Pac. Life Ins. Co. v. Bank of N.Y. Mellon

Full title: PACIFIC LIFE INSURANCE COMPANY And PACIFIC LIFE & ANNUITY COMPANY…

Court: UNITED STATES DISTRICT COURT SOUTHERN DISTRICT OF NEW YORK

Date published: Feb 22, 2021

17-CV-1388 (KPF) (RWL) (S.D.N.Y. Feb. 22, 2021)

Opinion

17-CV-1388 (KPF) (RWL)

02-22-2021

PACIFIC LIFE INSURANCE COMPANY And PACIFIC LIFE & ANNUITY COMPANY, Plaintiffs, v. THE BANK OF NEW YORK MELLON, Defendant.

ROBERT W. LEHRBURGER, United States Magistrate Judge.

DECISION AND ORDER: MOTIONS TO EXCLUDE EXPERT TESTIMONY

ROBERT W. LEHRBURGER, United States Magistrate Judge.

The parties in this RMBS trustee action both have moved to exclude the opinion testimony of the other party's experts in whole or in part pursuant to Fed. R. Evid. 702, Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579 (1993), and its progeny. On December 8, 2020, the Honorable Katherine P. Failla, District Judge, referred the motions to me as non-dispositive motions for decision and order. The Court has carefully reviewed the extensive briefing and supporting exhibits of the parties and the tendered reports for the thirteen experts at issue. The Court presumes the parties familiarity with all relevant facts. The parties' respective motions are GRANTED in part and DENIED in part as follows.¹ Plaintiffs will be referred to as PacLife; Defendant as BNYM.

1 The legal framework for addressing expert opinions is well known and fairly set forth by the parties; it need not be repeated at length here. In short, the court must act as gatekeeper to ensure that proffered expert testimony is relevant and reliable in that the expert is qualified and (a) his or her testimony will be helpful to the trier of fact; (b) the testimony provided is based on sufficient facts or data; (c) the testimony is the product of reliable principles and methods; and (d) the expert has reliably applied the principles and methods to the facts. See Fed. R. Evid. 702; see Daubert, 509 U.S. at 589-95, 597; Kumho Tire Co. v. Carmichael, 526 U.S. 137, 141 (1999); Amorgianos v. National Railroad Passenger Corp., 303 F.3d 256, 265-66 (2d Cir 2002). The Court has adhered to that framework in resolving the instant motions.

PacLife's Experts ²

2 The parties submitted separate sets of briefing for each of PacLife's experts. The Court uses the following terminology to refer to the parties' briefs: "[Expert Name] Mem." (BNYM's moving brief); "[Expert Name] Opp." (PacLife opposition brief); and "[Expert Name] Reply" (BNYM's reply brief).

A. Mark Adelson (Securitization)

BNYM seeks to exclude two opinions of PacLife's securitization expert Adelson, which are similar and in some respect the same, as opinions excluded by the Honorable Valerie E. Caproni, District Judge, in Phoenix Light SF Ltd. v. Bank of New York Mellon, No. 14-CV-10104, 2020 WL 1322856 (S.D.N.Y. March 20, 2020) ("Phoenix Light II"), modified on other grounds on reconsideration, 2020 WL 2765044 ("Phoenix III").³

3 Ten of the experts whose opinions are at issue here were also the subject of Daubert motions and rulings in Phoenix Light. Three decisions from that case are relevant to and discussed in this opinion: 2019 WL 5957221 (Nov. 13, 2019) ("Phoenix Light I"); 2020 WL 1322856 (March 20, 2020) ("Phoenix Light II"); and 2020 WL 2765044 (May 28, 2020) ("Phoenix Light III"). The Court recognizes that counsel (in choosing which opinions to challenge and what arguments to make) and the experts (in omitting, modifying, or revising certain opinions) have pivoted in some respects in an attempt to avoid the same or similar pitfalls that led to exclusion of particular opinions in Phoenix Light. The Court nevertheless finds Phoenix Light instructive in many instances, as cited throughout this decision and order.

1. Trustee's Post-Default Obligations

The first opinion concerns trustee obligations after an event of default ("EOD" or "default"), and paragraphs 214-216 of Adelson's report in particular. Paragraphs 214 and 215 contain prescriptive language as to whether or not it is "prudent" for a trustee to wait until investors request action by the trustee in response to an EOD rather than taking action proactively. Judge Caproni excluded similar prescriptive language in Phoenix Light. In particular, Judge Caproni determined that Adelson's opinion of what a trustee should do in response to an EOD was unreliable and not sufficiently tied to his experience. Phoenix Light II, 2020 WL 1322856 at *6-7. The same conclusion is warranted here; Adelson may not testify as to the issue of whether or not a prudent trustee "should" affirmatively take action (beyond sending a notice of default) in response to an EOD without direction. The inadmissible phrases are (a) "The trustee should choose among the reasonable alternatives based on what it expects to produce the best net result for investors" in paragraph 214, and (b) "because such a thing is not prudent" and (c) "(in the mode of a prudent person in the conduct of his own affairs)" in paragraph 215.

The remaining language of paragraphs 214 and 215 offer Adelson's opinions of what "market participants typically expect" and "do not expect" an RMBS trustee to do post-default. Phoenix Light does not address those opinions, which are not prescriptive and not Adelson's opinion of what "should" be done. As PacLife correctly notes, substantial earlier portions of Adelson's opinion are devoted to identifying sources of the expectations of various market constituents. (See Adelson Report ¶¶ 81-93, 90-91, 94-112, 115-134, 138-156, 213.) Neither BNYM's moving brief or its reply brief make mention of those portions of Adelson's report; nor do they lodge any objection to Adelson's opinion as to what market participants expect as stated other than in paragraphs 214-216. As long as Adelson specifically connects his opinions of market expectations to those sources and his experience, the opinions are admissible. See, e.g., Financial Guaranty Insurance Co. v. Putnam Advisory Company, LLC, No. 12-CV-7372, 2020 WL 4251229, at *12 (S.D.N.Y. Feb. 19, 2020) (denying motion to exclude expert testimony on industry standards and "investor expectations" with respect to structured securities); Securities and Exchange Commission v. Revelation Capital Management, Ltd., 215 F. Supp. 3d 267, 273-74 (S.D.N.Y. 2016) (permitting expert testimony offering characteristics "as understood by the securities industry"). Any remaining concerns about this testimony do not merit exclusion; instead, they go to weight, not admissibility, and can be dealt with on cross-examination.

In Paragraph 216, Adelson quotes from and opines that BNYM's policy manual confirms its post-default obligations. Adelson has no expert foundation for construing BNYM's policy manual, and the Court agrees with BNYM that the quoted policy is just as consistent with the opposite of what Adelson suggests it means. See Phoenix Light II, 2020 WL 1322856 at *7 (excluding Adelson opinion in part because source on which he relied "is as consistent with BNYM's position as it is with Plaintiffs' position"); American Home Assurance Co. v. Merck & Co., Inc., 462 F. Supp. 2d 435, 449 (S.D.N.Y. 2006) (allowing expert to testify about standard practices, customs, and industry expectations, but not about specific policy with which he had "nothing to do with the negotiating, drafting or performance of"). Adelson may not offer his opinion of what the policy means or requires.

2. Market Pressure

The second Adelson opinion at issue, found at paragraphs 249-256 of Adelson's report, asserts that proper disclosure of document defects in loans and noncompliance with so-called "Reg AB" would have led to "market pressure" to correct the problems. As Judge Caproni found, that opinion is inadmissible conjecture without sufficient basis. 2020 WL 1322856 at *7-8. PacLife's attempt to shore up the opinion with various articles, reports, and studies fails for the reason highlighted by BNYM: Adelson did not cite most of those sources in his report at all, and two articles that Adelson did cite were not cited for his "market pressure" opinion. (Adelson Reply at 6.)

B. Ingrid Beckles (Loan Servicing)

BNYM seeks to exclude three aspects of Beckles's opinions concerning loan servicing practices and analyses.

1. Foreclosure Timeline Analysis

The first is an analysis assessing foreclosure timelines, i.e., the length of time from loan default to foreclosure. Beckles determined that the foreclosure timelines for the loans at issue exceeded those for loans held by government sponsored entities ("GSE") Freddie Mac and Fannie Mae by an average of 527 days. (Beckles Report ¶ 176.) BNYM contends that Beckles's analysis is unreliable primarily because she selectively chose to include only two of more than a dozen "allowable delays," and she did not analyze individual loan files to determine what allowable delays likely would have applied to each individual loan. (Beckles Mem. at 4.) Beckles provided rationales for these choices at her deposition, explaining that the methods she used were consistent with those she employed during her tenure analyzing foreclosure timeline compliance at Freddie Mac. (See Beckles Opp. at 9-10 (discussing excerpts from Beckles's deposition).) BNYM lodges some salient criticisms of the foreclosure timeline analysis and the choices made by Beckles, but her opinions are sufficiently reliable to be admitted. Accordingly, Beckles is not precluded from testifying about her foreclosure timeline analysis.

