“And we played the Hustle music.  There were, you know, printed materials passed out,” with dance steps so “ideally we could all perform the Hustle in precision,” recalled the former Countrywide first vice president. “There was a lot of excitement.  There was a lot of fanfare. It was fun.”  He was describing events in the summer of 2007, when Countrywide decide to speed up its process for approving loans, using a program called the “High Speed Swim Lane,” or “HSSL” (or “Hustle”).  The music stopped after the global financial crisis.  Bank of America bought out the failing Countrywide Financial.  In 2013, federal prosecutors secured a $1.27 billion jury verdict against Bank of America for the “Hustle” scheme of the subprime operation of its Countrywide subsidiary.  Just a few weeks ago, the Second Circuit Court of Appeals reversed.

The ruling did not go unnoticed.  Some asked whether the ruling in US ex rel. O’Donnell v. Countrywide Home Loans was a blow against Government efforts to hold financial institutions accountable for fraud, which have themselves been criticized as anemic.  Others took it on face value as a narrow ruling, focusing on the requirement that prosecutors show intent during a fraudulent scheme.  Some viewed the ruling as sensibly distinguishing between a breach of a contract, which parties negotiate, and a separate tort or fraud.  On closer inspection, though, the Second Circuit’s ruling should not be taken on face value and it raises still greater concerns: not only does it narrow the law of fraud by picking and choosing doctrine selectively, but it delicately and misleadingly avoids the factual record in the case.  Overturning a jury verdict is no small matter, but overturning a billion-dollar jury verdict amply supported by the very facts that the Second Circuit claimed had been missing, raises still more troubling questions.

A qui tam plaintiff and whistleblower, Edward O’Connor, a former employee of Countrywide, initiated the lawsuit, which the Government joined, adding fraud claims under the Financial Institutions Reform, Recovery, and Enforcement Act of 1989 (“FIRREA”), 12 U.S.C. § 1833a, which establishes civil penalties for violations of criminal statutes, including the federal mail and wire fraud statutes, if the conduct was “affecting a federally insured financial institution.”  The suit alleged that the subprime operation at Countrywide systematically misrepresented the quality of mortgage loans sold to Fanny Mae and Freddy Mac, almost $5 billion dollars worth, using the “Hustle” program.  Lower level executives knew it and expressed concerns, sending notes up the chain like:  does “the request to move loans mean we no longer care about quality?” The Government summarized the evidence as follows in its closing arguments:

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.

As the trial court judge Jed Rakoff noted, “internal reports showed that more than half of the loans that the HSSL processing system had ‘cleared to close’ were ineligible for sale to any investor, even though those loans were to be sold to Government-sponsored entities.”  On appeal, though, the defendants argued that the evidence at trial showed at most an intentional breach of a contract, by selling mortgages known not to be of the quality promised—which they argued was insufficient as a matter of law to find fraud.

Was this just a lawyer’s trick?  If so, the Second Circuit swallowed it hook, line, and sinker.

The line between contract and fraud is an important one, but just because a party signs a contract does not mean the other party has free reign to lie, cheat, and steal.  The Second Circuit emphasized that the relevant contracts stated, “as of the date [of] transfer,” the mortgages sold would be an “[a]cceptable Investment.”  The mortgages sold were not of that quality and Countrywide employees knew it.  What was the problem with calling this fraud?  The Second Circuit said that “[t]he Government adduced no evidence and made no claim that Countrywide had fraudulent intent during the negotiation or execution of these contracts.”  But the Second Circuit claimed that “present-tense verb[s]” used like “makes” or “warrants and represents” and the use of “as of” all tell a story that the only warranties and representations made were at the time of the contract.

Yet the contracts themselves tell a completely different story, when one reads the entire documents and not just the snippets quoted by the panel.  The warranties were as of the date of transfer, not when the contract was initially signed.  The contract with Fannie Mae states clearly that the initial contract is just one of the sets of terms governing subsequent sales.  That contract allowed Countrywide to be an approved seller.  But each subsequent sale was also governed by its Guides, including amendments, and a written commitment to purchase.  The contract was crystal clear that the mortgages bought must meet the terms “on the day we make our written commitment,” that is on the date when the sale occurs.  The future-oriented timing could not be more clear.  Moreover, “the fact that we have signed this Contract does not mean that we must make a commitment to purchase any mortgage.”  And any sales require certain warranties to Fannie Mae, that “apply to each mortgage,” individually, and that “are made as of the date transfer is made to us.”   Again, the language is crystal clear that representations are made on the date of sale.  Indeed, those warranties also “continue after the purchase,” and the seller must “repurchase a mortgage” if the warranty made is untrue (something that Countrywide was unwilling to later do).

The evidence showed, as the Government pointed out on appeal, that “representations and warranties were made with each loan sale.”  The Contract itself showed it, but there was not just the one master contract.  Each new sale involved a separate deal, a “written commitment” by the buyer, and new representations made.  Moreover, the contracts stated that representations and warranties would apply at the time of sale or delivery.  The fraudulent misrepresentations were “made continually in the period after the contract was formed.” The Government added in its appellate brief: “That is sufficient to sustain a charge of fraud, regardless of whether there was fraudulent intent at the time of contracting.”  That is not a “contract breach, by itself.”  It was an ongoing and systematic deception.

