CMFG Life Insurance Company et al v. Credit Suisse Securities (USA) LLC

Filing 240

OPINION AND ORDER granting 173 Motion in Limine to Admit Credit Suisse's RMBS Settlement Agreement with the Department of Justice; granting in part and denying in part 176 Motion to Exclude the Expert Testimony of William N. Goetzmann, Ph .D.; denying 179 Motion in Limine to Exclude Evidence of Post-Transaction Loss Causation; granting 182 Motion to Partially Exclude the Expert Testimony of Leonard A. Blum; denying 184 Motion to Exclude the Expert Testimony of Dr. Charles Cowan Regarding the CSFB 2005-3 Certificate; denying 186 Omnibus Motions in Limine; denying 188 Motion to Exclude the Expert Testimony of Steven I. Butler. Signed by District Judge William M. Conley on 10/23/2017. (kwf)

IN THE UNITED STATES DISTRICT COURT
FOR THE WESTERN DISTRICT OF WISCONSIN
CMFG LIFE INSURANCE COMPANY,
CUMIS INSURANCE SOCIETY, and
MEMBERS LIFE INSURANCE COMPANY,

v.

Plaintiffs,

OPINION AND ORDER

CREDIT SUISSE SECURITIES (USA) LLC,

14-cv-249-wmc

Defendant.
This case is set for a bench trial commencing, October 30, 2017, on plaintiffs CMFG
Life Insurance Company, CUMIS Insurance Society and MEMBERS Life Insurance
Company’s (collectively “CUNA Mutual”) claim to rescind purchases of twelve separate
RMBS certificates from defendant Credit Suisse Securities (USA) LLC (“Credit Suisse”)
based on alleged misrepresentations in the offering documents. In advance of the final
pretrial conference scheduled for October 24, the court issues the following opinion and
order on the parties’ motions in limine.

OPINION
I. Plaintiffs’ Motions in Limine
A. MIL No. 1: Admit Credit Suisse’s RMBS Settlement Agreement with the
Department of Justice (dkt. #173)
In this motion in limine, plaintiffs seek an order admitting a January 2017
settlement agreement between Credit Suisse and the Department of Justice, including a
19-page statement of facts, which plaintiffs represent includes an acknowledgement on the

part of Credit Suisse of “broad misconduct in its creation and sale of [RMBS], including
the RMBS that it sold to CUNA Mutual.” (Pls. Br. (dkt. #174) 5.) Plaintiffs contend this
acknowledgement is admissible as a statement by a party opponent pursuant to Federal
Rule of Evidence 801(d)(2)(A). Unsurprisingly, defendant also filed its own motion in
limine to exclude the acknowledgement. (Dkt. #186.)
By way of background, on January 18, 2017, Credit Suisse agreed to pay $5.28
billion in exchange for the release of the DOJ’s civil claims relating to Credit Suisse’s RMBS
activities. (Strikis Decl., Ex. 1 (dkt. #175-1).) While the settlement concerned several of
the same RMBS certificates that CUNA Mutual purchased from Credit Suisse (see id., Ex.
2 (dkt. #175-2) (Annex 3 listing RMBS certificates subject to settlement)), it did not settle
individual claims like that asserted by CUNA Mutual here (see id., Ex. 1 (dkt. #175-1) ¶
6.b)).
As part of the settlement agreement, Credit Suisse “acknowledge[d]” the Statement
of Facts. (Id., Ex. 1 (dkt. #175-1) p.2 (citing Ex. 3 (dkt. #175-3) (Annex 1, Statement of
Facts))).) In this way, Credit Suisse neither expressly “admit[ted]” liability nor the truth
of the statements of facts, nor did it expressly dispute liability or the statement of facts.
Still, the statement contains several purported “facts,” including that: (1) the offering
documents were false; (2) Credit Suisse’s due diligence was inadequate; and (3) Credit
Suisse knew of misrepresentations in the offering documents, but failed to report that to
investors or rating agencies.

(See Pls.’ Br (dkt. #174) 6-7 (describing key factual

statements); Strikis Decl., Ex. 3 (dkt. #175-3)).)
Defendant argues that the Settlement and the Statement of Facts should be

2

excluded under Federal Rule of Evidence 408, which provides in pertinent part:
Evidence of the following is not admissible--on behalf of any
party--either to prove or disprove the validity or amount of a
disputed claim or to impeach by a prior inconsistent statement
or a contradiction:
(1) furnishing, promising, or offering--or accepting, promising
to accept, or offering to accept--a valuable consideration in
compromising or attempting to compromise the claim; and
(2) conduct or a statement made during compromise
negotiations about the claim--except when offered in a criminal
case and when the negotiations related to a claim by a public
office in the exercise of its regulatory, investigative, or
enforcement authority.
Fed. R. Evid. 408 (emphasis added).
Contrary to defendant’s argument, however, the Seventh Circuit has interpreted
“the claim” language narrowly in applying Rule 408. In particular, the court explained in
Zurich American Insurance Co. v. Watts Industries, Inc., 417 F.3d 682 (7th Cir. 2005), that
the balance between the “need for the settlement evidence” and the “potentially chilling
effect on future settlement negotiations” is “especially likely to tip in favor of admitting
evidence when the settlement communications at issue arise out of a dispute distinct from
the one for which the evidence is being offered.” Id. at 689. More recently, the Seventh
Circuit reiterated that holding after analyzing the text of Rule 408, concluding that “[t]he
Rules’ use of the singular term ‘claim’ suggests that settlements discussions concerning a
specific claim are excluded from evidence to prove liability on that claim, not on others.”
Wine & Canvas Dev., LLC v. Muylie, 868 F.3d 534, 541 (7th Cir. 2017).
While plaintiffs’ claim here certainly is similar to the claims underlying the Credit
Suisse-DOJ settlement, this alone does not satisfy the narrow construction of Rule 408.

3

Zurich Am. Ins., 417 F.3d at 689-90 (“Of course, the two actions are not totally unrelated.
Zurich did, after all, raise the deductible agreements as a defense to its duty to defend
Watts in Armenia and Rothschild. Still, the California action--based on the primary liability
insurance policies--is distinct from the Illinois petition to compel arbitration under the
deductible agreements.”).

