15-496-cv(L), 15-499-cv(Con)
United States ex rel. O’Donnell v. Countrywide Home Loans, Inc.

           UNITED STATES COURT OF APPEALS
               FOR THE SECOND CIRCUIT
                    ______________

                        August Term 2015

    (Argued: December 16, 2015            Decided: May 23, 2016)

                   Docket Nos. 15-496, 15-499



     UNITED STATES EX REL. EDWARD O’DONNELL,
                                   Plaintiff-Appellee,

                                 - v. -

         COUNTRYWIDE HOME LOANS, INC.,
             COUNTRYWIDE BANK, FSB,
    BANK OF AMERICA, N.A., and REBECCA MAIRONE,
                                 Defendants-Appellants,

            COUNTRYWIDE FINANCIAL CORP.
             and BANK OF AMERICA CORP.,
                                        Defendants. *
                    ______________




*The Clerk of Court is respectfully directed to amend the caption as set
forth above.
Before:
          RAGGI, WESLEY, and DRONEY, Circuit Judges.
                       _________________

       Appeal from a judgment entered by the United States
District Court for the Southern District of New York (Rakoff, J.).
A jury found Defendants-Appellants liable under the Financial
Institutions Reform, Recovery, and Enforcement Act of 1989 for
mail or wire fraud affecting a federally insured financial
institution, arising from the sale of mortgages to government-
sponsored entities. At the penalty stage, the District Court
imposed penalties exceeding $1.2 billion. On appeal,
Defendants-Appellants argue, inter alia, that the proof at trial is
insufficient under the mail and wire fraud statutes as a matter of
law. We agree and accordingly REVERSE the judgment of the
District Court.
                       _________________

      KANNON K. SHANMUGAM, Williams & Connolly LLP,
Washington, DC (Brendan V. Sullivan, Jr., Enu A. Mainigi, Craig
D. Singer, Williams & Connolly LLP, Washington, DC; Richard
M. Strassberg, William J. Harrington, Goodwin Procter LLP,
New York, NY, on the brief), for Defendants-Appellants Countrywide
Home Loans, Inc., Countrywide Bank, FSB, and Bank of America,
N.A.
       E. JOSHUA ROSENKRANZ, Orrick, Herrington & Sutcliffe
LLP, New York, NY (Robert M. Loeb, Kelsi Brown Corkran,
Orrick, Herrington & Sutcliffe LLP, Washington, DC; Marc L.
Mukasey, Michael C. Hefter, Ryan M. Philp, Seth M. Cohen,
Bracewell & Giuliani LLP, New York, NY, on the brief), for
Defendant-Appellant Rebecca Mairone.


                                1
       PIERRE G. ARMAND, Assistant United States Attorney
(Joseph N. Cordaro, Carina H. Schoenberger, Benjamin H.
Torrance, Assistant United States Attorneys, on the brief), for
Preet Bharara, United States Attorney for the Southern District of
New York, New York, NY, for Plaintiff-Appellee.
       Seth P. Waxman, Daniel Aguilar, Sina Kian, Wilmer
Cutler Pickering Hale and Dorr LLP, Washington, DC; Noah A.
Levine, Alan E. Schoenfeld, Wilmer Cutler Pickering Hale and
Dorr LLP, New York, NY, for Amici Curiae The Clearing House
Association, L.L.C., American Bankers Association, Financial Services
Roundtable, and Chamber of Commerce of the United States of
America.
       Dennis M. Kelleher, Better Markets, Inc., Washington, DC,
for Amicus Curiae Better Markets, Inc.
                         _________________

WESLEY, Circuit Judge:
        When can a breach of contract also support a claim for
fraud? This question—long an issue in common-law courts—
comes before us in the context of a judgment in the United States
District Court for the Southern District of New York (Rakoff, J.),
imposing civil penalties exceeding $1.2 billion on Defendants-
Appellants Countrywide Home Loans, Inc.; Countrywide Bank,
FSB; Bank of America, N.A. (collectively, “Countrywide”); and
Rebecca Mairone (together with Countrywide, “Defendants”)
under the Financial Institutions Reform, Recovery, and
Enforcement Act of 1989 (“FIRREA”), 12 U.S.C. § 1833a. As the
necessary predicate for these penalties, the Government alleged
that Defendants violated the federal mail and wire fraud statutes
by selling poor-quality mortgages to government-sponsored
entities. On appeal, Defendants argue that the evidence at trial
shows at most an intentional breach of contract—i.e., that they

                                 2
sold mortgages that they knew were not of the quality promised
in their contracts—and is insufficient as a matter of law to find
fraud. We agree, concluding that the trial evidence fails to
demonstrate the contemporaneous fraudulent intent necessary
to prove a scheme to defraud through contractual promises.
Accordingly, we reverse with instructions to enter judgment in
favor of Defendants.

                          BACKGROUND
       This case arises in the context of the post-financial-crisis
restructuring of the Full Spectrum Lending Division (“FSL”) of
Countrywide Home Loans. Prior to the events at issue in this
case, FSL had been the subprime lending division of
Countrywide; after the collapse of the subprime market in 2007,
Countrywide undertook a transformation of FSL into a prime
origination division with the goal of selling prime loans 1 to two
government-sponsored enterprises (“GSEs”): the Federal
National Mortgage Association (“Fannie Mae”) and the Federal
Home Loan Mortgage Corporation (“Freddie Mac”). The overall
reorganization of FSL was referred to as “Central Fulfillment,”
one component of which was a loan origination process 2 called
the “High Speed Swim Lane” or “HSSL,” introduced in August
2007 and expanded in October 2007. Rebecca Mairone, the only

1The terms “prime” and “subprime” refer to mortgage loans with
relatively lower and higher credit risks, respectively. See Pension Ben.
Guar. Corp. ex rel. St. Vincent Catholic Med. Ctrs. Ret. Plan v. Morgan
Stanley Inv. Mgmt. Inc., 

, 715 (2d Cir. 2013).
2A loan origination process refers to a “work flow”—i.e., a series of
operational steps taken to evaluate a particular loan application for
approval. It is distinct from loan origination guidelines—i.e., the
criteria for approving a particular loan.


