U.S. Bank, Nat'l Ass'n v. UBS Real Estate Sec. Inc.

Full title: U.S. BANK, NATIONAL ASSOCIATION, solely in its capacity as Trustee of…

Court: United States District Court, S.D. New York.

Date published: Sep 6, 2016

205 F. Supp. 3d 386 (S.D.N.Y. 2016)

Opinion

12-cv-7322 (PKC) (JCF)

09-06-2016

U.S. BANK, NATIONAL ASSOCIATION, solely in its capacity as Trustee of Mastr Adjustable Rate Mortgages Trust 2006-OA2, Mastr Adjustable Rate Mortgages Trust 2007-1, and Mastr Adjustable Rate Mortgages Trust 2006-3, Plaintiffs, v. UBS REAL ESTATE SECURITIES INC., Defendant.

Sean P. Baldwin, David L. Elsberg, Andrew R. Dunlap, Steven M. Edwards, Nicholas F. Joseph, Paul P. Hughes, Mario A. Pecora, QUINN EMANUEL URQUHART & SULLIVAN LLP, Attorneys for Plaintiff Scott D. Musoff, Robert A. Fumerton, Thomas J. Nolan, Charles F. Smith, Alexander C. Drylewski, Robert L. Dunn, Allison B. Holcombe, Abraham A. Tabaie, Lara A. Flath, SKADDEN, ARPS, SLATE, MEAGHER & FLOM LLP, Attorneys for Defendant

CASTEL, United States District Judge.

Sean P. Baldwin, David L. Elsberg, Andrew R. Dunlap, Steven M. Edwards, Nicholas F. Joseph, Paul P. Hughes, Mario A. Pecora, QUINN EMANUEL URQUHART & SULLIVAN LLP, Attorneys for Plaintiff

Scott D. Musoff, Robert A. Fumerton, Thomas J. Nolan, Charles F. Smith, Alexander C. Drylewski, Robert L. Dunn, Allison B. Holcombe, Abraham A. Tabaie, Lara A. Flath, SKADDEN, ARPS, SLATE, MEAGHER & FLOM LLP, Attorneys for Defendant

MEMORANDUM AND ORDER

CASTEL, United States District Judge.

OVERVIEW OF THE AGREEMENTS AND CLAIM ...400

SUBJECT MATTER JURISDICTION ...405

A. It Is Unlikely that the Court Ever Had Bankruptcy Jurisdiction ...406

B. The Court Has Diversity Jurisdiction ...408

1. Legal Standard ...408

2. Diversity Jurisdiction Existed When the Action Commenced ...408

3. Judge Baer Concluded that the Action Was Brought by U.S. Bank on Behalf of the Trusts...409

4. U.S. Bank Is the Real Party to the Controversy ...409

5. The Recent Americold Decision Does Not Alter This Analysis ...410

6. The Amended Complaint Expressly Alleges Diversity Jurisdiction ...411

CHOICE OF LAW ...411BURDEN OF PROOF ...411

RELEVANT NEW YORK CONTRACT PRINCIPLES ...412

A. The PSAs' Unambiguous Terms Are Construed to Reflect the Intent of the Parties as Manifested in the Agreements ...412

B. Parol Evidence Plays No Role in Interpreting an Unambiguous Contract ...413

C. No Party Asserts that Contra Preferentem Applies ...413

D. Contractual Warranties under New York Law ...414

THE TRUSTS MAY RECOVER THE MONEY DAMAGE EQUIVALENT OF THE REPURCHASE REMEDY FOR BREACHED LOANS WHERE REPURCHASE IS NOT POSSIBLE ...414

THE TRUSTS, THE TRUSTEE AND RELATED PERSONS ...415

THE NOTICE AND DISCOVERY REQUIREMENT ...416

A. The Requirements of Section 2.03 ...416

1. UBS Received Written Breach Notices of that Identified 4,869 Breached Loans, Including 4,462 Breaches that Were Timely Identified Before the Commencement of this Action ...417

a. The Assured Breach Notices ...418

b. The U.S. Bank Breach Notices...420

2. Upon Receipt of the Holt Report of August 2015, UBS Had Knowledge of Purported Breaches in the Loans Identified Therein...420

3. Lapse of the 90-Day Cure Period as Condition Precedent...422

4. Discovery of a Breach by UBS Triggers Its Cure and Repurchase Obligations...423

5. Knowledge on the Part of UBS Cannot Be Premised on "Pervasive Breach" or "Constructive Knowledge...424

6. UBS's Knowledge of a Breach May Be Proved Circumstantially or Through Evidence of Willful Blindness...425

a. The Trusts Have Proved Discovery of Breaches as to Certain Non-Noticed Loans...425

b. The Trusts Have Not Proved Willful Blindness on the Part of UBS...425

THE MORTGAGE LOAN SCHEDULE WARRANTY...427

A. Overview of the MLS Warranty...427

B. The Timing of Representation and Warranty Listed in the MLS...427

C. The MLS Warranty is Absolute and Does Not Merely Guarantee Accurate Transcription...428

D. The Trusts' Evidence Concerning MLS Warranty Breaches...430

E. The Seven Items of MLS Data at Issue...430

1. Occupancy Status...430

2. Property Type...431

3. Loan Purpose...431

4. LTV Ratio...432

a. Construction...432

b. AVM Data Is Not A Substitute for An Appraisal...434

5. Documentation Style...437

6. Credit Score...437

a. Construction...437

b. Application...438

7. DTI Ratio...439

a. Construction...439

b. Application...440

c. The MERS Database...441

d. Other Databases...441

e. Multiple Credit Inquiries...442

f. Credit Reports...442

g. Evidence of Multiple Misrepresentations as Evidence that a Borrower Also Misrepresented Debt...443

h. Social Security Administration Data and Pension Custodian Statements...443

i. Employer Income Statements, Including W-2 Forms...443

j. BLS Data...444

k. Other Salary Databases...444

l. Borrower Tax Returns...445

m. Borrower Bankruptcy Filings...445

n. Borrower Hardship Letters...446

o. Red Flags...447

THE GUIDELINES WARRANTY...447

A. Overview of the Guidelines Warranty and Its Relationship to the MLS Warranty...447

B. Automated Underwriting Systems...448

C. Exceptions to the Guidelines Must Be Documented...449

D. The Loan Files Are Provisionally Received Into Evidence...451

E. Permissible Use of Post-Origination Information by an Expert...455

F. Individual Guidelines Breaches...456

1. The Absence of "Core Documents" in Loan Files...456

2. Failure to Inquire As to the Stated Income That Does Not Appear Reasonable...457

3. Guidelines Exception...458

4. The Relationship between the MLS Warranty and the Guidelines Warranty...459

THE MAXIMUM LTV WARRANTY...460

THE MORTGAGE FILE WARRANTY...460

THE TITLE INSURANCE WARRANTY ...461

THE HAZARD INSURANCE WARRANTY...463

THE "MATERIALLY AND ADVERSELY AFFECTS" REQUIREMENT...464

A. "Materially and Adversely Affects" Is Measured from the Date of the Repurchase Demand, Not from the Date that the Warranties Were Made...464

1. The "Materially and Adversely Affects" Requirement of Section 2.03 Is Analyzed Separately from the Materiality Requirement Contained in Individual Representations and Warranties...467

2. A Material Adverse Effect under Section 2.03 May Be Proved with Evidence of an Increased Risk of Loss to the Certificateholders...467

3. The Trusts May Also Prove a Material Adverse Effect with Evidence that a Breach Resulted in Altered Loan Terms ...468

4. Multiple Breaches as Evidence of Material and Adverse Effect ...469

5. Proof of Intentional Misrepresentations by a Borrower Will Prove a Material and Adverse Effect...470

6. The Trusts' Proof of a Material and Adverse Effect...470

THE "MISSING" INDYMAC FILES...475

EXEMPLAR LOANS...476

A. Loan 1456451...478

1. Occupancy Misrepresentation...478

2. Income Misrepresentation with Red Flags...480

3. Material and Adverse Effect at the Time of Discovery or Notice...480

B. Loan 40577613...481

1. Income Misrepresentation...481

2. The Presence of Undisclosed Mortgages...482

3. Missing Title Insurance...483

4. Employment Verification...4835. Housing History Violation...483

6. Material and Adverse Effect at the Time of Discovery or Notice...483

C. Loan 40599698...484

1. Misrepresented Income...484

2. Undisclosed Mortgage...485

3. Resulting DTI Breaches...485

4. LTV Breaches...485

5. Occupancy Misrepresentation...485

6. Missing Title Insurance...486

7. Material and Adverse Effect at the Time of Discovery or Notice...486

D. Loan 40588193...486

1. Income Misrepresentation and/or Lack of Reasonableness Inquiry...487

2. Mortgage Payment History...487

3. Verification of Employment...488

4. Missing Title Insurance...488

5. No Breach...488

E. Loan 1447951...488

1. Employment Misrepresentation...489

2. Reasonableness of Stated Income and Income Misrepresentation...489

3. The LTV Ratio Defects...490

4. Material and Adverse Effect at the Time of Discovery or Notice...490

F. Loan 40595372...490

1. Employment Verification...490

2. Missing Note...491

3. Missing Title Insurance...491

4. No Breach...491

G. Loan 1416352...492

1. Undisclosed Mortgage...492

2. Income Misrepresentation and Reasonableness of Stated Income ...492

3. Improper Calculation of Debt...493

4. Housing History Violation...493

5. Arm's Length Violation...494

6. The LTV Ratio Defects...494

7. Material and Adverse Effect at the Time of Discovery or Notice...494

H. Loan 1450507...495

1. Employment Misrepresentation...495

2. Document Defect...495

3. Missing Title Insurance...496

4. Correlated DTI Discrepancy...496

5. Material and Adverse Effect at the Time of Discovery or Notice...496

I. Loan 40595823...497

1. Income Misrepresentation...497

2. Undisclosed Mortgage...498

3. Missing or Inadequate Title Insurance...498

4. The LTV Ratio Defects...498

5. Material and Adverse Effect at the Time of Discovery or Notice...498

J. Loan 158522879...499

1. Undisclosed Mortgage Debt...499

2. Asset Violations...500

3. Missing Loan Application...500

4. Employment Violation...501

5. Material and Adverse Effect at the Time of Discovery or Notice...501

K. Loan 124359279...502

1. Document Deficiency...502

2. The Trust Has Not Proved a Material and Adverse Effect on the Interests of the Certificateholders at the Time of Notice or Discovery...502

L. Loan 1423423...502

1. Income Misrepresentation...503

2. LTV Ratio Defects...504

3. Material and Adverse Effect at the Time of Discovery or Notice...504

M. Loan 138058233...504

1. Income Misrepresentation...504

2. Seasoning Violation...5053. Missing or Inadequate Title Insurance...506

4. LTV Ratio Defects...506

5. No Breach...506

N. Loan 1418398...506

1. Occupancy Misrepresentation and Income Violation...507

2. Breach of the MLS Warranty Based on Misstated DTI Ratio...508

3. Correlated DTI Discrepancy...509

4. No Breach...509

O. Loan 1477160...509

1. Unreasonable Stated Income...509

2. Undisclosed Debt...510

3. The Borrower's Occupancy Representation...510

4. Correlated DTI Ratio Defects...511

5. The LTV Ratio Defects...512

6. Material and Adverse Effect at the Time of Discovery or Notice...512

P. Loan 158514566...513

1. Income Misrepresentation and Reasonableness of Stated Income...513

2. DTI Ratio...514

3. Document Deficiency...515

4. Title Insurance...515

5. LTV Breaches...515

6. Material and Adverse Effect at the Time of Discovery or Notice...515

Q. Loan 1470345...516

1. Unreasonableness of Stated Income...516

2. Undisclosed Mortgage Debt...517

3. LTV Breaches...517

4. Material and Adverse Effect at the Time of Discovery or Notice...518

R. Loan 158609489...518

1. Income Misrepresentation...518

2. Occupancy Misrepresentation and the Related "Property Violation" Defect... 519

3. Undisclosed Mortgage Debt with Red Flags...521

4. DTI Ratio Defect...521

5. Material and Adverse Effect at the Time of Discovery or Notice...521

S. Loan 1416080...522

1. Undisclosed Mortgage Debt...522

2. Housing History Violation...522

3. DTI Ratio Defect...523

4. Material and Adverse Effect at the Time of Discovery or Notice...523

T. Loan 1437593...523

1. Income Misrepresentation...524

2. Correlated DTI Ratio Defects...524

3. Missing Title Insurance...524

4. LTV Breaches...525

5. Material and Adverse Effect at the Time of Discovery or Notice...525

CONCLUSION: STEPS TO FINAL JUDGMENT...525

A. Supplemental Submissions on Loan Files...525

B. Appointment of Masters...526

This is an action for breach of contract brought against UBS Real Estate Securities Inc. ("UBS") by U.S. Bank, N.A., as trustee of three trusts (the "Trustee"). The complaint asserts a single claim under three similarly-worded agreements. Specifically, the Trustee alleges that UBS breached certain representations and warranties made as to 9,342 of the 17,082 mortgage loans that were originally pooled in the three trusts, and that they are entitled to more than $2 billion in damages.

The action was tried to the Court without a jury. The Court now issues its findings of fact and conclusions of law pursuant to Rule 52, Fed. R. Civ. P. Citations to the record are for exemplary purposes and not intended to be an exhaustive listing of all record support for a statement.OVERVIEW OF THE AGREEMENTS AND CLAIM.

A home buyer seeking to finance the purchase of a home, or an owner seeking to refinance an existing mortgage, applied for a loan to an institution, directly or through an intermediary such as a mortgage broker. If the application was approved, the institution loaned money to the borrower secured by a mortgage on the borrower's home. These institutions, in the parlance of the contracts at issue, are loan "Originators." After making hundreds or thousands of residential loans, Originators would sell pools of loans to institutional buyers who either purchased and held the loans for their own accounts or placed them in a securities offering and sold the securities to investors.

In the 2006-07 time period, UBS was the successful bidder on several pools of loans offered for sale by Originators, including Countrywide Home Loans, Inc. ("Countrywide"), IndyMac Bank, F.S.B. ("IndyMac"), American Home Mortgage Investment Corp. ("American Home"), Residential Funding Company, LLC, MortgageIT, Inc. and Chevy Chase Bank, F.S.B. Most of the loans were "Alt-A" or "subprime" loans, which generally carried a higher risk of default than prime loans. (DX KJ, Lantz Trial Dec. ¶¶ 13-14; PX 99 at 42.) Many were originated under reduced-documentation programs. (DX KJ, Lantz Trial Dec. ¶ 14.) For example, an Originator who maintained a "stated income" program permitted the borrower to simply "state" his or her income on the loan application without verification by the Originator of the actual income. (Id. )

Typically, before making a bid on a pool of loans offered by an Originator, UBS conducted a due diligence review on about 25% of the loans offered. UBS retained vendors to examine the Originator's loan files, which contained information developed through the Originator's underwriting process. On occasion, the vendor would report to UBS that the loan did not comply with the underwriter's internal guidelines or that documents were missing from the file. This would prompt a discussion in which the selling Originator would either cure the problem or withdraw the loan from the pool. For the other 75% of the loans not selected for a due diligence review, UBS relied on a "data tape" or "loan tape" that provided information about the loans offered for sale, listing in words or code the type of loan (e.g. "full documentation" or "stated income"), whether the property was owner-occupied, the borrower's credit score, the debt-to-income ("DTI") ratio, the loan-to-value ("LTV") ratio and other data.

After completing the purchase of the loans from several unrelated Originators, UBS structured transactions in which a trust was established and, in turn, the trusts purchased the loans from UBS and held them. Principal and interest from the loans was paid into the trust, and certificates for a fractional share of the right to payment of principal and interest were sold to investors known in the language of the contracts as "Certificateholders." In essence, the right to payment of principal and interest in the pool of loans was converted into a security, as that term is used in the federal securities laws, and hence the generic label applied to transactions of this type of "Residential Mortgage-Backed Securities," or "RMBS." This action is brought by the Trustee on behalf of three Trusts created by UBS: MASTR Adjustable Rate Mortgages Trust 2006-OA2, MASTR Adjustable Rate Mortgages Trust 2007-1, and MASTR Adjustable Rate Mortgages Trust 2007-3 (collectively, the "Trusts")

To induce investors to purchase certificates, the UBS made certain representations and warranties in a Pooling and Service Agreement ("PSA") between UBS, the Trust and certain others. The PSAs' representations and warranties relieved the Trusts and the would-be Certificateholders of the need to conduct due diligence on the pool of loans. For example, the data or loan tapes obtained from Originators, with certain adjustments, became the Mortgage Loan Schedule ("MLS") incorporated into the PSA, and the PSA warranted that the information therein was true and correct as of certain dates "in all material respects." Another important warranty was that the loans were underwritten in accordance with the Originators' underwriting guidelines, with reasonable exceptions exercised by the Originator. Notably, unlike some transactions of this kind, there was no warranty against fraud on the part of an individual borrower.

Remarkably, the overwhelming majority of loans in the Trusts were for the purpose of refinancing a home already owned by the borrower, rather than for the purpose of purchasing new property. Of the approximately 12,400 loans held by the Trusts that have not been fully repaid, the parties agree that approximately 10,000 of them, or 80%, were obtained as some form of refinancing. (Tr. 1960-61.) Undoubtedly, many of these were for the purpose of monetizing and extracting rising equity for other use.

The Trusts' lead expert opined that as many as 38% of the loans at issue were "predicated on [borrower] fraud or misrepresentation of income, employment or occupancy." (Holt Dec ¶ 18.) While the Court does not accept the figure as established fact, the Court does find many proven instances of intentional misrepresentations by borrowers. There is no evidence in the record that any individual borrower was prosecuted for bank fraud or false statements to the lender.

The PSA provided the Trustee of a Trust with a narrow and specific remedy if it learned that UBS had breached a representation or warranty as to a given loan held by the Trust. Upon discovery or notice of a breach, section 2.03 of the PSAs obligated UBS to do one of the following within 90 days: (a) cure the breach; (b) replace the loan with a suitable substitute; or (c) repurchase the loan from the Trust.

This cure, replace or repurchase obligation was triggered if, but only if, the noticed or discovered breach "materially and adversely affects the interest of the Certificateholders ... in any Mortgage Loan...." (PSA § 2.03.) As will be discussed, in using the present-tense word "affects," the provision required an examination of whether the interest of the Certificateholders was affected at the time of discovery or notice, potentially years after origination of the loan, and not merely at the time the breach of warranty occurred. Notably, the PSA provided that breaches of certain representations and warranties "will be deemed automatically to materially and adversely affect the interests of the Certificateholders in such Mortgage Loan...." (PSA § 2.03.) But none of those "deemed automatically" representations and warranties are at issue in this action.

With regard to the discovery or notice requirement, this Court previously held that discovery of a breach by UBS triggers the cure, replace or repurchase 90-day clock, as does actual notice given by another party. MASTR Adjustable Rate Mortgages Trust 2006 – OA2 v. UBS Real Estate Sec. Inc. , 2015 WL 764665, at *6–10 (S.D.N.Y. Jan. 9, 2015), reconsideration denied , 2015 WL 797972 (S.D.N.Y. Feb. 25, 2015). But the Court rejected the Trusts' "pervasive breach" theory, which argued that the language of the PSA should be construed to deem UBS on notice of all defective loans based on its knowledge that several loans were defective. Id. at *10–12.The PSAs were explicit that the "sole remedies" in the event of a breach that materially affected the interests of the Certificateholders were cure, replacement or repurchase: "It is understood that the obligations under this Agreement of [UBS] to cure, repurchase or replace any Mortgage Loan as to which breach has occurred and is continuing shall constitute the sole remedies against [UBS]...." The late Judge Baer held—and this Court has since confirmed—that the money damages equivalent of the repurchase remedy is available for a loan that has been liquidated and, therefore, cannot be repurchased. See MASTR Adjustable Rate Mortgages Trust 2006 – OA2 v. UBS Real Estate Sec. Inc. , 2013 WL 4399210, at *2 (S.D.N.Y. Aug. 15, 2013) (Baer, J.); MASTR , 2015 WL 764665, at *17–18. The "sole remedies" provision, however, is a valid and enforceable contractual limitation that forecloses efforts by the Trust to fashion an expedient remedy based solely on the volume of loans at issue.

In this action, the Trusts seek to have UBS repurchase certain loans and, if the loan has been liquidated, to pay the money damages equivalent of repurchase. As to each mortgage loan, the Trust must prove that: (1) there was a breach of a warranty; (2) UBS either discovered the breach on its own or was placed on notice of the breach in a timely and proper manner and did not cure, replace or repurchase within 90 days; and (3) at the time of discovery or notice, the breach materially and adversely affected the interests of Certificateholders.

OVERVIEW OF THE TRIAL.

After hearing the parties' proposals, the Court allotted the two sides an evenly-divided total of 74 hours to present their case, a length slightly more than requested by the Trusts (70 hours) and somewhat less than requested by UBS (105 hours). At trial, neither side used all of its time or requested a modification of the time limit. The trial took place during the period from April 18 to May 13, 2016. The parties submitted direct testimony of all witnesses within their control through written declarations. Each witness was subject to cross-examination and redirect testimony.

The parties represented to the Court that the trial exhibits consumed about 2 terabytes of data.

The Trusts called five fact witnesses. Four were former UBS employees: Adrian Wu, a trader who purchased many of the Loans and structured the Trusts; William Twombly, the manager of UBS's Due Diligence Group, who oversaw the pre-securitization diligence that UBS performed on the Loans; Christopher Schmidt, the co-manager of UBS's Mortgage Finance Group, who oversaw the execution of the Trusts; and Jonathan Lantz, a manager in UBS's Surveillance Group, who monitored the performance of the Loans and, at times, conducted post-securitization reviews of certain Loans. The Trusts' final fact witness was Diane Reynolds, a senior vice president of the Structured Finance Group at the Trustee, who testified about the Trusts and the notices of breach that UBS received.

The Court received into evidence the deposition designations offered by the Trusts as to two unavailable witnesses: John Williams, the Rule 30(b)(6) designee for JCIII & Associates, a vendor hired by UBS to review the loans subject to Assured's repurchase demands, and Mujtaba Mohiuddin, a due diligence manager at UBS who reported to Twombly. UBS offered no deposition designations.

The Trusts called five expert witnesses at trial: Ira H. Holt, an employee of Analytic Focus LLC, who testified about his re-underwriting review and identification of defects in the loans; Nelson Lipshutz, Ph.D., President of Regulatory Corporation, an economics and statistical consulting firm, who testified about his statistical extrapolations of Holt's findings for Loans originated by IndyMac Bank; Charles M. Cowan, Ph.D., Managing Partner of Analytic Focus LLC, who testified about his use of an "automated valuation model" ("AVM") and statistical analysis to evaluate the reported appraisal values for the Loans and their corresponding LTV ratios; and Karl N. Snow Ph.D., a partner in the firm Bates White Economic Consulting, who testified about the methods of calculating the Trusts' recoveries consistent with the PSAs.

The Trusts called a witness, Charles Cipione, who testified about his methods for summarizing large volumes of relevant data in a series of databases. Those databases summarized expert reports and testimony, distribution reports and servicing data, repurchase demand letters, loan tapes and the results of reviews performed by UBS and its vendors in the course of due diligence, surveillance, and assessing loans for potential repurchase. The Court received these summaries pursuant to Federal Rule of Evidence 1006.

UBS called three expert witnesses: Deborah Grissom, a Senior Director at Treliant Risk Advisors, who testified in response to Holt's underwriting analysis; Arnold Barnett, Ph.D., a professor of statistics at the Sloan School of Management, Massachusetts Institute of Technology, who testified in response to Lipshutz and Cowan; and Andrew S. Carron, Ph.D., chairman of NERA Economic Consulting, who testified in response to Snow.

In identifying a breach of warranty, each side placed principal reliance on experts who, with the use of vendors employing hundreds of individual underwriters, examined the underwriting guidelines and the loan files. Holt, the Trusts' principal expert, used three firms: The Barrent Group, LLC, Opus Capital Markets Consultants, LLC and Digital Risk, LLC. (Tr. 850; Holt Direct ¶ 123.) "[S]everal hundred individuals" at the three companies were involved in the review of loan files. (Tr. 850.) Grissom, UBS's principal expert, relied on a team of approximately 135 people employed by Treliant Risk Advisors and FTI Consulting, Inc. (Tr. 302 & Grissom Dec. ¶ 34.)

At times, Holt and Grissom described the review process as a "re-underwriting" process. (See, e.g. , Holt Direct ¶¶ 3, 5, 6; Tr. 270, 819.) It was not, in fact, a re-underwriting process. The underwriting process seeks to answer the question of whether an application should be approved and a loan funded. Here, Holt's team looked for potential breaches of representations and warranties and recorded them when, in the opinion of the reviewer and Holt, breaches were found. Grissom's people had the benefit of Holt's work and endeavored to rebut many of the Trusts claims and Holt's opinions. Rather than traditional re-underwriting, these reviews were directed to the existence of potential breaches.

The backgrounds and professional experience of the individual reviewers has been only vaguely described. (See, e.g. , Tr. 303 ("Grissom: ... These were all experienced either underwriters or reunderwriters. I didn't feel that I needed to explain to them how to calculate a DTI or, you know, some of the basic tenets of reunderwriting. They all had that experience."); Holt Direct ¶ 123 (describing "a support team of re-underwriting professionals ....").) Neither party endeavored to show the extent to which the hundreds of individuals who worked on the loan review process were the same individuals employed in the process of underwriting subprime loans for other originators or institutions in the years leading up to the collapse of the financial markets in 2008. Perhaps neither side thought to explore the issue, or else thought that pursuit of the issue would lead to naught; then again, perhaps, each side thought it counterproductive because it would cast doubt on their own expert's work. The Court places no weight on this undeveloped issue.

Holt has a B.S. degree from the University of Alabama and a M.A. in Public and Private Management from Birmingham-Southern College. He has worked in the mortgage underwriting industry for 25 years. (Holt Direct ¶ 28.)

Holt also stated in his direct testimony that he "received ... a Graduate Degree in Retail Bank Management from the University of Virginia [ ("UVA") ] in 1994." (Holt Direct ¶ 28.) He conceded on cross-examination that the statement was not correct. (Tr. 752.) In fact, he attended a series of courses presented by the Consumer Banking Association ("CBA") on the UVA campus. (Id. ) In claiming a graduate degree from UVA, Holt overstated his credentials, aided by CBA, which labeled the certificate of completion a "Diploma of Graduation" and obtained the signature of the Dean of the UVA business school as a co-signer. (PX 1105.) It is a matter that the Court takes account of in assessing the overall credibility of Holt and the value of his opinions.

Holt also claimed in his direct testimony that he "found 39,654 Material Defects in these 9,831 Loans, consisting of 7,626 Data Defects, 7,439 Guideline Violation Defects and 3,785 Other Defects." (Holt Direct ¶ 16.) In fact, the three cited categories total 18,850 defects, less than half of the claimed total. When confronted, Holt adhered to the figure of 39,654 but could not explain the discrepancy in his direct testimony. (Tr. 762-64.) Holt was unable to explain other discrepancies in his overall findings, including an approximately 6,000-loan file discrepancy in the number of files that were subject to due diligence review by UBS. (Tr. 765-774.) Holt's inattentiveness to detail is taken into account in assessing his overall credibility and the value of his opinions. Other shortcomings in Holt's opinions are noted in context.

UBS's expert, Deborah Grissom holds a B.A. in Accounting and Business Administration from Carthage College and an MBA from the University of San Diego. She has 30 years of experience in the mortgage industry. (Ex. JD ¶ 1.)

Grissom's testimony was also flawed in several respects. Grissom reviewed approximately 1,500 of the loans at issue. (Tr. 325-26.) She testified that when she disagreed with the findings of a vendor whose employees reviewed loan files, she simply altered the conclusions on the document transmitted by the vendor. Grissom testified that she altered findings for approximately 20 loans, but was unable to identify which loans she revised. (Tr. 313-16.) Grissom also was unable to identify which 1,500 loans she personally reviewed (Tr. 333), and testified that "they all kind of run together at some point." (Tr. 578.) It is not possible to reconstruct which conclusions were proposed by the vendor and adopted by Grissom and which reflect her own, substituted view.

Grissom did not provide her reviewers with written guidance and testified that she never personally met with approximately half of the reviewing team. (Tr. 303-04, 556.) Grissom testified that for approximately 40 to 45 reviewers, she never provided any type of instruction, including oral instruction. (Tr. 312-13.) Her guidance to the reviewers principally directed them to use "common sense" and to undertake a "holistic review" of the loan files. (Tr. 317-19.)

Grissom at times engaged in circular reasoning. As explained more fully below, when she examined an Originator-approved loan that did not comply with the guidelines, she assumed, without any notation or other evidence in the loan file, that an authorized underwriter within the Originator had exercised an "exception" to the guidelines. (See, e.g. , Tr. 331 ("the exception was granted because the loan was made ....").) She looked to see whether an "exception" could have been retrospectively justified rather than whether an "exception" was, in fact, exercised during underwriting. In essence, she reasoned that when a loan failed to satisfy the guidelines but was approved and funded, some person in a position of authority at the Originator necessarily granted an exception.

Nevertheless, the Court finds that there is sufficient value and reliability to the work of Holt and Grissom to pass the threshold for admissibility. Rule 702, Fed. R. Evid.; Kumho Tire Co. v. Carmichael , 526 U.S. 137, 119 S.Ct. 1167, 143 L.Ed.2d 238 (1999). Holt and Grissom both have extensive experience in underwriting residential mortgage loans and the mortgage industry. Both are qualified to opine about the practices of reasonable underwriters employed in the industry, the contents of borrower loan applications and the considerations that an underwriter weighs during the loan-approval process. They have expertise in the application of underwriting guidelines and in identifying and interpreting the key data points and documents contained in borrower loan files. They are familiar with the resources and databases available to underwriters, and the uses and limitations associated with them. Given that the underwriting process draws heavily on judgment and takes into consideration the different qualities of each loan and borrower, the Court concludes that the opinions of Grissom and Holt applied their specialized knowledge using sufficiently reliable methods, and that their opinions are of assistance to the Court in its role as the finder of fact. See Rule 702 ; see also Assured Guar. Mun. Corp. v. Flagstar Bank, FSB , 920 F.Supp.2d 475, 505–06 (S.D.N.Y.2013) (describing the necessarily subjective and judgment-intensive nature of expert testimony on underwriting breaches).

The weight to be accorded and the degree of acceptance of each opinion are different matters, which will be discussed in the context of specific issues arising in this case.

The Court's findings are informed by the work of Holt and Grissom and their vendors, but that does not mean that the opinions of either are accepted or rejected wholesale. As the trier of fact, the Court may accept so much of an expert's opinion as accords with reason, common sense and the evidence, and reject the balance.

The parties were given the opportunity for closing arguments and post-trial submissions, the last of which was received on August 23, 2016.

SUBJECT MATTER JURISDICTION.

"Courts have an independent obligation to determine whether subject-matter jurisdiction exists, even when no party challenges it." Hertz Corp. v. Friend , 559 U.S. 77, 94, 130 S.Ct. 1181, 175 L.Ed.2d 1029 (2010). The initial Complaint invoked bankruptcy jurisdiction as the sole grounds for federal subject matter jurisdiction. After trial, the Court directed the parties to submit letter-briefs addressing the Court's subject-matter jurisdiction. The Trusts and UBS both contend that the Court has subject matter jurisdiction based on "related to" bankruptcy jurisdiction under 28 U.S.C. § 1334(b), as well as diversity of citizenship under 28 U.S.C. § 1332(a)(1), which was not alleged in the initial Complaint. With leave of the Court, on August 2, 2016, the Trusts filed an Amended Complaint that expressly invoked diversity jurisdiction.

For the reasons explained, the Court concludes that it has subject matter jurisdiction by reason of diversity of citizenship and an amount in controversy exceeding $75,000, exclusive of interest and costs. 28 U.S.C. § 1332(a).

A. It Is Unlikely that the Court Ever Had Bankruptcy Jurisdiction.

The Complaint invoked subject matter jurisdiction solely by reason of the jurisdictional grant over actions arising under title 11 of the U.S. Code or arising in or related to a bankruptcy proceeding. (Docket # 1.) In its entirety, the jurisdictional allegation states as follows: "This Court has jurisdiction over this proceeding pursuant to 28 U.S.C. § 1334(b)." (Compl't ¶ 12.) The Complaint makes no supporting allegations.

The Court, on its own, directed the parties to explain in writing the basis for invoking bankruptcy jurisdiction. (Docket # 422.) In response, the parties asserted that this Court had "related to" bankruptcy jurisdiction because certain loans held by the Trusts were originated by American Home. American Home filed for Chapter 11 bankruptcy protection in August 2007. See In re American Home Mortgage Holdings, Inc. , 07-11047 (Bankr. Del.) (CSS). On January 11, 2008, UBS submitted a Proof of Claim in the American Home bankruptcy proceedings for an aggregate amount of $8,498,893.06, plus "all damages, fees and related costs" arising from American Home's breaches of representations and warranties. (Docket # 430 Ex. A.) Prior to the commencement of this action, on February 23, 2009, the bankruptcy court confirmed the Amended Chapter 11 Plan of Liquidation for American Home. (Docket # 428 Ex. 7.) According to the parties, this Court had "related to" bankruptcy jurisdiction because at the time that the Trusts commenced this action, UBS had outstanding proofs of claim in the bankruptcy proceedings that sought indemnification from American Home. But they were claims that appeared facially barred by the confirmed plan of liquidation—and, as the Court came to learn, related to only one of the three Trusts.

In determining whether a court has bankruptcy jurisdiction, "[p]roceedings ‘related to’ the bankruptcy include ... suits between third parties which have an effect on the bankruptcy estate." Celotex Corp. v. Edwards , 514 U.S. 300, 307 n. 5, 115 S.Ct. 1493, 131 L.Ed.2d 403 (1995). The standard for exercising "related to" jurisdiction is broad, and looks to whether the outcome of litigation "might have any ‘conceivable effect’ on the bankrupt estate. If that question is answered affirmatively, the litigation falls within the ‘related to’ jurisdiction of the bankruptcy court." In re Cuyahoga Equip. Corp. , 980 F.2d 110, 114 (2d Cir.1992) ; see also Parmalat Capital Fin. Ltd. v. Bank of Am. Corp. , 639 F.3d 572, 579 (2d Cir.2011) (an action arising under state law falls within "related to" jurisdiction if its outcome could have "any ‘conceivable effect" on the bankrupt estate."). "One of the central purposes—perhaps the central purpose—of extending bankruptcy jurisdiction to actions against certain third parties, as well as suits against debtors themselves, is to ‘protect[ ] the assets of the estate’ so as to ensure a fair distribution of those assets at a later point in time." In re Quigley Co. , 676 F.3d 45, 57 (2d Cir.2012) (quoting In re Zarnel , 619 F.3d 156, 171 (2d Cir.2010) ).

It is doubtful that this Court ever had "related to" bankruptcy jurisdiction over the Trusts' claims. Approximately one month before the Trusts filed their Complaint in this action, Hon. Christopher S. Sontchi, U.S.B.J., who presided over the American Home proceedings, held that there was no "related to" jurisdiction for a claim that arose between third-party non-debtors in a dispute that arose "solely under state law, independent of the bankruptcy petition." In re American Home Mortgage Holding , 477 B.R. 517, 520 (Bankr.Del.2012). Judge Sontchi repeatedly observed that the plan's confirmation precluded the exercise of bankruptcy jurisdiction over the third-parties' dispute. See id. at 532 ("Confirmation of the Plan terminated the Debtor's estate. As such, there is technically no longer an estate to impact."); 522 n. 16 (claims that could affect former creditors " ‘no longer ha[d] a close nexus to bankruptcy plan or proceeding because they exchanged their creditor status to attain rights to the litigation claims.’ ") (quoting Resorts Int'l, Inc. v. Price Waterhouse & Co., LLP. (In re Resorts Int'l, Inc.), 372 F.3d 154, 169 (3d Cir.2004) ); 530 ("plan confirmation has the general effect of closing up the bankruptcy estate."); 526 ("the Debtor's estate has technically ceased to exist after confirmation ....").

Thus, the American Home bankruptcy judge expressly rejected the exercise of "related to" jurisdiction over third-party claims that arose independent of the bankruptcy petition, concluding that after the plan's confirmation, the third-party claims could have no conceivable effect on the bankruptcy estate. Judge Sontchi's reasoning applies to the Trusts' breach of contract claim against UBS, which postdates the plan's confirmation, arises under state law and does not implicate the bankruptcy petition.

In addition to Judge Sontchi's decision, a review of UBS's claims in the American Home proceedings shows that no outcome in this case could affect the bankruptcy proceedings. The American Home bankruptcy plan adopted a no-fault protocol pursuant to which all breach of warranty claims were liquidated and allowed as unsecured claims, and a table to calculate an award for each breach of warranty claim. (Docket # 427 Ex. 5 at 46-47.) UBS submitted the questionnaire required by the Plan, and on March 9, 2011, UBS's breach claims were modified pursuant to this protocol, at which point UBS's breach of warranty claim against American Home was extinguished. (Docket # 430 Exs. A, G.) Although UBS filed additional amended proofs of claim, including one that was outstanding at the time that this action commenced, none sought indemnification related to the present action, and the proofs of claim were ultimately withdrawn with prejudice. (Docket # 430 Exs. H, I.) Moreover, UBS's claim for contractual indemnification from American Home (Docket # 430 Ex. A ¶¶ 1, 4) was extinguished by the Plan. (Docket # 427 Ex. 5 Art. 2(B)(2) ("This treatment supersedes and replaces any agreements or rights those Entities have in or against the applicable Debtor or its property.").) The parties have been unable to explain how, at the time of filing, the outcome of this case could have had any conceivable effect on the American Home proceedings.

Lastly, UBS's proofs of claim in the bankruptcy proceeding referenced only the 2007-1 Trust. (Docket # 430 Ex. H.) In a letter-brief, UBS acknowledged that the 2006-OA2 Trust and the 2007-3 Trust "contain only a de minimis number of AHM-originated loans ...." (Docket # 441 at 4.) Specifically, for the 2007-1 Trust, 3,732 loans, totaling more than 75% of that Trust's holdings, were originated by American Home. (Id. ; Docket # 443 at 6.) For the 2006-OA2 Trust, four loans were originated by American Home, and for the 2007-3 Trust, only one loan was originated by American Home. (Docket # 443 at 6.) Both UBS and the Trusts have urged that, pursuant to 28 U.S.C. § 1367(a), the Court exercise supplemental jurisdiction over the 2006-OA2 Trust and the 2007-3 Trust. (Docket # 443 at 6; Docket # 441 at 5.) The Court is unaware of any authority that permits it to exercise supplemental jurisdiction over parties that would otherwise fall beyond its subject matter jurisdiction, as opposed to exercising supplemental jurisdiction over claims. Thus, even if the Court had "related to" jurisdiction over the claims of the 2007-1 Trust, there appears to be no basis to exercise that jurisdiction over the claims of the 2006-OA2 Trust and the 2007-3 Trust.

The Court need not reach the ultimate conclusion of whether this Court ever had "related to" jurisdiction because it concludes that diversity jurisdiction existed from the inception of this action.

B. The Court Has Diversity Jurisdiction.

1. Legal Standard.

"It has long been the case that ‘the jurisdiction of the court depends upon the state of things at the time of the action brought.’ " Grupo Dataflux v. Atlas Global Grp., L.P. , 541 U.S. 567, 570, 124 S.Ct. 1920, 158 L.Ed.2d 866 (2004) (quoting Mollan v. Torrance , 9 Wheat. 537, 539, 22 U.S. 537, 6 L.Ed. 154 (1824) ); see also Freeport – McMoRan, Inc. v. K N Energy, Inc. , 498 U.S. 426, 428, 111 S.Ct. 858, 112 L.Ed.2d 951 (1991) ("We have consistently held that if jurisdiction exists at the time an action is commenced, such jurisdiction may not be divested by subsequent events."). "Defective allegations of jurisdiction may be amended, upon terms, in the trial or appellate courts." 28 U.S.C. § 1653. Therefore, if a complaint does not adequately allege the grounds for subject matter jurisdiction, but the unpleaded reality supports the exercise of jurisdiction, "a federal court may simply allow a complaint to be amended to assert those necessary facts and then treat diversity jurisdiction as having existed from the beginning." Herrick Co. v. SCS Commc'ns, Inc. , 251 F.3d 315, 329 (2d Cir.2001).

2. Diversity Jurisdiction Existed When the Action Commenced.

The Court concludes that diversity jurisdiction existed at the commencement of this action, even though it was not expressly alleged in the initial Complaint, because the Trustee of the three Trusts is a citizen solely of Ohio, UBS is a citizen of both Delaware and New York and the jurisdictional threshold is exceeded.

The Complaint was filed on September 28, 2012. (Docket # 1.) The caption of identified the three Trusts as the plaintiffs and the Trustee was not included in the caption. The introductory paragraph of the Complaint, however, alleged that the Trusts were "acting through U.S. Bank National Association, solely in its capacity as Trustee ... for the transactions ...." It separately alleged that the "[p]laintiffs are acting through the Trustee, a national banking association organized and existing under the laws of the United States of America with its principal place of business in Minneapolis, Minnesota." (Compl't ¶ 7.)¹ The Complaint's signature block identified counsel as "Attorneys for U.S. Bank National Association, as Trustee ..." for the three Trusts.

1 The principal place of business of U.S. Bank was and is Minneapolis, Minnesota. But because of its status as a national banking association, its citizenship is determined by its registered main office, Wachovia Bank v. Schmidt , 546 U.S. 303, 318, 126 S.Ct. 941, 163 L.Ed.2d 797 (2006), which is Cincinnati, Ohio. U.S. Bank was a citizen of Ohio at the time that the action commenced and remains so.

The Complaint also alleged that UBS was a Delaware corporation with its principal place of business in New York. (Compl't ¶ 8.) It also alleged that the Trusts suffered more than $1.5 billion in losses as a result of UBS's breach of contract. (Compl't ¶ 41.)

Only the Trustee and not the Trusts could bring this action. The original allegation that the suit is brought by the Trusts, "acting through the Trustee," is of equivalent meaning to the Trustee suing solely in its capacity as Trustee of the Trusts. Both convey that the Trustee is acting for the Trusts in bringing the action. Further, the citizenship of U.S. Bank, as Trustee, is controlling for the purposes of the diversity analysis.

3. Judge Baer Concluded that the Action Was Brought by U.S. Bank on Behalf of the Trusts.

At an early stage of the case, UBS moved to dismiss the Complaint, arguing, among other things, that only the Trustee could sue to enforce UBS's obligations under the PSAs, and that the Complaint's caption identified only the Trusts as the plaintiffs. (Docket # 15 at 11-12.) Judge Baer denied the motion, and concluded that the Complaint alleged that U.S. Bank was the plaintiff in this action. Citing to the Complaint's opening paragraph, which alleged that the Trusts were "acting through U.S. Bank National Association," Judge Baer concluded that "the Complaint leaves no doubt that the Trusts' rights are being enforced by the Trustee." 2013 WL 4399210, at *2. To the extent that only the Trusts were named in the Complaint's caption, Judge Baer concluded that "it is ‘[t]he substance of the pleadings, not the caption, that determines the identity of the parties.’ " Id. (quoting Mateo v. JetBlue Airways Corp. , 847 F.Supp.2d 383, 384 (E.D.N.Y.2012) ); see also Ocasio v. Riverbay Corp. , 2007 WL 1771770, at *7 (S.D.N.Y. June 19, 2007) ("the caption of a complaint is not normally determinative of the identity of the parties or the pleader's statement of claim.") (quotation marks omitted).

This Court agrees with Judge Baer's conclusions. In both the introductory paragraph and paragraph seven, the Complaint alleged that the plaintiff Trusts are "acting through the Trustee ...." Paragraph seven of the Complaint falls under the heading "PARTIES." The Complaint states that the plaintiff Trusts "are acting through the Trustee," but there is no meaningful difference between the Trusts acting through a Trustee and a Trustee acting for the Trusts. The initial Complaint therefore adequately alleged that the action was brought by U.S. Bank on behalf of the three Trusts.

4. U.S. Bank Is the Real Party to the Controversy.

"[T]he ‘citizens' upon whose diversity a plaintiff grounds jurisdiction must be real and substantial parties to the controversy." Navarro Sav. Ass'n v. Lee , 446 U.S. 458, 460, 100 S.Ct. 1779, 64 L.Ed.2d 425 (1980) (citing McNutt v. Bland , 2 How. 9, 15, 43 U.S. 9, 11 L.Ed. 159 (1844) ). "[A] trustee is a real party to the controversy for purposes of diversity jurisdiction when he possesses certain customary powers to hold, manage, and dispose of assets for the benefit of others." Id. at 464, 100 S.Ct. 1779.

Section 2.01(c) of the PSAs establishes the Trusts and appoints U.S. Bank to act as Trustee. The PSA for the 2006-OA2 Trust states:

The Depositor does hereby establish, pursuant to the further provisions of this Agreement and the laws of the State of New York, an express trust (the ‘Trust’) to be known, for convenience, as ‘MASTR Adjustable Rate Mortgages Trust 2006-0A2’ and U.S. Bank National Association is hereby appointed as Trustee in accordance with the provisions of this Agreement.

(PX 49 at 0070-71.) Section 2.01(a) provides that the Depositor "hereby sells, transfers, assigns, sets over and otherwise conveys to the Trustee for the benefit of the Certificateholders and the Certificate Insurer, without recourse, all the right, title and interest of the Depositor in and to the Trust Fund ...." (PX 49 at 0061.) The "Trust Fund" is defined as "[t]he corpus of the trust," including all Mortgage Loans. (PX 49 at 0065-66.) Under section 2.02, "the Custodian and Trustee together declare that it holds or will hold such other assets as are included in the Trust Fund, in trust for the exclusive use and benefit of all present and future Certificateholders and the Certificate Insurer." (PX 49 at 0071.)

Section 2.03 of the PSAs authorizes the Trustee to enforce UBS's obligations in the event that a breach materially and adversely affects the interests of the Certificateholders. See PX 49 at 0074 ("if the Transferor [UBS] fails to correct or cure the defect within such period, and such defect materially and adversely affects the interests of the Certificateholders and the Certificate Insurer in the related Mortgage Loan, the Trustee shall enforce the Transferor's obligations hereunder ....") (emphasis added). The PSAs do not give the Trusts or the Certificateholders the authority to enforce the PSAs against UBS.

Further, section 2.03 states that UBS "hereby makes the representations and warranties set forth in Schedule II" to certain identified parties, specifically including "the Depositor, the Certificate Insurer and the Trustee ...." (PX 49 at 0073.) The warranties are not made to the Trust itself or to the Certificateholders.

The Court concludes that the Trustee is the real and substantial party to the controversy. It holds the corpus of each Trust "for the benefit of the Certificateholders" and is obligated to enforce UBS's remedial obligations in the event that a breach of warranty materially and adversely affects the interests of the Certificateholders. The warranties of Schedule II are made to the Trustee but not to the Trusts or the Certificateholders. These interests and obligations are sufficient to establish that the Trustee is the real and substantial party to the controversy. See, e.g. , Oscar Gruss & Son, Inc. v. Hollander , 337 F.3d 186, 195 (2d Cir.2003) (the party with "the express power to act on [others'] behalf with regard to their rights in the warrants" was the "master of the litigation" and the "real and substantial party"); U.S. Bank Nat. Ass'n v. Nesbitt Bellevue Prop. LLC , 859 F.Supp.2d 602, 608 (S.D.N.Y.2012) (Koeltl, J.) (trustee for mortgage loans is the real and substantial party when it "holds ‘all the right, title and interest’ in them under the PSA ... and has a fiduciary obligation to the certificateholders to see that the loans are paid and that the value of the collateral is maintained.").

