On Friday, the Second Circuit (Livingston, Carney, Sullivan) reinstated the conviction of a former Nomura Securities International, Inc. (“Nomura”) bond trader, Michael Gramins, in United States v. Gramins, No. 18-2007-cr, finding the lower court’s decision to grant Gramins a new trial pursuant to Rule 33 was based on an “overbroad” reading of the Circuit’s 2018 ruling in United States v. Litvak (“Litvak II”) that “cannot be located within the range of permissible decisions.” Litvak II was a major decision and the Gramins decision implicitly limits the reach of Litvak II to cases in which the government or its witnesses expressly described a broker as an agent when he or she is not, in fact, an agent. Future decisions will no doubt revisit these cases, looking for the dividing line between the two.

Background

Gramins worked as a broker-dealer of residential mortgage-backed securities (“RMBS”) at Nomura. The RMBS market is distinctive: RMBS are bought and sold by “large, sophisticated financial institutions” and are traded in a market in which the prices are relatively unknown. Prospective buyers and sellers must contact registered broker-dealers who find interested counterparties or who transact directly with the interested party from the broker-dealer’s own accounts. Accordingly, prospective buyers and sellers are largely dependent on information provided by the broker-dealer, who receives his or her compensation by taking “spread” or “margin” on the transaction, or by selling the bond from the broker-dealer’s inventory (presumably at a profit). Although the decision refers to this as a “commission,” this form of compensation is different from a transaction in which the broker is compensated by a fixed percentage commission on the sale price of the security. Once a buyer and seller are paired by a broker-dealer, the broker-dealer facilitates the trade by buying the RMBS from the seller at an agreed-upon price, and then selling it to the seller for a slightly higher price. The difference between those prices represents the broker-dealer’s compensation.

Between 2009 and 2013, on at least five occasions, Gramins was alleged to have misrepresented the prices at which buyers were willing to buy or sellers were willing to sell, in order to substantially increase his compensation for the trade. By making a buyer believe that the bond was purchased for a higher price than it was, the broker-dealer can claim a larger “spread” as his compensation. On September 3, 2015, Gramins and two alleged co-conspirators were indicted on charges of securities fraud, wire fraud, and conspiracy to commit securities and wire fraud in violation of 15 U.S.C. §§ 78j(b) and 78ff, 17 C.F.R. 240.10b-5, 18 U.S.C. § 1343, and 18 U.S.C. § 371.

On June 15, 2017, a jury convicted Gramins of conspiracy to commit securities and wire fraud. Gramins’ alleged co-conspirators were acquitted of all wrongdoing and Gramins was acquitted of the securities and wire fraud charges. During the trial, the government presented evidence of specific misrepresentations by Gramins, and elicited testimony from a witness, Joel Wollman—one of the clients whom Gramins deceived—that he “maintained heightened expectations of truthfulness from his broker-dealer” in such a RMBS trade.

After the trial, Gramins moved for a judgment of acquittal or, in the alternative, for a new trial on the sole count of conspiracy. On May 3, 2018, before the district court ruled on Gramins’ post-trial motions, the Second Circuit issued its opinion in Litvak II, in which it reversed a securities broker’s conviction for securities fraud because it was “tainted by testimony” from a counterparty as to that counterparty’s erroneous belief that Litvak acted as his agent in executing trades. The Litvak II decision held that the counterparty’s “idiosyncratic and unreasonable viewpoint” as to whether Litvak owed fiduciary duties to his clients was not indicative of a that of a reasonable investor, and was likely to confuse or mislead a jury.

The district court denied Gramins’ motion for acquittal but, in light of the decision in Litvak II, granted Gramins’ motion for a new trial, in part because Wollman’s testimony strongly implied that the role of a broker-dealer is one of an agent, when in fact it is not. In reality, the broker-dealer is not acting as an agent and he or she owes no fiduciary duties to the client. In addition, the client is a sophisticated party, like a hedge fund or an asset manager. These facts were relevant to both Litvak II and to this decision.