                                   3
named individual defendant, was the Chief Operating Officer of
FSL during 2007 and 2008 and was responsible for overseeing
FSL’s reorganization, including the implementation of HSSL.
        This case originated in February 2012 as a qui tam suit
under the False Claims Act (“FCA”), 31 U.S.C. § 3729 et seq.,
commenced by Edward O’Donnell, a former employee of
Countrywide. Subsequently, the Government intervened, added
claims under section 951 of FIRREA, 12 U.S.C. § 1833a—which
imposes civil penalties for violations of the federal mail and wire
fraud statutes that “affect[] a federally insured financial
institution”—and named Countrywide Home Loans, Inc.,
Countrywide Financial Corp., Countrywide Bank, FSB, Bank of
America Corp., Bank of America, N.A., and Mairone as
defendants. As a result of a later motion to dismiss and amended
complaint, the FCA claims, Bank of America Corp., and
Countrywide Financial Corp. were removed from the case,
leaving only FIRREA claims against the remaining defendants. It
is on these claims and against these defendants that the case
ultimately went to trial.
      At trial, the Government presented the following
evidence relevant to our consideration here. 3 Pursuant to
contracts with Fannie Mae, Countrywide as the seller of
mortgages represented that, “as of the date [of] transfer,” the
mortgages sold would be an “Acceptable Investment.” J.A. 5905,



3As discussed infra note 9, on this appeal we draw all inferences in
favor of the Government, do not make credibility determinations or
weigh the evidence, and disregard all evidence favoring Defendants
that the jury was not required to believe. See Reeves v. Sanderson
Plumbing Prods., Inc., 

, 149–51 (2000).


                                 4
5908, 5935, 5938. 4 Similarly, Freddie Mac’s selling guide 5
contained a representation by the seller—again, Countrywide—
that “all Mortgages sold to Freddie Mac have the characteristics
of an investment quality mortgage.” J.A. 6368; 6 see also J.A. 6366
(representing quality “[a]s of” the date the loans were delivered
to Freddie Mac). The Government adduced no evidence and
made no claim that Countrywide had fraudulent intent during
the negotiation or execution of these contracts.


4 The contracts define “Acceptable Investment” as one for which the
lender (i.e., Countrywide) knows “of nothing . . . that can reasonably
be expected to cause private institutional investors to regard the
mortgage as an unacceptable investment; cause the mortgage to
become delinquent; or adversely affect the mortgage’s value or
marketability.” J.A. 5908.
5Although the specific Countrywide–Freddie Mac contract
incorporating the selling guide does not appear in the record on
appeal, there was trial testimony that Countrywide’s contract
obligated it to make the representations contained within Freddie
Mac’s selling guide. J.A. 2976–77.
6The selling guide defines an “investment quality” mortgage as one
“that is made to a Borrower from whom repayment of the debt can be
expected, is adequately secured by real property and is originated in
accordance with the requirements of the Purchase Documents.” J.A.
6368. This definition is similar to the one contained in Countrywide’s
Technical Manual: “An investment quality loan is one that is made to a
borrower from whom timely repayment of the debt can be expected, is
adequately secured by real property, and is originated in accordance
with Countrywide’s Technical Manual (CTM) and Loan Program
Guides (LPGs).” J.A. 5959. Although slightly varying terms are used,
no party argues there is any substantive difference between the
representations; accordingly, we will refer to them all as “investment-
quality representations.”


                                  5
       The Government’s theory is that Countrywide sold loans
under these purchase agreements to the GSEs, knowing that the
loans were not investment quality and thus intending to defraud
them. To support this argument, the Government presented
extensive evidence of quality problems in the loans approved
through the HSSL program. See J.A. 1839–41, 1848–49, 1863–66,
2220–25, 2228–30, 3313–20, 4437–44, 5650, 5988, 5998. The
Government also identified three FSL officers (the “Key
Individuals”) as to whom they alleged fraudulent intent:
Mairone; Greg Lumsden, President of FSL; and Cliff Kitashima,
Chief Credit Officer of FSL. J.A. 3516; see also J.A. 5220. To
demonstrate the requisite intent, the Government presented
evidence that the Key Individuals were informed of the poor
quality of HSSL loans by FSL employees and internal quality
control reports and nonetheless sold them to the GSEs. See J.A.
1890–900, 2237–43, 2250–53, 2255–62, 3416–18, 3364–70, 3486–91,
6063–66, 6716–22, 7002–05, 7019–22, 7178. 7
       With respect to the Key Individuals, the Government also
presented evidence that at least Kitashima and Mairone knew of
the investment-quality representations made in the contractual

7On appeal, Defendants also challenge the exclusion of certain defense
witnesses, arguing that the District Court inconsistently applied its
determination that only testimony from witnesses that was relevant to
the Key Individuals’ intent would be admitted. However, they do not
argue that the testimony of the Government’s witnesses should have
been excluded, merely that they should have been given an
opportunity to rebut that testimony with comparable witnesses.
Because on sufficiency review we must disregard any evidence
contesting the Government’s proof in any event, see, e.g., 

–51, and because we answer the sufficiency question in
Defendants’ favor, resolving Defendants’ evidentiary arguments is not
necessary for our decision here.


                                  6
documents between Countrywide and the GSEs. See J.A. 3800–
01, 4324. The Government presented no evidence that any of the
Key Individuals were involved in the negotiation or execution of
these contracts, nor did it present evidence that any of them
communicated with either GSE regarding the loans sold; in fact,
Defendants elicited testimony from GSE witnesses to the
contrary. See J.A. 2764–65, 3041; see also J.A. 3800–01, 4304–05. 8
The Government’s case rested upon facts showing that the Key
Individuals knew of the pre-existing contractual representations,
knew that the loans originated through HSSL were not
consistent with those representations, and nonetheless sold
HSSL loans to the GSEs pursuant to those contracts. For
example, in its closing argument, the Government summarized
as follows:
           And now that all the evidence has come in,
           this case still comes down to a few simple
           facts. First, the Hustle loans were bad.
           Second, the defendants knew the Hustle
           loans were bad. And third, the defendants
           passed the Hustle loans off as good loans
           anyway to cheat Fannie and Freddie out of
           money.
J.A. 5009; see also J.A. 5006–07, 5020, 5041, 5049, 5147–48, 5153.

