Relevant Period') routinely prepared packages of generic disclosure documents, including HUD-1s, irrespective of whether the loan was subject to RESPA." Id. ¶ 15 (footnote omitted). Mr. Grice also reported that WMB used HUD-1 forms for non-RESPA loans during the Relevant Period based on interviews with WMB staff. Id. ¶ 29.

Mr. Grice states that "a necessary first step in assessing whether a loan transaction is in fact covered by RESPA is a review of the full loan file. This review should focus not only on the borrower's responses to standardized questions in the loan application and related documents (e.g., is the property an 'investment' or 'primary residence?'), but should also look to other documents evidencing the borrower's intentions. If the borrower's own admissions and intentions as reflected in the loan file demonstrate no RESPA coverage, it is appropriate to conclude that the loan is not covered by RESPA, and the analysis is concluded." Id. ¶ 32 (footnote omitted). After reviewing the loan file, Mr. Grice would then "review factors outside the loan file." Id. ¶ 33.

Based upon the type of review that Mr. Grice submits is necessary, he concludes that Scholl's loan was not a RESPA-qualifying loan. Id. ¶ 35.

Thus, not surprisingly, defendant argues that determining RESPA status will require numerous individual issues to determine whether a loan is for personal or business use.

II. Analysis

A. Defendant's Motion for Summary Judgment

Defendant moves for summary judgment that named plaintiff Sidney Scholl does not have a RESPA-qualified loan, that plaintiffs have presented no evidence of a conspiracy at the time of Scholl's or Spears' loan or that appraisals were inflated in the aggregate, and that there is no evidence that class members' loans are covered by RESPA. The court GRANTS the motion for summary judgment as to Scholl, and otherwise DENIES the motion.

1. Legal Standard on Summary Judgment

Summary judgment is proper where the pleadings, discovery, and affidavits demonstrate that there is "no genuine issue as to any material fact and that the moving party is entitled to judgment as a matter of law." Fed. R. Civ. P. 56(c); see also Celotex Corp. v. Catrett, 477 U.S. 317, 322 (1986). At the summary judgment stage, the Court "does not assess credibility or weigh the evidence, but simply determines whether there is a genuine factual issue for trial." House v. Bell, 547 U.S. 518, 559-60 (2006). Material facts are those which may affect the outcome of the case. Anderson v. Liberty Lobby, Inc., 477 U.S. 242, 248 (1986). A dispute as to a material fact is genuine if there is sufficient evidence for a reasonable jury to return a verdict for the nonmoving party. Id. When determining whether there is a genuine issue for trial, "inferences to be drawn from the underlying facts . . . must be viewed in the light most favorable to the party opposing the motion." Matsushita Electric Industrial Co. v. Zenith Radio Corp., 475 U.S. 574, 587 (1986) (citation omitted).

2. Plaintiff Sidney Scholl's Loan is Not Covered by RESPA

RESPA exempts from its coverage any credit transaction involving extensions of credit primarily for business or commercial purposes. 12 U.S.C. § 2606(a)(1). As explained by the Ninth Circuit, the court must look to the "interpretations of [12 C.F.R.] § 226.3 in the Official Staff Commentary on Regulation Z" to determine whether a loan has a business purpose. Johnson v. Wells Fargo Home Mortg., Inc., 635 F.3d 401, 417-18 (9th Cir. 2011). A loan to acquire a "non-owner-occupied rental property," that is "rental property (regardless of the number of housing units) that is not owner-occupied is deemed to be for business purposes." 12 C.F.R. Pt. 226, Supp. I, Cmt. 3(a)(4).

Scholl's loan falls directly within this exemption. Ms. Scholl's loan involves a mortgage on a property in Edmond, Oklahoma that she then leased-back to a developer. Ms. Scholl never occupied, and never intended to occupy, the property. Dkt. No. 385-5 at 5 (citing Owens Decl. Ex. 4, Dkt. No. 390-8, at SCHOLL 005082). Plaintiffs do not contest this, but argue that Ms. Scholl bought the property "as a second home." Dkt. No. 402-4 at 14. Therefore, the plaintiffs argue that the court must apply a five-factor test for owner occupied rental property. See Thorns v. Sundance Props, 726 F.2d 1417, 1419 (9th Cir. 1984); Bergman v. Fidelity Nat. Financial, Inc., No. 12-cv-05994-ODW(MANx), 2012 WL 6013040 (C.D. Cal. Dec. 3, 2012). The plaintiffs do not cite any evidentiary support for their "second home" theory, and do not rebut the undisputed evidence that Ms. Scholl never intended to, and in fact did not, occupy the property. Ms. Scholl's loan was clearly for a non-owner occupied rental property, and therefore had a business purpose and is not covered by RESPA. The court GRANTS summary judgment in favor of defendants as to named plaintiff Sidney Scholl.

3. There Are Material Disputes of Fact as to Whether There Was an Agreement to Exchange Inflated Appraisals for Business Referrals

To prevail on their RESPA § 8(a) claim, plaintiffs must show that EA and WMB had an "agreement or understanding" to exchange inflated appraisals for business referrals. EA argues that plaintiffs have failed to present evidence of any agreement or conspiracy.

At this stage, plaintiffs' evidence can be summed up as: (1) Dr. French's and Dr. Courchane'se showing of inflation in the aggregate through sampling EA appraisals and conducting retrospective reviews, (2) EA and WMB's business arrangements that created an environment that allowed for and encouraged inflation of appraisal values, and (3) the awareness of both parties that EA was providing inflated appraisals. From this evidence, a fact finder could draw the reasonable inference that EA and WMB actually agreed to exchange inflated appraisals for business referrals. The court therefore DENIES summary judgment on the agreement and inflation issue.

a. Evidence Tending to Show an Agreement

"An agreement or understanding for the referral of business incident to or part of a settlement service need not be written or verbalized but may be established by a practice, pattern or course of conduct. When a thing of value is received repeatedly and is connected in any way with the volume or value of the business referred, the receipt of the thing of value is evidence that it is made pursuant to an agreement or understanding for the referral of business." 12 C.F.R. § 1024.14(e) (C.F.P.B. RESPA Regulation X). Thus, plaintiffs need not provide direct evidence of an agreement, but may show that an agreement existed through circumstantial evidence.

Plaintiffs set forth their arguments and evidence tending to show an agreement through the expert report of Dawn Molitor-Gennrich. Dkt. No. 385-12 (Molitor Rep.). Ms. Molitor-Gennrich provides several examples of activities or circumstances that either provided WMB with the ability to influence appraisal values, encouraged EA to inflate appraisal values, or showed that EA and WMB knew that appraisals were being inflated.

Ms. Molitor-Gennrich's findings include:

• The WMB-EA agreement provided that EA would not warranty "reconsiderations of value" ("ROV"), and would only warranty an original appraisal. Ms. Molitor concluded that "this was a 'loophole' in the Warranty that allowed abuses by EA, contending that the Warranty did not cover any resulting changes of the ROV. The result was that EA was less concerned about their conduct when it came to ROVs, allowing this technique to be used
by WMB loan origination staff to pressure for higher appraised values, enabling more loans to close." Molitor Rep. at 19. The Warranty was in place by November 1, 2006. Id. at 15.



• WMB told EA appraisers loan amount and refinancing values for potential loans. Id. at 35-36. This may have biased the appraiser to conclude that the home was worth the amount suggested by the loan value.



• WMB applied "sales pressure" to EA appraisers and WMB was aware of "improper requests" from WMB sales staff to appraisers. Id. at 36. An April 6, 2006 memo documented some of these requests. Id.



• ROVs were used to secure higher values for appraisals, sometimes without proper documentation to support a higher value. Id. at 37.



• EA supervisors, known as "appraisal business managers" ("ABMs") would conduct "desk reviews" of appraiser work, even in geographic areas outside of their expertise or licensure. Id. at 40-41. The ABMs were sometimes former WMB employees. Dkt. No. 410-4 at 12.



• WMB used a "champion-challenger" model to pit EA and its competitor, LSI, against each other to win business. Molitor Rep. at 41. Under the champion-challenger model, the company that had fewer ROVs would get more business. In September of 2006, WMB reduced EA's share of appraisal business in certain regions because EA appraisers were not providing high enough appraisals. Id.; see also Merlo. Tr. at 152:24-153:17. The use of the champion-challenger suggests that WMB had a method for enforcing the alleged referral agreement, and in fact did enforce it by taking some business away from EA.



• In February 2007, EA agreed to use a Proven Appraiser List ("PAL") for all of WMB's appraisals. Id. at 42. The PAL list was implemented in April 2007.



