Rolling Stone’s Matt Taibbi described Goldman Sachs as “a great vampire squid wrapped around the face of humanity,” a phrase that, while defamatory of a uniquely adapted cephalopod minding its own business 3,000 feet under the sea, rang true. Yesterday, the intermediate appellate court for New York state agreed: Goldman Sachs is so obviously dishonest that you cannot sue them for fraud unless you get them to specifically agree that they aren’t lying to you.


First, the facts. In essence, Goldman Sachs brought in a hedge fund (Paulson & Co.) to put together a group of horrible investments (called “Abacus”) that they expected to fail — and even bet against — and then set about finding rubes to invest in it, thereby helping Goldman and Paulson make a tidy profit off the investor’s losses. One other banker who passed on the deal described it as “like a bettor asking a football owner to bench a star quarterback to improve the odds of his wager against the team.” (He’s quoted in the dissent.)


ACA Financial Guaranty Corporation was one of the rubes Goldman Sachs found. As Reuters reported, ACA’s lawsuit against Goldman Sachs “alleged that Goldman misrepresented the role of the hedge fund Paulson & Co, which supposedly selected underlying mortgage-backed securities that doomed the [collateralized debt obligation] to fail, thereby assuring Paulson of big profits on its undisclosed Abacus short.” The scam was so blatant the Securities and Exchange Commission brought its own case against Goldman Sachs, which settled for $550 million.


Sounds simple enough; as James Surowiecki wrote about the scandal three years ago, echoing the thoughts of many financial journalists, there was ample reason to believe that ACA was both a “dupe” hoodwinked by Goldman and a “dope” that failed to perform adequate due diligence on a complicated investment. Being a “dope” is a problem, but one would assume that a duped dope would be allowed to present evidence to a jury arguing that the fraud was a bigger problem than the lack of due diligence.


Except that the New York courts won’t let ACA get to a jury. As the New York Appellate Division just affirmed in ACA Financial Guaranty Corp. v Goldman, Sachs & Co., “a fraud claim is barred where a sophisticated and well-counseled entity fails to include an appropriate prophylactic provision in the agreement governing the transaction from which the legal dispute arises to ensure against the possibility of misrepresentation.” It didn’t matter that Goldman Sachs had repeatedly misrepresented to ACA that Paulson was actually investing in the deal (rather than engineering the deal to fail), because, the majority of the Court held:


(1) such misrepresentations were specifically contradicted by the offering circular’s disclosure that no such equity position was being taken [and]

(2) plaintiff’s alleged reliance on such misrepresentations would have been contrary to its acknowledgment in the offering circular that, in entering into the transaction, it was “not relying upon any representations (whether written or oral) of Goldman Sachs other than in the final offering circular


Opinion, p. *3 (some edits made for formatting and clarity). Putting aside the affirmative misrepresentations for a moment, there’s nothing unusual about the above analysis. That sort of language is called an “integration clause” or “merger clause,” and it’s taught in 1L Contracts alongside the “parol evidence rule.” In short, if a contract says something like, “you agree that nothing we wrote or said outside of this contract itself mattered to you,” then courts will hold sophisticated parties to that agreement. There’s a straightforward explanation of these clauses in UAW-GM Human Resources Center v. KSL Recreation Corp., 579 N.W.2d 411, 414 (Mich. 1998).


But that holding wouldn’t have been enough to dismiss the lawsuit, and so the court also held that ACA couldn’t sue Goldman for lying about the hedge fund’s role because “[ACA] was in direct contact with the hedge fund, [but] failed to ask the hedge fund what position it intended to take in this investment.” In other words: it is the law of New York that Goldman Sachs is so dishonest that anyone who trusts their word deserves what they get.


This part of the ruling was, in a word, nuts. It has long been the law that an integration or merger clause will not protect a fraud when the fraud is sufficient to set aside the whole contract. Indeed, one of the leading cases on this issue comes from New York: Sabo v. Delman, 3 N.Y.2d 155, 164 N.Y.S.2d 714, 143 N.E.2d 906 (1957): “Indeed, if it were otherwise, a defendant would have it in his power to perpetrate a fraud with immunity, depriving the victim of all redress, if he simply has the foresight to include a merger clause in the agreement. Such, of course, is not the law.” See also 3 Williston on Contracts, §§ 811-811A; 3 Corbin on Contracts, § 578; 2 Restatement, Contracts, § 573. The dissent notes that was still the law of New York state, too, per DDJ Management, LLC v Rhone Group, L.L.C., 15 NY3d 147, 153, 156 (2010)(“Where, however, a plaintiff has taken reasonable steps to protect itself against deception, it should not be denied recovery merely because hindsight suggests that it might have been possible to detect the fraud when it occurred.”).


But, just like there is financial risk, there is legal risk, too. Most lawyers wouldn’t have said that ACA could have certainly sued Goldman for fraud, or that demanding the integration clause be removed would have saved them, only that such a measure would have put ACA in a better position. You never know when a slim majority of a Court is going to hold that the joke’s on you.


I suppose the lesson is quite simple: don’t do business with a vampire squid.


[Update: Matt Levine at DealBreaker has a thorough post on the case, pointing out that, while ACA was indeed duped and lost money because of it, “ACA is perhaps the least sympathetic of all the financial crisis victims” given that they were “as much co-conspirators as they were victims.” I wouldn’t go that far — they lost money due to Goldman’s misrepresentations, making them more a victim than a co-conspirator in my book — but their role in the transaction was certainly dubious and worthy of scrutiny.


Nonetheless, there are still two big problems with the court’s opinion. First, ACA’s role and their ability to discover the fraud are exactly the sort of messy factual issues that juries resolve at trials, not the sort of clean-cut indisputable legal issues judges resolve on briefs. Second, the court’s opinion isn’t narrowly limited to the precise facts of this transaction; indeed, their argument could be used to punt all of the lawsuits brought by every investor, including those who were clearly victims, not co-conspirators. Taken at face value, it’s almost a repudiation of the Sabo rule, at least for sophisticated businesses, and while that might be justifiable for ACA in this case, it certainly isn’t justifiable for all business cases.]