8Although we construe all evidence in favor of the Government, we
must also review the record as a whole and credit “uncontradicted and
unimpeached” evidence put forth by the moving party, “at least to the
extent that that evidence comes from disinterested witnesses.” 

–51; accord Cameron v. City of New York, 

, 59–
60 (2d Cir. 2010). The Government has not disputed—either at trial or
on appeal—this characterization of the Key Individuals’ interactions
with the GSEs.


                                   7
       After closing arguments, the jury was charged as to the
elements of federal mail and wire fraud. In particular, the jury
was instructed that it had to find a scheme to defraud, which
was defined as “a plan or design to obtain money or property by
means of one or more false or misleading statements of a
material fact.” J.A. 5219. The District Court defined a false
statement as “an outright lie” and a misleading statement as
“true as far as it goes but creat[ing] a false impression by
omitting information necessary to correct the false impression.”
J.A. 5219. The jury was charged that the Government’s theory
was that “the defendants devised a scheme to induce [the GSEs]
to purchase mortgage loans originated through [HSSL] by
misrepresenting that the loans were of higher quality than they
actually were,” and was further charged that “the fact that some
of these alleged misrepresentations may have constituted
breaches of the contracts . . . is neither here nor there.” J.A. 5219–
20. Second, the jury was charged that it needed to find that “the
defendant you are considering participated at some point in the
scheme knowingly and with a specific intent to defraud”—that
is, “act[ed] consciously and deliberately . . . [with] knowledge
that that defendant was participating in a fraudulent scheme”
and “purposely intended to deceive and harm [the GSEs] by
seeking to sell them mortgage loans . . . through false or
misleading representations.” J.A. 5220. The jury was also
charged that, as to Countrywide, it could only find fraudulent
intent if “at least one of [the Key Individuals] participated in
such a fraudulent scheme with such intent.” 

the jury returned a general verdict in
favor of the Government, whereupon the District Court imposed
civil penalties of $1 million against Mairone individually and
$1.27 billion against Countrywide. See 12 U.S.C. § 1833a(b)(1),
(b)(3)(A).


                                  8
                             DISCUSSION
         The provision of FIRREA under which Defendants were
found liable provides for civil penalties against “[w]hoever”
violates or conspires to violate, inter alia, the federal mail or wire
fraud statutes, see 18 U.S.C. §§ 1341, 1343, in a manner “affecting
a federally insured financial institution.” 12 U.S.C. § 1833a(a),
(c)(2). 9 Defendants inform us that this suit is the first in the
federal courts of appeals to consider the validity of a FIRREA
action brought against a financial institution for so-called “self-
affecting” conduct. Much of the parties’ and amici’s briefing
concerns the validity of such an action. Ultimately, however, we
need not reach the issue, because Defendants have persuaded us
to reverse with another argument: the Government has failed to
prove the necessary FIRREA prerequisite—i.e., a violation of (or
conspiracy to violate) § 1341 or § 1343.10

9 Our interpretation of the federal statutes in question is de novo.
Auburn Hous. Auth. v. Martinez, 

, 143 (2d Cir. 2002).
Following our articulation of the legal standards, however, we draw
all evidentiary inferences in favor of the Government and do not make
credibility determinations or weigh the evidence. See Stampf v. Long
Island R.R., 

, 197–98 (2d Cir. 2014). Considering the
evidence in this manner, we determine de novo whether judgment as a
matter of law is warranted—i.e., whether “‘a reasonable jury would
not have a legally sufficient evidentiary basis to find for the [non-
movant] on that issue.’” 

(alteration in original)
(quoting Fed. R. Civ. P. 50(a)(1)). Because the proper interpretation of
the contracts at issue is a question of law, we review them de novo as
well. See Magi XXI, Inc. v. Stato della Città del Vaticano, 

, 720
(2d Cir. 2013).
10Our conclusion here thus renders immaterial the other grounds for
appeal put forth by Defendants, including challenges to the District
Court’s evidentiary rulings and penalty calculations. As for


                                     9
       A simple hypothetical presents the central issue in this
case. Imagine that two parties—A and B—execute a contract, in
which A agrees to provide widgets periodically to B during the
five-year term of the agreement. A represents that each delivery
of widgets, “as of” the date of delivery, complies with a set of
standards identified as “widget specifications” in the contract.
At the time of contracting, A intends to fulfill the bargain and
provide conforming widgets. Later, after several successful and
conforming deliveries to B, A’s production process experiences
difficulties, and the quality of A’s widgets falls below the
specified standards. Despite knowing the widgets are subpar, A
decides to ship these nonconforming widgets to B without
saying anything about their quality. When these widgets begin
to break down, B complains, alleging that A has not only
breached its agreement but also has committed a fraud. B’s fraud
theory is that A knowingly and intentionally provided
substandard widgets in violation of the contractual promise—a
promise A made at the time of contract execution about the
quality of widgets at the time of future delivery. Is A’s willful but
silent noncompliance a fraud—a knowingly false statement,
made with intent to defraud—or is it simply an intentional
breach of contract?
       This question, not an unusual one at common law, poses a
novel issue in the context of the federal fraud statutes before us.
Supreme Court precedent instructs us to apply the common-law
understanding of fraud principles to these statutes, absent
inconsistency with their text. Once we do so, however, the trial
record reveals a basic deficiency in proof under the statutes, and

Defendants’ request for judicial reassignment following any remand,
we need not consider that request, as our remand will require only
entry of a new judgment.


                                 10
accordingly, we conclude the evidence is insufficient to sustain
the jury’s verdict.
I.     The Common Law’s Treatment of Fraud Claims Based
       Upon Breaches of Contract
        On appeal, Defendants argue—as they did in the District
Court—that the conduct alleged and proven by the Government
is, at most, a series of intentional breaches of contract. The
common law, they contend, does not recognize such conduct as
fraud, and as a result, the federal statutes do not either.
Specifically, Defendants argue that—because the only
representations involved in this case are contained within
contracts—to demonstrate fraud, rather than simple breach of
contract, under the common law and federal statutes, the
Government had to prove that Defendants never intended to
perform those contracts—i.e., at the time of contract execution,
Defendants knew and intended that they would not perform
their future obligations thereunder.
       In both pre- and post-trial decisions, the District Court
concluded that the federal fraud statutes do not incorporate the
common-law principle that actions brought in fraud cannot be
premised solely upon evidence of contractual breaches—or, in
the alternative, that the scheme alleged here fell into one of the
recognized exceptions to this principle for actions premised on
contractual breaches that nonetheless can sustain an action for
fraud. See United States ex rel. O’Donnell v. Countrywide Fin. Corp.
(Countrywide II), 

, 533–34 (S.D.N.Y. 2015);
United States ex rel. O’Donnell v. Countrywide Fin. Corp.
(Countrywide I), 

, 607–08 (S.D.N.Y. 2013).
However, the law compels a different analysis that would not
permit a reasonable jury to find a § 1341 or § 1343 violation on
the facts of this case.