• EA and WMB had meetings and discussed monthly quality control reports which sometimes showed that "almost 40% of the tech reviews produced unsupportable values." Id. at 43. Between February and June 2007, the monthly quality reports showed problems in over 10% of the reviews. Id. at 44; see also Kravec Decl. Ex. 11, Dillon Depo., at 23:22-24:13, 25:16-26:24, 40:16-42:19 and 83:9-25.

EA argues that this evidence is not sufficient to create a material issue of fact because plaintiffs have not obtained any testimony confirming the existence of an agreement to inflate. Dkt. No. 416-4 at 5-6. Although the court agrees that plaintiffs have not proved an agreement existed, they have presented evidence that would allow a reasonable fact finder to conclude that an agreement did exist. Plaintiffs are not required to present any testimony confirming an agreement existed, and EA is welcome to present testimony that an agreement did not exist in its case. See 12 C.F.R. § 1024.14(e) ("An agreement or understanding for the referral of business incident to or part of a settlement service need not be written or verbalized but may be established by a practice, pattern or course of conduct."); see also Henson v. Fidelity Nat. Financial Inc., Case No. 14-cv-01240-ODW (RZx), 2014 WL 1246222, at *7-8 (C.D. Cal. Mar. 21, 2014) (denying motion to dismiss based on circumstantial evidence of referral agreement, citing prior HUD regulations); Toldy v. Fifth Third Mortg. Co., 721 F. Supp. 2d 696, 704-05 (N.D. Ohio 2010) (same, denying defendant's motion for summary judgment); Walter v. Clarion Mortg. Capital, Case No. 08-cv-0536-W-GAF, 2009 WL 909594, at *10-11 (W.D. Mo. Apr. 2, 2009) (same, denying defendant's motion for summary judgment); Krupa v. Landsafe, Inc., 514 F.3d 1153, 1156 (11th Cir. 2008) (noting that a "threat that business being referred would be discontinued or diminished if the thing of value was not conveyed" could be circumstantial evidence of an illegal referral agreement, but finding no such evidence was presented). Plaintiffs' evidence here is consistent with an agreement, and tends to show how an agreement to inflate could have been carried out and enforced.

Finally, the evidence detailed above shows that an agreement may have existed from the inception of the EA-WMB relationship (i.e. the ROV warranty provision), prior to the PAL list, and therefore Spears' appraisal may fall within the scope of the agreement. Accordingly, the court DENIES the motion for summary judgment on this issue.

b. Evidence Tending to Show Inflation in the Aggregate or a "Thing of Value"

Plaintiffs have also presented evidence that would allow a fact finder to conclude that appraisals were inflated in the aggregate. Plaintiffs' two economics experts, Dr. French and Dr. Courchane, both concluded that EA appraisals were inflated in the aggregate. See Libeu Decl. Ex. 4, French Rep., Dkt. No. 385-12, ¶ 23; Libeu Decl. Ex. 9, French Supp. Rep., Dkt. No. 385-13, ¶ 13; Libeu Decl. Ex. 5, Courchane Rep., Dkt. No. 385-12, ¶¶ 71-75.

Dr. French drew a sample of 450 loans from five Metropolitan Statistics Areas (MSAs) and concluded that "nationwide at least 14.6 percent of EA appraisal values were inflated with virtually certain probability in all areas and . . . that nationwide at least 20.3 percent of EA appraisals were inflated either with a virtually certain probability or at least a strong probability in all areas." French Supp. Rep. at ¶ 13. In the MSAs, which account for 38% of the appraisals EA conducted for WMB, 38.2% were virtually certain to be inflated, and 53.1% had a strong or virtually certain probability of inflation. Id. at ¶ 12.

Dr. Courchane compared the appraisal values of EA to median home prices in the same zip code, to values from "retrospective automated valuation models" ("AVM"), and to the homeowners' own estimates of value. Courchane Rep.at ¶ 14. Dr. Courchane found that 61 % of properties had an appraised value above the median home price for homes sold in the same quarter and in the same zip code. Only 37.2 % of EA appraisals have values lower than the median. Id. at ¶ 14. Dr. Courchane also found that EA appraisals exceeded the AVM value 52.9 % of the time, and were below the AVM value 41.9 % of the time. The excess value was 9.2 % on average. Id. at ¶ 15. Dr. Courchane also found that EA appraised the property at "nearly exactly" the homeowners' or buyers' estimate 13.7 % of the time, which is "highly unlikely." Id. at ¶ 18.

EA complains that plaintiffs' experts did not use a control group to "isolate the impact of the EA-WaMu relationship on appraisal values." Dkt. No. 385-5 at 2. EA's rebuttal report, not surprisingly, showed no inflation. Id. at 3 (citing Owens Decl. Ex. 48, Dkt. No. 390-19, James Rep.). EA's expert Dr. James replicated Dr. Courchane's AVM analysis and found no inflation when compared with the rest of the market, but rather found that EA's appraisals were less inflated than the rest of the market. James Rep. at ¶¶ 33-39. Dr. James also criticized Dr. Courchane's median home price analysis, noting that WMB tended to target large borrowers, which would naturally lead to larger loans. Id. at ¶¶48-49. As to Dr. French, Dr. James found numerous flaws in Dr. French's statistical analysis, including lack of a control group. Id. at ¶¶63-102.

The control group argument is not necessarily valid. Plaintiffs' experts compared the EA appraisal to a new, retrospective appraisal from the standpoint of a person at the time of the original appraisal. Therefore, to the extent that the entire market was inflated, this inflation would have been reflected in both appraisals, and any measureable difference would be attributable to the alleged agreement.

The court finds that the evidence on whether appraisals were inflated in the aggregate boils down to a classic battle of the experts. Each side has presented its own analysis on the issue, and reached different conclusions. None of the experts are patently unreasonable, and none are conclusive beyond a doubt. Summary judgment is therefore improper. Anderson, 477 U.S. at 248; Celotex, 477 U.S. at 322-32.

In addition to attacking plaintiffs' evidence of inflation in the aggregate, EA also argues that because WMB did not securitize all of the loans related to an EA appraisal, WMB did not receive a "thing of value." However, WMB did securitize two-thirds of the loans, which is sufficient for a fact finder to determine that the inflated appraisals were a thing of value. Dkt. No. 410-4 at 8. As plaintiffs argue, "if at the time of an appraisal WMB knew that there was a 70% chance that the loan would be securitized, a jury could conclude that incentivized WMB to issue that loan, even if the property sold only because an inflated appraisal was provided. An inflated appraisal for a property that is 70% likely to be resold is a thing of value just as would be an inflated appraisal for a property certain to be securitized." Id.

Because plaintiffs have presented evidence that would allow a jury to make a reasonable inference that EA provided inflated appraisals to WMB, the court DENIES the motion for summary judgment on this issue.

4. Plaintiffs Have Presented Evidence that Class Members' Loans are Covered by RESPA

EA also argues that plaintiffs have not presented evidence that class members' loans are covered by RESPA. The RESPA issue is discussed in detail in the motion to decertify, Part II(C)(3), infra. As relevant to the motion for summary judgment, plaintiffs have presented sufficient evidence to create a material dispute of fact over whether class members' loans are covered by RESPA to justify denying summary judgment.

5. EA's Request for Partial Summary Judgment

EA also requests partial summary judgment on three issues. Dkt. No. 385-5 at 24-25. As discussed above, plaintiffs have presented evidence from which an agreement prior to the PAL list could be inferred, and therefore EA's first request for summary judgment "as to any class member with an EA service performed prior to the April 13, 2007 implementation of PAL" is DENIED. EA's requests for summary judgment "as to all potential class members who received a loan from WaMu's wholesale channel" and potential class members who "did not receive a funded loan" are unopposed and GRANTED, although the court notes that these loans were never a part of the class definition. Dkt. No. 410-4 at 1, n.1.

B. Plaintiffs' Motion for Summary Judgment on Affirmative Defenses

Plaintiffs' move for summary judgment on defendant's four affirmative defenses raised in the First Amended Answer. Dkt. No. 383 (Pltf. MSJ); Dkt. No. 310 (FAA). Defendant filed an opposition, Dkt. No. 404, and plaintiffs filed a reply, Dkt. No. 423. For the reasons stated below, the court GRANTS plaintiffs' motion as to EA's first, second, and fourth affirmative defenses and DENIES the motion as to the third affirmative defense.