2. Liquidated Loans

The second issue concerns Beckles's opinion that 4,847 at-issue loans with outstanding document exceptions were imprudently liquidated. Beckles drew her conclusion from review of "final exception reports" and other material. BNYM asserts that Beckles's opinion is inadmissible because she made an assumption - that none of the outstanding document exceptions were cured prior to liquidation - that is unsupported and false. The problem with that argument is that the evidence produced by BNYM on cures apparently is sparse and disputed. Indeed, BNYM does not dispute PacLife's assertions that BNYM did not produce its available cure data and instead obtained a protective order on the purported basis that it was too burdensome to produce.⁴ (Beckles Reply at 6.) BNYM should not be rewarded with excluding Beckles opinion based on its own refusal to produce the relevant data that supposedly undermines it.

4 BNYM's refusal to produce cure reports appears to distinguish this case from Judge Caproni's ruling in Phoenix Light regarding Beckles's opinion that thousands of loans in the trusts there at issue remained uncured. Indeed, there, Beckles opinion faltered because it was based on an incorrect reading of BNYM's cure reports. Phoenix Light, 14-CV-10104, Dkt. 365, at 2. Judge Caproni rejected plaintiffs' efforts to rehabilitate Beckles's opinion and observed that the burden to show that document exceptions were not cured lied with plaintiffs, not BNYM. Id. Left "[w]ith no evidence that the loans with previously identified exceptions had not been corrected, Beckles ha[d] no basis to make that assertion." Id. at 3. Here, in contrast, Beckles cites to what limited evidence there is and explains the basis for her opinion. (Beckles Report ¶¶ 148-151.) Notably, BNYM does not cite to Judge Caproni's ruling on Beckles in Phoenix Light to support its motion.

In reply, BNYM says it does not fault Beckles for not considering "additional" data but rather for not considering information that BNYM did produce. (Beckles Reply at 6-7.) BNYM points in particular to an internal BNYM email referring to "progress" in curing document exceptions and states that Beckles "would have found evidence of the considerable progress" in curing document exceptions "[h]ad [she] sought out additional information." But BNYM does not identify what that evidence is other than the vague email; nor does BNYM indicate whether that progress included any of the loans at issue.

BNYM argues that Beckles did receive but improperly ignored trailing exception and cure reports that she reviewed but deemed unreliable. BNYM then faults Beckles as not making it "entirely clear" why she deemed the reports unreliable. But in attempting to illustrate that contention, BNYM cites to two explanations given by Beckles, one in her deposition and one in her report. (Beckles Mem. at 9.) The explanations may be different, but that does not render them incompatible. Ultimately, the parties disagree about what to conclude from the limited information that BNYM produced regarding document exception cures. BNYM can make its points on cross-examination of Beckles. It is up to the fact-finder, however, to resolve the dispute about what the evidence shows regarding document exception cures and to assess whether Beckles's assumption that none of the 4,847 loans had document exception cures is reasonable. Accordingly, Beckles's opinion in that regard is not excluded.

3. Loss Severity Differential Opinion

BNYM's third challenge seeks to exclude Beckles's opinion concerning the cause - which she identifies as servicing performance - of the differential found by another expert, Spencer, in loss severity between loans at issue and the benchmark loans established by Beckles. Rather than setting forth that argument in its brief in support of its motion to exclude Beckles's opinion, BNYM incorporates by reference its arguments addressing the issue in its brief supporting its motion to exclude Spencer's opinions. This particular challenge thus goes to Spencer's opinions. As discussed next, the Court finds Spencer's analysis to be admissible. Accordingly, Beckles's opinions will not be excluded as being based on Spencer's analysis.

C. Bruce Spencer (Loss-Severity Differential Analysis and Damages)

Spencer is another expert on whose opinions Judge Caproni ruled in Phoenix Light. There, Judge Caproni found Spencer's opinions admissible and rejected BNYM's arguments that Spencer's loss severity differential analysis was unreliable. Phoenix Light II, 2020 WL 1322856 at *9-12. BNYM now has retooled its arguments in an attempt to exclude Spencer's loss severity differential analysis in this case.

1. Endorsement of Beckles's Loss Severity Benchmark

Spencer relies on the loss severity benchmark determined by expert Beckles. BNYM does not challenge Spencer's doing so but does seek to exclude, as impermissible bolstering or vouching, any endorsement by Spencer that the loss severity benchmark Beckles used is "appropriate." (Spencer Mem. at 4.) Spencer's report does not include such an opinion; rather, that assessment emerged during his deposition in response to a BNYM question. PacLife represents that Spencer is not and will not be offering an opinion on the propriety of Beckles's loss severity benchmark. (Spencer Opp. 7-8.) Accordingly, the issue is moot, and Spencer will not, and may not, offer the opinion.

2. Spencer's Loss Severity Differential Analysis

According to BNYM, Spencer's analysis of loss severity differential (between loans at issue and GSE loans benchmarked by Beckles) attributable to servicing is not reliable because his regression model did not include all requisite variables. In particular, BNYM refers to debt-to-income ratio ("DTI"), loan product (framed by BNYM in terms of whether a loan is prime, subprime, Alt-A, or such), "REO disposition methodology," and certain other loan "features" all of which are variables that Beckles recognized "could" impact loss severity. (Spencer Mem. at 7.)

That argument does not stand up to scrutiny. The variables that Spencer incorporated in his regression analysis are the variables that Beckles identified as those which "systematically impact" loss severity between the at-issue and benchmark loans. (Beckles Report at ¶¶ 113-114.) BNYM's challenge, however, is based on variables that are either not included in Beckles's list, or variables that Spencer did account for in some fashion. Both parties cite case law holding that to be admissible, a regression analysis must include all "major factors" but need not include other less significant factors. See, e.g., Reed Construction Data Inc. v. McGraw-Hill Cos., 49 F. Supp. 3d 385, 400-01 (S.D.N.Y. 2014), aff'd, 638 F. App'x 43 (2d Cir. 2016). The variables that BNYM brings to the fore are not among the "major" variables that Beckles considers important to include. At the very least, the arguments and sources marshalled by BNYM, and the counter-arguments and sources advanced by PacLife, show that reasonable experts can disagree about the significance of the variables used and not used in Spencer's analysis. If BNYM wants to take Spencer or Beckles to task in this respect, it can do so through cross-examination.

3. Spencer's Reply Declaration

One of the sources cited by PacLife to rebut BNYM's criticism of Spencer's analysis is a declaration from Spencer prepared for and submitted in opposition to BNYM's motion to exclude. In reply, BNYM requests that the Spencer declaration be stricken as belatedly submitted after expert discovery and in conflict with the parties' agreement not to submit reply reports.

The governing legal principles require courts to exclude expert declarations submitted in response to Daubert motions that set forth "a wholly new and complex approach designed to fill a significant and logical gap in the first report," but allow courts to consider "evidentiary details that a declaration provides in support of opinions already expressed in the expert's report." Phoenix Light I, 2019 WL 5957221 at *2 (internal quotation marks and citations omitted). Proper declarations include "declarations that address concerns raised in a Daubert motion about the reliability and application of an expert's methodology." Id. "The key issue is whether the expert declaration is sufficiently within the scope of the initial expert report ... so that an opposing party is not sandbagged with new evidence." Id. (internal quotation marks, citations, and brackets omitted). The Spencer declaration fits comfortably within those boundaries as its contents respond directly to BNYM's attacks on the reliability of Spencer's methodology. See id. at *2-5 (denying BNYM's motion to exclude the responding declarations of Spencer, as well as those of Adelson, Beckles, and Bitner, in similar circumstances).

BNYM also takes issue with paragraph 7 in Spencer's declaration because it expands on his background experience. (Spencer Reply at 3-4.) BNYM does not suggest that the information contradicts anything Spencer said at his deposition or that Spencer failed to disclose the information in response to a question at his deposition. As Judge Caproni explained in Phoenix Light, "Rule 26(a)(2) does not require [an expert] to discuss every anecdote from his past experience that would support the opinions in his Report." Phoenix Light I, 2019 WL 5957221 at 4. The paragraph will not be stricken.

Additionally, BNYM contends that Spencer's statement that "Alt-A" lacks an industry-accepted meaning contradicts his deposition testimony. (Spencer Reply at 4-5.) The Court does not agree. The statements cited are not mutually exclusive. The Court agrees, however, that Spencer is not qualified to opine about industry definitions of Alt-A, subprime, etc. He is free, however, to rely on his understanding based on information provided by other sources, such as Beckles or industry articles.

BNYM also challenges portions of the declaration that merely "take up the mantle of Plaintiffs' attorneys to advance arguments in response to those made in BNYM's memorandum of law." (Spencer Reply at 4.) The Court agrees with respect to Paragraphs 18 and 21 of Spencer's declaration. Accordingly, those two paragraphs are stricken. See id. at *4 n.4 (striking paragraph in expert declaration for "usurping counsel's role in responding to Defendant's Daubert motion").