                                11
       The federal mail and wire fraud statutes, in relevant part,
impose criminal penalties on “[w]hoever, having devised or
intending to devise any scheme or artifice to defraud, or for
obtaining money or property by means of false or fraudulent
pretenses, representations, or promises” uses the mail, 18 U.S.C.
§ 1341, or wires, 

for such purposes. Thus, the essential
elements of these federal fraud crimes are “‘(1) a scheme to
defraud, (2) money or property as the object of the scheme, and
(3) use of the mails or wires to further the scheme.’” United States
v. Binday, 

, 569 (2d Cir. 2015) (quoting Fountain v.
United States, 

, 255 (2d Cir. 2004)). “The gravamen of
the offense is the scheme to defraud, and any ‘mailing that is
incident to an essential part of the scheme satisfies the mailing
element,’ even if the mailing itself ‘contain[s] no false
information.’” Bridge v. Phoenix Bond & Indem. Co., 

,
647 (2008) (alteration in original) (citation omitted) (quoting
Schmuck v. United States, 

, 712, 715 (1989)). The exact
contours of what kinds of conduct constitute a “scheme to
defraud” have been the subject of some judicial discussion.
       It is well established that statutes employing common-law
terms are presumed, “unless the statute otherwise dictates, . . . to
incorporate the established meaning of these terms.” Nationwide
Mut. Ins. Co. v. Darden, 

, 322 (1992) (internal
quotation marks omitted); accord United States v. Castleman, 134 S.
Ct. 1405, 1410 (2014). The Supreme Court has expressly applied
this rule to the term “scheme to defraud,” holding that the
statutes require proof—as at common law—that the
misrepresentations were material, notwithstanding the fact that
a solely “natural reading of the full text” would omit such an
element. Neder v. United States, 

, 21, 25 (1999) (internal
quotation marks omitted). The Court rejected certain
requirements of common-law fraud (i.e., reliance and damages)


                                12
as clearly “inconsistent” and “incompatible” with “the language
of the fraud statutes,” which prohibit “the ‘scheme to defraud,’
rather than the completed fraud.” 

By contrast, the Court
incorporated the common-law requirement of materiality into
the statutes because it was neither inconsistent nor incompatible.

task here is to determine whether the common-law
principles on which Defendants rely are incompatible with the
language of the federal statutes.
       As we summarized above, Defendants rely on the
common-law rule that parties cannot allege or prove fraud solely
on the basis of a contractual breach—i.e., the common law
requires more than simply “proof that a promise was made and
that it was not fulfilled” to sustain a fraud claim, Tenzer v.
Superscope, Inc., 

, 30 (1985); see also United States v.
D’Amato, 

, 1261 n.8 (2d Cir. 1994). By contrast, the
Government argues that any contractual relationship between
the defendant and an alleged fraud victim is “irrelevant,” citing
as examples decisions in which this Court and others recognized
a fraud claim where the parties were engaged in a contractual
relationship. Gov’t Br. 43–44. These cases are distinguishable,
however, in that none recognize a contract breach, by itself, to
constitute fraud. Rather, in each, the defendants made
affirmative fraudulent misrepresentations to their contractual
counterparties in the course of performance or to feign
performance under the contract. See, e.g., United States v. Naiman,

, 44, 49 (2d Cir. 2000) (submitting false certifications
of compliance required by contracts with the government).
      Durland v. United States, 

(1896), relied upon
by the District Court, is also inapt because it dispensed with a
completely different common-law rule—that promises of future
performance       could       never     constitute    fraudulent
misrepresentations—on the basis of statutory language clearly

                                 13
designed to reach both fraudulent statements as to the present
and fraudulent promises as to the future. See 

As the
Supreme Court more recently clarified, Durland did not disturb
what fraud at common law requires the Government to prove,
except to the extent it is inconsistent with the statutory language.
See 

. Thus, Durland has little application to
the question posed by this case: what is required to prove a
scheme to defraud when alleged misrepresentations concerning
future performance are contained within a contract?
       In some sense, both the Government and Defendants are
correct: the common law does not permit a fraud claim based
solely on contractual breach; at the same time, a contractual
relationship between the parties does not wholly remove a
party’s conduct from the scope of fraud. What fraud in these
instances turns on, however, is when the representations were
made and the intent of the promisor at that time. As explained
below, where allegedly fraudulent misrepresentations are
promises made in a contract, a party claiming fraud must prove
fraudulent intent at the time of contract execution; evidence of a
subsequent, willful breach cannot sustain the claim. Far from
being “arcane limitations,” Countrywide 

,
these principles fall squarely within the core meaning of
common-law fraud that neither the federal statutes nor Durland
disrupted. See 

(“[Durland] did not hold, as
the Government argues, that the [mail fraud] statute
encompasses more than common-law fraud.”).
      It is emphatically the case—and has been for more than a
century—that a representation is fraudulent only if made with
the contemporaneous intent to defraud—i.e., the statement was
knowingly or recklessly false and made with the intent to induce
harmful reliance. While on the New York Court of Appeals,
then–Chief Judge Benjamin Cardozo wrote that “[a]

                                14
representation even though knowingly false does not constitute
ground for an action of deceit unless made with the intent to be
communicated to the persons or class of persons who act upon it
to their prejudice.” Ultramares Corp. v. Touche, 

, 187
(1931) (emphasis added); see also RESTATEMENT (FIRST) OF TORTS
§§ 526, 531 (1938). Even earlier, the highest common-law courts
in the country routinely espoused the view that any party
wishing to claim fraud must prove that the representation was
actually made with contemporaneous fraudulent intent:
           The representation upon which [a fraud
           claim] is based must be shown not only to
           have been false and material, but that the
           defendant when he made it knew that it was
           false, or not knowing whether it was true or
           false and not caring what the fact might be,
           made it recklessly, paying no heed to the
           injury which might ensue.
Kountze v. Kennedy, 

, 129 (1895) (emphasis added)).
           There can be no question at this date that, in
           an action of deceit, the scienter must not only
           be alleged, but proved, and the jury must be
           satisfied that the defendant made a
           statement knowing it to be false, or with
           such conscious ignorance of its truth as to be
           equivalent to a falsehood. This is the general
           rule, and it has been declared with notable
           emphasis in several recent cases in this state.
Griswold v. Gebbie, 

, 363 (1889); see also, e.g., Shackett v.
Bickford, 

(1906); Nw. S.S. Co. v. Dexter Horton & Co., 

, 568–69 (1902).



