Opinion
Civil Action No. 3:12-cv-1538 (CSH)
12-30-2015
LAW DEBENTURE TRUST COMPANY OF NEW YORK, solely in its capacity as Separate Trustee of the Securitized Asset Backed Receivables LLC Trust 2006-WM2, Plaintiff, v. WMC MORTGAGE, LLC, f/k/a WMC MORTGAGE CORP., Defendant.
RULING ON DEFENDANT'S MOTION TO EXCLUDE OR , IN THE ALTERNATIVE, LIMIT TESTIMONY OF NELSON R. LIPSHUTZ, Ph.D., AN EXPERT WITNESS HAIGHT, Senior District Judge :
This is a residential mortgage backed securities case, one of a current legion of such cases generically referred to as an "RMBS" case.
Plaintiff Law Debenture Trust Company of New York ("Law Debenture") is the trustee of an RMBS trust. Defendant WMC Mortgage, LLC (WMC") is the originator of the mortgage loans comprising that trust. WMC now moves pursuant to Rule 702 of the Federal Rules of Evidence to exclude or limit the opinion testimony of an expert witness proffered by Law Debenture. See Doc. 74. That sort of application is known colloquially as a Daubert motion, in tribute to the Supreme Court decision which inspired Rule 702's amendments into its present form. This Ruling resolves that motion.
Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579 (1993).
I
Familiarity is assumed with each of the Court's prior rulings during the course of this energetically conducted litigation. For present purposes, it is sufficient to state that the Plaintiff at bar, like many plaintiffs in RMBS cases, asserts "put-back breach of contract claims" against the originator of residential mortgages that were bundled together into a trust that then issued certificates purchased by the investing public.
RMBS investors anticipate they will share in a stream of cash generated by a universe of faithful and solvent home owners making their mortgage payments on time. That anticipation was dashed when an epidemic of failed and defaulted mortgages afflicted the economy. On behalf of investors, trustees of RMBS trusts sued the arrangers and sponsors of the trust to recover damages for the certificate holders. The case at bar is such a case. Stated briefly, the typical plaintiff in an RMBS case (plaintiff at bar is one of them) alleges that the defendant originator or sponsor breached representations and warranties contained in interlocking agreements controlling the securitization process which reassured investors as to loan characteristics concerning the likelihood of repayment of each mortgage. These representations and warranties were contained in, among other places, a Pooling and Service Agreement ("PSA") under which the bundled loans were transferred into the trust. The PSA provided that if the defendant was found to be in breach of these representations or warranties, it would cure the breach, repurchase the defective loan, or substitute a qualifying loan for the defective one.
The phrase "put-back breach of contract claims" reflects and summarizes the claims Law Debenture makes against WMC: specifically, that WMC committed pervasive breaches of representations and warranties throughout the mortgage loans contained in the trust, and thereafter breached the PSA contract's "put-back" obligation by failing to take any of the curative steps required by that contract with respect to any of the breaching loans.
II
This is all by way of background. The case now comes before the Court as Law Debenture, having survived WMC's motion to dismiss the complaint, has made a motion for an expedited determination of the validity of statistical sampling as a source of proof at the trial. That subject has generated brisk exchanges of opposing views by the parties and counsel.
See the Court's opinion in the companion case of Deutsche Bank National Trust Co. v. WMC Mortgage, LLC, No. 3:12-cv-00933, 2014 WL 1289234 (D. Conn. Mar. 31, 2014).
Specifically, the controversy began when counsel for Law Debenture proffered the expert opinion of Nelson R. Lipshutz, Ph.D. Dr. Lipshutz is a statistical consultant. Law Debenture identifies him as its expert in sampling. Dr. Lipshutz submitted a report setting forth the testimony he intended to present at the bench trial of this case. Counsel for WMC deposed Dr. Lipshutz about his report and intended testimony, and then attacked them in what may be characterized as a Daubert cross-motion.