The Appellate Decision

The Second Circuit, reviewing the district court’s decision under an abuse of discretion standard, affirmed the denial of acquittal, but reversed and reinstated Gramins’ conviction. In an opinion that emphasized that this was a close case that raised issues of first impression, and that the district court had committed no legal error, the panel concluded that the district court’s decision resulted from an “overbroad reading of Litvak II.”

In the panel’s view, the testimony at Gramins’ trial did not rise to the level of that at Litvak’s trial. At Gramins’ trial, Wollman described Gramins as a “broker” and said that he was “brokering” trades but never referred to Gramins as an “agent”; in fact, he did not utter the word “agent” at all. Nor did Wollman express an “erroneous or idiosyncratic” view of his relationship with Gramins. Wollman did not express that Gramins owed him, or any customers, any fiduciary duties. While Wollman’s testimony supported the Government’s theory that Gramins’ misrepresentations were material to his counterparties, so did the testimony of all the witnesses with whom Gramins transacted. That is, the Government elicited proper testimony as to whether Gramins’ counterparties reasonably relied on his misrepresentations without eliciting testimony that would have impermissibly suggested a fiduciary relationship between Gramins and those counterparties. Because “nothing that occurred at Gramins’ trial conferred an undue advantage on the government in the battle over the issue of materiality,” the Second Circuit held that Wollman’s testimony was not unduly prejudicial or likely to mislead the jury, and that the district court’s decision to the contrary was based on an impermissible extension of Litvak II.

The Second Circuit then determined that, even if the admission of Wollman’s testimony was in error, any such error was harmless because “the prosecutors actively took steps to disabuse the jury of any mistaken notion that Gramins acted in a fiduciary capacity.” The Government expressly told the jury that “‘Nobody’s claiming here that anybody is a fiduciary.’” Indeed, “every legal authority present in the courtroom—prosecutors, defense counsel, and judge—expressly and repeatedly informed the jury that no agency relationship existed between Wollman and Gramins.” These facts set the case apart from Litvak II. Moreover, the Court determined that Wollman’s testimony was “cumulative” of other properly admitted evidence, and was not the lynchpin of his conviction: indeed, the record provided an independent basis for supporting Gramins’ conviction aside from Wollman’s testimony. It could be argued that this aspect of the Court’s holding rendered its discussion of Litvak II dicta.

Commentary

In this decision, the Court went to great lengths to distinguish the facts of Litvak II, in which a witness expressly suggested that the defendant owed fiduciary duties to his customers—which he clearly does not—and the facts here, in which a witness suggested that he expects a broker’s statements to him to be truthful and used more ambiguous phrases like “brokering.” It is not hard to imagine that the testimony offered by Wollman might have led some jurors to believe that an agency relationship did, in fact, exist between Wollman and Gramins. But the omission of an express statement to that effect, coupled with the Government’s repeated insistence to the jury that it was not alleging an agency relationship or fiduciary duty, was enough for the panel to find the cases distinguishable.

The Second Circuit recognized that this case involved a “novel form of prosecution,” perhaps signaling that it would be preferable for the government to address novel issues in the marketplace through regulation rather than through prosecution (even where the evidence was sufficient and complied with the governing law). The panel’s ruling aimed to resolve “issues of first impression” based on “two prior precedents on the subject [that] are at times obscure.” Yet it is possible that this decision may unintentionally create more uncertainty, as the panel reached a decision opposite to Litvak II on a similar set of facts. In a future case, the word “agency” may not be used at trial, but there be more testimony than here that was evocative of an agency relationship. Will this be allowed or will the next panel find that it is too similar to Litvak II? For lawyers advising clients, the guidance remains the same as before Gramins: any false statement made to a client, whatever the circumstances, is not only bad business ethics but also possibly subjects the speaker and his firm to investigation or worse. For lawyers representing someone at trial after Gramins, continue to object to any “agency-like” testimony or arguments in order to make a record for a future appeal.