                                  15
       Of course, “fraudulent intent is rarely susceptible of direct
proof, and must instead be established by legitimate inferences
from circumstantial evidence.” United States v. Sullivan, 

, 186 (2d Cir. 1969). Nonetheless, where the relevant
representation is made within a contract, the common law rejects
any attempt to prove fraud based on inferences arising solely
from the breach of a contractual promise:
          [T]hat proof that a promise was made and
          that it was not fulfilled is sufficient to prove
          fraud . . . is not, and has never been, a
          correct statement of the law.

. This rule exists because, at common law,
a post-agreement intent to breach the contract is not actionable
as fraud:
          [I]f the promises or representations were
          made in good faith at the time of the
          contract, and the defendant subsequently
          changed its mind, and failed or refused to
          perform the promises, then such conduct of
          the company, originally or subsequently,
          would not constitute such fraud, in legal
          acceptation, as would justify the rescission
          of the contract or the cancellation of the
          deed.
Chi., Tex. & Mex. Cent. Ry. Co. v. Titterington, 

, 224
(1892); see also, e.g., Hoyle v. Bagby, 

, 781 (1961);
Citation Co. Realtors, Inc. v. Lyon, 

, 790–91 (Okla.
1980); Lloyd v. Smith, 

, 145–46 (1928). This principle
has been applied in the context of fraud not only by our Court
but by our sister circuits as well. See 

n.8;
Cohen v. Koenig, 

, 1172 (2d Cir. 1994); Mills v. Polar


                                 16
Molecular Corp., 

, 1176 (2d Cir. 1993); DiRose v. PK
Mgmt. Corp., 

, 632–33 (2d Cir. 1982); see also Corley v.
Rosewood Care Ctr., Inc. of Peoria, 

, 1007 (7th Cir.
2004); McEvoy Travel Bureau, Inc. v. Heritage Travel, Inc., 

, 791–92 (1st Cir. 1990); Lissmann v. Hartford Fire Ins. Co., 

, 53 (4th Cir. 1988); United States v. Kreimer, 

,
128 (5th Cir. 1980). This prohibition is more than, as the
Government attempts to characterize it, “the uncontroversial
view that breach of a contract, without further evidence of
fraudulent intent, does not establish a fraud claim,” Gov’t Br. 49
n.7. Instead, as certain of our sister circuits have convincingly
explained, this principle exists because, at its core, fraud requires
proof of deception, which is absent from ordinary breach of
contract. See McEvoy Travel 

–92; 

; see also Kehr Packages, Inc. v. Fidelcor, Inc., 

, 1417 (3d Cir. 1991).
       Accordingly, the common law requires proof—other than
the fact of breach—that, at the time a contractual promise was
made, the promisor had no intent ever to perform the obligation:
           It is the preconceived design of the buyer,
           formed at or before the purchase, not to pay for
           the thing bought, that constitutes the
           fraudulent concealment which renders the
           sale voidable, and not an intent formed after
           the purchase. If the purchaser forms the
           intent not to pay for the goods after he has
           received them and the title has passed, it is a
           mere intended breach of contract, and not
           such a fraud as to authorize a rescission of
           the sale. . . . This intent never to pay for the
           goods has sometimes been treated as a
           fraudulent representation, and sometimes as

                                   17
          a fraudulent concealment, but in either event
          it must precede the sale. The distinction is
          between an intent not to pay according to
          the terms of the contract and an intent to
          obtain goods under color of a formal sale,
          upon a sham promise to pay, but with the
          design of never paying for them. The former
          is a mere intent to break a contract; the
          latter, an intent to defraud.
Starr v. Stevenson, 

, 218 (Iowa 1894) (emphases
added) (citations omitted); accord 

–24.
          To constitute the fraud, there must be a
          preconceived design never to pay for the
          goods. A mere intent not to pay for the
          goods when the debt becomes due, is not
          enough; that falls short of the idea. A design
          not to pay according to the contract is not
          equivalent to an intention never to pay for
          the goods, and does not amount to an
          intention to defraud the seller outright,
          although it may be evidence of such a
          contemplated fraud.
Burrill v. Stevens, 

, 399–400 (1882). More recently, our
sister circuit expressed the principle succinctly:
          Fraud requires much more than simply not
          following through on contractual or other
          promises. It requires a showing of deception
          at the time the promise is made. A
          subsequent breach, although consistent with
          deceptive intent[,] is not in and of itself
          evidence of such an intent.


                                18


.
        As already observed, our Court has consistently applied
this principle: “A breach of contract does not amount to mail
fraud. Failure to comply with a contractual obligation is only
fraudulent when the promisor never intended to honor the
contract.” 

n.8 (emphasis added)); see
also Murray v. Xerox Corp., 

, 122 (2d Cir. 1987)
(holding that “a showing of fraudulent intent fails as a matter of
law” where no evidence demonstrates the promisor “did not
intend to comply with his promise from its inception”); Ford v.
C.E. Wilson & Co., 

, 617 (2d Cir. 1942) (A. Hand, J.)
(rejecting common-law fraud claim where there was no evidence
of intent not to perform at the time the contract was entered
(citing, inter alia, In re Levi & Picard, 

(S.D.N.Y. 1906);
Starr, 

; Burrill, 

)). The alternate approach—
proving intent only as to the act of breaching the promise,
instead of making the promise—contravenes the fundamental
common-law requirement of contemporaneity between
representation and fraudulent intent.
        More than thirty years ago, our Court discussed the
interaction between fraud and contractual promises in Thyssen,
Inc. v. S.S. Fortune Star, 

(2d Cir. 1985) (Friendly, J.).
That case concerned whether a “deviation” constituted an
“independent, willful tort in addition to being a breach of
contract” for purposes of awarding punitive damages. 