1. First Affirmative Defense: EA Acted in Good Faith

Defendant's first affirmative defense states:

Plaintiff's claims and/or the claims of any members of the class whom Plaintiffs purport to represent are barred, in whole or in part, because EA acted at all times in good faith and without knowledge or intent to violate RESPA, Section 8(a), and did not directly or indirectly participate in, or induce, any unlawful acts by others.

a. RESPA Does Not Include an Intent, Knowledge, or Bad Faith Requirement

Plaintiffs are correct that RESPA § 8(a) does not include any express intent requirement. Section 8(a) only requires an "agreement or understanding." 12 U.S.C. § 2607(a). Several courts have concluded that RESPA § 8(a) does not include an intent component and that defendants may be liable for even inadvertent violations. See Dkt. No. 383 at 6-7, and cases cited therein. This reasoning is based at least in part on the "safe harbor" provision in § 2607(d)(3), which provides that "[n]o person or persons shall be liable for a violation of the provisions of subsection (c)(4)(A) of this section if such person or persons proves by a preponderance of the evidence that such violation was not intentional and resulted from a bona fide error notwithstanding maintenance of procedures that are reasonably adapted to avoid such error." Subsection (c)(4)(A), not at issue here, deals with "affiliated business arrangements." The inclusion of a specific safe harbor provision for subsection (c)(4)(4), but not subsection (a), suggests that Congress did not include any implied (and obviously not any explicit) intent requirement for subsection (a) violations.

Defendants argue that because plaintiffs' theory of liability is based on a civil conspiracy between EA and WMB, they must prove "a unity of purpose or a common design or understanding, or a meeting of the minds in an unlawful arrangement." Dkt. No. 404 at 3, quoting Transgo, Inc. v. AJAC Transmission Parts Corp., 768 F.2d 1001, 1020 (9th Cir. 1985). Defendants thus argue that plaintiffs must produce evidence that EA "knew that a wrongful act was planned and intended to aid in its commission." Dkt. No. 404 at 4.

§ 8(a) of RESPA does not require a showing that EA knew that it was engaging in illegal activity in order to establish liability for entering into an agreement to inflate appraisals in exchange for business referrals. The statutory language of § 8(a) only requires an agreement or understanding and does not include any "bona fide error" or other good faith defense to an alleged violation. Therefore, the court DISMISSES EA's first affirmative defense based upon good faith or the lack of knowledge of illegality. However, as a practical matter in this case, the dismissing of the defense is of no apparent consequence. An agreement to inflate appraisals in exchange for business referrals could not have been made in good faith. EA's argument is that no agreement or understanding existed between it and WMB to inflate appraisals. EA can argue that it had legitimate business reasons to engage in the various practices that plaintiffs suggest support the alleged agreement to inflate.

b. The Second Portion of EA's Affirmative Defense Is an Improper Negative Defense

The portion of EA's defense that reads EA "did not directly or indirectly participate in, or induce, any unlawful acts by others" is an improper negative defense. As the court explained in the JOP Order, Dkt. No. 309 at 11:

"A defense which demonstrates that plaintiff has not met its burden of proof as to an element plaintiff is required to prove is not an affirmative defense." Zivkovic v. S. California Edison Co., 302 F.3d 1080, 1088 (9th Cir. 2002). This court has previously stricken defenses that are not actually affirmative defenses where the defenses "allege defects in plaintiff's claims and raise issues that are plaintiff's burden to prove." Joe Hand Promotions, Inc. v. Nguyen, 2012 WL 1183738 (N.D. Cal. Apr. 6, 2012) (Whyte, J.).

Here, the second portion of EA's defense is simply a general statement that EA did not violate any laws. As this is not a defense on which defendants have the burden of proof, Barnes v. AT &T Pension Ben. Plan-Nonbargained Program, 718 F. Supp. 2d 1167, 1174 (N.D. Cal. 2010), it is STRICKEN from the FAA. As previously noted, EA is free to argue "at trial that plaintiffs' claims are deficient" for failing to adequately show any element of their case, including an agreement. JOP at 11.

2. Second Affirmative Defense: Safe Harbor Under § 2607(c)(2)

EA's second affirmative defense is that plaintiffs' claim is barred by 12 U.S.C. § 2607(c)(2), which states that "nothing in this section shall be construed as prohibiting . . . (2) the payment to any person of a bona fide salary or compensation or other payment for goods or facilities actually furnished or for services actually performed."

Plaintiffs argue that the court has already ruled that § 8(c)(2) does not apply to their claim. Dkt. No. 383 at 11, citing March 9, 2009 Order, Dkt. No. 147 at 5; August 30, 2009 Order, Dkt. no. 169 at 6; January 8, 2010 Order, Dkt. No. 182, at 5; July 2, 2010 Order, Dkt. No. 209, at 8-9.

In the prior orders cited by plaintiffs, the court distinguished between the payment of the appraisal fee by the borrower and the exchange of the inflated appraisal for business referrals. The plaintiffs are not alleging that charging borrowers a fee for an inflated appraisal violated RESPA, but rather are arguing that exchanging inflated appraisals for business referrals violated RESPA § 8(a). Thus, "this inflation of appraisals was not payment for goods or services actually rendered but rather was payment for business referrals." Dkt. No. 182 at 5. Because this was not payment for goods or services actually performed, the safe harbor provision does not apply. Defendant argues that the intervening case of Martinez v. Wells Fargo Home Mortgage, Inc., 598 F.3d 549 (9th Cir. 2010) casts doubt on the court's prior analysis. Martinez involved an alleged violation of RESPA § 8(b) which prohibits "splitting charges." Martinez is not applicable here. The Ninth Circuit held that § 8(b) only applies to "the practice of giving or accepting money where no service whatsoever is performed in exchange for that money" and not to "charging fees, excessive or otherwise, when those fees are for services that were actually performed." Id. at 55354. Therefore, courts will not split a fee into permissible and excessive components, and find defendants liable for only the "excessive" or "unreasonable" portion. Instead, defendants have either charged some amount for a service performed, and cannot be liable under § 8(b), or they have charged for a service never performed, and are liable for the full amount of the charge. See Martinez, 598 F3.d at 554; Kruse v. Wells Fargo Home Mortgage, Inc., 383 F.3d 49, 56 (5th Cir. 2005); Santiago v. GMAC Mortgage Grp., Inc., 417 F.3d 384, 387 (3d Cir. 2005).

EA argues that the interpretation of § 8(b) should inform the interpretation of § (c)(2) because both deal with "payments for 'services actually performed.'" Dkt. No. 404 at 7 n.4, citing § 2607. EA thus concludes that the court cannot divide the appraisal service into two components: a service for payment of a fee and a service in exchange for referrals.

This is not persuasive because EA is actually receiving two separate "payments" related to the appraisals: payment by the borrower for the appraisal, and payment from WMB in the form of referrals for the inflation. Dkt. No. 182 at 5. Thus, the appraisal fee is not split or divided into permissible and impermissible parts, which would be prohibited by Martinez, but rather the practice of receiving two separate payments—one a permissible fee for the service provided, and one an impermissible business referral arrangement—is at issue here. See also Dkt. No. 423 at 8.

EA also argues that the safe harbor provision will apply if plaintiffs' fail to prove a conspiracy or agreement to inflate appraisals, i.e. if plaintiffs fail to prove their affirmative case,. Dkt. No. 404 at 8. The court fails to see how the safe harbor provision would apply to behavior that is not a RESPA violation. Accordingly, because the safe harbor provision does not apply to plaintiffs' claim that EA provided inflated appraisals to WMB in exchange for referrals, the court GRANTS plaintiffs' motion summary judgment on EA's second affirmative defense.

3. Third Affirmative Defense: Statute of Limitations

EA's third affirmative defense is based on the one-year statute of limitations for violations of RESPA. 12 U.S.C. § 2614. The court has already held that RESPA's statute of limitations is not jurisdictional and may be equitably tolled. JOP, Dkt. No. 309, at 5. In the prior JOP Order, the court allowed plaintiffs to proceed on the theory that they could not have known about EA and WMB's referral arrangement until the release of the New York Attorney General's report ("NYAG Report"). Id. at 7. Specifically, the court was persuaded that "[p]laintiffs, unlike the New York Attorney General, are not a prosecutorial body, and thus could not have obtained access to the internal EA documents necessary to discover the existence of a claim." Id. The court concluded that "equitable tolling applies until the release of the NYAG report" for the purposes of a motion to dismiss, and denied EA's motion to dismiss without prejudice. Id. The NYAG Report was released November 1, 2007, and plaintiffs filed their complaint on February 8, 2008. Id.