In contrast, defendant directs the court to a District of

Massachusetts opinion excluding as evidence the very same settlement agreement as that
at issue here under Rule 408. See Mass. Mut. Life Ins. Co. v. DLJ Mortg. Cap., Nos. 1130047-MGM, 11-30048-MGM, 2017 WL 1709594, at *1-2 (D. Mass. May 2, 2017). In
excluding the agreement, the Massachusetts court, however, followed First Circuit
precedent as it is required to do. Under its own precedent, that court found the Rule 408
prohibition “applies equally to settlement agreements between a defendant and third party
and between a plaintiff and a third party.” Id. at *1 (quoting Portugues-Santana v. Rekomdiv
Int’l, 657 F.3d 56, 63 (1st Cir. 2011)). Because the Seventh Circuit adopted a narrow
interpretation of Rule 407, however, the Mass Mutual Life Insurance decision holds no
weight. 1
From this, the court concludes that Rule 408 does not outright bar consideration of
the Settlement Agreement, including the Statement of Facts, but this does not end the
inquiry. Rather, the question remains on what basis would the court admit this document?
Plaintiffs contend that the Settlement Agreement and Statement of Facts are admissible

Defendant also argues that the CUNA Mutual has no need for the Settlement Agreement because
“all of the underlying emails and referenced documents have been produced to CUNA Mutual.”
(Defs.’ Opp’n (dkt. #227) 10.) This argument cuts both ways: if the underlying documents are
admissible, what prejudice would Credit Suisse suffer from admitting the Settlement Agreement
and the Statement of Facts? On the other hand, if the court credits defendant’s argument, then
what independent purpose does the Settlement Agreement serve?
1

4

generally as the statement of a party opponent under Federal Rule of Evidence 801(d)(2),
which provides in pertinent part:
(2) An Opposing Party's Statement. The statement is offered
against an opposing party and:
(A) was made by the party in an individual or representative
capacity; [or]
(B) is one the party manifested that it adopted or believed to
be true;
...
Fed. R. Evid. 801(d)(2). Relying on dictionary definitions of the word “acknowledge,” or
other cases interpreting that word in other contexts, plaintiffs argue that by acknowledging
the Statement of Facts, defendant “admit[ed] or recognize[d] those facts as true.” (Pls.’
Br. (dkt. #174) 8.)
As defendant points out in its response brief, however, the cases cited by plaintiff
in support of this definition of “acknowledge” involve settlement agreements in which the
party opponent expressly admitted liability. (See Def.’s Opp’n (dkt. #227) 15-16.) Here,
Credit Suisse’s acknowledgement of the Statement of Facts stops short of an admission -- as
mentioned above, the Settlement Agreement is silent as to whether Credit Suisse admitted
liability. Indeed, the word choice is obviously one of art at best and legalese at worst,
which could be construed as an admission or simply notice of the statements that are
contained in this compromise document. Absent more, this court has no way of telling for
certain which it is.
As such, the Statement of Facts constitutes a “statement” “made by the party in an
individual or representative capacity” under Rule 801(d)(2)(A), while the court is hardpressed to find that the statement was one Credit Suisse “manifested that it adopted or

5

believed to be true.” Fed. R. Evid. 801(d)(2)(B). On the other hand, Credit Suisse did
not insist that the Settlement Agreement contain the word disavowal of any liability, and
the best evidence as to why it paid a fortune to resolve its dispute with the Justice
Department would seem to be the so-called Statement of Facts.
As such, the motion is GRANTED in so far as Rule 408 does not bar Credit Suisse’s
statements in the Settlement Agreement and the Statement of Facts is admitted under
Rule 801(d)(2)(A), although the court expresses no opinion without more context as to
what weight, if any, the court should attach to it in light of Credit Suisse’s having
“acknowledge[d]” those facts.

B. MIL No. 2: Exclude Expert Testimony of William N. Goetzmann, Ph.D.
(dkt. #176)
In this motion in limine, plaintiffs seek an order excluding expert testimony. The
admissibility of expert testimony in federal courts is principally governed by Federal Rule
of Evidence 702, as elucidated by Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579
(1993). 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.

6

In applying Rule 702, a district court is to function as a “gatekeeper,” determining
whether a party’s proffered expert testimony is relevant and reliable. Daubert, 509 U.S. at
589; see also United States v. Johnsted, 30 F. Supp. 3d 814, 816 (W.D. Wis. 2013) (expert
testimony must be “not only relevant, but reliable”). Although “liberally admissible under
the Federal Rules of Evidence,” Lyman v. St. Jude Med. S.C., Inc., 580 F. Supp. 2d 719, 723
(E.D. Wis. 2008), expert testimony must, therefore, satisfy the following three-part test:
(1) the witness must be qualified “as an expert by knowledge,
skill, experience, training, or education,” Fed. R. Evid. 702;
(2) the expert’s reasoning or methodology underlying the
testimony must be scientifically reliable, Daubert, 509 U.S. at
592-93; and
(3) the testimony must assist the trier of fact to understand the
evidence or to determine a fact in issue. Fed. R. Evid. 702.
Ervin v. Johnson & Johnson, Inc., 492 F.3d 901, 904 (7th Cir. 2007). Still, “[v]igorous crossexamination, presentation of contrary evidence, and careful instruction on the burden of
proof are the traditional and appropriate means of attacking shaky but admissible
evidence.” Daubert, 509 U.S. at 596.
Given that this case will be tried to the bench, where the court need not worry about
protecting a jury, “the court’s gatekeeping role is necessarily different.” In re Salem, 465
F.3d 767, 771 (7th Cir. 2006). In particular, “the need to make [reliability] decisions prior
to hearing the testimony is lessened.” Id. In deciding this motion and the other motions
challenging expert testimony, therefore, the court’s focus will be on the relevance or
usefulness of the expert opinion, rather than its reliability. Although for the reasons
explained below, the court will strike some proposed testimony that is so unreliable the

7

court need not wait to rule. 2
As for William Goetzmann’s statistical analysis that is the subject of plaintiff’s MIL
No. 2, he opines that the loans underlying the Certificates at issue in this case performed
as expected based on the loan characteristics disclosed in the Offering Documents and
prevailing macroeconomic conditions during the relevant period. More specifically, relying
on his statistical analysis and other economic studies, Goetzmann concludes that
macroeconomic conditions caused a historic rise and fall in U.S. housing prices that
ultimately led to the decline in value of the RMBS at issue in this case. Defendant argues
that Goetzmann’s testimony, therefore, refutes plaintiffs’ position that: (1) there were
misrepresentations in the offering documents; and (2) these misrepresentations caused the
underperformance of the Certificates.
1. Regression Analysis
Plaintiffs challenge two regression analyses. The first regression analysis uses a linear
regression model to compare the performance, measured by default rates, of the loans in
the supporting loan groups (“SLGs”) underlying the Certificates at issue in this case to a
“comparison sample” consisting of millions of Alt-A, subprime and second-lien mortgage
loans taken from approximately 1,546 non-agency RMBS offerings that were issued
between 2004 and 2006, excluding all RMBS that were underwritten, issued or sponsored
by Credit Suisse. (Pls.’ Br. (dkt. #177) 10.) Perhaps anticipating plaintiffs’ motion,

To the extent excluded, the court is treating the expert’s report as the proponent’s proffer for
purposes of the record, consistent with its intent to rely on those reports in lieu of the experts’
direct testimony on this case, but will not preclude the losing party from making a further proffer
at the Final Pretrial Conference.
2