Finally, BNYM requests, in the alternative, that it be allowed to take further discovery of Spencer concerning his declaration. The Court agrees that PacLife should produce all additional materials, if any, on which Spencer relied in his declaration that have not previously been produced. Given the limited scope of the declaration and the detailed explanations provided, however, the Court does not find further deposition testimony to be merited. See id. at *1-2 (denying BNYM's alternative request to re-depose plaintiffs' experts who submitted reply declarations). BNYM has ample expert firepower with which to assess Spencer's responsive opinions and provide fodder for cross-examination.

D. Richard Bitner (Reunderwriting)

Bitner reviewed 1,147 mortgage loans in eight of the trusts at issue and determined that 841 of the loans had at least one material breach of a representation and warranty ("R&W"). He found 2,752 breaches in total. BNYM now seeks to exclude certain of the breaches Bitner found. BNYM challenges four categories of breaches.

1. Regulatory Compliance R&W

Bitner determined that 18 loans contained material breaches of the "Regulatory Compliance" R&W stating that the loans were originated in compliance with all applicable local, state, and federal laws. The basis for Bitner's finding a breach in each instance is the absence or defective nature of a "HUD-1" statement, which lists all charges and credits to the buyer in a sale or refinance. BNYM contends Bitner's findings in that regard are inadmissible because Bitner fails to identify any particular law with which any of those loans failed to comply.

BNYM is correct. Bitner and PacLife respond by saying that HUD-1 statements are "in effect mandatory" - without citing any law saying so (Bitner Opp. at 5); that Bitner's report does address various lending laws such as the Truth in Lending Act - without citing any provision in those laws requiring a HUD-1 (Bitner Opp. at 5); that no originator would purchase a loan without a complete and correct HUD-1 - again without identifying any legal violation as opposed to a market expectation (Bitner Opp. at 5-6); and that "it is not possible to check whether there has been a violation of these laws" without a HUD-1 - which only confirms that Bitner does not know if there was any violation to begin with and that his breach findings based on missing or defective HUD-1 statements are speculative and unreliable (Bitner Opp. at 5). PacLife does not cite any case law or other legal authority that a loan file must contain a completed, non-defective HUD-1 statement. Having failed to identify any legal requirement on which Bitner can reliably base his findings of breach of the Regulatory Compliance R&W, Bitner may not testify to these alleged breaches.

2. No-Default R&W

Bitner found breaches of the "No-Default" R&W in 444 loans. The No-Default R&W includes two different types of default, one specific to a "material monetary default," and the other referring to knowledge of an event of default or breach in a more general sense. (Bitner Mem. at 10 (characterizing the first clause as a "monetary default" and the second as referring to "default[s] generally") (emphasis and modification in original).) BNYM argues that Bitner's No-Default breach findings should be excluded because, at his deposition, Bitner appeared to confuse the two provisions. Specifically, BNYM contends that Bitner improperly interpreted the scope of the monetary default clause to be as broad as the second default clause. (Bitner Mem. at 10-11.)

The Court does not agree. The deposition excerpt quoted by BNYM, even if scrambling the first and second clauses, does not state that Bitner based his breach findings on the first clause as distinct from the second clause. Nor does it state that Bitner did not base his findings on the second clause. More to the point, because there is no dispute that the No-Default R&W contains both the more specific reference to monetary defaults and the more general reference to defaults, BNYM's quibbling about Bitner's deposition testimony is trivial. BNYM further contends that Bitner's opinion is incompatible with his recognition that a No-Default R&W is distinct from a No-Fraud R&W, and that at his deposition, he essentially equated the latter with the former. (Bitner Mem. at 11-12.) That is a basis for cross-examination, not exclusion.

3. Prospectus Supplement Representations

The R&W's on which Bitner based his findings are found in the RMBS transaction documents known as pooling and servicing agreements ("PSAs"). BNYM faults Bitner for including references in his chart of "Contractual Provision(s) Breached" not only to the PSA provisions, but also to statements made in prospectus supplements ("ProSupps"). As BNYM argues, with legal support, the ProSupps are not contracts. See, e.g., Stichting Pensioenfonds ABP v. Credit Suisse Group AG, No. 653665/2011, 2012 WL 6929336, at *5 (N.Y. Sup. Ct. Nov. 30 2012) ("A prospectus, or a prospectus supplement, is indeed not a contract").

PacLife does not say otherwise, instead explaining that Bitner's references to the ProSupps merely provide further "support" and "detail" for the "underlying breach." (Bitner Opp. at 9.) Regardless, as PacLife does not dispute the non-contractual nature of the ProSupps, Bitner may not testify that the ProSupps create contractual obligations and may not designate statements made in the ProSupps as "contractual provisions." Of course, this does not preclude Bitner from testifying about provisions in the ProSupps if otherwise relevant to his opinions. Nor does it negate Bitner's breach findings that are based on PSA provisions even if Bitner additionally cited ProSupp provisions.

4. MLS R&W

Lastly, BNYM takes issue with Bitner's finding breaches of the "MLS" R&W. MLS refers to the "Mortgage Loan Schedule," which contains specified information about each loan (e.g., loan number, maturity date, LTV, etc.). The MLS R&W states that the information set forth in the MLS is "true and correct in all material respects as of the Closing Date." At the RMBS transaction closing, the MLS data is contained in what is known as the Mortgage Loan Tape, or simply the "Loan Tape." The Loan Tape may also include data beyond the specified data points required by the PSA. Loan Tapes have been described as "the principal source of data for the disclosures" to investors, the SEC, and ratings agencies. Federal Housing Finance Agency v. Nomura Holding America, Inc., 104 F. Supp. 3d 441, 462-63, 465 (S.D.N.Y. 2015).

Bitner examined the Loan Tapes for the loans he reviewed and found breaches where he deemed the information on the Loan Tapes to be not true and correct. BNYM seeks to exclude any asserted breach based on data discrepancies for data point categories that are not among those specified on the MLS as designated by the PSA. In particular, BNYM takes issue with Bitner's finding a breach of the MLS R&W based on Loan Tape data discrepancies regarding DTI, CLTV Ratio, Appraised Value, and Credit Score. BNYM asserts that there are 201 such loans.

The Court agrees with BNYM. Data on the Loan Tape that extends beyond the data points specified by the PSA no doubt are important in at least some respects to some or all of the relevant constituents (i.e., investors, SEC, rating agencies, etc.). But PacLife does not point to anything showing that discrepancies with respect to those additional data points are encompassed by a contractual R&W. In support of its position, PacLife cites Bank of New York Mellon v. WMC Mortgage, LLC, 136 A.D.3d 1, 3, 6, 22 N.Y.S.3d 3 (1st Dep't 2015). In that case, however, the MLS R&W expressly referred to the Loan Tape, not just the MLS. Id.at 4-5, 22 N.Y.S.3d at 4-5 (MLS R&W referring to "information set forth in the Mortgage Loan Schedule and the tape"). Here, however, the MLS R&W refers only to information set forth in the MLS and makes no mention of the tape. (See CWALT 2006-32CB PSA Schedule III-A(1).) Accordingly, Bitner may not testify to breaches based on discrepancies for data points that are not included in the PSA-defined MLS. That said, any loan that has such an "out-of-bounds" discrepancy may still be treated as breaching based on discrepancies regarding "in bounds" data points.

E. Christopher Milner (Damages)

BNYM seeks to exclude Milner's damages opinions in two respects. The first condemns Milner's analysis for failing to properly account for BNYM's recovery of proceeds from its participation in a settlement with Countrywide Financial and its successor Bank of America (the "Countrywide Settlement"). The other criticism concerns calculation of pre-judgment interest.

1. The Countrywide Settlement

Milner's damages model is based on comparison of a but-for world - what PacLife would have received if BNYM had performed as PacLife contends it should have - to an actual world of what actually happened and what PacLife actually received. The methodology is reliable, and BNYM does not argue otherwise.

Instead, BNYM asserts that Milner's calculation is fatally flawed because the actual world modeled by Milner does not accurately represent the actual world. Specifically, Milner did not factor into the data entered into the model the settlement proceeds that PacLife received from the Countrywide Settlement, which involved some of the same trusts at issue here.

BNYM's criticism is well taken. Judge Caproni's rulings in Phoenix Light again provide guidance. There, Judge Caproni granted BNYM's motion to exclude the damages opinion of plaintiffs' damages expert insofar as it had not appropriately accounted for settlement payments. Phoenix Light III, 2020 WL 2765044. The expert used a similar model to what Milner uses here. As here, the expert's alleged error also involved failure to properly account for the Countrywide Settlement. In particular, he accounted for the Countrywide Settlement in both the but-for and actual worlds. The problem with that approach is that it effectively negated any accounting for the Countrywide Settlement because it was included "on both sides of the ledger." Id. at *1. Referring to the Countrywide Settlement payments, Judge Caproni explained, "[the expert] must not only include those payments in the actual experience, he must also omit them from the but-for experience." Id. at *2 (emphases added). Having failed to do so, the expert's analysis was "absurd" and "not grounded in reality." Id. Here, Milner failed to include the Countrywide Settlement payment on either side of the ledger, which is the equivalent of including it on both sides and no less absurd. Just as in Phoenix Light, "the model's reliability is not in question; the problem is that one of its assumptions has corrupted its output beyond reason." Id. at *3.