EA argues that plaintiff Scholl was on notice that EA's appraisals were questionable before the NYAG report. Scholl testified at her deposition that she was "suspicious" of a "process problem" related to WMB appraisals around June 2007. Dkt. No. 385-4 (Scholl Dec. 1 2009 Depo.) at 97:15-21. However, she did not suspect "anything specifically with the 194 Terrace property" at that time. Id. at 97:22-23. Scholl's suspicions appear to be based on her experiences in trying to sell properties that had been appraised in the past, and discovering that she "had overpaid for these properties." Id. at 98:6-17. Scholl then concluded that the appraisals must not have been accurate. Id. at 98:21-23. However, none of those appraisals were done by EA. Id. at 99:4-6. Scholl's suspicions about her "problem" appraisals and the "reliability and unprofessionalism of the appraisals" led her to research the issue, and she "found" the NYAG Report. Id. at 99:6-12.

EA argues that Scholl's deposition testimony creates a material issue of fact as to whether she was on notice of a potential claim in June 2007, when she first suspected problems with WMB-related appraisals. The court does not find this persuasive because Scholl did not suspect anything related to EA, the only defendant in this case, until the NYAG Report.

The burden is on plaintiffs to show entitlement to tolling. Although plaintiffs have presented evidence that facts were not available to put plaintiffs on inquiry notice before November 1, 2007, the court is reluctant to grant summary judgment in their favor on this issue at this stage. EA may still present evidence showing that some individual plaintiff is not entitled to equitable tolling. EA's evidence regarding now-dismissed plaintiff Scholl is not sufficient to negate tolling, and EA has not suggested that it has any other relevant evidence to negate tolling. Accordingly, the court DENIES plaintiffs' motion for summary judgment on EA's third affirmative defense, but does not anticipate that equitable tolling will create a significant individualized issue.

4. Fourth Affirmative Defense: Unjust Enrichment

EA's fourth affirmative defense reads:

EA did not enter into a conspiratorial agreement to inflate appraisal values. Each class member who obtained a loan covered by RESPA in connection with an appraisal service provided by EA (and paid for that appraisal service) received the appraisal service for which the class member paid. Plaintiffs' claims and/or the claims of any members of the class whom Plaintiffs purport to represent are therefore barred because any class member who obtains a recovery of the amount he or she paid for an appraisal service, or any portion of that payment, would obtain a windfall and be unjustly enriched.

Neither party presented any case law addressing unjust enrichment as a defense to a RESPA violation, and the court did not discover any on its own. EA argues that plaintiffs will be unjustly enriched by return of their appraisal fee because (1) plaintiffs may have shared the cost of the fee with others (i.e. Scholl may have shared the fee with her investment partners), and (2) some borrowers committed mortgage fraud in connection with their loans. The court does not find these issues persuasive because Scholl is no longer a part of the class, and the borrowers that committed mortgage fraud are likely not a part of the class because they do not have RESPA-qualifying personal loans.

Plaintiffs argue that unjust enrichment is not a defense to a RESPA action. Because RESPA has a statutory damages provision, § 2607(d)(2), plaintiffs argue that recovery of damages is mandatory. Furthermore, it is not clear how any plaintiff benefited from paying for an inflated appraisal.

Finally, the majority of the Fourth Affirmative Defense is simply a denial of liability. FAA at 12. As explained above with respect to the First Affirmative Defense, the court strikes the improper negative defense. Here, because there is no basis for defendant to assert an unjust enrichment defense and the defense is primarily negative, the court GRANTS plaintiffs' motion as to the Fourth Affirmative Defense.

C. Defendant's Motion to Decertify

Defendant argues that the class no longer meets the ascertainability requirement of Rule 23(a) or the predominance requirement of Rule 23(b)(3) because of individualized issues in determining class membership. For the reasons explained below, the court DENIES the motion to decertify.

While the court reaffirms the definition of the Class, the court modifies the description of its members to make it more explicit and reduce the chance of a misunderstanding of who is a class member. The court now defines the class as composed of:

All consumers throughout the United States who, on or after June 1, 2006, received home loans for personal, as opposed to business or commercial purposes, originated by Washington Mutual Bank, F.A., utilizing appraisals that they paid for and obtained from defendant eAppraiseIT. Excluded from the Class are employees, officers, and directors of defendant and their subsidiaries and affiliates.

1. Standard on a Motion to Decertify

Plaintiffs retain the burden of demonstrating that class certification is warranted. Marlo v. United Parcel Serv., Inc., 639 F.3d 942, 947 (9th Cir. 2011). Federal Rule of Civil Procedure 23(a) lists four conjunctive criteria that must be met to certify a class action: numerosity, commonality of issues, typicality of the representative plaintiffs' claims, and adequacy of representation. Fed. R. Civ. P. 23(a). A class may only be certified if the court is "satisfied, after a rigorous analysis, that the prerequisites of Rule 23(a) have been satisfied." Gen. Tel. Co. of the Southwest. v. Falcon, 457 U.S. 147, 161 (1982). In addition to fulfilling the four prerequisites of Rule 23(a), a class action must also meet one of the disjunctive requirements of Rule 23(b) by satisfying the criteria set forth in at least one of the three types of class actions. Fed. R. Civ. P. 23. The court previously certified this class action under Rule 23(b)(3), which requires that "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."

2. Waiver Arguments

Plaintiffs argue that EA waived its arguments related to loan funding, fee payment, and RESPA purpose by failing to raise them in its oppositions to the plaintiffs' earlier motions for class certification. See Dkt. Nos. 201, 231 (EA opp'ns to class cert.). This is not persuasive, because EA did raise at least the appraisal fee argument. Dkt. No. 231 at 24-25. At the time, plaintiffs responded to EA's concern by stating that the information would be readily available from WMB. Dkt. No. 238 at 15. Now, EA is renewing its concerns, and adding other concerns, because the Chase/WMB data has not proven as useful or readily available as plaintiffs believed it would be at the class certification stage. A court is free to revisit its class certification order based on subsequent litigation developments. Fed. R. Civ. P. 23(c)(1)(c); Gen. Tel. Co. of the Southwest v. Falcon, 457 U.S. 147, 160 (1982); Rodriguez v. West Publishing Corp., 563 F.3d 948, 966 (9th Cir.2009).

3. Plaintiffs Have Proposed an "Administratively Manageable Method" of Determining Class Eligibility and Individual Issues Do Not Predominate

The court discusses the overlapping issues of ascertainability and predominance together. Despite plaintiffs being unable to provide the court with a list of class members, as defendant would prefer, the class-wide issue central to liability—an agreement to exchange referrals for inflated appraisals resulting in inflation in the aggregate—still predominates.

As the court has explained,

the overarching inquiry [on agreement] still relates to the course of conduct between EA and WMB. A factfinder will likely have to determine not only whether there was a conspiracy but, if so, the scope of that conspiracy. But these are still common questions, with a single answer that pertains to the entire class, to be determined from common proof. Once the scope of the conspiracy - in time or as to other parameters - is determined, the individual question of whether an appraisal falls within that scope will likely be easy to answer. Moreover, even if plaintiffs' claims were tried on an individual basis, the entirety of EA's relationship with WMB would still be relevant to whether an agreement existed at a particular point in time. Thus, to prove only their own claims, plaintiffs could and might still introduce a large portion of the evidence that would be needed to try the case as a class action.

As the court's prior order indicates, this is the type of case that calls for class treatment. The central liability issue— whether EA appraisals were inflated in the aggregate as a result of an agreement between WMB and EA—is a complex, expert-dominated inquiry. That evidence would be needed to prove any individual case, and is equally applicable to each class member. See also Fed. R. Civ. Pro. 23(b)(3) Advisory Committee's Note ("a fraud perpetrated on numerous persons by the use of similar misrepresentations may be an appealing situation for a class action, and it may remain so despite the need, if liability is found, for separate determination of the damages suffered by individuals within the class."). The cost of putting on such a case is likely to deter an individual plaintiff, whose recovery is capped at around $1,000 (three times the appraisal fee, with an average appraisal fee of $334 dollars). Even though a plaintiff may recover attorney's fees if he or she prevails, RESPA § 8(d), the intervening expense of litigation and uncertainty of outcome are significant deterrents to individual actions. In addition, what the court has characterized as the "liability issue" is substantially more complex than the several eligibility issues. Determining eligibility—and thus ascertaining the class—is likely to be answerable through straightforward inquiries.