8

Goetzmann then created a “modified comparison sample,” which removed any loans
subject to litigation. (Id. at 11.) In its response brief, defendant solely relies on the
modified sample to oppose the motion in limine, apparently conceding, at least in part, the
validity of plaintiffs’ criticism of Goetzmann’s analysis based on his original comparison.
As such, the court limits its review of the parties’ arguments to the modified sample.
In his report, Goetzmann purports to “calculate a predicted rate for each of the
[SLGs],” after selecting a comparison sample, relying on the stated loan and borrower
characteristics of each SLG at-issue and the “changes in macroeconomic conditions” to
which each was exposed and assuming that the SLGs had been truthfully described by
Credit Suisse in the offering documents and otherwise. (Goetzmann Rept. (dkt. #207) ¶
73.) Goetzmann then compared this predicted default rate calculated based on the modified
comparison sample against the actual default rate of the SLGs at issue. Based on the
modified comparison sample, Goetzmann concluded that for five of the seven Certificates,
“the actual default rates did not exceed the predicted default rate at a statistically significant
level.” (Pls.’ Br. (dkt. #177) 12 (quoting Goetzmann Rept. (dkt. #207) ¶ 77).) From this,
Goetzmann opined that “the default and delinquencies experienced by the [SLGs] are (1)
consistent with their stated characteristics and the changes in macroeconomic conditions
and (2) are consistent with the hypotheses that alleged defects did not exist or did not
affect the performance of the [SLGs].” (Id. (quoting Goetzmann Rept. (dkt. #207) ¶ 78).)
In challenging this analysis, plaintiffs direct the court to the decisions of two other
district courts, one of which excluded Goetzmann’s testimony based on the same regression
analysis, Nat’l Credit Union Admin. Bd. v. UBS Sec., LLC, Nos. 12-2591-JWL, 12-2648-

9

JWL, 2016 WL 7373857, *6-8 (D. Kans. Dec. 20, 2016) (“NCUA”), and the other of
which excluded the same benchmark regression analysis of another expert in a similar
RMBS lawsuit, Fed. Hous. Fin. Agency v. Nomura Holding Am., No. 11cv6201 (DLC), 2015
WL 539489, *6-7 (S.D.N.Y. Feb. 10, 2015) (“Nomura”). As those courts found, plaintiffs
argue, that the comparison samples underlying Goetzmann’s analysis were not proper
controls, making the analysis itself unreliable.
Stated another way, Goetzmann would opine that the underwriting defects did not
“cause” the underperformance of the Certificates by purportedly comparing the
performance to a group of underlying loans that did not suffer from the same underwriting
errors. Assuming for purposes of argument that such a comparison has a bearing on
causation here, however, Goetzmann apparently made no effort to ensure that the original
comparison sample contained loans that did not have underwriting errors. See NCUA,
2016 WL 7373857, at *6 (“If the other loans suffered from similar defects found in the
subject loans, then a comparison using those other loans could not determine whether
factors other than the defects caused the loans.”); Nomura, 2015 WL 539489, at *5
(“[W]hen designing an experiment to test whether an observed result was caused by [a]
given variable, the control or benchmark group must lack that variable. That is the whole
point of a control group.”)
While the modified sample removed all loans subject to litigation, defendants
concede that even this modified sample is not “clean.” (Def.’s Opp’n (dkt. #228) 21
(citing Goetzmann Dep. (dkt. #170) 53 (stating that creating a clean empirical sample “is
not something that is easily done” and not necessary to his analysis”).)

10

See also NCUA,

2016 WL 7373857, at *7 (“Dr. Goetzmann conceded in his testimony that the absence of
litigation was not a clear proxy for a lack of defects, and he refused to express an opinion
as to whether the absence of litigation is an accurate predictor of compliance.”). The court
agrees with the other district courts that while removing loans that are subject to litigation
may be a “good start for creating a clean benchmark,” Nomura, 2015 WL 539489, at *7,
Goetzmann’s modified sample falls short of providing a reliable control group. See also
NCUA, 2017 WL 7373857, at *7 (rejecting analysis based on modified benchmark
excluding loans subject to litigation).
In opposition, defendant repeats Dr. Goetzmann’s position that a clean sample
“does not bear on the reliability or scientific validity of [his] methodology” (Def.’s Opp’n
(dkt. #228) 22), but in support merely offers a convoluted argument involving
Goetzmann’s second regression analysis as described below. Regardless, for the reasons
explained above and in the opinions of other district courts striking the same or similar
expert testimony, the court agrees that absent a clean control group, Goetzmann’s first
regression analysis is not a reliable basis to opine that the alleged underwriting defects were
not the cause the certificates’ underperformance. As such, the court will grant plaintiffs’
motion to exclude from trial the first analysis and Goetzmann’s opinions based on that
analysis.
As for Goetzmann’s second regression analysis, he purports to have compared the
actual performance of loans in the SLGs categorized by plaintiffs’ expert Stephen Butler as
“materially defective” with the actual performance of loans in the SLGs that Butler did not

11

categorize as “materially defective.” 3 Goetzmann then ran a “logit regression” on the 686
sampled loans, using whether a loan had ever been “in default” as its dependent variable
and several independent variables, including owner occupancy, LTV, housing prices, along
with the “binary variable” indicating whether Butler categorized the loan as materially
defective. From this, Goetzmann found loans deemed “materially defective” were more
likely to default, but also found this variable was not statistically significant. From this,
Goetzmann again opined that “the alleged underwriting defects identified by Mr. Butler in
the loans he reviewed did not contribute to the defaults and delinquencies experienced by
the loans from the [SLGs].” (Pls.’ Br. (dkt. #177) 13 (quoting Goetzmann Rept. (dkt.
#270) ¶ 85).)
In challenging this opinion and underlying testimony, plaintiffs take issue with
Goetzmann’s reliance on Butler’s own categorization, claiming that he was “intentionally
conservative and under-inclusive” in categorizing loans as “materially defective.” (Pls.’ Br.
(dkt. #177) 9.) From this, plaintiffs argue that “Dr. Goetzmann improperly equates the
absence of a finding that a loan is “Materially Defective” with the presence of an affirmative
finding that the loan is free of defects.” (Id. at 21.) At most, this criticism goes to weight,
not admissibility. Even if Butler’s categorization falls short of creating a completely clean
sample, Goetzmann’s reliance on plaintiff’s expert Butler’s own system of categorizing
loans as “materially defective” against those not categorized as “materially defective,”
seems reasonable in a “but for” world. While plaintiffs may argue that Butler’s analysis

Butler purported to categorize loans based on underwriting defects at the time the loans were
originated.
3