A “deviation” is a term of art in admiralty law, originally
meaning “a departure from the agreed course of the voyage”; it
“amount[ed] to a breach of warranty or condition precedent”
and thus was considered “‘no more than a breach of the contract
of carriage,’” albeit “‘ipso facto a more serious breach than if it
had occurred on land.’” 

(quoting Farr v. Hain S.S. Co.,

, 944 (2d Cir. 1941) (L. Hand, J.)). Notwithstanding

                                 19
the gravity—and obvious materiality—of such a breach, Judge
Friendly, writing for the Court, concluded that even the
intentional and willful deviation at issue could not constitute
fraud absent “an element essential to fraud, namely, an intention
not to perform the promise when made.” 

11 In doing so,
he explained why common law courts do not consider even
serious, intentional, or malicious contractual breaches to be
tortious—notably, he relied on Justice Holmes’s articulation of
the common law’s view of contracts as “simply a set of
alternative promises either to perform or to pay damages for
nonperformance,” and on the common law’s tolerance for, even
encouragement of, so-called “‘efficient breaches’” that increase
overall wealth. 

(citing, inter alia, OLIVER WENDELL
HOLMES, THE COMMON LAW 235–36 (Mark DeWolfe Howe ed.
1963); RESTATEMENT (SECOND) OF CONTRACTS ch. 16, reporter’s
note (1981)); see also 

(“A contract may be
breached for legitimate business reasons. Contractual breach, in
and of itself, does not bespeak fraud, and generally does not give
rise to tort damages.” (citation omitted)).
      Although Thyssen concerned the availability of punitive
damages, not the application of the federal fraud statutes, the
prerequisite question was whether a breach of contract—
acknowledged to be intentional and willful at the time of
breach—could be tortious as a fraud at common law. 

63.
The distinctions Judge Friendly identified in Thyssen were not, as
the Government argues, merely “rooted in the desire not to

11The relevant “promise,” in the context of a deviation, was the
warranty contained in the contract of carriage. See 

–64. Judge Friendly concluded that the case “clearly lacked”
evidence of an intent not to perform this promise when the contract
was entered. 

                                  20
inappropriately expand the scope of civil remedies under
contract law,” Gov’t Br. 49. To the contrary, the decision not to
expand civil contract remedies appears rooted in the nature of
contracts and torts at common law—particularly, the nature of
fraud as deceptive, see, e.g., McEvoy Travel 

–92—and the common law’s reason for treating them
differently. In essence, the Government’s theory would convert
every intentional or willful breach of contract in which the mails
or wires were used into criminal fraud, notwithstanding the lack
of proof that the promisor intended to deceive the promisee into
entering the contractual relationship. 12 The reasons identified by
Judge Friendly in Thyssen counsel with persuasive force against




12As we noted above, deception is the core of fraud and the key
distinction between fraud and a contractual breach. See, e.g., Kehr

; McEvoy Travel 

–92;

. The cases cited by the Government all involve
deceptive conduct that was employed in a contractual relationship to
hide breaches of contract or nonperformance. See, e.g., 

(issuing false certifications required by contracts and
misrepresenting included information); United States v. Frank, 

, 334–36 (2d Cir. 1998) (falsifying billing records for contractual
services); First Bank of the Ams. v. Motor Car Funding, 

,
289, 291–92 (N.Y. 1st Dep’t 1999) (misrepresenting the characteristics
of loans on loan tapes to mask noncompliance with representations
and warranties). In all of these cases, the purported fraudulent
statements or conduct were made outside the four corners of the
contract, albeit related to performance thereunder. The law of these
cases is perfectly consistent with our holding today that fraudulent
intent must be found at the time of the allegedly fraudulent conduct,
and the results are consistent because, as we explain infra Part II, the
Government only proved representations wholly within the contract.


                                  21
such a dramatic expansion of fraud liability in circumstances like
the case before us. 13
       In sum, a contractual promise can only support a claim
for fraud upon proof of fraudulent intent not to perform the
promise at the time of contract execution. Absent such proof, a
subsequent breach of that promise—even where willful and
intentional—cannot in itself transform the promise into a fraud.
Far from being an arcane limitation, the principle of
contemporaneous intent is, like materiality, one without which
“the common law could not have conceived of ‘fraud.’” 

.
       Although Neder does not require that a common-law
principle promote the interests of the federal statute but instead
presumes the common-law meaning is incorporated unless
inconsistent, see 

we note that the contemporaneity
principle does, in fact, promote those interests. Unlike fraud at
common law, the federal statutes require neither reliance by nor
injury to the alleged victim. Compare, e.g., Small v. Lorillard
Tobacco Co., 

, 57 (1999) (injury); Jones v. Title Guar. &
Tr. Co., 

, 419 (1938) (reliance), with 

–25. So, unlike the common law, the statutes punish “the
scheme, not its success.” United States v. Helmsley, 

, 94
(2d Cir. 1991). What gives a scheme its fraudulent nature is, as
Durland explained, “the intent and 

.
Thus, what matters in federal fraud cases is not reliance or injury
but the scheme designed to induce reliance on a known

13Justice Holmes’s theory of alternative promises seems particularly
persuasive where, as here, the parties have a contractually determined
remedy—repurchase—in place for breached obligations. In essence,
the parties bargained precisely in an alternative fashion to provide
investment-quality loans or to repurchase defective loans sold.