A class is ascertainable if the class is defined with "objective criteria" and if it is "administratively feasible to determine whether a particular individual is a member of the class." See Wolph v. Acer America Corp., No. 09-1314, 2012 WL 993531, at *1-2 (N.D. Cal. Mar. 23, 2012) (certifying a class where "the identity and contact information for a significant portion of these individuals can be obtained from the warranty registration information and through Acer's customer service databases"); see also Hofstetter v. Chase Home Finance, LLC, No. 10-01313, 2011 WL 1225900, at *14 (N.D. Cal. Mar. 31, 2011) (certifying class where "defendants' business records should be sufficient to determine the class membership status of any given individual"); Xavier v. Philip Morris USA Inc., 787 F. Supp. 2d 1075, 1089 (N.D. Cal. 2011) (denying the ascertainability of a class that smoked cigarettes for "at least twenty years"); Tietsworth v. Sears, Roebuck & Co., No. 09-288, 2013 WL 1303100, at *3-4 (N.D. Cal. Mar. 28, 2013) (denying certification where "ascertaining class membership would require unmanageable individualized inquiry").

Therefore, to meet the ascertainability requirement, plaintiffs must propose a reliable, manageable method of determining class membership by answering (1) whether the loan was originated and funded by WMB; (2) whether the borrower paid the appraisal fee; (3) the amount of the appraisal fee; and (4) whether theloan is a RESPA-qualifying loan.

Plaintiffs have proposed that they can answer question (1) through information provided by Chase in the December spreadsheets. See Libeu Decl. Ex. 9, French Supp. Rep., Dkt. No. 385-13, at ¶ 27 (MTM_SENT_TO_ML origination methodology). Plaintiffs have proposed to answer question (2) through the use of class members' HUD-1 forms, or, alternatively, through verified claims from class claimants averring that they paid the appraisal fee. Question (3) can also be answered from the HUD-1 or a class member verification form, and possibly from the Chase spreadsheets.

Defendant primarily argues that determining which loans were subject to RESPA, question (4), will swallow the litigation. As defendant notes, a few other courts, primarily district courts within the Sixth Circuit, have denied class certification on this basis. See Henson v. Fid. Nat. Fin. Inc., 2:14-CV-01240-ODW, 2014 WL 2765136 (C.D. Cal. June 18, 2014); Toldy v. Fifth Third Mortgage Co., 1:09 CV 377, 2011 WL 4634154 (N.D. Ohio Sept. 30, 2011); Powers v. Fifth Third Mortgage Co., 1:09-CV-2059, 2011 WL 3811129 (N.D. Ohio Aug. 12, 2011) report and recommendation adopted, 1:09 CV 2059, 2011 WL 3812634 (N.D. Ohio Aug. 29, 2011); but see Edwards v. First American Corp., 289 F.R.D. 296, 305 (C.D. Cal. 2012).

In Henson, a case from within the Ninth Circuit, the RESPA qualification issue was only one of several reasons for denying class certification. 2014 WL 2765136. Other reasons included delay in filing the motion for certification more than eight months, the necessity of individualized tolling inquiries extending back 15 years, and the unlikelihood of finding records extending through the class period. Here, the only difficult individual question to be answered is RESPA status, and the court envisions the claim form being a first (and perhaps dispositive) step in that process. The court will not be asking lay class members to determine whether their loan invoked a RESPA exemption, the claim form will ask putative class members to disclose information so that the parties or the court can make that determination.

In contrast to Henson, in Edwards v. First American Corp., 289 F.R.D. 296, 305 (C.D. Cal. 2012), the court declined to decertify a RESPA class and concluded that "[i]f necessary, the purpose of class members' loans may be determined via a claim form sent to class members asking them 'a single question to determine whether they are entitled to relief.'" (citation omitted).

After reviewing the cases and the parties' arguments, the court finds Edwards more persuasive than Powers, Toldy, and Henson. Here, similar to Edwards, it appears that simply asking a claimant to disclose, under penalty of perjury, the intended use for his or her loan funds will suffice in most cases to answer the RESPA-purpose question. The court has already determined that superiority weighs in favor of class treatment, and this circuit has found that class actions may be a superior method of adjudicating RESPA claims. See, e.g., Edwards v. First American Corp., 385 Fed. App'x. 629 (9th Cir. 2010) (non-precedential).

The court also finds persuasive plaintiffs' argument that EA is essentially taking the position that no RESPA actions should be certified, as individual issues would always predominate. Dkt. No. 402-4 at 4. This is not the law, as every class action requires identification of class members, and most require individual proof of loss. See Yokoyama v. Midland Nat'l Life Ins. Co., 594 F.3d 1087, 1094 (9th Cir. 2010) ("[T]he amount of damages is invariably an individual question and does not defeat class action treatment." (internal quotation marks omitted)).

A verified claim form and review process can be used to determine class member eligibility. Although defendant needs to be given the opportunity to challenge eligibility, and if request before a jury, the court does not foresee that such challenges would be frequent and predominate over the fundamental liability issue of whether there was inflation of appraisals in the aggregate caused by an agreement between WMB and EA to inflate appraisals in exchange for referrals of business.

4. Equitable Tolling Does Not Require Decertification

EA argues that equitable tolling requires an individualized inquiry into each class member's circumstances (for those class members who obtained a loan on or before February 8, 2007). The court does not agree. As previously noted, "[p]laintiffs, unlike the New York Attorney General, are not a prosecutorial body, and thus could not have obtained access to the internal EA documents necessary to discover the existence of a claim." Dkt. No. 309 at 7. The court does not anticipate that equitable tolling will create an issue that predominates over the liability issue, as the evidence to date supports tolling for a reason that applies to all plaintiffs who filed after the statute of limitations would have run absent tolling. Although plaintiffs have the burden of showing that the statute of limitations should be tolled, defendant has offered no evidence that there was anything to put plaintiffs on reasonable inquiry notice before information about the New York Attorney General's investigation became publically known. To the extent that defendant wishes to challenge some putative class member's eligibility for equitable tolling, defendant may raise that challenge after there is a determination, if any, of an agreement to inflate appraisals in exchange for the referral of business. See also Part II(B)(3), supra.

5. Spears is an Adequate Class Representative

Because the court has granted summary judgment as to Scholl, Part II(A)(2), supra, the court revisits whether Spears is an adequate class representative. The court earlier expressed concern at Spears' lack of familiarity with the case. Order Denying Class Cert. (Dkt. No. 209) at 5:25-26. There does not appear to be any new evidence that Spears is unfamiliar with the case, and he attended, but did not participate in, the hearing on these motions. EA also argues that Spears cannot represent class members subject to equitable tolling. This is not persuasive because the tolling issues can be addressed, if necessary, after the predominate liability issue is answered. Further, Spears' representation of plaintiffs who do require tolling to make their claims timely does not conflict with his duty to represent other plaintiffs. See Ruppert v. Alliant Energy Cash Balance Pension Plan, 255 F.R.D. 628, 634 (W.D. Wis. 2009).

D. Plaintiffs' Motion to Bifurcate

Plaintiffs move to bifurcate the trial to have a single trial on liability and aggregate damages, and then have a special master or claims administrator distribute any aggregate award. Dkt. No. 392. Plaintiffs propose to show aggregate damages by paring down the number of potential class members based on the Chase spreadsheets and through extrapolation from the 450 sample loans. See French Supp. Rep. Table 3. In the alternative, plaintiffs move to bifurcate liability from damages entirely. Because the practice of extrapolating damages is questionable, the court bifurcates the common liability issue from any damages calculations.

1. Legal Standard

A district court has "broad discretion" in deciding whether to order separate trials under Rule 42(b). Zivkovic v. S. Cal. Edison Co., 302 F.3d 1080, 1088 (9th Cir. 2002)."Factors to be considered when determining whether to bifurcate a trial include: avoiding prejudice, separability of the issues, convenience, judicial economy, and reducing risk of confusion." Bates v. United Parcel Service, 204 F.R.D. 440, 448 (N.D. Cal. 2001).

In this case, the common issue of whether there was inflation of appraisals on an aggregate basis as a result of an agreement between WMB and EA is separable, and bifurcation of it from the eligibility and damages questions will promote judicial economy, convenience, and avoid prejudice to EA.

2. The Issue of Whether There Was Inflation of Appraisals on an Aggregate Basis as a Result of an Agreement Between WMB and EA is Separable From Damages

As discussed with regard to the motion to decertify, plaintiffs' proof of inflation of appraisals on an aggregate basis as a result of an agreement between WMB and EA must be through common evidence and predominates over the individual issues. This proof does not require any individualized inquiries and applies to all class members. See also Dkt. No. 249 (Cert. Order) at 5. EA agrees, and stated at the hearing that "defendant would never contend that the core issue of conspiracy is not something that makes sense to try first, because it could completely eradicate the case." Dkt. No. 429, Hearing Tr. at 75:12-15.