5. The Recent Americold Decision Does Not Alter This Analysis.

The Supreme Court's recent opinion in Americold Realty Trust v. ConAgra Foods, Inc. , ––– U.S. ––––, 136 S.Ct. 1012, 194 L.Ed.2d 71 (2016), does not alter this analysis. Americold concluded that a "real estate investment trust" formed under a Maryland statute was an "artificial entity" that took on the citizenship of all of its members. Id. at 1015–16. Under the Maryland statute, the purported trust was an "unincorporated business trust or association" whose members were "shareholders" that had "ownership interests" and voted on the trust "by virtue of their ‘shares of beneficial interest.’ " Id. The Maryland statute also treated the trust as "a ‘separate legal entity’ that itself can sue or be sued." Id. Americold concluded that the trust's shareholders were analogous to the members of a limited partnership or the shareholders of a joint-stock company—"artificial entities" whose citizenship is determined by the citizenship of all members. Id. at 1015–16. Americold's holding "coexists" with Navarro , which "reaffirmed a separate rule that when a trustee files a lawsuit in her name, her jurisdictional citizenship is the State to which she belongs— as is true of any natural person." Id. at 1016.

The Trusts in this case are not analogous to the investment trust in Americold , and the citizenship of the Trusts' individual members is not relevant to determining diversity jurisdiction. In contrast to a Maryland real estate trust, the Trusts have no power to sue on their own behalves and the Trustee alone is responsible for the corpus of the Trusts. The Court therefore concludes that only the Trustee's citizenship is relevant to this diversity analysis. See also Halley v. Deustche Bank Nat'l Trust Co. , 2016 WL 3855872, at *1 (S.D.Tex. July 15, 2016) (Americold "reaffirmed the holding in Navarro " that a trustee's citizenship governs when the trustee sues in its own name); The Bank of New York Mellon v. Townhouse S. Ass'n, Inc. , 2016 WL 3563503, at *2 (D.Nev. June 29, 2016) (when a trustee sues in its own name and has legal title to the trust's assets, the trustee's citizenship " ‘is all that matters for diversity.’ ") (quoting Ame r ic o ld , 136 S.Ct. at 1016 ).

6. The Amended Complaint Expressly Alleges Diversity Jurisdiction.

On August 1, 2016, the Court granted the Trusts' motion to file an Amended Complaint to allege diversity jurisdiction under 28 U.S.C. § 1332(a)(1). (Docket # 494.) The Amended Complaint was filed on August 2, 2016. (Docket # 495.) UBS filed its answer on August 19, 2016. (Docket # 499.)

The Amended Complaint expressly alleges that this Court has diversity jurisdiction pursuant to 28 U.S.C. s 1332(a). (Am. Compl't ¶ 13.) It alleges that the Trustee, U.S. Bank, is a national banking association organized under the laws of the United States, with its registered main office in Ohio. (Am. Compl't ¶ 7.) The citizenship of a national banking association is determined by "the State designated in its articles of association as its main office." Wachovia Bank v. Schmidt , 546 U.S. 303, 318, 126 S.Ct. 941, 163 L.Ed.2d 797 (2006) ; see also 28 U.S.C. § 1348 ("All national banking associations shall, for the purposes of all other actions by or against them, be deemed citizens of the States in which they are respectively located.").

UBS is alleged to be a Delaware corporation with its principal place of business in New York. (Am. Compl't ¶ 8.) The amount in controversy is alleged to exceed $75,000, exclusive of interest and costs. (Am. Compl't ¶ 13.)

The Court concludes that because plaintiff U.S. Bank is a citizen of Ohio, defendant UBS is a citizen of New York and Delaware, and the amount in controversy exceeds the jurisdictional threshold, this Court has had diversity jurisdiction over the action since its commencement.

CHOICE OF LAW.

Each PSA includes a choice-of-law provision that states, in all-uppercase letters:

THIS AGREEMENT SHALL BE CONSTRUED IN ACCORDANCE WITH AND GOVERNED BY THE SUBSTANTIVE LAWS OF THE STATE OF NEW YORK APPLICABLE TO AGREEMENTS MADE AND TO BE PERFORMED IN THE STATE OF NEW YORK AND THE OBLIGATIONS, RIGHTS AND REMEDIES OF THE PARTIES HERETO AND THE CERTIFICATEHOLDERS SHALL BE DETERMINED IN ACCORDANCE WITH SUCH LAWS.

(Fact Stip. ¶ 13; PX 49 § 11.03; PX 110 § 12.03; PX 182 § 11.03.) New York law therefore governs the PSAs.

BURDEN OF PROOF.

The Trusts have the burden of proving each disputed element of their claims by a preponderance of the evidence. See, e.g. , Diesel Props. S.r.1. v. Greystone Business Credit II LLC, 631 F.3d 42, 51 (2d Cir.2011). The Trusts bear this burden with respect to each alleged breach for each loan as to which they seek relief. See, e.g. , Ret. Bd. of the Policemen's Annuity & Benefit Fund of the City of Chi. v. Bank of N.Y. Mellon , 775 F.3d 154, 162 (2d Cir.2014) (RMBS defendant's "alleged misconduct must be proved loan-by-loan and trust-by-trust").

Competent expert testimony is required where certain complex or technical issues would not be obvious to the finder of fact. See Wills v. Amerada Hess Corp. , 379 F.3d 32, 46 (2d Cir.2004) (expert testimony to prove causation because "the nexus between the injury and the alleged cause would not be obvious to the lay juror"). Courts have required expert testimony on issues related to a breach of warranty, injury and causation. See, e.g. , Russo v. Keough's Turn of the River Hardware, LLC , 529 Fed.Appx. 50, 52 (2d Cir.2013) (affirming exclusion of expert witness and holding that "[w]ithout the testimony of their expert witness, [plaintiff's breach of warranty] claims fail"); Barnes v. Anderson , 202 F.3d 150, 160 (2d Cir.1999) ("[W]e do not see how a jury could rationally decide causation without the aid of expert testimony.").

Here, with the exception of a small number of claims that turn on borrower FICO scores, the Trusts' case turns on expert testimony, and specifically the breaches identified by Holt's review and the LTV ratios that were re-calculated using Cowan's AVM model.

RELEVANT NEW YORK CONTRACT PRINCIPLES.

A. The PSAs' Unambiguous Terms Are Construed to Reflect the Intent of the Parties as Manifested in the Agreements.

This case turns on application of the plain and unambiguous language of the PSAs. "Where the terms of a contract are clear and unambiguous, the intent of the parties must be found within the four corners of the contract, giving a practical interpretation to the language employed and reading the contract as a whole." Ellington v. EMI Music, Inc. , 24 N.Y.3d 239, 244, 997 N.Y.S.2d 339, 21 N.E.3d 1000 (2014) ; accord Beal Sav. Bank v. Sommer , 8 N.Y.3d 318, 324, 834 N.Y.S.2d 44, 865 N.E.2d 1210 (2007) ("Construction of an unambiguous contract is a matter of law, and the intention of the parties may be gathered from the four corners of the instrument and should be enforced according to its terms."); Greenfield v. Philles Records, Inc. , 98 N.Y.2d 562, 569, 750 N.Y.S.2d 565, 780 N.E.2d 166 (2002) ("The best evidence of what parties to a written agreement intend is what they say in their writing.").

" ‘[C]ourts may not by construction add or excise terms, nor distort the meanings of those used and thereby make a new contract for the parties under the guise of interpreting the writing.’ " ACE Sec. Corp. v. DB Structured Products, Inc. , 25 N.Y.3d 581, 597, 15 N.Y.S.3d 716, 36 N.E.3d 623 (2015) (quoting Vermont Teddy Bear Co. v. 538 Madison Realty Co. , 1 N.Y.3d 470, 475, 775 N.Y.S.2d 765, 807 N.E.2d 876 (2004) ). "Courts will give effect to the contract's language and the parties must live with the consequences of their agreement. If they are dissatisfied, the time to say so is at the bargaining table." Eujoy Realty Corp. v. Van Wagner Commc'ns, LLC , 22 N.Y.3d 413, 424, 981 N.Y.S.2d 326, 4 N.E.3d 336 (2013) (quotation marks and alterations omitted). "The words and phrases used by the parties must, as in all cases involving contract interpretation, be given their plain meaning." Ellington , 24 N.Y.3d at 244, 997 N.Y.S.2d 339, 21 N.E.3d 1000 (quotation marks omitted); accord Greenfield , 98 N.Y.2d at 569, 750 N.Y.S.2d 565, 780 N.E.2d 166 ("[A] written agreement that is complete, clear and unambiguous on its face must be enforced according to the plain meaning of its terms.") (quotation marks and citations omitted).

"A reading of the contract should not render any portion meaningless. Further, a contract should be read as a whole, and every part will be interpreted with reference to the whole; and if possible it will be so interpreted as to give effect to its general purpose." Beal Sav. Bank , 8 N.Y.3d at 324–25, 834 N.Y.S.2d 44, 865 N.E.2d 1210 (quotation marks and citations omitted); see also S. Rd. Assocs., LLC v. Int'l Bus. Machines Corp. , 4 N.Y.3d 272, 277, 793 N.Y.S.2d 835, 826 N.E.2d 806 (2005) (it is "important to read the document as a whole to ensure that excessive emphasis is not placed upon particular words or phrases."). A party "may not pick and choose which provisions suit its purposes .... A contract should be read to give effect to all its provisions." God's Battalion of Prayer Pentecostal Church, Inc. v. Miele Assocs., LLP , 6 N.Y.3d 371, 374, 812 N.Y.S.2d 435, 845 N.E.2d 1265 (2006) (quotation marks omitted). Courts "must be careful not to add new terms or alter the terms of the contract in the guise of interpreting it." Rosenthal v. Quadriga Art, Inc. , 69 A.D.3d 504, 507, 894 N.Y.S.2d 32 (1st Dep't 2010).

B. Parol Evidence Plays No Role in Interpreting an Unambiguous Contract.

A court may consider extrinsic evidence of the parties' intent only if it concludes that the contract is ambiguous. Greenfield , 98 N.Y.2d at 569, 750 N.Y.S.2d 565, 780 N.E.2d 166. " ‘[A] contract is not rendered ambiguous just because one of the parties attaches a different, subjective meaning to one of its terms.’ " Bank of New York Mellon v. WMC Mortgage, LLC , 136 A.D.3d 1, 9, 22 N.Y.S.3d 3 (1st Dep't 2015) (quoting Bajraktari Mgt. Corp. v. American Int'l Grp., Inc. , 81 A.D.3d 432, 432, 916 N.Y.S.2d 771 (1st Dep't 2011) ); accord Rosenthal v. Quadriga Art, Inc. , 69 A.D.3d 504, 506, 894 N.Y.S.2d 32 (1st Dep't 2010) ("[T]he existence of a disagreement about the ‘plain meaning’ of the words does not necessarily render those words ambiguous for purposes of construing the contract."); Johnson v. Lebanese Am. Univ. , 84 A.D.3d 427, 435, 922 N.Y.S.2d 57 (1st Dep't 2011) ("[T]hat one party to an agreement may attach a particular, subjective meaning to a term that differs from the term's plain meaning does not render the agreement ambiguous.").

A contract is unambiguous if its language has "a definite and precise meaning, unattended by danger of misconception in the purport of the agreement itself, and concerning which there is no reasonable basis for a difference of opinion." Greenfield , 98 N.Y.2d at 569–70, 750 N.Y.S.2d 565, 780 N.E.2d 166 (quotation marks and alteration omitted). "It is well settled that extrinsic and parol evidence is not admissible to create an ambiguity in a written agreement which is complete and clear and unambiguous upon its face." W.W.W. Associates, Inc. v. Giancontieri , 77 N.Y.2d 157, 163, 565 N.Y.S.2d 440, 566 N.E.2d 639 (1990).

C. No Party Asserts that Contra Preferentem Applies.

Neither party contends that the PSAs should be construed against their drafter pursuant to the canon of contra preferentem. See generally Matter of Riconda , 90 N.Y.2d 733, 740, 665 N.Y.S.2d 392, 688 N.E.2d 248 (1997) ; see also Schering Corp. v. Home Ins. Co. , 712 F.2d 4, 10 (2d Cir.1983) (New York courts apply canon of contra preferentem only "as a matter of last resort, after all aids to construction have been employed but have failed to resolve the ambiguities in the written instrument."). No party has contended that the PSAs were other than arm's length transactions agreed to by sophisticated parties represented by counsel. There is no basis in the record before this Court to construe the PSAs against their drafters. See Riconda , 90 N.Y.2d at 740, 665 N.Y.S.2d 392, 688 N.E.2d 248.

D. Contractual Warranties under New York Law.

A warranty is " ‘an assurance by one party to a contract of the existence of a fact upon which the other party may rely. It is intended precisely to relieve the promisee of any duty to ascertain the fact for himself; it amounts to a promise to indemnify the promisee for any loss if the fact warranted proves untrue, for obviously the promisor cannot control what is already in the past.’ " CBS Inc. v. Ziff – Davis Pub. Co. , 75 N.Y.2d 496, 503, 554 N.Y.S.2d 449, 553 N.E.2d 997 (1990) (quoting Metropolitan Coal Co. v. Howard , 155 F.2d 780, 784 (2d Cir.1946) (Hand, J.)). An "express warranty is as much a part of the contract as any other term." Id. at 503, 554 N.Y.S.2d 449, 553 N.E.2d 997.

The Trusts contend that UBS breached certain warranties set forth in Schedule II to the PSAs. These warranties set forth the characteristics of the loans held by the Trusts. Schedule II contains 49 representations and warranties, six of which the Trusts claim were breached by UBS

THE TRUSTS MAY RECOVER THE MONEY DAMAGE EQUIVALENT OF THE REPURCHASE REMEDY FOR BREACHED LOANS WHERE REPURCHASE IS NOT POSSIBLE.

Section 2.03 provides for three potential remedies in the event of a discovered or noticed breach: cure, replacement by an equivalent substitute loan or repurchase. The language of section 2.03 does not contemplate a money damages remedy. Indeed, the remedies are expressly denominated as the "sole remedies": "It is understood and agreed that the obligation under this Agreement of [UBS] to cure, repurchase or replace any Mortgage loan as to which a breach has occurred and is continuing shall constitute the sole remedies against [UBS] respecting such matters available to Certificateholders ... or the Trustee on their behalf." (PX 49 at 0075-76.)

Many loans for which a breach is claimed are existing, outstanding loans. If and to the extent that the Trusts prove their claims as to these loans, they are entitled to the specific performance of section 2.03's repurchase remedy. (Tr. 1884.) UBS does not dispute this point. (Id. ) But some loans have been liquidated with a loss. They are no longer in the Trusts and cannot be repurchased by UBS. UBS, in effect argues that as to those loans, there is no remedy because there is no loan to repurchase. The Trusts urge and the Court agrees that a court may award the money damage equivalent of the repurchase remedy in an instance where a court sitting in equity may not award specific performance.

In deciding UBS's motion to dismiss, Judge Baer concluded that the sole-remedies provision of section 2.03 does not preclude an award of money damages in the event of a breach. See MASTR Adjustable Rate Mortgages Trust 2006 – OA2 v. UBS Real Estate Sec. Inc. , 2013 WL 4399210, at *3–4 (S.D.N.Y. Aug. 15, 2013). He noted that "when specific performance of the agreement is not possible, the parties are left to whatever legal or equitable remedies they may have," including an award of money damages. Id. at *3 (quotation marks omitted). Any money damages award is "limited to ‘the Purchase Price’ under the PSAs," and thus cannot exceed in value the remedies created by the PSAs. Id. In its summary judgment decision, this Court agreed that, "consistent with the reasoning of New York courts," the Trusts may recover money damages in lieu of the repurchase remedy. 2015 WL 764665, at *17. The Court concluded that damages must be commensurate with any actual loss suffered by the Trusts, and that the Trusts could not receive rescissory damages that exceed actual damages. Id. at *17. See also Nomura Home Equity Loan, Inc. v. Nomura Credit & Capital, Inc. , 133 A.D.3d 96, 106, 19 N.Y.S.3d 1 (1st Dep't 2015) ("In the RMBS context, most courts have repeatedly held that while a provision providing for equitable relief as the ‘sole remedy’ will generally foreclose alternative relief, where the granting of equitable relief appears to be impossible or impracticable, equity may award damages in lieu of the desired equitable remedy.") (quotation marks omitted) (collecting cases); Morgan Stanley Mortgage Loan Trust 2006 – 13ARX v. Morgan Stanley Mortgage Capital Holdings LLC , 143 A.D.3d 1, 9, 2016 WL 4217793, at *5 (1st Dep't Aug. 11, 2016) ("In Nomura , we recognized that the remedy of specific performance in put-back cases might be impossible to fulfill. It is for this reason we left open the possibility that, even for ordinary breach of contract claims, equity may require an award of monetary damages in lieu of specific performance.") (internal citation omitted).

The Court therefore concludes that for loans that cannot be repurchased, it may award the money damage equivalent of the repurchase remedy.

THE TRUSTS, THE TRUSTEE AND RELATED PERSONS.

UBS acquired home mortgage loans from third-party Originators, which it pooled and securitized into the three plaintiff Trusts: Master Adjustable Rate Mortgages Trust 2006-OA2 (the "2006-OA2 Trust"), MASTR Adjustable Rate Mortgages Trust 2007-1 (the "2007-1 Trust") and MASTR Adjustable Rate Mortgages Trust 2007-3 (the "2007-3 Trust"). (Fact Stip. ¶¶ 1-2.) The vast majority of the loans held by the three Trusts were originated by Countrywide, American Home and IndyMac. (Fact Stip. ¶ 3.)

The Trusts issued Certificates, which were sold to Certificateholders. (Fact Stip. ¶ 2.) Certificateholders were entitled to payment of the cash flows from principal and interest payments made by the borrowers on the loans. (Fact Stip. ¶ 2.) The Trusts' main source of funds for making distributions on the certificates was the stream of income flowing into the Trusts from payments made by borrowers on the securitized loans each month. (See, e.g. , PX 264 (2007-3 Prospectus Supplement) at 1 ("The trust's main source of funds for making distributions on the certificates will be collections on closed end, adjustable-rate loans secured by first mortgages or deeds of trust on residential one– to four-family properties, all of which are negatively amortizing loans ....").) The outflow of these payments to Certificateholders was determined by the PSAs, which created different classes of certificates, and established a "waterfall" in which holders of senior certificates generally were paid the full amount of their monthly income before the holders of junior certificates. (See PX 49 at 0015-16 (creating "waterfall").) The Trusts issued the certificates pursuant to prospectus supplements filed with the Securities Exchange Commission ("SEC"). (Fact Stip. ¶ 4.)

The 2006-OA2 Trust closed on November 15, 2006. (Fact Stip. ¶ 5.) It was backed by approximately 5,660 loans, with an aggregate principal balance of approximately $2,013,321,248. (Fact Stip. ¶ 5.) Loans originated by Countrywide made up 47.51% of the principal balance, and loans originated by IndyMac made up another 37.47% of the principal balance. (Fact Stip. ¶ 5.)

The 2007-1 Trust closed on January 16, 2007. (Fact Stip. ¶ 6.) It was backed by approximately 4,944 loans, with an aggregate principal balance of approximately $2,099,439,579. (Fact Stip. ¶ 5.) Loans originated by American Home made up approximately 78% of the principal balance, and loans originated by IndyMac made up another 16% of the principal balance. (Fact Stip. ¶ 6.)

The 2007-3 Trust closed on May 15, 2007. (Fact Stip. ¶ 7.) It was backed by approximately 6,478 loans, with an aggregate principal balance of approximately $2,582,881,408. (Fact Stip. ¶ 7.) Loans originated by Countrywide made up approximately 52% of the principal balance, and loans originated by IndyMac made up approximately 40% of the Principal Balance. (Fact Stip. ¶ 7.)

U.S. Bank National Association ("U.S. Bank") is the Trustee for each of the three Trusts. (Fact Stip. ¶ 8.) Under the PSAs, UBS acted as the sponsor (or "Transferor") of each Trust, and U.S. Bank functions as the Trustee for each Trust. (Stip. Fact ¶¶ 1, 8.) Wells Fargo Bank, N.A. served as Master Servicer, Trust Administrator, Custodian and Credit Risk Manager. (PX 49 at 007; PX 007 at 1; PX 182 at 007.) Assured Guaranty Municipal Corp. ("Assured") provided financial guaranty insurance on certain classes of Certificates that the Trusts issued. (Fact Stip. ¶ 9.)

The three Trusts were each formed and are governed by separate PSAs. (Fact Stip. ¶¶ 10-12; PX 49 (2006 OA-2), PX 110 (2007-1), PX 182 (2007-3).) The PSAs govern the rights and obligations of the various parties to the transaction, specifically including U.S. Bank, UBS and Assured. (Fact Stip. ¶¶ 10-12.) The terms of the three PSAs that are relevant to the present controversy are materially identical. Schedule II of each PSA contains 49 representations and warranties. (Fact Stip. ¶¶ 16-24.)

THE NOTICE AND DISCOVERY REQUIREMENT.

Before turning to issues relating to breach, the Court will address the threshold notice or discovery requirement. Because UBS was on timely notice of some breaches more than 90 days before the commencement of this action, supplemental notice, including through the service of an expert report, are deemed, under New York law, to be timely. Nomura Home Equity Loan, Inc. v. Nomura Credit & Capital, Inc. , 133 A.D.3d 96 , 19 N.Y.S.3d 1(1st Dep't 2015). Hence, UBS was on proper and timely notice of all breaches for which the Trusts seek recovery.

A. The Requirements of Section 2.03.

Central to the Trusts' entitlement to relief in this case are the provisions of one particular section of the PSAs, section 2.03. The excerpted language is quoted to set forth those portions necessary to understand the notice and discovery provisions:

The Transferor [UBS] hereby makes the representations and warranties set forth in Schedule II hereto, and by this reference incorporated herein, to ... the Certificate Insurer and the Trustee, as of the Closing Date, or if so specified therein, as of the Cut-off Date. With respect to any representation and warranties set forth on Schedule II hereto which are made to the best of the Transferor's knowledge if it is discovered by any of the ... the Certificate Insurer, ..., the Transferor, ... [or] the Trustee ... that the substance of such representation and warranty is inaccurate and such inaccuracy materially and adversely affects the value of the related Mortgage Loan or the interests of the Certificateholders or the Certificate Insurer

therein, notwithstanding the Transferor's lack of knowledge with respect to the substance of such representation or warranty, such inaccuracy shall be deemed a breach of the applicable representation or warranty.

Upon discovery by any of the ... the Certificate Insurer [or] the Transferor ... of a breach of a representation or warranty made by the Transferor pursuant to this Section 2.03 that materially and adversely affects the interests of the Certificateholders or the Certificate Insurer in any Mortgage Loan, the party discovering such breach shall give prompt notice thereof to the other parties and the Trustee .... The Trustee shall enforce the obligations of the Transferor in accordance with this Section 2.03 to correct or cure any such breach of a representation or warranty made herein, and if the Transferor fails to correct or cure the defect within such period, and such defect materially and adversely affects the interests of the Certificateholders and the Certificate Insurer in the related Mortgage Loan, the Trustee shall enforce the Transferor's obligations hereunder to (i) purchase such Mortgage Loan at the Purchase Price or (ii) substitute for the related Mortgage Loan an Eligible Substitute Mortgage Loan ....

The Transferor hereby covenants that within ninety (90) days of the earlier of its discovery or its receipt of written notice from any party of a breach of any representation or warranty made pursuant to this Section 2.03 which materially and adversely affects the interest of the Certificateholders or the Certificate Insurer in any Mortgage Loan, it shall cure such breach in all material respects, and if such breach is not so cured, shall, (i) if such ninety (90) day period expires prior to the second anniversary of the Closing Date, remove such Deleted Mortgage Loan from the Trust Fund and substitute in its place an Eligible Substitute Mortgage Loan or Mortgage Loans into the Trust Fund, in the manner and subject to the conditions set forth in this Section; or (ii) repurchase the affected Mortgage Loan or Mortgage Loans from the Trustee at the Purchase Price in the manner set forth below. ...

(Fact Stip. ¶¶ 25-28 & PX 49 at 0073-74.) Thus, under the terms of section 2.03, UBS covenants that it will cure a breach "within ninety (90) days of the earlier of its discovery or its receipt of written notice from any party of a breach of any representation or warranty ...." Following the expiration of the 90-day period, within the first two years of the Closing Date, UBS could remedy the breach by substituting the breached loan with an eligible substitute loan. The two-year period for the substitution remedy expired long before the commencement of this action. Therefore, repurchase is the only remaining remedy for a loan that meets all breach requirements.

As will be explained, this Court construes section 2.03 to mean that UBS's cure or repurchase obligation is triggered upon receipt of a written notice from the Trustee. But UBS also has a duty to provide prompt notice to the Trustee if it discovers that a warranty has been breached, and if it fails to do so, UBS is deemed to have received notice of the breach that it discovered.

1. UBS Received Written Breach Notices of that Identified 4,869 Breached Loans, Including 4,462 Breaches that Were Timely Identified Before the Commencement of this Action.

Section 2.03 provides that UBS's obligation to cure, replace or repurchase a defective loan can be triggered through notice brought by a party to the PSA. It states:

Upon discovery by any of the Depositor, the Certificate Insurer [Assured], the Transferor [UBS], the Master Servicer [Wells Fargo], the Trust Administrator [Wells Fargo] or the Custodian [Wells Fargo] of a breach of a representation or warranty made by the Transferor pursuant to this Section 2.03 that materially and adversely affects the interests of the Certificateholders or the Certificate Insurer in any Mortgage Loan, the party discovering such breach shall give prompt notice thereof to the other parties and the Trustee.

(Fact Stip. ¶¶ 25-27 & PX 49.) It also provides that "[t]he Trustee [U.S. Bank] shall enforce the obligations of the Transferor in accordance with this Section 2.03 ...." (Id. )

a. The Assured Breach Notices.

The "Certificate Insurer" who provided certain guarantees to Certificateholders was, in the case of the three Trusts, Assured. Assured was identified as a person who, under the PSA, could trigger the obligation to cure, replace or repurchase a specific mortgage loan. From August 9, 2010 through September 27, 2012, Assured sent UBS a series of letters asserting that UBS had breached its representations and warranties, and demanding that UBS cure or repurchase purportedly defective loans held by the three Trusts. (Fact Stip. ¶¶ 37-57.) In total, Assured sent 21 letters to UBS demanding the cure or repurchase of loans alleged to be in breach of the representations and warranties. (Fact Stip. ¶¶ 37-57.) Collectively, these breach notices identified 4,642 mortgage loans alleged to have breached one or more of the representations and warranties set forth in Schedule II.

The breach notices stated that they were submitted pursuant to section 2.03 of the PSAs. (See, e.g. , PX 334.) They identified particular loans alleged to have breached the representations and warranties of Schedule II, and stated that the breaches materially and adversely affected the interests of the Certificateholders and Assured. (See id. ) They stated that pursuant to section 2.03, UBS was required to repurchase or cure the identified loans within 90 days of the letters' date. (See id. ) In their stipulation of facts, the parties identify the following breach notices sent by Assured to UBS:

1. An August 9, 2010 notice identifying 270 breached loans in the 2007-1 Trust, with a total original principal balance of $137,009,951. (Fact Stip. ¶ 37.)

2. An August 13, 2010 notice identifying 325 breached loans in the 2007-3 Trust, with a total original principal balance of $129,403,675. (Fact Stip. ¶ 38.)

3. An August 13, 2010 notice identifying 112 breached loans in the 2007-3 Trust, with a total original principal balance of $55,219,616. (Fact Stip. ¶ 39.)

4. An August 13, 2010 notice identifying 55 breached loans in the 2007-3 Trust, with a total original principal balance of $19,644,848. (Fact Stip. ¶ 40.)

5. A June 17, 2011 notice identifying 412 breached loans in the 2007-1 Trust, with a total original principal balance of $184,153,505. (Fact Stip. ¶ 41.)

6. A June 30, 2011 notice identifying 289 breached loans in the 2007-1 Trust, with a total original principal balance of $72,081,142. (Fact Stip. ¶ 42.)

7. A June 30, 2011 notice identifying 329 breached loans in the 2007-1 Trust, with a total original principal balance of $164,780,597. (Fact Stip. ¶ 43.)

8. A July 5, 2011 notice identifying 236 breached loans in the 2007-1 Trust,

with a total original principal balance of $83,524,093. (Fact Stip. ¶ 44.)

9. A July 7, 2011 notice identifying 87 breached loans in the 2007-3 Trust, with a total original principal balance of $30,816,660. (Fact Stip. ¶ 45.)

10. A December 14, 2011 notice identifying 167 breached loans in the 2006-OA2 Trust, with a total original principal balance of $61,295,315. (Fact Stip. ¶ 46.)

11. A December 15, 2011 notice identifying 78 breached loans in the 2007-3 Trust, with a total original principal balance of $25,402,002. (Fact Stip. ¶ 47.)

12. A December 15, 2011 notice identifying 80 breached loans in the 2007-1 Trust, with a total original principal balance of $31,172,576. (Fact Stip. ¶ 48.)

13. A March 28, 2012 notice identifying 361 breached loans in the 2006-OA2 Trust, with a total original principal balance of $238,335,720. Assured reiterated these breach claims in a separate demand letter dated April 27, 2012. (Fact Stip. ¶ 49.)

14. A May 7, 2012 notice identifying 179 breached loans in the 2006-1 Trust, with a total original principal balance of $79,628,709. Assured reiterated these breach claims in a separate demand letter dated May 30, 2012. (Fact Stip. ¶ 50.)

15. A May 31, 2012 notice identifying 261 breached loans in the 2006-OA2 Trust, with a total original principal balance of $98,863,185. Assured reiterated these breach claims in a separate demand letter dated June 19, 2012. (Fact Stip. ¶ 51.)

16. A June 27, 2012 notice identifying 183 breached loans in the 2006-OA2 Trust, with a total original principal balance of $54,941,581. Assured reiterated these breach claims in a separate demand letter dated May July 19, 2012. (Fact Stip. ¶ 52.)

17. A June 28, 2012 notice identifying 192 breached loans in the 2007-3 Trust, with a total original principal balance of $172,706,965. Assured reiterated these breach claims in a separate demand letter dated July 19, 2011 [sic]. (Fact Stip. ¶ 53.)

18. An August 17, 2012 notice identifying 354 breached loans in the 2007-3 Trust, with a total original principal balance of $179,395,420.99. Assured reiterated these breach claims in a separate demand letter dated September 14, 2012. (Fact Stip. ¶ 54.)

19. An August 17, 2012 notice identifying 264 breached loans in the 2006-OA2 Trust, with a total original principal balance of $89,465,828. Assured reiterated these breach claims in a separate demand letter dated September 14, 2012. (Fact Stip. ¶ 55.)

20. An August 30, 2012 notice identifying 180 breached loans in the 2006-OAS Trust, with a total original principal balance of $36,216,561. (Fact Stip. ¶ 56.)

21. A September 27, 2012 notice identifying 228 breached loans in the 2007-3 Trust, with a total original principal balance of $77,841,954.84. (Fact Stip. ¶ 57.)

U.S. Bank, in its capacity as Trustee, separately sent a series of letters that incorporated by reference all breach notices sent by Assured. (PX 401-403, 437, 496, 502-03, 506, 509, 510.)

The Court finds that these breach notices sent by Assured satisfied the notice requirements of section 2.03. These breach notices were sufficient to provide UBS with "written notice" under section 2.03, and to commence the 90-day period to cure, replace or repurchase of any defective loans that materially and adversely affected the interests of the Certificateholders or Assured.

The Court's summary judgment decision concluded that claims arising out of the August 30, 2012 breach notice were untimely. See 2015 WL 764665, at *8. However, as discussed below, subsequent New York case law has addressed the relationship between the 90-day cure period and New York's six-year limitations period for contract actions and undermines the reasoning behind the Court's prior dismissal of these claims on timeliness grounds. See, e.g., ACE Sec. Corp. v. DB Structured Prods., Inc. , 29 N.Y.S.3d 139, 158 (N.Y.Sup.Ct.N.Y.Cnty.2016) (questioning whether this Court's summary judgment analysis as to the August 30, 2012 breach notice remained viable in light of subsequent New York authority). For reasons that will be further explained, the Court now concludes that the Trusts may pursue relief as to the loans identified in this breach notice.

b. The U.S. Bank Breach Notices.

The Trusts contend that UBS received notice as to an additional 407 loans that were identified in breach notices sent by U.S. Bank in its capacity as Trustee. On September 26, 2012, U.S. Bank sent a breach notice to UBS identifying 180 loans that it claimed were in breach of the representations and warranties. (PX 504.) On January 14, 2013, U.S. Bank sent a breach notice to UBS identifying 227 loans that it claimed were in breach of the representations and warranties. (PX 512.) This breach notice claimed that the breaches were discovered "from a forensic review of the Transaction collateral pool, separate from prior breach notices." (PX 512.)

For the reasons explained below, the Court concludes that these breach notices are timely under the reasoning of Nomura and UBS was placed on notice of the purported breaches.

2. Upon Receipt of the Holt Report of August 2015, UBS Had Knowledge of Purported Breaches in the Loans Identified Therein.

On August 7, 2015, plaintiffs served UBS the Third Supplemental Expert Report of Ira H. Holt, Jr. (the "Holt Report"). (PX 552 (Holt Report); PX 947 (Transmittal e-mail).) Holt was the expert retained by the Trusts to conduct a full analysis of the loans claimed to be in breach of the representations and warranties. (Holt Report ¶¶ 1-3.) The Holt Report purported to identify all breaches contained in the loans held by the three Trusts.

Appendix 1 to the Holt Report purports to identify each loan that the Trusts claimed to be in breach. (PX 553-Appendix 1.) Each loan contained a line entry that identified which Trust contained that loan, the loan number, the UBS Loan ID and the Servicer Loan Number. (PX 553-Appendix 1.) The additional appendices listed the breaches contained in each loan.

It is undisputed that UBS received the Holt Report and its annexed appendices on or about August 7, 2015. UBS's expert, Deborah J. Grissom, purported to rebut on a loan-by-loan basis the conclusions of the Holt Report.

The Court finds that upon receipt of the Holt Report, UBS was placed on notice of the purported breaches identified therein.

In reaching this conclusion, the Court applies the reasoning of the New York Appellate Division, First Department, in Nomura Home Equity Loan, Inc. v. Nomura Credit & Capital, Inc. , 133 A.D.3d 96, 19 N.Y.S.3d 1 (1st Dep't 2015). This Court addressed the effect of Nomura in MASTR Adjustable Rate Mortgages Trust 2006 – OA2 v. UBS Real Estate Sec. Inc. , 2016 WL 1449751 (S.D.N.Y. Apr. 12, 2016), and adheres to that decision.In Nomura , various trustees brought claims against the sponsor of RMBS trusts, alleging the breach of representations and warranties. 133 A.D.3d at 98–102. The underlying PSAs included notice provisions similar to those in dispute here. Id. at 101–02.

The First Department concluded that trustees could bring claims alleging the breach of representations and warranties "that plaintiffs failed to mention in their breach notices or that were mentioned in breach notices sent less than 90 days before plaintiffs commenced their actions." 133 A.D.3d at 108. It reasoned that because "there were some timely claims" brought by the trustees, "a complaint amended to add the claims at issue would have related back to the original complaints." Id. In addition, "[p]laintiffs' presuit letters put defendant on notice that the certificateholders whom plaintiffs (as trustees) represented were investigating the mortgage loans and might uncover additional defective loans for which claims would be made." Id.

Applying the reasoning of Nomura to this action leads to the conclusion that notices of breach served during the pendency of this action relate back to the pre-suit notices for statute of limitations purposes.

Under Nomura , the post-suit notice of additional breaches of warranty that were not individually identified in pre-suit breach notices are not ineffective because timely pre-suit notices were sent by Assured to UBS that reserved the right to assert future additional breaches. For example, in a breach notice of August 9, 2010, Assured stated: "Please be advised that this repurchase request letter reflects the Insurer's findings after reviewing loan files relating to only a small portion of the Mortgage Loans included in the Transaction. The Insurer reserves all of its rights and remedies, including its right to give notice of additional breaches relating to any Mortgage Loans, as well as its rights under all other agreements, not referenced herein, related to the Transaction." (PX 334.)

UBS argues that the lack of pre-suit notice dooms the Trusts' claims for loans not identified in the Assured notices and that the Trusts' arguments is a failed effort to rewrite the terms of section 2.03. However, the notice provision quoted in Nomura was substantially similar to that set forth in the PSAs at issue here, and Nomura concluded that the existence of timely, pre-suit breach notices was sufficient for the trustees to pursue later-noticed breach claims.

Also, UBS's argument conflates the notice of breach given to the Trustee with the Trustee's obligation "in turn" to provide notice to UBS. The relevant language in section 2.03 states:

Upon receiving notice of a breach, the Trustee shall in turn notify the Transferor of such breach. The Trustee shall enforce the obligations of the Transferor in accordance with this Section 2.03 to correct or cure any such breach of a representation or warranty made herein, and if the Transferor fails to correct or cure the defect within such period, and such defect materially and adversely affects the interests of the Certificateholders and the Certificate Insurer in the related Mortgage Loan, the Trustee shall enforce the Transferor's obligations hereunder to (i) purchase such Mortgage Loan at the Purchase Price ....

(PX 49 at 0074.) This provision does not mandate any specific form of notice or procedure for notice by the Trustee is to UBS. Indeed, it does not expressly require that the notice be in writing. It only requires that "the Trustee shall in turn notify the Transferor" of a breach, and that "the Trustee shall enforce the Transferor's obligations hereunder ...." (PX 49 at 0074.)

Section 2.03 establishes notice-related obligations of the Trustee that are separate and distinct from those of other parties to the PSAs. These other parties are obligated to "give prompt notice" to all other parties, but are not tasked with enforcing UBS's remedial obligations. See PX 49 at 0074 ("Upon discovery by any of the Depositor, the Certificate Insurer, the Transferor, the Master Servicer, the Trust Administrator or the Custodian of a breach of a representation or warranty made by the Transferor pursuant to this Section 2.03 that materially and adversely affects the interests of the Certificateholders or the Certificate Insurer in any Mortgage Loan, the party discovering such breach shall give prompt notice thereof to the other parties and the Trustee."). Thus, section 2.03 contains both 1.) a "prompt notice" requirement whereby certain parties to the PSA, not including the Trustee, must notify the other parties of a breach, and 2.) a separate requirement that the Trustee specifically notify UBS of a breach and enforce the Trustee's obligations to remedy the breach.

UBS does not assert that the Holt Report failed to provide particularized notice of all breaches claimed by the Trusts. For the reasons explained above, as well as those discussed in the Court's prior decision applying Nomura, see 2016 WL 1449751, the Court finds that the Holt Report properly placed UBS on notice of all breaches claimed by the Trusts.

3. Lapse of the 90-Day Cure Period as Condition Precedent.

Section 2.03 establishes a 90-day period during which UBS is obligated to cure, repurchase or replace any breached mortgage loan. The relevant portion of section 2.03 states:

The Transferor hereby covenants that within ninety (90) days of the earlier of its discovery or its receipt of written notice from any party of a breach of any representation or warranty made pursuant to this Section 2.03 which materially and adversely affects the interest of the Certificateholders or the Certificate Insurer in any Mortgage Loan, it shall cure such breach in all material respects, and if such breach is not so cured, shall, (i) if such ninety (90) day period expires prior to the second anniversary of the Closing Date, remove such Deleted Mortgage Loan from the Trust Fund and substitute in its place an Eligible Substitute Mortgage Loan or Mortgage Loans into the Trust Fund, in the manner and subject to the conditions set forth in this Section; or (ii) repurchase the affected Mortgage Loan or Mortgage Loans from the Trustee at the Purchase Price in the manner set forth below.

(PX 49 at 0074.)

As discussed, section 2.03 contains a 90-day period during which UBS must cure the defective loan. UBS contends that the lapse of the 90-day cure period acts as a condition precedent to bringing suit, and that the 90-day period must have elapsed before the Trusts could bring a viable claim as to a given loan. (Def. FF ¶ 346.) Nomura is again instructive. In Nomura, the First Department held that the plaintiff trustees could bring claims as to loans "that plaintiffs failed to mention in their breach notices or that were mentioned in breach notices sent less than 90 days before plaintiffs commenced their actions ." 133 A.D.3d at 108 (emphasis added). It reasoned that because the complaint included claims based on written notices for which the full 90-day period had lapsed, these additional, unexpired claims all related back to the fully-lapsed claims: "Unlike the situation in [ACE Securities Corp. v. DB Structured Products, Inc. , 112 A.D.3d 522–23, 977 N.Y.S.2d 229 (1st Dep't 2013) ], there were some timely claims in these cases. Hence, a complaint amended to add the claims at issue would have related back to the original complaints." Nomura , 133 A.D.3d at 108 (citing Koch v. Acker, Merrall & Condit Co. , 114 A.D.3d 596, 597, 981 N.Y.S.2d 70 (1st Dept. 2014) ).

Thus, Nomura expressly holds that when a plaintiff asserts timely claims brought after the lapse of the cure period, a plaintiff may thereafter bring additional claims directed to breaches that had not yet satisfied the 90-day cure period. Here, the Complaint alleged that UBS received timely written breach notices as to 4,460 loans prior to the commencement of this action, and that the 90-day cure period had expired at the time that this action commenced. Applying Nomura , the Court finds that the additional breach claims are not disqualified on the basis of the cure period set forth in section 2.03.²

2 Further, New York Supreme Court Justice Marcy S. Friedman in a series of decisions has concluded that the lapse of PSAs' cure periods "are not substantive conditions precedent that must be satisfied in order for the cause of action even to come into existence." ACE Sec. Corp. v. DB Structured Products, Inc. , 29 N.Y.S.3d 139, 157 (N.Y.Sup.Ct.N.Y.Cnty.2016) (amended complaint could add claims for additional loans with newly expired cure periods); Wilmington Trust Co. v. Morgan Stanley Mortg. Capital Holdings LLC , 2016 N.Y. Slip Op. 31130(U) (N.Y.Sup.Ct.N.Y.Cnty. June 14, 2016) (applying ACE ); Deutsche Bank Nat'l Trust Co. v. Barclays Bank PLC , 2016 N.Y. Slip Op. 31056(U), at 5 (N.Y.Sup.Ct.N.Y.Cnty. June 8, 2016) (applying ACE ).

4. Discovery of a Breach by UBS Triggers Its Cure and Repurchase Obligations.

While the foregoing discussion of the Trustee's notice to UBS in the Holt report disposes in its entirety the issue of timely notice to UBS of all loans as to which the Trustee asserts a breach, the Court will address the Trustee's alternate arguments that turn upon UBS's own discovery of breaches.

Section 2.03 expressly states that "[u]pon discovery by ... the Transferor [UBS] ... of a breach of a representation or warranty made by the Transferor pursuant to this Section 2.03 that materially and adversely affects the interests of the Certificateholders or the Certificate Insurer in any Mortgage Loan, the party discovering such breach shall give prompt notice thereof to the other parties and the Trustee." In this case, the Trustee alleges that UBS discovered certain breaches but failed to give the Trustee notice. The logical consequence of the failure of UBS to give the Trustee the required notice upon UBS's own discovery of a breach is that the required notice by the Trustee is dispensed with and UBS's repurchase obligation is nevertheless triggered. This is because UBS may not thwart the notice requirement, a condition precedent to its obligation to cure or repurchase, by deliberately refraining from giving the required notice.

As the Court observed in its summary judgment ruling, courts interpreting similar provisions have held that the transferor has an obligation to cure any breach that it independently discovers. See, e.g. , ACE Sec. Corp. Home Equity Loan Trust, Series 2007 – HE3 v. DB Structured Prods., Inc. , 5 F.Supp.3d 543, 560 (S.D.N.Y.2014) (Nathan, J.) (permitting plaintiff to pursue non-noticed claims for breaches that defendant independently discovered, but noting that denial of motion to dismiss "does not relieve Plaintiff of its burden of proving loan-by-loan breaches at later stages of litigation.").

The Court construes section 2.03 to mean that "[u]pon discovery" of a breach by UBS and a failure by UBS to give the Trustee "prompt notice thereof," the obligation of the Trustee to give UBS notice of its cure or repurchase obligation is excused and UBS is deemed to have received prompt notice through its own discovery, thereby triggering its obligation to cure or repurchase.

5. Knowledge on the Part of UBS Cannot Be Premised on "Pervasive Breach" or "Constructive Knowledge".

In its summary judgment decision, this Court concluded that the Trusts could not prove discovery by UBS based on a theory of "pervasive breach." See 2015 WL 764665, at *10–12. That conclusion stands.

The Trusts previously urged that the breach notices sent by Assured had the cumulative effect of placing UBS on notice that a significant number of loans placed in the pools were in breach of the warranties, beyond those specifically identified. Id. The Court concluded "that the terms of the PSAs foreclose such a broad and improvised remedy," and observed that the PSAs provided for (1) an individualized, loan-specific obligation to cure, replace or repurchase a breached loan, (2) remedies only as to those breaches that materially and adversely affected the interests of Certificateholders, and not as to all breaches, and (3) the "sole remedies" of cure, replacement or repurchase, as opposed to other PSAs that provided for broader remedies in certain defined circumstances. Id.

Although the Trusts no longer use the term "pervasive breach," they argue that UBS had "constructive knowledge" of "most" breaches. (Pl. FF ¶¶ 188-92.) According to the Trusts, in conducting due diligence prior to the closing of the three Trust transactions, UBS knew or should have known about widespread breaches in the pools based on 1.) higher-than-average levels of risk identified in samples drawn from the loans, 2.) vendor reports that certain loans had breached the PSAs and 3.) UBS's review of its vendors' reports. (Pl. FF ¶ 192.) As described by the Trusts: "UBS had the ability to investigate its suspicions before closing ... and, as a sophisticated entity with significant resources, UBS could have conducted a broader review of the Loan pools in the Trusts by the Closing Dates." (Pl. FF ¶ 192.)

The Trusts' constructive knowledge argument is little more than a reformulation of their "pervasive breach" theory. (See Pl. FF ¶¶ 188-92.) In raising this argument, the Trusts do not contend that UBS discovered breaches that triggered its remedial obligation under section 2.03. Rather, they urge that UBS "could have conducted a broader review" across the loan pools to identify specific breaches, at which point, it might have discovered specific breaches that triggered remedial obligations. This understanding is not based in the language of the PSAs, and turns on what the plaintiffs believe UBS would have learned if it had adopted a different diligence protocol. But as the Court discussed in its summary judgment decision, the PSAs provide for "sole remedies" that apply to breaches on an individualized loan-by-loan basis. See 2015 WL 764665, at *11. Section 2.03 provides for a loan-specific repurchase remedy, triggered "[u]pon discovery" by UBS or upon notice to UBS sent by another party to the PSAs. The Trusts' "constructive knowledge" theory would create a broader obligation for UBS than the one negotiated by the parties. As the Court noted in its summary judgment decision: "Other sophisticated parties have bargained for a broader obligation to repurchase the entire pool of loans under specified circumstances. These parties did not. Nor did these parties, unlike others, bargain for a broad remedy provision making the obligation to cure or repurchase a nonexclusive remedy." 2015 WL 764665, at *11.

The parties could have, but did not, bargain for additional remedies or a notice provision that did not turn on loan-specific knowledge. The Trusts therefore may not rely on evidence of "constructive knowledge" or "pervasive breach" to prove UBS's knowledge of breached warranties.