12

was so under inclusive as to undermine Goeztmann’s analysis, the court sees no reason to
strike his testimony as unreliable on that basis. On the contrary, using plaintiffs’ expert to
draw a line between loans that suffer from material underwriting defects and those that do
not, at least brings the parties’ competing analysis closer to talking about an “apples to
apples” comparison. 4
2. Macroeconomic Conditions
Finally, plaintiffs seek to exclude Goetzmann’s testimony more generally as to larger
macroeconomic forces being the cause of the underperformance of the Certificates at issue.
Plaintiffs contend that this testimony is only relevant to defendant’s loss causation defense,
which is subject to its own motion in limine, and to prove it, defendant would need to
prove that “the loss in the value of the security was proximately caused by events unrelated
to the phenomena underlying the alleged misrepresentations.” (Pls.’ Br. (dkt. #177) 23
(quoting Fed. Hous. Fin. Agency v. Nomura Holding Am., Inc., 104 F. Supp. 3d 441, 589
(S.D.N.Y. 2015), aff’d sub nom. No. 15-1872-CV, 2017 WL 4293322 (2d Cir. Sept. 28,
2017).)
As an initial matter, plaintiffs cite to Judge Cote’s post-trial opinion in Nomura,
which found that the defendant failed to prove the affirmative defense of loss causation.
The Second Circuit has since articulated a different, arguably lower standard of proof,

In fairness, Judge Cote in the Nomura case agreed with plaintiffs and struck similar analysis of
another expert. See Nomura, 2015 WL 539489, at *8-9. This court may ultimately agree with
Judge Cote’s reasoning, but will consider defendant’s challenge after hearing both sides’ testimony,
particularly since unlike the expert in Nomura, plaintiffs’ expert will be available for crossexamination, thus distinguishing this case from the circumstances in Nomura. Id. at *9.
4

13

however, which requires only a showing that “the risk that caused the loss[es] was [not]
within the risk concealed by the misrepresentations and omissions.” Fed. Hous. Fin. Agency
v. Nomura Holding Am., Inc., No. 15-1872-CV, 2017 WL 4293322, *51 & n.73 (2d Cir.
Sept. 28, 2017).
macroeconomic

Regardless of the precise standard of proof, however, whether the
conditions

Goetzmann

describes

operated

independently

from

defendant’s alleged misconduct is an area for cross-examination and not a basis for
excluding this portion of Goetzmann’s testimony.
Accordingly, plaintiffs’ motion is GRANTED IN PART AND DENIED IN PART
consistent with this opinion.

C. MIL No. 3: Exclude Evidence of Post-Transaction Loss Causation (dkt.
#179)
Related to plaintiffs’ challenges to Goetzmann’s testimony, plaintiffs move to
exclude defendant’s evidence of post-transaction loss causation generally. (Pl.’s Mot. (dkt.
#179.)) Plaintiffs argue that loss causation is neither an element nor an affirmative defense
to a rescission claim under Wisconsin law, and therefore causation evidence should be
barred. Since this evidence is relevant to the court’s consideration of the equities here, as
well as for other purposes relevant to plaintiff’s rescission claim, the motion is DENIED.
In

formalist

terms,

plaintiffs’

argument has

some

merit.

A

material

misrepresentation results in the failure of mutual assent, and therefore, the failure to form
a contract. Restatement (Second) of Contracts § 163. Thus, a party may unilaterally void
a contract if (1) its assent was induced by a material misrepresentation; and (2) it was
justified in relying upon that misrepresentation. Id. at § 164(1). The party seeking

14

rescission does not need to show causation to make a prima facie case.
Here, plaintiffs seek rescission based upon material misrepresentations made at the
time of sale.

Allegedly, the RMBS certificates were substantially less valuable than

defendant represented, making the contract flawed from the very beginning. Plaintiffs
argue that since the contract was voidable at the time of sale, any consideration of the
housing market or other forces later affecting the value of the certificates would be
inappropriate -- a true contract never existed.
However, rescission is an equitable remedy. When exercising this discretionary
power, the court must balance the equities of the parties. As noted by the Seventh Circuit,
Wisconsin courts have previously drawn from the Restatement of Restitution and are likely
to adopt its rules for granting rescission. See CMFG Life Ins. Co. v. RBS Sec., Inc., 799 F.3d
729, 736 (7th Cir. 2015) (“In deciding cases involving rescission, the Wisconsin Supreme
Court has drawn from restatements, including the Restatement of Restitution”) (citations
omitted). 5 Restatement (Third) of Restitution and Unjust Enrichment § 54(3) (2011)
states:
Rescission is limited to cases in which counter-restitution by
the claimant will restore the defendant to the status quo ante,
unless
(a) the defendant is fairly compensated for any deficiencies
in the restoration made by the claimant, or
(b) the fault of the defendant or the assignment of risks in
Plaintiffs argue that under Admiral Ins. Co. v. Paper Converting Mach. Co., 811 NW.2d 351 (Wis.
2012), the Restatement of Restitution only applies when a party asserts an unjust enrichment claim.
In Admiral, however, the court was addressing the issue of whether or not an insurer could make an
unjust enrichment claim, and it was directly addressing the substance of that claim. Id. at 361.
Since the claim was barred, citations to the Restatement were considered unpersuasive, but Admiral
did not limit the Restatement of Restitution to claims for unjust enrichment.
5

15

the underlying transaction makes it equitable that the
defendant bear any uncompensated loss.
In this case, subsection (b) may apply. If so, defendant may end up bearing some or all of
the loss caused by the RMBS certificates’ underperformance based on fault. Loss causation
may, therefore, be a factor in the equitable allocation of fault or risk.
In arguing that causation evidence is nevertheless barred, plaintiffs rely heavily on
First National Bank & Trust Company of Racine v. Notte, 97 Wis. 2d 207, 293 N.W.2d 530
(Wis. 1980), a case that -- as plaintiffs admit -- is about scienter. Under Notte, innocent
misrepresentation is not a defense to rescission claims. For example, defendant could not
argue that rescission is legally barred because, unbeknownst to defendant, mortgagors lied
on their mortgage applications. Again, however, the Notte decision cannot be read so
broadly as to bar all evidence of causation, nor is Notte in conflict with the Restatement of
Restitution.
While disagreeing on the relevance of loss causation itself, both parties recognize
that some form of harm may be important to establish a claim for rescission. So, too, does
this court as explained in a 2014 opinion, issued in a substantially similar case, also
involving plaintiffs.
[Defendant] relies on the Wisconsin Court of Appeals’
decision in Tam v. Luk, 453 N.W.2d 158 (Wis. Ct. App. 1990),
which held that “a showing of pecuniary loss is not necessary
to establish a claim for rescission based on a seller’s
misrepresentation of the status of the subject property, . . .
[but] there must be some showing of prejudice, damage or
detriment to the buyer.” Id. at 288. While this proposition
makes some intuitive sense, CUNA Mutual is correct that
Wisconsin’s law on rescission arises from the Restatement
(Second) of Contracts, see Notte, 97 Wis. 2d at 222 (adopting
Restatement approach), and the Restatement expressly