The remaining questions are individual—receipt of a WMB loan, RESPA status, payment of the appraisal fee, and the amount of the appraisal fee. A fact finder can answer the common question without delving into any individual questions, and vice versa. Although plaintiffs must secure favorable answers on both the individual and common questions to recover, the facts and evidence needed are separable.

3. Bifurcation Will Promote Judicial Economy and Convenience, and Avoid Extrapolation of Damages and Prejudice to EA

Bifurcation of the common question from the individual questions appears to be the best solution to preserving EA's "right to litigate individual defenses to RESPA," Dkt. No. 406-3 at 8, and avoid the extrapolation of damages.

It is clear that EA must have the opportunity to challenge a putative class members' RESPA status, but, as described above, the court does not find that the RESPA inquiry will predominate over the common question. It is less clear whether plaintiffs could rely on extrapolation to determine aggregate damages if aggregate damages were made part of the initial liability phase.

"Because the Rules Enabling Act forbids interpreting Rule 23 to abridge, enlarge or modify any substantive right, a class cannot be certified on the premise that [the defendant] will not be entitled to litigate its statutory defenses to individual claims." Wal-Mart Stores, Inc. v. Dukes, 131 S. Ct. 2541, 2561 (2011) (internal citation and quotation marks omitted).

As explained in United States v. City of New York,

Wal-Mart's rejection of 'Trial by Formula' means that the underlying substantive law determines whether individual proceedings are required; a litigant may not convert an individual question into a common question by concocting a method of classwide proof that subverts rights created by the underlying substantive law. When determining whether a question is common to the class, the court must look to the underlying substantive law to determine whether the proposed method of classwide proof prevents the party opposing class certification from asserting its substantive rights.

EA argues that to award damages in the aggregate would deprive it of its substantive rights, because RESPA status must be determined on a case-by-case basis. Dkt. No. 406-3 at 10, citing Martin v. Litton Loan Servicing LP, 2014 WL 977507, at *8 (E.D. Cal. Mar. 12, 2014) and Thorns v. Sundance Properties, 726 F.2d 1417, 1419 (9th Cir. 1984). Plaintiffs argue that because they are using common evidence to show RESPA status (i.e., the Chase spreadsheets) and loan payment (i.e. the extrapolation), EA would have the opportunity to contest the aggregate damages claimed without the need for individual inquiry.

As noted by Judge Alsup prior to Dukes, "[t]he extent to which fluid recovery [or extrapolation] can be obtained in a litigated class action on an aggregate basis rather than by tallying up individual claims (even if on a formulaic basis using a computerized database) has been a point of contention for decades." Gutierrez v. Wells Fargo & Co., C 07-05923 WHA, 2009 WL 1247040 *3 (N.D. Cal. May 5, 2009) (emphasis in original). More recently, the Ninth Circuit stated "[s]ince Dukes and Comcast were issued, circuit courts including this one have consistently held that statistical sampling and representative testimony are acceptable ways to determine liability so long as the use of these techniques is not expanded into the realm of damages." Jimenez v. Allstate Ins. Co., No. 12-56112, 2014 WL 4338841 (9th Cir. Sept. 3, 2014). It thus appears that using the 450 sample loans to extrapolate or estimate the total amount of damages the class is entitled to may run afoul of applicable precedent. Separating the common question from the calculation of damages will avoid this potential problem and will force a "tallying up of individual claims." Gutierrez at *3.

In addition to avoiding prejudice to EA, bifurcation promotes judicial economy. Trying the individual damages questions in a single trial with the common question would be potentially confusing, as the question of an agreement to inflate is different from the question of eligibility. Separating the individual questions will promote judicial economy and convenience, as it is unlikely that there will be numerous "mini-trials" on specific individual class members. It still appears that most class members can be identified through use of a claim form. And, to the extent necessary, EA will have the opportunity to challenge individual class members on each of the individual questions. Accordingly, the motion to bifurcate liability from damages is GRANTED, because of the predominance of the common issue of whether there was an agreement to inflate appraisals in exchange for the referral of business over the individualized issues of damages. Further, if EA prevails on the bifurcated issue there will be no need to determine the individual issues.

4. Plaintiffs May Use the HUD-1 Forms in the Claims Period

Plaintiff proposes using HUD-1 forms obtained from Chase after liability has been determined, or requesting the HUD-1 form from the putative class members to be submitted with their verified claims. Dkt. No. 392 at 5 n.4. EA objects because plaintiffs did not secure the HUD-1 forms before the close of discovery. The court will allow plaintiffs to use the HUD-1 forms from both sources during the damages phase. From the papers, the best the court can conclude is that plaintiffs and Chase attempted to use discovery other than the expensive gathering and copying of the HUD-1 forms, but that plan ultimately did not work. See Dkt. No. 359. It appears that plaintiffs and Chase worked throughout the discovery period to find a workable solution, and it was only at the last moment before discovery closed that plaintiffs were more or less forced to move to compel the production of the HUD-1 forms. Id. at 2. Plaintiffs did seek to compel the production of the Chase HUD-1 forms before the close of discovery when it became clear that they were not going to obtain the forms voluntarily. Id. The court previously warned when it extended the deadline for discovery of the forms from Chase until December 22, 2013 that "Spears will have to make his case with whatever information he can obtain by December 22, 2013." Dkt. No. 350 at 2. However, since the case has now been bifurcated and plaintiffs appear to have acted in good faith and the magistrate judge ordered production of the forms, the court will allow plaintiffs to obtain and use the late acquired HUD-1 forms in the damages phase.

E. Defendant's Motion to Strike

Pursuant to the court's scheduling order, Dkt. No. 350, plaintiffs served their opening expert reports on January 31, 2014. EA chose not to serve any opening reports. Plaintiffs then served "supplemental" expert reports on May 14, 2014, the date listed in the scheduling order for plaintiffs' "Supplemental and Rebuttal Expert Reports." EA then served their Rebuttal Expert Reports on June 6, 2014, in accordance with the scheduling order.

EA now moves to strike all or parts of plaintiffs' supplemental reports as allegedly including materials that were available when plaintiffs served their opening reports. Dkt. No. 388. EA argues that including material previously available at the time of the opening reports is not a proper basis for supplementation.

Plaintiffs argue that their supplemental reports were served on the deadline specified in the court's scheduling order, and therefore are timely regardless of their content. Moreover, plaintiffs point out that some of the material in the supplemental reports is based on Chase spreadsheets that plaintiffs could not decode until May 3, 2014, only eleven days prior to the supplementation deadline. Plaintiffs also argue that EA is not prejudiced by the supplemental reports because EA still had an additional three weeks to review plaintiffs' supplemental reports before serving their own rebuttals.

Overall, the court agrees with EA that the supplemental reports were improper because they included materials available at the time of the opening expert reports. The court's scheduling order did not intend to provide plaintiffs with two dates to serve opening expert reports. Defendant's strategic choice not to serve opening reports did not entitle plaintiffs with the chance to bolster their expert case through supplemental reports containing analyses that should have been produced earlier. However, EA has not shown that it was prejudiced by all of the supplementation, as discussed in more detail below. The court therefore GRANTS IN PART AND DENIES IN PART defendant's motion to strike.

1. Legal Standard for Supplementation

Federal Rule of Civil Procedure 26(e) provides that a party's expert reports must be supplemented "if the party learns that in some material respect the disclosure or response is incomplete or incorrect, and if the additional or corrective information has not otherwise been made known to the other parties during the discovery process or in writing." Fed. R. Civ. P. 26(e). "Supplementation under the Rules means correcting inaccuracies, or filling the interstices of an incomplete report based on information that was not available at the time of the initial disclosure." Keener v. United States, 181 F.R.D. 639, 640 (D. Mont. 1998).

"If a party fails to provide information or identify a witness as required by Rule [26(e)], the party is not allowed to use that information or witness to supply evidence on a motion, at a hearing, or at a trial, unless the failure was substantially justified or is harmless." Fed. R. Civ. P. 37(c)(1).