Counsel for Law Debenture characterize their motion as "seeking an expedited determination of whether [Plaintiff] may establish the extent of Defendant [WMC's] liability and damages based on a statistically significant, random sample of the loans at issue." Brief [Doc. 61], at 1. The substance of Dr. Lipshutz's opinion and testimony has been usefully summarized in counsel's submissions. An early brief for Law Debenture refers to "Dr. Lipshutz's opinion regarding the use of statistical sampling to streamline this litigation." Doc. 77, at 1. WMC does not quarrel with this use of the verb "to streamline," and of course the proposition is obvious, if one gives that verb its dictionary definition of "to make simpler or more efficient." Webster's New Collegiate Dictionary (1976) at 1151. If one undertakes to prove a case involving a number of nonperforming mortgages by sampling and examining some (but not all) of them, those responsible for the litigation will have to read fewer mortgages than would be necessary if there was no sampling.
In its citations, the Court follows the original pagination of the parties' briefs, rather than the page numbers generated by filing these briefs as Court Documents.
In aid of this motion for expedited consideration, counsel for Law Debenture cite a number of cases for the proposition that
statistical sampling should be approved before discovery goes into full swing, so that the parties can tailor their discovery and other pretrial efforts to the approximately 400 loans in the [Plaintiff's] proposed sample, and avoid the much more costly and time-consuming task of re-underwriting and litigating thousands of loans. To maximize judicial economy and cost savings, sampling should be addressed now.Brief [Doc. 61], at 10. The procedure Law Debenture envisioned on this motion called for it to "submit its statistically significant sample of [m]ortgage [l]oans and supporting expert report at this early stage of the litigation," with the understanding that "[t]o the extent WMC opposes the sample's validity for establishing liability and damages in this litigation or seeks to disqualify the [Plaintiff's] expert on Daubert or other grounds, WMC shall file any such objections within 30 days, so this issue will be ripe for determination by the Court well in advance of trial." Id. at 10-11. To no one's surprise, WMC timely opposed the sample's validity for the stated purpose, its opposition taking the form of a Daubert cross-motion.
To turn from the preliminaries to the particulars of the issues: Counsel for Law Debenture say in their Court-directed offer of proof that "Dr. Lipshutz has devised a randomly selected, statistically significant sample of 400 loans out of 4,170 loans not paid in full." Doc. 89, ¶ 4. In its most recent brief, counsel reiterate that a ruling sustaining "the statistical validity" of Dr. Lipshutz's sample "will greatly aid the parties and the Court in streamlining this litigation" and then break down their intended trial presentation into a series of steps which I will quote, adding bracketed progressive step numbers for the sake of clarity:
The use of statistical sampling necessarily requires several steps. The first step is to [1] draw a representative sample of loans from the Trust. That is what Dr. Lipshutz has done at this stage. The second step is to [2] conduct a review of the loans in the sample to determine whether the loans are in breach of WMC's representations and warranties, which includes an analysis of whether the original loan files comply with WMC's own underwriting guidelines. That process will be far more efficient—for the Court and the parties—when focused on the sample of 400 loans, rather than thousands of mortgage loans, as WMC proposes. After the number of loans in breach has been identified through the loan file review process, Dr. Lipshutz will [3] mathematically determine the breach rate among the loans in the sample. Dr. Lipshutz will then be able to [4] extrapolate the breach rate among the remainder of the loans in the Trust to a reasonable degree of scientific certainty. Dr. Mason will then [5] calculate the damages using that breach rate.Doc. 105, at 4-5.
The "Dr. Mason" referred to in this passage is Joseph R. Mason, Ph.D., Law Debenture's proffered expert witness on the subject of damages. As the quotation in text reflects, Law Debenture's trial plan envisions Dr. Mason basing his calculation of damages caused by WMC's breaches upon the breach rate determined by Dr. Lipshutz through statistical sampling.
Law Debenture's game plan for the trial, then, is to call Dr. Lipshutz as a statistical expert whose testimony based on sampling will, if the trial judge accepts it, establish WMC's liability for breaches of contractual representations and warranties across the broad scope of nonperforming mortgages within the Trust. Dr. Mason will follow up as a damages expert whose damages calculations will proceed from the breach rate statistically established by Dr. Lipshutz.