6. UBS's Knowledge of a Breach May Be Proved Circumstantially or Through Evidence of Willful Blindness.

No direct proof is required to establish that UBS discovered breaches of additional non-noticed loans; it may be proved by circumstantial evidence. "Knowledge may be established by circumstantial evidence, in the face even of professions of ignorance, but knowledge there must be, or negligence so reckless as to betoken indifference to knowledge." Woloszynowski v. New York Cent. R. Co. , 254 N.Y. 206, 208–09, 172 N.E. 471 (1930) (internal citations omitted); see also Reed v. Fed. Ins. Co. , 123 A.D.2d 188, 195, 510 N.Y.S.2d 618 (1st Dep't 1987) (jury was properly instructed that insurer could prove insured's knowledge of loss through circumstantial evidence).

Plaintiffs also may rely on evidence of willful blindness on the part of UBS to prove that UBS had knowledge that the representations and warranties had been breached. Under New York law, where a party "had sufficient information to impose a duty upon it to make further inquiry ... its failure to do so constituted ‘willful blindness.’ " Scher Law Firm, LLP v. DB Partners I, LLC , 97 A.D.3d 590, 591–92, 948 N.Y.S.2d 335 (1st Dep't 2012). " ‘[A] willfully blind defendant is one who takes deliberate actions to avoid confirming a high probability of wrongdoing and who can almost be said to have actually known the critical facts.’ " Id. at 592, 948 N.Y.S.2d 335 (quoting Global – Tech Appliances, Inc. v. SEB, S.A. , 563 U.S. 754, 769, 131 S.Ct. 2060, 179 L.Ed.2d 1167 (2011) ); see also Viacom Int'l, Inc. v. YouTube, Inc. , 676 F.3d 19, 35 (2d Cir.2012) ("A person is ‘willfully blind’ or engages in ‘conscious avoidance’ amounting to knowledge where the person was aware of a high probability of the fact in dispute and consciously avoided confirming that fact.") (describing willful blindness in federal common law).

a. The Trusts Have Proved Discovery of Breaches as to Certain Non-Noticed Loans.

As will be discussed in greater detail below, the Trusts have proved that UBS independently discovered breaches at to 41 non-noticed loans. Adrian Wu, who worked at UBS as a "structurer" in the securitization of mortgage pools, learned prior to the Closing Date of the 2007-3 Trust that borrower FICO scores were inaccurately listed in the Trust's MLS, and that the updated FICO scores for those borrowers were materially lower than listed. (Tr. 206, 215, 217 & PX 208.) Wu testified that a "significant" decline in FICO scores indicated an increased risk of borrower default. (Tr. 215.) However, the MLS was not revised to reflect the revised FICO scores prior to the Closing Date.

The Court finds that UBS independently discovered these breaches of the MLS Warranty. UBS had knowledge that the MLS Warranty listed borrower FICO scores that were materially different from these borrowers' updated FICO scores, which triggered UBS's obligation to cure the defects within 90 days, or thereafter repurchase or replace the defective loans.

b. The Trusts Have Not Proved Willful Blindness on the Part of UBS.

UBS's surveillance group monitored the performance of the loans that had been transferred to the Trusts. (Tr. 1397-98.) UBS had a policy to monitor any loan if it became delinquent by 30 days, at which point it would gather information from the entity servicing the loans, including notes on discussions with borrowers. (Tr. 1400-01, PX 41 at 99.) If loans were delinquent by 60 days, UBS's policy was to submit a subset of them for a re-underwriting review. (Tr. 1401-02; PX 41 at 99.) UBS's policy required it to conduct such a review of all loans that were 90 days delinquent. (Tr. 1408; PX 41 at 99.) UBS hired third-party firms to conduct these reviews, primarily Clayton Fixed Income Services and Wells Fargo Bank. (Tr. 1404.) These vendors reviewed the loan files for the loans identified by UBS, compared them to the Originators' underwriting guidelines to determine if the loans met the guidelines and then reported their findings to UBS. (Tr. 1405, 1410-11.)

John Lantz, who managed surveillance at UBS, testified that "a purpose of doing surveillance was reducing Certificateholders' losses by removing nonperforming loans from securitizations and putting them back to the originators[.]" (Tr. 1411.) Lantz testified that "[t]he surveillance group was in place to look at the performance of the loans because, you know, if we had loans that were, you know, going delinquent or going default or experiencing high losses, it would be very easy for investors to go to some other bank and buy there." (Tr. 1398.) Lantz testified that once the loans had been packaged and sold to investors, surveillance continued of those loans because, in essence, UBS needed an awareness of the quality of its product in order to maintain a competitive position in the investor marketplace for future sales to investors. (Tr. 1398-1400.)

Lantz testified that in the course of performing underwriting reviews, he learned that certain loans had breached underwriting guidelines. (Tr. 1454.) Lantz also testified that UBS monitored the loan performance with an awareness of its put-back right, and would demand an Originator's repurchase if a loan was deemed to be defective and non-performing. (Tr. 1406-07, 1411.) He testified that the presence of a guideline breach did not prompt him to look for additional breached guidelines in the rest of the pool. (Tr. 1457-58.)

Delinquencies in loan repayments increased throughout 2007. (Tr. 1467.) American Home, which originated a substantial number of loans in the 2007-1 Trust, filed for Chapter 11 bankruptcy protection in August 2007, which, Lantz testified, limited any put-back right UBS had as to American Home-originated loans. (Tr. 1467-71.)

Lantz testified that UBS closed its surveillance group in January 2008. (Tr. 1412.) In the lead-up to the surveillance group's closing, internal UBS e-mails discussed significant losses within its securitized RMBS pools. (Tr. 1417-26.) Toward the end of January 2008, Lantz was asked by two UBS traders if UBS could repurchase loans out of the MARM 2007-1 and put them back to Originators that remained in solid financial condition. (PX 309 at 2; Tr. 1483-84.) Lantz told them "we do not conduct R&W reviews on loans in our deals anymore," and explained "the various reasons why." (PX 309 at 2.)

Lantz testified that by January 2008, UBS had ceased surveillance monitoring of the three Trusts at issue in this case. (Tr. 1482-83.) Lantz testified that the surveillance group laid off "hundreds of people, almost our entire business in the Fall of 2007, and then essentially everyone else was laid off in the Spring of 2008." (Tr. 1486.) He said that this was because "the entire business was imploding.... We already had laid off hundreds of traders." (Tr. 1486-87.) Lantz also testified in his deposition that he did not recall the reasons why UBS shut down its surveillance program other than that "it was a decision informed by counsel." (Tr. 1494-95.)

The Trusts have not established that UBS was willfully blind to widespread breaches of warranties across the loans in the three Trusts. The evidence submitted by the Trusts, including the testimony of Lantz, reflects that UBS ceased its surveillance operations around the same time that it wound down its business of structuring and selling RMBS pools. Lantz's testimony reflects that the principal purpose of the surveillance group was to monitor loans with an eye toward the marketing of future RMBS products to investors. The surveillance group exercised put-back rights and would demand the repurchase of defective loans, and performed reunderwriting reviews as part of that process.

As the housing market began its collapse, the surveillance group would have become aware of growing delinquencies on the part of buyers, and learned of guidelines breaches. The closing of the surveillance group also dovetailed with the collapse of the broader housing market and investor interest in RMBS deals. This does not prove that UBS took "deliberate actions to avoid confirming a high probability of wrongdoing," or that UBS "can almost be said to have actually known the critical facts" concerning widespread breaches across the three trusts. Scher Law Firm , 97 A.D.3d at 591–92, 948 N.Y.S.2d 335. The Trusts have not proved that UBS took deliberate actions to avoid knowledge that widespread breaches had occurred throughout the loans held by the Trusts, thereby triggering UBS's repurchase obligations.

THE MORTGAGE LOAN SCHEDULE WARRANTY.

A. Overview of the MLS Warranty.

The Mortgage Loan Schedule ("MLS") is a listing of the loans transferred to the Trusts pursuant to the PSA, together with key data about the loans, including the amount of the loan, whether the property is owner-occupied, whether it is a single family dwelling, the Loan-to-Value ("LTV") ratio, the Debt-to-Income ("DTI") ratio and the borrower's credit (or FICO) score. It is the principal source of information regarding the loans provided by UBS to the Trust and its investors.

UBS made the following representation and warranty concerning the MLS: "The information set forth in the Mortgage Loan Schedule was true and correct in all material respects at the date or dates respecting which such information is furnished as specified in the Mortgage Loan Schedule ...." (PX 49 at 0191 & subsection (i) to Schedule II of each PSA.) Notably, it is not limited to information known or believed to be true. It is an unqualified warranty of the truth and correctness of the information on the MLS.

The MLS is defined in the PSAs as: "The list of Mortgage Loans ... transferred to the Trustee as part of the Trust Fund and from time to time subject to this Agreement ... setting forth the following information with respect to each Mortgage Loan ...." (PX 49 at 0045.) The definition lists 40 items of information that must be included in the MLS for each loan. (PX 49 at 0045-46.) Each MLS is a lengthy spreadsheet containing information about thousands of loans. (See PX O01-O04.) Items of information include the original balance on the mortgage loan and whether the mortgage loan was an adjustable rate mortgage. According to the Trusts, there are 9,439 breaches contained in the Mortgage Loan Schedules of the three Trusts.

B. The Timing of Representation and Warranty Listed in the MLS.

The MLS Warranty is made "at the date or dates respecting which such information is furnished as specified in the Mortgage Loan Schedule ...." (PX 49 at 0191.) To the extent that "the date or dates" related to such information is "specified" in the PSAs' definitions for "Mortgage Loan Schedule," UBS warranted the MLS data as of those given date or dates.

Of the forty items of information required for the MLS, the Trusts assert that UBS failed to supply information that was "true and correct in all material respects" as to the following seven items. Each item is identified by the corresponding number set forth in the PSAs' definition of "Mortgage Loan Schedule":

(13) a code indicating whether the Mortgaged Property is owner occupied, a second home or an investor property;

(14) a code indicating whether the Mortgaged Property is a single family residence, a two-family residence, a three-family residence, a four-family residence, a planned-unit development, a condominium or a Cooperative Unit;

(15) a code indicating the loan purpose (i.e., purchase, rate/term refinance, cash-out refinance); ...

(20) the Loan-to-Value Ratio at origination ; ...

(22) a code indicating the documentation style of the Mortgage Loan; ...

(26) the credit score (or mortgage score) of the Mortgagor;

(27) the debt-to-income ratio of the Mortgage Loan ....

(PX 49 at 0045.) Of these seven items, only the Loan-to-Value Ratio (or "LTV") is warranted to be true at correct "at origination." None of the other items specifies a point in time at which their contents are warranted to be true.

For these remaining six items, Schedule II establishes that all representations and warranties are true as of the "Closing Date" defined in the PSAs, unless specified that they are true as of the "Cut-off Date." All individual representations and warranties are preceded by this provision:

UBS Real Estate Securities Inc. (the "Transferor ") hereby makes with respect to those Mortgage Loans sold by it to the Depositor pursuant to the Mortgage Loan Purchase Agreement, the following representations and warranties as of the Closing Date or, if so specified herein, as of the Cut-off Date.

(PX 49 at 0191.) Thus, all representations and warranties in Schedule II are made as of a PSA's "Closing Date," unless specified that they are made as of the PSA's "Cut-off Date."

The MLS Warranty does not incorporate the Cut-off Date, so the truth and correctness of all disputed MLS items, other than LTV ratio, are determined with reference to the Closing Date. Each PSA separately defines a Closing Date. For the 2006-OA2 Trust, the Closing Date is November 15, 2006. (PX 49 at 0029.) For the 2007-1 Trust, the Closing Date is January 16, 2007. (PX 110 at 0039.) For the 2007-3 Trust, the Closing Date is May 15, 2007. (PX 182 at 0030.)

UBS warranted the LTV "at origination," i.e. the time the borrower's loan was funded by the Originator. For all other disputed items in the MLS, UBS warranted the information as it existed at each PSA's Closing Date which are November 15, 2006 (2006-OA2), January 16, 2007 (2007-1) and May 15, 2007 (2007-3).

C. The MLS Warranty is Absolute and Does Not Merely Guarantee Accurate Transcription.

UBS argues that the MLS Warranty is "more reasonably read as a transcription rep," and should be interpreted as "referring only to information ‘furnished’ to UBS RESI regarding loans it did not originate ...." (Def. FF ¶ 161.) It argues that UBS could not guarantee the absolute truth of the thousands of MLS data points, and that the more plausible reading is that UBS accurately transcribed the information as transmitted to it by the Originators. (Def. FF ¶¶ 162-65.)

UBS's argument is not supported by the language of the PSAs. As noted, the MLS Warranty is not limited to UBS's knowledge or belief; it is an unqualified warranty. The PSAs could have been drafted to state that the MLS reflected only an accurate transcription of the information communicated by the Originators. But there is no such limitation, and courts "must be careful not to add new terms or alter the terms of the contract in the guise of interpreting it." Rosenthal , 69 A.D.3d at 507, 894 N.Y.S.2d 32. Without qualification, the MLS Warranty states that the MLS "was true and correct in all material respects at the date or dates respecting which such information is furnished as specified in the Mortgage Loan Schedule ...." This language expressly warrants the truth and correctness of the information "furnished" in the MLS on the dates specified therein.

It is true that the MLS Warranty imposes a form of strict or absolute liability for a materially untrue or incorrect statement on the MLS. While at first blush this may appear harsh because a breach may be premised upon a post-origination but pre-Closing change in a borrower's circumstance, e.g. FICO score, income, debt, that was unknown and, in some cases, unknowable to the Originator or UBS. But there is some symmetry between the absolute nature of the warranty and the limited nature of the remedies in the event of breach. UBS conducted due diligence on only 25% of the loans it bought from Originators and otherwise used equivalent data tapes from the Originators to price its bid. If the information UBS, in turn, provided to the Trusts was wrong, it was only exposed to the possibility of cure, replacement or repurchase. For misstatements based upon pre-origination events, UBS also could protect itself with remedies against the Originator. For post-origination, pre-Closing changes, UBS simply assumed the economic risk. Symmetrical or not, harsh or not, this is the bargain to which UBS, a sophisticated financial institution, agreed.

UBS argues that, unlike other RMBS transactions, the PSAs here do not contain a warranty against borrower fraud. (Def. FF ¶ 163.) UBS argues that to interpret the MLS Warranty as directed to more than the accuracy of transcription is the equivalent to imposing a no-fraud warranty that the parties did not negotiate. UBS overstates the case. The MLS may contain an incorrect item of information as a result of an innocent mistake and yet be liable to the Trusts for the untrue and incorrect information. Imputing fraud or fault on the part of a borrower, underwriter or any other person or entity, including UBS, is not an element of the claim.

UBS asserts that many of its employees and others in the RMBS field thought of a MLS Warranty as merely a transcription warranty and that it should be interpreted as a warranty that UBS accurately transmitted the information received from the Originator with no assurance as to its accuracy. The interpretation is flatly contradicted by the plain and unambiguous terms of the MLS Warranty. It also reads out of the PSAs the temporal distinction made in the PSA between, for example, an LTV ratio listed on the MLS, which is warranted as of the date of origination of the loan, and a DTI ratio listed on the same MLS that is warranted as of the Closing Date of the PSAs. UBS's attempt to rely on parol evidence or custom and usage to vary, modify or contradict the terms of a plain and unambiguous provision is meritless. See, e.g. , Johnson , 84 A.D.3d at 435, 922 N.Y.S.2d 57 ("[T]hat one party to an agreement may attach a particular, subjective meaning to a term that differs from the term's plain meaning does not render the agreement ambiguous.").

D. The Trusts' Evidence Concerning MLS Warranty Breaches.

The Trusts' evidence concerning breaches of the MLS Warranty was introduced through their experts. Holt testified that in reviewing the MLS data, he "focused on the characteristics that are strongly associated with risk ...." (Holt Direct ¶ 134.) Specifically, he considered the accuracy of MLS data concerning occupancy, borrower credit or FICO score, LTV ratio, loan documentation type, DTI ratio, property type and loan purpose. (Holt Direct ¶ 134.)

In determining whether a breach materially and adversely affected the interest of the Certificateholders, Holt considered whether the MLS defect significantly increased the risk of loss. (Holt Direct ¶ 135.) Holt testified that when a data defect is the result of fraud or misrepresentation by a borrower, broker, appraiser or underwriter, the breach is material because it undermines the integrity of the underwriting process. (Holt Direct ¶ 135.) He also testified that when an MLS defect did not result from fraud, and the loan remained within the applicable guidelines despite the MLS defect, a material and adverse effect could arise because the defect could still increase the risk of loss. (Holt Direct ¶ 135.) As an example, he notes that the risk of loss on a loan with a 35% DTI ratio is higher than the risk of loss on a loan with a 20% DTI ratio, even if both loans are within the guidelines. (Holt Direct ¶ 135.) Lastly, Holt testified that when an MLS breach caused a loan to fall outside of the guidelines, but an exception would likely have been granted based on compensating factors, the data defect would still significantly increase the risk of loss. (Holt Direct ¶ 135.)

Holt testified that there may be several reasons why data in an MLS may be inaccurate. (Holt Direct ¶ 131.) He stated that information may have been misrepresented during origination; the underwriter may have erred in applying the applicable guidelines; the underwriter may have erred in calculating a metric; or UBS may have included information that did not correspond with the information in the loan file. (Holt Direct ¶ 131.)

As is apparent, some but not all MLS data defects maybe the result of deceit by the borrower. An intentional misrepresentation by a borrower calls into question all information furnished by the borrower and casts doubt on both the borrower's ability to repay and willingness to repay. MLS data defects that are the result of innocent mistakes by the borrower, the underwriter or third-parties require a closer examination to determine whether they materially increased the risk of loss at the date of discovery or notice of breach.

E. The Seven Items of MLS Data at Issue.

As noted, the Trusts contend that UBS breached the MLS Warranty as to seven items of information required. The Court reviews and, where necessary, construes the contractual requirements for each item.

1. Occupancy Status.

The MLS must include "a code indicating whether the Mortgaged Property is owner occupied, a second home or an investor property ...." (PX 49 at 0045.) This information must be true and correct as of the Closing Date for the applicable PSA. (See, e.g. , PX 49 at 0191.) For the 2006-OA2 Trust, Column BC of the MLS is headed "Occupancy," and describes the relevant property with phrases like "Owner Occupied," "Investor Occupied," and "Second Home." (See, e.g. , PX O01.) The Trusts contend that UBS breached the MLS Warranty as to occupancy status for 822 loans. (Pl. FF ¶ 252, 207.)

The Trusts may prove a breach of the MLS Warranty by establishing that it is more likely than not that the listed occupancy status was not "true and correct in all material respects" as of the Closing Date.

The Trusts must also prove that "such defect materially and adversely affects the interests of the Certificateholders and the Certificate Insurer in the related Mortgage Loan." (PX 49 at 074.) As to property occupancy, there will be a difference in the kind and quality of proof between a refinance transaction and a new home purchase. With respect to a refinance transaction, the borrower likely knows to a certainty whether at the time of origination he or she is occupying the property; there is little room for innocent mistake. With respect to a new home purchase, the statement regarding occupancy is inherently a statement of future intention for which there could be the possibility of an innocent explanation for a delayed or changed plan (sickness, death, loss of employment).

The cause of the incorrect statement (for example, intentional misstatement by the borrower vs. delayed plan by the borrower for unforeseen reasons) may have bearing on whether the incorrect statement materially and adversely affected the interests of the Certificateholders at the time of discovery or notice.

2. Property Type.

The MLS must include "a code indicating whether the Mortgaged Property is a single family residence, a two-family residence, a three-family residence, a four-family residence, a planned-unit development, a condominium or a Cooperative Unit ...." (PX 49 at 0045.) This information must be true and correct as of the Closing Date for the applicable PSA. (See, e.g. , PX 49 at 0191.) For the 2006-OA2 Trust, Column BT of the MLS is headed "Property Type," and describes the relevant property with phrases like "Single Family," "Two Family," "Low-Rise Condominium" and "Planned Unit Development." (See, e.g. , PX O01.)

The Trusts may prove a breach of the MLS Warranty of property type if they prove that it is more likely than not that the property type listed on the MLS was not "true and correct in all material respects" as of the Closing Date. The Trusts must also prove that "such defect materially and adversely affects the interests of the Certificateholders and the Certificate Insurer in the related Mortgage Loan." (PX 49 at 074.) The Trusts contend that UBS breached the MLS Warranty as to the property type for 88 loans. (Pl. FF ¶ 218.)

3. Loan Purpose.

The MLS must include "a code indicating the loan purpose (i.e., purchase, rate/term refinance, cash-out refinance) ...." (PX 49 at 0045.) This information is warranted as true and correct as of the Closing Date for the applicable PSA. (See e.g. , PX 49 at 0191.) Column BA of the MLS is headed "Loan Purpose," and describes the relevant loan with phrases like "Purchase," "Cash Out Refinance" and "Rate/Term Refinance." (See, e.g. , PX O01.)

The Trusts may prove a breach of the MLS Warranty as to loan purpose if they prove that it is more likely than not that the loan purpose as stated in the MLS was not "true and correct in all material respects" as of the Closing Date. The Trusts must also prove that "such defect materially and adversely affects the interests of the Certificateholders and the Certificate Insurer in the related Mortgage Loan." (PX 49 at 074.) The Trusts contend that UBS breached the MLS Warranty as to the loan purpose for 98 loans. (Pl. FF ¶ 252.)

4. LTV Ratio.

a. Construction.

The MLS must include "the Loan-to-Value Ratio at origination ...." As discussed, unlike the other items of information, the LTV ratio is warranted to be true and correct "at origination." (PX 49 at 0045.)

"Loan-to-Value Ratio" is defined in the PSAs as follows:

With respect to any Mortgage Loan and as to any date of determination, the fraction (expressed as a percentage) the numerator of which is the principal balance of the related Mortgage Loan at such date of determination and the denominator of which is the Appraised Value of the related Mortgaged Property. For purposes of representation (xxxi) of Schedule II, the Loan-to-Value Ratio will be the loan-to-value ratio calculated in accordance with applicable state laws regarding primary mortgage insurance.

(PX 49 at 0042-43.) The "Appraised Value," is necessary to calculate the LTV and it is defined as follows:

With respect to any Mortgage Loan, the Appraised Value of the related Mortgaged Property shall be: (i) with respect to a Mortgage Loan other than a Refinancing Mortgage Loan, the lesser of (a) the value of the Mortgaged Property based upon the appraisal made at the time of the origination of such Mortgage Loan and (b) the sales price of the Mortgaged Property at the time of the origination of such Mortgage Loan; and (ii) with respect to a Refinancing Mortgage Loan, the value of the Mortgaged Property based upon the appraisal made at the time of the origination of such Refinancing Mortgage Loan as modified by an updated appraisal.

(PX 49 at 0023.) Finally, the PSA defines a "Refinancing Mortgage Loan" as "[a]ny Mortgage Loan originated in connection with the refinancing of an existing mortgage loan." (PX 49 at 0058.)

The PSA requires different methods for calculating the LTV ratio depending on the type of underlying mortgage loan. The LTV ratio for a Refinancing Mortgage Loan is calculated using the value of the mortgaged property "based upon the appraisal made at the time of the origination of such Refinancing Mortgage Loan as modified by an updated appraisal." For all other types of mortgage loans, the LTV is calculated using either "the lesser of" the appraisal "at the time of origination" and the property's sales price "at the time of origination of such Mortgage Loan ...."

Unlike other items of data, the "V" or value component of the LTV ratio is not based upon a fact, but an opinion as to the value of property. The opinion is generally based upon the condition and location of comparable properties, but it is nevertheless an opinion. A property of equal size and condition next to a gas station will be worth less than one in the midst of other homes with an ocean view. An opinion as to value is based on weighing and evaluating the total mix of information about the property, including conflicting and contradictory data, intangible considerations and considerations unique to the property.

At trial, Dr. Charles Cowan, the Trusts' main witness on the LTV ratio, acknowledged that an appraisal, whether done manually in accordance with industry standards or based upon computer-generated data, was nevertheless an opinion:

THE COURT: A value generated by an AVM is an opinion as to the value of the property; is that your testimony?

THE WITNESS: Yes.

THE COURT: Okay. And an appraisal is also an opinion, correct?

THE WITNESS: Yes, sir.

THE COURT: Thank you.

THE WITNESS: Okay.

THE COURT: So when we talk about loan-to-value ratio, to the extent that ratio incorporates an appraised value, whether it comes from an appraisal or from an AVM, that number, it captures an opinion, correct?

THE WITNESS: Yes, sir.

(Tr. 1595.) Consistent with the foregoing, he also testified that he was offering no opinion about whether the appraisals that were incorporated into the LTV followed industry standards for a properly conducted appraisal:

Q. You're not offering an opinion as to whether any of the appraisals here were performed consistently with the appraisal standards, are you?

A. No, I didn't address that at all in my report. I addressed values of the appraisals.

(Tr. 1574-75.)

The MLS Warranty as to LTV ratio is premised upon an opinion and, thus, stands on a different footing than the other MLS warranties. The truth or correctness of an opinion is analyzed differently than the truth or correctness of an objectively verifiable fact.

In Fait v. Regions Fin. Corp. , 655 F.3d 105 (2d Cir.2011), the plaintiffs alleged that the corporation misstated the value of its goodwill. It was brought not as fraud case requiring proof of an intent to deceive, but as a material misstatement case under section 11 of the Securities Act of 1933. The Second Circuit held that, because the value of goodwill was a matter of opinion, to state a claim for an actionable misstatement, the plaintiffs were required to "plausibly allege that defendants did not believe the statements regarding goodwill at the time they made them." Id. at 112 ; accord City of Omaha, Neb. Civilian Employees' Ret. Sys. v. CBS Corp. , 679 F.3d 64, 66 (2d Cir.2012) (per curiam). New York law is not different. Sidamonidze v. Kay , 304 A.D.2d 415, 416, 757 N.Y.S.2d 560 (1st Dep't 2003) (distinguishing non-actionable "opinions of value" from actionable "false statements of value"); N. Grp. Inc. v. Merrill Lynch, Pierce, Fenner & Smith Inc. , 135 A.D.3d 414, 422, 22 N.Y.S.3d 421 (1st Dep't 2016) (distinguishing non-actionable "opinions of value" for commercial mortgage-backed securities from actionable "false statements of value" as to the same) (citing Kay ); cf. Collins v. Hydorn , 12 N.Y.S. 581, 582, 34 N.Y.St.Rep. 574 (3d Dep't. 1890) ("Difference in opinion as to values naturally exists. It would be unjust to impeach a transaction as fraudulent because of an honest difference of opinion as to values."), aff'd , 124 N.Y. 641, 27 N.E. 412 (1891).

In context, an LTV ratio appearing on the MLS was a statement of opinion known by the Trust or purchaser of Certificates to be reflective of an opinion of value and was not a statement of an objectively verifiable fact.

With respect to the amount of the loan—the "L" in the LTV ratio—the Trusts need only prove that it is more likely than not that at the time of origination the loan amount was not the amount implied by the LTV ratio on the MLS. The Trusts do not contend that the LTV ratio listed on the MLS was untrue or incorrect because the amount of the loan was misstated.

With respect to a claim that the LTV is not accurate because the appraised value is wrong, the Trusts must prove that it is more likely than not that the appraiser, the underwriter and/or UBS did not honestly believe that the appraised value reflected in the schedule. Circumstantial evidence may be used to satisfy this burden.

The Trusts must also prove that it is more likely than not that any LTV breach "materially and adversely affects the interests of the Certificateholders ... in the related Mortgage Loan." (PX 49 at 0074.)

At trial, the Trusts did not endeavor to prove and thus failed to prove by a preponderance of the evidence that any appraiser failed to follow accepted industry practices in conducting an appraisal, that any appraiser acted corruptly or dishonestly, or that any appraiser, underwriter or Originator did not honestly believe in the accuracy of the appraisal as a statement of opinion. No evidence was presented that UBS did not honestly believe any particular appraisal. The Trusts claims premised upon a breach of the MLS Warranty as it relates to LTVs fail in their entirety.

b. AVM Data Is Not A Substitute for An Appraisal.

Separately, as an alternative to the Court's findings and conclusions that the Trusts failed to prove that it is more likely than not that the appraiser, the underwriter and/or UBS did not honestly believe the appraised value reflected in the MLS, the Court considers the probative value of the AVMs relied upon by the Trusts to endeavor to prove that the LTV ratios were incorrect.

The Trusts rely on the testimony of Dr. Charles M. Cowan, managing partner of Analytic Focus LLC. Cowan, who is not a licensed appraiser in any state, testified about his use of an "automated valuation model" ("AVM") to calculate the appraisal values for the Trusts' loans.

An AVM is a valuable analytical tool that is widely-used for certain limited purposes. In the underwriting or due diligence process, an AVM could be of assistance in drawing conclusions about a large pool of properties or even identifying a specific property for further review.

The Trusts offered no evidence that, for the type of loans at issue in this action, the guidelines of any Originator permitted the underwriter to dispense with a conventional appraisal in favor of an AVM, which does not include an evaluation of the condition of the property, amenities or disamenities, atypical circumstances, or political boundaries in which the property is situated, including school district or taxing authority.

As will be discussed, the AVMs used in this action cannot tell whether the roof of a home has a gaping hole or, instead, a beautiful skylight. It cannot tell whether the kitchen and bathrooms were recently remodeled or date back to construction from the 1950s. It cannot distinguish between a home with an ocean view and one with the view of a parking lot. While the AVM methodology takes account of the physical distance of a comparable property to the subject property, it draws no distinction based upon local political boundaries and may treat as comparable a home located in a school district substantially inferior to the school district in which the subject property is located. These are important considerations in valuing a residential property.

At Cowan's direction, Phoenix Advisors & Managers USA LLC ("Phoenix"), which maintains tax-assessment data obtained from more than 3,000 counties and parishes in the United States, ran an algorithm that identified the sales price for ten comparable properties for each at-issue property, using a metric called the "Mahalanobis distance." The concept of distance, as used in statistics, does not refer to physical distance of locations but to the number of standard deviations between a point and the mean of a distribution of data. The algorithm purports to identify properties comparable to the subject property using the following equally weighted factors: square footage, age, lot size, date of sale, number of bedrooms and number of bathrooms. Based on that metric, Phoenix provided ten comparable properties and their sales prices, which were adjusted for their date of sale and square footage.

Cowan then used the AVM results to recalculate the loans' LTV ratios. For refinance loans, he used the AVM mean in the denominator, in place of the appraised value; for purchase loans, he used the lesser of the AVM mean or the purchase price. (Cowan Direct ¶ 91.) Cowan then reported the Loans with recalculated LTV ratios that exceeded any of the three following thresholds selected by Holt: (i) more than 80% and more than 5 percentage points greater than the LTV ratio stated in the MLS; (ii) more than the maximum LTV ratio permitted by the applicable underwriting guideline for the Loan by at least 5 percentage points; and (iii) more than the maximum LTV ratio represented in the relevant PSA, plus one percentage point.

Cowan agreed that human appraisers have "certain advantages over AVMs," and that appraisers can observe traits that are "not recorded or knowable on a large database." (Tr. 1556.) He agreed that AVMs "cannot know for an individual property the condition a property or unique feature of the property," for example, whether there was a "hole in the roof." (Tr. 1562, 1599).

Cowan testified that with respect to an individual property, as distinguished from a "mass basis," AVMs do not account for localized considerations that might alter the value of a given property, such as flooding risks, school district, proximity to nuisances or a hole in the roof. He testified as follows:

THE COURT. But if you're talking about a house that's in a flood plain or not in a flood plain, that's not necessarily true; or close to a parking lot, for example, it would make a big difference if it's adjacent to the parking lot rather than one house away from the parking lot, no?

A. Oh, I agree, and you are correct. It doesn't take account of that. There isn't a way to do that on a mass basis.

THE COURT. All right. And what about whether it is a property that is situated on a hill or on a flat parcel, is that taken into account?

A. No, sir.

* * * *

THE COURT. And let's assume that in terms of school districts, there is a great advantage in being in Polk County. It may be that it works the exact opposite. And the identity of the school district in which a home is situated has a material impact on the price of the home. How does the modeling take that into account if the modeling does not distinguish between, in developing comparables, a home on one side of a county line and a home on another side of a county line? I don't know if my question is clear enough for you.

A. Oh, that was—no, I understand your question perfectly. It's a good question. So to answer your—I have a two-part answer to your question. First of all, within the span of the modeling, there is no distinction between those two counties. The computer program just says, hey, look, they're right next to each other, although technically they're across the street so they're in different counties. The computer program doesn't know that they're in different counties.

* * * *

MR. MUSOFF.... you would agree that, for example, an appraiser may go to the property to see if there is a hole in the roof, correct?

A. Yes.

Q. An AVM cannot check if there is a hole in the roof, correct?

A. That is correct.

* * * *

A.... I am saying the outcomes of the appraisals are random because I don't know what properties they're looking at, I am just comparing the subject property to 10 comparable properties on similar characteristics, there is bound to be some variation that is just the result of what we were just discussing, which is sometimes there is a hole in the roof, sometimes there is a skylight.

(Tr. 1600-01, 1683-84, 1557, 1709.)

At trial, Cowan testified that his application of the AVM model could not determine the validity or accuracy of any specific appraisal incorporated into the LTV ratio listed in the MLS. Instead, according to Cowan, on average, the value reached through an AVM calculation was more accurate than any individual appraisal:

Q. You're saying you're using statistics because on the whole you think it averaged out, but for any individual property you don't know whether the Phoenix AVM average mean for those 10 comps is more accurate than the appraisal, correct?

A. Well, let me point out that I understood your question and you're correct. I don't know, but the appraisal is based on three comparables that the appraiser chose, hopefully out of the 10 that I chose, but apparently not necessarily.

(Tr. 1609.)

The limited utility of an AVM for valuing a particular residential property is illustrated when the AVM calculated by Cowan is compared to the sales price in a transaction. Cowan stated that the purchase price in such an arm's length transaction is the "gold standard" for determining fair market value.

Q. Turning back to a comparison between the average mean of the 10 comparables that you generated through this Mahalanobis distance metric and sales price, you agree with me a sales price is the gold standard for market value, correct?

A. I would.

* * *

Q. Many of the actual sales prices are considerably different from the Phoenix AVM estimate that you generated using the average mean of the 10 comparables, correct?

A. Yes, sir. You established earlier that there was a wide range.

(Tr. 1604.) Yet Cowan still used the mean AVM value to recalculate the LTV and identify an alleged material defect. For example, Loan 1381590 was appraised at $495,000, with a purchase price of $478,228, and the original LTV of 80% calculated using the actual purchase price. (Pl. Appendix 40.) The Phoenix AVM generated ten comparable properties with a range of $282,322 to $422,999, such that the original appraisal fell outside of the ten comparables, with a mean of $351,676. (Id. ) Cowan and the Trusts used the $351,676 mean generated by the AVM—a sum approximately $127,000 less than the actual purchase price—to recalculate a LTV of 108.8% and find a breach of the MLS Warranty. (Id. )

As another example, Loan 124546076 was originally appraised at $605,000 and purchased at origination for $603,308. (Pl. Appendix 40.) The purchase price was used to calculate the original LTV of 80 percent. (Id. ) The Phoenix AVM generated ten comparables valued between $231,308 and $531,921, such that the original appraisal was above the range of comparables, with a mean of $398,353. (Id. ) Cowan and the Trusts used the AVM mean to recalculate an LTV ratio of 121 percent. (Id. )

The Phoenix AVM sometimes assigned a value that was far from the actual sales price, but Cowan nevertheless relied on the AVM. For example, Loan 1492376 had a purchase price of $1.7 million, but the Phoenix AVM predicted a value of $980,000. (DX NU; Tr. 1606.) As another example, Loan 10749375 had a purchase price of $710,000, but the Phoenix AVM predicted that its market value should be approximately $485,000. (DX NU; Tr. 1608.)

The Court finds that the Trusts have failed in their burden of proof in proving any breaches of the MLS Warranty as to LTV ratios. The Court rejects Cowan's opinions and the Trusts' evidence as to MLS breaches concerning LTV ratios because they rely exclusively on AVMs to prove that an LTV ratio listed on the MLS for a specific residential property is not true and correct The Trusts offered no analysis of the actual appraisal conducted at the time of origination and no evidence that it was conducted dishonestly, negligently or not in conformity with industry standards. The Trusts offered no conventional appraisals as a cross-check to establish the reliability of the AVMs.

The Trusts have failed to prove that any LTV ratio listed on the MLS was not true or correct as of the date of origination.

5. Documentation Style.

The MLS must include "a code indicating the documentation style of the Mortgage Loan ...." (PX 49 at 0045.) This information must be true and correct as of the Closing Date for the applicable PSA. (See, e.g. , PX 49 at 0191.) For example, as to the 2006-OA2 Trust, Column BB of the MLS is headed "Documentation," and describes the relevant loan with phrases like "Stated Income / Verified Assets (Reduced)," "Full / Alternate," and "No Doc." (See PX O01.) The Trusts contend that UBS breached the MLS Warranty as to documentation style for 38 loans. (Pl. FF ¶ 228.)

While the word "style" may imply a category of information lacking in real substance, the name understates the importance of the category. For example, a "stated income" loan may require verification of the fact of employment but not verification of the borrower's actual income. According to one witness, "stated income" loans were known pejoratively in the mortgage industry as "liar loans" because they had "the inherent risk of borrower misrepresentation regarding income and/or assets." (Lantz Direct ¶ 14.) In contrast, a "full documentation" loan may require verification of several categories of information. The accuracy of these descriptions is important because an individual deciding whether to invest in a pool of loans may view the loan's "document style" as a significant consideration.

The Trusts will have established a breach of the MLS Warranty if they can prove that it is more likely than not that the listed "documentation style" was not "true and correct in all material respects" as of the Closing Date. The Trusts must also prove that "such defect materially and adversely affects the interests of the Certificateholders and the Certificate Insurer in the related Mortgage Loan." (PX 49 at 074.)

6. Credit Score.

a. Construction.

The MLS includes "the credit score (or mortgage score) of the Mortgagor ...." (PX 49 at 0045.) The PSA defines the "Mortgagor" as "[e]ach obligor on a Mortgage Note." (PX 49 at 0046.) The Mortgagor's credit score must be true and correct as of the Closing Date for the applicable PSA. (See, e.g. , PX 49 at 0191.) For the 2006-OA2 Trust, Column BY of the MLS is headed "FICO Score," and lists the FICO score of each borrower. (See PX O01.) The Trusts contend that UBS breached the MLS Warranty as to credit scores for 41 loans in the 2007-3 Trust. (Pl. FF ¶ 238.)The Trusts will have established a breach of the MLS Warranty if they can prove that it is more likely than not that a listed credit score was not "true and correct in all material respects" as of the Closing Date. The Trusts must also prove that "such defect materially and adversely affects the interests of the Certificateholders and the Certificate Insurer in the related Mortgage Loan." (PX 49 at 0074.)

b. Application.

In contrast to other claimed breaches, the Trusts do not rely exclusively on expert evidence to prove that the MLS failed to accurately list borrower credit scores. At trial, the Trusts called Adrian Wu and endeavored to prove through him that UBS failed to accurately list the credit scores for 41 loans contained in the 2007-3 Trust.

Wu was employed by UBS from 2004 through 2007. (Tr. 183.) He described his job duties as that of "a structurer," and testified that he worked on the securitizations of all three Trusts. (Tr. 177, 183.) He testified that he developed structures for securitizations underwritten by UBS, meaning that he developed different tranches of mortgage loans for the pools offered by UBS. (Tr. 183-185.) Wu also testified that in his capacity as a trader, he priced and sold RMBS securitizations to investors. (Tr. 186.)

The 2007-03 Closing Date was May 15, 2007, which is the date as of which the MLS Warranty as to credit scores is made. Shortly before then, on April 26, 2007, Wu received an e-mail stating that the FICO credit scores for "[a]round 40" borrowers in the 2007-3 pools had dropped below 600 points. (PX 208 & attachment.) Wu testified that the e-mail arrived before the transaction closed. (Tr. 217.) The e-mail attached a spreadsheet detailing 300 borrowers who UBS had reviewed for changes to their FICO scores, including 41 whose revised scores fell below 600. (PX 208.) Wu testified that there was a "general understanding" that a change to borrower FICO scores would negatively affect rating agencies' evaluations of the Trust. (Tr. 206.) He also testified that "significant" declines in FICO scores indicated an increased likelihood of borrower default. (Tr. 215.)

The Trusts proved that the reassessed, lower FICO scores of April 26, 2007 were known to UBS but not included in the Mortgage Loan Schedule attached at Schedule I to the 2007-3 Trust PSA. (See PX 8 (2007-3 Trust Mortgage Loan Schedule); PX 208 attachment (FICO reassessments).) For instance, as to the borrower for Loan Number 6058625, the MLS reflected a FICO score of 699. (PX 8, line 2012, column CD.) However, this borrower's FICO score had actually been revised downward to 507, a full 192 points lower than what was recorded in the Mortgage Loan Schedule. (PX 28, line 186, column C.)

The Trusts have proved that the 2007-3 Trust MLS did not accurately reflect the FICO scores of these 41 loans, and that UBS therefore breached the MLS Warranty as to those loans. The Court further finds that the breaches materially and adversely affected the interests of the Certificateholders at the time of discovery or breach. As Holt testified, a lower FICO score is associated with a greater risk of loss on the loan. (Holt Direct ¶ 76.) At trial, Holt opined that FICO scores are usually assessed within 40-point ranges, such that a five-point change in FICO score may not be material, whereas a larger drop would likely alter the loan terms available to a borrower, e.g. the rate of interest charged for the loan. (Tr. 1126-27.)

The Trusts have proved that a MLS Warranty breaches as to 41 loans containing FICO scores that were 40 points or more lower than reflected in the MLS for the 2007-3 Trusts. The Court further finds that these breaches had a material and adverse effect on the interests of the Certificateholders because they increased the risk of loss for the underlying loans.

7. DTI Ratio.

a. Construction.

The PSAs required UBS to list on the MLS "the debt-to-income ratio of the Mortgage Loan ...." (PX 49 at 0045.) The debt-to-income ratio, or "DTI," reflects the percentage of a borrower's monthly debt obligations as to the borrower's overall monthly income. For example, a borrower earning $10,000 per month with monthly debt obligations of $2,000 would have a debt-to-income ratio of 20 percent. As Holt explains, as an individual's DTI ratio increases, he or she is left with less disposable income, whereas a borrower with a low DTI ratio has a greater ability to make payments on a loan and presents an overall lower risk to the lender. (Holt Direct ¶ 73.) The Trusts contend that UBS breached the MLS Warranty as to the stated DTI ratio for 4,753 loans. (Pl. FF ¶ 228.)

UBS points out that it was required to list on the MLS the DTI ratio "of the Mortgage Loan," and not of the Mortgagor. (Def. FF ¶ 171.) Based on this language, it argues that a failure to correctly list the DTI ratio of the borrower would not be a breach.

As the Court previously noted, "Mortgage Loans" is defined as "the mortgage loans and cooperative loans transferred and assigned to the Trustee," and specifically "the mortgage loans so held being identified in the Mortgage Loan Schedule ...." (PX 49 at 0046.) By contrast, the Mortgagor is defined as "[e]ach obligor on a Mortgage Note." (PX 59 at 0046.) The PSAs do not use the terms "Mortgagor" and "Mortgage Loans" interchangeably. To use two above-discussed examples, the MLS is required to include "the credit score (or mortgage score) of the Mortgagor ," as well as "the documentation style of the Mortgage Loan ...." UBS also accurately observes that in their submissions discussing the DTI ratio, the Trusts often use the words "mortgagor" and "mortgage loan" interchangeably. (See, e.g. , Tr. 869 (question from plaintiffs' counsel referring to "DTI of the mortgagor.")

As used in the MLS definition, however, the phrase "of the Mortgage Loan" can only be understood to mean the DTI ratio related to the mortgage loan, which, in this case, necessarily makes reference to the borrower's DTI. The word "of" is defined as "relating to; with reference to; as regards." Webster's Third New International Dictionary Unabridged (2002) at 1565. It is also "used as a function word to indicate the place or thing from which anything moves, comes, goes or is directed or impelled." Id. The DTI ratio "relating to" a Mortgage Loan is necessarily the DTI ratio of its borrower, given that a Mortgage Loan cannot, itself, have a DTI ratio.

UBS does not propose an alternative construction for the phrase "debt-to-income ratio of the Mortgage Loan," and merely argues that the Trusts should not conflate the DTI ratio of the Mortgagor with that of the Mortgage Loan. (Def. FF ¶ 171.) It would have been a simple matter for UBS to demonstrate that the parties had some other understanding of the DTI ratio on the MLS and that UBS consistently applied that understanding to the MLS. For example, if it thought that it was required to only include in the "debt" component the actual mortgage loan amount, one would expect to see that consistently used throughout the MLS. UBS offers no such evidence. The Court infers that UBS has not proffered an alternative construction because it would expose UBS to a claim that the MLS did not uniformly conform to that alternative construction and, therefore, UBS is in breach under the alternative construction.

Because a loan, itself, cannot have a DTI ratio but a borrower can have such a ratio, the most reasonable construction is that the phrase "the debt-to-income ratio of the Mortgage Loan ...." means the debt-to-income ratio relating to the mortgage loan which is the borrower's debt-to-income ratio, calculated using all of the borrower's monthly debt in relation to all of the borrower's monthly income.

Separately, Holt's direct testimony identifies and defines two categories of DTI ratio. (Holt Direct ¶ 197.) He defines "front-end ratio" as "the percentage of a person's stable monthly income that is paid toward monthly mortgage payments" and the "back-end ratio" as "the percentage of a person's stable monthly income that is paid toward all monthly debts." (Holt Direct p. 83 n.83 & n.84.) Each MLS contains a column for "Front Ratio" and "Back Ratio." (See PX O01 col. BZ & CA; PX O02 col. CE & CF; PX O03 col. CE & CF.)

For the MLS Warranty, Hold identifies a breach if the recalculated front-end ratio exceeds the MLS listing by 2% or if the recalculated back-end ratio exceeds the MLS listing by 1%. (Holt Direct ¶ 197.) Holt identifies a breach of the Guidelines Warranty if the recalculated front-end ratio exceeds the guidelines maximum by 2%, or if the recalculated back-end ratio exceeds the guidelines maximum by 1%. (Holt Direct ¶ 197.) Even where the recalculated DTI ratio does not exceed a guidelines maximum, Holt identifies a defect if the recalculated DTI was within 5% of the guidelines maximum or more than 5% higher than the DTI listed in the MLS. (Holt Direct ¶ 197.)

Holt's opinions as to the individual DTI-related breaches, as set forth in Appendix 1, do not expressly distinguish between the front-end ratio and back-end ratio. When he cites to a guidelines maximum or cites a recalculated DTI ratio, it is unclear for any given loan whether he is referencing the front-end ratio or the back-end ratio. It is nevertheless the case that when a borrower misstates income or has undisclosed mortgage debt, any recalculation of income and debt alters both the front-end and back-end ratios. Thus, while the precise sub-category of Holt's DTI-based breach opinions is not explicit, the Court generally can make findings as to whether a Trust has proved a breach as to inaccurately calculated DTI ratio. Further, the Court is unaware of any instances in which Grissom opines that Holt incorrectly relied on a guidelines maximum for the front-end ratio when he should have been applying the maximum for the back-end ratio, or vice versa.

The Trusts will have established a breach of the MLS Warranty if they can prove that it is more likely than not that the listed DTI ratio, as defined herein, was not "true and correct in all material respects" as of the Closing Date. The Trusts must also prove that "such defect materially and adversely affects the interests of the Certificateholders ... in the related Mortgage Loan."

b. Application.

Through Holt, the Trusts contend that 4,753 loans breached the MLS Warranty because borrowers' DTI ratios exceeded those listed in the MLS. According to Holt, these borrowers had undisclosed debts, including other undisclosed mortgages, that were not reflected in the DTI ratio, incomes that were lower than listed in the DTI ratio, or both.