                                 22
misrepresentation. See 

–57; United States
v. Regent Office Supply Co., 

, 1180–81 (2d Cir. 1970).
       Accordingly,      we    deem     the   common        law’s
contemporaneous fraudulent intent principle incorporated into
the federal mail and wire fraud statutes. Applying these
principles to a fraud claim based on the breach of a contractual
promise, we conclude that the proper time for identifying
fraudulent intent is contemporaneous with the making of the
promise, not when a victim relies on the promise or is injured by
it. Only if a contractual promise is made with no intent ever to
perform it can the promise itself constitute a fraudulent
misrepresentation.
II.   The Evidence Was Insufficient as a Matter of Law to
      Prove Fraud
        Having described the proof that the federal fraud statutes
require, we conclude the Government’s proof at trial failed to
meet its burden. The only representations alleged to be false
were guarantees of future quality made in contracts as to which
no proof of contemporaneous fraudulent intent was introduced
at trial. The Government did not prove—in fact, did not attempt
to prove—that at the time the contracts were executed
Countrywide never intended to perform its promise of
investment quality. Nor did it prove that Countrywide made
any later misrepresentations—i.e., ones not contained in the
contracts—as to which fraudulent intent could be found.
       Although the Government was not always clear as to
what theory of fraud applied in this case, see, e.g., J.A. 4861–64,
the record shows that the jury was charged only as to a theory of
fraud through an affirmative misstatement, i.e., a statement that
was either “an outright lie” or partially true but “omitt[ed]
information necessary to correct [a] false impression.” J.A. 5219.


                                23
Thus, we review the proof at trial only by reference to this
charged theory, see Yates v. Evatt, 

, 409 (1991), and we
do not address whether other situations, such as silence without
any affirmative statement while under a duty to disclose material
information, can constitute fraud under the federal statutes,
particularly in the context of a breach of contract, cf. United States
v. Gallant, 

, 1228 (10th Cir. 2008) (nondisclosure is
actionable under the federal fraud statutes where there is a duty
to speak); United States v. Altman, 

, 102 (2d Cir. 1995)
(failure to disclose material information while in a fiduciary
relationship constituted a scheme to defraud). 14
       Both to the jury and to this Court, the Government
identified provisions in the contracts between Countrywide and
the GSEs—and only those provisions—as the representations
underlying its fraud claim, despite acknowledging that the
contracts’ execution pre-dated the alleged scheme to defraud. See
Gov’t Br. 43 (arguing that “the government’s claims of mail and
wire fraud were valid despite the preexisting contracts between
the Bank and the GSEs”). In summation, the Government argued
that     these   representations     were     the     “lies”  and
“misrepresentations” that formed “the kernel of the case here.”
J.A. 5147; see also 

5153–54. Before this court, the
Government contends that this proof was sufficient because “no
case cited by defendants holds that fraudulent intent must have
existed at the time of contracting, when the alleged fraud
(inducing the other party to the contract to take action through a

14While the case law reaffirms that fraudulent intent must accompany
silence to constitute fraud, see Sanchez v. Triple-S Mgmt., Corp., 

, 10 (1st Cir. 2007) (citing cases in six circuits); 

,
we need not decide here how fraud through silence in the context of a
contractual relationship would operate.


                                     24
scheme to defraud) occurred later.” Gov’t Br. 44. Of course,
freestanding “bad faith” or intent to defraud without
accompanying conduct is not actionable under the federal fraud
statutes; instead, the statutes apply to “everything designed to
defraud by representations as to the past or present, or suggestions
and promises as to the future.” 

(emphases
added); see also 

(“Fraud never consists in
intention, unless it be accompanied by some act.”). Thus, on the
affirmative misrepresentation theory charged to the jury, the
Government needed to show false or misleading statements
made with fraudulent intent.
       Critically, the Government presented no proof at trial that
any quality guarantee was made with fraudulent intent at the
time of contract execution. Nor did it offer evidence of any other
representations, suggestions, or promises—separate from and
post-dating execution of the initial contracts—that were made
with fraudulent intent to induce the GSEs to purchase loans. In
fact, at oral argument before this Court, counsel for the
Government identified no representations or statements other
than those contained in the contracts and instead argued that the
contractual representations were “made” not at contract
execution but at the point of sale. 15



15See Oral Argument at 1:37:00, United States v. Mairone, Nos. 15-496-
cv(L), 15-499-cv(Con) (argued Dec. 16, 2015) (Mr. Armand, arguing
misrepresentations were made “continuously” after contract execution
at each point of sale); 

(Mr. Armand, arguing that fraud
occurred in the performance of the contract); see also 

(Mr.
Armand, answering in the negative Judge RAGGI’s question whether
there were any representations that would not support a breach of
contract claim).


                                   25
         The plain language of the contracts does not admit this
characterization. 16 In the relevant contractual provisions,
Countrywide “makes” or “warrants and represents” certain
statements (i.e., present-tense acts), including that the future
transferred loan will be investment quality “as of” the transfer or
delivery date. J.A. 5905, 5935, 6366, 6368; see also 

5938.
The use of a present-tense verb in a contract indicates that the
parties intend the act—here, the making of the representation—
to occur at the time of contract execution, not in the future. See
VKK Corp. v. Nat’l Football League, 

, 130 (2d Cir. 2001);
Aspex Eyewear, Inc. v. Altair Eyewear, Inc., 

, 215
(S.D.N.Y. 2005); Fed. Home Loan Mortg. Corp. v. Kopf, No. CV 90-
2375 (RR), 

, at *2 (E.D.N.Y. Jan. 30, 1991) (Raggi,
J.); Ellington v. EMI Music, Inc., 

, 246–47 (2014); see
also Rubenstein v. Mueller, 

, 232 (1967) (holding a
present-tense clause in a will indicated a present intention).


16 We conduct the same inquiry when reviewing motions for judgment
as a matter of law as we do reviewing motions for summary judgment.
See 

(“[T]he standard for granting summary
judgment ‘mirrors’ the standard for judgment as a matter of law, such
that ‘the inquiry under each is the same.’” (quoting Anderson v. Liberty
Lobby, Inc., 

, 250–51 (1986))). When interpreting contractual
language, we “accord that language its plain meaning giving due
consideration to the surrounding circumstances and apparent purpose
which the parties sought to accomplish,” and “[o]nly where the
language is unambiguous” may we construe it as a matter of law.
Palmieri v. Allstate Ins. Co., 

, 187 (2d Cir. 2006) (Sotomayor,
J.) (internal quotation marks omitted). “The mere assertion of an
ambiguity does not suffice to make an issue of fact. Ambiguity resides
in a writing when—after it is viewed objectively—more than one
meaning may reasonably be ascribed to the language used.” 

marks omitted).