PacLife does not put forward an effective response to that flaw in Milner's analysis. To start, PacLife defends Milner's omission of the Countrywide Settlement payments from both the actual and but-for worlds based on the argument that the Countrywide Settlement compensation was predicated on Countrywide's misconduct, not BNYM's misconduct. PacLife falls back on an earlier decision in this case in which the Court denied BNYM's motion to dismiss, Pacific Life Insurance Co v. Bank of New York Mellon, No. 17-CV-1388, 2018 WL 1382105 (S.D.N.Y. March 16, 2018). At that juncture, BNYM argued that the Countrywide Settlement precluded PacLife's claims here. In rejecting that argument, the Court held that "the damages sought are distinct" and that "the present action pertains to conduct not directly at issue in the Countrywide Settlement." Id. at *12. PacLife now seeks to turn the Court's rationale into a basis to argue that the damages analysis need not account for the Countrywide Settlement at all and posits that Milner's damages calculations "must be evaluated within the context of plaintiffs' stated theory of the case." (Milner Opp. at 11.) Whatever theory PacLife may embrace, however, it cannot escape the requirement that its damages analysis account for the Countrywide Settlement payments and do so in a manner "grounded in reality."

PacLife asserts that even though Milner's analysis did not account for Countrywide Settlement payments in the data fed into the model, he nevertheless can account for the payments on the output side by subtracting the payment total from the damages total. (Milner Opp. at 10.) There are at least two significant problems with that argument. First, Milner did not perform any such set-off calculation or describe any such accounting in his report. At the very end of his report, Milner does quantify the proceeds that the trusts received from the Countrywide Settlement but he does not at all discuss what should be done with those figures. (Milner Report ¶ 111.) Second, even if Milner's report had put forth a post-output set-off calculation, it would remain flawed because simply subtracting the overall gross numbers on the output side of the model does not reflect the but-for world where the Countrywide Settlement payment proceeds would have been fed through the so-called "waterfall" prescribed by the transaction documents for making payments to investors. Perhaps the simple subtraction suggested by PacLife would be a rough, and even reasonable, approximation of the impact of the Countrywide Settlement proceeds, but PacLife has not presented any such analysis.

PacLife also seeks to minimize Milner's improper assumption by explaining that BNYM's criticism relates to Milner's treatment of approximately only "10 of the nearly 2 million monthly calculations he makes as part of his damages model." (Milner Opp. at 12.) Put another way, BNYM's criticism concerns only .000005% of the inputs in Milner's model. (Milner Opp. at 12.) That argument has some appeal, but does not provide insight into what the actual impact of the omitted inputs have on the outcome. For instance, according to BNYM, the Countrywide Settlement includes more than $30 million in recoveries related to the trusts at issue. (Milner Reply at 1.)

Notwithstanding the significance of Milner's failure to account for the Countrywide Settlement, PacLife suggests that it would be "simple" to rerun the numbers with the Settlement proceeds accounted for in the data fed into Milner's model. (Milner Opp. at 12.) Given the reliability of Milner's model, the Court does not believe it would be appropriate or proportional to exclude Milner from offering his opinions on damages. Rather, he should be given the opportunity to execute the "simple" process of rerunning the numbers to take account of the Countrywide Settlement proceeds. Cf. Phoenix Light III, 2020 WL 2765044 at *3 (giving the parties the opportunity to brief any issues or concerns in having Milner do so). The parties shall meet and confer as to the timing for doing so and raise any unresolved issues with the Court.

2. Pre-Judgment Interest

BNYM faults Milner's determination of pre-judgment interest for calculating monthly interest separately on principal and on interest rather than on the monthly total of principal and interest. BNYM contends that methodology gives PacLife a windfall because it awards pre-judgment interest whenever either principal or interest is greater in the but-for world than the actual world, thereby awarding pre-judgment interest for months in which PacLife actually made more in combined loan principal and interest than it would have absent BNYM's alleged breaches. As underscored by BNYM, Milner's breaking out principal and interest separately for purposes of calculating pre-judgment interest conflicts with his not having broken them out separately in determining underlying damages.

In Milner's defense, PacLife points to the fact that BNYM's remittance reports to investors break out principal and interest separately. (Milner Opp. at 14.) That is no surprise. But that also has no bearing on whether the two components should be separated for purposes of calculating pre-judgment interest, and PacLife has provided no plausible explanation of why it would. PacLife also invokes N.Y. C.P.L.R. § 5001(b), which governs calculation of pre-judgment interest. (Milner Opp. at 13.) That provision, however, does not concern the issue presented here. Rather, it speaks to the date from which interest should be calculated.

The Court finds suspect Milner's calculation of pre-judgment interest on principal and interest as separate components. That said, the criticism ultimately goes to weight, not admissibility, and is an issue that the Court should resolve at a later time. The Court may take testimony on the issue out of presence of the jury and then make a determination of the appropriate calculation in the event that PacLife obtains a verdict in its favor. See Soley v. Wasserman, No. 08-CV-9262, 2013 WL 526732, at *9 (S.D.N.Y. Feb. 13, 2013) (denying motion to strike expert's calculation of pre-judgment interest); N.Y. C.P.L.R. § 5001(c) (indicating that either the jury or court may determine the date from which interest is calculated, but the calculation shall be made by the clerk of court). Accordingly, Milner's opinion on pre-judgment interest will not be excluded.

BNYM's Rebuttal Experts ⁵

5 The parties submitted omnibus briefs addressing all of BNYM's experts together in each brief. The Court refers to the parties' briefing as follows: "PacLife Mem." (PacLife's moving brief); "BNYM Opp." (BNYM's opposition brief); and "PacLife Reply" (PacLife's reply brief).

A. Christopher S. Hillcoat (RMBS Trustee Custom and Practice)

Hillcoat's primary testimony focuses on corporate trust custom, practice, and law. PacLife moves to exclude all of Hillcoat's opinions as irrelevant, and, failing that, several discrete aspects of Hillcoat's opinions.

1. Relevance of RMBS Trustee Custom and Practice

PacLife argues that the entirety of Hillcoat's report should be excluded because trustee custom and practice is irrelevant to the PSA's prudent person standard, as distinguished from a prudent trustee standard. The Court does not agree. Custom and practice of trustees at the time remains relevant even though not controlling. Phoenix Light II, 2020 WL 1322856 at *7. To be sure, trustee custom and practice is not sufficient alone for the jury to evaluate whether BNYM complied with the prudent person standard. To the contrary, as Judge Caproni put it, "following one's industry colleagues in a race to the bottom does not mean that the bottom reflects prudent conduct." Id.; see also MBIA Insurance Corp. v. Credit Suisse Securities LLC, Index No. 603751/2009, 2020 WL 7041787, at *7-8 (N.Y. Sup. Ct. Nov. 30, 2020) (distinguishing "prudent" as an objective "qualitative indicator" and "customary" as "quantitative," and recognizing that loan underwriting practices during 2006 "may have been customary" but "they certainly were not prudent"). Testimony about trustee custom and practice nevertheless will be helpful to the jury, and PacLife will be able to engage in robust cross-examination on whether BNYM blindly followed a "race to the bottom" and acted imprudently even if customarily. See Phoenix Light, 14-CV-10104, Dkt. 342 (Transcript of Oral Argument, March 9, 2020) ("Phoenix Light Tr.") at 7 (denying plaintiff's motion to exclude Hillcoat's testimony of RMBS trustee custom and practice).

As a second reason to exclude Hillcoat's opinions, PacLife argues that Hillcoat "cannot establish any well-settled meaning" of what the custom and practice actually was, if any. (Hillcoat Mem. at 4.) More particularly, PacLife contends that Hillcoat focuses on only the trustee viewpoint to the exclusion of other industry participants. (Hillcoat Mem. at 4.) That criticism may be valid, but it does not render the opinion unreliable or unhelpful; rather, it goes to weight, not admissibility. Afterall, Hillcoat's opinions largely counter those of PacLife's expert Adelson who opines about the expectations of various market constituents.

PacLife also argues for exclusion on the basis that the PSA language is unambiguous, and admitting Hillcoat's testimony would impermissibly allow the jury to consider extrinsic evidence to vary the meaning of the PSA's prudent person standard. To be sure, Hillcoat's opinions of custom and practice are extrinsic evidence. But those opinions are not being offered to vary contract language and may prove to be evidence of industry-wide imprudent conduct. See Phoenix Light Tr. at 7 ("Evidence of custom and practice in the RMBS industry will be helpful to a jury deciding whether BNYM acted as a prudent person as required by the governing agreements"). Accordingly, Hillcoat's opinions of custom and practice will not be excluded.