16

provides that “the recipient of a misrepresentation need not
show that he has actually been harmed by relying on it in order
to avoid the contract.” Restatement (Second) of Contracts §
164 cmt. c. In light of the Restatement, it seems more accurate
to say that a showing of harm, while not mandatory, is
nevertheless an important and sometimes dispositive factor in
a court’s determination of whether it should exercise its
discretion to order rescission given the facts and circumstances
of a particular case. See Ott v. Peppertree Resort Villas, Inc., 716
N.W.2d 127, ¶ 55 (Lundsten, P.J., dissenting).
CMFG Life Ins. Co. v. RBS Sec. Inc., No. 12-CV-037-WMC, 2014 WL 3696233, at *22
(W.D. Wis. July 23, 2014), aff’d in part, rev’d in part, 799 F.3d 729 (7th Cir. 2015). The
reasoning from RBS remains applicable, as both plaintiffs’ reliance and defendant’s related
causation of harm may well be important factors in the court’s decision to exercise its
discretion in awarding total or partial rescission.
Finally, defendant argues that the facts underlying loss causation are relevant to
plaintiffs’ affirmative claims. Most significantly, defendant intends to rely affirmatively on
much of their expert Goetzmann’s opinion not excluded by the court to show that the
performance of the RMBS certificates matched their stated characteristics. (Goetzmann
Rept. (dkt. #142) ¶¶ 68-86.) As far as this is offered to rebut plaintiffs’ claims of material
misrepresentation, it is admissible. Of course, the parties will still have an opportunity at
trial to argue which evidence should, and should not, inform the court’s exercise of its
discretion under rescissionary principles.

II. Defendant’s Motions in Limine
A. MIL No. 1: Exclude Settlement Agreement Between the Department of
Justice and Credit Suisse and Similar Settlements (dkt. #186)
This motion is DENIED for the same reasons provided in granting plaintiffs’ motion

17

in limine no. 1.

B. MIL No. 2: Exclude Evidence Concerning the 2012 Settlement Between
Credit Suisse and the SEC (dkt. #186)
This motion touches on markedly similar concerns as raised in the admissibility of
the 2017 Settlement Agreement between Credit Suisse and the DOJ in plaintiffs’ motion
in limine no. 1 and defendant’s motion in limine no. 1. Here, defendants seek to exclude
a 2012 settlement between Credit Suisse and the SEC regarding loan repurchases by Credit
Suisse and the use of those proceeds. Defendant offers two, core reasons why this evidence
should be excluded under Federal Rules of Evidence 401, 402, 403 and 408. First, in its
2012 agreement with the SEC, Credit Suisse consented to entry of the cease-and-desist
order “[s]olely for the purposes of these proceedings . . . and without admitting or denying
the findings herein.” (Def.’s Br. (dkt. #187) 15 (quoting Valdes Decl., Ex. 15 (dkt. #1909).) Second, the allegations underlying that settlement agreement concerned repurchase
of certain obligations under the Pooling and Servicing Agreements and had no bearing on
the alleged misstatements at issue in this lawsuit.
For the same reasons described above in rejecting Rule 408’s application to the
Credit Suisse-DOJ settlement, the court agrees with plaintiff that the 2012 SEC settlement
is not barred by Rule 408. This leaves defendant’s challenges to the possible relevance of
the 2012 settlement, and whether it is outweighed by undue prejudice to Credit Suisse.
Plaintiffs contend that the settlement is relevant to the efficacy of Credit Suisse’s due
diligence for two reasons:

(1) “the SEC Order shows that Credit Suisse willfully

disregarded red flags that should have alerted Credit Suisse to the fact that many of the

18

loans it had securitized violated underwriting guidelines”; and (2) “the SEC’s findings
demonstrate Credit Suisse’s perverse financial incentives to avoid conducting thorough due
diligence and quality control.” (Pls.’ Opp’n (dkt. #222) 13-14.)
As set forth in the 2012 SEC Agreement, when Credit Suisse purchased loans from
originators, it generally received representations and warranties that allowed Credit Suisse
to demand that the originator repurchase the loan if the borrower missed one of the first
three mortgage payments after Credit Suisse’s purchase -- a so-called “early payment
default.” When this occurred after Credit Suisse had already securitized the loan, however,
rather than requiring the originator to repurchase the loan, Credit Suisse instead demanded
a cash payment from the originator, without passing along those proceeds or otherwise
informing the RMBS investor about the early payment default under that loan or
repurchasing that loan from the securitized trust. The SEC found that from 2005 to 2012
Credit Suisse had “improperly obtained” more than $55 million in such payments.
To put those payments in context, plaintiffs intend to offer the 2012 SEC
settlement as evidence that Credit Suisse knew of defective loans as early as 2005, but
opted to ignore the problems with the underlying loans and not improve its due diligence
efforts because it was financially lucrative not to do so. While the evidence underlying the
2012 SEC settlement, as well as the statement that defendant neither admits nor denies
the SEC’s findings, may be admissible, the same issue as to the weight the court should
assign to the 2012 settlement is the same as the 2017 settlement with the DOJ.
Nevertheless, the court will DENY defendant’s motion in limine.

19

C. MIL No. 3: Exclude Evidence Concerning the Financial Crisis Inquiry
Commission Report (dkt. #186)
In the next motion, defendants seek to exclude a 2010 Report by the Financial Crisis
Inquiry Commission (“FCIC Report”) on relevance grounds under Federal Rules of
Evidence 401 and 402 and as inadmissible hearsay under Rule 802. 6 Defendant argues
that a report “on the causes of the financial crisis are not relevant to the disputed issued in
this case,” and even if the report touched on relevant issues, that it should be excluded
because it is “not the product of a reliable adjudicatory process.” (Def.’s Br. (dkt. #187)
17.)
In their response, plaintiffs offer various ways that the report is relevant:

(1)

demonstrating falsity and materiality of the underwriting defects to “a reasonable investor”
who “would not have known (but would have considered important) the kind of
information about mortgage quality that Credit Suisse withheld from investors”; (2)
rebutting any argument that conforming with industry practices should be deemed
adequate in light of widespread due diligence problems; and (3) responding to Credit
Suisse’s loss causation defense with the report’s finding that “Credit Suisse’s conduct in
securitizing defective mortgage was part and parcel of the macroeconomic factors that
Credit Suisse seeks to blame for [] CUNA Mutual’s losses.” (Pl.’s Opp’n (dkt. #222) 18.)
The court finds each of these bases adequate to find that the report is relevant to the claims

Defendant also sought to exclude a 2011 Report by the U.S. Senate Permanent Subcommittee on
Investigations into the origins of the 2008 financial crisis. (Def.’s Br. (dkt. #187) 16.) In their
response brief, however plaintiffs disavowed any intent to introduce this report at trial. (Pls.’ Resp.
(dkt. #222) 17 n.5.) As such, this portion of defendant’s motion will be GRANTED AS
UNOPPOSED.
6

20

at issue in this lawsuit.
As for defendant’s challenge to the reliability of the report, the court agrees with
plaintiffs that this challenge simply goes to the weight the court may place on the report
and not to its admissibility. Furthermore, because this case will be tried to the bench,
concerns about prejudice under Rule 403 are largely abated. See United States v. Lim, 57 F.
App’x 701, 704 (7th Cir. 2003) (“[W]e reject Lim’s Rule 403 claims, which are inapposite
in a bench trial, where there is no risk of jury prejudice.”); United States v. Shukri, 207 F.3d
412, 419 (7th Cir. 2000) (“In a bench trial, we assume that the district court was not
influenced by evidence improperly brought before it unless there is evidence to the
contrary.”).
This leaves defendant’s argument that the report is inadmissible hearsay.
Anticipating plaintiffs’ response, defendant specifically asserts that the report does not fall
under the exception to hearsay for “factual findings from a legally authorized
investigation,” because the “source of information or other circumstances indicate lack of
trustworthiness.” (Def.’s Br. (dkt. #187) 18 (quoting Fed. R. Evid. 803(8)).) As support,
defendant directs the court to a statement by the Vice Chairman of the FCIC “that the
Commission was created for political purposes with a partisan structure and a partisan
agenda.” (Id. (quoting Statement of Bill Thomas, Testimony Before the House Committee
on

Financial

Services,

Feb.