WMC's counsel considered Dr. Lipshutz's written report, took his deposition, and then filed a Daubert cross-motion "to exclude, or in the alternative, limit the testimony" of Dr. Lipshutz at trial. Notice of Motion [Doc. 74], at 1. WMC's brief in support of this motion makes three contentions. Doc. 75, at 1-2. First: The PSA limits Law Debenture to the "sole remedy" of repurchase of a loan whose performance deficiencies cannot be cured, a limitation which precludes "money damages via sampling or otherwise" and "requires loan-by-loan proof of both breaches of representations and damages," and "Dr. Lipshutz has conceded that his sample cannot be used to provide such loan-by-loan proof." Id., at 1 & n1. Second: Dr. Lipshutz has conceded that his sample cannot be used "to estimate damages to any specific level of precision, and that any opinion he offered on that subject at this time would be based upon 'speculation.'" Id., at 1. Third: With respect to the use of Dr. Lipshutz's sample to estimate the breach rate (defined as the percentage of loans in the population as to which there is a material breach), "Dr. Lipshutz has presented no sound scientific basis supporting his choice of sample size or margin of error." Id., at 2.
Law Debenture rejected these contentions in a brief which argued that the "sole remedy" provision in the PSA did not preclude proof of liability or damages by sampling, and WMC's other contentions either did not fall within a Daubert motion at all or went to the weight of Dr. Lipshutz's opinion testimony rather than its admissibility (a Daubert analysis being confined to the latter consideration). Doc. 77.
The case came on for oral argument. Prior to the hearing, the Court entered a Scheduling Order which eliminated WMC's first stated objection to Dr. Lipshutz's report from the subjects to be considered as part of a Daubert hearing. Doc. 84. That Order said: "While an argument can be made for shoehorning the 'sole remedy' substantive merits issue into a Daubert inquiry into the validity of the methodology by which Dr. Lipshutz arrived at his statistical opinion, I think the better practice is to require WMC to make a separate, free-standing motion in limine if it wishes to press the sole remedy theory (as I suspect it does)." Id., at 5. This suspicion was well founded: WMC's most recent brief asserts that "[d]amages are not available under the contract's sole remedy clause." Doc. 104, at 1 n.1. That continuing assertion, to which Law Debenture expresses continuing disagreement, remains an open question for future adjudication.
III
These emerging issues implicate the proper interpretation and application of Rule 702 of the Federal Rules of Evidence, which was amended in 2000 to respond to the Supreme Court's 1993 decision in Daubert.
The Advisory Committee Notes to the 2000 amendments begin with the observation that "In Daubert the Court charged trial judges with the responsibility of acting as gatekeepers to exclude unreliable expert testimony." Daubert's most striking contribution to modern legal parlance, the casting of federal district judges as "gatekeepers" with its evocation of an earlier, gentler, more rustic time, was fashioned by Justice Blackmun in his majority opinion: "We recognize that, in practice, a gatekeeping role for the judge, no matter how flexible, inevitably on occasion will prevent the jury from learning of authentic insights and innovations," 509 U.S. at 597, a somewhat daunting prospect which led Justice Blackmun to reassure his readers with the additional observation: "We are confident that federal judges possess the capacity to undertake this review," id. at 593.
Rule 702 provides: "A witness who is qualified as an expert by knowledge, skill, experience, training, or education may testify in the form of an opinion or otherwise if:
(a) the expert's scientific, technical, or other specialized knowledge will help the trier of fact to understand the evidence or to determine a fact in issue;
(b) the testimony 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 of the case."
It is important to recognize what function Rule 702 performs and what it does not. The Rule is concerned with whether the proffered expert opinion is admissible, not whether it is correct, or persuasive to the factfinder. The Advisory Committee's extended Notes on the 2000 amendment emphasize that distinction:
[T]he Rule as amended provides that all types of expert testimony present questions of admissibility for the trial court in deciding whether the evidence is reliable and helpful. . . . Courts both before and after Daubert have found other factors relevant in determining whether expert testimony is sufficiently reliable to be considered by the trier of fact. . . . Vigorous cross-examination, presentation of contrary evidence, and careful instruction on the burden of proof are the traditional and appropriate means of attacking shaky but admissible evidence. . . . When a trial court, applying this amendment, rules that an expert's testimony is reliable, this does not necessarily mean that contradictory expert testimony is unreliable. The amendment is broad enough to permit testimony that is the product of competing principles or methods in the same field of expertise. . . . [P]roponents do not have to demonstrate to the judge by a preponderance of the evidence that the assessments of their experts are correct, they only have to demonstrate that their opinions are reliable. . . . The evidentiary requirement of reliability is lower than the merits standard of correctness. . . . Daubert neither requires nor empowers trial courts to determine which of several competing scientific theories has the best provenance.Fed. R. Evid. 702 advisory comm. nn. (emphases added, citations and internal quotation marks omitted).