2. Legal Opinions and Interpretation of Contract Terms

PacLife seeks to exclude Hillcoat's testimony to the extent it comments on what the PSA or other agreements require or what an RMBS trustee must do, is obligated to do, or should do. The parties do not dispute that their experts may not opine on what the law requires or the meaning of unambiguous contract language. See, e.g., United Talmudical Academy of Kiryas Joel, Inc. v. AP Gas & Electric (NY), LLC, No. 15-CV-1666, 2017 WL 4350609, at *7 (S.D.N.Y. May 17, 2017) (expert may not opine on unambiguous contract language); In re Rezulin Products Liability Litigation, 309 F. Supp. 2d 531, 541 (S.D.N.Y. 2004) (expert may not opine on law). Rather, the parties disagree about whether Hillcoat's opinions transgress those boundaries.

The Court agrees that Hillcoat may not offer opinions about what duties BNYM was obligated to perform, just as Adelson is precluded from opining prescriptively.⁶ See Phoenix Light Tr. at 7-8 (excluding Hillcoat testimony about what duties BNYM had as trustee). The Court has reviewed the paragraphs of Hillcoat's report that PacLife designated for exclusion by that principle. Of the paragraphs identified by PacLife, the following are not problematic and may be admitted, either because they do not contain legal opinion, or because they provide context or are direct rebuttal to PacLife's expert Adelson: ¶¶ 23-25, 49, 107. The following paragraphs contain objectionable material that should be excluded in part or in whole: ¶¶ 15iii, 28 (2d sentence), 33-42, 45-48, 54-64, 103 (last sentence), 109, 111, 120, 140. Except for where the Court has indicated that a specific sentence is not admissible, the Court expects PacLife to parse the other paragraphs to identify objectionable material.

6 In defending Hillcoat's testimony, BNYM points out that one paragraph of Hillcoat's "no duty" testimony merely rebuts PacLife expert Adelson's opinion about a trustee's responsibilities. (BNYM Opp. at 2.) BNYM distorts what Adelson said. He does not opine that "the trustee ... [is] responsible for determining the materiality of missing mortgage loan documents." (BNYM Opp. at 2.) Rather, he opines that "market participants would expect the trustee of an MBS transaction to be the party responsible." (Adelson Report ¶ 236.)

PacLife also argues for exclusion of the section of Hillcoat's report titled "Limited Role of the RMBS Trustee," comprised of paragraphs 34-66. PacLife refers to that section as an improper and incompetent "legal-history lecture." (Hillcoat Reply App. A at 2.) Most of that section is comprised of paragraphs already excluded for containing prescriptive, legal opinion. The remaining paragraphs largely address trustee custom and practice. Accordingly, the Court will not exclude the section in its entirety.

3. State-of-Mind

PacLife challenges numerous portions of Hillcoat's opinions as improperly speculating about what others "understood" and looked at. (PacLife Mem. at 6; Hillcoat Reply App. A at 2.) Paclife is correct that expert opinion about what someone thought is improper. See, e.g., Board of Trustees of AFTRA Retirement Fund v. JPMorgan Chase Bank, N.A., No. 09-CV-686, 2011 WL 6288415, at *8 (S.D.N.Y. Dec. 15, 2011) ("opinions concerning state of mind are an inappropriate topic for expert opinion"). But as Judge Caproni recognized in Phoenix Light, an expert may testify as to broad industry understandings and expectations. Phoenix Light Tr. at 33 (testimony about investor expectations generally and how they understood events in the RMBS market "is not impermissible state-of-mind testimony about beliefs, intents or motives"); see also Matter of Trusts Established Under Pooling and Servicing Agreements Relating to Wachovia Bank Commercial Mortgage Trust Commercial Mortgage Pass-Through Certificates, Series 2007-C30, No. 17-CV-1998, 2020 WL 1304400, at *52 (S.D.N.Y. March 19, 2020) (Failla, J.) (explaining that cases precluding expert testimony about someone's state of mind are irrelevant to expert testimony about typical understanding in an industry). Hillcoat's opinions of what others "understood" is of the same kind as PacLife's expert Adelson's opinions of what various market constituents - almost everyone except for trustees - "expected." As the old saw goes, what's good for the goose is good for the gander. Accordingly, the portions of Hillcoat's report challenged based on state-of-mind will not be excluded.

4. Fact Narratives

Finally, just as BNYM did with respect to certain PacLife expert testimony, PacLife seeks to exclude Hillcoat testimony that it characterizes as "improper factual narrative or advocacy." (PacLife Mem. at 5.) The specific paragraphs targeted by PacLife do not fit that description, at least not as set forth by Hillcoat. Rather, they are fair and appropriate rebuttals to Adelson's opinions pointing out what Adelson allegedly did not account for, ignored or mischaracterized. (Hillcoat Report ¶¶ 114-117, 128, 132.) That testimony will not be excluded.

B. John J. Richard (Investor Expectations)

Richard offers opinions about investor expectations, and, like Hillcoat, also offers rebuttal to PacLife expert Adelson's opinions. PacLife's challenges to Richard's opinions are thus similar to its challenges to Adelson's opinions, and the Court's rulings are similar as well.

1. Relevance of Investor Expectations

PacLife seeks to exclude the entirety of Richard's opinions on the basis that his testimony about investor expectations is irrelevant and unreliable. PacLife's request strikes the Court as disingenuous given that PacLife put forward expert Adelson to opine on the same subject. Richard may testify about investor expectations for the same reasons that Adelson may do so. See Phoenix Light Tr. at 33 (denying plaintiff's motion to exclude Richard's testimony about investor expectations).

2. Legal Opinion and Interpretation

As with Adelson and Hillcoat, Richard may not opine on what duties trustees legally have or what they "should" do. The following paragraphs targeted by PacLife for exclusion are not objectionable and will not be excluded on that basis: Richard Report ¶ 39; Richard Rubuttal ¶¶ 5, 12, 14, 15, 33. Paragraph 15 includes a legal assumption from counsel, and PacLife has not, at least on the instant motion, challenged that assumption. The other paragraphs either do not contain legal opinion or interpretation, or are fair rebuttal to Adelson's opinions and references to documents.

However, paragraphs 34-37, which address how investors "generally understood" the No-Default R&W will be excluded as they address investor expectations without grounding those expectations in any particular experience. See Phoenix Light II, 2020 WL 1322856 at *7 (excluding opinion that expert failed to connect to his experiences). Rather, the opinions are entirely conclusory and nothing more than a conduit for legal interpretation. See United States v. Bilzerian, 926 F.2d 1285, 1295 (2d Cir. 1991) ("testimony encompassing an ultimate legal conclusion based upon the facts of the case is not admissible, and may not be made so simply because it is presented in terms of industry practice").

3. State of Mind

As with Hillcoat and PacLife expert Adelson, Richard's testimony about investor expectations, which directly rebuts Adelson's opinions on the same subject, will not be excluded (other than paragraphs 34-37 as explained above).

4. Fact Narratives

The two paragraphs (16-17) in Richard's Rebuttal report that PacLife contends are impermissible factual narratives are short and appropriate rebuttals to opinions from Adelson. They will not be excluded.

C. Richard Johns (Regulation AB)

Johns offers rebuttal opinions to those of PacLife experts Adelson and Beckles regarding Regulation AB, which sets forth SEC rules regarding registration, disclosure, and reporting requirements for publicly registered asset-backed securities, including RMBS. See generally 17 C.F.R. § 229.1100 et seq. PacLife seeks to exclude Johns's opinions in their entirety.

1. Unverifiable, Untestable, and Unreliable

PacLife contends that all of Johns's opinions are inadmissible as unreliable because they are unverifiable and untestable. PacLife asserts that Johns purports to know the "typical industry view" of every market participant and that the only support he can cite to is unspecified experience and unidentified industry participants. (PacLife Mem. at 8.) To some extent those criticisms are justified. At deposition, Johns admitted that he did not know how to objectively verify that there is an industry understanding that there is "no tie between Reg AB disclosures and EOD triggers." (Johns Dep. at 70) He testified that he believes various industry white papers, comment letters, and the like support his views regarding "industry-consensus opinion," but he knows of nothing objective beyond that. (Johns Dep. at 152).

That said, Johns is qualified to speak from an industry viewpoint given his experience. For example, from 2013-2018, he led the trade association known as the Structured Finance Industry Group. (Johns Report ¶ 3.) He has almost 25 years in the securitization industry and took a "lead role in building industry consensus across both the interpretation and the implementation" of the newly issued Regulation AB. (Johns Report ¶¶ 3, 7.) His opinions are based to a large degree on his experience building consensus among industry participants.

Federal Rule of Evidence 702 "'specifically contemplates that expert testimony may be based upon experience.'" Phoenix Light II, 2020 WL 1322856 at *6 (quoting Pension Committee of University of Montreal Pension Plan v. Bank of America Securities, LLC, 691 F. Supp. 2d 448, 473 (S.D.N.Y. 2010)). To be admissible, "the witness must show why that experience provides a reliable foundation for his or her testimony." Id. (internal quotation marks omitted). Johns did exactly that at his deposition in response to specific questions asking him what experience served as a basis for particular opinions. (See Johns Dep. 10-16, 21-26, 29-35.) Whether he did so convincingly can be tested on cross-examination. His testimony will not be excluded.