16,

2011,

available

at

https://financialservices.house.gov/media/pdf/021611thomas.pdf).)
As plaintiffs point out in reply, however, this is the only evidence defendant offers
to support a finding that the report is untrustworthy. Indeed, Vice Chairman Thomas

21

included a 27-page dissenting statement in the report itself, explaining his disagreement
with certain findings, thus ameliorating any concern that only one political view was
represented in the report.

Moreover, the dissent of one individual member of the

committee, simply raising questions about whether the investigation was colored by
politics, does not by itself render the report unreliable. See Daniel v. Cook Cty., 833 F.3d
728, 740 (7th Cir. 2016) (“We assume that public officials, in crafting such a report, acted
properly and without bias.”). For these reasons, other courts have found the FCIC report
sufficiently trustworthy to be admissible under the Rule 803(b) exception. (Pl.’s Opp’n
(dkt. #222) 19-20 (citing cases).)
Finally,

defendant

challenges

as

hearsay-within-hearsay

those

third-party

statements contained within the reports, citing cases from the Seventh Circuit affirming
exclusion as hearsay statements made by third parsons contained in official public reports.
(Def.’s Br. (dkt. #187) 19.) For their part, plaintiffs concede that statements of individuals
quoted in the report are inadmissible hearsay, but contend that they seek to submit the
report for the truth of its findings, not for the truth of the matters asserted in individual
statements. As such, the court will DENY the motion in limine, subject to rejecting any
third party statements for the truth of the matter asserted.

D. MIL No. 4: Exclude Testimony of Two Witnesses Whose Prior Testimony
Was Found by a Court in Another RMBS Case to be Tainted and
Improperly Procured (dkt. #186)
Defendant also seeks to exclude the deposition testimony of Ron Szukala and Diane
Johnson under Rule 403. Szukala and Johnson were employees of third-party vendors with
whom Credit Suisse’s due diligence department contracted in connection with loan

22

reviews. CUNA Mutual requested that defendant produce transcripts and exhibits of two
depositions in another RMBS action. While producing both, defendant did so without
waiving its right to seek their exclusion at trial.
In support of its motion to exclude these witnesses’ testimony, defendant directs
the court to an opinion by a New York Supreme Court judge in another RMBS action
expressing “serious concerns . . . about the veracity of their testimony and the manner in
which it was procured.” (Def.’s Br. (dkt. #187) 21) (citing MBIA Ins. Corp. v. Credit Suisse
Sec. (USA) LLC, No. 603751/2009 (N.Y. Sup. Ct. Oct. 3, 2013)).) (See Valdes Decl., Ex.
17 (dkt. #190-10).) In that case, the defendants, which included Credit Suisse, sought to
compel the production of plaintiff’s counsel’s communications with certain third-party
witnesses, employees of third-party vendors of Credit Suisse’s due diligence department.
Plaintiff relied on affidavits of those third-party witnesses in alleging fraud. During the
witnesses’ depositions, defendants discovered that plaintiff had “paid substantial sums of
money (in certain cases more than $10,000) to witnesses and flew the witnesses from
across the country to New York so that they could recount their knowledge of (and possible
participation in) defendants’ fraudulent business practices.” (Id. at 4.) Plaintiff’s counsel
had also “worked with these witnesses to draft affidavits,” “extensively prepped them prior
to deposition and defended their testimony against cross-examination by defendants.”
(Id.) Moreover, during the depositions “several of the witnesses began to recant their
testimony or indicated that what was written in their affidavits was the work of [plaintiff’s]
counsel and was not entirely an accurate description of their knowledge.” (Id. at 5.) While
the court granted the defendants’ motion to compel, finding a “reasonable suspicion of a

23

witness dissembling” (id. at 7), the court did not strike or exclude their deposition
testimony, including the testimony of the two witnesses challenged here.
As an initial matter, Rule 403 has less traction in the context of a bench trial. See
Lim, 57 F. App’x at 704. Here, the two witnesses at issue were specifically asked about the
role of the plaintiff’s counsel in drafting their affidavits and preparing them for trial.
Accordingly, the court will consider that testimony in assessing what weight, if any, their
testimony deserves as a whole. However, the court sees no basis to strike the testimony
out of hand, even though defendant has painted a compelling picture as to how the
testimony has been tainted. Because this determination is best made after reviewing the
parties’ deposition designations, this motion is DENIED.

E. MIL No. 5: Partially Exclude the Expert Testimony of Leonard A. Blum
(dkt. #182)
In this motion, defendant seeks to exclude certain testimony of plaintiff’s expert
Leonard A. Blum.

In his report, Blum purported to opine as to the materiality of

information contained in the initial offering documents for the Certificates at issue. At his
deposition, however, Blum indicated that the opinion in his report is limited to the
Certificates purchased in the initial offering or primary market, while the majority of the
Certificates at issue were purchased on the secondary market.
In reviewing one of the Certificates at his deposition in the case in particular, Blum
discovered, apparently for the first time, that it was issued on the secondary market:
By the way, I just want to mention when I answered your
questions about would someone buy something in the
anticipation of it being downgraded, I thought we were talking

24

about primary offerings of securities.
secondary trade.