The factors for the court to consider when determining whether a violation of the expert discovery rules was harmless, include: (1) prejudice or surprise to the party against whom the evidence is offered; (2) the ability of that party to cure the prejudice; (3) the likelihood of disruption of the trial; and (4) bad faith or willfulness involved in not disclosing the evidence. See David v. Caterpillar, Inc., 324 F.3d 851, 857 (7th Cir. 2003); Lanard Toys, Ltd. v. Novelty, Inc., 375 Fed. Appx. 705, 713 (9th Cir. 2010); Transbay Auto Service, Inc. v. Chevron U.S.A., Inc., No. 09-cv-04932 SI, 2010 WL 4591596 (N.D. Cal. Nov. 3, 2010).

Plaintiffs do not seriously argue that the inclusion of previously available information into the supplemental reports was substantially justified. EA does not seriously argue any bad faith or willfulness in not disclosing the supplemental opinions earlier.

2. Retrospective Appraisal Review Reports

After plaintiffs' supplemental report deadline passed, the plaintiffs produced five retrospective appraisal review reports on properties that were not previously disclosed. The five reports were all complete before the January 31, 2014 deadline for opening reports. Plaintiffs' respond that the appraisal reviews were sent to EA "in response to a larger discovery request from Defendants for data underlying each Expert's reports" and "[t]he data from those appraisal reviews had already been incorporated into Dr. French's and Frank Gregoire's Supplemental Reports which was [sic] tendered in a timely manner." Dkt. No. 409-5 at 9. The timeliness of the retrospective appraisal review reports therefore depends on the sufficiency of the supplementation of the French and Gregoire reports.

As discussed below, the court finds that EA is not prejudiced by the substitution of the 21 loans, because it actually reduced plaintiffs' damages calculations. The five new retrospective appraisal reviews arise out of the 21 substitute loans, and were required to be produced as information that Mr. Gregoire and Dr. French relied on in their supplemental reports. Therefore, the court DENIES the motion to strike the retrospective appraisal review reports.

3. Ms. Molitor-Gennrich

Dawn Molitor-Gennrich is a purported expert on appraisals. Libeu Decl. Ex. 1, Molitor Rep., Dkt. No. 385-12, at 1. Ms. Molitor-Gennrich was retained by plaintiffs to discuss the relationship between EA and WMB and its tendency to produce inflated appraisals, and "to review and offer an opinion regarding the conclusions of" Danny Wiley, another purported appraisals expert. Id. at 1-2. With respect to Mr. Wiley, Ms. Molitor-Gennrich "agree[s]" with his summary conclusion. Id. at 49.

Ms. Molitor-Gennrich's supplemental report "provide[s] additional fact-based examples, details, and citations for opinions and conclusions I made in my original expert report, dated January 31, 2014." Libeu Decl. Ex. 6, Molitor Supp. Rep., Dkt. No. 385-12, at 4. Ms. Molitor-Gennrich also "offer[s]" certain opinion of Mr. Wiley "as [her] own." Id. at 70. She then "adopt[s] as [her] own the summary conclusion presented by Mr. Wiley." Id. at 73.

EA argues that the inclusion of additional supporting documents is improper because all documents (except one) were available to Ms. Molitor-Gennrich at the time of her original report.

EA also argues that Ms. Molitor-Gennrich used her supplemental report to adopt the opinions of Mr. Wiley. EA alleges this is improper because her "adoption" of Mr. Wiley's opinions "was a direct result of EA's criticism of Plaintiffs' refusal to produce Mr. Wiley for deposition to defend his own work." Dkt. No. 385-9 at 15.

Plaintiffs respond that Ms. Molitor-Gennrich's supplemental report is perfectly consistent with her opening report, and only adds additional clarity and support for her original opinions. Plaintiffs do concede that the additional information Ms. Molitor-Gennrich cited was previously available. Dkt. No. 409-5 at 11.

EA argues generally that is was prejudiced by plaintiffs' supplemental reports because "Defendants had no time to address the new opinions in their rebuttal reports" and the reports "fundamentally revis[ed] their expert case." Dkt. No. 385-9 at 23. The court does not agree that Ms. Molitor-Gennrich "fundamentally revised" her expert report by either including additional examples or confirming that she agreed with Mr. Wiley's opinions. However, plaintiffs failed to present any justification for Ms. Molitor-Gennrich's supplementation. Plaintiffs have the burden of showing that their failure to comply with Rule 26(e) (i.e. failing to file a proper supplemental report, and instead filing a second opening report), was substantially justified or harmless. See Torres v. City of Los Angeles, 548 F.3d 1197, 1213 (9th Cir. 2008). Here, plaintiffs have not shown any substantial justification for Ms. Molitor-Gennrich's supplemental expert report, and must rely on her original report, which supposedly contains the same substantive opinions. Accordingly, the motion to strike is GRANTED as to Ms. Molitor-Gennrich's supplemental expert report.

4. Mr. Gregoire

Francois Gregoire is another purported expert on appraisals. Libeu Decl. Ex. 2, Gregoire Rep., Dkt. No. 385-12, at 1. Mr. Gregoire's opening expert report detailed his method of selecting "competent and qualified expert review appraisers" based on "the minimum requirements" specified by Mr. Wiley. Id. at 12. Mr. Gregoire concurred with Mr. Wiley's qualification requirements, and also eliminated appraisers that might have a conflict of interest. Id. Mr. Gregoire also developed a plan for having the appraisers conduct retrospective reviews, based on Mr. Wiley's prior Expert Report. Id. at 13. Mr. Gregoire's supplemental report states that "I understand that Mr. Wiley is not a testifying expert at trial, and therefore I will not be relying on his report in this supplemental report," but that Mr. Gregoire's opinions on the minimum requirements of his retrospective appraisal surveys remain the same. Libeu Decl. Ex. 7, Gregoire Supp. Rep., Dkt. No. 385-12, at 1-3.

EA characterizes Mr. Gregoire's supplemental report as an abrupt change in course, but acknowledges that Mr. Gregoire offers a "substantively identical review framework as [Mr. Wiley]." Dkt. No. 385-9 at 16.

EA also argues that Mr. Gregoire's new "service type" analysis is improper because it is based on the July 2013 Chase spreadsheet and therefore was available at the time of the opening reports.

EA's next complaint is that Mr. Gregoire's supplemental report includes 21 appraisals that were not used in his original report. At the time of the opening reports, the 450 Sample Loans included at least 21 loans that were not actually funded by WMB. Gregoire Supp. Rep. at 6. EA argues that "Mr. Gregoire could have avoided mistakenly including these twenty-one appraisals in the sample if he had properly prepared for his initial report by reviewing publicly available mortgage deeds beforehand." Dkt. No. 385-9 at 17-18.

Plaintiffs' respond that Mr. Gregoire's substitution of the 21 appraisals was required because Chase failed to provide the WMB loan files in a timely manner. Dkt. No. 409-5 at 14. Plaintiffs were in the process of confirming the WMB loan status of all 450 sample loans, but could not obtain confirmation for 21 of the loans by the January 31, 2014 deadline. Id. All 21 of the substitute loans "either had no review at all or the review they had didn't meet the thresholds to be inflated" and therefore actually reduced the aggregate inflation analysis calculated by Dr. French. See Libeu Decl. Ex. 9, French Supp. Rep., Dkt. No. 385-13, at ¶ 11; Braun Decl. Ex. G, French Dep., Dkt. No. 4096, at 98:21-99:1.

The court finds that plaintiffs' substitution of the 21 loans into the sample was both harmless and substantially justified. It was harmless because it actually reduced the aggregate inflation calculations, which is part of the key liability issue. It was substantially justified because plaintiffs believed that the Chase information was reliable on the issue of loan funding up until the deposition of Mr. Nagamastu, and plaintiffs began searching for public records to confirm loan status shortly thereafter. Dkt. No. 409-5 at 19. The service type analysis, on the other hand, was not substantially justified because it was based on the July 2013 spreadsheet which plaintiffs' expert had more than enough time to analyze prior to the January 31, 2014 opening reports. Finally, Mr. Gregoire's adoption of Mr. Wiley's methods is harmless but not substantially justified, as discussed with regard to Ms. Molitor-Gennrich. The explanation of Mr. Wiley's methods contained in Mr. Gregoire's opening report is sufficient. Accordingly, the court GRANTS IN PART AND DENIES IN PART the motion to strike the supplemental report of Mr. Gregoire. The portions of the supplemental report based on "service type" and adoption of Mr. Wiley's methods are stricken; the portions of the supplemental report based on the 21 substitute loans are not stricken.