Holt based his analysis on a variety of databases and sources. Some were publicly available databases and some were contained in the borrowers' loan files. Certain of these sources were admitted into evidence at trial, and others were not admitted but informed Holt's opinions as to whether UBS breached the MLS ratio. The Court now addresses each of these sources.

c. The MERS Database.

For 733 loans, Holt concluded that borrowers' debts were misstated on the basis of one or more mortgages that they did not list on their applications, which Holt uncovered through the Mortgage Electronic Research System ("MERS"). MERS is an electronic database that provides information for registered mortgages, and was available at the time that the loans were originated.

Both Holt and UBS's re-underwriting expert Deborah Grissom testified that MERS is considered to be a reliable source of information. Grissom testified:

Q. And it's a reliable source of information about registered loans, correct?

A. Yes, it reflects loans that have been recorded in county recorders' offices.

Q. Have you ever accessed MERS?

A. In another project for another reason. It wasn't reunderwriting, but yes.

Q. And it's relatively straightforward to use?

A. My understanding is, yes, it is.

(Tr. 460.) Holt testified that he often relies on MERS in the course of his work as an underwriter and re-underwriter, and described MERS data a "reliable" and "easily available." (Tr. 1185.) He testified that the MERS database is available to the general public and are generally relied upon by the members of his profession in the course of doing business. (Tr. 1181.) It is the type of material that an expert in underwriting relies upon in the ordinary course of his work as an underwriter and thus may form the basis for the expert witness's opinion.

Further, the Court finds that MERS data is admissible for the truth of its content under the hearsay exception of Rule 803(17), Fed. R. Civ. P. ("The following are not excluded by the rule against hearsay, regardless of whether the declarant is available as a witness: ... Market quotations, lists, directories, or other compilations that are generally relied on by the public or by persons in particular occupations.").

d. Other Databases.

For 777 loans, Holt relied on commercial data sources including Accurint, Data Verify, Data Tree, Sitex and Lexis-Nexis.

Grissom testified that Accurint is commonly used among underwriting professionals and was in use at the time that the loans in this case were originated. (Tr. 548, 550.) Holt testified that he used the Accurint database to form his opinion as to the existence of any breach. (Holt Dec. ¶ 225 n.102.) The Court finds that Accurint was a reliable source for Holt's opinions, and that evidence from the Accurint database is admissible under the hearsay exception of Rule 803(17).

Grissom testified that Data Verify was an online data verification and fraud prevention database that was available at the time that the loans originated. (Tr. 395.) Holt testified that he was unfamiliar with Data Verify: "Q. What is Data Verify? You're not familiar with that? A. I am not familiar with that, no, sir. We have used it, I have seen it in the past. Again a lot of these sources, a lot of these firms and third-party sources provide information from public records." (Tr. 1013.) Holt's answer contradicted itself: he testified both that he was not familiar with it, but that "[w]e have used it" before stating that it provides information drawn from public records. (Id. ) While Holt's testimony is muddled, the Court concludes that the testimony of both Holt and Grissom are sufficient to find that Data Verify is a reliable source for Holt's opinions, and that evidence from Data Verify is admissible under the hearsay exception of Rule 803(17). Alternatively, it is the type of material that an expert in underwriting relies upon in the ordinary course of his work as an underwriter and thus may form the basis for the expert witness's opinion.

Holt identifies Data Tree as a database that includes information on properties by owner and transaction history. (Holt Direct ¶ 225 n.102.) Holt testified that he and other underwriters and re-underwriters commonly used the database as a source of information. (Tr. 1185) (in the trial transcript, Data Tree appears to have been mis-transcribed as "Data Trade.") Grissom testified that she had no independent knowledge of Data Tree. (Tr. 395.) The Court finds that Data Tree was a reliable source for Holt's opinions, and that evidence from the Data Tree database is admissible under the hearsay exception of Rule 803(17).

Holt testified that Sitex is publicly available and commonly used by underwriters and re-underwriters. (Tr. 1180-81, 1185.) Grissom testified that Sitex was available to underwriters at the time that the loans were originated. (Tr. 567-68.) The Court finds that Sitex was a reliable source for Holt's opinions, and that evidence from the Sitex database is admissible under the hearsay exception of Rule 803(17). Alternatively, it is the type of material that an expert in underwriting relies upon in the ordinary course of his work as an underwriter and thus may form the basis for the expert witness's opinion.

Grissom testified that Lexis-Nexis "provides a lot of information" to help identify fraud, and is commonly used by underwriters to identify fraud. (Tr. 394.) She testified that Lexis-Nexis was available to underwriters at the time that the loans originated. (Tr. 567, 577.) Holt also testified that Lexis-Nexis is publicly available and commonly used by underwriters and re-underwriters to access personal information about borrowers. (Tr. 918, 1181, 1185.) The Court finds that Lexis-Nexis was a reliable source for Holt's opinions, and that evidence from the Lexis-Nexis database is admissible under the hearsay exception of Rule 803(17).

e. Multiple Credit Inquiries.

Holt identified 374 loans for which the loan files contained evidence of multiple credit inquires. Grissom testified that "credit inquires could have been a red flag" as to the existence of undisclosed debt held by a borrower. (Tr. 487, 527-28.) She testified: "Certainly credit inquiries, particularly of a mortgage due, produce a red flag for the underwriter. And it should be investigated." (Tr. 484.) Holt testified that multiple credit inquiries are a "red flag" that indicate the presence of additional debt and potential misrepresentation of income. (Tr. 1045.) According to the Trusts, the presence of multiple credit inquiries make it "especially likely" that DTI ratios for the related loans were misstated in the MLS. (Pl. FF ¶ 275.)

The Court finds that multiple credit inquiries are a red flag that would cause a prudent underwriter to engage in further inquiry. Standing alone, it provides an insufficient basis for a finding that a borrower had undisclosed debt or that the DTI ratio was misstated. It is, however, data that an expert may take into account along with the totality of information in forming an opinion on whether the MLS accurately listed a borrower's DTI ratio.

f. Credit Reports.

Holt identified 373 loans for which he concluded that the borrowers' debts were misstated based on his analysis of credit reports. These credit reports and associated materials are contained in the loan file pertaining to the borrower. As discussed below, the loan files, including the credit reports contained therein, are received into evidence not for the truth of their contents, but as evidence of the underwriter' knowledge and actions and inactions, including whether the underwriter complied with the Originator's underwriting guidelines.

The Court concludes that it was appropriate for Holt to consider the content of a credit report in a loan file as part of the mix of information used in forming an opinion on whether the MLS accurately listed a borrower's DTI ratio.

g. Evidence of Multiple Misrepresentations as Evidence that a Borrower Also Misrepresented Debt.

Holt based his opinion that UBS breached the MLS Warranty relating to the DTI ratio as to an additional 1,675 loans based upon his conclusion that the borrower misrepresented his or her debt and made an additional misrepresentation concerning occupancy, employment or income. "Such evidence that a borrower made multiple misrepresentations makes it more likely that he misrepresented his debts," the Trusts argue. (Pl. FF ¶ 277.) The Trusts cite to two exemplar loans in which it contends that a borrower's misrepresentation of debt was made alongside additional misrepresentations. (Id. )

Proof that a borrower made an intentional misrepresentation as to one item of information may form part of the expert's opinion, to the extent that such an opinion is relevant or material, that a different misstatement by the same borrower was intentional. It is reasonable to infer that a borrower who misrepresents income, occupancy or employment would also be inclined to make additional misrepresentations as to debt. Such misrepresentations may be considered in the total mix of information that informs an opinion as to whether the borrower misrepresented his or her debts.

h. Social Security Administration Data and Pension Custodian Statements.

Holt identified 10 loans for which he concluded that borrower income was misstated based on his review of Social Security Administration ("SSA") benefit statements and/or retirement pension income statements issued by pension custodians. (Pl. FF ¶¶ 279-80.) Federal Rule of Evidence 803(8)(B) recognizes a hearsay exception for "[a] record or statement of a public office if ... the opponent does not show that the source of information or other circumstances indicate a lack of trustworthiness." There is no indication that that the SSA data or pension statements lack trustworthiness. They therefore are received into evidence under the hearsay exception of Rule 803(B).

It was appropriate for Holt to rely on these material in forming an opinion as to whether the income figure used in the DTI ratio was accurately listed in the MLS.

i. Employer Income Statements, Including W-2 Forms.

For 282 loans, Holt concluded that borrower income was misstated based on information contained in their W-2 forms and paystubs. These items were included in the borrower loan files. The Trusts contend that these materials are admissible under the business records exception of Federal Rule of Evidence 803(6), but have laid no business record foundation for these materials. While they are not admissible evidence, they nevertheless were appropriately considered by Holt, as an expert, as the type of hearsay that an expert in the field of underwriting would use in the ordinary course of their work in forming opinions. Thus, they may form a part of the basis for Holt's opinions whether DTI ratio data was accurately listed in the MLS.j. BLS Data.

For 72 loans, Holt concluded that borrower income was misstated based on his review of information regarding salaries that was included on the website of the Bureau of Labor Statistics ("BLS"). As described in Holt's direct testimony:

The BLS website is free, and has been available to and used by underwriters since 1997. It contains historical statistics provided by the United States Department of Labor on the range of incomes for a wide and detailed list of professions and locations, by year, and provides information on the distribution of different salary ranges based on job title, location, and year.

(Holt Direct ¶ 152.) Holt testified that he used the 90th percentile of occupation salary listed on BLS, and then multiplied that figure by 125 percent to use a higher-baseline salary. (Tr. 1016.) Holt concluded that the listed DTI ratio was inaccurate only if it exceeded the grossed-up BLS figure by 10 percent or more. (Tr. 2039.) Holt explained his use of BLS data as follows:

As an example of how this information was used, for stated income loans, for salaried borrowers only, I assessed a finding for Stated Income Not Reasonable if the stated income was above the 90th percentile provided by the BLS, grossed up (i.e., multiplied by) 125%, for the borrower's occupation, position, and geographic location, as reflected on the original verification (verbal or written VOE) or stated on the final 1003, and retroactive to the year of income tested. I believe this to be a very conservative approach because it gives the Originator the benefit of an assumption that all loan applicants earn at the 90th percentile of their stated occupation, which is, of course, not the case. However, to be even more conservative, when using BLS data to estimate a reasonable income for the borrower, I based my estimate of the borrower's reasonable income on 125% of the highest available income within the range up to the 90th percentile income.

(Holt Direct ¶ 153.)

Grissom testified that BLS data was available to underwriters and re-underwriters, and was a source "that we use because it's very common." (Tr. 387-88.) Grissom also testified that BLS data "suffers from serious flaws and can be readily misused." (Grissom Dec. ¶ 74.) On cross-examination, she testified that the underwriter makes the judgment call as to whether to cross-check stated income against BLS data: "An underwriter always has to make that determination. If they are unfamiliar with how much a mechanic makes, it certainly was in their ability to go out and check that, or look at something, call a friend, whatever they do to convince themselves that that [income] is reasonable." (Tr. 388.)

Based on the testimony of Grissom and Holt, the Court concludes that BLS data was widely used by underwriters and re-underwriters at the time that the loans were originated. The Court finds that BLS data is admissible under the hearsay exception to Rule 803(17). Alternatively, the Court concludes that BLS data was properly considered by Holt in forming his opinions as to whether DTI ratio was misstated in the MLS because it was the type of material that an expert in underwriting relies upon in the ordinary course of his work as an underwriter.

k. Other Salary Databases.

Holt concluded that there were four instances where borrowers misstated income based on information at salary.com, which lists salary information including bonuses, commissions and overtime. (Tr. 1016.) Grissom testified that a reasonable underwriter would consult salary.com to consider income information. (Tr. 433.) Based on the testimony of Grissom and Holt, the Court concludes that salary.com data was widely used by underwriters and re-underwriters at the time that the loans were originated. The Court finds that salary.com data is admissible under the hearsay exception to Rule 803(17). Alternatively, it is the type of material that an expert in underwriting relies upon in the ordinary course of his work as an underwriter. It was appropriate for Holt to rely on this data in forming an opinion as to whether the DTI ratio was accurately listed in the MLS.

For 37 loans, Holt concluded that borrower income was misstated based on information listed at the website Work Number. Grissom testified that Work Number is a service that underwriters access to determine whether a borrower's information was accurate, describing it as "a rather automated service, yes." (Tr. 463.) Based on the testimony of Grissom and Holt, the Court concludes that Work Number data was widely used by underwriters and reunderwriters at the time that the loans were originated. The Court finds that Work Number data is admissible under the hearsay exception to Rule 803(17). Alternatively, it is the type of material that an expert in underwriting relies upon in the ordinary course of his work as an underwriter and thus may form the basis for the expert witness's opinion. The Court concludes that it was appropriate for Holt to rely on this data in forming an opinion as to whether the DTI ratio was accurately listed in the MLS.

l. Borrower Tax Returns.

For 543 loans, Holt concluded that borrower income was misstated based on his (or his vendor's review) of borrower tax returns or Form 1099s that were contained in the borrower loan files. Holt testified that in calculating borrower income, he would rely on "the borrower's tax return for the relevant year" when a tax return was included in the loan file. (Holt Direct ¶ 154.) Grissom testified that she relied on tax return information, "if there is, you know, W-2 or tax returns in the file." (Tr. 729.) The Court concludes that it was appropriate for Holt to rely on these documents in forming an opinion as to whether the DTI ratio was accurately listed in the MLS.

m. Borrower Bankruptcy Filings.

For 554 loans, Holt concluded that borrowers misstated their incomes based on his analysis of borrower bankruptcy filings, which he found through searches of PACER and Lexis-Nexis. The bankruptcy filings included statements by borrowers concerning their incomes. Holt concluded that the bankruptcy filings were a more reliable source of borrower's income than the stated-income figures that the borrower included in their applications to the Originators.

UBS argues that bankruptcy filings are unreliable sources of truthful information concerning income. In bankruptcy proceedings, UBS argues, debtors have an incentive to understate their incomes. They cite to a report by the U.S. Department of Justice that concluded that in 23% of individual bankruptcy cases, debtors understate their incomes, assets or pre-petition property transfers. (Def. FF ¶ 192.) At trial, Holt testified that he assumed the truth of a borrower's statement in a bankruptcy filing, even when they were contradicted by other sources that he considered reliable.

Q. In other words, you were assuming that the borrower's lying on the loan application but telling the truth in the bankruptcy filing. Is that correct?

A. Yes, sir, that's correct.

(Tr. 872.)

Holt testified that he credited the truth of a bankruptcy filing even when it was refuted by a borrower's W-2 statement contained in the same loan file. (Tr. 906-08.) When questioned by the Court, Holt confirmed that he credited statements in bankruptcy filings as true, even if there was no additional corroborating data:

THE COURT: And if the bankruptcy filing showed a lower income than that on the W-2 in the loan file, would that cause you to credit the bankruptcy filing, without more?

THE WITNESS: To credit it—we would accept that.

THE COURT: To accept the bankruptcy?

THE WITNESS: Yes.

THE COURT: So it wasn't dependent on a red flag in the loan file, unless you consider the fact that the borrower filed bankruptcy after origination—there was no other red flag—you would still credit the bankruptcy filing, correct?

THE WITNESS: If there was no other red flags?

THE COURT: Yes.

THE WITNESS: Yes, sir

(Tr. 907-08.)

In some instances, Holt relies on a borrower's W-2 form as evidence of income, but in others, he ignores the W-2 form and credits the borrower's statements made in bankruptcy filings. For example, as to Loan 138058233, Holt opined that the W-2 contained in the loan file was fraudulent, and he credited the borrower's statement of income made in bankruptcy filings. (See Holt Direct Appx. 1 & Tr. 882-83.) As discussed in greater detail below, Holt's conclusion that the W-2 statement was fraudulent was based on insubstantial reasons, and it was unreasonable in that instance to credit the bankruptcy filing over the W-2 statement.

As part of the total mix of information, a borrower's statement in a bankruptcy proceeding may be relevant and persuasive in forming an opinion that, for example, a borrower misstated income on a loan application. A borrower has an incentive to overstate income on a "stated income" loan application but there is also an incentive to understate income in a bankruptcy petition. Forming an opinion as to the truth of one or the other may require consideration of any W-2s, 1099s, tax returns in the loan files and databases such as the BLS database in forming an opinion. But the Court rejects the notion advanced by Holt that a statement in a bankruptcy proceeding trumps all other available data and per se proves a misstatement.

The Court finds that Holt did not use a consistent or reliable methodology in determining the weight to be applied to data found in a bankruptcy proceeding. Because Holt did not evaluate the total mix of information and gave controlling or near-controlling weight to statements made in the course of a bankruptcy proceeding, the Court does not credit Holt's opinions as to DTI ratio breaches that were based solely on a borrower's statement in a bankruptcy proceeding.

However, in some circumstances, statements made in bankruptcy proceedings were consistent with additional relevant information, such as servicing records or tax-related documents. In these instances, bankruptcy submissions are properly considered as part of the total mix of information that goes toward a borrower's income.

n. Borrower Hardship Letters.

For 30 loans, Holt relied on borrower "hardship letters" that were submitted by borrowers and included in loss-mitigation servicing records. (Pl. FF ¶ 294.) These letters postdate the loans' origination. (Pl. FF ¶ 294.) To the extent that these letters reflect borrower income prior to the Closing Date of the relevant PSA, it was appropriate for Holt to rely on their contents to identify a breach of the MLS Warranty. The Court recognizes that information post-Closing could be some evidence relating to the pre-Closing period depending upon the nature of the employment (self-employed, variable hours or salaried position) and the temporal proximity to the Closing. But if a reasonable reading of the letter indicates that it discloses only the borrower's income after the Closing Date, and there is no reason to conclude that this subsequent income reflects on the borrower's income as of the Closing Date, then, without more, a breach of the MLS Warranty relating to the DTI ratio may not be proved based on the contents of the letter.

o. Red Flags.

Holt relied on a variety of "red flags" for identifying inaccuracies in the DTI ratios listed in the MLS. These included "relatively low liquid assets"; "low bank deposits or no deposits at all;" unsourced bank deposits; "payment shock," which he describes as a risk that a borrower's payments will greatly increase as a result of a change in a mortgage loan's interest rate; prior bankruptcy; low credit limit; high credit utilization; and changed income in applications. (Pl. FF ¶¶ 295-303.) He testified that he did not use these red flags alone as a basis for an opinion that the DTI ratio was in breach, but rather, treated them as factors that increased the likelihood that DTI ratio was misstated in an MLS.

While Holt's testimony largely discusses the presence of red flags in the context of best practices for applying the underwriting guidelines, the Court concludes that red flags may be considered in the total mix of information on which an opinion is formed as to whether a DTI ratio in the MLS is incorrect. Standing alone, red flags are not sufficient to prove that a borrower falsely stated DTI, but viewed alongside additional evidence of a breach, they strengthen an opinion that DTI ratio was misstated.

The Court therefore concludes that Holt properly considered these red flags as part of the total mix of information in forming an opinion whether the MLS misstated a borrower's DTI ratio.

THE GUIDELINES WARRANTY.

A. Overview of the Guidelines Warranty and Its Relationship to the MLS Warranty.

In each PSA, UBS warranted that, as of the Closing Date, "[t]he Mortgage Loan was underwritten in accordance with the underwriting guidelines of the related Loan Seller [i.e. , Originator] in effect at the time of origination with exceptions thereto exercised in a reasonable manner." (PX 49 at 0196.)

Underwriting guidelines are a standard methodology utilized by an Originator to determine whether to approve and fund a loan to a particular borrower. Management of an originating institution quite reasonably would want uniformity of standards among personnel and locations. The guidelines are unique to each Originator, although, as one might expect, there are often similarities from Originator to Originator. The guidelines used by American Home, Countrywide and IndyMac were received into evidence at trial.

The MLS Warranty and the Guidelines Warranty are fundamentally different even though they address many of the same criteria, such as DTI ratio and whether a property is or will be owner occupied. UBS warranted that the statements in the MLS are "true and correct in all material respects," and that warranty does not excuse a materially incorrect statement because it was derived from compliance with a standard or set of procedures that permitted the incorrect statement to go undetected. But, with respect to the guidelines, UBS warranted that the loan was underwritten in compliance with the Originator's guidelines, and it is the guidelines that determine whether it was appropriate for the underwriter to rely on a borrower's statement.

The guidelines present the underwriter with a series of steps and procedures to determine whether a loan should be approved and funded. It admits the possibility that borrower-supplied information may be wrong, but contains certain steps and procedures to be taken to reduce the possibility of that occurrence. Holt agreed that "one of the goals of reunderwriting is to step into the shoes of the original underwriter who made the loan," and that re-underwriters "try to go back and look at the loan files [as] if you were taking the application right from the borrower across the desk and reviewing the file." (Tr. 864-65.)

Many loans for which the Trusts have claimed a breach of the Guidelines Warranty are also claimed to have breached the MLS Warranty. But a breach of one will not necessarily result in the breach of the other.

The DTI ratio provides a good example, because the DTI ratio is listed in the MLS and also used in applying the guidelines. First, the two warranties speak to different points in time. With the exception of LTV, the MLS Warranty speaks as of the Closing Date of the PSAs, and the Guidelines Warranty speaks to an earlier point in time, i.e. , the time of origination and funding of the loan to the individual borrower. But, more importantly, the nature of the warranty is very different. The MLS Warranty is breached by showing that debt or income used to calculate the DTI ratio was incorrectly stated. But a breach of the Guidelines Warranty is proved by showing that the loan was originated not "in accordance with" the guidelines.

To use specifics, an Originator's underwriting guidelines for "stated income" loans typically do not require income verification of a borrower unless the "stated income" appears unreasonable. A borrower's "stated income" could be materially understated but not bear the indicia of unreasonableness. In that circumstance, there would be no breach of the Guidelines Warranty, even though there is a proven breach of the MLS Warranty.

To prove a breach of the Guidelines Warranty, the Trusts must prove that it is more likely than not that the loan was originated without compliance with the Originator's underwriting guidelines, unless an exception to the guidelines was actually exercised in a reasonable manner at the time of origination.

B. Automated Underwriting Systems.

The application of the guidelines to a particular borrower's loan application was performed either manually, or through an "automated underwriting system," or "AUS." Virtually all loans presently at issue were underwritten utilizing an AUS, except for a subset of loans originated by American Home that were manually underwritten. (See Tr. 967.) Countrywide had its own proprietary AUS program, called "CLUES," but other Originators also used AUS programs. An AUS may vary from Originator to Originator, and an individual underwriter would personally read and evaluate the output of an AUS. Holt described the AUS approval process as follows:

Once the processor and underwriter collects all the information from the application, income, assets, amount, all the information pertinent to the loan, they would then input it into the system or the automated underwriting system, and then there would be an output of that information once the engines within the automatic system do their work. Then you get a response whether or not the loan was accepted or not and the reasons

why. So it is a way of expediting the underwriting process.

(Tr. 1054.) An AUS would issue recommendations as to whether a loan should be approved, approved with caution contingent on satisfying certain conditions, or whether the data was insufficient for it to make a recommendation. (Tr. 1053-54.) Holt testified that employees of the originating institution were generally responsible for reviewing AUS recommendations, checking the accuracy of data and ensuring that conditions for approval were satisfied. (Tr. 1055-56.)

C. Exceptions to the Guidelines Must Be Documented.

From time to time, the strict or literal application of an underwriting guideline may indicate that a loan application should be denied. But underwriters and originating institutions believed that there could be offsetting factors—known as "compensating factors"—that counseled in favor of approving and funding the loan. Originators and the underwriting community understood that there could be such circumstances, and they were referred to as "exceptions" to the guidelines.

The PSAs contemplated that there could be good and sufficient reason to grant an exception to the guidelines. The PSAs warranted that "[t]he Mortgage Loan was underwritten in accordance with the underwriting guidelines of the related Loan Seller [i.e. Originator] in effect at the time of origination with exceptions thereto exercised in a reasonable manner." (PSA Schedule II at xxx.) An "exception[ ]" to the Guideline had to have been "exercised," meaning actually invoked at the time of origination. This means that there must be some indication in the loan file that an exception was actually considered and granted by the Originator.

As Holt testified: "In certain limited circumstances, underwriting guidelines expressly permit ‘exceptions' from their requirements, provided specific ‘compensating factors' are present that offset the increased risk associated with the exception—e.g. , cash reserves, owner equity, credit history, and property marketability." (Holt Direct ¶ 8.) Holt further testified:

[I]t was (and continues to be) an industry practice to require that each exception be supported by a "compensating factor," or a borrower attribute that offsets the risk presented by the variance from the loan program guideline. As an example, consider a loan program having a DTI maximum limit of 40% where the borrower at the time of origination had a DTI of 45%. If the originator wanted to make this loan, it would need to grant an exception to its guideline requirement. In order to obtain the exception, the borrower would need to exhibit at least one compensating factor—possibly, for example, sufficient savings or a sufficiently low LTV—offsetting the risk of his or her paying a percentage of monthly income on debts greater than allowed by guidelines.

(Holt Direct ¶ 119.)

The Guidelines themselves, which include certain technical manuals used by the underwriters, required the individual underwriter to document guidelines exceptions in writing. See, e.g. , PX U479 at 1 (Countrywide Guidelines requiring that "[t]he Underwriter must also clearly describe any guideline exceptions and rationale for approval."); PX U248 at 1 (IndyMac Guidelines stating that "Loans that require an exception must first be submitted for manual underwriting .... When submitting a Loan that will require an exception, the Seller should so indicate by checking the appropriate box ...."); PX U576 at 1 (American Home Mortgage Guidelines stating "[o]nly exceptions that have been approved in writing ... will be honored").

Grissom testified, however, that in forming the opinion that an exception had been exercised she did not disqualify any loan where there was not written indicia that an exception had been considered and allowed. She opined that if a loan was not guideline-compliant but had been approved, and she was able to identify compensating factors, then an exception must have been considered and allowed, even if there was no record of such allowance:

Q. So the fact that an exception—sorry—the fact that this loan was approved with a DTI breach supports your view that the underwriter did not need to seek formal approval of an exception for DTI?

A. That's one of the reasons. And I also—if you read back to my report or my declaration that was submitted, I feel like exceptions are one of those things that are a core underwriting document, and if they were required by the guidelines, that it's not something that would have been overlooked. I mean it's in the same category as a loan approval, as an appraisal, as a credit report. It's those things that are necessary to reunderwrite a loan.

So, the fact that ten years later if it was required by the guidelines and it's not there, the loan funded. You know, the underwriter was—you know, it's something that—and that might not be the best example for this particular loan, but in general we are talking about exceptions, and exceptions are something that they wouldn't have funded it without that.

(Tr. 642-43.)

Grissom also testified that where an approved loan contained a strong indication that a relevant qualifying factor like LTV fell outside of the guidelines, she concluded that "the exception was granted because the loan was made ...." (Tr. 331.) In such circumstances, she instructed the individuals working at her direction "to make the decision whether or not that loan made sense at the time of origination." (Tr. 331.) She testified that if the loan made sense, "it was clear the loan—the exception was granted because the loan was made, and there's documentation in the file that shows, you know, the LTV was granted, approved above the guideline level." (Tr. 331.) Thus, under Grissom's view, a loan falling outside the guidelines must have had an approved exception because it could not have otherwise been funded.

The Court rejects Grissom's circular reasoning and finds that it is inconsistent with the language of the PSAs as informed by the Originator's guidelines. Grissom's reasoning that the exception was granted because the loan was made begins with a loan's approval and works backward in search of a justification. Absent written indicia that an exception was exercised, there is no basis to determine whether a loan was approved in violation of the guidelines or consistent with the guidelines. Similarly, without documentation of the exception, there is no way to determine whether the exception was exercised "in a reasonable manner," as promised in the Guidelines Warranty.

Based on the Court's review of certain individual loan files, the written exercise of an exception required the underwriter to expressly identify the basis for the exception, and attached certain conditions to the exception. (See, e.g. , PX L13607 at 0223-25.) In this example, the written exception named the borrower, the individual underwriter requesting the exception and listed key data concerning the borrower (i.e. , FICO score, monthly income, employer name, et cetera ). (Id. at 0223.) The exception identified any "Rules Violations" with "Comments" that explained the basis for granting the exception; the document also contained a checklist of compensating factors. (Id. at 0223–24.) The exception stated that the document was "not a loan approval," and that "This exception will be null and void if: All Exceptions to program guidelines are not disclosed." (Id. at 0224.) While this document is just one example of a written exception, and the requirements and format may vary across loan types and originators, it illustrates that the grant of an exception adhered to certain formalities and conditions, and was not, as Grissom opines, an action that can be inferred based on a loan's eventual approval.

The Court finds that it would be unreasonable for an underwriter to exercise an exception without providing some record of the exception's existence and the basis for exercising it. This is because the Originators intended to sell the loans. The Originators knew that prudent bidders and buyers would subject the loans and the loan files to due diligence and that it was important to demonstrate to an institution buying the loan that while the strict letter of the guidelines was not followed, there were good and sufficient reasons at the time of origination for making an exception to the guidelines. The Court finds that an unexpressed and undocumented guideline exception is not an "exception ... exercised in a reasonable manner."

The Court finds that in order for an exception to guidelines to be "exercised in a reasonable manner," there must be evidence of a contemporaneous expression of an intent to exercise an exception and documentation that an exception was exercised.

D. The Loan Files Are Provisionally Received Into Evidence.

The Trusts have offered 14,402 exhibits, described as loan files, into evidence in order for the Court "to determine whether statements in the Loan Files are inconsistent with the Guidelines and therefore the Guideline Warranty was breached." (Docket # 434 at 3.) The Trusts further urge that the loan files "are also admissible for the non-hearsay purpose of determining whether there was a valid basis for the expert re-underwriting testimony in this case." (Docket # 434 at 4.) The Trusts do not offer the loan files for the truth of their content. UBS objects to their admission for any purpose.

The phrase "loan files" is not a defined term in the PSAs. Holt defines the term as follows: "loan files are created and maintained by originators whenever a borrower applies for and is issued a loan. They should contain all of the documentation that the borrower submits in support of his or her application, and any documents gathered or generated by the originator during the underwriting process that are necessary to the underwriter's decisions." (Holt Direct ¶ 4.) He also testified: "As described above, a loan file should contain all documents necessary to establish that the original underwriter obtained and analyzed the information as required by the Applicable Underwriting Guidelines." (Holt Direct ¶ 169.) "The loan files are the originator's proof that the loans were originated in accordance with guidelines, industry standards, and all applicable representations and warranties. By reviewing the documentation in the loan file supporting approval in conjunction with the applicable underwriting guidelines, a re-underwriter can determine, among other things, whether the originator's guidelines were correctly applied, and if not, whether the deviation from the guidelines was supported by legitimate compensating factors." (Holt Direct ¶ 57.)

Holt and Grissom both relied heavily on the loan files to perform their so-called re-underwriting analyses. (See, e.g., Holt Direct ¶ 5; Holt Direct Appendix 1; Grissom Direct ¶¶ 35, 37.)

UBS objects to the admission of the loan files because, it asserts, a proper foundation has not been laid for their admission. UBS asserts that the Trusts have not proved that the proffered loan files are in the condition they were in at the time of origination. (Def. FF ¶ 251.) They also object to the admission of the loan files on relevance grounds. (Tr. 2097.)

The Trusts have submitted to the Court in electronic form 14,403 exhibits that use a "PX L" prefix and purport to be the loan files. Each individual exhibit often totals more than 1,000 pages in length. The files for any individual loan may be spread across as many as four separate exhibits. (See, e.g. , PX L2310, 6823, 7363, 8190 (loan file documents for Loan 40577613).)

A joint letter from the parties dated June 6, 2016 attached at Exhibit 2 a table listing all disputed trial exhibits, which includes all loan file exhibits. (Docket # 438.) Exhibit 2 summarizes UBS's objection as follows:

Admissibility dependent on Parties' briefing on the issue. Plaintiffs will submit letter briefing setting forth its full position on the subject on June 6, 2010. Plaintiffs are not moving to admit these documents for their truth. UBS objects to the admission of this document because (1) it is hearsay and (2) Plaintiffs never attempted to move it into evidence before resting their case.

(Docket # 438 Ex. 2.) UBS does not object to the admission of PX L1 through PX L14403 on grounds of authenticity. In a joint stipulation filed on February 4, 2016, the parties stated as follows concerning the authenticity of certain documents, including the loan files:

Absent evidence suggesting that a document is not what it purports to be, the loan origination files (the "Loan Files"), post-origination servicing records ("Servicing Records"), underwriting guidelines ("Guidelines"), and matrices ("Matrices") produced in this Action are authentic for purposes of Federal Rule of Evidence 901 ; provided , however, that the parties do not agree that the Loan Files are complete or are the same as the loan files that existed at the time of each corresponding loan's origination.

(Docket # 339 ¶ 3.) At trial, UBS made the following explanation for its objection:

We stip[p]ed to authenticity. We stip[p]ed whoever had the loan files produced a copy to plaintiffs, but we have no idea in what condition these loan files are in. We have heard the testimony there is multiple versions of these loans files. So to the extent Mr. Dunlap is saying we want to make sure the loan files are part of the record the court can consider, we don't understand the relevance or even what they're trying to use these loan files for.

(Tr. 2097.)

Holt testified that it was industry practice for Originators to maintain a full and complete loan file whenever possible. (Holt Direct ¶ 175.) The incentive for Originators to include all critical documents in a loan file was particularly strong for loans that were to be sold in the secondary market, where purchasers required review of the loan file prior to purchasing it. (Holt Direct ¶ 175.) As Holt testified: "These protocols meant it was unlikely that documents could accidentally disappear from loan files, and that even if they did somehow disappear from the manual file, the digitized version of the file should still reflect the complete version of the file as it existed at origination. Although it is true that the digitized version of the loan file could be corrupted, it is not likely that this would result in the selective deletion of critical documents such as I have observed here; rather, it would result in the entire digital loan file being destroyed or becoming inaccessible." (Holt Direct ¶ 175; see also Tr. 801 ("THE COURT: ... [A]s I understand your testimony, loan files were digitized by a vendor retained by an originator shortly after the origination of a loan; is that correct? Is that your testimony? THE WITNESS: Yes, sir.").)

At trial, Holt testified that based on his years working as an underwriter and on conversations with others in the industry, he understood that it was an industrywide practice to digitize loan files at the time of origination. (Tr. 780-81.) Holt testified that underwriters "might be sending them on a weekly basis to get them out of the office, digitized, and then they would warehouse them somewhere outside of their offices." (Tr. 781.) Holt also acknowledged that he had "no firsthand knowledge" of the Originators' digitization practices. (Tr. 783.) He testified that his review of the loans was based exclusively on digitized versions of the loan files, and that sometimes a loan had differing versions of a loan file. (Tr. 802.)

Grissom testified that the loan files she received for review were digitized at the time of origination. (Tr. 658-59.) She testified, "If the digital information was scanned appropriately to begin with, barring any extraordinary event, I wouldn't think that a digital loan file would be changed." (Tr. 659.)

Standing alone, the testimony of Grissom and Holt weigh in favor of a conclusion that the loan files consist of documents used in the underwriting process that were scanned and digitized contemporaneously to the loans' underwriting. This also would support a conclusion that, more likely than not, the loan files reviewed by the experts were consistent with their condition at origination.

A limited review of the "loan files" submitted as the Trusts' PX L exhibits paints a different picture. Upon the Court's own review of the exhibits submitted under the PX L prefix, the Court observed that they contained docket sheets related to borrowers' post-origination bankruptcy proceedings. (Loan 1447951, PX L10722 at 0457-61.) They contain extensive servicing records and materials that post-date origination, including borrowers' post-origination Form 1040 filings and utility bills, documents containing fax timestamps dated in 2011 and correspondence dated as recently as 2013. (See, e.g. , PX L6166 at 0001-0010, 0132-44, 0968-1004, 1036-37; PX L13588 at 0003-12; 0053-72, 0391-465; 1028-37.)

The Court located in one PX L "loan file" an internal, undated analysis authored by an individual named Natalie Cohen. (PX L792 at 0011-12.) Her analysis discussed whether certain guidelines breaches identified for Loan 1456451 materially and adversely affected the interests of the Certificateholders, and concluded that they did not. (PX L792 at 0011-12.) Another document relating to a different borrower, in Loan 1423423, summarized purported guidelines breaches related to a borrower's income misrepresentation, and the reasonableness of the borrower's stated income. (PX L300 at 1076.)

In another instance, the loan file relating to Loan 1477160 contains a document summarizing a breach allegation, the response to the allegation and a recommendation to rebut. (PX L1621 at 0087-89.) It includes discussion of the reasonableness of the borrower's stated income, the misrepresentation of income, layered risk and "excessive" DTI. (PX L1621 at 0087-89.) It also includes a reference to the Mortgage Loan Schedule. (PX L1621 at 0088.) This document contains an additional anomaly: while the borrower for Loan 1477160 is identified as a regional manager for Kay Jewelers with an address in Maryland, the borrower in this document is described as holding "a position in Food Service with the Department of Veteran Affairs" in Palo Alto, California. (PX L1621 at 0087.)

How this document, seemingly relating to the breach allegations of the present lawsuit, made its way into the loan file for a different, unrelated borrower, remains a mystery to the Court. At least in this instance, the exhibit, PX L1621, includes information that could not have been part of the borrower's loan file at origination. In this instance, the exhibit contains documents that do not relate to the relevant, identified loan number.

It is doubtful that the Trusts intended to submit to the Court these analyses and critiques. (PX L792 at 0011-12; PX L300 at 1076; PX L1621 at 0087.) They fall well outside the parties' description of the "loan files"—which, it has been maintained, consist of "any documents gathered or generated by the originator during the underwriting process that are necessary to the underwriter's decisions." (Holt Direct ¶ 4.) Neither the Trusts nor UBS has alerted the Court to the presence of these documents or explained why they were interspersed within the "loan files."

There is no apparent logic to the organization of the loan file exhibits. They are not arranged chronologically, and there is no clear basis to distinguish which materials were part of the "loan files" digitized at origination, and which were gathered or created after origination. For example, PX L6166 alternates post-origination servicing records with documents that appear to have been generated during the underwriting process.

Holt also testified that when he referred to a "loan file," he sometimes was referring to materials from post-closing servicing files:

THE COURT: When you say "the file," what file are you referring to, sir?

THE WITNESS: The loan file that was provided to us, digitized.

THE COURT: And that does not include the servicing file?

THE WITNESS: It could, yes, sir.

(Tr. 907.) The inclusion of servicing-related documents in a "loan file" contradicts the definition of "loan file" used in Holt's direct testimony, which he limited to "documents gathered or generated by the originator during the underwriting process that are necessary to the underwriter's decisions." (Holt Direct ¶ 4.)

The Court will require the Trusts to submit an affidavit or affidavits within 14 days explaining the anomalies, which exist despite the repeated representation that the loan files were digitized shortly after origination and/or during underwriting. The Court reserves the right to amend its findings, including by striking the loan files, any findings dependent upon the contents of the loan files, and, where appropriate, any opinions of Holt that are premised on the loan files.

Subject to this supplemental submission, the loan file exhibits are provisionally admitted, not for the truth of their contents but for the fact that certain statements contained therein were made. The loan files are received into evidence (1) for the fact that certain statements therein were made by the borrower, third-parties and/or the underwriter; and (2) to prove the actions and/or inactions of the underwriter. Because they are not being received into evidence for the truth of their contents, they are not hearsay. See generally Crawford v. Tribeca Lending Corp. , 815 F.3d 121, 126 (2d Cir.2016) (loan documents were not hearsay because they were received to show receipt of Truth in Lending Act disclosures, not for the truth of their contents).

The loan files contain the type of materials than an underwriter such as Holt relies on in the ordinary course of his business, and may support his opinions. Both Grissom and Holt relied heavily on the contents of the loan files. (See, e.g. , Holt Direct ¶ 5; Holt Direct Appendix 1; Grissom Direct ¶¶ 35, 37.) Rule 703, Fed. R. Evid., states that "[i]f experts in the particular field would reasonably rely on those kinds of facts or data in forming an opinion on the subject, they need not be admissible for the opinion to be admitted. But if the facts or data would otherwise be inadmissible, the proponent of the opinion may disclose them to the jury only if their probative value in helping the jury evaluate the opinion substantially outweighs their prejudicial effect." Rule 703 contains no such limitation on a non-jury trial, and when an action is tried to a court, the Supreme Court has stated that "the Federal Rules place no restriction on the revelation of such information to the factfinder. When the judge sits as the trier of fact, it is presumed that the judge will understand the limited reason for the disclosure of the underlying inadmissible information and will not rely on that information for any improper purpose." Williams v. Illinois , 567 U.S. 50, 132 S.Ct. 2221, 2235, 183 L.Ed.2d 89 (2012). It is therefore appropriate for the Court to consider the contents of the loan files not for the truth of their contents, but for the purpose of reviewing the opinions of Holt and Grissom.

There remains the separate issue of breaches identified by the Trusts that relate to certain "trailing documents" that they argue are not contained in the loan files. As understood by the Court, "loan file" refers to the documents that the underwriters for the Originator received, considered and/or generated in the course of deciding whether to approve and fund a borrower's application for a loan. The term is not a catch-all that encompasses any and all materials related to a loan, but only those materials that "are necessary to the underwriter's decisions." (Holt Direct ¶ 4.)

The Trusts have not proved that loan files were customarily present at the time of closing of title. They have not proved whether documents delivered at or after the closing of title—so called "trailing documents" such as title insurance policies, hazard insurance policies, recordable documents and documents executed or delivered at closing of title—would as a matter of custom and practice find their way into the loan files at or before the time the loan files were digitized or imaged. The Trusts rely on Holt's testimony to show that loan files were digitized or imaged shortly after the loan was originated and funded. This was an important point of rebuttal to UBS's argument that loan files lost their integrity through the passage of time and handling by servicers and others. But the Trusts offered no proof of when trailing documents, as a matter of custom and practice, were typically placed in a loan file and whether there was a custom and practice to await trailing documents before digitizing the loan file. In the case of trailing documents, the Trusts may not prove that the document never existed from the mere fact that it was not in the loan file on the date the file was digitized.

E. Permissible Use of Post-Origination Information by an Expert.

UBS argues that in identifying a breach of the Guidelines Warranty, Holt improperly relied on post-origination documents not available to underwriters at the time of origination. The PSAs warrant that each mortgage loan "was underwritten" in accordance with the applicable underwriting guidelines "in effect at the time of origination," which requires that any breach analysis be based on information that was available, or should have been available, to the underwriters. The analysis cannot be based on hindsight.For example, if a loan was underwritten in 2006, documents that prove a borrower's income in 2008 could not have been a red flag to the underwriter when examining the application in 2006. On the other hand, there may be documents that came to light in 2008 that speak directly to the borrower's circumstances in 2006, and contain information that should have informed the underwriting process under the guidelines.

A post-origination document showing, for example, that a borrower was reporting $30,000 per year income for the period 2007-10 may cast serious doubt on the same borrower's claim in 2006 that he was paid $100,000 per year for the same type of work with the same employer. Also, a post-origination document could reveal information about the borrower's income during an earlier period.

Of course, proving by whatever means that a borrower falsely stated his income on a loan application does not prove that the loan was underwritten in violation of the Originator's underwriting guidelines. For a stated-income loan, the amount of salary claimed, e.g. $100,000, may have been outright lie but it is an entirely separate question whether the stated income raised red flags for the underwriter. The analysis under the Guideline Warranty has a different focus than a breach of the MLS Warranty relating to DTI ratio. A MLS Warranty breach may be established by proving the income information was untrue as of the Closing Date, regardless of whether the Originator's underwriter or UBS knew or should have known the untruth at that time. A Guideline Warranty breach, however, requires proof of the actions and inactions of the underwriter at the time the loan "was underwritten."

The Court concludes that post-origination documents are not per se inadmissible or immaterial and may have some relevance to satisfying part of the Trust's burden. But proving that the borrower's income was inaccurately or falsely stated does not prove that the underwriter violated any guideline in failing to detect the inaccuracy or falsity. That is a separate inquiry on which a document that did not exist at the time of the origination is likely to be of little use.

As will be seen, post-origination documents may also be relevant on the issue of whether a proven breach materially and adversely affected the interests of the Certificateholders at the time of discovery or notice of breach.

F. Individual Guidelines Breaches.

The Trusts contend that the Originators failed to comply with underwriting guidelines in four principal respects. These purported infirmities include the absence of critical documents, unreasonable stated income, the application of unreasonable guidelines exceptions and the approval of loans that violated guidelines without the exercise of exceptions. The Court addresses each of them in turn.

1. The Absence of "Core Documents" in Loan Files.

The parties agree that certain "core documents" were required in order for Originators to issue a home loan. The Trusts contend that 1,276 loan files were missing these documents, and that the loans therefore were issued in violation of the guidelines.

According to the Trusts, these missing documents include property appraisals (absent for 97 loans); credit reports, including borrowers' FICO scores, payment histories, defaults and bankruptcy filings (absent for 109 loans); HUD-1 settlement statements, which itemize charges imposed on a borrower (absent for 331 loans); loan approvals, which document an underwriter's decision to issue a loan, including the presence of any compensating factors (absent for 297 loans); a Form 4506-T, which are forms that request a borrower's past tax returns and are required to corroborate a borrower's employment history and stated income (absent for 38 loans); and complete loan applications, including information about borrower income, assets, liabilities, sources of funds used for closing and the date of the property's acquisition (absent for 235 loans). (See Pl. FF ¶¶ 368-73.)

To the extent that the applicable guidelines required these materials to be considered by or prepared by an underwriter, the Court concludes that their absence from the loan files makes it more likely than not that they were not used in the underwriting process. While the Guidelines Warranty is not to be equated to a "complete loan files warranty," the contents of the loan files were the basis for the loans' underwriting and approval. The absence of these documents from the loan files means that it was probable that they were not considered by the underwriters. Therefore, to the extent that they were required pursuant to the underwriting guidelines, their absence from the loan files proves a breach of the Guidelines Warranty.

2. Failure to Inquire As to the Stated Income That Does Not Appear Reasonable.

The Trusts contend that 969 stated income loans were funded despite borrower applications that listed a stated income that was unreasonably high and should have raised a red flag and further inquiry.

The Originators' guidelines required an underwriter to evaluate a borrower's stated income for reasonableness. (See PX U578 at 31 (American Home Mortgage Guidelines stating that "Careful consideration must be provided relative to the income stated by the borrower and a reasonableness test is made by the AHM underwriter through various means such as ‘Salary.com’ or other providers of such information."); PX U012 at 12 (Countrywide Guidelines stating that "[l]imited income verification does not eliminate the need to analyze and evaluate the borrower's ability and willingness to repay the mortgage debt. The analysis must include a judgment about the reasonableness of the income stated on the application in relation to the borrower's occupation and credit information."); PX U374 at 19 (IndyMac guidelines stating: "The Borrower's stated income must be reasonable for the Borrower's type of employment, line of work, and assets.").)

Grissom and Holt both testified that they reviewed stated-income loans for reasonableness. Grissom testified:

Q. So to reunderwrite a stated income loan in these trusts, you must confirm that the underwriter determined that stated income was reasonable, correct?

A. I must confirm that the stated income was reasonable.

(Tr. 422.) Holt testified:

In general, the Originators' guidelines for their stated income programs required that the underwriters determine whether the stated income was reasonable before approving the loan. For example, one of the Originator's guidelines provided that "[s]tated income must be deemed reasonable and consistent with the borrower's profession," thus requiring underwriters to make such an evaluation. The Originators also provided general instructions to underwriters to assess the reasonableness of stated income by asking whether the "stated income [was] reasonable for the Borrower's type of employment, line of work, and assets?"

(Holt Direct ¶ 65.)