The operation of these principles is illustrated by Justice Bransten's decision and order in SACO I Trust 2006-5 v. EMC Mortgage LLC, Index No. 651820/2012 (N.Y. Sup. Ct., N.Y. Cty. Dec. 2, 2015). The record of the hearing before Justice Bransten on November 4, 2015 shows that SACO I is an RMBS put-back case in which the PSA contractual provisions are the same as those in the case at bar and the plaintiff-trustees' action follows the same pattern of nonperforming mortgages. Id., NYSCEF Doc. No. 564 ("Tr."). There, the trustees moved for partial summary judgment seeking a ruling that "the use of statistical sampling to prove liability and damages on their claims is consistent with the terms of the contract governing the transactions, including but not limited to the PSAs." Tr., at 16. Justice Bransten granted that motion over the defendants' objection. Her oral opinion reflects the distinction, comparable to that noted by the federal Rule 702 Advisory Committee, between admitting statistical sampling into evidence at the trial and what factfinders may make of that evidence when they come to consider it. Justice Brantsen observed trenchantly that
In SACO I, three of the four plaintiffs were the master servicers of the trusts, rather than trustees. As this distinction carries no relevant distinction for present purposes, the Court refers to the SACO I plaintiffs collectively as "trustees."
there is no guaranty that the jury will accept the sampling or that it's a mechanism of proving the liability and/or damages in the case. So, I mean, we don't know. We still don't know what a jury is going to think about using sampling. So that is always the danger of doing it. But I certainly agree that you could do it. . . . Again, whether a jury will find this presentation persuasive is a separate and - is a separate matter.Tr., at 8, 15.
This analysis is useful, even though the case at bar will be tried to the Court rather than a jury. My function in ruling on WMC's Daubert motion is to decide if Dr. Lipshutz's proffered statistical sampling passes muster under Rule 702 and can accordingly be admitted into evidence at the trial. Whether at the end of a full plenary trial I find his presentation persuasive is a separate matter, to be decided at that future time. To someone without the advantage (or disadvantage) of a legal education, it may seem odd that a trial judge must read a document in order to decide if he is allowed to do so, but those decisional gymnastics are inherent in the nature of a bench trial, they come with the territory.
Thus it came to pass that District Judge Rakoff, that able and experienced jurist, had occasion during an RMBS bench trial to evaluate the expert opinion on statistical sampling offered by the same Dr. Lipshutz. See Assured Guar. Mun. Corp. v. Flagstar Bank, FSB, 920 F.Supp.2d 475 (S.D.N.Y. 2013). Judge Rakoff noted that "in accordance with the liberal admissibility standards of the Federal Rules of Evidence, only serious flaws in reasoning or methodology will warrant exclusion," and added: "This is especially true in the context of a bench trial, where there is no possibility of prejudice, and no need to protect the factfinder from being overawed by 'expert' analysis." Id. at 502 (citations and internal quotation marks omitted). I prefer to think of myself as almost as impervious to awe of experts as Judge Rakoff.
The admissibility of Dr. Lipshutz's proffered opinion and intended trial testimony on statistical sampling must now be addressed.
IV
RMBS cases present distinct difficulties, but the admissibility of statistical sampling opinion testimony such as that proffered by Dr. Lipshutz in the case at bar is not one of them. In Flagstar, Judge Rakoff held that Dr. Lipshutz's opinions satisfied the Rule 702 and Daubert conditions of reliability and admitted them as probative evidence at the trial. That ruling resonates in this case because Dr. Lipshutz's opinions and methodology in Flagstar and in this case are essentially the same.