2. Conflict with SEC Statements

PacLife argues that Johns's opinions are inadmissible in their entirety because Johns expressed views that either "ignore" or "contradict" the SEC's statements. (PacLife Mem. at 8-9.) PacLife identifies five instances of deposition testimony that it claims demonstrate that flaw. The Court does not agree. First, none of the items alone or together would merit excluding all of Johns's opinions; none are identified as being the fulcrum of everything he says. Second, PacLife either mischaracterizes or exaggerates snippets of Johns's deposition testimony to give the impression of contradiction or lack of knowledge. Third, the points made by PacLife go to weight, not admissibility, and readily can be addressed on cross-examination.

3. Legal Opinions and Interpretations

As with its challenges to several other experts, PacLife requests exclusion of Johns's opinions about what duties a trustee legally has or should have or that otherwise contain or equate to legal opinion. And, as with its rulings on the other experts, the Court will exclude testimony that oversteps these boundaries. Of the specific paragraphs in Johns's report identified by PacLife, paragraphs 26 ("trustees generally did not have a duty"), 36-38 (legal interpretation), and 59 (whether or not trustee duties are implied or explicit) will be excluded. The other paragraphs will not.

D. Charles H. Grice (Reunderwriting)

Grice is BNYM's reunderwriting expert. He essentially rebuts Bitner's review of loans at issue. PacLife moves to exclude Grice's opinions in two respects.

1. Rejection of Post-Origination Information

In assessing to what extent loans in the trusts manifest material breaches, Bitner considered information that was not available at the time the loans were originated. Examples include post-origination tax returns and bankruptcy filings. In rebuttal, Grice rejects the practice of considering post-origination material to reunderwrite loans. PacLife argues that all of Grice's opinions should be excluded because he did not consider post-origination materials. In support of its motion, PacLife cites cases in which trustees and courts accepted evaluation of loan breaches based on post-origination information. (PacLife Mem. at 10-11.)

Putting aside that much of Grice's report does not implicate the post-origination issue, the issue is one on which reasonable experts can disagree. Although Grice does not support his opinion with any particular source or set of facts, he does offer a rationale for why he does not endorse the practice; namely, that it does not reflect the state of the underwriter's knowledge at the time of origination. (Grice Report ¶¶ 104-106.) Whether that rationale stands up to scrutiny before the jury is a matter of weight, not admissibility. PacLife, and expert Bitner, no doubt will put Grice's opinion to the test.⁷

7 PacLife cites to the New York Supreme Court's recent decision in MBIA Insurance Corp., 2020 WL 7041787 at *5-6, where the court recognized the relevance of post-origination information to determining whether borrowers made material representations with respect to information appearing on the Loan Tape. The Court rested that part of its decision on other cases but also because "Credit Suisse relied on [post-closing information] in the ordinary course of its business, including when it demanded repurchase from originators." Id. at *10. The court's reliance on Credit Suisse's own practice demonstrates the factual nature of the inquiry, not the inadmissibility of Grice's opinion, even if it turns out to be one that is not persuasive and ultimately rejected.

2. Meaning of No-Default R&W

The second aspect of Grice's opinions challenged by PacLife is his opinion that the PSAs' No-Default R&W is limited to payment defaults and does not include misrepresentations made by borrowers during the mortgage origination process. (Grice Report ¶¶ 103, 107-112.)

In his report, Grice expressly disavows having any opinion of the correct legal interpretation of the No-Default R&W. (Grice Report ¶ 109.) He then proceeds, however, to opine, based on his experience, what a "typical industry participant during the relevant period would have understood the representation to mean." (Grice Report ¶ 109.) He thus adopts the mantra that experts on both sides in this case do, which is to opine based on his experience about industry custom and practice, understandings, and expectations. As with his opinion regarding post-origination information, Grice does not base his opinion on much, but he does offer a rationale tied to specific industry experience. According to Grice, if participants wanted to protect against borrower misrepresentations - for which post-origination information may be relevant - they could and, sometimes did, bargain for a separate "no fraud" R&W. (Grice Report ¶ 109.) In Grice's view, Bitner applies the No-Default R&W as if it were a No-Fraud R&W, which the relevant transaction documents do not include.

Whether that opinion is relevant and admissible, however, likely will depend on whether the Court determines the No-Default R&W to be unambiguous. The opportunity to address that question likely will arise on summary judgment. But absent a determination that the No-Default R&W is unambiguous, Grice may offer an opinion as to industry understanding or application of the No-Default R&W.⁸ See MBIA Insurance Corp. v. Credit Suisse Securities (USA) LLC, 165 A.D.3d 108, 115, 84 N.Y.S.2d 157, 162 (1st Dep't 2018) (holding that the trial court erred in interpreting the No-Default R&W as a matter of law to include borrower misrepresentation); Ambac Assurance Corp. v. Countrywide Home Loans, Inc., 151 A.D.3d 83, 89, 56 N.Y.S.3d 21 (1st Dep't 2017) (explaining in regard to No-Default R&W that "the better course is to hold a trial to inquire into and develop the facts to clarify the relevant legal principles and their application to [the] representations and warranties" (internal quotation marks and alteration omitted)); MBIA Insurance Corp., 2020 WL 7041787 at *5-6 (after trial, rejecting bank's argument that No-Default R&W did not include non-monetary defaults).

8 The Court recognizes that Grice's opinion about the No-Default R&W as compared to a No-Fraud R&W is not dissimilar from Richard's opinion, which the Court excluded above. Richard's opinion, however is even less grounded in experience on that subject than is Grice's, and Richards engages in speculation when he says that investors generally were not aware of the No-Default R&W but that "even if investors were aware ... investors would not have (Richards Report ¶ 34.) Grice's opinion does not suffer from similar defects.

E. Marcel Bryar (Servicing)

Bryar, BNYM's servicing expert, primarily rebuts PacLife expert Beckles's opinions and opines on servicing industry custom, practice, and standards. PacLife seeks to exclude five aspects of Bryar's opinions.

1. Prudent Servicing Standard

PacLife contends that Bryar substitutes a "subjective" servicing standard for an objective standard recognized by Beckles. Not so. Bryar disagrees with Beckles that GSE guidelines are the industry standard for private non-GSE loans. That is fair ground for disagreement. On cross-examination, PacLife can question Bryar on the extent to which the standards he endorses are objectively prudent. Bryar's opinions on the applicable servicing standards are admissible.

2. What Servicers Think or Believe

In keeping with its challenges to other of BNYM's experts, PacLife argues that Bryar's opinions about what servicing professionals think or believe is inadmissible. Again, that is true to a certain extent but not to the extent that it precludes testimony of industry custom or practice. Bryar may not testify to what other individuals "thought" or "believed," but he may testify to what servicers and industry professionals typically did or did not do and what they typically would have expected or understood. See Matter of Series 2007-C30, 2020 WL 1304400 at *52; Phoenix Light Tr. at 33.

3. Materiality of Document Defects

PacLife asks to exclude Bryar's opinions that tens of thousands of document defects are immaterial because "the plain terms of the PSAs do not require that a Document Defect be material to trigger repurchase obligations." (PacLife Mem. at 13.) That is another instance where relevance of the opinions turns on contractual interpretation. Bryar's opinions will be admissible unless the Court determines on summary judgment or otherwise that the contractual language unambiguously resolves the issue. If the issue is not so resolved, then Bryar's opinions are direct and appropriate rebuttal to Beckles's assessment of the materiality of document defects.

4. Opinions About Evidence of Cures

PacLife faults Bryar for impermissibly opining on the "credibility" of evidence. The relevant parts of Bryar's report address the question of to what extent document defects may have been cured. (Bryar Report ¶¶ 170-173.) Bryar merely rebuts Beckles's opinion that questioned the reliability of certain cure evidence. (See Beckles Report ¶ 149.) To the extent Beckles offers opinions about the reliability of BNYM information, Bryar may offer a counter opinion.

PacLife also challenges Bryar's finding fault with Beckles for failing to obtain additional evidence of cures even though BNYM did not produce cure reports as requested by PacLife. Bryar asks rhetorically "why did Ms. Beckles not seek additional information regarding the significance of the cure records?" (Bryar Report ¶ 170.) As set forth above in regard to BNYM's application to exclude Beckles's testimony, that criticism is inadmissible. Accordingly, Bryar may not opine about Beckles's failure to obtain additional cure information. Bryar may, however, rebut Beckles's consideration of the cure evidence to which Beckles had access.

5. Servicer Duties vis-à-vis Modified Loans

Finally, PacLife seeks to exclude opinions that it characterizes as legal conclusions about what duties servicers did or did not have and about the PSA's provisions regarding repurchase of modified loans. The opinions are ones that Bryar offers in rebuttal to PacLife expert Beckles's opinion, including that the master servicer "was required to repurchase all modified loans." (Beckles Report ¶ 181.) Bryar disagrees with Beckles's statement as inconsistent with how mortgage servicers would understand the relevant contractual provisions. (Bryar Report ¶¶ 259, 261.) Bryar then sets forth four factors as the bases for his counter-opinion. (Bryar Report ¶¶ 262-268.) As a direct rebuttal to Beckles's understanding tied to specific factors beyond Bryar's mere say-so, Bryar's opinion is appropriate and admissible. In the event, however, that the relevant contractual language is addressed and determined to be unambiguous on summary judgment or otherwise, then neither Beckles nor Bryar may offer opinions as to what would be required or expected as to repurchase of modified loans.