Holt has opined that 969 loans breached the Guidelines Warranty because borrowers' stated incomes were unreasonable. He states:

In conducting this analysis, the standard I applied was whether the stated income was plainly so improbable that a reasonable and responsible underwriter could not have accepted it without question; I believe this to be a conservative approach, giving the Originator the benefit of the doubt on close calls. In re-underwriting the Loans, I regularly saw stated incomes that were contradicted by other evidence within the loan file and met this standard.

(Holt Direct ¶ 148.)

UBS disputes Holt's opinions because they are premised, at least in part, on post-origination information. (See, e.g. , Def. FF ¶ 206.) As previously discussed, to the extent that this post-origination information proves facts that could have been known at the time of origination, they may constitute a valid basis for Holt to conclude that stated income was unreasonable.

The Court finds that the guidelines required the Originators to determine the reasonableness of stated income, and that the failure to do so constituted a breach of the Guidelines Warranty. Compliance with the guidelines is examined through the lens of information that would have been available to an underwriter at the time of origination. Information that was only available post-origination may bolster or rebut the conclusion that the underwriter failed to assess reasonableness of stated income at the time of origination or funding, but the focus of the inquiry remains on what the underwriter did or did not do based upon the facts before him or her at the time of origination.

As has been noted, the Guidelines Warranty is breached only if the Trusts prove that the loan was not "was underwritten in accordance with" the guidelines. The warranty speaks in the past tense and refers to the time when the loan "was underwritten" and not to a later period.

The issue of whether a failure to assess the reasonableness of stated income materially and adversely affects the interests of the Certificateholders stands on a different footing. Post-origination information may be relevant to this inquiry. This is because it is part of the Trusts' burden to prove that the breach affected the interests of the Certificateholders at the time of discovery or notice, points in time that are post-origination.

3. Guidelines Exception.

The Guidelines Warranty required that exceptions to guidelines be "exercised in a reasonable manner." (PX 49 at 196.) According to the Trusts, the Originators unreasonably exercised exceptions for 597 loans. Holt testified: "If an exception was approved, I reviewed the compensating factors identified by the underwriter to determine whether they were valid (i.e. , whether they existed) and whether they sufficiently compensated for the risk created by the exception. If they did, I concluded that the Loan was originated in compliance with the Applicable Underwriting Guidelines, and did not have a Material Guideline Defect." (Holt Direct ¶ 139.)

The Guidelines Warranty expressly requires that any exception be exercised in a reasonable manner. To the extent that the Trusts prove that any exception was unreasonably exercised, they have established a breach of the Guidelines Warranty. Holt could form a valid opinion that an exception was unreasonably exercised if the purported exception was based on compensating factors that did not exist or were not documented. Holt also could form a valid opinion that an exception was unreasonably exercised if he articulates a persuasive explanation as to why a particular exception was contrary to the best practices of a reasonable underwriter, given the total mix of information in a loan file.

As the Court previously discussed, only an exception that was exercised qualifies as an exception. To have been "exercised," there must be a contemporaneous expression of the decision to grant an exception. In practice, this means there must be some contemporaneous documentation. This conclusion is based on the text of the Originators' guidelines, as well as the requirement of the Guidelines warranty that any exception be "exercised in a reasonable manner."

According to the Trusts, the Originators approved 3,142 loans that violated underwriting guidelines, without any exception being exercised. For the reasons previously explained, Grissom's opinion that exceptions may be inferred from the fact that a loan not otherwise in compliance with the Guidelines was approved and funded is afforded no weight because it is contrary to the guidelines' documentation requirement and the Guideline Warranty's requirement to exercise exceptions "in a reasonable manner." The Court finds that an undocumented exception is nothing more than a post-hoc attempt to rationalize an Originator's noncompliance with the guidelines and is not an exception exercised in a reasonable manner.

4. The Relationship between the MLS Warranty and the Guidelines Warranty.

The Trusts argue that, to the extent that the Court finds a violation of certain MLS Warranties, those breaches necessarily prove a violation of the Guidelines Warranty. Specifically, the Trusts argue that to the extent that the MLS inaccurately lists data for DTI ratios, LTV ratios and owner occupancy, those defects would cause a breach of the Guidelines Warranty because this inaccurate information formed the basis for issuing loans that fell outside of the underwriting guidelines. The Trusts contend that 2,583 loans have breached the Guidelines Warranty based on a breach of the MLS Warranty as DTI ratio, LTV ratio or owner occupancy.

A breach of the MLS Warranty does not without more establish a breach of the Guideline's Warranty. The Court has already explained that an inaccurate statement of income, to the extent incorporated into the MLS's listing of the DTI ratio, may establish a breach of the MLS Warranty. But, in the case of a stated income loan, it may be provable with hindsight that the income number given by the borrower was false yet the underwriter may have fully complied with the requirements of the Guidelines. Incorrectness of the income figure proves a breach of the MLS Warranty, but only non-compliance with the guidelines proves a breach of the Guidelines Warranty. The same principles apply to LTV ratio, and owner occupancy. Non-compliance with the guidelines must be proved.

Moreover, the MLS guaranteed the accuracy of information at certain defined points in time, primarily as of the Closing Date set in the PSAs. (Of disputed items in the MLS Warranty, only the LTV ratio was warranted to be true at origination.) True, the Guidelines Warranty speaks as of the Closing Date and not as of the date of origination, but what is being warranted is past event, i.e. "[t]he Mortgage Loan was underwritten in accordance with the underwriting guidelines of the [originator] in effect at the time of origination with exceptions thereto exercised in a reasonable manner." Thus, in substance the Guidelines Warranty is directed to the facts available to the Originator during the underwriting period, and not as of the Closing Date.The time difference may have significant consequences for evaluation of individual loans. For instance, as discussed, Wu testified about borrower FICO scores that had declined significantly between origination and Closing Date. In finding a breach of the MLS Warranty, those lower FICO scores were gauged as of the Closing Date, and are not evidence of a breach of the Guidelines Warranty at the time of origination. DTI ratio might change significantly between "the time of origination" and the Closing Date, during which time a borrower might either accumulate or pay off a large amount of consumer debt. That higher DTI ratio would not have been a factor in underwriting the loans. Standing alone, a breach of the MLS Warranty as to DTI ratio would therefore not be evidence that the Guidelines Warranty was breached.

A breach of the MLS Warranty does not establish a breach of the Guidelines Warranty. Non-compliance with the guidelines must be proved to establish a breach of the Guidelines Warranty.

THE MAXIMUM LTV WARRANTY.

The PSAs each contain a warranty of the range of LTV ratios in the pool of loans held by the particular Trust. For example, the PSA governing the 2006-OA2 Trust states that "the range of original Loan-to-Value Ratios of the Mortgage Loans is 7.82% to 95.00% ...." (PX 49 at 196 (Sched. II(xxxiii))); PX 110 at 236 (Sched. II(xxxiii)) (same, but with LTV ratios ranging from "14.29% to 100%"); PX 182 at 230 (Sched. II(31)) (same, but with LTV ratios ranging from "17.71% to 95.00%").

According to the Trusts, UBS breached this MLS Warranty based upon Cowan's opinion that LTV ratios that exceeded those warranted for each Trust. (Pl. FF ¶¶ 334-36.) For the reasons discussed, however, the Court rejects Cowan's method for recalculating LTV ratios as inconsistent with the PSAs' prescribed approach to determining LTV ratio. Because the Trusts rely entirely on Cowan's AVM recalculations to prove a breach of the Maximum LTV Warranty, they have failed to prove a breach of that warranty.

THE MORTGAGE FILE WARRANTY.

Each PSA warrants that, as of the Closing Date, "[w]ith respect to each Mortgage Loan, the Transferor [UBS] is in possession of a complete Mortgage File except for the documents which have been delivered to the Trustee or which have been submitted for recording and not yet returned." (PX 49 at 196 (Sched. II(xxxii))); PX 110 at 236 (Sched. II(xxxii)); PX 182 at 230 (Sched. II(30).)

"Mortgage File" is a defined term, and is not to be confused with the shorthand phrase "loan files" used by the parties and the Court. The PSAs define "Mortgage File" as: "The mortgage documents listed in Section 2.01 hereof pertaining to a particular Mortgage Loan and any additional documents delivered to the Custodian to be added to the Mortgage File pursuant to this Agreement." (PX 49 at 0045.)

Section 2.01 is headed "Conveyance of Mortgage Loans." In addition to creating the Trusts, section 2.01 governs the mortgage loans' transfer from UBS to U.S. Bank in its capacity as Trustee. (PX 49 at 0067-71.) Section 2.01(b)(i) requires the Depositor (Mortgage Asset Securitization Transactions, Inc.) to deliver to the Custodian (Wells Fargo) a set of documents, including "the Original Mortgage Note," "the original recorded Mortgage or a copy of such Mortgage," "a duly executed assignment of the Mortgage," "the original copies of each assumption, modification, written assurance or substitution agreement, if any," and "the original or duplicate original lender's title policy and all riders thereto." (PX 49 at 0068.) Additional categories of documents are required for a cooperative mortgage loan. (PX 49 at 0068-69.) The Court concludes that the documents listed in section 2.01 are the contents of the "Mortgage File" that UBS warranted was in its possession at the Closing Date.

Shortly after the Closing Date, the "Mortgage File" was delivered by UBS to Wells Fargo in its capacity as Custodian. Section 2.02 states that "[t]he Custodian, on behalf of the Trustee ... declares that it holds and will hold such related documents and the other documents delivered to it constituting the Mortgage Files ...." (PX 49 at 0071.) Section 2.02 further provided that, within 90 days after closing, the Custodian would provide a "Final Certification" that these documents had been received. (PX 49 at 0072.) It states that "[u]pon receiving each Final Certification from the Custodian, [U.S. Bank] shall notify [UBS] and the Certificate Insurer of any document defects listed as exceptions in each such Final Certification." (PX 49 at 0072.) Further, "[t]he Custodian shall retain possession and custody of each related Mortgage File in accordance with and subject to the terms and conditions set forth herein." (PX 49 at 0072.)

The Mortgage File Warranty must be read in harmony with sections 2.01 and 2.02. See, e.g. , Beal Sav. Bank , 8 N.Y.3d at 324–25, 834 N.Y.S.2d 44, 865 N.E.2d 1210. The two provisions allocate responsibility for custody of a complete Mortgage File, with UBS responsible for such files as of the Closing Date, and the Custodian responsible for custody within 90 days thereafter. Section 2.02 provides that Wells Fargo, the Custodian, "holds and will hold" the Mortgage Files. Within 90 days of closing, Wells Fargo was to issue a "Final Certification," and U.S. Bank, in its capacity as Trustee, was to notify UBS and Assured of any document defects. Thereafter, Wells Fargo was required to maintain custody of the Mortgage Files.

The Trusts have not proved that UBS breached the Mortgage File Warranty. They cite to no evidence concerning Wells Fargo's custody of the files, the Final Certification that Wells Fargo was obligated to issue or U.S. Bank's obligation to notify UBS of any defects in the Mortgage Files, nor do they cite to any evidence that UBS did not a complete Mortgage File as of the Closing Date. The PSAs prescribed a procedure and timeline for the transfer of the Mortgage Files and the procedures for noticing any defects they contained, which the Trusts have failed to address.

Moreover, the Trusts appear to equate the loan files used in underwriting with the Mortgage Files. As previously discussed, the loan files are limited to those materials that are created when a borrower applies for and is issued a loan, and includes documents submitted by a borrower or gathered or generated by the underwriter; the loan files are literally a snapshot as of the date the loan file was digitized. The "Mortgage File," as defined in the PSAs, is specific category of documents, including documents that would not be expected to have been received until after the origination of the loan.

The Trusts have failed to prove that UBS breached the Mortgage File Warranty as to any loan.

THE TITLE INSURANCE WARRANTY.

As has been noted, the Mortgage File Warranty includes a warranty that the file contains and "the original or duplicate original lender's title policy and all riders thereto." (PX 49 at 0068.) Separate and apart from this warranty, UBS warranted that, as of the Closing Date, the property "is covered by an American Land Title Association mortgagee title insurance policy or other generally acceptable form of policy or insurance ...." (PX 49 at 0193 (section (xvii) to Schedule II).)The Title Insurance Warranty requires that the property in fact be covered by appropriate title insurance. The Mortgage File Warranty speaks not to coverage put to the actual policy and riders. The two warranties apply to different but related circumstances.

The Trusts contend that that they have proved a breach of the Title Insurance Warranty because title insurance policies are entirely missing from 1,170 loan files, and because for another 1,498 loans, the title insurance policy is missing a negative amortization rider. (Pl. FF ¶ 354.)

Grissom testified, and the Trusts do not dispute, that a title insurance policy is issued as of the closing of title, and that the policy itself may be delivered after the closing and may not have been available to the underwriters at origination. (Pl. FF ¶ 354; Tr. 72 ("we don't dispute what Ms. Grissom says about title insurance, that it often was obtained after closing.").) The Trusts argue that, even if a title insurance policy was not available at the closing of title, it "is irrelevant, because even if title insurance was not available at origination, the Title Insurance Warranty is made as of the Closing Dates [of the PSAs], which occurred later." (Pl. FF ¶ 354.) But Holt and Grissom have both used the phrase "loan files" to refer to documents used by underwriters to evaluate the borrowers' loan applications and to fund the loans. Schedule II warrants that each property is covered by a title insurance policy, but it does not warrant that documentation of the title insurance was required to be included in loan files used by Originators.

In closing argument, counsel to the Trusts stated that "[m]ortgage file is defined in the PSA as including ... the documents listed in 2.01. And 2.01 identifies the documents which are, among others, title insurance policy, mortgage note and deed." (Tr. 2073.) Counsel made the same observation in his opening:

MR. BALDWIN: The mortgage file is a defined term, your Honor. And the mortgage file includes title insurance, the mortgage note and the deed. Those are the critical ones here.

THE COURT: So it's not an underwriting file?

MR. BALDWIN: It's not an underwriting file. It's not the full loan file.

(Tr. 72.)

The title insurance policy was therefore required to be included in the Mortgage File defined in the PSAs, which, as discussed, is separate from the loan files used by the underwriters. As discussed, the Mortgage Loan Files are held by Wells Fargo and at trial the Trusts did not offer proof of the content of those files. As previously noted, they have failed to prove a violation of the Mortgage File Warranty.

The Trusts have shown that loan files that they have offered into evidence were digitized or imaged at or about the time of origination of the loan. They have offered no evidence that a loan file would contain documents received after origination but before the Closing Date of the PSAs. The Trusts have not shown that it would be unreasonable, improper or contrary to industry custom and practice for a copy of the title policy or proof of title insurance to be placed in a loan file after the date of closing of title and after the digitization or imaging process was complete.

The Trusts have credibly argued, and Holt has testified, that loan files were digitized or imaged very soon after origination. This is a point they emphasize to rebut UBS's argument that some of the contents of the loan files may have been misplaced or destroyed in the post-origination period. But the Trusts have not shown that it would be reasonable to expect a uniform practice that title insurance policies or other proof of title insurance would have made its way into the loan file at the time of digitization. Indeed, they have not endeavored to prove that the loan file was customarily present at the time and place of closing of title, nor have they offered evidence as to how long it would take for documents delivered at closing to make their way into the loan files.

Because the Trusts have failed to prove that the title policy (or its negative amortization rider) was required to be placed in the loan file at or about the time the loan files were digitized, the Trusts have failed to prove a breach of the Title Insurance Warranty.

But whether viewed as a breach of the Guidelines Warranty or of the Title Insurance Warranty, the Trusts will have proved a violation of either or both of these warranties if there is no indication in the loan file that title insurance was obtained prior to the funding of the loan. That indication in the loan file need not be a specimen of the actual policy, but may be other preliminary confirmation of coverage. See, e.g. , IndyMac Guidelines § 2402.03 at PX U261-020 ("The title commitment (or other form of title evidence applicable in the jurisdiction, such as a title binder, preliminary title report, attorney's opinion letter) must be included in the Loan file....") As discussed below, Grissom has repeatedly identified instances where evidence of title insurance is reflected in a HUD-1 Form or where a copy of the title insurance policy is included in the loan file.

THE HAZARD INSURANCE WARRANTY.

In the PSAs, UBS warranted that, as of the Closing Date, the property for each loan "is insured" against "loss by fire" and, if flood hazards were present, had "a flood insurance policy." (PX 49 at 0194 (section xviii of Schedule II) (the "Hazard Insurance Warranty").) This language unambiguously warrants that, as of the Closing Date of the PSAs, each loan was insured by the required hazard insurance policy or policies. The Trusts argue that the absence of evidence of hazard insurance in 242 loan files proves that there was no such insurance.

Apart from the Hazard Insurance Warranty in section xviii of Schedule II, there is a separate provision of the PSA, section 3.11 of the PSAs, which is labeled "Maintenance of Hazard Insurance." Section 3.11(a) states that "[f]or each Mortgage Loan," Wells Fargo, as the Master Servicer, is required "to maintain or cause to be maintained standard fire and casualty insurance and, where applicable, flood insurance, all in accordance with the provisions of the related Servicing Agreements." (PX 49 at 093.) The policy must be issued by insurers who meet certain eligibility requirements set forth in the Servicing Agreement. (Id. )

The Trusts make no claim that the maintenance of hazard insurance provision, section 3.11, has been breached as to any loan. The Hazard Insurance Warranty and section 3.11 speak to different time periods. The Hazard Insurance Warranty provides that a hazard insurance policy was in place as of the Closing Date. Separately, section 3.11(a) assigns the Master Servicer [Wells Fargo] a responsibility to ensure that hazard insurance is maintained "[f]or each Mortgage Loan ...."

Hazard policies, like title insurance policies, ought to become effective contemporaneous with the closing of title. Before closing of title, the borrower does not have an insurable interest in the property. Effectiveness of the policy as of the closing ensures that there is no temporal gap. The actual specimen of the hazard policy, like a title insurance policy, is a trailing document because there may be a time lag before the actual policy is received by an Originator. As in the case of Title Insurance, the Trusts have not shown that it is logical or reasonable to assume that actual policy should exist in a loan file as of the date the loan files were digitized or imaged.

The Trusts may not prove a breach of the Hazard Insurance Warranty because a fully endorsed policy is not contained in the loan file. However, the Trusts may prove that there was a breach of either the Guidelines Warranty or the Hazard Insurance Warranty or both by proving that there is no proof of a commitment by an insurer to cover the property such as a cancelled check for the policy amount or a binder or memorandum or letter of commitment from the insurer.

THE "MATERIALLY AND ADVERSELY AFFECTS" REQUIREMENT.

A. "Materially and Adversely Affects" Is Measured from the Date of the Repurchase Demand, Not from the Date that the Warranties Were Made.

Under section 2.03, UBS has the obligation to repurchase a breached loan only if the breach "materially and adversely affects" the interests of Certificateholders. The relevant portions of section 2.03 state:

With respect to any representation and warranties set forth on Schedule II hereto which are made to the best of the Transferor's knowledge if it is discovered by any of the Depositor, the Certificate Insurer, the Master Servicer, the Transferor, any Servicer, the Trustee or the Trust Administrator that the substance of such representation and warranty is inaccurate and such inaccuracy materially and adversely affects the value of the related Mortgage Loan or the interests of the Certificateholders or the Certificate Insurer therein, notwithstanding the Transferor's lack of knowledge with respect to the substance of such representation or warranty, such inaccuracy shall be deemed a breach of the applicable representation or warranty.

* * * *

The Trustee shall enforce the obligations of the Transferor in accordance with this Section 2.03 to correct or cure any such breach of a representation or warranty made herein, and if the Transferor fails to correct or cure the defect within such period, and such defect materially and adversely affects the interests of the Certificateholders and the Certificate Insurer in the related Mortgage Loan, the Trustee shall enforce the Transferor's obligations hereunder ....

(PX 49 at 0073, 0074.)

"[M]aterially ... affects" means that the breach at issue would have altered the price that a willing purchaser would pay for the loan or otherwise changed the risk of loss on the loan. "[A]dversely ... affects" means that the impact would be detrimental to the financial interests of the Certificateholders; in other words, the alteration in price would mean a lower price, or the change in risk would be an increased risk.

In its summary judgment decision, the Court concluded that a material adverse effect arises when a breach results in an increased risk of loss to the Certificateholders. 2015 WL 764665, at *15. The Court observed that "not all breaches trigger a cure or repurchase obligation. The breach must ‘materially and adversely affect the interests of the Certificateholders in such Mortgage Loan.’ " 2015 WL 764665, at *10. The Court also concluded at summary judgment that, pursuant to the express contractual language quoted above, that "the determination of whether the breach ‘materially and adversely affects the interest of the Certificateholders ...’ is assessed as of the cure-repurchase period." Id. The timing of a material adverse effect is grounded in the language of the PSAs, which uses the word "affects" in the present tense. The present-tense use of the word "affects" expressly applies to the time when notice is given: section 2.03 states that in order for an inaccurate warranty to be "deemed a breach," it is necessary that "such inaccuracy materially and adversely affects" the Certificateholders "with respect to the substance of such representation or warranty ...." Section 2.03 might have been worded differently: it could have provided that a warranty is "deemed a breach" if it affected Certificateholders as of the Closing Date, or it might have provided that an inaccuracy is "deemed a breach" if it "has affected" the interests of the Certificateholders. But, in this context, to construe the word "affects" as having an expansive meaning that extends back to the Closing Date would rewrite the express terms of the PSAs.

The parties were on notice of this construction of section 2.03 more than a year before trial because of the summary judgment decision, see 2015 WL 764665, at *10, and the point was repeated in the course of the trial. (PX 49 at 0074; see also Tr. 128 (the Court) ("But I'm saying as a matter of construction of the word ‘affects' it is speaking as of the moment of the triggering of the cure or repurchase obligation."), Tr. 128 (the Court) ("The contract was written with the cure or repurchase remedy in mind, and so it says there must be a breach of a rep and a warranty which has been discovered and noticed, which triggers a cure or repurchase obligation if the defect materially and adversely affects the interest of the certificate holder.").) To impose a repurchase obligation on UBS, the Trusts must prove that the alleged breach had a material adverse effect on Certificateholders' interests at the time that the repurchase obligation was triggered.

This reading of the PSAs is further supported by other language in section 2.03. Section 2.03 states: "It is understood and agreed that the obligation under this Agreement of the Transferor to cure, repurchase or replace any Mortgage Loan as to which a breach has occurred and is continuing shall constitute the sole remedies against the Transferor respecting such matters available to Certificateholders ...." (PX 49 at 0075-76.)

Further, certain types of breaches not at issue in this action were expressly exempted from the requirement to prove that the breach materially and adversely affects the interests of the Certificateholders. Section 2.03 provides in part that: "Notwithstanding the foregoing ... a breach of any of the representations and warranties set forth in clauses (xiii), (xiv), (xv) and (xxxv) through (1) of Schedule II, in each case, will be deemed automatically to materially and adversely affect the interests of the Certificateholders in such Mortgage Loan ...." (PX 49 at 0074.) None of the foregoing clauses are at issue in this action.

The Trusts do not meaningfully address this language. Instead, they urge that whether a defect "materially and adversely affects" the Certificateholders should be determined from the point in time that the representations and warranties were made. (Pl. FF ¶ 165.) They base this argument on a series of decisions that discuss New York's six-year accrual period for breach of contract claims, principally ACE Securities Corp. v. DB Structured Products, Inc. , 25 N.Y.3d 581, 15 N.Y.S.3d 716, 36 N.E.3d 623 (2015), which held that the limitations period begins to run from the date that the representations and warranties are made, and not at the time the defendant failed to satisfy its repurchase obligation. ACE explained the importance of a bright-line enforcement of the limitations period, and held that the representations and warranties applied to the facts as they existed at the time that the representations and warranties were made. Id. at 593–97, 15 N.Y.S.3d 716, 36 N.E.3d 623. Because the breach of contract claim is directed toward the breach of representations and warranties, the limitations period is not separately triggered when the defendant fails to repurchase a defective loan. Id. at 596–97, 15 N.Y.S.3d 716, 36 N.E.3d 623 ; see also Deutsche Bank Nat. Trust Co. v. Quicken Loans Inc. , 810 F.3d 861, 866 (2d Cir.2015) ("A representation of present fact is either true or false—and the contract therefore performed or breached—if the underlying fact was true or false at the time the representation was made.") (applying ACE ).

However, ACE says nothing about when the required material and adverse effect of such breaches is assessed. In Deutsche Bank , the Second Circuit, applying New York law, followed ACE 's conclusion that the representations and warranties are breached, if at all, "upon effectiveness," and thus the statute of limitations began to run at that time. 810 F.3d at 869. The Second Circuit also said nothing about establishing an entitlement to the contractual repurchase remedy, which is a separate inquiry that, under the plain language of the contracts, must be made at the time the remedy is sought. Indeed, in a later summary order, the Second Circuit clarified that, under the reasoning of ACE , "[t]he ‘material[ ] and adverse[ ]’ effect language does not create an element of an actionable breach. Rather, the language referring to material adverse effects is part of the provision through which [plaintiff] might seek to repurchase as one recourse for a breach of representations and warranties made in the MLPA." Wells Fargo Bank, NA v. JPMorgan Chase Bank, N.A. , 643 Fed.Appx. 44, 47 (2d Cir.2016) (summary order). The Second Circuit added: "the material and adverse effect requirement is a component of the remedy and not of the breach ; thus, like the cure period in Quicken Loans , it does not delay accrual" of the statute of limitations. Id. (emphasis added). Relying on ACE and the Second Circuit's Deutsche Bank decision, the Appellate Division, First Department recently drew the same distinction between the accrual of a claim and the contractual notice requirement. See Deutsche Bank Nat'l Trust Co. v. Flagstar Capital Markets Corp. , 143 A.D.3d 15, 36 N.Y.S.3d 135, 141 (1st Dep't 2016) ("the accrual provision's requirement that a plaintiff make a demand on defendant for performance of the agreement does not constitute a substantive condition precedent that could delay accrual of the breach of contract claim.").

The Court therefore concludes that ACE and its related authority have no bearing on the timing of the PSAs' material and adverse effect requirement.

The Trusts also argue that the plain reading of the PSAs as requiring proof of a material and adverse effect the time of notice or discovery would be "absurd" and "commercially unreasonable" (Pl. Br. at 30), but they do not articulate why. To the contrary, it is reasonable for the parties to bargain for a limitation on the representations and warranties such that the repurchase obligation is triggered only where a material breach at the time of contracting continues to have a material adverse effect at the time the breach is noticed or discovered and a remedy is sought. To conclude otherwise would give the Trusts a unilateral ability to put back loans that, after many years of performance, may have had breaches even if those breaches no longer affect the Certificateholders' interests.

A breach at the time of origination or at the Closing Date may have an effect that carries on indefinitely, including up to the time of discovery or notice, but that need not always be the case. An intentional misrepresentation of income by a borrower, if known, would have resulted in the loan never having been approved, funded or sold; this is an effect that continues to the time of discovery or notice. In contrast, a failure of an underwriter to obtain a verification of one of two jobs held by a borrower would be a breach of the Guideline Warranty but would have no effect continuing effect if, for example, the verification would have confirmed employment.

The Trusts must prove that a breach materially and adversely affects the interest of the Certificateholders at the time of the breach's discovery or notice.

1. The "Materially and Adversely Affects" Requirement of Section 2.03 Is Analyzed Separately from the Materiality Requirement Contained in Individual Representations and Warranties.

Section 2.03 of the PSA incorporates an express materiality requirement for certain representations and warranties. For example, in subsection (i) to Schedule II of the PSAs, UBS warrants that "[t]he information set forth in the Mortgage Loan Schedule was true and correct in all material respects at the date or dates respecting which such information is furnished as specified in the Mortgage Loan Schedule." (PX 49 at 0191; emphasis added) Thus, a claim of breach of the MLS Warranty requires the Court to consider whether the purported breach rendered the MLS untrue or incorrect in a material way.

If the MLS Warranty was breached—meaning that information was materially untrue and incorrect—then, in order for UBS's repurchase obligation to apply, the Trusts must also prove that "such inaccuracy materially and adversely affects" the interests of the Certificateholders.

In contrast, the Guidelines Warranty states: "The Mortgage Loan was underwritten in accordance with the underwriting guidelines of the related Loan Seller in effect at the time of origination with exceptions thereto exercised in a reasonable manner." (PX 49 at 0196.) Thus, there is no express materiality requirement incorporated into the Guidelines Warranty. However, in deciding whether failure to comply with guidelines breached the Guidelines Warranty, the materiality analysis of section 2.03 turns on whether "such inaccuracy materially and adversely affects" the interests of the Certificateholders.

2. A Material Adverse Effect under Section 2.03 May Be Proved with Evidence of an Increased Risk of Loss to the Certificateholders.

In its summary judgment decision, the Court concluded that a material adverse effect arises "when a breach of representation and warranty increases the risk of loss, and that the loan need not be in default." 2015 WL 764665, at *15 (collecting cases). The Court noted that section 2.03 does not require a loan to be in default in order for a material and adverse effect to arise, and does "not specify any other limitations on which events constitute a material and adverse effect arising from an inaccuracy in the representations and warranties.... The Trusts may rely upon proof that as to a specific loan, there is a material or significant increase in the risk of loss." Id.

In denying the Trusts' motion for reconsideration, the Court noted that any material adverse effect arising out of a breach could be negated if the "significantly increased risk of loss" was eliminated by other circumstances. As the Court explained: "[F]or example, the failure to obtain written verification of the salary and employment of a borrower may be a material deviation from underwriting standards, but if the borrower, in fact, was paid the exact salary and had the exact employment that he claimed, the material deviation would not ‘materially and adversely affect the interests of the Certificateholders ....’ " 2015 WL 797972, at *3. It further explained: "While the Summary Judgment Opinion rejects the notion that the plaintiffs must prove an actual loss or default, it does require proof of a significant increase in the risk of a loan's default." Id.

The Court adheres to the reasoning in its summary judgment opinion. In order to prove that a breach "materially and adversely affects" the Certificateholders, the Trusts need not show that the breach cause an actual loss, nor that any actual loss suffered was caused by the breach. They may instead prove that the breach increased the risk of loss to the Certificateholders.

3. The Trusts May Also Prove a Material Adverse Effect with Evidence that a Breach Resulted in Altered Loan Terms.

The Trusts may establish a material and adverse effect if they prove that the proper application of the Originator's guidelines would have dictated different terms for the loan. For example, the guidelines may have permitted the approval or funding of a loan but at a higher rate of interest. The effects continue past origination of the loan because the borrower's monthly payment would have been correspondingly higher; those funds would have been paid into the Trust, ultimately affecting the interests of the Certificateholders. There may be other loan terms that would have been different for which the Trusts, at least in theory, may endeavor to prove an ongoing economic effect.

As the Trusts raised in argument, the Certificateholders also may suffer a material and ongoing harm because the Trusts paid a higher price than they should have to acquire a given loan in their respective pools. (Tr. 2007.) If there is proof that the Trusts overpaid for the purchase of an individual loan because of a breach, then the repurchase remedy will be available.

Because section 2.03 provides a cure or repurchase remedy, the Trusts need not quantify the amount of the ongoing economic impact. They need only prove that the breach materially and adversely affected the interest of the Certificateholders at the time of notice or discover. If it did (and other elements are satisfied), then the repurchase obligation is triggered.

No witness for the Trusts offered credible evidence as to how a defect would have caused a higher rate of interest to be charged or other loan terms to be different. The closest to come to that was Holt. Using a hypothetical example, Holt testified that he was uncertain as to whether the exercise of an exception would lead to higher interest rates for a borrower, or, alternatively, additional fees that were collected by the Originator. On re-direct examination from the Trusts' counsel, he testified as follows:

Q. Let's say an exception had been granted, sought and obtained for this loan. In your experience, would that exception have required an additional interest rate adjustment?

A. For a deviation that high, yes, it should have received some kind of increase in interest rate or an increase in originating fees that were collected at the time of closing, some kind of increase.

Q. Do you have a view whether it would have been the additional price would have been reflected in higher fees or interest rate?

A. Say that again.

Q. Do you have an [sic] view whether the additional price would have been reflected in higher fees or a higher interest rate, or both?

A. I don't. It could be either one them or it could be both. The rate sheet could explain that, but you could see it either way.

THE COURT: To the banks, money is pretty much fungible, right?

THE WITNESS: Right.

(Tr. 1128-29.)

The Trusts, through Holt, engaged in a systemic review of all loan files at issue, except for certain IndyMac files. It could have offered evidence of whether a defect resulted in an approval of a loan but a higher rate of interest. But, with the exception of the one loan discussed in Holt's testimony quoted above, they did not do so.

As Holt noted, an Originator's "rate sheet" and other content of the Originator's underwriting guidelines would likely show whether a specific deviation from a guideline would result in the loan issuing but at a higher rate of interest or fees. But the Trusts have not pointed to specific content of any Originator's "rate sheet" or guidelines that show that any loan, among the thousands at issue, would have issued at a higher interest rate or fee.

4. Multiple Breaches as Evidence of Material and Adverse Effect.

The Trusts contend that the presence of multiple, material breaches strengthens and reinforces any findings as to materiality. (Pl. FF ¶ 412.) For example, if more than one item of information in the MLS is incorrect, those multiple misstatements should be weighed collectively in determining whether a breach of the MLS Warranty had a material and adverse effect on the interests of the Certificateholders. (Pl. FF ¶ 412.) The notion that multiple defects increase the risks associated with a loan is referred to by the shorthand phrase "layered risk."

Holt testified that underwriters should consider "layered risk associated with the cumulative impact of other defects or risk factors associated" with each loan. (Holt Direct ¶ 188.) In its diligence reviews of the loans, UBS considered "layered risk," which it said "occurs when several credit and property attributes together, increase the frequency and seventy of default. Independently, these attributes do not appear to be concerning, however, when a seller layers the risk, the potential for loss increases considerably." (PX 19 at 15; see also PX 12 at 18.) Both Twombly and Lantz testified to the importance of assessing layered risk. (Tr. 1295 ("[S]ome of the criteria could be layered. So I may be looking for—as an example, I may be looking for high LTV with a low FICO."); 1459 (Lantz testimony that UBS diligence considered layered risk).) This practice was consistent with the Originators' assessment of layered risk. Countrywide guidelines stated that its AUS was designed to identify loans that "present[ ] excessive layering of risk" such as "insufficient liquid assets, high LTV, credit history, and/or high debt ratios for the loan program selected." (PX U008 at 2.) American Home instructed its underwriters to "[d]etermine if layering of risk factors is acceptable" for loans referred for failure to meet Fannie Mae's eligibility requirements. (PX U578 at 10.)

The Court finds that the concept of layered risk is accepted in the guidelines and was accepted in the underwriting industry. It should be applied in assessing whether the material and adverse standard has been met.

The Court concludes that the cumulative effect of multiple breaches may support a finding that a breach materially and adversely affected the interests of Certificateholders. Of course, one serious and significant breach may be sufficient to satisfy the materially and adversely standard, while several minor deviations by the underwriter which does not reflect an increased risk of loss may not. There is no mechanical formula that can reliably be applied.

5. Proof of Intentional Misrepresentations by a Borrower Will Prove a Material and Adverse Effect.

In assessing a credit risk, an underwriter looks at the borrower's ability and intention to repay the loan. It is important to the underwriter to have an understanding of the borrower's true financial picture with respect to the property and the loan. Deceit by a borrower undermines the ability to trust other statements by a borrower. For example, a borrower who intentionally concealed outstanding mortgage debt on other non-disclosed properties may be acquiring the subject property purely for investment purposes and would be more willing to walk away from the property in a falling market.

The Court concludes that it is more likely than not that the level of risk would increase substantially and materially if a loan was procured by borrower fraud, deceit or misrepresentation, because a borrower who made knowing misrepresentations in the underwriting process would be less reliable in making scheduled payments and more likely to default. Therefore, if the Trusts can prove that it is more likely than not that a borrower intentionally misstated information as part of the underwriting process, they will have proved a material and adverse effect on the interests of Certificateholders.

6. The Trusts' Proof of a Material and Adverse Effect.

The Trusts principally, if not exclusively, relied on Holt's testimony to prove that a breach materially and adversely affected the interests of the Certificateholders at the time of discovery or breach.

Holt's direct testimony focused on the entire universe of over 9,000 allegedly defective loans and drew no important distinctions between the several different warranties in the PSA:

For each Loan for which I identified defects, I then evaluated whether the defects were material and would adversely affect the interest of Certificateholders, who held the risk of loss on the Loan. In my opinion, a material defect is a defect that significantly increased the risk of loss with respect to the Loan. Throughout this declaration, I refer to such defects as "Material Defects" (and to loans with Material Defects as "Materially Defective"). In residential mortgage lending generally, and the Loans specifically, such a risk of loss includes (a) the risk that the borrower will default on the Loan, (b) the risk that the property securing the Loan will not constitute adequate collateral in the event of default, and (c) the risk that the Loan and/or collateral documentation will not be fully enforceable against the borrower, in whole or in part.

In assessing materiality, I relied on my many years of experience as an underwriter, and in particular my experience with underwriting guidelines, which distill the collective experience of underwriters accumulated over many decades and many millions of loans. As explained further below, in Section III, the purpose of underwriting guidelines is to ensure the quality of loans, thereby minimizing the risk of loss on the loans. ...

(Holt Direct ¶¶ 7-8.)

Holt assessed materiality at the date of origination and not at the time when the cure or repurchase obligation arose, which may have been 3 to 9 years after origination. Holt steadfastly maintained, despite the language of section 2.03 and the guidance of this Court's summary judgment decision, that materiality "can and should be assessed at the date the loan was originated, not retrospectively":

For an underwriter, materiality can and should be assessed at the date the loan was originated, not retrospectively. In other words, an imprudently underwritten loan cannot subsequently be rendered prudent with the benefit of hindsight. For example, if the underwriting guidelines required that the underwriter obtain a verification of the borrower's income and the underwriter failed to obtain such verification, and if it were subsequently determined—post-origination—that the borrower had such income, that would not cure the Material Defect. The risk of loss with respect to the loan was still significantly higher as a result of the failure to obtain the required income verification. More importantly, as with fraud or misrepresentation, an underwriter's failure to comply with basic underwriting requirements casts doubt on the integrity of the entire underwriting process. However, all of the defects I found to be Material Defects would also be Material Defects, i.e., they would still significantly increase the risk of loss, even if materiality were assessed at a later date, including at the date the Loans defaulted, or at the time notice of the defects was provided to UBS.

(Holt Direct ¶ 10.)

Holt's direct testimony overstates the case for the flat-out conclusion that all "material" breaches of a warranty at the time of origination have an ongoing effect that continues into the future. The example he cites (originating with the Court's opinion denying reconsideration of its summary judgment opinion, see 2015 WL 797972, at *3 ) is as follows: in breach of the underwriting guidelines, an underwriter fails to verify a borrower's income but, had he done so, it would have confirmed the accuracy of the borrower's income. Holt concludes that this would have a lingering and ongoing impact on the Certificateholder because an underwriter's failure to comply with basic underwriting requirements casts doubt on the integrity of the entire underwriting process.

The Court finds Holt's testimony on this point unpersuasive. In Holt's example, the borrower's circumstance have remained constant and, indeed, truthfully reported. The borrower's ability and willingness to repay are exactly as they appear to be on the face of the loan application. The breach of the Guideline Warranty resulted from the underwriter's failure to verify an item of information which, in the case of the hypothetical, was truthful and accurate information from the borrower. The interests of the Certificateholders are not materially or adversely affected by this particular breach.

It does not follow logically or comport with common sense that the work of the underwriter cannot be trusted on other aspects of the same loan because of one oversight. If this single act of oversight in failing to verify perfectly truthful information infected the entire underwriting process conducted by the negligent underwriter, then logically the infection would extend to all loans underwritten by the negligent underwriter in the same time period. Holt and the Trusts make no such argument.

Holt does not explain the logic of his opinion and uses no examples from his work in the field of underwriting to justify this opinion. It is conclusory and contrary to common sense. It fails to take account of the overall context of the borrower's loan approval and what other data may have been available.

The Court rejects Holt's attempt to equate underwriter inattentiveness with borrower fraud, deceit or misrepresentation. An intentionally deceptive act by the borrower speaks to the borrower's intent. Underwriter negligence is unintentional conduct and its significance to a materiality analysis is context specific.

It appears that, at some point in time, Holt or the persons working under his direction attempted to make a nuanced, breach-by-breach analysis of whether each individual breach materially and adversely affected the interests of the Certificateholders. As the Court has already noted, the "loan files" submitted by the Trusts as to Loan 1456451 contained a written analysis by an individual named Natalie Cohen, which reviewed whether three breaches—based on the misrepresentation of income, unreasonable stated income and the resulting DTI breaches—materially and adversely affected the interests of the Certificateholders. (PX L792 at 011-12.) In each instance, she concluded that they did not, and instead attributed any resulting loss to the "severe financial downturn." (Id. )

Such conclusions were not reflected in the summary of opinions that Holt submitted as Appendix 1 to his direct testimony, however. Instead, in his direct testimony, Holt attempted to explain his methodology for determining whether a given breach had a material and adverse effect on the Certificateholders. These explanations are sometimes muddled and contradictory, with no cogent explanation of how he identified a breach versus how he evaluated the effect of the breach on Certificateholders.

Broadly, the Court understands Holt to be stating that his analysis was directed to only those breaches that he considered to present a heightened risk of loss. When he identified such a breach, he classified it as material and adverse to Certificateholders' interests unless compensating factors showed otherwise. Whether a breach materially and adversely affected the interests of the Certificateholders was largely incorporated into Holt's choice of which items of information to analyze.

When questioned at trial, Holt's testimony contradicted important parts of his direct testimony. Holt testified that he never performed an analysis as to whether a breach materially increased the risk of loss to the lender or Certificateholders:

THE COURT. Let me understand that you did not, you were not asked to, and you did not do an analysis of whether a defect materially increased the risk of loss to the lender, is that correct, or the certificateholder in this case?

A. I did not do any analysis, no, sir.

THE COURT. Right. Okay. You weren't asked to?

A. Just to verify the information, yes.

THE COURT. That would be true across the board as to all of the underwriting files you reviewed?

A. Yes.

(Tr. 928.) Holt later testified that in reviewing for a breach of the Guidelines Warranty, he did not consider whether the breach materially and adversely affected the Certificateholders' interests:

THE COURT. Now I want you to assume hypothetically that that is the borrower's actual income. How would the breach of that underwriting guideline materially and adversely affect anyone, if at all?

A. Because the lender was not assessing the risk going through the determination capacity, character and collateral based on the guideline as stated to do that test to prove reasonableness, and because of that, that's why I would determine a guideline breach.

THE COURT. I understand that, there is a breach, but you don't have an opinion, or do you, on whether or not the breach which you're testifying about has

any adverse effect on investors if, for example, it was the individual's income?

A. That's correct, just the violation.

THE COURT. Just the violation? Okay, thank you.

(Tr. 1192.)

When questioned by the Court as to how he went about identifying whether a breach was material, Holt confirmed that he was "looking at the significance or importance of the breach ...." (Tr. 1001.) Outside the presence of the witness, the Trusts' trial counsel argued that Holt had not understood the question, and that Holt had, in fact, reviewed the loans for increased risk of loss, and that he would further inquire to that effect. (Tr. 1002-05.) Though counsel thereafter questioned Holt in extensive detail about his methods for identifying the existence of a breach, there was no additional testimony as to how a breach affected the interests of the Certificateholders.

The Trusts argue that "Holt appeared to be confused" by the Court's questions. (Pl. FF ¶ 95.) As the trier of fact, the Court finds that Holt was not confused by the Court's questions. The Trusts had ample opportunity to question Holt further if they thought his testimony was inaccurate or confused. The Court finds as a fact that Holt looked at the materiality of the breach but made no systematic and separate assessment of whether the breach had affected the interests of the Certificateholders at the time the cure or repurchase obligation was triggered. The absence of such analysis is most apparent in Appendix 1 to Holt's direct testimony, which purports to identify and explain each breach on a loan-by-loan basis but omits any discussion of the resulting effect on the Certificateholders.

Holt also contradicts himself when describing how he assessed the materiality of a breach of the Guidelines Warranty. At one point he indicates that all loans written in violation of underwriting guidelines are materially defective. He states that "Loans originated in violation of underwriting guidelines are Materially Defective." (Holt Direct ¶ 8.) This blanket conclusion treats any failure to comply with underwriting guidelines as having a material and adverse effect on the Certificateholders.

But Holt later paints a more nuanced picture. He testified that a violation of the guidelines was material if it affects "one of the 3Cs"—the "3Cs" being "capacity risk," "credit or character risk" and "collateral risk." (Holt Direct ¶¶ 141, 40.) He defines capacity risk as the ability to repay a loan, credit or character risk as a borrower's willingness to repay a loan, and collateral risk as the value of the property and the borrower's equity in that property. (Holt Direct ¶ 40.) "All of the Guideline Violation Defects I discuss in this declaration are ones that meet these criteria and as such are Material unless they are offset by compensating factors." (Holt Direct ¶ 141.)

In describing a breach of the MLS Warranty, Holt uses the term "data defect." Holt states that when he reviewed the MLS for data defects, he "focused on the characteristics that are strongly associated with risk .... I found a Material Data Defect only where my review of the Loan established that, as a result of the Data Defect, the Loan was significantly riskier than represented on the MLS." (Holt Direct ¶ 134.) Holt states that, "to be conservative," he then considered compensating factors that might offset the overall risk of a loan. (Holt Direct ¶¶ 136, 186.) He opines that in assessing the materiality of an MLS breach, "the question should be whether, as a result of the Data Defect, the loan was significantly riskier than represented on the MLS," such that "investors were materially misled as to the loan's risk profile by the data on the MLS." (Holt Direct ¶ 136.)According to Holt, compensating factors cannot decrease risk if the compensating factors were also shown in the MLS. (Holt Direct ¶ 136.) Holt states that the purpose of the MLS is to permit the loan's purchasers to reliably assess the loan's risk profile. (Holt Direct ¶ 136.) He states, "If a DTI misrepresentation on the MLS made the loan significantly riskier than represented, that conclusion is not changed by pointing to a low LTV or high FICO that was accurately represented on the MLS." (Holt Direct ¶ 136.) Thus, in Holt's view, for a compensating factor to offset the risk associated with the MLS breach, that compensating factor must be undisclosed to the loan's purchasers.

Holt described his process for weighing the compensating factors that tend to offset an MLS breach. His testimony relied heavily on the reports of outside vendors who reviewed loan files at his direction:

Based on the Vendor Reports, I was able to determine which Loans were Materially Defective. Specifically, where the Vendor Report indicated that a Loan had no defects, I did not find that the Loan was Materially Defective. Similarly, where the Vendor Report indicated that the Loan had defects that would otherwise have been Material Defects, but also had valid compensating factors that I determined were sufficient to offset the increased risk associated with the defects, I did not find that the Loan was Materially Defective. On the other hand, where the Vendor Report indicated that the Loan had Material Defects and no compensating factors, or insufficient compensating factors, I concluded that the Loan was Materially Defective. For each Loan I reached my own conclusion, based on my own judgment and experience, as to whether the Loan was Materially Defective.

(Holt Direct ¶ 191.) Holt afforded weight to specific compensating factors, including whether a DTI or LTV were significantly below an Originator's guideline limits, whether a FICO score significantly exceeded a guidelines minimum and was above 700, whether a borrower had significant assets or reserves and whether a borrower made a down payment of 20 percent or more. (Holt Direct ¶ 189.) Holt generally, although not always, relied on vendor reports to identify material defects:

In most cases, the Vendor Reports were more than sufficient to enable me to determine whether the Loans were Materially Defective. However, where I felt that manual review of the loan file was necessary to make a final determination as to the status of a Loan, I conducted such a review with the help of my team of reunderwriters at Analytic Focus. Prior to filing my Report, I conducted such a manual review of approximately 3,400 Loans.

(Holt Direct ¶ 193.)