                                    26
Similarly, the phrase “as of” is “used to indicate a time or date at
which something begins or ends.” As of, Webster’s New Third
International Dictionary, Unabridged, http://unabridged.
merriam-webster.com; see also As of, Oxford English Dictionary
Online, www.oed.com (“[A]s things stood on (a date); (orig.
U.S.) (in formal dating) reckoning from; from, after.”). As these
definitions indicate, “as of” describes the timing of a state of
affairs, and a state of affairs—i.e., the investment-quality status
of particular loans—is precisely what is being represented in the
contracts at issue.
       Accordingly, the only reasonable interpretation of the
contracts is that the date contained in the “as of” clause identifies
the moment at which the promised fact will exist—i.e., when the
representation becomes effective. Where a party makes a
contractual representation of quality that is effective as of a
future date rather than the time of contract execution, the date of
future effectiveness determines the date of performance (and,
thus, breach), see Deutsche Bank Nat’l Tr. Co. v. Quicken Loans Inc.,

, 866 (2d Cir. 2015), but the promisor’s intent to
perform on that promise is fixed as of contract execution, see Sabo
v. Delman, 

, 160 (1957) (concluding that a party’s
intent with respect to representations of future acts is a “material
existing fact” at the time of contract execution upon which a
fraud claim may lie). 17 The Government urges us to read the


17Not all representations of fact are made with a future effectiveness
date. For example, the warranty in ABB Industries Systems, Inc. v. Prime
Technology, Inc., 

, 360 (2d Cir. 1997), promised “in the land-
sale contract” that a piece of real property was in compliance with
state and federal environmental laws as of the date of sale. In such a
case, a party’s intent to perform and its performance (or breach) are
simultaneous—both take place at contract execution.


                                   27
relevant contract provisions as, in essence, promises at execution
to make future representations as to quality. The language of the
provisions, however, constitutes a present promise, made at the
time of execution, to provide investment-quality loans at the
future delivery date. The plain and objective meaning of the
contract simply does not support the Government’s contention
that Countrywide actually made these representations—rather
than merely set their performance date—at the time of the
subsequent sales of loans. Thus, to the extent its fraud claim is
based on these contractual representations of quality, it
necessarily fails for lack of proof that, at the time of contract
execution, Defendants had no intent ever to honor these
representations.
       Because we conclude that the contracts unambiguously
make the representations at the time of contract execution,
extrinsic evidence—such as witness testimony—cannot vary that
meaning. See Seiden Assocs., Inc. v. ANC Holdings, Inc., 

, 428 (2d Cir. 1992). Thus, we examine the other evidence
presented at trial solely for the purpose of determining whether
the jury had a sufficient basis for concluding that other,
noncontractual fraudulent misrepresentations occurred to
induce the sale of HSSL loans. 18
      The testimony of the GSE employees, as well as former
Countrywide employees, focused on the meaning and

18As we noted above, the jury was not charged as to—and therefore
could not have found—liability as a result of fraudulent silence.
Accordingly, the Government had to prove some affirmative
statement—either wholly false or partially true but misleading—as to
which the requisite scienter was present. See also supra note 12
(discussing the Government’s reliance on cases involving fraudulent
conduct separate from contractual promises).


                                28
importance of the contractual representations but did not
identify any promise, statement, or representation outside of the
contract made to induce loan sales or to mask nonperformance.
For example, an employee of Freddie Mac testified that he
understood the contractual representation to mean that “the
information that they’re presenting to us at time of sale is
accurate,” which describes the timing of the representation’s
content, not the underlying promise itself. J.A. 2974. Other
testimony from Fannie Mae and former Countrywide employees
emphasized the importance of these representations to the GSEs’
business models and their applicability to each loan sold—
ostensibly to prove the materiality of the misrepresentations,
which was hotly contested at trial. No witness identified
additional promises or statements that could serve as the “false
or misleading statements” the jury was charged to find. J.A.
5219. Nor, as we have noted, does the Government identify any
on appeal, relying solely—as it did at trial—on the contracts
themselves. Accordingly, we conclude that any finding by the
jury that a post-execution representation occurred to induce the
sale would be premised on a legally erroneous reading of the
contracts or “the result of sheer surmise and conjecture.” 

(internal quotation marks omitted). 19

19Where, as we conclude here, the only misrepresentations alleged and
proven are wholly contained within the contract, there is no factual
basis to find, as the District Court did, that the exception in New York
common law for “collateral misrepresentations” applies. See, e.g.,
Torchlight Loan Servs., LLC v. Column Fin., Inc., No. 11 Civ. 7426(RWS),

, at *9–10 (S.D.N.Y. July 25, 2012); Varo, Inc. v. Alvis
PLC, 

, 265 (N.Y. 1st Dep’t 1999). We therefore need not
determine whether this exception or others at New York common law,
see Bridgestone/Firestone, Inc. v. Recovery Credit Servs., Inc., 

,
19–20 (2d Cir. 1996), are incorporated into the federal statutes. We note


                                    29
       In sum, the Government has never argued—much less
proved at trial—that the contractual representations at issue
were executed with contemporaneous intent never to perform,
and the trial record contains no evidence that the three Key
Individuals—or anyone else—had such fraudulent intent in the
contract negotiation or execution. Instead, the Government’s
proof shows only post-contractual intentional breach of the
representations. Accordingly, the jury had no legally sufficient
basis on which to conclude that the misrepresentations alleged
were made with contemporaneous fraudulent intent. Because we
construe the federal mail and wire fraud statutes to require such
proof, consistent with the common law, the Government has not
proven the prerequisite violation necessary to sustain an award
of penalties under FIRREA.

                          CONCLUSION
       For the reasons stated above, we REVERSE the judgment
of the District Court and REMAND the case with instructions to
enter judgment for Defendants.



that the two cases from our Circuit on which the Government relies—
Frank and Naiman—present fact patterns similar to those in cases
falling within the “collateral misrepresentations” exception. However,
neither Frank nor Naiman confronted an argument based on the
fraud/contract distinction that Defendants make here: Frank concerned
challenges to the jury instructions and sufficiency of the evidence on
the intent to cause harm, 

–37, and Naiman concerned
sufficiency challenges as to materiality and intent to cause harm, 

. Accordingly, while Frank and Naiman are certainly
consistent with a theory of fraud through “collateral
misrepresentations,” we cannot say they have expressly approved the
exception.


                                 30