Judge Rakoff's post-trial decision on the merits described Dr. Lipshutz's testimony in Flagstar: "Dr. Nelson Lipshutz, Assured's expert on statistical sampling, created a random sample of 400 loans from each of the two securitization pools, for a total of 800 loans." 920 F.Supp.2d at 486 (footnote omitted). "In determining the size of the random samples for the two Trusts," Dr. Lipshutz applied a formula whose objective was that "the estimate derived from the sample has a 95% chance of being within . . . plus or minus five percentage points of the actual population proportion, a standard measure for a high level of statistical confidence. In particular, Dr. Lipshutz designed the samples with the understanding that they would be used 'to make a binary decision on each loan,' that is, whether the loan file conforms to the representations and warranties made in the Transaction Documents, or not." Id. (citations to trial transcript and internal quotation marks omitted). "Dr. Lipshutz concluded that the random samples are representative of the total loan populations in both the 2005-1 and 2006-2 loan pools." Id. at 487.
In Flagstar, the random samples produced by Dr. Lipshutz were submitted to Rebecca Walzak, "Assured's mortgage origination and underwriting expert," who "concluded that 606 of the 800 loans in the sample contained material breaches of Flagstar's representations and warranties." Id. at 487. Dr. Joseph Mason, "Assured's damages expert," id. at 500, "extrapolated pool-wide" the samples produced by these other experts to calculate the amount of plaintiff's damages, id. at 503.
In the case at bar, Plaintiff Law Debenture appears to be modeling its trial strategy upon that of the successful plaintiff in Flagstar. Dr. Lipshutz, the same statistical sampling expert, has prepared what he opines to be a randomly selected, statistically significant sale of 400 loans out of 4,170 loans not paid in full. That is the present state of this case. On Law Debenture's approach, those 400 loans in the sample will thereafter be re-underwritten to determine the number of them (if any) where WMC's representations or warranties were breached; Dr. Lipshutz will then calculate a "breach rate" applicable to the entire pool; and Dr. Mason, the same damages expert, is waiting in the wings to calculate total damages on the basis of the breach rate that emerges. Law Debenture seeks by its present motion to obtain the Court's validation of Dr. Lipshutz's statistical sampling in the case. WMC seeks by its Daubert cross-motion to exclude Dr. Lipshutz's opinion and testimony on the point.
That is the key element in the case. In Flagstar, the opinion evidence on the existence of breaches of representations and warranties was supplied by the plaintiff's underwriting expert, Ms. Walzak, of whom Judge Rakoff said: "The most contested testimony in this case, however, was that of Ms. Walzak, without which the opinions of Dr. Lipshutz and Dr. Mason would be irrelevant," 920 F.Supp.2d at 504.
WMC fails to show that Dr. Lipshutz's opinion should be excluded at this stage of the litigation under Daubert or that decision's incarnation in Rule 702, Fed. R. Evid. In this Court's decision in Deutsche Bank, 2014 WL 1289234, at *12, I praised "Judge Rakoff's reasoned approval of sampling as a source of probative evidence in Flagstar," and quoted his ruling: "Sampling is a widely accepted method of proof in cases brought under New York law, including cases relating to RMBS and involving repurchase claims," id. (quoting Flagstar, 920 F.Supp.2d at 512). More recently, Justice Bransten reached the same conclusion in SACO I, stating that "the use of sampling is well-accepted by courts in RMBS cases," and "the overwhelming majority of courts with RMBS cases that have considered sampling have permitted it as a means to present proof at trial." Tr., at 15. In its main brief on these motions, [Doc. 77] at 4-5 and n. 1, Law Debenture collected a number of federal and state cases in support of its assertion that "the overwhelming majority of courts" in the Second Circuit and elsewhere considering the question "have accepted statistical sampling as an appropriate method of establishing liability and damages in RMBS cases."
In contrast, WMC does not cite any case holding or suggesting that the sort of statistical sampling prepared by Dr. Lipshutz for the stated purpose fails to pass muster for admissibility under Rule 702 or Daubert. On the contrary: Dr. Lipshutz's sampling methodology in this case would appear to be in material respects the same as that he employed in the comparable circumstances of Flagstar, where Judge Rakoff denied a defense motion during trial to exclude the testimony under Rule 702.