F. Philip B. Stark (Statistics)

As a statistician, Stark offers opinions rebutting certain opinions of PacLife experts Beckles, Bitner, and Spencer. PacLife challenges the admissibility of Stark's opinions in all three respects.

1. Rebuttal to Beckles Analysis of Foreclosure Delay

PacLife seeks to exclude Stark's opinion that Beckles "does not establish any causal relationship between uncured document exceptions and the length of the foreclosure process." (See PacLife Mem. at 15 (citing Stark Report ¶ 45).) Stark comes to that conclusion based on his opinion that Beckles's analysis may show an association between document exceptions and length of the foreclosure process, but it does not establish a causal relationship. (Stark Report ¶ 45.)

PacLife asserts that Stark's causation opinion is irrelevant as contrary to the PSAs, which, according to PacLife, do not require that a document defect cause foreclosure delays to be subject to repurchase. That argument, however, ignores that PacLife expert Beckles suggests a cause-and-effect relationship. (See Beckles Report ¶ 135(vi) ("an incomplete or defective Mortgage File can result in many negative consequences," including "prevent[ing] or delay[ing] a foreclosure proceeding").) Stark merely opines on the statistical limits of that opinion. In Phoenix Light, plaintiffs' counsel made the same challenge to Stark's rebuttal of the same opinion from Beckles, and Judge Caproni found that "[i]f Beckles testifies that there is a correlation between foreclosure delays and documentation defects, Stark's testimony will be relevant and helpful to a jury to understand that correlation is not causation." Phoenix Light Tr. at 75-76. PacLife has provided no good reason for a different conclusion here.

PacLife also faults Stark for not analyzing the data on which Beckles's opinion is based in that Stark based his opinion on the number of purportedly uncured document defects as of April 2011 rather than the number on BNYM's final exception reports. (PacLife Mem. at 15.) At deposition, however, Stark explained why he thought the conclusion would be the same regardless. (Stark Dep. at 53, 59.) That particular criticism therefore goes to weight, not admissibility, and is not a basis for exclusion.

2. Rebuttal to Bitner

Stark opines that the 1,147 loans that Bitner analyzed are not representative of other loans in the trusts, and a breach rate therefore cannot be extrapolated to them. (Stark Report ¶ 33.) PacLife says that opinion is irrelevant because Bitner did not opine that the loans he analyzed were a representative sample that could be extrapolated. Nonetheless, Stark's opinion is a fair rebuttal to what a reasonable juror might otherwise infer from Bitner's opinions.

Stark also opines that, for varied reasons, Bitner should not have used the 90th percentile income for professions on the BLS database when determining reasonableness of a borrower's stated income. (Stark Report ¶¶ 26-32, 72-89.) Stark asserts that there are statistical fallacies and many uncertainties about BLS data and Bitner's use of it. PacLife seeks to exclude Stark's opinions on this subject on the basis that Bitner did not use the BLS data in the manner that Stark claims and that Stark is not qualified to testify on use of BLS data in mortgage reunderwriting, which Stark readily admits. (Stark Dep. at 21.) Bitner appears to have used BLS data in reunderwriting primarily to assess, along with other data, whether borrowers misrepresented their income. (See, e.g., Bitner Report ¶¶ 97 n.4, 117 n.5, 133.) Nothing in what PacLife has presented suggests that Stark's opinions about BLS data do not apply when used for that purpose. And, while there is no dispute that Stark has no experience in underwriting, he is qualified to offer opinions about statistical flaws in the data or statistically flawed assumptions in using the data. Accordingly, Bitner's opinions about the BLS data will not be excluded. See Phoenix Light Tr. 77 (making same determination regarding attempt to exclude Stark's opinions of BLS data used by Bitner).

3. Rebuttal to Spencer

PacLife asks to exclude Stark's opinions criticizing PacLife expert Spencer's use of GSE loan data for "matching" with at-issue loans to determine differential loss severity. PacLife argues that Stark's criticisms are incompetent because they go to servicing issues that are outside his expertise, and PacLife servicing expert Beckles selected the loan groups that Spencer used for his analyses based on variables that Beckles determined are significant. That may be so, but the opinions Stark offers go to statistical issues. See Phoenix Light II, 2020 WL 1322856 at *10 (Plaintiffs' position "would, in essence, disqualify all [statisticians] from conducting a [statistical] assessment unless they first acquire 'expertise' in the specific industry in which they purport to opine" (internal quotation marks omitted, first alteration in original). Stark's opinions about Spencer's analyses will not be excluded. PacLife will be able to challenge Stark's opinions as inapt through examination of Beckles and Spencer, and cross-examination of Stark.

G. Faten Sabry (Damages)

Sabry primarily offers rebuttal opinions about damages. While not criticizing Milner's basic damages model, Sabry opines on several topics that affect what goes into the model.

1. Loan Performance

Sabry analyzed performance of loans with R&W breaches as compared to loans without breaches. Her opinions in that respect rebut Beckles's and Bitner's opinions about loan performance. (Sabry Report ¶ 75 (citing Beckles Report ¶ 135), ¶ 81 (citing Bitner Report ¶ 76).) PacLife argues that Sabry's opinions are irrelevant and unreliable because they "ignore the operative standard." (PacLife Mem. at 17.) Specifically, PacLife contends that the PSAs require repurchase of loans with document defects without regard to actual loss or materiality, and because materiality is demonstrated by increased risk of loss, rather than actual loss. The first proposition has not yet been determined as a matter of law, and therefore the relevance of opinions about materiality of document defects cannot yet be determined. But unless the matter is resolved as a matter of law in PacLife's favor, Sabry's rebuttal remains relevant and admissible.

The second proposition - that materiality is based on increased risk of loss, not actual loss - is supported by case law. See Assured Guaranty Municipal Corp. v. Flagstar Bank, FSB, 920 F.Supp. 2d 475, 511 (S.D.N.Y. 2013); Home Equity Mortgage Trust Series 2006-1 v. DLJ Mortgage Capital, Inc., 175 A.D.3d 1175, 1177, 109 N.Y.S.3d 231, 233 (1st Dep't 2019). But that does not render Sabry's opinions either irrelevant or unreliable. Even if actual loan performance is not the "standard" for determining materiality of defects, loan performance may still be probative of whether, in retrospect, there was an increased risk of loss. Sabry's opinions in that regard therefore are admissible.

2. Loan Repurchase Rates

Sabry faults PacLife expert Milner for assuming that 100% of loans put back for repurchase are actually repurchased. Sabry bases her opinion on repurchase data from so-called ABS-15G Reports filed with the SEC ("15G Reports"). According to Sabry's analysis, only 9.5% of loans put back for repurchase were successful. (Sabry Report ¶ 88.) PacLife seeks to exclude Sabry's opinions about loan repurchase rates, arguing that the 15G Reports do not concern document defect loans or modified loans and also that the data is incomplete.

The first objection has merit, although it largely is a strawman. Sabry expressly connected her repurchase rate analysis directly to R&W repurchase demands. (Sabry Report ¶¶ 87-88.) To the extent the opinion is offered with respect to other types of repurchase demands, particularly those based on document defects or modifications, the opinions are not relevant and will be excluded. See Phoenix Light Tr. 82 (ruling that Sabry may not opine about repurchase rates for document defect loans based on 15G Reports).

The second objection is more apt but has less merit. PacLife asserts that the 15G Report data is incomplete because it does not indicate the source of repurchase demands, reasons that requests were withdrawn or rejected, how aggressively repurchases were pursued, or why requests may otherwise have failed. Those issues go to weight, not admissibility, and are fodder for cross-examination.

3. Servicing Opinions

Sabry rebuts Milner's extrapolation of PacLife expert Spencer's loss severity differential calculations to unmatched loans, opining that the unmatched loans had higher balances and therefore fixed costs will not contribute as meaningfully to loss severity. (Sabry Report ¶¶ 117-123.) Paclife challenges Sabry's rebuttal opinion as unreliable for multiple reasons. First, PacLife contends that Sabry "conflates level of loss severity ... with the loss-severity differential between poorly serviced Trust loans and better serviced GSE loans." (PacLife Mem. at 20.) That criticism has no merit. Sabry expressly addressed the relationship between loss severity and its impact on loss severity differential. (Sabry Report ¶ 122.) Whether she has a sound basis to do so can be tested on cross-examination.

PacLife next faults Sabry for not analyzing many other factors impacting loss severity. That may be, but the omission does not warrant exclusion. Sabry demonstrated a particular distinction between matched and unmatched loans influencing the analysis of loss severity differential. The limited nature of that analysis goes to weight, not admissibility.