Holt describes breaches of representations and warranties other than breaches of the Guidelines Warranty and the MLS Warranty as "other defects." (Holt Direct ¶¶ 142-43.) For example, item (xviii) of Schedule II warrants that each loan "is insured" against "loss by fire" and has a flood insurance policy, if necessary. According to Holt, a breach of these representations and warranties by failing to have fire or flood coverage would affect the risks associated with the loans. (Holt Direct ¶ 142.) As with the MLS Warranty, when Holt identified a breach, he also reviewed compensating factors that might offset increased risk, provided that those compensating factors had not been disclosed to investors and factored into the loans' risk profile. (Holt Direct ¶ 143.)

The Court has reviewed certain individual loans and loan files to assess whether, based on the opinions offered by Holt, the Trusts have proved that a breach materially and adversely affected the interests of the Certificateholders at the time of discovery or notice. As noted, the Court accepts Holt's conclusion that proven borrower deceit materially and adversely affects the interests of the Certificateholders. But in other instances, the Court has examined the nature of the breach in the context of the total mix of information to determine whether Holt's opinion that the breach materially and adversely affected the interests of the Certificateholders should be accepted.

THE "MISSING" INDYMAC FILES.

As of the Closing Date of the Trusts, there were 5,982 IndyMac files among the three Trusts. (Barnett Dec. ¶ 18.) IndyMac filed for Chapter 7 bankruptcy protection in 2008.

According to the Trusts, the loan files are in the hands of OneWest, FSB ("OneWest"), the entity that assumed the loan servicing obligations of IndyMac. During the initial discovery period in this action (Tr. 1240-41), the lawyers for the Trusts and the lawyers for OneWest agreed that OneWest would produce to the Trusts a requested sample of loans—422 loan files, or about 7% of all IndyMac loans; the Trusts and OneWest further agreed that the Trusts would not seek additional loan files from OneWest. (Tr. 1236.)

No court precluded the Trusts from demanding more loan files. The Trusts never sought relief from any court concerning its private agreement and, indeed, the agreement has not been produced to the Court for inspection. There is no claim that the IndyMac files have been destroyed or impaired. Presumably, they remain securely in the hands of OneWest.

Nelson R. Lipshutz, Ph.D., one of the Trusts testifying experts, randomly selected 422 of the 5,982 IndyMac loans, consisting of 168 IndyMac loans from the 2006-OA2 Trust, 85 IndyMac loans from the 2007-1 Trust and 169 IndyMac loans from the 2007-3 Trust. (Tr. 1226-27.) The files were then produced by OneWest. Holt examined the 422 loans and, where Holt found breaches, Lipshutz used Holt's figure to extrapolate over the balance of the universe of the 5,982 IndyMac loans. Lipshutz has no opinion on the validity of the breaches in the 422 IndyMac loans; his work is entirely dependent upon the rate of breaches that Holt opined existed in the IndyMac loans. (Tr. 1229.) As Lipshutz testified: "The results that I have for breach rates are wholly dependent on Mr. Holt's results. If they change, my results change." (Id.)

Lipshutz has adjusted his opinions for some of the changes in Holt's opinions. He now opines that there is a 95% chance that there is a breach rate between 83.79% and 90.16% for the IndyMac loans. Using accepted statistical methods for expressing the range in a single number, he opines that one or more warranties were breached in 87.2% of the IndyMac loans.

There is a fundamental problem with statistical extrapolation from the work of Holt. The Court has not accepted Holt's opinions across the board in finding breaches of warranties. Some of Holt's opinions have been accepted by the Court and some have not.

The Court, for example, has held that a proven intentional misrepresentation by a borrower inherently increases the risk of loss to the Certificateholder. Lipshutz extrapolates (based on Holt's work) that 22.04% of the IndyMac loans without files have borrower fraud-based breaches of the MLS Warranty. (Revised Direct Testimony of Lipshutz at Ex. 6.) But, even as to this 22.04%, there are a multitude of types of potential fraud-based MLS breaches (owner occupancy, FICO scores, undisclosed debt, inflated income) and a multitude of ways of proving the fraud (MERS, income tax returns, bankruptcy filings). The Court has explained why some categories of documents, such as bankruptcy filings, must be treated with caution and cannot be accepted at face value. Also, Holt found that 76.78% of the 422 loans had missing documents, thereby establishing a breach; but the Court often has found Holt's opinions in this category to be either unsubstantiated (e.g. lack of proof of mortgage payment for the last month prior to origination) or without a proven material effect on the Certificateholders' interests. Holt's unfounded opinions fatally infect Lipshutz's work.

Separately, the Court cannot determine whether the Trusts have proved that UBS received notice or otherwise discovered that a loan was in breach unless the loan is identified. This Court has rejected the Trusts "pervasive breach" argument for reasons explained in prior opinions. This Court has concluded that consistent with the New York rule expressed in Nomura , the August 2015 expert reports may be vehicle through which UBS is placed on notice of a loan alleged to be in breach. But a blanket statement by Lipshutz or Holt that defective loans exist among the pool of 5,982 loans without identifying the defective loans is no notice at all. It reads the notice requirement out of the PSA and suffers from the same flaws as the rejected "pervasive breach" theory.

Further, only equitable relief, and not a money damages remedy, is available for non-liquidated loans, and the Court has no way of knowing whether an IndyMac loan without files has been liquidated or not. The Court declines the Trusts' invitation to fashion a new remedy by disallowing relief to the Trusts on any loan later shown not to have been liquidated, and limiting relief to only liquidated loans. (Tr. 2084.) The Trusts have not offered the Court a reliable means to distinguish between the two categories. The cure, replace or repurchase remedies were addressed to existing, outstanding loans. The Court, at the Trusts' urging, has allowed the money damage equivalent of the repurchase remedy if a loan has been liquidated. Accepting the Trusts' latest argument would turn the parties' agreement on its head and make the exception the rule.

Finally, the Trusts have not been left without a remedy for many of the IndyMac loans for which they have not obtained loan files. A MLS Warranty breach, for example, may be proven without reviewing a loan file through reliance upon third-party sources. Indeed, Holt examined the MLS data for over 4,094 additional IndyMac loans (above the 422 loans for which he had files) and compared them with third-party information (e.g. bankruptcy filings, MERS data, FICO scores). Holt opined that there were MLS warranty breaches as to about 1,500 additional loans. (Tr. 1237.) If UBS was placed on notice of those breaches, and the evidence of breach has been placed before the Court such that UBS had a fair opportunity to respond at trial, then the Trusts may pursue a claim that the MLS Warranty was breached based on that evidence in the trial record.

Statistical sampling is an important tool for proving or disproving a case, particularly one involving a large pool of data. It is used in a wide range of actions in federal court. But, in the specific context of the IndyMac loan files in the hands of OneWest, the Trusts have not proved that Holt's opinions or a subset thereof may be reliably applied across the universe of IndyMac loans to establish a breaches that materially and adversely affect the interests of the Certificateholders.

EXEMPLAR LOANS.

Certain loans were selected by the Trusts and UBS for use in cross-examination of the opposing side's witnesses and/or redirecting their own witnesses. Because testimony was elicited at trial concerning these loans, they provide the Court with the ability to make exemplar rulings that then may be applied to different loans.

The parties did not represent to the Court that the loans used in cross-examination or redirect were representative of all loans. Indeed, a loan may have been selected to illustrate a single, isolated point. There is no basis to conclude that the exemplar loans discussed below are representative of the universe of loans.

Much of the live examinations of Holt and Grissom focused on an exhibit, offered by the Trusts and introduced through Holt, which summarized and updated Holt's opinions concerning all individual loans based upon his review (and his vendors' review) of the MLS, the loan file, the underwriting guidelines, third-party sources and other relevant data pertaining to the loans. The Trusts supplemented the exhibit with Grissom's opinions concerning the same loans based on her review or that of her vendors. Finally, the Trusts inputted any reply or rebuttal to Grissom from Holt. That exhibit was marked as Appendix 1 to Holt's direct testimony and was received as a summary of opinion evidence, subject to the objection of UBS. (Tr. 748-49.) It is identified in the record as PX 1103, Appendix 1.

Holt's opinions speak to facts that may support a finding that there was a breach of warranty as to a specific loan. He offers no loan-by-loan opinion on whether the breaches materially and adversely affected the interests of the Certificateholders.

Appendix 1 was submitted to the Court in the form of an Excel spreadsheet which, if printed in hard copy, would total 203,400 pages in length. At the Court's request, the Trusts also submitted in hard copy those portions of Appendix 1 that were used in the cross-examinations of Grissom and Holt, consisting of the summaries of Holt's and Grissom's opinions as to each purported breach. Unless otherwise noted, all discussion of the exemplar loans is drawn from the contents of Appendix 1.

At trial, the Court advised counsel that it intended to review these exemplar loans as part of its findings of fact. (Tr. 1652 ("THE COURT: It may very well be that my findings of fact will have exemplars, so, for example, I think it is Exhibit B to the Holt declaration, the 25,000 page exhibit which was the subject of extensive discussion in the examination of Ms. Grissom and also in the examination of Mr. Holt. It seems to me that I may in the course of my ruling use some of the exemplars that were used and cross-examination or examination to illustrate my rulings, and then it is going to be up to the parties to apply those illustrative rulings to the balance of the loan portfolio.").) Counsel to the Trusts described this as "actually a very sensible approach" and stated in his summation that "rulings on exemplars would be very helpful ...." (Tr. 1652, 2044.) The Court notes that in their proposed findings of fact and conclusions of law, neither the Trusts nor UBS discussed the exemplar loans. They offered no proposed analysis as to how the Court should review the experts' opinions on the individual breaches or how these exemplar loans should be assessed in the larger context of the trial.

The Court has considered the totality of the evidence relating to a loan in making findings on any specific issue relating to that loan. The evidence most directly applicable to the claimed breach has not been considered in isolation but in conjunction with the totality of the evidence concerning the loan. For example, evidence of one type of breach, e.g. an owner-occupancy misstatement, has been considered with respect to another breach, e.g. an income misstatement. Thus, a proved misstatement on one point, taken in the total mix of other evidence, may make it somewhat more likely that there was a misstatement on another point. Evidence concerning a borrower's miniscule assets may bolster the conclusion that the borrower's income was unreasonable or misstated. The absence of a specific discussion of all evidence relating to the loan does not mean it was not considered in making a finding on specific breach.

The Court makes the following findings as to the specifically identified loans:

A. Loan 1456451.

Loan 1456451 was a stated-income loan originated by American Brokers Conduit (a division of American Home), and was included in the 2007-1 Trust.³ The loan was for the stated purpose of the borrower's purchase of a second home. The MLS lists an "Original DTI" of 24.13%. The funding date for the subject property was October 6, 2006. For the 2007-1 Trust, the Closing Date was January 16, 2007. (PX 110 at 33, 36.)

3 Appendix 1 lists American Brokers Conduit as the originator of several loans. In the MLS, American Home is listed as the originator of those same loans. The Court takes judicial notice that American Brokers Conduit was a division of American Home, and identifies all subsequent loans originated by American Brokers Conduit as originated by its parent company, American Home.

Holt identified breaches of the MLS warranties as to occupancy, LTV and DTI ratios and violations of American Home's underwriting guidelines as to occupancy and income.

1. Occupancy Misrepresentation.

On her loan application, the borrower, who resided in California, stated that she was purchasing the subject property, a condominium unit located in Florida, as a second home. (PX L00792 at 0005.) According to Holt, documents from the servicing file, i.e. post-origination documents from the files of the entity responsible for monitoring repayment and other issues of loan compliance, included a letter from the borrower stating that the property was an investment property, even though it was listed in the MLS as a second home. (PX 1103 Appx. 1.)

Based on the servicing documentation, Holt opined that the property was an investment home and not a second home, and therefore that the loan was funded even though it exceed the guideline for ratio for a second home. Holt opined that, under the applicable underwriting guidelines, the required LTV ratio for a second home was 90%, with a qualifying credit score of 688, but for an investment property, the LTV ratio was 80%. In other words, if the loan were truly a second home, the institution would consider granting a loan up to 90% of the appraised value, but if it were an investment property, American Home would only lend up to 80% of the appraised value. Here, American Home originated a loan for 90% of the appraised value in the belief that the property was being purchased as a second home.

At trial, Grissom did not dispute that the servicing records stated that the property was an investment property and was not, in fact, used as a second home: "I did not dispute his findings in my response." However, she opined that the underwriters properly underwrote the loan given the borrower's stated intention to occupy the property as a second home. Grissom contended that post-origination servicing records should not be considered in a re-underwriting analysis on the theory that it was not known to the underwriter at the time of origination.

The Court has reviewed the loan files for this loan and the post-origination servicing materials. (PX L792 and PX L10988.) A letter dated July 7, 2010 from the borrower advises that she has three properties in Florida that have remained vacant for 16 months. (PX L 792 at 0010.) Sixteen months prior to the date of the letter would be March 7, 2009. Holt, as an expert, could reasonably conclude that, at least as of March 7, 2009, the 3 properties, including the subject property, were investment properties.

The MLS Warranty is breached if UBS's statement on the MLS, in some material respect, was not true and correct as of the Closing Date of the PSA. A breach of the MLS Warranty turns on the correctness of the information on the MLS and not on what was known or what inquiry should have been made.

There is no apparent dispute between Holt and Grissom that the borrower held the property as an investment property. Grissom focused on the borrower's intention at origination. The question, in terms of the MLS Warranty, is whether the borrower held the property as an investment property as of the Closing Date of the PSA, January 7, 2007. While it is a close question, the Court finds that Holt, an underwriting expert, could fairly conclude from the acquisition of a total of three rental or investment properties in Florida that they were part of an overall strategy to acquire Florida properties for investment, and reflects on the borrower's intent as of the Closing Date and before. The Court concludes that it is more likely than not that the borrower purchased the property as an investment property and not as a second home. The MLS Warranty therefore was breached because the MLS incorrectly listed the property as a second home, a statement that was materially untrue and incorrect.

In contrast to the MLS Warranty, the Guidelines Warranty does not assure the accuracy of the information considered by the underwriter. It simply warrants that the loan was originated in accordance with the Originator's guidelines. Grissom opined that the guidelines were not violated because the "borrower declared her intent to occupy the subject property as a second home...." (Appendix 1.)

But the American Home Desktop Underwriter report for the loan shows that more was required in order to comply with the guidelines:

Verify that this second home property is located a reasonable distance from the borrower's primary residence and is suitable for year-round occupancy. The borrower must occupy the property for some portion of the year and must have exclusive control over the property; therefore, the borrower must not enter into any rental agreements that require the property to be rented or give a management firm control over the property. Rental income may not be used as qualifying income on a second home.

(PX L792 at 14.)

While the guideline does not define "reasonable distance," the Court understands it to mean that the two properties are neither so close nor so far as to make it implausible that one is a second home. A cursory review of the loan application would have disclosed that the primary residence was in California and the second home in Florida; even when the ease of air travel is taken into account, the exceptionally great distance was a red flag that would have prompted either a further inquiry or an "exception" from a higher underwriting authority. There is no indication of a further inquiry or the exercise of an exception. The result of the failure to make inquiry was that the loan was underwritten at a 90% loan to value ratio that is higher than the 80% that would have been the maximum for an investment property.2. Income Misrepresentation with Red Flags.

The borrower's loan application stated that she was a self-employed realtor who had worked as a realtor for 25 years, including with Remax for eight years. She claimed a monthly income of $20,000, i.e. , annual income of $240,000. Holt opines that her income was misstated based upon his reliance upon a 2006 tax return submitted by the borrower during loss-mitigation servicing showed an annual income in 2006 of $51,860.

Grissom disputed Holt's conclusion that the borrower's income was misrepresented because, she asserts, the borrower's 2006 tax returns were not available to the underwriters at origination. Grissom opined that compensating factors justified the loan.

The Court finds that the Guidelines Warranty was breached because there is no indication that the underwriter performed a reasonableness assessment on the stated income of a realtor earning $240,000 per annum. Implicit in Holt's opinion is that this level of income from a self-employed realtor should have prompted a further inquiry as to reasonableness. The Court finds that it is more likely than not that there was a breach of the Guidelines Warranty in failing to inquire further as to reasonableness of income.

Apart from the failure to conduct a reasonableness inquiry, the actual income of the borrower, in Holt's opinion, was that reported on the income tax return for 2006, $51,860. This was, based upon Holt's experience, a misstatement of the borrower's income.

Holt had a valid basis for opining that the loan was originated and funded in violation of the maximum guideline DTI ratio of 45%. Holt had a valid basis to opine that the DTI ratio was 110.35%, utilizing the figure in the 2006 tax return.

The borrower's stated DTI ratio was listed in the MLS as 24.13%, but, in Holt's opinion, the actual DTI ratio was 110.35%. The Court finds that the MLS Warranty was breached because the income was misstated also breached because it is more likely than not that the DTI ratio was materially misstated as of the Closing Date of the PSAs.

3. Material and Adverse Effect at the Time of Discovery or Notice.

To recap, there was a breach of the Guidelines Warranty because: (1) there was not an adequate assessment of reasonableness of income; (2) the DTI ratio was improperly calculated based on income that was inflated; and (3) the wrong LTV was applied because the underwriter ignored the red flag of the geographic distance between the primary residence and the purportedly second home. The MLS Warranty was breached because the MLS incorrectly listed the property as a second home and the DTI ratio was incorrectly stated.

The loan file offered by the Trusts in evidence contains a written critique and analysis of the underwriter's adherence to the guidelines and the effect of any breach; it concludes that the breaches did not materially and adversely affected the interests of the Certificateholders, noting that the borrower made 23 timely payments before defaulting. (PX L00792 at 10-11.) Notably, neither Holt nor Grissom separately addresses whether any particular breach materially and adversely affected the interests of the Certificateholders at the time of discovery or notice.

With due consideration of the total mix of information, including the opinions of Holt and Grissom, the Court finds that the established breaches of the Guidelines Warranty and the MLS Warranty materially and adversely affected the interests of the Certificateholders at the time of discovery or notice. This is because an intentional misrepresentation of income and of an intention to occupy the property as a second home goes to the borrower's commitment to repay the loan. A borrower is more likely to walk away from investment property than from a second home. Also, a lie about income calls into the question an intention to repay. The 2007-1 Trust has established that this loan would not have been originated or funded on the existing terms had the truth been known.

Because a reasonableness assessment would have revealed that the income was overstated, the guidelines breach also materially and adversely affected the Certificateholders because compliance with the guidelines would have revealed an income in excess of the Guideline DTI. The loan would not have been funded on the existing terms.

The 2007-1 Trust has proved each element of its claim against UBS with respect to Loan 1456451 and UBS is obligated to repurchase the loan or, if it has been liquidated, then to pay the Trust the money damage equivalent.

B. Loan 40577613.

Loan 40577613 was a stated-income loan originated by MortgageIT, and was included in the 2006-OA2 Trust. The loan was made for the purpose of a cash-out refinance of an owner-occupied property. The borrower's original DTI ratio is not separately listed in Appendix 1, but the "Back Ratio" listed on the MLS is 44.57%. (See PX O01, line 4427, col. CA.) The funding date for the subject property was June 7, 2006. The Closing Date for the 2006-OA2 Trust was November 15, 2006. (PX 49 at 23, 27.)

Holt concluded that the loan included multiple defects, including income misrepresentation, misstated DTI ratio, missing or inadequate title insurance, "employment violation," "housing history violation" and the existence of an undisclosed mortgage held by the borrower.

1. Income Misrepresentation.

Holt opined that the borrower's stated monthly earnings of $6,750, or $81,000 annually, for the position of staff nurse were misstated in light of the borrower's 2006 Form 1040, which was provided during loss-mitigation servicing. The Form 1040 reflected a 2006 monthly income of approximately $3,700, or $44,000 annually.

Holt opined that the borrower's misrepresentation of income and undisclosed mortgages altered the loan's DTI ratio. Holt opines that the 26.05% DTI ratio listed in the MLS was, in reality, 251.84%, nearly ten times higher.

The Court has been unable to discern how Holt identified a DTI ratio of 26.05% listed in the MLS. In the MLS submitted at trial, the borrower's "Front Ratio" is listed as 44.38% and the "Back Ratio" is listed as 44.57%. (See PX O01, line 4427, cols. BZ, CA.) Elsewhere, Holt notes that the underwriter calculated a DTI ratio of 44.57%, which is consistent with the MLS's listing for "Back Ratio." The 26.05% figure may be an error on Holt's part, or perhaps it is drawn from a source other than the 2006-OA2 MLS submitted as PX O01. In this particular instance, however, the seemingly erroneous MLS data cited by Holt is of no significance because the recalculated DTI ratio far exceeds either figure.

The Court finds that the 2006-OA2 Trust has proved that it is more likely than not that UBS breached the MLS Warranty by stating a DTI ratio for this loan as of the Closing Date of the PSAs that was materially untrue and incorrect.

With respect to the Guidelines Warranty, the guidelines required a reasonableness assessment, and there is no indication in the loan file that such an assessment was made. A reasonableness assessment would likely have led to a further investigation and either a withdrawal of the application or the underwriter learning the borrower's true income.

Given the difference between the borrower's stated income and the income reported on the borrower's Form 1040, the Court concludes that Holt's opinion that the borrower misrepresented income at the time of underwriting was well founded. The maximum permissible guidelines DTI ratio for this loan was 38%, but the true ratio was 251.84%. The Trust has proved that it was more likely than not that the underwriters made no assessment of the reasonableness of income, and there is no evidence in the file of a weighing of any compensating factors as of the time of origination.

2. The Presence of Undisclosed Mortgages.

Using MERS, Holt identified three additional undisclosed mortgages held by the borrower at the time of origination. One of the borrower's other properties had a first mortgage of $472,000 and a second mortgage of $118,000. Holt notes that the borrower's purchase of this property occurred within 30 days of the subject property's funding date. The borrower had an additional property with a mortgage of $302,000, and Holt opines that the purchase of this property occurred prior to the subject property's funding date. Holt did not identify any red flags associated with these undisclosed mortgages.

Grissom contended that compensating factors justified the origination of this loan, including a credit score of 673 and reserves of approximately $12,000. There is no indication that the underwriters weighed these compensating factors at the time of origination or that there could have been a reasonable exercise of an exception in light of these undisclosed mortgages.

While it is undisputed that the borrower had additional undisclosed mortgages, which likely would have caused the DTI ratio to exceed the maximum DTI ratio of 38% allowed under the guidelines, Holt has not identified any failure on the part of the underwriter. For example, he does not opine that, without the presence of a red flag, the guidelines required the underwriter to search for these mortgages using MERS or other comparable databases. In his direct testimony, Holt states that MERS was a resource available to underwriters at the time the disputed loans were originated (Holt Direct ¶ 132), but he does not opine that a reasonable underwriter would, as a matter of course, perform a MERS search to locate additional loans. He also does not opine that the borrower acted fraudulently in not disclosing these post-origination mortgages. Grissom noted that the borrower's credit report reflects only one credit inquiry within 90 days of the subject transaction, and that the underwriter requested a letter of explanation as to the inquiry. (See PX L8190 at 0042 (credit inquiry by Equidata).)

Because Holt did not cite to any violation of the guidelines or opine that the underwriter acted unreasonably in not independently discovering the additional mortgage debt, the Trusts have not proved that it is more likely than not that the underwriter breached the Guidelines Warranty.

However, while the 2006-OA2 Trust has not proved a breach of the Guidelines Warranty, the Court finds that it is more likely than not that UBS breached the MLS Warranty because the debt used to calculate the DTI ratio was not correct. The Trust has proved that the DTI ratio listed in the MLS was not true and correct in all material respects as of the Closing Date. Using the undisclosed mortgage debt, Holt calculates that the borrower's actual DTI ratio was 251.84%, which far exceeds the 44.57% DTI ratio listed in the MLS.

3. Missing Title Insurance.

Holt opined that the loan files did not contain a title insurance policy. For the reasons discussed above, the Court concludes that the Trust has not proved a violation of the Title Insurance Warranty.

4. Employment Verification.

Holt opined that the lender did not receive a verification of employment for the borrower, which he states violated the underwriting guidelines.

Grissom identified a verification of employment in the loan file. Holt acknowledged that Grissom located the borrower's employment verification, but stated that "this document was not present at the time of the initial audit reunderwriting review." During cross-examination at trial, Holt testified that he could not recall whether this breach had been withdrawn, but stated that the presence of other defects could still give rise to a breach. (Tr. 809-13.) Holt has not disputed that Grissom located a verification of employment for this borrower.

The Court finds that the 2006-OA2 Trust has not proved a breach of the Guidelines Warranty as to failure to verify employment.

5. Housing History Violation.

Holt opined that the loan file did not contain evidence of mortgage payments on the existing mortgage for three of the preceding twelve months, as required by the applicable underwriting guidelines. The record does not indicate whether the borrower in fact missed or was delinquent in any of his mortgage payments. Grissom did not dispute that the loan file was missing payment history documents, but opines that there were additional compensating factors, including the borrower's credit score, reserves and perfect payment history on real-estate credit lines, which may have made the loan's approval compliant with the guidelines.

The Trust has proved that it is more likely than not that a payment history was not reviewed and placed in the loan file in the course of origination of the loan. The failure to obtain a payment history was a violation of the guidelines and hence a breach of the Guidelines Warranty.

6. Material and Adverse Effect at the Time of Discovery or Notice.

To recap, the 2006-OA2 Trust has proved that that UBS breached the MLS Warranty by stating a DTI ratio for this loan as of the Closing Date of the PSAs that was materially untrue and incorrect. The Trust also has proved that there was a Guidelines Warranty breach because it was more likely than not that the underwriters made no assessment of the reasonableness of income or obtain a payment history.

The Court finds that the 2006-OA2 Trust has proved that the breach of the MLS Warranty in misstating the DTI ratio materially and adversely affected the interests of the Certificateholders at the time of discovery or notice. The Trust has established that this breach materially and adversely affected the interests of the Certificateholders because the Trust would have rejected the inclusion of the loan or would have priced the transaction differently.

In addition, the Court finds that the borrower intentionally misrepresented his income. This reflects on the borrower's willingness and ability to repay the loan. The Trust has established that this loan would not have been originated or funded on the existing terms had the truth been known.

The Trust has not proven that the failure to obtain a payment history materially and adversely affected the interests of the Certificateholders because there has been no showing of what such a history would have disclosed.

The 2006 OA-2 Trust has proved that UBS is obligated to repurchase Loan 40577613 or, if it has been liquidated, then to pay the money damage equivalent.

C. Loan 40599698.

Loan 40599698 was a stated-income loan originated by MortgageIT, and was included in the 2006-OA2 Trust. The loan was made for the purpose of a cash-out refinance of an owner-occupied home. The borrower's original DTI ratio is not listed in Appendix 1, but the MLS lists a "Front Ratio" of 9.44% and a "Back Ratio" of 32.58%. (See PXO01, line 4722, cols. BZ, CA.) The funding date for the subject property was May 31, 2006. The Closing Date for the 2006-OA2 Trust was November 15, 2006. (PX 49 at 23, 27.)

Holt concluded that the loan had multiple defects, including occupancy misrepresentations, unreasonable stated income, an undisclosed mortgage, income misrepresentation with red flags, missing title insurance, ineligibility under the guidelines and inaccurate LTV ratio and DTI ratio determinations.

1. Misrepresented Income.

The borrower's loan application stated that the borrower worked as a Contact Representative in the Social Security Administration and earned $14,000 each month or $168,000 per year. Based on Holt's review of the borrower's 2006 and 2007 tax returns, which were included in a loss-mitigation servicing file, he opined that the borrower actually earned $2,052 each month, or an annual income of $24,634. For the 2006-OA2 Trust, the Closing Date is November 15, 2006. (PX 49 at 23, 27.) Thus, a review of the 2006 income tax return properly covered the period covered by the MLS warranty.

The intentional understatement of the borrower's income by a material amount necessarily resulted in a breach of the MLS Warranty because the stated DTI ratio was untrue and incorrect in material respects.

The underwriting guidelines required that the borrower's stated income be reasonable and consistent with occupation. According to Holt, the borrower's loan file included an employment verification stating that the borrower worked as a Contact Representative in the Social Security Administration. Holt opined that the borrower's profile did not support a reasonable stated income of $14,000 a month, and that the loan file contains no indication that the underwriters attempted to assess the reasonableness of the stated income. Grissom cited to compensating factors in the loan file, but there is no written evidence that they were considered at the time, and the Court affords them no weight.

It is not self-evident that BLS data for a "Customer Service Representative," as relied upon by Holt, is relevant to an employee of the Social Security Administration. But a salary for a Social Security Contact Representative at the borrower's stated level would raise a red flag. In 2006, according to the federal government's widely-published general scale of pay, a GS-15 (the highest grade), at step 10 (the highest step), working in New York or San Francisco (cities with the highest pay differentials), would only earn $143,000 per year. (https://archive.opm.gov/oca/06tables/txt/gstbls.txt) The title Contact Representative is not suggestive of one who is either part of the Senior Executive Service of the federal government or a Presidential appointee. The claimed salary would be sufficiently high to require further inquiry.

Considering the mix of information and the opinions of the experts, the Court finds a breach of the Guidelines Warranty based on the underwriter's failure to conduct an inquiry as to the reasonableness of income.

2. Undisclosed Mortgage.

Holt opined that the borrower did not disclose an existing mortgage, which he located utilizing MERS. According to Holt, the borrower had previously opened a first mortgage on the subject property in an amount of $198,000. Holt states: "The recalculated [DTI], including the mortgage payment for the undisclosed mortgage, which closed prior to the subject loan transaction, was 33.33%."

Under the guidelines, the maximum allowable DTI ratio was 38%. Holt opined that a prudent underwriter should have discovered this additional mortgage because the borrower's credit report contained red flags in the form of credit inquiries made prior to the closing date thereby prompting further review.

The Court accepts Holt's opinion and concludes that the existence of this additional undisclosed mortgage caused the DTI ratio in the MLS to be untrue and incorrect. The Trust has proved a breach of the MLS Warranty.

Because there was a red flag in the form of credit inquiries, a reasonable underwriter would have inquired further and identified the borrower's additional mortgage debt. There is no indication that the underwriter did so at the time of origination. The 2006-OA2 Trust has therefore also proved a breach of the Guidelines Warranty.

3. Resulting DTI Breaches.

Holt opined that when the borrower's actual, lower income and the undisclosed mortgage are taken into account, the borrower's DTI ratio jumps from the DTI ratio originally disclosed on the MLS to 232.93%. The Court accepts Holt's opinion on this point.

The 2006-OA2 Trust has proved a breach of the MLS Warranty relating to the DTI ratio because the DTI has been proved to be untrue and not correct as of the Closing Date.

Because an underwriter should have but did not conduct a reasonableness of income inquiry and should have checked MERS for other mortgages, the Guidelines Warranty was breached because the maximum DTI ratio permitted under the guideline was 38%.

4. LTV Breaches.

Holt states that Cowan's AVM analysis establishes that the loan violated the guidelines based on misstated LTV data. For the reasons previously explained, the appraised value is an opinion which has not been shown to be other than honestly held. Moreover, the Court has rejected exclusive reliance upon an AVM analysis as a basis for recalculating LTV ratio.

5. Occupancy Misrepresentation.

As noted, the borrower applied for the loan as a refinancing for an owner-occupied single-family property under a stated-income documentation program. Holt concluded that the borrower did not reside at this property based on (1) 2010 filings in Chapter 7 bankruptcy proceedings and (2) the 2006 and 2007 tax returns contained in the borrower's servicing records. Grissom counters that the borrower submitted a signed occupancy affidavit and a final loan application stating that he resided at the subject property.

Here, the statement in the bankruptcy filing is consistent with the borrower's 2006 and 2007 tax return and supports a finding that it was more likely than not that the subject property was not the borrower's residence at the Closing Date of the PSAs. The Court finds that that the MLS Warranty regarding occupancy was, in material respects, not true and correct as of the Closing Date. The nature of the borrower's misrepresentation regarding occupancy, in an existing home that the borrower was refinancing, is such that it bespeaks of an intentional misrepresentation.

While the borrower's statements regarding occupancy was false, the Trust has not explained why or how the guidelines would have required the underwriter to have engaged in a more probing inquiry as to the borrower's occupancy. Thus, the Trust has proved a breach of the MLS Warranty, but not a breach as to the Guidelines Warranty as to the guidelines' owner-occupancy standard.

6. Missing Title Insurance.

Holt concludes that UBS violated the Title Insurance Warranty because the loan files did not contain a title insurance policy. For the reasons already discussed, the Court concludes that the Trust has not proved a violation of the Title Insurance Warranty. Moreover, Grissom stated that she located the title insurance policy upon her review of the loan file, and Holt does not address the purportedly missing title insurance in reply.

7. Material and Adverse Effect at the Time of Discovery or Notice.

The Court finds that the 2006-OA2 Trust has proved that the breach of the MLS warranty in misstating the DTI ratio and in misstating occupancy materially and adversely affected the interests of the Certificateholders at the time of discovery or notice. The nature of the borrower's false statements about his own income, other outstanding debt and his occupancy of the subject property reflect on the borrower's willingness and ability to repay the loan. The Trust has established that this loan would not have been originated or funded on the existing terms had the truth been known.

This Court has found a breach of the Guidelines Warranty based on the underwriter's failure to review the reasonableness of the borrower's stated income and to follow up on red flags to the existence of additional mortgage debt. This breach would have materially and adversely affected the interests of the Certificateholders at the date of discovery or notice because a reasonableness check and further inquiry as to additional mortgage debt would have revealed information that placed the loan at a level above the DTI ratio maximum permitted by the guidelines. The loan would not have been approved, or else would not have been approved on the same terms had the guidelines been applied properly.

The 2006-OA2 Trust has therefore proved that UBS is obligated to repurchase Loan 40599698 or, if it has been liquidated, then to pay the money damage equivalent.

D. Loan 40588193.

Loan 40588193 was a stated-income loan originated by MortgageIT, and was included in the 2006-OA2 Trust. The loan was made for the purpose of a cash-out refinance on an owner-occupied home. The borrower's original DTI is not separately listed in Appendix 1, but the MLS lists a "Front Ratio" of 13.77% and a "Back Ratio" of 34.21%. (See PXO01, line 4405, cols. BZ, CA.) The funding date for the subject property was June 1, 2006. The Closing Date for the 2006-OA2 Trust was November 15, 2006. (PX 49 at 23, 27.)

Holt opined that the loan had multiple guidelines defects, including a failure to obtain employment verification, inadequate title insurance, income misrepresentation, housing history violation, unreasonable stated income and DTI ratio violation.

1. Income Misrepresentation and/or Lack of Reasonableness Inquiry.

The borrower's stated monthly income as a nurse for Kaiser Permanente was $10,000, or $120,000 per year. Holt opines that the income was misrepresented and therefore the DTI ratio on the MLS was not correct. While the MLS Warranty is determined as of the Closing Date of the 2006-OA2 Trust, November 15, 2006, the principal evidence that the income was not correct is a W-2 from the borrower reporting an annual income for 2008 of $55,777. Because the 2008 W-2 speaks to a time period well beyond both origination and the Closing Date of the PSA and the nature of employment lent itself to shift work rather than a steady monthly salary, the change in income does not alone prove an inaccurate statement on the MLS.

Holt identifies a red flag because the loan was a cash-out refinance of an owner-occupied property, in which the borrower paid $4,987 at closing to receive only $4,705 in cash on a negative amortization, adjustable-rate mortgage. Holt does not adequately explain why this circumstance would reflect specifically on the borrower's income, as distinguished from some other misstatement or nefarious purpose.

Holt also asserts that the underwriter failed to review the reasonableness of the borrower's stated income of $10,000 per month for the position of a nurse at Kaiser Permanente. According to Holt, BLS date indicates that the 90th percentile of income for registered nurses in 2006 was $95,300 annually. "Utilizing this monthly income, grossed up 125%, the borrower's [DTI] ratio would have been 31.36%." This figure would be within the guidelines maximum DTI ratio of 38%. Also, applying Holt's approach to "grossed up" BLS data, the Court calculates that the borrower's projected annual income could be as high as $119,125, which is slightly below the figure stated by the borrower.

Holt also notes that the borrower had stated assets of only $1,500, though Grissom notes that the borrower had verified assets of $13,219, which Holt does not dispute in his reply.

There is no proof in the record as to the borrower's actual income, other than the borrower's statement that she made $120,000 per annum. Holt's calculation of 125% of the 90th percentile of BLS data suggesting a possible income as high as $119,125 and a W-2 for 2008 (the Closing Date of the PSA was November 15, 2006, a considerably earlier period) showing an income of $55,777.

Taking into account the full mix of information and the opinions of the experts, the Trust has failed to prove that the borrower's income was misstated or that there was a failure to test the reasonableness of income (because a check of BLS data would not have suggested unreasonableness). While the transaction may not have been in the borrower's best economic interest, this does not prove the borrower's misrepresentation of income.

The Trust has not proved a breach of the MLS Warranty relating to the stated DTI ratio because it has not proved that it was materially untrue or incorrect as of the Closing Date. Nor has it proved a breach of the Guidelines Warranty with regard to a failure to inquire further as to the reasonableness of income or violation of the guidelines DTI ratio.

2. Mortgage Payment History.

Holt concluded that there was no documented mortgage payment history for the subject property contained in the loan file. He notes that under the applicable guidelines, the borrower was not permitted to have a record of any late payments within the preceding 12 months. However, Grissom states that the borrower's credit report shows that the borrower had no history of late payments on the property through March 2006, and that additional documentation shows a payment of April 2006. Holt does not dispute Grissom's findings, but notes that the loan closed on June 1, 2006. Thus, Holt appears to assert that the lack of documentation for a May 2006 payment amounts to a guidelines violation.

But Holt's opinion that the guidelines were violated is premised on the requirement that a borrower have no late payments in the preceding 12 months. At most, he has shown that the loan file does not include evidence of a mortgage payment in May 2006, which was the month before the transaction closed. This is not proof of a late payment, and thus does not breach the guidelines requirement that Holt summarizes.

The Trust has failed to prove a breach of the Guidelines Warranty based on the borrower's history of mortgage payments.

3. Verification of Employment.

Holt originally concluded that the borrower failed to obtain an employment history for this borrower, resulting in a guidelines violation. However, Grissom located a verbal verification of employment in the loan file. The employment verification was consistent with the borrower's loan application and met the guidelines requirements. In reply, Holt stated that the document "was not present in the loan files from original audit review," but does not dispute its existence. At trial, Holt stated that this breach claim should have been withdrawn. (Tr. 1197-98.)

Because there is evidence that the borrower's employment was verified, the Trust has failed to prove a breach of the Guidelines Warranty in relation to verification of the borrower's employment.

4. Missing Title Insurance.

Holt asserts that the loan files did not contain a title insurance policy and from this opines that the Title Insurance Warranty was breached. For the reasons already discussed, the Court concludes that the Trust has not proved a violation of the Title Insurance Warranty. Moreover, Grissom stated that she located the title insurance policy upon her review of the loan file, and which Holt acknowledges in reply.

5. No Breach.

Taking into account all evidence relating to the loan and giving due consideration to all actual or potential red flags and the cumulative effect of the purported breaches, the Court finds that the 2006 OA-2 Trust has not proved a breach of any warranty as to Loan 40588193 and, thus, it is unnecessary to address whether material and adverse effect at the time of discovery or notice.

E. Loan 1447951.

Loan 1447951 was a stated-income loan originated by American Home and was included in the 2007-1 Trust. It was a cash-out refinance loan for an owner-occupied home. Appendix 1 lists an "Original DTI" of 27.52%. The funding date for the subject property was October 5, 2006. The Closing Date of the 2007-1 Trust was January 16, 2007. (PX 110 at 33, 36.)

Holt opined that the loan had multiple defects, including unreasonable stated income, employment misrepresentations with red flags, income misrepresentation, incorrect application of LTV ratio to the guidelines, incorrect application of DTI ratio to the guidelines and incorrect DTI ratio listed on the MLS.1. Employment Misrepresentation.

The borrower's final loan application stated that he worked at "Signature Drywall as a Customer Care Specialist for 7 years," but the verification of employment identified his work as "Drywall Detail." Because of this purported discrepancy, Holt opined that the borrower's employment was misrepresented at origination. As Grissom pointed out, the two are not necessarily inconsistent. Holt failed explain why it would not be plausible to assume that a Customer Care Specialist in the drywall business spends much of his time on completing the detail or punch-list work on a drywall job.

Taking account of the mix of information and the opinion of the experts, the Trust has not proved that it is more likely than not that he misrepresented the nature of his employment.

Holt also notes that the borrower filed for bankruptcy on August 15, 2008 and certain of the borrower's 2008 bankruptcy filings list three years of employment at Home Pointe Property Management. Holt states: "A prudent underwriter should have questioned the job title discrepancy and required the borrower to provide full income documentation."

The Court's review of the loan file reveals two bankruptcy filings, but neither was filed in August 2008. (08-22281, U.S. Bankruptcy Court, E.D. Calif (Sacramento) filed Feb. 27, 2008; 08-25048, filed April 21, 2008; PX L10722 at 0457-61.) The bankruptcy filings that form the basis for Holt's opinion have not been flagged as part of the trial record. The Court cannot determine whether Holt's point is that the drywall employment was omitted or whether Home Pointe was listed as an additional employer.

The Court concludes that the Trust has failed to prove that it is more likely than not that the borrower misrepresented the nature of his employment on his loan application.

2. Reasonableness of Stated Income and Income Misrepresentation.

The borrower's loan application stated that he had worked at Signature Drywall as a customer care specialist for seven years, and reported a monthly income of $11,690 or $140,280.

Using BLS data, Holt opined that a customer service representative in the 90th percentile of income would have earned $3,484 per month or $41,808 per year in 2006. Holt states that there is no indication in the loan file that the underwriter attempted to verify the borrower's stated income. Utilizing a "grossed up" calculation of BLS data, Holt calculates that the borrower's DTI ratio would have been 76.93%, under guidelines that required a maximum DTI ratio of 45%.

Grissom notes the presence of compensating factors that would have supported the reasonable exercise of a guidelines exception. However, there is no indication in the loan file that an exception was exercised. She also notes that a "customer service representative" is not necessarily equivalent to the borrower's occupation as a "customer care specialist."

In the absence of additional detail, the Court concludes that it is more likely than not that a "customer service representative" and a "customer care specialist" would have the same duties, and that Holt selected the appropriate BLS employment category.

The 2007-1 Trust has proved that it is more likely than not that the borrower's stated income was unreasonable, and that the underwriter made no inquiry as to the reasonableness of stated income. The borrower's stated income should have prompted the underwriter to take further steps such as to verify income or convert the loan to a full documentation loan. The Trust has further proved that it is more likely than not that an inquiry would have shown a much lower income than stated, which would have caused the borrower's DTI ratio to exceed the guidelines maximum. The Trust, therefore, has proved a breach of the Guidelines Warranty.

The Trust also asserts that there was a misrepresentation of income that affected the DTI ratio reported on the MLS. The Trust relies on BLS data as evidence of income misrepresentation. As discussed, the borrower's unverified stated income fell substantially above the top end of the BLS data for a person with this job title. The Court therefore concludes that it is more likely than not that the borrower misrepresented his income in the application process and that there was a breach of the MLS Warranty because the DTI ratio stated on the MLS was materially untrue and incorrect as of the Closing Date.

3. The LTV Ratio Defects.

Holt states that Cowan's AVM analysis establishes that the loan violated the guidelines based on misstated LTV data. For the reasons explained, the Court rejects the AVM analysis as a basis for recalculating LTV ratio.

4. Material and Adverse Effect at the Time of Discovery or Notice.

The Court finds that the 2007-1 Trust has proved that the breach of the MLS Warranty in misstating the DTI ratio materially and adversely affected the interests of the Certificateholders at the time of discovery or notice. The nature of the borrower's materially false statements about his own income reflects on the borrower's willingness and ability to repay the loan. The 2007-1 Trust has established that this breach materially and adversely affected the interests of the Certificateholders because the Trust would have rejected the inclusion of the loan or would have priced the transaction differently.

This Court has found to have been a breach of the Guidelines Warranty in failing to conduct an inquiry as to the reasonableness of income. This breach materially and adversely affected the interests of the Certificateholders at the date of discovery or notice because a reasonableness check would have revealed lower income, which would have placed the loan at a level above the DTI ratio permitted by the Guidelines. The loan would not have been approved or not approved on the same terms had the Guidelines been applied properly.

The 2007-1 Trust has proved that UBS is obligated to repurchase the Loan 1447951 or, if it has been liquidated, then to pay the money damage equivalent.

F. Loan 40595372.

Loan 40595372 was a stated income/stated assets loan originated by MortgageIT and was included in the 2006-OA2 Trust. The loan was made for the cash-out refinance of an owner-occupied home. Appendix 1 lists an "Original DTI" of 33.33%. The funding date for the subject property was May 25, 2006. The Closing Date of the 2006-OA2 Trust was November 15, 2006. (PX 49 at 23, 27.)

Holt identified three defects in this loan: failure to verify employment, failure to obtain a necessary document and absence of title insurance.

1. Employment Verification.

The borrower claimed that, as the owner of investments, he was self-employed for 3 years, with a stated monthly income of $22,500 or $270,000 per year. Holt states that the relevant guidelines required verification of two years of employment history for a self-employed borrower. Relying on unspecified material from a 2010 bankruptcy petition, Holt concluded that the borrower had been in business for only two years. Inadequately described references to statements in a bankruptcy filings are insufficient alone to establish the length of time that the borrower was in business.

In her rebuttal, Grissom identified a letter verifying that his business had been in existence for two years. Grissom also pointed to a statement from a CPA that verified the borrower's investment income from two partnerships. In reply, Holt merely noted that the CPA letter was not included during his initial review, and does not dispute Grissom's finding.

At trial, Holt testified, "Based on this right here, it looks like the verbal verification has satisfied the employment violation." (Tr. 813.)

The Court concludes that the 2006-OA2 Trust has not proved a breach of the Guidelines Warranty based on the underwriters' failure to verify this borrower's employment.

2. Missing Note.

Holt described one defect as follows:

The following document was not available at the time of this loan review:

Note for the simultaneous Second Mortgage.

A review of this document may change the Auditor's analysis for the loan file.

This is the entirety of Holt's description for this breach. The Court understands Holt to have opined that the loan was in breach of the Guidelines Warranty because it did not contain documentation related to a second mortgage that was issued simultaneous with this one.

In reply, Grissom states that the applicable guidelines do not require documentation of a second lien to be place in the loan file for the first-lien mortgage loan. She also notes that the underwriter had access to the second-lien documents because the second lien was simultaneously being underwritten by the same lender. In reply, Holt cites to no guidelines requirement about second-lien documentation.

The Court finds that the Trust has failed to prove a breach of the Guidelines Warranty. The information provided is too vague and imprecise for the Court to find for the Trust. The Trust cites no relevant guideline requirement concerning the documentation of a second-lien mortgage when a first-lien mortgage is being underwritten.

3. Missing Title Insurance.

Holt concludes that UBS violated the Title Insurance Warranty because the loan files did not contain a title insurance policy. For the reasons already discussed, the Court concludes that the Trust has not proved a violation of the Title Insurance Warranty. Moreover, Grissom stated that she located the title insurance policy upon her review of the loan file. Holt acknowledges that the loan file contained a title policy, but opined that the policy was incomplete. The presence of the title policy in the loan file also is indication that a title insurance policy was obtained prior to the funding of the loan, and therefore does not support a breach of the Guidelines Warranty.

4. No Breach.

Taking into account all evidence relating to the loan and giving due consideration to all actual or potential red flags and the cumulative effect of the purported breaches, the Court finds that the 2006 OA-2 Trust has not proved a breach of any warranty as to Loan 40595372 and, thus, it is unnecessary to address whether any breach had a material and adverse effect at the time of discovery or notice.G. Loan 1416352.