This looks like a

. . . So there it’s a different situation because . . . you’ve got an
established history for the bond which is information that’s . .
. different. When you’re buying in the primary market, that
doesn’t exist.
(Blum Dep. (dkt. #195) 340.) Based on this exchange, counsel for defendant then asked:
“Are you saying that the materiality of purchases bought on the secondary market are not
within the scope of your report?,” to which Blum responded, “When you’re talking about
the secondary market purchases, the information is different. It’s a different dynamic than
a primary security purchase.” (Id. at 341; see also id. at 344 (again reiterating that “I’m not
opining about anything in the secondary market.”; “Secondary market is a totally different
market;” and “Prospectuses don’t have the importance as they do of -- of a primary offering
of securities.”).)
In the face of this unambiguous disavowal of any application of his materiality
opinion to Certificates purchased on the secondary market, counsel for plaintiffs attempted
to rehabilitate Blum’s testimony:
Q. So my understanding is that eight of the ten deals at issue
in this case were purchased by Mr. Prusha on the secondary
market and not on the primary market. So assuming that is
true, should your testimony be read to suggest that your
opinions don’t apply to those eight deals?
A. No, I don’t believe so. I mean, that’s a legal question and
I’m not an attorney. Investors are -- are perfectly -- it’s
perfectly reasonable for an investor to rely on a -- on a
prospectus and there are period -- you know, periods specified
in the laws that I will let the lawyers fight about.
(Id. at 359-60.) While the garbled response falls far short of providing assurance that
Blum’s opinion testimony applies to the certificates sold on secondary market, it arguably

25

suggests some possible relevance. More critically, however, Blum offers no basis for finding
relevance in light of his earlier statements unequivocally distinguishing the primary market
from the secondary market and limiting his opinion to the former.
Accordingly, this motion is GRANTED and Blum’s testimony as to the materiality
of information contained in the offering documents for the Certificates purchased on the
secondary market is excluded.

F. MIL No. 6: Exclude the Expert Testimony of Dr. Charles Cowan regarding
the CSFB 2005-3 Certificate (dkt. #184)
Defendant further seeks to exclude certain testimony of another of plaintiffs’ expert,
Dr. Charles Cowan, as it relates to one of the Certificates at issue in this case -- the CSFB
2005-3 Certificate.

Dr. Cowan selected an initial random sample of loans that were

representative of the supporting loan group (“SLG”) underlying the CSFB 2005-3
Certificate. That initial sample failed five out of eleven of Cowan’s own representative
tests. 7 As a result, Cowan “re-weighted” the sample to “correct” for this issue. He then
provided that final sample to another one of plaintiffs’ experts, Steven Butler, to reunderwrite the loans in that sample. After Butler performed his analysis, Cowan then
extrapolated from the final sample of loans to the whole SLG to determine the percentage
of “materially defective” loans to a 95% confidence level with a maximum margin of error
of +/- 10 percent. (See Pls.’ Resp. (dkt. #224) 7.)
Defendants contend that: (1) Cowan failed to disclose his methodology in re-

The variables were FICO score, debt-to-income ratio, LTV ratio, CLTV ratio, note rate, original
loan amount, original term, documentation type, property type and loan purpose. (Cowan Rept.
(dkt. #219) ¶ 61.)

7

26

weighting the sample “instead relying on nothing more than his say-so”; and (2) the final
sample still fails to be representative. (Def.’s Br. (dkt. #185) 6.) Based on these flaws,
defendant contends that Cowan’s testimony on this Certificate is scientifically unreliable
and should be excluded. As for the first contention, defendant devotes much of its brief to
the initial sample having failed several of the representative tests, but Cowan freely
admitted this, explaining in his deposition that he used re-weighing “to make the sample
look like the population . . . [and] reduc[e] the variability of the sample.” (Id. at 10-11.)
Still, defendant criticizes Cowan for never explaining “precisely” why he engaged in reweighting for the CSFB 2005-3 Certificate and not for the SLGs underlying any of the
other Certificates. (Id. at 11.) While defendant obviously disputes its validity, however,
the reason Cowan gave for re-weighting could hardly be more straightforward -- the initial
sample of the SLG for the CSFB 2005-3 Certificate was not a representative sample.
Next, defendant contends that Cowan failed to disclose his method for reweighting
the Certificate, but that argument, too, is contradicted by Cowan’s report itself, which
discloses that he not only applied a “re-weighting adjustment,” but includes the appendices
to his report detailing his methodology, which were already provided to Credit Suisse. (See
Pls.’ Resp. (dkt. #224) 8 (citing Cowan Rept. (dkt. #219) ¶ 62 n.8).) Moreover, Credit
Suisse had the opportunity to, and, indeed, did, question Cowan about re-weighting at his
deposition. (See id. at 9 & n.15 (citing Cowan Dep. (dkt. #194) 72-76, 118-20, 125-29,
136-43, 166-69).)

The court, therefore, rejects defendant’s first basis for excluding

Cowan’s testimony, although defendant remains free to cross-examine Cowan about his reweighting methodology at trial.

27

As for its second contention, defendant argues that Cowan’s testimony regarding
CSFB 2005-3 is unreliable generally because the final sample still fails the representative
testing. In support, defendant simply points to Cowan’s own admission that the final
sample fails one of the representative tests. (Def.’s Br. (dkt. #185) 12 (citing Cowan Rept.
(dkt. #219) ¶ 43 n.56).) Cowan, however, testified at his deposition that failing one or
two of the 11 tests would be an expected result and does not indicate the sample is not
representative. (Pls.’ Resp. (dkt. #224) 15-16 (citing Cowan Dep. (dkt. #194) 133).)
Defendant’s remedy is again to cross-examine Cowan as to the soundness of his final
sample as representative of the loans in the SLG. Accordingly, this motion is DENIED.

G. MIL No. 7: Exclude Expert Testimony of Steven I. Butler (dkt. #188)
Finally, defendant seeks to exclude the testimony of plaintiffs’ expert Steven I.
Butler. As described in the context of other Daubert challenges, plaintiffs retained Butler
as a mortgage-banking expert to “reunderwrite” a sample of loans from the SLGs
underlying each of the Certificates at issue. Based on his analysis, Butler opines that more
than 60% of the sampled loans were materially defective, meaning that “they either
violated applicable underwriting guidelines or deviated from their represented
characteristics in a way that materially increased credit risk.” (Pls.’ Opp’n (dkt. #226) 7.)
Defendant proffers several reasons for striking Butler’s testimony. First, defendant
argues that Butler’s testimony is unreliable because he “bases his opinion in whole or in
part on the purported absence of a particular document from the file that he reviewed”
without knowing, or having any basis to know, whether the document was missing at the
time of loan origination or simply was lost from the loan file over the decade that followed

28

its origination. (Def.’s Br. (dkt. #189) 10-13.) As Butler explained in his report, however,
the industry-standard practice was to retain all documents considered in approving a loan
in the file. (Pls.’ Resp. (dkt. #226) 22 (citing Butler Rept. (dkt. #205) ¶¶ 58, 60, 174).)
Indeed, this was Credit Suisse’s own practice. Certainly, this is a fair basis to criticize
Butler’s ultimate opinion, but based on industry practice, he had at least some basis to
assume that the loan was not properly documented at the time of its origination. See Nat’l
Credit Union Admin. Bd. v. UBS Sec., LLC, No. 12-2591-JWL, 2017 WL 235013, at *11
(D. Kan. Jan. 19, 2017) (denying motion to exclude evidence based on same challenge,
finding reasonable basis to assume documents were missing at time of origination). Even
if defendant were correct that Butler had no basis for this factual assumption, this would
be a reason to give the opinion no weight, not to exclude the opinion. See Williams v.
Illinois, 567 U.S. 50, 132 S. Ct. 2221, 2228 (2012) (“Under settled evidence law, an expert
may express an opinion that is based on facts that the expert assumes, but does not know,
to be true. It is then up to the party who calls the expert to introduce other evidence
establishing the facts assumed by the expert.”). Of course, defense counsel remains free to
explore this further in cross-examination but the criticism still goes to weight, not
admissibility. 8
Second, defendant similarly argues that Butler relies on post-origination information