As the former vice president put it in his deposition testimony, there was “a separate swim lane for loans with reduced documentation.”  It was understood that there was a “reasonability test” for every loan to be approved.  Yet the loans were being signed up en masse with shoddy review.  Loan specialists “couldn’t leave for the night” if they didn’t hit their goals to “love” loans.  And as the vice president testified at trial, “if you follow the policies in the selling guide and you sell a loan to Fannie Mae, you are rep[resenting] and warranting that you have followed their – their guidelines.”  The vice president certainly understood that selling loans that did not meet the representation and warranties “would be considered a fraud.”

The Second Circuit could not have it both ways, and the panel knew it, conceding: “a contractual relationship between the parties does not wholly remove a party’s conduct from the scope of fraud.”  But the panel concluded: “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.”  Thus, since the intentionally deceptive contract also touched on contract provisions, the panel found that it was immunized from subsequent fraud.  The Second Circuit relied upon its conclusion that: “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.”  That is not factually correct.  Not only did the Contract clearly call for future representations, truthful ones, but the Government relied substantially on representations made when each mortgage was sold, even running through individual examples of such false information in its complaint.

The Second Circuit panel then said “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.”  That is not correct either—the Government alleged and proved to the jury an ongoing fraud.  Shouldn’t an ongoing fraud that occurs, on a grand billion dollar scale, despite signing a contract promising to perform truthfully be treated as more serious, far more so, than a one-shot fraudulent inducement to enter a contract?  To be sure, the ongoing representations were linked to the earlier master contract.  But does that make them somehow “inside the contract”?  The Contract itself rules that interpretation out.  As the lower court explained in one of its rulings, each sale of a loan represented that it complied with “all the applicable requirements in [the] Guides and this [Master] Contract” and that the seller knew “of nothing … that [could] reasonably be expected to… cause the mortgage to become delinquent; or adversely affect the mortgage’s value or marketability.”  Yet those representations that attempted to place this inside the contract were false.  And these were new sales and new representations.

The Second Circuit was right to say “deception” is the key difference between mere breach of a contract and fraud.  And the Government proved deception here in spades—to a jury, whose verdict should not be overturned lightly.  For that reason, as the lower court explained, New York state courts have held in very similar factual circumstances that fraud claims “premised on false representations about quality of mortgages” are not “impermissibly ‘duplicative’ of common-law breach of contract claims.”  Nor is it correct, as the Second Circuit had to acknowledge that the common law of fraud is necessarily or even usually incorporated into the meaning of the federal mail fraud statutes.  Nor, as the Government pointed out, is the common law of fraud from the time when the mail fraud statute was adopted supportive; cases, including New York cases at the time often stated that proof of a contract breach may in fact be highly useful evidence of fraud, far from being an obstacle to proving fraud.  The Second Circuit then dodged in a footnote the caselaw holding that “collateral misrepresentations” can render post-contract conduct fraud.  Yet those are squarely on point: finding fraud in the use of deception post-contract to give the false impression that a contract is being performed.  Legally, the Second Circuit went out on a limb, picking and choosing from fraud caselaw to reach a result.

Still more troubling, and giving an even stronger impression that the panel was on a $1.27 billion mission, the Second Circuit went out on a limb factually.  Even if the common law of fraud requires contemporaneous intent to deceive, what is the relevant time period?  Why should it be the time when the initial master contract was signed and not when sales with accompanying representations occurred?  That is the time period when the fraud occurred.  The panel said “fraudulent intent must be found at the time of the allegedly fraudulent conduct” but it played a shell game, since the allegedly fraudulent conduct occurred with each loan sale.

It is no surprise that the Second Circuit, with its prominence and importance in developing federal law concerning financial claims, would re-interpret the law in a significant ruling on the definition of fraud.  It is surprising that the Second Circuit would appear to be trying to re-interpret the facts.  To be sure, the ruling does not affect the settlements in other mortgage fraud related cases.  The panel did not interpret the meaning of FIRREA itself, either, as some had anticipated.  In future prosecutions, perhaps the ruling will not pose an obstacle, since contracts can be drafted with this in mind; who would want to sign a “master” contract that might later expose a person to outright deceit?  And as Greg Klass has pointed out, a False Claims theory might fare better as well.  Nevertheless, if and when other judges consider the question, perhaps this legally unwarranted and factually confused ruling will not long stand.

Brandon Garrett is the Justice Thurgood Marshall Distinguished Professor of Law, at the University of Virginia School of Law.  His recent book, “Too Big to Jail: How Prosecutors Compromise with Corporations,” was published in 2014.  He is currently working on a new book, “The Triumph of Mercy: How the Demise of the Death Penalty can Revive Criminal Justice,” for Harvard University Press.