Loan 1416352 was a stated-income/verified assets loan originated by American Home and was included in the 2007-1 Trust. (See PXO03, line 3294.) The loan was issued for a rate-and-term refinance of a non-owner occupied property. Appendix 1 lists "Original DTI" as 22.26%. The loan closed on September 5, 2006. The 2007-1 Trust closed on January 16, 2007. (PX 110 at 33, 36.)

Holt opined that the loan had multiple defects, including improper calculation of debt, housing history violation, non-arm's length violation, undisclosed mortgage debt with red flags, income misrepresentation with red flags, unreasonable stated income, and violations related to the DTI ratio and LTV ratio.

1. Undisclosed Mortgage.

Holt opines that on September 15, 2006, within 30 days of the subject property's funding on September 6, 2006, the borrower secured another mortgage on a second property. The borrower took out a first mortgage on the second property of $412,930 and a second mortgage on the second property of $147,435. Holt opines that, as a result of these additional mortgages, the borrower's DTI ratio was 78.17%, which exceeded the guidelines maximum DTI ratio of 45%. Holt states that he learned of the additional loans through the MERS database and post-origination bankruptcy filings, and that additional credit inquiries about the borrower were red flags that should have prompted closer scrutiny.

The UBS's MLS Warranty as to the DTI ratio speaks as of the Closing Date of the PSAs. For the 2007-1 Trust, the Closing Date was January 16, 2007. Grissom testified that the mortgage on the second property would have appeared on the MERS system within 30 or 60 days, i.e. by December 15, 2006 before the Closing Date. (Grissom Tr. 491) The borrower's DTI ratio as of the Closing Date, and taking account of the two mortgages on the second property, was 78.17%, when it was inaccurately listed on the MLS as 22.26%. (PXO03, line 3294, col. CE.)

Separate from the MLS breach, Holt observed and Grissom did not dispute that there were 12 credit inquiries concerning the borrower in the 90 days before origination. Inquiries concerning a borrower to credit agencies are suggestive that a borrower may be applying elsewhere for additional loans. Multiple credit inquiries are a red flag to an underwriter that would call for a follow-up inquiry of the borrower. Grissom reasonably observes that sometimes a credit inquiry may be double counted.

Weighing the opinions of the experts, the Court concludes that the sheer number of credit inquiries so close in time to the funding of the loan should have caused the underwriter to make further inquiry. The 2007-1 Trust has proved that it is more likely than not that the Guidelines Warranty was breached in not pursuing the red flag of 12 credit inquiries with the 90 days preceding the funding of the loan.

2. Income Misrepresentation and Reasonableness of Stated Income.

The borrower stated that he earned $30,000 per month or $360,000 per year as a sales manager at Freestand Financial for the origination year of 2006. According to Holt, the borrower's Chapter 7 bankruptcy filing in 2008 stated that he and his spouse had a combined monthly income of $14,328 or $171,936 per year in 2007. Holt opines that because the income for 2007 was "near-year" to the 2006 income, it proves that the income was misstated.

But Holt provides no detail about the nature of the borrower's compensation at Freestand Financial and whether it was commissioned-based or otherwise subject to fluctuation due to market conditions. In context, Holt has not set forth a cogent basis for his opinion that the statement in the 2008 bankruptcy petition as to the borrower's 2007 income should be credited as proving that the 2006 income was misstated. Income for calendar year 2007 could not have impacted the DTI ratio on the MLS because for the 2007-1 Trust, the Closing Date was January 16, 2007.

The Trust has not proved that it is more likely than not that the income reflected in the DTI ratio on the MLS was materially untrue or incorrect.

There remains the issue of whether the underwriter complied with the guidelines as to the reasonableness of stated income. Holt opines that the borrower's stated monthly income of $30,000 was not reasonable, and that the loan file contains no indication that the originator assessed the reasonableness of the stated income. Grissom cites to the presence of compensating factors, but there was no written indication that the underwriter exercised an exception. Moreover, Holt notes that in 2005 and 2006, the borrower had refinanced the loan for his residence in 2005 and 2006, a "trend of equity stripping" that indicated the borrower's monthly obligations exceeded his income.

Given that the borrower stated income of $360,000 per year, evidence of his affiliation with the Originator and red flags that the borrower was engaging in the "equity stripping," the Court finds that the Trust has proved that it is more likely than not that there was a breach of the Guidelines Warranty due to the underwriter's failure to review the reasonableness of the borrower's stated income.

3. Improper Calculation of Debt.

Holt opined that the underwriter used the incorrect interest rate for calculating principal, interest, taxes and insurance ("PTI") payment in the underwriting process for this loan: "The underwriter used the negative amortization payment of $1,580 for this property and the interest rate of 1.4% versus the qualifying interest rate of 4.75% for the loan amount of $580,400 when the payment should have been $2,703 plus taxes and insurance." Holt opined that this error resulted from the underwriter's failure to correctly calculate DTI ratio, and that the loan approval listed an inaccurate monthly PTI payment. In rebuttal, Grissom argues that there is no evidence to find that DTI ratio was inaccurately calculated, a conclusion that the Court rejects.

Based on the borrower's undisclosed mortgages and the underwriter's failure to review the reasonableness of the borrower' stated income, the Court concludes that Trust has proved that the DTI ratio was inaccurately calculated during underwriting.

The Court finds that the Trust has proved that the underwriter inaccurately calculated the monthly PTI payment, which was derived from an inaccurate calculation of DTI ratio. The Trust therefore has proved a breach of the Guidelines Warranty.

4. Housing History Violation.

Holt observed that there was no documented mortgage payment history in the loan file for the borrower's investment property. The guidelines required that the borrower have no late mortgage payments within the preceding 12 months. Holt stated that he saw no indication of the underwriter's verification of the borrower's housing payment history and, therefore, the underwriter could not determine whether the borrower met the guidelines requirement in this respect.

Grissom noted that the borrower's loan file included a credit report that showed no late payment by the borrower for the preceding 24 months. In reply, Holt states that because the credit report reflects that the most recent reporting date was August 2006, and the subject property's closing was September 11, 2006, the credit report was insufficient to prove the payment history for all preceding 12 months. But again, as previously discussed with Loan 40588193, Holt has shown, at most, that the loan file does not contain payment history for the month preceding the closing of the subject property. The record shows evidence of a credit check, but no evidence of a late payment.

The Court finds that the Trust has failed to prove that it is more likely than not that underwriter failed to review the borrower's payment history in a manner complaint with the guidelines and therefore failed to prove that UBS breached the Guidelines Warranty in this respect.

5. Arm's Length Violation.

According to Holt, the borrower for this loan was employed by the loan's origination firm, and acted as the loan officer on his own loan. The American Home underwriting guidelines dated July 12, 2006 stated that borrowers involved in the construction or financing of the subject property were eligible for financing on a case-by-case basis, provided that the loan was underwritten by a Level 5 underwriter and the underwriter was satisfied that the transaction "makes sense." Documentation within the loan file must show that all services must have been provided by third-party providers, and that borrowers who are employed in real-estate trades may not apply commissions earned on the transaction toward closing costs or down payment.

Holt opined that, based on his review of the loan file, he could find no indication that the underwriting was performed by a Level 5 underwriter, or that the required "Drive-By Appraisal" for the property was completed. Grissom opines that the relevant Guidelines do not require either a Level 5 underwriter or a Drive-By Appraisal, and, in reply, Holt opines that Grissom relied on the underwriting guidelines of a different and irrelevant originator.

The Court finds that the 2007-1 Trust has proved that it is more likely than not that UBS breached the Guidelines warranty because of a failure to comply with the guideline requirements governing loans to employees, i.e. a non-arm's length transaction.

6. The LTV Ratio Defects.

Holt states that Cowan's AVM analysis establishes that the loan violated the guidelines based on misstated LTV data. For the reasons explained, the Court rejects the AVM analysis as a basis for recalculating LTV ratio.

7. Material and Adverse Effect at the Time of Discovery or Notice.

The Court has found that the 2007-1 Trust has proved that the breach of the MLS warranty in misstating the DTI ratio by reason of the two additional mortgages existing as of the Closing Date. The Court further finds that the breach materially and adversely affected the interests of the Certificateholders at the time of discovery or notice because the Trust would have rejected the inclusion of the loan or would have priced the transaction differently.

This Court has found that the Trust has proved breaches of the Guidelines Warranty in failing to review the reasonableness of the borrower's stated income and for non-compliance with the "non-arm's length" provision. While perhaps a close question, the Court finds that the Trust has not proved that it is more likely than not that these guidelines breaches would have materially and adversely affected the interests of the Certificateholders at the date of discovery or notice. They have failed to prove that the loan would not have been approved or not approved on the same terms had the Guidelines been applied properly.

Nevertheless, based upon the MLS Warranty breach, the Trust has proved that UBS is obligated to repurchase the Loan 1416352 or, if it has been liquidated, then to pay the money damages equivalent.

H. Loan 1450507.

Loan 1450507 was a stated-income loan originated by American Home and was included in the 2007-1 Trust. The loan was issued as a stated-term refinance loan for an owner-occupied property. Appendix 1 lists "Original DTI" as 26.96%. The funding date for the subject property was October 18, 2006. The Closing Date for the 2007-1 Trust was January 16, 2007. (PX 110 at 33, 36.)

Holt concluded that the loan had multiple defects, including employment misrepresentation with red flag, missing deed of trust, inadequate title insurance, DTI discrepancy and missing title insurance.

1. Employment Misrepresentation.

The borrower's loan application stated that the borrower had been self-employed as the owner of ET and Associates for three years, with a stated monthly income of $22,500 per month, or $270,000 per year. Holt opines that the borrower's stated income was not reasonable. He states that the borrower's loan file did not contain any business license documentation and that the borrower's credit profile did not include information on the borrower's annual income. Holt also relied upon the borrower's 2010 Chapter 7 bankruptcy filings, which indicated that he was self-employed for less than two years.

In rebuttal, Grissom notes that the loan file contained a letter from the borrower's tax preparer verifying that the borrower had been self-employed for two years. Of course, this letter does not explain the discrepancy between the claim in the loan application of three years of self-employment and the tax preparer's statement of two years self-employment. She cites to various compensating factors, but there is no evidence in the loan file that an exception to the guidelines was exercised.

The Court finds that Holt's reliance upon the absence of documentation in the loan file concerning the borrower's business license, and that the borrower's credit profile did not reflect annual earnings, is well grounded.

Weighing the opinions of the experts and the mix of information in the record, the Court finds that a reasonable underwriter applying the guidelines would have made further inquiry of the borrower concerning the stated income given the size of income, its derivation from self-employment, the variance in statements concerning the length of self-employment, the absence of any business license and the absence of other documentation.

The Court finds that the 2007-1 Trust has proved that it is more likely than not that UBS breached the Guidelines Warranty in the foregoing respects.

2. Document Defect.

Next to the heading "Missing Mortgage/DOT Violation," Holt states, "N/A for American Home—No applicable guideline found. Please pursue as Document Deficiency. Missing a recorded Deed of Trust." At trial, Holt explained that this notation was intended to state that the American Home guidelines did not require a recorded deed of trust from the borrower, and that the defect should be classified as a document deficiency. (Tr. 825-27.)In her rebuttal, Grissom notes that the loan file included an unrecorded deed of trust, and that the loan file included a Final HUD 1 form that reflects a deed recording fee of $92. Holt did not dispute her conclusions in his reply. At trial, Holt testified that the deed of trust "could have" been recorded and that "[w]e were looking at the four corners of the loan file to see if it was in there." (Tr. 828.) He stated that "if there was a supporting document or looking at this, I can't tell you ...." (Tr. 829.) He stated, "I might drop this particular breach, but I don't know if it would eliminate this loan as being materially deficient." (Tr. 829.)

As previously discussed, the deed was one of the documents required to be contained in the Mortgage File, as the term is described in the PSAs. Based upon the record presented, the Trust has not proved that it is more likely than not that there was a breach of the Mortgage File Warranty.

The Trust has failed to demonstrate how the absence of the deed of trust supports a separate breach finding under the Guidelines Warranty or other warranty.

3. Missing Title Insurance.

Holt concluded that UBS violated the Title Insurance Warranty because the loan files did not contain a title insurance policy. Grissom stated that the HUD-1 Form shows that a lender title policy had been purchased. Indeed, the HUD-1 Form lists the borrower's expenses in acquiring title insurance and summarizes policy coverage. (See PX L612 at 0228.) For the reasons discussed above, the Court concludes that the Trust has not proved a violation of the Title Insurance Warranty. Moreover, the HUD-1 Form is some evidence that a title insurance policy was obtained prior to the funding of the loan, and therefore does not support a breach of the Guidelines Warranty.

4. Correlated DTI Discrepancy.

Holt also identifies a breach of the MLS Warranty based on a misstatement of DTI ratio. Holt states that the MLS lists a DTI ratio of 26.96%, but that the actual DTI ratio was 39.29%. However, Holt does not explain the basis for such a recalculation. For other loans, Holt has identified an underlying defect in the borrower's debt profile or income, and recalculated the DTI ratio accordingly. This particular loan contains no such information.

The Trust has not proved a breach of the MLS Warranty based on materially incorrect DTI ratio on the MLS.

5. Material and Adverse Effect at the Time of Discovery or Notice.

The Court has found a Guidelines Warranty breach in not assessing the reasonableness of stated income. But the 2007-1 Trust has not come forward with evidence to demonstrate what information would have been known to the underwriter and the originating institution at the time of origination had there been full and complete compliance with the guidelines. Finding that income was unreasonably stated or that there was a discrepancy in the length of self-employment does not establish that further inquiry would have revealed facts that would cause the loan not to be approved or approved on different terms.

The nature of self-employment is such that there can be reasonable disagreement as to when self-employment began. For example, one person might decide from the inception and first discussion of the concept, while another might mark it from the rental of space and installation of telephone lines. No claim is made that there is a guideline prohibition on extending loans to those with fewer than three years self-employment. Certainly a $270,000 income sounds very high, but there is no evidence that this figure is materially wrong.

The underwriter failed to take the next required step to inquire further as to the reasonableness of stated income but the Trust has failed to prove what the underwriter would have learned had he or she taken that step and how it would have affected the Originator's willingness to approve and fund the loan.

The Court finds that the Trust has not proved that it is more likely than not that the guidelines breach materially and adversely affected the interests of the Certificateholders at the date of discovery or notice. It has failed to prove that the loan would not have been approved or not approved on the same terms had the guidelines been applied properly, or that the breach has increased the risk of loss to the Certificateholders.

I. Loan 40595823.

Loan 40595823 was a stated-income loan originated by MortgageIT and was included in the 2006-OA2 Trust. (See PXO01, line 4090.) The loan was issued for the purpose of a cash-out refinance on an owner-occupied home. Appendix 1 does not separately list the borrower's DTI ratio, but the MLS lists a "Front Ratio" of 36% and a "Back Ratio" of 41%. (PX O01, line 4090, cols. BZ, CA.) The funding date for the subject property was May 30, 2006. The Closing Date for the 2006-OA2 Trust was November 15, 2006. (PX 49 at 23, 27.)

Holt opined that the loan had multiple defects, including missing title insurance, document deficiency based on the absence of a title insurance policy, unreasonable stated income, undisclosed mortgages, income misrepresentation and incorrect LTV ratio and DTI ratio.

1. Income Misrepresentation.

The borrower's loan application stated that he had earned $12,000 per month (or $144,000 per year) for six years as a professional skater employed by the clothing company Billabong. However, tax returns for 2006 submitted by the borrower during loss-mitigation servicing indicated that the borrower's annual income was only $15,313, or $1,276 per month. The 2006 returns would have been filed in 2007, i.e. , post-origination.

At trial, Grissom testified that she believed the variance between the borrower's stated income and the income reflected in the later-filed tax returns was due to the unpredictable nature of professional athletes' income, and that the borrower might have sincerely expected to earn $12,000 a month in 2006. (Tr. 513-14.) At trial she opined that the underwriters "assumed the income was the stated income, yes." (Tr. 519.) This may have some bearing on the underwriting process, but does not impact the truthfulness of the data on the MLS, including specifically the DTI ratio.

Although the borrower's tax return was located in the post-origination servicing file, it reflected the borrower's income at the time of underwriting and was appropriately relied upon by Holt. The Court finds that the 2006-OA2 Trust has proven that it is more likely than not that the DTI ratio listed on the MLS, in material respects, was not true and correct because it used inaccurate income information, and that UBS therefore breached the MLS Warranty.

The Court addresses the related but separate inquiry of whether the Guidelines Warranty was breached with respect to a failure to inquire adequately as to the reasonableness of stated income of $12,000 per month for a professional skater. The Court credits Holt's opinion that the borrower's stated monthly income was unreasonable and was not verified by the underwriter.Holt identified red flags in the borrower's loan file, including an unexplained federal tax lien and a recent history of refinancing existing mortgage loans, for what appear to Holt to be an intent to strip them of equity. Holt also notes that the loan was a cash-out refinance that increased the loan amount from $385,000 to $468,000.

Grissom notes, however, that the borrower had a verified employment history of 7.58 years with the same employer, as well as 18 years of industry experience. At trial, she testified that she was unsure how to interpret the verification of employment, given that the borrower listed himself as self-employed, and speculated that it might have referred to an endorsement contract. (Tr. 515-16 ("Looking at this, I couldn't tell you.").) She lists various compensating factors related to the borrower, but there is no documentation that indicates an exception was exercised.

The Court finds that the Trust has proved that it is more likely than not that the stated income was unreasonable and should have been the subject of inquiry by the underwriter but was not. The Court further finds that this failure was a breach of the Guidelines Warranty.

2. Undisclosed Mortgage.

Holt states that, using the MERS database, he identified two mortgage loans opened by the borrower shortly before the funding date of the subject property on a second property. The borrower opened a first mortgage of $1,000,000 on this second property, and, on May 22, 2006, opened a second mortgage in the amount of $395,000 secured by this property. Neither was disclosed by the buyer to the loan originator.

The undisclosed mortgages would have increased the back-end DTI ratio listed on the MLS at 41% to 121.14%. If the correct monthly income is also considered, the DTI ratio increases to 1,139.25%. The Court finds that the Trust has proven that it is more likely than the DTI ratio listed on the MLS was, in material respects, not true and correct because it used incorrect debt and income information, and that UBS breached the MLS Warranty.

With respect to the separate question of whether there was a Guidelines Warranty breach, Holt states that there were multiple credit inquiries for the borrower, which, in his opinion, should have been a red flag to a prudent underwriter. The Court credits Holt's testimony on this point and finds that the Trust has proven that it is more likely than not that the Guidelines Warranty was breached because the underwriter should have pursued the red flag of multiple credit inquiries and would have been led to the undisclosed mortgages.

3. Missing or Inadequate Title Insurance.

Holt concluded that UBS violated the Title Insurance Warranty because the loan files did not contain a title insurance policy. For the reasons already discussed, the Court concludes that the Trust has not proved a violation of the Title Insurance Warranty. Moreover, Grissom stated that the loan file contained the final title insurance policy.

4. The LTV Ratio Defects.

Holt states that Cowan's AVM analysis establishes that the loan violated the guidelines based on misstated LTV data. For the reasons explained, the Court rejects the AVM analysis as a basis for recalculating LTV ratio.

5. Material and Adverse Effect at the Time of Discovery or Notice.

The Court has found that the Trust has proved that the breach of the MLS Warranty in misstating the DTI ratio by reason of the two additional mortgages existing as of the Closing Date and the borrower's misstated income. The Court further finds that the breach materially and adversely affected the interests of the Certificateholders at the time of discovery or notice because the Trust would have rejected the inclusion of the loan or would have priced the transaction differently.

The Court also has found that the Trust has proved that the breach of the Guidelines Warranty by reason of the failure to make due inquiry concerning the reasonableness of stated income and the missing additional mortgages. The Court further finds that the breach materially and adversely affected the interests of the Certificateholders at the time of discovery or notice because the Trust would have rejected the inclusion of the loan or would have priced the transaction differently. Unlike other exemplar loans, there is no speculation as to what a further income inquiry would likely have discovered or what a follow up on the multiple credit inquires would have uncovered. Holt opines that at what he views as the borrower's true income combined with certain undisclosed mortgage loans discussed below, the borrower's DTI ratio was 1,139.25%, where the guidelines set a maximum DTI ratio of 38%. In addition, the borrower's materially false statements about his or her own income reflect on the borrower's willingness and ability to repay the loan.

Based upon the MLS warranty breach and the Guidelines Warranty breach, the 2006-OA2 Trust has proved that UBS is obligated to repurchase the Loan 40595823 or, if it has been liquidated, then to pay the money damage equivalent.

J. Loan 158522879.

Loan 158522879 was a reduced-documentation loan originated by Countrywide and was included in the 2007-3 Trust. It was a cash-out refinance loan for a non-owner occupied home. Appendix 1 lists "Original DTI" as 6.44%. The funding date for the subject property was February 20, 2007. The Closing Date of the 2007-3 Trust was May 15, 2007. (PX 182 at 24, 26.)

Holt concluded that the loan had multiple defects, including undisclosed mortgage debt, failure to verify asset reserves, missing loan application and employment violation.

1. Undisclosed Mortgage Debt.

Using Lexis-Nexis and the MERS database, Holt identified six mortgages that the borrower did not disclose on her application. She refinanced five existing mortgages on January 19, 2007 and a sixth on February 16, 2007, with the balances on the mortgages ranging from $349,217 to $827,045. The subject property's note date was February 20, 2007.

The MLS Warranty as to the DTI ratio was made as of May 15, 2007, and at that point the DTI ratio had dramatically changed. The 2007-3 Trust has proved that it is more likely than not that the DTI ratio listed in the MLS was untrue and incorrect, and that the MLS Warranty therefore has been breached.

With respect to the related but separate issue of whether the Guidelines Warranty was breached by reason of these undisclosed mortgages, Holt's explanation in the hard-copy version of the summary submitted to the Court is incomplete. However, in the electronic version of Appendix 1, Holt states:

It should be noted, the borrower provided a letter of explanation, dated 3/7/2007, to explain the multiple credit inquiries present on her credit report, which indicated it was due to her mortgage needs. The borrower advised, "After searching for a lender, I finally decided

on your company as the right lender for my refinancing needs". The letter of explanation did not indicate the borrower had not obtained any new or additional debt.

In rebuttal, Grissom states that even when the borrower's undisclosed debt is considered, the resulting DTI ratio of 35.52% falls below the 38% maximum allowable DTI under the guidelines. In reply, Holt does not dispute this conclusion by Grissom.

Based on Holt's summary of the borrower's letter contained in Appendix 1, it appears that the borrower intentionally misled the underwriter as to the reason that these additional credit inquiries appeared in her credit report. While this is evidence that the borrower intentionally misstated the basis for the credit inquiries, Holt does not opine that either the guidelines or the practices of a reasonable underwriter required further inquiry or follow up. In this instance, the credit inquiries were a red flag, which was addressed through an apparently untruthful letter of explanation by the borrower. Holt does not opine that a reasonable underwriter thereafter would have taken additional steps to locate the source of the credit inquiries.

The Court finds that the 2007-3 Trust has not proved that there was a breach of the Guidelines Warranty as a result of the undiscovered mortgages. First, Holt does not dispute that when the additional mortgage loans are considered, the borrower still fell below the maximum allowable DTI ratio. Second, the borrower submitted a letter of explanation which, although apparently untruthful, offered an explanation of recent credit inquiries, and Holt has not identified any additional obligation on the part of the underwriter to either refute or confirm the contents of that letter. Therefore, while the Trust has proved a breach of the MLS Warranty, it has not proved a breach of the Guidelines Warranty.

2. Asset Violations.

Holt concluded that the borrower's cash asset reserves were in the name of the borrower's business, and that, under Countrywide's guidelines, an exception approval was required in order for such a business to satisfy the borrower reserve requirements. The exception could be exercised only if the borrower was a sole proprietor or sole shareholder of the business. According to Holt, no exception was exercised for the borrower to use business funds and the lender did not verify that the borrower was the sole proprietor or sole shareholder of the business. Because the only verified assets for this loan were the borrower's business funds, Holt concluded that the guidelines had been violated.

In her rebuttal, Grissom appears to conclude that the loan was properly documented based on its ultimate approval. As previously discussed, the Court concludes that the fact of a loan's approval does not establish that it was underwritten in compliance with the guidelines or that an exception was exercised. Similarly, Grissom recites various compensating factors, but there is no written indication that an exception was exercised.

The Court finds that the 2007-3 Trust has proved a breach of the Guidelines Warranty based on the underwriter's failure to comply with the guidelines requirement where assets are not held in the name of the borrower but held by a related company. No verification was made that the borrower was the sole shareholder or sole proprietor of the business and no exception was actually exercised for the use of business funds to meet reserve requirements.

3. Missing Loan Application.

According to Holt, Countrywide's guidelines required a final loan application to be signed by the borrower no later than at closing. Typically, a loan application is submitted and signed when the initial loan application is submitted. To protect against changed circumstances, the borrower is asked to again sign the application at or before the closing of title and funding of the loan. Holt states that the loan file did not contain a final signed loan application, which this Court construes to mean the re-execution of the loan application shortly before closing. The Court previously noted that the loan application was a "core document" that, if absent from the loan file, could give rise to a breach of the Guidelines Warranty.

In rebuttal, Grissom stated that a loan file audit document "indicated" that a final signed loan application was in the loan file prior to closing and that the loan file audit is contained in the loan file. Holt offers no response to Grissom's assertion except to note that there was no signed application in the file.

The Court finds that the 2007-3 Trust has failed to prove that it is more likely than not that the final loan application was obtained in connection with the underwriting and origination of the loan and, thus, they have failed to prove a breach of the Guidelines warranty in this respect.

4. Employment Violation.

According to Holt, the loan application stated that the borrower had been self-employed as a business owner for ten years and one month. He stated monthly earnings of $40,000, or $480,000 per year. According to Holt, the guidelines required that when a borrower reported income from a self-employed business, the underwriter must call directory assistance to verify the business's existence, and if the business is not listed, the loan should be denied. Holt concluded that there was no indication in the loan file that the underwriter called directory assistance to validate the business's existence.

In rebuttal, Grissom states that the loan file included a written verification of employment, the business's name and phone number, business licenses and a credit report that listed the borrower's business name going back to 1998. In reply, Holt opines that the written documentation is not a substitute for the requirement in the guidelines to call directory assistance.

The Trust has failed to prove that the Countrywide guidelines were understood to require the call to directory assistance even when the underwriter secured the business's name and phone number, business licenses and an independent credit report for the business. The Court finds that the Trust has not proved a Guidelines breach in this respect.

5. Material and Adverse Effect at the Time of Discovery or Notice.

The Court has found that the Trust has proved that the breach of the MLS Warranty in misstating the DTI ratio by reason of the additional mortgages existing as of the Closing Date. The Court further finds that this breach materially and adversely affected the interests of the Certificateholders at the time of discovery or notice because the Trust would have rejected the inclusion of the loan or would have priced the transaction differently.

The Court also has found that the Trust has proved that the breach of the Guidelines Warranty with regard to the failure to verify assets used from the borrower's related company. However, the Trust has not identified any resulting effect on the interests of the Certificateholders. It is not clear how the absence of verification would have affected the underwriting or loan approval process, and whether the loan would not have been approved or approved on different terms.The Court has not found a breach arising from the failure to place a call to directory assistance to verify the existence of the business. But if the Court were to find this breach, it would not have materially and adversely affected the Certificateholders because of the verification of the same information, as well as a business license, though an independent credit report.

However, based upon the MLS Warranty breach, the 2007-3 Trust has proved that UBS is obligated to repurchase the Loan 158522879 or, if it has been liquidated, then to pay the money damages equivalent.

K. Loan 124359279.

Loan 124359279 was a full-documentation loan originated by IndyMac and was included in the 2006-OA2 Trust. The loan was for the rate-and-term refinancing of an owner-occupied home. Appendix 1 does not list a DTI ratio for the loan, but the MLS lists a "Front Ratio" of 0% and a "Back Ratio" of 37.7%. (PX O01, line 457, cols. BZ, CA.) The funding date for the subject property was September 22, 2006. The Closing Date of the 2006-OA2 Trust was November 15, 2006. (PX 49 at 23, 27.)

The sole breach identified by Holt is a document deficiency based on a missing loan approval.

1. Document Deficiency.

In describing the breach, Holt's opinion states in its entirety: "The Loan Approval was not available at the time of this loan review. A review of the document may change the Auditor's analysis of the loan file."

As previously discussed, the Court concludes that the loan approval was one of the "core documents" used in underwriting, and its absence from the loan files constitutes a breach of the guidelines. In rebuttal, Grissom does not dispute that the loan approval is missing from the loan file, and at trial, she testified that the loan approval was a core document that was necessary to funding the loan. (Tr. 651.)

The Court therefore finds that the 2006-OA2 have proved that the loan approval was absent from the loan file, and the Guidelines Warranty was breached.

2. The Trust Has Not Proved a Material and Adverse Effect on the Interests of the Certificateholders at the Time of Notice or Discovery.

Although the Trust have proved that the loan approval is missing from the loan file, it is not apparent how this defect had a material and adverse effect on the interests of the Certificateholders at the time of notice or discovery. Holt's opinion is that the loan approval "was not available at the time of this loan review." Holt does not contend that the loan should not have been approved or should have been issued on different terms. Standing alone, this breach does not provide the Court with any basis to conclude that it is more likely than not that the Certificateholders suffered a material and adverse effect at the time of discovery or notice due to the absence of a loan approval from the loan file.

The Court therefore finds that the 2006 OA-2 Trust has not proved that it is more likely than not that the breach had a material and adverse effect on the interests of the Certificateholders. UBS is therefore not obligated to repurchase or pay the money damages equivalent as to Loan 124359279.

L. Loan 1423423.

Loan 1423423 was a stated-income loan originated by American Home and was included in the 2007-1 Trust. The loan was for the rate-and-term refinancing of an owner-occupied home. Appendix 1 lists "Original DTI" as 13.28%. The funding date for the subject property was October 2, 2006. The Closing Date of the 2007-1 Trust was January 16, 2007. (PX 110 at 33, 36.)

Holt concluded that the loan had multiple defects, including misrepresentation of income, incorrect DTI ratio and incorrect LTV ratio.

1. Income Misrepresentation.

The borrower's loan application stated that the borrower worked for 15 years at the "Foster Grandparent Program" as a grant manager. The borrower's application stated a monthly income of $18,500, or $222,000 per year. The subject property transaction closed on October 2, 2006.

The borrower filed for Chapter 7 Bankruptcy protection on January 31, 2008. The borrower's bankruptcy petition, according to Holt, stated that the borrower's 2006 annual income was $51,670, or $4,306 per month. Holt also states that the borrower's loan file included 2006 tax returns and W2 statements that support the conclusion that the borrower misrepresented his income, though Holt does not specify the income reflected on those returns.

Holt also notes that there is no indication that the underwriter assessed the reasonableness of the borrower's stated income. The Court concludes that a stated monthly income of $18,500 for the position of grants manager for a non-profit program was sufficiently high to have been a red flag causing a reasonable underwriter to investigate further.

Grissom opines that the bankruptcy filings were created post-origination and not available to the underwriters, but the bankruptcy filings reflected the borrower's income at the time origination. At trial, however, Grissom testified that she had no reason to dispute Holt's conclusions. (Tr. 531.) Grissom also notes compensating factors such as credit score, verified reserves and mortgage history, but the Court does not consider these compensating factors because there is no written evidence that an exception was exercised.

The Court finds that there was a breach of the Guidelines Warranty because the loan was approved and funded without inquiry as to reasonableness of income, a guideline requirement.

The Court finds that it is more likely than not that the borrower misrepresented a monthly income of $18,500. This finding is further supported by the especially high income claimed by the borrower—$222,000 per year—for the job of managing grants at a foster grandparents' organization in the Reno, NV area. While the bankruptcy filing post-dated origination, it is evidence that the borrower's annual income for 2006 was $51,670, in a transaction that closed on October 2, 2006. Further, the borrower's income misrepresentation raises the borrower's DTI ratio from 13.28% to 77.43%.

As noted, Holt concluded that the borrower's income misrepresentation resulted in an inaccurate DTI ratio, thereby resulting in a breach of the MLS Warranty and the Guidelines Warranty. Holt opines that when the borrower's actual, lower income is taken into account, the borrower's DTI ratio jumps from the 13.28% disclosed in the MLS to 77.43%.

The Court finds that it is more likely than not that UBS breached the MLS Warranty because the borrower's DTI ratio as listed on the MLS was in material respects not true and correct as of the Closing Date.

Because an underwriter should have but did not conduct a reasonableness of income inquiry, the Guidelines Warranty was also breached because the maximum DTI ratio permitted under the guideline was 40%. The recalculated DTI ratio of 77.43% significantly exceeds the guidelines maximum, and there is no indication that an exception was exercised by the underwriter.

2. LTV Ratio Defects.

Holt states that Cowan's AVM analysis establishes that the loan violated the guidelines based on misstated LTV data. For the reasons explained, the Court rejects the AVM analysis as a basis for recalculating LTV ratio.

3. Material and Adverse Effect at the Time of Discovery or Notice.

The Court has found that the 2007-1 Trust has proved the breach of the MLS Warranty in misstating the DTI ratio by reason of the borrower's income misrepresentation as of the Closing Date. The Court further finds that this breach materially and adversely affected the interests of the Certificateholders at the time of discovery or notice because the Trust would have rejected the inclusion of the loan or would have priced the transaction differently.

This Court has found a breach of the Guidelines Warranty based on the borrower's misrepresentation of income and the underwriter's failure to review the reasonableness of the borrower's stated income. This breach materially and adversely affected the interests of the Certificateholders at the date of discovery or notice because an income reasonableness check would have revealed lower income, which would have placed the loan at a level above the DTI ratio permitted by the Guidelines. The loan would not have been approved, or would have been approved under different terms had the guidelines been applied properly.

Based upon the MLS Warranty breach and the Guidelines Warranty breach, the 2007-1 Trust has proved that UBS is obligated to repurchase Loan 1423423 or, if it has been liquidated, then to pay the money damage equivalent.

M. Loan 138058233.

Loan 138058233 was a full-documentation loan originated by Countrywide, and was included in the 2006-OA2 Trust. The loan was for a cash-out refinancing of an investment property. Appendix 1 lists the "Original DTI" as 41.55%. The funding date for the subject property was July 28, 2006. The Closing Date of the 2006-OA2 Trust was November 15, 2006. (PX 49 at 23, 27.)

Holt concluded that the loan had multiple defects, including missing title insurance, a seasoning violation, income misrepresentation, missing final title, incorrect DTI ratio and incorrect LTV ratio.

1. Income Misrepresentation.

The borrower's final loan application stated that the borrower had worked as a supervisor for 3 years and 10 months, with a stated monthly income of $8,166.47, or about $98,000 annually. The subject property transaction closed on July 28, 2006.

The borrower filed for Chapter 7 bankruptcy protection on September 23, 2008. The borrower's bankruptcy petition stated that the borrower earned a total of $10,024 in 2008.

Grissom notes that the borrower's loan was issued under an "alternate documentation program," which required income verification. She states that the loan file included current year-to-date paystubs and W-2 forms from 2004 and 2005. She states that the underwriter calculated the borrower's income by averaging 2004 and 2005 earnings and year-to-date paystubs ending on April 30, 2006, which allowed for the calculation of monthly income over a 28-month period. The loan file also documented a three-year employment history with the same employer. Thus, she concludes, at origination, the underwriter properly documented and verified the borrower's monthly income of $8,166.47.

In reply, Holt states that the "W2s at origination were found to be fraudulent," but he includes no additional details to support that conclusion. At trial, Holt testified that he believed the W-2 forms were fraudulent because of their font style and the absence of commas in numerical figures:

Well, looking at the W-2s, one thing you look at is whether or not the—the numbers match up with the rest of the form. Typically the numbers will match up with the font. The size of the font will stand out as being different. And you see on here the font is different than, say, you would see on—the font with the form, everything's done the same. There is—there's typically commas in there. There's no commas. I see there's commas on this one. There's no commas on that one either.

* * *

Typically you'd see commas in the numbers like in cells one, two, three, four—anything over 1,000, you'd typically see the—a comma.

(Tr. 882, 883.)

Holt's testimony about the use of commas in W2 forms is without basis in logic and experience. It is also contrary to guidance for employers that is published by the IRS, of which the Court takes judicial notice. See Topic 752—Filing Forms W2 and W3, available at https://www.irs.gov/taxtopics/tc752.html ("Make all dollar entries without the dollar sign and comma but with the decimal point (00000.00).").

In testifying about the basis for labeling the W2 forms as fraudulent, the Court found Holt's demeanor was halting and confused, and his explanations appeared to be improvised on the stand. The Court rejects Holt's opinion on the authenticity of the W2.

The Court finds that the Trust has not proved that the borrower misrepresented income. Grissom identified documentation in the loan file that verified the borrower's stated income. Holt did not articulate a persuasive or credible explanation as to why he concluded that the W-2 forms were fraudulent, and did not address why the borrower's bankruptcy filings speaking to a later time period (2008) should be credited over the contents of the loan files.

Because the Court concludes that the Trust has not proved that the borrower's income was misrepresented, it also has failed to prove correlated violations of the Guidelines Warranty and the MLS Warranty based on a purportedly inaccurate DTI ratio.

2. Seasoning Violation.

Holt also identified a breach of the Guidelines Warranty based on what he called a "seasoning violation." At trial, Holt described a seasoning violation as follows:

It's where the—when you're buying a house, and you have—it's a purchase price, say, of $100,000 but the appraised value is $105,000, you can only use the lesser of the two in calculating the loan to value. You can't use the higher of the two.

(Tr. 877.) Holt concluded that the Countrywide guidelines required the LTV ratio to be calculated using the lesser of the appraised value or purchase price. Holt states that the actual purchase price of the home was $215,700, but that LTV was calculated using an appraised value of $275,000. Because the loan was for $247,500, Holt concluded that the true LTV ratio was 114.74%, as opposed to the 90% calculated by the underwriter.

In contrast to many of the disputed loans, in this instance, there was a written exception to the guidelines, which approved the use of the higher appraisal value. As summarized by Holt: "The Lender noted and approved the exception citing no late payments on the credit report and time on job as compensating factors." Grissom's rebuttal states, "as Mr. Holt even acknowledges, the loan file included an exception that approved the use of the appraised value of $275,000, rather than the purchase price, to calculate a 90% LTV."

As noted, the Guidelines Warranty provides that each mortgage loan was underwritten according to the originator's underwriting guidelines "with exceptions thereto exercised in a reasonable manner." (PX 49 at 196.) Holt does not dispute that there is a written exception, and he does not opine that the exception was unreasonable. The Court therefore concludes that the Trust has not proved a breach of the Guidelines Warranty due to the so-called "seasoning violation" and the underwriter's exercise of an exception. Under the exception that was exercised, the underwriter used the higher appraisal value of $275,000, and calculated the LTV accordingly.

The Court finds that the Trust has not proved a breach of the MLS Warranty based on a misstated LTV ratio. As discussed in the construction of the MLS Warranty, the LTV ratio is calculated based on "Appraised Value," which, for a Refinancing Mortgage Loan, is "based upon the appraisal made at the time of the origination of such Refinancing Mortgage Loan as modified by an updated appraisal." (PX 49 at 023.) Based on Holt's summary of his opinion, the LTV ratio listed in the MLS was based on the property's appraisal value of $475,000, and thus was calculated in a manner consisted with the "Appraised Value" defined by the PSAs.

3. Missing or Inadequate Title Insurance.

Holt concluded that UBS violated the Title Insurance Warranty because the loan files did not contain a title insurance policy. For the reasons already discussed, the Court concludes that the Trust has not proved a violation of the Title Insurance Warranty.

Moreover, Grissom stated that the loan file contained the final title insurance policy. The presence of the title policy in the loan file also is indication that a title insurance policy was obtained prior to the funding of the loan, and therefore does not support a breach of the Guidelines Warranty.

4. LTV Ratio Defects.

In addition to the purported LTV defect based on the claimed seasoning violation, Holt states that Cowan's AVM analysis establishes that the loan violated the guidelines based on misstated LTV data. For the reasons explained, the Court rejects the AVM analysis as a basis for recalculating LTV ratio.

5. No Breach.

Taking into account all evidence relating to the loan, including the opinion of the experts, and giving due consideration to all actual or potential red flags and the cumulative effect of the purported breaches, the Court finds that the 2006-OA2 Trust has not proved a breach of any warranty as to Loan 138058233 and, thus, it is unnecessary to address whether any breach had a material and adverse effect at the time of discovery or notice.

N. Loan 1418398.

Loan 1418398 was a stated-income loan originated by American Home, and was included in the 2007-1 Trust. The loan was for the purchase of an owner-occupied property. Appendix 1 lists "Original DTI" as 37.76%. The funding date for the subject property was October 25, 2006. The Closing Date of the 2007-1 Trust was January 16, 2007. (PX 110 at 33, 36.)

Holt concluded that the loan had multiple defects, including occupancy misrepresentation, income violation and incorrect DTI ratio.

1. Occupancy Misrepresentation and Income Violation.

Holt has opined that there were two somewhat interrelated guidelines violations based on the status of what he describes as the borrower's "departure home," which the Court understands to mean the existing home that the borrower is occupying at the time of application, but will cease to occupy, once the borrower closes on a new home purchase.

First, Holt observed that the borrower's final loan application indicated that the borrower would be retaining his departure home, but also stated that he was not using rental income from that property to qualify for the subject property. However, the loan approval listed rental income from the departure home. Holt opines that the failure to separately identify this rental income was an "income violation" under the guidelines.

Somewhat relatedly, Holt opines that the borrower continued to reside in the departure home, and misrepresented that the subject property would be owner-occupied. Holt based this conclusion on a Lexis-Nexis Advanced Person report, which "indicated the borrower's departure home as current and primary from 2004 through 5/2015." Further, he stated that Miami-Dade County property tax records from 2006 through 2008 reflected that the borrower maintained a homestead exemption for the departure home. Holt concludes that the Lexis-Nexis data and the tax records establish that the borrower continued to reside in the departure home and that he misrepresented that the subject property would be owner-occupied.

Holt states that there was a red flag at origination because the departure home was refinanced one month prior to the transaction's closing.

Grissom opines that evidence concerning the borrower's post-origination occupancy is irrelevant because it was not available to the underwriters. She also notes that the borrower had declared his intention to occupy the subject property. At trial, she testified that the borrower may have refinanced the departure home because of his intention to use it as a rental property, and that refinancing would have lowered the carrying costs of the departing home. (Tr. 577.)

With respect to a MLS Warranty breach concerning owner-occupancy, post-origination data may have some relevance. But, here, Holt and the Trust overstate the case. There is no dispute that the underwriter knew that the borrower was retaining the departure home and, indeed, the underwriter is faulted because he or she took into account future rental income on the departure property. The mere fact that the borrower retained ownership of the departure property does not prove that the borrower did not become an owner-occupier of the newly-acquired property.

Because the borrower was retaining the departure property, its refinancing would not necessarily be a red flag to the underwriter. The departure property was subject to a homestead exemption for the two years following funding of the loan on the subject property but this could be attributed to any number of reasons—including inadvertence on the part of the borrower or officials, or a conscious decision to receive favorable tax treatment to which the borrower was not entitled. It also is unclear from Holt's summary how and why Lexis-Nexis designates the departure property as a primary residence.

While it is plausible that the borrower continued to occupy the departure property as his primary residence, it is also plausible that he continued to hold that property as a source of rental income and occupied the subject property. The Trust has failed to prove that it is more likely than not that, at the time of origination and funding of the loan, the borrower did not intend the subject property to be the borrower's primary residence. The Trust has not proved that the loan was not originated and funded "in accordance with" the Originator's guidelines as it relates to an occupancy misrepresentation.

Further, the Court finds that the Trust has not proved that, as of the Closing Date of the PSA, the MLS incorrectly listed the subject property as owner-occupied. The Trust therefore has not proved a violation of the MLS Warranty as to occupancy status.

The Trust also has not proved a guidelines violation based on rental income from the departure home. Grissom identified documentation in the loan file listing rental income from the departure home, which Holt does not dispute. The amount of monthly rental income listed in the loan application totaled $419 a month. Holt does not contend that the guidelines barred use of rental income to qualify for the loan, but merely that the source of rental income was not labeled as such in the final loan application. Based on Holt's summary, the Court is unable to identify a guidelines breach based on the failure to separately denote the source of the $419 in monthly rental income on the final loan application. Moreover, even if the Court were to determine that the Trust proved an income violation, Holt's summary describes no basis for identifying a material and adverse effect on the interests of the Certificateholders based on the borrower's failure to specifically identify the $419 figure as monthly rental income.

2. Breach of the MLS Warranty Based on Misstated DTI Ratio.

The Trust claims that there is a breach of the MLS Warranty because, as Holt observes, the MLS listed a DTI ratio of 37.76%, whereas the desktop underwriter software used a DTI ratio of 40.67% and there is no explanation for the disparity.

As the Court previously discussed, the MLS Warranty states that the DTI ratio listed in the MLS "was true and correct in all material respects" as of the Closing Date of the PSA. In the case of the 2007-1 Trust, the closing date was January 16, 2007. The subject loan transaction closed on October 25, 2006. Thus, there was a two-and-a-half month gap between the last date on which the underwriting guidelines were applied and the date of the MLS Warranty as to DTI ratio.

It is possible that as of the Closing Date, a borrower's DTI ratio could have changed from the DTI ratio used in the underwriting process. In this case, the borrower's listed DTI ratio was 2.91% lower in January 2007 than it was in October 2006. This change could be due to the borrower receiving a salary increase, a new job, or having paid off personal debt.

Holt's conclusion that UBS breached the MLS Warranty is based solely on the difference between the DTI ratio listed in the MLS and the DTI ratio used in underwriting. He does not cite to any evidence in the loan file that raises an inference that the MLS listed an incorrect calculation of the DTI ratio.

Because the MLS Warranty requires that the DTI ratio be true and correct in all material respects as of the Closing Date (in this case, January 16, 2007), the Court concludes that the Trust has not proved a breach of the MLS Warranty based solely on the fact that the underwriter used a different DTI ratio for a loan that closed on October 25, 2006.

3. Correlated DTI Discrepancy.

Holt opines that if the borrower's rental income is excluded from the DTI calculation, the borrower's DTI ratio would increase from 40.67% to 52.66%, which exceeds the guidelines maximum DTI ratio of 45%. Because the Court finds that the Trust has not proved an income violation based on the borrower's rental income, it also has not proved a guidelines breach as to the DTI ratio.

4. No Breach.

Taking into account all evidence relating to the loan, including the cumulative effect of the purported breaches and giving due consideration to the opinions of the experts, the Court finds that the 2007-1 Trust has not proved a breach of any warranty as to Loan 1418398 and, thus, it is unnecessary to address whether any breach had a material and adverse effect at the time of discovery or notice.

O. Loan 1477160.

Loan 1477160 was a stated-income loan originated by American Home, and was included in the 2007-1 Trust. The loan was a rate-and-term refinance for an owner-occupied home. Holt lists an "Original DTI" of 27.47% (Appendix 1). The funding date for the subject property was October 31, 2006. The Closing Date of the 2007-1 Trust was January 16, 2007. (PX 110 at 33, 36.)

Holt identified numerous defects in the loan, including unreasonable state income, occupancy misstatement with red flags, unreasonable occupancy statement, undisclosed mortgage debt with red flags and incorrect DTI ratio and LTV ratio.

1. Unreasonable Stated Income.

The borrower's final loan application stated that the borrower had worked for ten years as a "Regional Manager" for Kay Jewelers, and earned $14,000 per month, or $168,000 per year. The guidelines required a reasonableness assessment of the borrower's stated income, and Holt found no evidence that such an assessment was made.

Holt identified certain characteristics of the borrower that called into question the reasonableness of stated income, including a vested 401K balance of $48,903 and a loan balance of $12,540 on the 401K statement; a prior bankruptcy; and a revolving account with a balance of $3,729 with 100% credit utilization.