Sabry opines that Spencer should have considered loan product type as a factor in his differential analysis. (Sabry Report ¶¶ 108-116.) PacLife contends that opinion is unreliable because her assertion that "borrowers self-select into non-traditional mortgages based on unobserved risk attributes, such as greater risk tolerance" is unsupported by any evidence or independent analysis. (PacLife Mem. at 20 (quoting Sabry Report ¶ 111).) As BNYM points out, however, that simply is not so. Sabry analyzed the effect of loan product type using Spencer's model (Sabry Report ¶¶ 108, 112, 113, 115), and relied on third-party sources (e.g., Sabry Report ¶ 111).

Accordingly, Sabry's rebuttal opinions concerning Spencer's loss severity differential analysis are not excluded.

4. Countrywide Settlement

Sabry opines about the way in which PacLife expert Milner accounted for the Countrywide Settlement, discussed above in the context of BNYM's motion to exclude Milner's opinion. As explained there, Milner's treatment of the Countrywide Settlement is flawed. Sabry explains that flaw and recalculates damages accounting for Countrywide Settlement payments in cash flows entering into the distribution waterfall. That is classic rebuttal and perfectly appropriate. PacLife contends that Sabry's analysis nevertheless is unreliable because "she does not know if Plaintiffs received these payments," which could impact her calculations. (PacLife Mem. at 20 (citing Sabry Dep at 51-52).) That is disingenuous - Milner himself acknowledged that Plaintiffs had received proceeds from the settlement. (Milner Dep. at 131.) To the extent there are any settlement proceeds that should not be fed into the but-for world waterfall, PacLife can address that on cross-examination, and damages can be recalculated. In any event, Sabry's opinions about the relationship between the Countrywide Settlement and damages is admissible.

5. Pre-Judgment Interest

Lastly, PacLife challenges Sabry's rebuttal to Milner's calculation of pre-judgment interest. PacLife argues that Sabry "used the wrong measure," offers a legal opinion, and that her opinion is "unnecessary." (PacLife Mem. at 20.) As explained earlier, the parties and their experts dispute whether interest should be calculated on principal and interest separately or together. That is an issue for the Court to determine at a later time based, in part, on testimony from the experts to be taken out of the presence of the jury. Sabry's rebuttal opinion about pre-judgment interest is no more or less admissible than Milner's.

H. Thomas Z. Lys (Countrywide Wherewithal)

Lys, a specialist in financial economics working with Analysis Group, provides opinions about the extent to which Countrywide had the wherewithal to repurchase loans and how that may affect a damages analysis. The crux of his opinions is that if, in the but-for world, PacLife had put back the loans at issue for repurchase, then so would certificate holders of hundreds of other Countrywide trusts resulting in such a financial strain on Countrywide that it would not have been able to pay all repurchase claims and may have gone into bankruptcy. (Lys Report at ¶¶ 51-53.)

The Court finds that Lys's opinions are unreliable, speculative, indefinite, and unhelpful to the jury.

At deposition, Lys betrayed a considerable lack of familiarity with his report. Lys conceded that he could not identify in his report anything specific that he wrote; could not recall how much time he spent editing the report or any details of directions he gave for its preparation; could not identify which Analysis Group individuals performed what tasks or even describe the supervision he purportedly provided other than to say "they did what I told them to," which he could not specify; and "rel[ied] on their hours to collect data, to analyze data and to, you know, produce the product." (Lys Dep. at 21-25, 28-29.) Lys also betrayed a lack of understanding as to why certain statements were included in the report. (Lys Dep. ¶¶ 46, 63, 152-53.) BNYM counters that Lys reviewed the entire report and testified that "whatever was typed in by others, I reviewed, edited or rejected" and "all the words are mine." (Lys Dep. at 22.) But that does not by itself earn the price of admission. See, e.g., Bank of New York Mellon v. WMC Mortgage, LLC, No. 12-CV-7096, 2015 WL 4887446, at *6, 8 (S.D.N.Y. Aug. 17, 2015) (excluding expert opinions on reunderwriting because the expert was unaware of the process and degree of care taken by the reunderwriters). Lys's lack of involvement and familiarity with his own report renders its reliability highly suspect.

Second, BNYM gave Lys instructions and assumptions that artificially constrain the nature of the inquiry. The entity that is subject to potential liability here is Countrywide Home Loans, Inc. ("CHL"). Lys did not review the financial statements of CHL. Rather, Lys analyzed the financial statements and condition of CHL's parent entity, Countrywide Financial Corp. ("CFC") because that is what BNYM instructed him to do. Lys, however, did not analyze the financial statements of CFC's parent entity, Bank of America ("BOA"), because BNYM instructed him to look only at CFC's wherewithal as a stand-alone entity. The purpose for which Lys's opinion is being offered is to undermine PacLife's damages analysis by demonstrating that in the but-for world, PacLife's putback demands would not have been satisfied due to lack of financial wherewithal. Merely evaluating CFC's wherewithal alone does not answer that question.

For instance, Lys acknowledges that even if not legally obligated, BOA "contributed to a number of settlements related to legacy Countrywide mortgage issues." (Lys Report ¶ 55 n.90.) In an effort to cast that fact aside, Lys opines "that does not mean [BOA] would continue to contribute to additional claims, especially if they were significant." (Lys Report ¶ 55 n.90.) Of course, BOA's contributions to settlements in other cases also does not mean that BOA would not have funded additional claims, let alone the claim at issue in this lawsuit. Lys's opinion is mere speculation.⁹

9 PacLife's expert Milner renders an opinion about BOA's wherewithal to fund repurchases arising from this case. (Milner Report ¶¶ 108-110.) As noted, Lys does not opine about BOA's ability to pay and does not rebut Milner in that respect. Whether and to what extent BOA would have funded the repurchases is a separate question about which Lys's opinion is, as noted, entirely speculative. And notwithstanding exclusion of Lys's opinions, BNYM may properly challenge PacLife's repurchase rate assumptions with data, insofar as it goes, compiled and opined on by BNYM damages expert Sabry. (See Sabry Report ¶¶ 84-90.)

Third, Lys does not provide any opinion of the extent to which Countrywide's limited resources would impact damages in this case. He only vaguely opines that "Countrywide's ability to pay the alleged put-back claims related to the At-Issue Trusts as a stand-alone entity is a matter of real economic significance that must be considered in assessing Plaintiffs' damages." (Lys Report ¶ 55.) To offer an opinion that Countrywide's wherewithal or lack thereof is significant without any quantification of the impact on damages gives the jury no guidance and invites wild speculation.¹⁰ See United States v. McAllister, No. 96-1292, 1996 WL 730502, at *2 (2d Cir. Dec. 19, 1996) (unpublished opinion) ("It is also clear that a district court has discretion to exclude evidence that would require the jury to speculate"); Securities and Exchange Commission v. Lek Securities Corp., 370 F. Supp. 3d 384, 412 (S.D.N.Y. 2019) (excluding portion of expert testimony in part because the expert "improperly invites the jury to speculate"). As such, the opinion will not be helpful to the jury.

10 The report of BNYM's damages expert Sabry also does not quantify the impact on damages that would be attributed to Countrywide's wherewithal. Rather, Sabry addresses "the possibility that Countrywide and its successor, even if these entities were solvent, may not have agreed to repurchase all the submitted demands or that claims pursued in litigation may have failed." (Sabry Report ¶ 84.)

Finally, Lys's opinion facilitates an argument that PacLife should not be able to recover damages for BNYM's alleged misconduct because so many other investors were similarly victimized that in the but-for world there would have been an unknown amount left to pay repurchase claims for the trusts at issue. That is too clever by half. Such a proposition not only is ill defined but if embraced would incentivize actors to maximize, not minimize, their breaches.

Lys's opinions should be excluded in their entirety.¹¹

11 In Phoenix Light, Judge Caproni admitted portions of Lys's opinions. (Phoenix Light Tr. at 21-25.) But the issues she addressed are not quite the same as those presented here. For instance, in Phoenix Light, plaintiffs sought to exclude Lys's testimony that BOA would not have funded the entirety of Countrywide's liability. Lys does not offer such an opinion here other than the speculative statement, for which the obverse is just as plausible, that just because BOA had contributed with respect to some claims, does not mean it would do so with respect to other claims. In Phoenix Light, plaintiffs unsuccessfully sought to exclude as unreliable Lys's testimony that Countrywide would have faced other litigation and claims that would have limited its ability to pay. Here, the Court does not fault Lys for his assumptions in that regard, but rather finds it speculative and unhelpful because Lys fails to quantify the impact of that fact on PacLife's potential recovery. And, opinions such as "everyone acts in there self-interest," which were at issue in Phoenix Light, are not the basis of PacLife's motion or the Court's decision.

Conclusion

The parties' motions to exclude expert testimony are GRANTED in part and DENIED in part as set forth in the foregoing opinion.

The Clerk of Court is respectfully requested to terminate the motions at Dkts. 150, 152, 154, 156, 158, 161.

SO ORDERED,

/s/_________

ROBERT W. LEHRBURGER

UNITED STATES MAGISTRATE JUDGE Dated: February 22, 2021

New York, New York