5. Ms. Ghiglieri

Catherine Ghiglieri is a purported expert on "the manner in which banks makes [sic] loans." Libeu Decl. Ex. 3, Ghiglieri Rep., Dkt. No. 385-12, at 2. Ms. Ghiglieri was retained to "opine on the manner in which bank examiners determine if a loan is subject to [RESPA]." Id. Ms. Ghiglieri's opening report is limited to the conclusions that if a loan file includes a HUD-1 form or other RESPA disclosures, it is presumptively a RESPA-qualifying loan, and that named plaintiffs' loans were subject to RESPA. Id. at 12-14. Ms. Ghiglieri did not attempt to estimate the number of class members, or specific class members beyond named plaintiffs, who had RESPA-qualifying loans.

In her supplemental report, Ms. Ghiglieri presents two methods of determining which or how many class members have RESPA-qualifying loans. First, Ms. Ghiglieri reviewed the 450 sample loans and extrapolated the proportion of RESPA loans from the sample for Dr. French to apply to the entire population of loans. Second, Ms. Ghiglieri used the Chase spreadsheets and codes to group loans by type and opine on whether certain loan types were RESPA-qualified or not. EA moves to strike only the former analysis. Libeu Decl. Ex. 8, Ghiglieri Supp. Rep., Dkt. No. 385-13 at 4-6.

EA argues that Ms. Ghiglieri's analysis of the 450 sample loans is clearly improper because other experts (Mr. Gregoire and Dr. French) analyzed the sample loans in their opening reports.

Plaintiffs' respond that Ms. Ghiglieri's analysis of the 450 sample loans was delayed by the Chase spreadsheets. Dkt. No. 409-5 at 18 n.10. The court does not find this persuasive because Ms. Ghiglieri's review of the 450 sample loans is not based on the Chase spreadsheet code, although she does include this information in her analysis. See Ghiglieri Supp. Rep. at App. D-H (columns "Chase Spreadsheet" and "Data in LoType field of JPMChase_WaMu_000003"). Ms. Ghiglieri "directed Plaintiffs' counsel to pull the loan files and to record information from those loan files into the columns shown in Appendices D to H." Id. at 4. There is no mention in her supplemental report that any information from the Chase spreadsheets was used to evaluate the RESPA-status of the 450 sample loans.

Plaintiffs also argue that Ms. Ghiglieri's review of the sample loans is "consistent with her earlier opinions of how bank examiners would determine the existence of a RESPA-covered loan." Dkt. No. 409-5 at 18 n.10. The court also finds this unpersuasive because Ms. Ghiglieri's supplemental report goes well beyond the review suggested in her opening report, which seemed to consist only of checking for RESPA documents, not reviewing other information in the loan file.

The court GRANTS the motion to strike the supplemental report of Ms. Ghiglieri as to the review of the 450 sample loans only, as specified in EA's motion. The entirely new review of the sample loans, already produced at the time of the opening reports, was not substantially justified and was not harmless to EA's preparation of their rebuttal reports or deposition of Ms. Ghiglieri.

6. Dr. French

Dr. Gary French, a purported economics expert, was tasked with determining the "prevalence of EA appraisal inflation and to calculate class-wide damages." Libeu Decl. Ex. 4, French Rep., Dkt. No. 385-12, at ¶ 4. Dr. French's opening report calculates aggregate inflation and class-wide damages. Dr. French extrapolates the number of WMB-funded loans from the 450 sample loans. French Rep. ¶¶ 17, 25. Dr. French's supplemental report updates his analyses and reviews the Chase spreadsheets as an alternate method of calculating class-wide damages. Libeu Decl. Ex. 9, French Supp. Rep., Dkt. No. 385-13, p.3 ¶ 5.

EA argues that Dr. French's updated liability analysis "is a thinly veiled attempt to correct the same sampling mistakes made by Mr. Gregoire" regarding the 21 substitute loans. Dkt. No. 385-9 at 19. EA argues that Dr. French's updated damages analysis is faulty for the same reason, and because he makes a calculation to exclude non-RESPA loans, based on Ms. Ghiglieri's supplemental report. French Supp. Rep. ¶¶ 14, 17-21. EA also argues that Dr. French's new damages analysis, which identifies specific class members that received funded loans, is improper because it was based on the July 2013 Chase spreadsheets. Finally, EA argues that Dr. French improperly included the "service type" opinions of Mr. Gregoire into his supplemental report. French Supp. Rep. ¶ 26. EA does not move to strike Dr. French's "MTM Sent to ML origination methodology." Dkt. No. 385-9 at 21.

Plaintiffs respond that Dr. French relied on information in the later Chase spreadsheets that had to be decoded. Dkt. No. 409-5 at 20. It is unclear whether the service type information was included in the later spreadsheets. Dr. French's supplemental reports only references JMPC000000, the July 2013 spreadsheet, as relevant to the service type analysis. French Supp. Rep. at ¶¶ 23-26.

EA also argues that Dr. French required the RESPA-status information to perform his service type analysis. Id. This is somewhat misleading, however, because Dr. French actually testified that he needed the RESPA information to arrive at the final damage numbers, not to perform the service type analysis. French Dep. at 223:22-23, 224:22-225:3. Plaintiffs further argue that because Dr. French relied on the reports provided by Ms. Ghiglieri and Mr. Gregoire, he had to update his own report in response to their updates.

As discussed above, the court is satisfied that the substitution of the 21 loans was substantially justified. Dr. French's analysis based on RESPA status, however, is based entirely on Ms. Ghiglieri's review of the 450 sample loans, and therefore must be striken. The court has also already addressed the service type analysis done by Mr. Gregoire. Finally, the identification of specific class-member loans based only on the July 2013 spreadsheets is not substantially justified and is struck. The court GRANTS IN THE PART AND DENIES IN PART the motion to strike Dr. French's supplemental expert report, denying as to the 21 substitute loans and granting as to the remainder of the motion.

7. Dr. Courchane

Dr. Marsha Courchane, another purported economics expert, compared values of EA-WMB appraisals to median house prices and to retrospective automated valuation models ("AVMs"). Libeu Decl. Ex. 5, Courchane Rep., Dkt. No. 385-12, at ¶ 14. Dr. Courchane's supplemental report includes the same comparisons, but restricts her analysis to loans funded by WMB based on the both the July 2013 spreadsheet and the later Chase spreadsheets. Libeu Decl. Ex. 10, Courchane Supp. Rep., Dkt. No. 385-13, at ¶ 6, n.3 (referring to both JPMC000000 and JPMChase_WaMu_000003 and MTM_SEND_TO_ML field).

Dr. Courchane's supplemental report states her analysis is dependent on the late Chase spreadsheets. Dkt. No. 409-5 at 21. However, Dr. Courchane admitted at her deposition that her report was only based on the July 2013 spreadsheet and the October 7, 2013 deposition of Chase's Mr. Jei Nagamatsu. Dkt. No. 416-6 at 7, citing Supp. Libeu Decl. Ex. 15, Courchane Dep. at 224:918, 226:11-227:6 (confirming that Dr. Courchane's only basis was the transcript of Mr. Nagamatsu's deposition and the July 2013 spreadsheet). Both of these documents were available to Dr. Courchane at the time of her opening report.

As discussed above, the supplemental analysis based solely on review of the July 2013 spreadsheet, even in combination with Mr. Nagamatsu's deposition, is not substantially justified. The court GRANTS the motion to strike as to Dr. Courchane's supplemental report.

III. ORDER

For the reasons stated above, the court:

• On defendant's motion for summary judgment, the court GRANTS the motion as to plaintiff Sidney Scholl, and otherwise DENIES the motion.



• On plaintiffs' motion for summary judgment on affirmative defenses, the court GRANTS as to defenses 1, 2, and 4; and DENIES as to defense 3.



• On plaintiffs' motion to bifurcate, the court GRANTS IN PART and bifurcates the question of whether EA appraisals were inflated in the aggregate as a result of an agreement between WMB and EA from any class member eligibility or damages questions.



• On defendant's motion to decertify the class, the court DENIES the motion.



• On defendant's motion to strike, the court GRANTS the motion to strike as to: the supplemental report of Dr. Courchane, the supplemental report of Ms. Molitor-Gennrich, and the supplemental report of Ms. Ghiglieri; DENIES the motion to strike the retrospective appraisal reports; GRANTS the motion to strike the supplemental report of Mr. Gregoire regarding "service type" and adoption of Mr. Wiley's method, DENIES the motion to strike the supplemental report of Mr. Gregoire as to changes based on the 21 supplemental loans; DENIES the motion to strike the supplemental report of Dr. French as to changes based on the 21 supplemental loans, and GRANTS the motion as to the remainder of the supplemental report.

/s/_________

Ronald M. Whyte

United States District Judge