Grissom lists various factors that support the borrower's claimed income, including the borrower's verified employment history of 10 years and 20 years of industry experience, and verified cash reserves of $36,293.36. She opines that such factors would support the underwriter's reasonableness decision.

Holt also cites to BLS data, which stated that for "Sales and Related Occupations" in the borrower's region, the 90th-percentile income was $8,017 per month, or $96,212 annually. Holt does not calculate the "grossed up" income for such a position (as he does elsewhere), and it is not apparent that a "Regional Manager" would necessarily fall within the category of "Sales and Related Occupations." There is no indication of the breadth of responsibilities or number of stores encompassed within the responsibilities of a "Regional Manager" and whether the borrower's job might have been more appropriately evaluated in the category of corporate management.

While the Court affords minimal weight to the Holt's use of BLS data to evaluate the stated income of this borrower, it is nevertheless the case that neither Grissom nor Holt identified a written reasonableness assessment in the loan file, and that the guidelines required the underwriter to consider whether the income was reasonable and consistent with the borrower's profession, occupation and income source. The Court accepts Holt's opinion that there were red flags that should have prompted further inquiry: a loan against a relatively small 401K balance, a prior bankruptcy and 100% utilization on a revolving credit account. The Court concludes that a reasonable underwriter would have made a reasonableness assessment as to the borrower's stated income, as required under the guidelines, and documented same in the loan file.

On balance, the Court finds it more likely than not that UBS breached the Guidelines Warranty based on the underwriter's failure to assess the reasonableness of the borrower's stated income.

2. Undisclosed Debt.

Holt concluded that the borrower failed to disclose an auto loan acquired one month prior to the closing of the property transaction. The borrower also failed to disclose a loan taken against the borrower's 401K account. Holt, in Appendix 1, does not list the amount of the auto loan or the amount of the undisclosed 401k loan.

Holt states that the credit report at origination showed that 25 credit inquiries were made as to the borrower on or about October 5, 2006, which was a red flag to underwriters about the possible existence of additional, undisclosed debt. Grissom argues that Holt inappropriately relies on post-origination information to identify undisclosed debt, but the credit report that indicating the 25 credit inquiries was available to the underwriter while evaluating the application.

The Court concludes that the Trust has proved a breach of the Guidelines Warranty based upon the underwriter's failure to pursue red flags that were suggestive of the existence of additional, undisclosed debt incurred by the borrower.

3. The Borrower's Occupancy Representation.

Holt opines that there was a breach of the Guidelines Warranty based on what he identifies as the borrower's occupancy misrepresentation. (Holt appears to characterize it as two breaches: "Unreasonable Occupancy" and "Occupancy Misrepresentation with Red Flags.") The transaction was originated for the purpose of making a rate-and-term refinance of an owner-occupied home. Holt opines, however, that the transaction was in fact a cash-out refinancing of a non-owner occupied property.

Holt states that there were red flags indicating that the property was not owner-occupied. The transaction closed on October 31, 2006. The original appraisal reflected that the purportedly owner-occupied property to be refinanced was vacant at the time of inspection. A bank statement of June 30, 2006 reflected that the borrower resided at a prior address, and the "Uniform Underwriting and Transmittal Summary" stated that the property was listed for sale as of April 27, 2006.

Grissom responds identifying a letter of explanation in the loan file which stated that the subject property was vacant at the time of underwriting because the property was undergoing construction. The loan file also included proof of homeowner's insurance for the property, which did not include landlord coverage, and showed that the policy's billing address matched the subject property's address. Holt does not address these items in his reply.Separately, the borrower filed for Chapter 7 bankruptcy protection in August 2009. According to Holt, the borrower's bankruptcy petition stated that in 2007, the borrower resided at an address that was different from the subject property. Holt also concluded that a Lexis-Nexis search indicated that eight months after the subject loan transaction closed, the borrower registered a motor vehicle at the property previously listed as the borrower's primary residence and not the subject property.

Holt's conclusions based on the bankruptcy filings and Lexis-Nexis data were undermined at trial. Counsel to UBS presented Holt with several items in the borrower's bankruptcy filing reflecting that the borrower and the borrower's co-debtor, did, in fact, reside at the subject property. (Tr. 916-17.) Holt conceded that the bankruptcy submissions established that the borrower resided at the subject property:

Q. So you'll agree with me, won't you, that this filing shows that your statement that the borrower never occupied 5503 Justina Drive was a false statement?

A. Based on the bankruptcy court, yes.

(Tr. 918.) Holt also conceded that a Lexis-Nexis report showed that "every single one" of the borrower's addresses was listed as the subject property, including the report for the borrower's motor vehicle registration. (Tr. 919.)

Holt's opinions were based on an erroneous and unfounded summary of bankruptcy filings and Lexis-Nexis data. Contrary to Holt's summaries contained in Appendix 1, these sources were evidence that the subject property was, in fact, owner-occupied, and contradict Holt's conclusion that occupancy was misrepresented.

This evidence also makes more plausible the letter of explanation's assertion that the residence was vacant during appraisal due to construction on the premises. Holt does not address the letter of explanation in his reply. At trial, Holt testified that there were other "pretty strong red flags" as to occupancy, but did not identify them. (Tr. 921.)

This example further supports the Court's findings that Holt's opinions, including those based upon bankruptcy filings, may not be accepted at face value and are appropriately subject to scrutiny; they may not be applied across the board to all loans of a certain type or category.

The 2007-1 Trust has failed to prove that it is more likely than not that the borrower misstated or misrepresented the occupancy status of the subject property. The Trust has therefore failed to prove a breach of the Guidelines Warranty or MLS Warranty premised upon such a misstatement or misrepresentation.

4. Correlated DTI Ratio Defects.

Holt concludes that the borrower's actual DTI ratio was 187.71%, as opposed to the DTI ratio of 27.47% that was listed in the MLS. Holt bases this recalculation on the existence of the additional, undisclosed auto debt and 401K loan, and on his use of BLS data to recalculate the borrower's income.

The Court concludes that the income component of the DTI has not be proved by the Trust to be untrue or incorrect. Although the Court found that the borrower failed to assess the reasonableness of stated income, the Court does not accept Holt's use of BLS data in this instance: it is not apparent that the salary of a "Regional Manager" for a jewelry-retail chain should be weighed against BLS data for someone working in "Sales and Related Occupations." It also is not apparent whether Holt used the methodology that he applied elsewhere of "grossing up" BLS salary data by 125%, or whether, as indicated in Appendix 1, he calculated salary only according to the 90th percentile of persons working in "Sales and Related Occupations." Again, while the Court finds a guidelines breach based on the underwriter's failure to conduct a reasonableness assessment of income, the Trust has not proved that it is more likely than not that the borrower's stated income was untrue or incorrect.

The undisclosed loans stand on a different footing. There was an undisclosed auto loan of unknown size. There also was an undisclosed loan from the 401k. The Court is unable to determine whether the amount of the 401k loan is included in the loan balance against the $12,540.

The omission of some or all of this additional debt made the DTI ratio on the MLS untrue and incorrect. But to establish a breach of the MLS Warranty, it is not enough to prove that information on the MLS was untrue and incorrect. The MLS Warranty has a separate materiality requirement and the Trust has not proved that it is more likely than not that the additional debt was material to the listed DTI ratio on the MLS. Nor has the Trust established that the true and correct DTI, whatever it may have been, would have placed the loan outside the guidelines DTI ratio.

The Trust has not proven that it is more likely than not that the DTI on the MLS was materially untrue or incorrect. Nor has it proved that the Guideline Warranty was breached by reason of exceeding the guideline's DTI maximum.

5. The LTV Ratio Defects.

Holt states that Cowan's AVM analysis establishes that the loan violated the guidelines based on misstated LTV data. For the reasons explained, the Court rejects the AVM analysis as a basis for recalculating LTV ratio.

6. Material and Adverse Effect at the Time of Discovery or Notice.

The Court has found that the Trust has proved a breach of the Guidelines Warranty based on the underwriter's failure to review the reasonableness of the borrower's stated income, and that the Trust has separately proved a breach of the Guidelines Warranty based on a failure to make further inquiry concerning the existence of the additional undisclosed debt.

The Trust has not proved that these breaches materially and adversely affected the interests of the Certificateholders at the time of discovery or notice. The Court rejects the BLS data applied by Holt in this instance because it is somewhat plausible that a "Regional Manager" for a national jewelry retailer may have earned a salary that approximated the borrower's stated income. It also is not clear how much additional debt was accumulated as a result of the undisclosed auto loan, or what repayment terms were applied to the borrower's loan of $12,500 taken against the 401K account. The guidelines set a maximum DTI ratio of 50%, and with no reliable measure for recalculating the borrower's income or total debt, the Court is unable to determine whether the borrower's actual DTI ratio exceeded the guidelines maximum.

Holt's opinion does not persuade the Court that it is more likely than not these breaches of the Guidelines Warranty materially and adversely affected the interests of the Certificateholders as to Loan 1477160. It is plausible that the loan would have issued under different terms, or not have been issued at all, but it is equally plausible that the borrower's loan would have been approved on the same terms under the existing guidelines. The Court concludes that the 2007-1 Trust has not proved a material and adverse effect on the interests of the Certificateholders at the time of discovery or notice.

P. Loan 158514566.

Loan 158514566 was a stated-income loan originated by Countrywide, and was included in the 2007-3 Trust. Holt lists an "Original DTI" of 43.58% (Appendix 1). He reports "Data Not Available" in the column listing the original stated purpose of the loan (e.g. , whether it was for a cash-out refinance, purchase, etc.) (Id. ). Elsewhere, Holt describes the loan as a rate–and-term refinance of an owner-occupied single family property. The funding date for the subject property was February 13, 2007. The Closing Date for the 2007-3 Trust was May 15, 2007. (PX 182 at 24, 26.)

Holt identified numerous defects in the loan approval process, including income misrepresentation, missing or inadequate title insurance, inconsistent information in the loan file, unreasonable stated income, a document deficiency, incorrect DTI ratio listed in the MLS and used in underwriting and incorrect LTV ratio.

1. Income Misrepresentation and Reasonableness of Stated Income.

The borrower's loan application stated that the borrower worked at Naples Community Hospital as a secretary for 27 years. A loan application signed by the borrower and dated January 25, 2007 stated an income amount of $7,600. (PX L13588 at 0600.) A second loan application signed by the borrower and dated January 26, 2007 stated an income amount of $9,800. (PX L6166 at 0361; PX L13588 at 0629.) A third loan application signed by the borrower and dated February 13, 2007 was consistent with the first application, and stated an income amount of $7,600. (PX L6166 at 0012.) The loan file also contains unsigned, undated loan applications for the borrower, which stated an income amount of $9,800. (PX L6166 at 0367, 377; PX L13588 at 0635, 0645.) In one instance, that $9,800 listing is circled; the Court is unable to discern whether that marking was made by an original underwriter or an individual who reviewed the file post-origination. (PX L6166 at 0367.)

Holt identifies a guidelines breach based on the borrower's inconsistent statements as to reported income. Holt quotes the Countrywide Technical Manual, section 1.5.4, revised June 26, 2006, which states in part, "Files containing application inconsistencies will be required to be processed using full or alternative documentation." Holt opines that the borrower's inconsistent statements as to income required full or alternative documentation, but that the loan was nevertheless underwritten as a reduced documentation loan.

While standing alone the inconsistencies do not establish borrower misrepresentation, they do establish an inconsistency in statements that should have required further scrutiny by the underwriter, including requiring further documentation. The first-dated and last-dated loan applications are consistent in stating an income of $7,600. In between, the borrower signed a loan application that listed the higher income of $9,800, which was subsequently revised downward to the original figure.

Separate from the apparent inconsistency, Holt asserts that there was no assessment of the reasonableness of income. The BLS data indicates that the 90th percentile of income for a medical secretary in the borrower's region, grossed up by 125%, was $4,038.54 monthly, or $48,462.50 annually. This is contrasted with the lower of the two figures on the loan applications of $7,600 per month or $91,200.

Grissom asserts that the guidelines did not require a check of BLS data. Though it may be true that the guidelines did not specifically reference BLS data, she does not contend that there was a requirement to assess the reasonableness of income. She offers no other metric that would render the stated income to be apparently reasonable.

Grissom's rebuttal also lists several compensating factors, including 27 years of employment, a FICO score of 741, history of perfect payments on real-estate credit lines and revolving credit lines and $201,846 in verified asset reserves. True, the assets, credit score and payment history are among factors that an underwriter may consider on a reasonableness of an income of $91,200. But, giving due consideration to the opinions of both experts, these factors alone do not provide cause to dispense with a further assessment of reasonableness of income.

The Court finds Holt's opinion in this instance to be more persuasive. The combination of a BLS inquiry and the inconsistent statement of monthly income ($7,600 vs. $9.800) should have caused the underwriter to inquire further.

While not known to the underwriter at the time of origination, the borrower's loss-mitigation servicing file contains an employment letter stating that, in 2008, the borrower earned $33,827.04 annually. Holt notes that the borrower's 2007 income is not disclosed in the employment letter, but that the borrower held the same job, secretary in a hospital, in both 2007 and 2008.

Grissom challenged Holt's use of the borrower's 2008 income as reflected in the loss-mitigation servicing employment letter, stating that post-origination salary information should not be considered when assessing the borrower's 2007 income. When questioned about this conclusion at trial, however, Grissom testified that the borrower's income likely did not change significantly from 2007 to 2008. (Tr. 33: "Q. Would you agree that the nature of the borrower's employment was such that you would not expect a substantial change in income from year to year? A. Barring any leave of absence or anything like that, I think you are correct.") Grissom also testified that she drew a bright line against ever considering post-origination information. (Tr. at 534-35.) The Court finds that the underwriter breached the guidelines by not assessing the reasonableness of the borrower's stated income. The 2007-3 Trust has therefore proven a breach of the Guidelines Warranty.

It is more likely than not that the income of this borrower who worked as a hospital secretary was, at the time of origination, at or about $34,000, and not $91,200 claimed as the lower of the two income figures. The sheer size of the disparity bespeaks of an intentional misrepresentation of income. The Court finds that it is more likely than not that the borrower's income was intentionally misstated.

2. DTI Ratio.

Holt opines that when the borrower's actual, lower income is taken into account, the borrower's DTI ratio jumps from the 43.58% disclosed on the MLS to 134.11%.

For reasons discussed more fully above, the Court accepts Holt's opinion and concludes that the borrower's income as a hospital secretary as of the Closing Date of the PSA was more likely at or about the $34,000 annually and not the $91,200 claimed (the lower of the two inconsistent figures) during the loan application process. This income discrepancy in income rendered the MLS DTI ratio to be materially untrue and incorrect. The Trust has proved a breach of the MLS Warranty relating to the DTI ratio because the DTI has been proved to be untrue and not correct as of the Closing Date.

The underwriting guidelines set a maximum allowable DTI ratio of 38%. The recalculated DTI ratio may have been as high as 134.11% but was certainly above the 38% guidelines maximum. The Trust has proved that is more likely than not that there was a breach of the Guidelines Warranty relating to DTI ratio.

3. Document Deficiency.

Holt was not able to locate a simultaneous second-lien note in the loan file. He states: "The simultaneous 2nd lien Note was not available at the time of this loan review. A review of this document may change the Auditor's analysis of the loan file."

In rebuttal, Grissom states that the applicable guidelines do not require documentation of a second lien to be placed in the loan file for the first-lien mortgage loan. She also notes that the underwriter had access to the second-lien documents because the second lien was simultaneously being underwritten by the same lender. In reply, Holt cites to no guidelines requirement about second-lien documentation.

The Court finds that the Trust has not proved a breach of the Guidelines Warranty based on this document deficiency. The Trust cites no relevant guideline requirement concerning the documentation of a second-lien mortgage when a first-lien mortgage is being underwritten.

4. Title Insurance.

Holt identifies a breach of the Title Insurance Warranty and a separate breach of the Guidelines Warranty based on the absence of a title insurance policy in the loan files. For the reasons discussed above, the Court concludes that the Trust has not proved a violation of the Title Insurance Warranty.

In addition, Grissom identified a copy of the final title policy in the borrower's loan file. The presence of the title policy in the loan file also is indication that a title insurance policy was obtained prior to the funding of the loan, and therefore does not support a breach of the Guidelines Warranty.

5. LTV Breaches.

Holt states that Cowan's AVM analysis establishes that the loan violated the Guidelines Warranty and the MLS Warranty based on misstated LTV data. For the reasons previously explained, the appraised value is an opinion which has not been shown to be other than honestly held. Moreover, the Court has rejected exclusive reliance upon an AVM analysis as a basis for recalculating LTV ratio.

6. Material and Adverse Effect at the Time of Discovery or Notice.

The Court finds that the established breaches of the MLS Warranty and the Guidelines Warranty materially and adversely affected the interests of the Certificateholders at the time of discovery or notice. The Court has found that the borrower intentionally misrepresented income, and that misrepresentation goes to the borrower's commitment to repay the loan. This increased the risk of loss to the Certificateholders at the time of discovery or notice.

Because a reasonableness assessment would have revealed that the income was overstated, the Guidelines breach also materially and adversely affected the Certificateholders because compliance with the Guidelines would have revealed a DTI ratio that exceeded the guidelines maximum, and the loan would either not have been funded or funded on different terms more favorable to the owner of the loan.

The Court finds that the breach of the MLS Warranty as to the misstated DTI ratio materially and adversely affected the interests of the Certificateholders because the Trust would have rejected the inclusion of the loan or would have priced the transaction differently.

The 2007-3 Trust has proved each element of its claim against UBS with respect to Loan 158514566 and UBS is obligated to repurchase the loan or, if it has been liquidated, then to pay the Trust the money damages equivalent.

Q. Loan 1470345.

Loan 1470345 was a stated-income loan originated by American Home, and was included in the 2007-1 Trust. The loan was for the rate-and-term refinance of an owner-occupied home. The original DTI ratio was calculated as 28.82%. The funding date for the subject property was October 18, 2006. The Closing Date for the 2007-1 Trust was January 16, 2007. (PX 110 at 33, 36.)

Holt identified numerous defects in the loan, including income misrepresentation, unreasonable stated income, undisclosed mortgage debt with red flags, incorrect DTI ratio listed in the MLS and used in underwriting, and incorrect LTV ratio.

1. Unreasonableness of Stated Income.

The borrower stated monthly earnings of $26,000 (or annual income of $312,000) based on his self-employment as the owner of a handyman remodeling business, which he had operated for 15 years. The co-borrower stated monthly earnings of $10,500 based on employment as a nurse at a hospital and a senior living facility. The combined annual income for the two borrowers would have been $438,000. At trial, Holt opined that instead of listing his actual personal income, the borrower likely listed the total gross revenue of his handyman business. (Tr. 888.)

In loss-mitigation servicing, the two borrowers submitted documentation showing that their 2007 annual income was $57,077, and that the two borrowers continued in the same employment as stated at origination. Holt opined that although the 2007 income post-dated origination, the borrowers' incomes were likely stable from 2006 to 2007, and that "it was reasonable to assume" that the borrowers earned similar income in 2006.

At trial, counsel questioned Holt as to whether the borrower's income through a handyman and remodeling business may have decreased from 2006 to 2007 based on broader economic trends:

Q. If the nation were to experience an unprecedented decline in the price of homes, do you think that could affect someone's home remodeling business?

A. I don't know. He's doing handyman type business, so I don't know exactly—you know, what he's doing. That's a good question. If you're an underwriter and you're the owner of a handyman remodeling business, my first inclination is to document that in the loan file. What is this business? Is he out cutting trees or cleaning out gutters, or is he actually gutting a kitchen and redoing it?

So on the face of this, I'd want to look further. But in terms of answering your question, it's possible. But I want to see exactly what business is he in.

(Tr. 890.)

Holt concluded that there was no evidence that the borrowers' stated income was reviewed for reasonableness, as required by the guidelines. Holt identified red flags that he opines called into question their stated monthly incomes, including five revolving credit cards with active balances of $44,344, totaling 60% of the borrowers' credit limits, as well as total liquid assets of $5,040. Grissom notes, however, that the borrowers maintained a separate account with $108,507.98 in liquid assets.

Grissom also cites to various compensating factors that would justify an exception to the guidelines, including the borrowers' lengthy credit history, history of perfect payment on 19 real estate trade lines and strong FICO scores, but there is no written indication that an "exception" was "exercised." These same factors are some evidence that could bear on the reasonableness of stated income on the theory that a person with a good credit history may be making an income sufficient to meet his or her obligations.

But giving due consideration to both experts, the Court finds that the Trust has proved that it is more likely than not that the underwriting guidelines were breached because the underwriter should have but did not conduct an inquiry as to the reasonableness of combined stated income of $438,000. This finding is based on several facts identified by Holt. First, a decline of the housing market does not fully explain the disparity between the borrowers' 2006 stated annual income of $438,000 and their 2007 annual income of $57,077. Even if the housing market affected the bottom line of the borrower's self-owned handyman business, the co-borrower continued her employment, for which she claimed to have earned $126,000 in 2006—more than double the income of both borrowers in 2007. The significant disparity between the borrowers' 2006 and 2007 incomes cannot be explained solely by a change in the demand for handyman services. Additionally, the borrowers' high revolving credit debt and 60% utilization further support the conclusion that the borrowers misstated an annual income of $438,000, and raised a red flag that should have prompted the underwriter to review the reasonableness of stated income.

The Court finds that the Trust has proved that a breach of the Guidelines Warranty because there is no evidence that the borrowers' stated income was reviewed for reasonableness.

2. Undisclosed Mortgage Debt.

Holt concluded that the borrowers failed to disclose mortgages that Holt identified using MERS and a servicing credit report. Prior to the subject property's funding date, the borrower's refinanced an existing $768,365 mortgage and opened a second mortgage on the same property for $230,000. Holt notes that the credit report obtained during the underwriting process reflected multiple mortgage inquiries, which should have raised a red flag leading the underwriter to inquire further.

The Court concludes that the additional credit inquiries raised a red flag that should have led the underwriter to inquire further as to these undisclosed mortgages. An inquiry through MERS would have disclosed the mortgages. As discussed below, the undisclosed mortgages would have materially increased the borrowers' DTI ratio.

The Court finds that the Trust has proved a breach of the MLS Warranty because the DTI ratio listed in the MLS, in material respects, was not true and correct in all material respects as of the Closing Date for the 2007-1 Trust.

The Court also finds that it is more likely than not the materially inaccurate DTI ratio breached the guidelines, which set a maximum DTI ratio of 45%. There is no evidence that an exception was exercised. The Trust therefore has proved a breach of the Guidelines Warranty as to DTI ratio.

3. LTV Breaches.

Holt states that Cowan's AVM analysis establishes that the loan violated the Guidelines Warranty and the MLS Warranty based on misstated LTV data. For the reasons previously explained, the appraised value is an opinion which has not been shown to be other than honestly held. Moreover, the Court has rejected exclusive reliance upon an AVM analysis as a basis for recalculating LTV ratio.

4. Material and Adverse Effect at the Time of Discovery or Notice.

The Court finds that the established breaches of the MLS Warranty and the Guidelines Warranty materially and adversely affected the interests of the Certificateholders at the time of discovery or notice. The Court has found that the borrower concealed the existence of mortgage debt that undermines the borrower's commitment to repay the loan. Also, had the mortgages been disclosed, the loan either would not have funded or would have been funded on different terms more favorable to the owner of the loan.

The Guidelines Warranty breach also materially and adversely affected the Certificateholders because compliance with the guidelines would have revealed a DTI ratio that exceeded the guidelines maximum, and the loan would either not have been funded or funded on different terms more favorable to the owner of the loan.

The Court finds that the breach of the MLS Warranty as to the misstated DTI ratio materially and adversely affected the interests of the Certificateholders because the Trust would have rejected the inclusion of the loan or would have priced the transaction differently.

The 2007-1 Trust has proved each element of its claim against UBS with respect to Loan 1470345 and UBS is obligated to repurchase the loan or, if it has been liquidated, then to pay the Trust the money damage equivalent.

R. Loan 158609489.

Loan 158609489 was a full-documentation loan originated by Countrywide, and was included in the 2007-3 Trust. The loan was for a cash-out refinance of the borrower's primary residence. Appendix 1 lists "Original DTI" as 23.92%. The funding date for the subject property was February 28, 2007. The Closing Date of the 2007-3 Trust was May 15, 2007. (PX 182 at 24, 26.)

Holt identified numerous defects in the loan, including occupancy misrepresentation with red flags, income misrepresentation, undisclosed mortgage debt, property violation, DTI ratio defects and breaches of the MLS Warranty as to DTI ratio and occupancy.

1. Income Misrepresentation.

The borrower's final loan application stated that the borrower worked as a sales representative at an automobile dealership for four years and had monthly earnings of $11,482, or $137,784 per year. As noted, the subject property transaction closed on February 28, 2007.

In October 2010, the borrower filed for Chapter 13 bankruptcy protection. In the borrower's bankruptcy filings, the borrower reported a 2007 annual income of $116,725.55 and a 2008 annual income of $130,854. Holt opines that these bankruptcy filings are evidence that the borrower misstated his income.

With due consideration of Holt's opinion, the Court finds that the 2007-3 Trust has not proved that the borrower misrepresented his income. The bankruptcy filings reflect the borrower's annual income for 2007 and 2008, but the subject loan closed in February 2007. Presumably, the borrower's annual income of $137,784 was calculated based on previous earnings. The disparity of near-year earnings, as reflected in the bankruptcy filings, is not so large as to bespeak of misrepresentation. This is particular true because the borrower worked as a salesman at an auto dealership and not at a steady salaried position.Perhaps most telling of all, the W-2 form in the loan file for 2006 stated that the borrower earned $142,457 for the year. Holt was questioned about the W-2 and a January 2007 paystub for the borrower, which reflected a monthly income of $14,484, a sum that was higher than the monthly income claimed by the borrower. Holt agreed that he and his team "ignored" the 2006 W-2 form contained in the loan file. (Tr. 900-01.) When questioned by the Court, Holt maintained his opinion that the borrower had nevertheless misrepresented income, and testified that he had a practice of relying on bankruptcy filings even when they were contradicted by other documents in the loan file. (Tr. 902-08.) This reasoning makes little sense and is not credited by the Court.

Holt acknowledges (Appendix 1) that this was a full-documentation loan and not a stated-income loan. Holt does not opine that the income was insufficiently documented, and instead relies entirely on bankruptcy filings for subsequent years, which reflect varying incomes that do not significantly depart from the income claimed by the borrower during underwriting.

The Court finds that the 2007-3 Trust has not proved that the borrower made a misrepresentation as to income. Indeed, given that a W-2 Form for 2006 and a paystub from January 2007 both reflected incomes that exceeded the figure listed by the borrower, it appears that the borrower's claimed income was conservatively calculated. Even if these conflicting materials in the loan file are not considered, the bankruptcy materials cited by Holt do not reflect a variance of income so substantial that they would support a finding that the borrower misrepresented income when applying for the loan.

2. Occupancy Misrepresentation and the Related "Property Violation" Defect.

Holt identifies breaches related to the borrower's stated intent to occupy the subject property, and the borrower's obtaining a mortgage on an additional property one month after the subject loan was funded. Holt opines that borrower misrepresented that the subject property would be owner-occupied, and that the subject loan was unreasonably approved given that the borrower had recently listed the subject property for sale, thereby amounting to a "Property Violation."

The subject loan was underwritten for the purpose of a cash-out refinance of an owner-occupied property. In the Dataverify database, Holt determined that on June 9, 2006, approximately eight months before the subject transaction closed, the borrower changed the address of his driver's license to reflect an address that was different from the subject property. Holt also identified within the loan file a statement from a "Multiple Listing Service" prepared for marketing purposes, which listed the subject property as being for sale.⁴ This "Multiple Listing Service" entry was withdrawn on January 15, 2007, approximately one-and-a-half months before the subject loan issued.

4 The findings of Holt and Grissom abbreviate this "Multiple Listing Service" item as "MLS." In the context of discussing the "Multiple Listing Service," their reference to "MLS" are not to be confused with the Mortgage Loan Schedule.
--------

In the loan file, Grissom identifies a letter of explanation from the borrower, which states that the borrower had previously listed the subject property for sale, but had decided against relocating. The letter states:

I've been asked to explain my reason for trying to sell my house recently. I was seriously considering relocating. The frequency of hurricanes and subsequent Insurance increases left me feeling the

desire to move out of Florida. After having my house on the market for three months and not having any offers I'd accept, I've decided to stay.

(PX L13607 at 0091.)

In addition, Grissom has identified a written exercise of an exception to the guidelines. (PX L13607 at 223-25.) Under the guidelines (Countrywide Technical Manual section 1.6.2, revised July 5, 2005), a cash-out refinance of an owner-occupied home was not permitted if the property was listed for sale within the preceding six months, unless a written exception was exercised. In this instance, the loan file contained the written exercise of an exception. In identifying a "First Lien Non-Mustang Rules Violation," the exception states, "Properties Listed for Sale," with "Comments," "Requesting exception for property being listed for sale w/I last 6 months. Listing cancelled 1/15/07." (PX L13607 at 0223-24.)

In reply, Holt opines that the exercise of an exception was unreasonable "when taking into consideration the entire loan file," but the countervailing factors he is referring to are not made clear.

There is conflicting evidence as to whether the borrower intended to occupy the subject property at the time of underwriting. Holt cites to the borrower's change of address on a driver's license and the listing of the subject property in a "Multiple Listing Service" until about six weeks prior to the issuing of the subject loan. But in this case, the loan file also contains a letter of explanation stating that the borrower had intended to relocate from the subject property before deciding to remain. It also contains a formal approved exception for the listing of the property for sale.

After reviewing the total mix of information and the opinions of Holt and Grissom, the Court finds that a reasonable underwriter could have properly concluded that the borrower would, in fact, continue to occupy the subject property. The underwriters did not overlook any red flags, but instead considered them and concluded that the borrower intended to occupy the subject property. With due regard to the opinions of the two experts, the Trust has failed to prove that the exception was unreasonably exercised.

The Court therefore finds that the 2007-3 Trust has not proved that the borrower misrepresented the property's occupancy. The Trust has failed to prove a breach of the Guidelines Warranty because due consideration was given to the explanation furnished by the borrower and a formal exception was exercised in a reasonable manner with regard to the MLS listing.

However, the Court accepts Holt's identification of a breach of the MLS Warranty. The MLS listed the subject property as owner-occupied. On March 28, 2007, the borrower closed on a $462,673 mortgage for a separate property. This property had the same address that the borrower used when he changed his address on his government-issued driver's license approximately eight months before the subject loan was funded. As noted, the Closing Date for the 2007-3 Trust was May 15, 2007. Therefore, between the February 28 funding of the subject property and the May 15 Closing Date for the 2007-3 Trust, the borrower purchased an additional home at the same address he had previously listed when changing his government-issued ID.

The Court accepts Holt's opinion, and finds that it is more likely than not that the MLS misstated that the subject property was owner-occupied and therefore the MLS Warranty was breached. There is no inconsistency between finding a breach of the MLS Warranty as to occupancy but no breach of the Guidelines Warranty as to the same: the underwriters considered the data indicating that the borrower had contemplated vacating the property and exercised a reasonable exception, but after the subject transaction closed, the borrower proceeded with the purchase of the additional property. It is more likely than not that the borrower's occupancy status changed in the three months between the funding date of the subject loan and the Closing Date of the 2007-3 Trust. Considering the total mix of information and the opinions of Holt and Grissom, the Court therefore finds a breach of the MLS Warranty as to occupancy status, but no breach of the Guidelines Warranty.

3. Undisclosed Mortgage Debt with Red Flags.

Holt states that during Countrywide's post-closing review of the loan, it learned that on March 28, 2007, the borrower opened a $462,673 loan for the purchase of an additional property. Holt identified multiple recent inquires on a credit report, with no indication that the borrower submitted a letter of explanation or that the underwriter inquired further. Grissom does not dispute that the borrower opened this additional mortgage or that the credit inquiries were red flags.

Based on the opinions of Holt and Grissom, the Court finds that a reasonable underwriter would have inquired further as to the basis of the credit inquiries. Because the underwriter failed to do so, the borrower had $462,673 in additional undisclosed debt, which increased the borrower's DTI ratio accordingly.

Further, the Court finds that it is more likely than not that the borrower intentionally misrepresented his debt. It is more likely than not that he was in the process of applying for this additional loan while the subject loan was being underwritten. Because he had submitted a letter of explanation related to his occupancy, he knew or should have known that the status of additional property or mortgages was important to the underwriting process.

The Court therefore finds that the 2007-3 Trust has proved a breach of the Guidelines Warranty as to this undisclosed mortgage debt, and that the borrower's failure to disclose the debt was due to his intentional misrepresentation.

4. DTI Ratio Defect.

Holt opines that when the borrower's additional undisclosed mortgage debt and misstated income are taken into account, the borrower's DTI ratio increases from 23.92% to 57.58%.

As discussed, the Court finds that the Trust has not proved that the borrower had misstated his income, but that it has proved that the borrower had $462,673 in mortgage debt. While the Court is unable to calculate with precision a new DTI ratio that does not weigh the lower income amount, it is nevertheless the case that the additional undisclosed debt would cause a significant increase in the borrower's DTI ratio, likely above the guidelines maximum of 50%, and well above the MLS's listed DTI ratio of 23.92%.

The Court therefore finds that the Trust has proved a breach of the MLS Warranty as to the DTI ratio, and that it is more likely than not that the loan was issued in violation of the guidelines maximum of 50%.

5. Material and Adverse Effect at the Time of Discovery or Notice.

The Court finds that the established breaches of the MLS Warranty and the Guidelines Warranty materially and adversely affected the interests of the Certificateholders at the time of discovery or notice. The Court has found that the borrower concealed the existence of mortgage debt that undermines the borrower's commitment to repay the loan. Also, had the mortgages been disclosed, the loan either would not have funded or would have been funded on different terms more favorable to the owner of the loan.

The Guidelines Warranty breach also materially and adversely affected the interests of the Certificateholders because compliance with the guidelines would have revealed a DTI ratio that exceeded the guidelines maximum, and the loan would either not have been funded or funded on different terms more favorable to the owner of the loan.

The Court finds that the breach of the MLS Warranty as to the misstated DTI ratio materially and adversely affected the interests of the Certificateholders because the Trust would have rejected the inclusion of the loan or would have priced the transaction differently.

The 2007-3 Trust has proved each element of its claim against UBS with respect to Loan 158609489 and UBS is obligated to repurchase the loan or, if it has been liquidated, then to pay the Trust the money damage equivalent.

S. Loan 1416080.

Loan 1416080 was a stated-income loan originated by American Home, and was included in the 2007-1 Trust. The loan was for the purchase of the borrower's second home. Appendix 1 lists "Original DTI" as 34.57%. The funding date for the subject property was October 9, 2006. The Closing Date for the 2007-1 Trust was January 16, 2007. (PX 110 at 33, 36.)

Holt identified numerous defects in the loan, including undisclosed mortgage debt with red flags, undisclosed additional property, housing history violation and DTI ratio defects.

1. Undisclosed Mortgage Debt.

Using the Accurint and Site X databases, and the borrower's servicing credit report, Holt identified mortgages that were not disclosed by the borrower. On October 2, 2006, the borrower refinanced a mortgage on an undisclosed property for the amount of $130,000. The borrower also refinanced the mortgage of the borrower's primary residence on January 17, 2006, for the amount of $246,400. As noted, the funding date for the loan was October 9, 2006.

Holt states that recent credit inquiries should have been a red flag that prompted the underwriter to inquire further, but that there was no letter of explanation contained in the loan file.

Using Site X and MERS, Holt identified one property owned by the borrower that was not identified in the underwriting process. This property had one of the undisclosed mortgages also identified by Holt. Grissom does not dispute that this property was not considered during the underwriting.

The Court finds that a reasonable underwriter would have recognized the credit inquiries as a red flag and inquired further. The underwriter did not do so, and the borrower's additional mortgage debt therefore was not considered during underwriting. The Court concludes, after considering the opinions of the experts and the mix of information, that the failure to inquire further as to the credit inquires was a violation of the guidelines and that the Trust has therefore proved a breach of the Guidelines Warranty. The Court finds persuasive Holt's opinion that if "underwriter questioned this red flag, the undisclosed property and mortgage would have been discovered."

2. Housing History Violation.

Holt concluded that the borrower's loan file did not adequately document that the borrower had missed no payments in the preceding 12 months on the borrower's primary residence, as required by the relevant guidelines (the Products and Programs Guidelines, dated May 18, 2006). However, Grissom identified the borrower's credit report in the loan file, which showed a 16-month mortgage history with no late payments.

In reply, Holt notes that the subject loan closed in October 2006, and that the credit report does not reflect borrower payment information for September 2006. But again, as previously discussed with Loans 40588193 and 1416352, Holt has shown, at most, that the loan file does not contain payment history for the month preceding the closing of the subject property. This is evidence of a credit check, but is not evidence of a late or missing payment.

The Court finds that the Trust has failed to prove that it is more likely than not that underwriter failed to review the borrower's payment history and therefore failed to prove that UBS breached the Guidelines Warranty in this respect.

3. DTI Ratio Defect.

Using the borrower's undisclosed mortgage debt, Holt has recalculated the borrower's actual DTI ratio as 60.27%, which varies from the 34.57% DTI ratio listed in the MLS. It also exceeds the maximum qualifying DTI ratio under the guidelines, which was 45%.

Because the DTI ratio listed in the MLS was not true and correct as of the Closing Date, the Trust has proved a breach of the MLS Warranty. It also has proved a breach of the Guidelines Warranty because the loan's true DTI ratio exceeded the guidelines maximum.

4. Material and Adverse Effect at the Time of Discovery or Notice.

The Court finds that the breach of the MLS Warranty materially and adversely affected the interests of the Certificateholders at the time of discovery or notice. Had the MLS Warranty accurately listed the borrower's DTI ratio, the Trust would have rejected the inclusion of the loan or would have priced the transaction differently.

Moreover, the nature of the borrower's failure to include other property and the mortgages on that property was not likely to have been the subject of inadvertence or mistake; instead, the omission was likely intentional deceit on the part of the borrower. This intentional misrepresentation further bolsters the conclusion that the breaches materially and adversely affected the interests of the Certificateholders because they affect the borrower's willingness to pay.

The Court also finds that the breach of the Guidelines Warranty materially and adversely affected the interests of the Certificateholders at the time of discovery or notice because the presence of the undisclosed mortgages and the correlated effect on DTI ratio resulted in a DTI ratio that exceeded the guidelines maximum. Had the mortgages been disclosed, and the DTI ratio calculated accordingly, the loan either would not have funded or would have been funded on different terms more favorable to the owner of the loan.

The 2007-1 Trust has proved each element of its claim against UBS with respect to Loan 1416080 and UBS is obligated to repurchase the loan or, if it has been liquidated, then to pay the Trust the money damages equivalent.

T. Loan 1437593.

Loan 1437593 was a stated-income loan originated by American Home, and was included in the 2007-1 Trust. The loan was for a cash-out refinance of non-owner occupied property. Holt identifies the "Original DTI" as 31.3%. (Appendix 1.) The funding date for the subject property was October 18, 2006. The Closing Date of the 2007-1 Trust was January 16, 2007. (PX 110 at 33, 36.)

Holt identified numerous defects in the loan, income misrepresentation with red flags, missing final title, DTI ratio defects and incorrect LTV ratio.

1. Income Misrepresentation.

The borrower's final loan application stated that for 4.5 years, the borrower had been self-employed as the owner of a home-sales business. The borrower's stated monthly income was $9,850, or $118,200 annually.

The borrower filed for Chapter 7 bankruptcy protection on December 17, 2007. In bankruptcy filings, the borrower stated that his annual income for 2006, the year in which the loan was funded, was $9,509.76, which would average to $792.48 per month.

Under the relevant guidelines (Products and Programs Guidelines, dated September 27, 2006), the underwriter was required to review the stated income for reasonableness. There is no indication in the loan file that the borrower's income was reviewed for reasonableness. Holt identifies certain red flags as to the reasonableness of the borrower's stated income, including revolving debt balances of $30,724, which was 90% of the borrower's credit limit.

The Court finds that a reasonable underwriter would have recognized the borrower's high revolving credit balance and the status as a self-employed sales person, as a red flag and reviewed the stated income for reasonableness. There is no indication that the underwriter performed such a review.

The Court therefore finds that the 2007-1 Trust has proved a breach of the Guidelines Warranty based for the failure to assess the reasonableness of the borrower's income.

The Court finds that it is more likely than not that the borrower intentionally misrepresented his or her stated income, which was approximately ten times higher than the 2006 total income that was listed in the borrower's bankruptcy filing. There is no apparent explanation for this extreme discrepancy.

2. Correlated DTI Ratio Defects.

When the borrower's actual 2006 income is considered, as reflected in bankruptcy filings, the borrower's DTI ratio increases to 436.76%. This significantly exceeds the 31.3% DTI ratio listed in the MLS and the 35.729% DTI ratio used in the American Home automated underwriting software.

The Court finds that the 2007-1 Trust has proved a breach of the MLS Warranty because the actual DTI ratio varied materially from the DTI ratio listed in the MLS.

The Court also finds that the Trust has proved a breach of the Guidelines Warranty because the actual DTI ratio was far higher than the DTI ratio calculated during underwriting. The relevant guidelines (Products and Programs Guidelines, dated September 27, 2006) established a maximum allowable DTI ratio of 45%. Because the DTI ratio was significantly higher, the loan was issued in violation of the guidelines, and the Trust has therefore proved a breach of the Guidelines Warranty.

3. Missing Title Insurance.

Holt opined that the loan files did not contain a title insurance policy. For the reasons discussed above, the Court concludes that the Trust has not proved a violation of the Title Insurance Warranty.

In addition, Grissom cites to the HUD-1 statement contained in the loan file, which includes a line entry reflecting that the borrower obtained title insurance. (See PX L10566 at 0138.) Grissom also identified a title commitment to issue a policy. (See PX L10566 at 0174-80.) These materials are indication that a title insurance policy was obtained prior to the funding of the loan, and therefore do not support a breach of the Guidelines Warranty.

4. LTV Breaches.

Holt states that Cowan's AVM analysis establishes that the loan violated the guidelines based on misstated LTV data. For the reasons previously explained, the appraised value is an opinion which has not been shown to be other than honestly held. Moreover, the Court has rejected exclusive reliance upon an AVM analysis as a basis for recalculating LTV ratio.

5. Material and Adverse Effect at the Time of Discovery or Notice.

The Court comfortably finds that the breaches of the MLS Warranty and the Guidelines Warranty materially and adversely affect the interests of the Certificateholders at the time of discovery or notice. The borrower's materially false statements about his or her own income reflect on the borrower's willingness and ability to repay the loan.

The Court finds that the breach of the MLS Warranty materially and adversely affected the interests of the Certificateholders at the time of discovery or notice. Had the MLS Warranty accurately listed the borrower's DTI ratio, the Trust would have rejected the inclusion of the loan or would have priced the transaction differently.

The Court also finds that the breach of the Guidelines Warranty materially and adversely affected the interests of the Certificateholders at the time of discovery or notice because the DTI ratio far exceeded the guidelines maximum. Had the borrower's income been accurately calculated, and the DTI ratio calculated accordingly, the loan either would not have funded or would have been funded on different terms more favorable to the owner of the loan.

The 2007-1 Trust has proved each element of its claim against UBS with respect to Loan 1437593 and UBS is obligated to repurchase the loan or, if it has been liquidated, then to pay the Trust the money damage equivalent.

CONCLUSION: STEPS TO FINAL JUDGMENT.

The trial is over. With the exception of calculation of damages and withdrawal of claims of breach by the Trusts, the evidence is closed. But findings of fact and conclusions of law based upon the existing trial record remain to be made as to thousands of loans. The Court will outline the steps to advance this case to entry of final judgment.

A. Supplemental Submissions on Loan Files.

As noted above, the 14,403 loan files have been provisionally received into evidence and for limited purposes. Troubling anomalies exist in the loan files that call into question whether they are in fact the loan files that existed at the time of origination, as represented to this Court. These anomalies include, but are not limited to, the inclusion in the files of analyses, seemingly prepared for litigation, of guideline compliance and whether a violation of the guidelines materially and adversely affected the Certificateholders. (PX L792 at 0011-12; PX L300 at 1076; PX L 1621 at 0087.) In one instance, an analysis for one borrower was included with the origination documents of a different borrower. (PX L1621.)

Within 14 days hereof, the Trusts, by affidavit, shall explain (1) how the analyses found their way into files purportedly digitized at or near the time of origination; (2) whether the inclusion of materials not generated at the time of origination is systemic, aberrational or limited to one or more subset of loan files; and (3) why it is not reasonable to assume that a purportedly missing document may simply have been placed in a different borrower's file or lost. UBS may respond 7 days thereafter. The Court, as noted, has reserved the right to amend its findings, including by striking the loan files, any findings dependent upon the content of the loan files and, where appropriate, any opinions that are premised upon the loan files.

B. Appointment of Masters.

Rule 52(b), Fed. R. Civ. P., empowers the Court to make additional findings in an action not tried to a jury. Rule 53(a)(1)(B), Fed. R. Civ. P., further empowers a Court to appoint a master to "hold trial proceedings and make or recommend findings of fact on issues to be deiced without a jury if appointment is warranted by: (1) some exceptional condition...." The exceptional condition is the sheer volume of loans, several thousand for which findings and conclusions are necessary. Rule 53(a)(1)(C) allows the appointment of a master to "address... postrial matters that cannot be effectively and timely addressed by an available district judge or magistrate judge of the district." Again, the sheer volume proves that the process "cannot be effectively and timely addressed" by the available resources in this District.

The Court will enter an order, after hearing from the parties, appointing a master (the qualifier "special" has been abandoned by the Federal Rules) who will serve as the Lead Master in this action. The order of appointment, to be drafted by the parties, will delineate the powers and authority of the Lead Master but a brief overview of the Court's intended scope of responsibilities is appropriate.

The Lead Master will develop a plan for the timely entry of findings of fact and conclusions of law relating to all loans not covered by the rulings herein. The plan will provide for the appointment by the Court of additional masters to prepare recommended findings and conclusions consistent with the rulings herein. The Lead Master may solicit candidates for appointment as masters and make recommendations to the Court as to appointments and compensation of these masters. The Lead Master may consider a plan for uploading trial evidence into a secure cloud facility for ready access by the masters and a web portal for the masters to upload their findings and conclusions. The Lead Master may propose uniform formats and time schedules for masters to report their proposed findings and conclusions to the Court. The Lead Master will be empowered to meet with the parties in an effort to secure stipulations to conform this Court's findings and conclusions to the universe of loans and otherwise to promote settlement, including by directing the parties to resume discussions before their privately-retained mediator.

The Lead Master may meet and confer with masters but will not be empowered to alter or modify the finding or conclusions of a court-appointed master. The Lead Master may issue orders consistent with the powers vested in the order of appointment. Rule 53(d). The Lead Master but not the masters recommending proposed findings of fact and conclusions of law, may communicate with the Court ex parte on any matter within the order of appointment but shall not communicate any position of a party on settlement. Rule 53(b)(2)(B). The Lead Master may explore whether there is an agreement as to the standard of review of the masters' findings and conclusions, as provided for in Rule 53(f)(3)(A) & (B).

The compensation of the Lead Master and any other master to be appointed will be shared equally by both sides. The Court retains the power to amend this interim allocation. Rule 53(g).

The Trusts and UBS shall meet and confer on recommendations of persons for appointment as the Lead Master and the content of the order of appointment. Each side may provide the names and resumes of up to five candidates for appointment of Lead Master within 21 days, together with a marked copy of an order of appointment showing the points on which there is disagreement.

SO ORDERED.