In support of this argument, defendant directs the court to one example where Butler opined that
the origination underwriter failed to obtain a fully complete version of the “VVOE form,” but “the
file loan’s approval noted a cleared condition reflecting that fully complete VVOE must have been
present before closing.” (Def.’s Br. (dkt. #189) 12.) Defendant’s own assumption that the original
underwriter would have only marked that condition as cleared if the form actually existed in the
first instance is itself subject to challenge.
8

29

in evaluating the sample loans, thus undermining his opinion that the origination
underwriters missed “red flags” in reviewing a mortgage application. Defendant essentially
points to two types of “post” origination information: (1) information that did not exist
at the time of the origination; and (2) information that existed but the underwriter did not
have. As to the former, the court agrees that Butler’s reliance on information only gained
after the origination of the loan may well undermine his opinion about misrepresentations
at the time of the origination. In its brief, however, defendant only points to a single
example of this involving a single loan: “Butler faulted the underwriter for not discovering
the borrower’s purchase of a property with first and second liens two days before the loan
closed,” while acknowledging that “there can be a lag time in information appearing on
public-record reports.” (Def.’s Br. (dkt. #189) 14 (citing Butler Dep. (dkt. #196) 286).)
Defendant is again free to cross-examine Butler about this one example, but absent more,
the court declines to strike Butler’s testimony as unreliable.
Indeed, the vast majority of post-origination evidence the parties discuss in their
briefing concern evidence that the original underwriter could have accessed at the time of
the origination of the loan if, at least in plaintiffs’ view, he or she had performed sufficient
due diligence. Defendant argues that Butler’s list of “red flags” is overbroad, is inconsistent
with industry standards and covers instances where the underwriter complied with industry
standards, but the borrower made misrepresentations.

Once again, however, these

criticisms do not go to the admissibility of Butler’s testimony. Instead, they concern
questions as to the weight his opinions should be given in light of: (1) whether Butler’s reunderwriting process demanded more than is required by industry standards for

30

underwriting at the time; (2) the scope of the representation in the offering documents
that the loans “were originated generally in accordance with the [applicable] underwriting
criteria”; and (3) whether that representation concerns process only or a promise that the
loans within the SLGs met the criteria set forth in the guidelines. See Fed. Hous. Fin. Agency
v. Nomura Holding Am., Inc., 74 F. Supp. 3d 639, 653 (S.D.N.Y. 2015) (discussing the role
of post-origination information in determining whether underwriters complied with
industry guidelines). No doubt, these are important questions, but ones the court need
not, and will not, address in the context of a motion in limine.
Third, defendant contends that Butler’s purported “industry standards” are
irrelevant and unreliable. Specifically, defendant takes issue with (1) Butler’s use of Bureau
of Labor Statistics data to assess the reasonableness of stated income loans and (2) the
rigor with which Butler and his team applied that data in reviewing loans. Yet again, this
is an area defendant may explore in cross-examination, as well as ultimately argue that
Butler’s standards are too strict and do not reflect industry standards. Defendant has,
however, offered no basis for excluding Butler’s testimony.

Moreover, the thrust of

defendant’s argument concerns the nature and scope of the promise made in offering
documents, which is an issue the court will not address in this opinion and order on the
parties’ motions in limine.
Finally, defendant challenges Butler’s comparison of the credit characteristics of
each individual loan as listed in the tapes or mortgage loan schedules (“MLS”), with the
loan’s “true” credit characteristics.

(He concluded that the true credit characteristics

differed from those in the MLS, approximately 25% of the time.) Defendant contends

31

that this evidence was irrelevant because the MLS was not used to construct the collateral
tables in the prospectus supplements and that, even if it were, the loan characteristics were
presented in an aggregate form, not at the individual loan level. (Def.’s Br. (dkt. #189)
20.) In response, plaintiffs argue that a separate promise in the offering documents,
representing that the MLS data was accurate, make Butler’s MLS findings relevant. (Pls.’
Resp. (dkt. #226) 18 (citing language from PSAs and Mass Mut. Life Ins. Co. v. DB
Structured Prods., Inc., No. CIV.A. 11-30039-MGM, 2015 WL 3964560, at * 12 (D. Mass.
June 19, 2015) (holding that PSAs were “incorporated by reference into the prospective
supplements”)).) In the alternative, plaintiffs contend that a reasonable investor would at
least read the offering documents as implicitly representing that the MLS data was accurate.
As evidenced by the parties’ briefing, this challenge similarly concerns a dispute as
to the nature and scope of promises made in the offering documents, rather than the
content of Butler’s opinion. As such, the court will deny this motion as well, and will
consider Butler’s testimony if the court finds that Credit Suisse made representations about
the MLS data at the individual loan level.
Having rejected all of defendant’s bases for excluding Butler’s testimony,
defendant’s final motion in limine is DENIED.

ORDER
IT IS ORDERED that:
1) Plaintiffs CMFG Life Insurance Company, CUMIS Insurance Society and
MEMBERS Life Insurance Company’s motion in limine no. 1 to admit Credit
Suisse’s RMBS settlement agreement with the Department of Justice (dkt.
#173) is GRANTED as set forth above.

32

2) Plaintiffs’ motion in limine no. 2 to exclude the expert testimony of William N.
Goetzmann, Ph.D. (dkt. #176) is GRANTED IN PART as to Goetzmann’s first
regression analysis and DENIED IN PART as to his second regression analysis
and macroeconomic conditions.
3) Plaintiffs’ motion in limine no. 3 to exclude evidence of post-transaction loss
causation (dkt. #179) is DENIED as set forth above.
4) Defendant Credit Suisse Securities (USA) LLC’s omnibus motions in limine
(dkt. #186) are
a) motion in limine no. 1 is DENIED as set forth above;
b) motion in limine no. 2 is DENIED as set forth above;
c) motion in limine no. 3 is DENIED, except that the court will ignore any
statement by individuals in the report for the truth of the matter asserted;
and
d) motion in limine no. 4 is DENIED.
5) Defendant’s motion in limine no. 5 to partially exclude the expert testimony of
Leonard A. Blum (dkt. #182) is GRANTED as set forth above.
6) Defendant’s motion in limine no. 6 to exclude the expert testimony of Dr.
Charles Cowan regarding the CSFB 2005-3 certificate (dkt. #184) is DENIED.
7) Defendant’s motion in limine no. 7 to exclude expert testimony of Steven I.
Butler (dkt. #188) is DENIED.
Entered this 23rd day of October, 2017.
BY THE COURT:

/s/
__________________________________
WILLIAM M. CONLEY
District Judge

33