The arguments WMC puts forward do not support its requested conclusion. The core objection is that Dr. Lipshutz's statistical sampling does not furnish loan-by-loan proof of breaches of representations or warranties and damages. That is true, but it does not matter because Dr. Lipshutz was not asked to form such opinions. Loan-by-loan proof would be required only if WMC is correct in its contention that the "sole remedy" provision in the PSA entirely precludes proof of liability or damages by sampling. That question is reserved for later in limine motion practice; the Court intimates no present opinion on the point; and the issue does not fall within this Daubert - Rule 702 exercise.
WMC's related criticism, that any opinion Dr. Lipshutz might give at this time on the amount of damages would be impermissibly speculative, misses the mark because Dr. Lipshutz does not offer any opinion about the quantum of damages: that will be the trial function of Dr. Mason.
WMC's criticism of Dr. Lipshutz's choice of sample size or margin of error may be advanced through cross-examination at trial, but does not rise to the level of excluding his testimony, particularly since Dr. Leamer, WMC's expert, at his deposition stopped well short of condemning outright Dr. Lipshutz's 5 percent margin of error. Brief [Doc. 105-1], Ex. A at 228: "So the answer is that a 5 percent margin of error could be appropriate depending upon the costs and benefits."
It follows from these considerations that the Court hereby GRANTS Plaintiff's request to expedite determination of statistical sampling and DENIES Defendant's Cross-motion to Exclude or Limit the Testimony of Nelson Lipshutz [Doc. 74].
The Court notes that, upon the parties' prior stipulation, in which Plaintiff agreed to withdraw its "Motion for An Expedited Determination of Statistical Sampling," the Court denied that motion as moot. See Doc. 78 (denying as moot Doc. 60 "in light of the Court's [71] Order adopting the parties' [70] Revised Stipulation and Amendment to Scheduling Order"). In the stipulation, the parties set a jointly proposed Scheduling Order, setting a deadline for WMC to file its anticipated Daubert motion, which, as a practical matter, maintained Plaintiff's request for an expedited determination regarding its "use of statistical sampling in this litigation" in an attempt "to prove liability, damages, or entitlement to any other relief with respect to any or all of [its] claims." Doc. 70, ¶ 3. Thus, in considering the present motion regarding the testimony of Dr. Lipshutz, the Court contemporaneously grants Plaintiff's request for an expedited determination of the validity of statistical sampling as a source of proof at the trial.
This resolution was delayed because, while decision regarding his proffered testimony was pending, Dr. Lipshutz discovered that he had inadvertently included in the total pool of loans, from which the 400-loan sample had been taken, a number of paid-in-full loans, which he had intended to exclude. Dr. Lipshutz assembled another pool, conducted another sample, and issued a revised opinion, developments which triggered another round of briefs of counsel and a renewed deposition of Dr. Lipshutz. I have considered these circumstances, and conclude that they do not affect the analysis or conclusion contained in this Ruling. In particular, while these latest submissions have led to a discussion of whether Dr. Lipshutz's eventual breach rate will be count-based or dollar-based, there is no showing that it will make any material difference to the accuracy or reliability of the rate.
Lastly, it is advisable to reiterate the Court's reasoning and the effect of this Ruling. As the discussion supra stresses, the Court's sole function on these motions is to determine if Dr. Lipshutz's statistical sampling opinions and testimony have about them sufficient indicia of reliability to be admissible at trial, and consequently considered by the Court as finder of fact. This Ruling says nothing about whether, at the conclusion of a full plenary trial, replete with cross-examination, opposing expert opinion testimony, and all the other weapons, stratagems, and devices of advocacy, the Court will find or decline to find Dr. Lipshutz's opinions and testimony to be correct and persuasive. If Law Debenture, in making this motion, sought a more broad validation by the Court of Dr. Lipshutz's opinions than that conferred by this Ruling, it must accept this result and its chances - run the remaining risk, Justice Bransten would say - at trial.
V
Conclusion
For the foregoing reasons, Plaintiff's request to expedite determination of statistical sampling, as set forth in its former motion [Doc. 60] is GRANTED. Defendant's "Motion to Exclude or, in the Alternative, Limit the Testimony of Nelson Lipshutz, Ph.D." [Doc. 74] is DENIED.
See n. 7, supra. --------
It is SO ORDERED. Dated: New Haven, Connecticut
December 30, 2015
/s/Charles S . Haight, Jr.
CHARLES S. HAIGHT, JR.
Senior United States District Judge