Summary
admitting testimony from a participant in a recorded conversation interpreting the recording when the testimony "would be helpful to the jury within the meaning of Rule 701"
Summary of this case from Towner v. HoganOpinion
16 Cr. 483 (JSR).
04-18-2017
UNITED STATES of America v. Stefan LUMIERE, Defendant.
Ian Patrick McGinley, Joshua Arditi Naftalis, A. Damian Williams, United States Attorney's Office, New York, NY, for United States of America. Eric M. Creizman, Amanda Shoffel, Melissa Madrigal, Creizman PLLC, Jonathan Hale Friedman, Jonathan Nassau Halpern, Foley & Lardner LLP, New York, NY, for Defendant.
Ian Patrick McGinley, Joshua Arditi Naftalis, A. Damian Williams, United States Attorney's Office, New York, NY, for United States of America.
Eric M. Creizman, Amanda Shoffel, Melissa Madrigal, Creizman PLLC, Jonathan Hale Friedman, Jonathan Nassau Halpern, Foley & Lardner LLP, New York, NY, for Defendant.
OPINION AND ORDER
JED S. RAKOFF, U.S.D.J.
On January 19, 2017, following a six-day trial, a jury found defendant Stefan Lumiere guilty of wire fraud, securities fraud, and conspiracy to commit those offenses. See Verdict, ECF No. 65. The basic charge was that the defendant, a former employee of Visium Capital Management ("Visium"), had participated in a two-year scheme to dupe investors into believing that the Visium Credit Opportunities Fund (the "Credit Fund") was performing far better than it actually was. Lumiere now moves for a new trial pursuant to Rule 33(a) of the Federal Rules of Criminal Procedure. For the reasons stated below, Lumiere's motion is hereby denied.The evidence at trial (taken, along with all reasonable inferences, most favorably to the Government) showed as follows:
Visium was a hedge fund manager that operated several different hedge funds and that, at its peak, had $8 billion of investor money under management. See Tr. 40, 42. Among the hedge funds it managed was the Credit Fund, which invested principally in bonds, bank debt, and other credit instruments, and which managed up to $500 million. See Tr. 42–43, 630–31. The Credit Fund performed well through 2010, but in 2011, its performance began to suffer. See Tr. 248–49; Government Exhibit ("GX") 117.
In 2011, Christopher Plaford was the senior portfolio manager for the Credit Fund, and was responsible for directing the fund's investment process. See Tr. 45, 630–31, 651; see also GX 106. Lumiere, who reported to Plaford, was a senior analyst and also a portfolio manager responsible for around $100 million of the Credit Fund's investments. See Tr. 146–47, 244–46, 664–65. Jason Thorell was the trader for the Credit Fund, which meant that he executed trades as directed by Plaford, Lumiere, and others. See Tr. 241–42, 652–53. When the Credit Fund's investments began to suffer in 2011, Plaford and Lumiere, with assistance from Thorell, hatched a scheme to defraud investors in the Credit Fund into believing it was still performing well. Plaford pleaded guilty to participating in the fraud and testified at trial under a cooperation agreement; Thorell, who was not charged, testified under a grant of immunity. See Tr. 222–23, 640–41.
The key metric by which Visium reported the Credit Fund's performance to investors was its net asset value ("NAV"), which represented the aggregate fair market value of all investments and cash held by the hedge fund. See Tr. 60, 662–63. Investors also monitored the proportion of the Credit Fund's investments that were classified as illiquid "Level 3" assets under Accounting Standard Codification 820, and strongly preferred that the Credit Fund hold no Level 3 assets at all. See Tr. 663, 703–04; see also Tr. 63–66. Plaford and Lumiere also had personal stakes in the level of the Credit Fund's NAV, because the higher the NAV, the higher their compensation would be, whereas, by contrast, poor performance could result in their termination. See Tr. 248, 664. Visium reported the Credit Fund's NAV and liquidity to investors on a monthly basis, see Tr. 661; thus, to achieve their goal of duping investors into believing the struggling Credit Fund was performing well and held no illiquid investments, Plaford and Lumiere crafted a plan to systematically manipulate both the NAV and the liquidity calculations, month-by-month, from 2011 to 2013.
Plaford and Lumiere did not directly control the level at which the NAV was reported, since it was calculated by the back-office Operations group. However, to make the calculation, Operations relied to a significant degree on securities prices reported by the investment team—i.e., by Plaford and Lumiere. At the end of each month, the investment team would send Operations a list of prices for each security held by the Credit Fund. See Tr. 560–61. The next business day, Operations would obtain market prices for each security from a third-party pricing source such as Markit, and then send to the investment team a spreadsheet that compared Visium's in-house prices with the third party prices. See Tr. 561–63. In order to cause Operations to use an in-house price instead of the third party's price in calculating the NAV, the investment team was required to provide independent verification that its in-house price was fair and accurate. This was known as "overriding" the third-party price. See Tr. 563–65, 667–68. A quote from an outside broker indicating that it would transact in a security at the investment team's in-house price was, however, sufficient to override a third-party price. See Tr. 564–65. Thus, beginning in mid–2011, Plaford and Lumiere began suborning outside brokers to provide, on demand, assurances regarding whatever back-up securities price quotes Lumiere asked for. Lumiere personally selected and corrupted the two main brokers used for this purpose: Scott Vandersnow of Princeridge and Jonathan Brook of Janney Montgomery Scott. See Tr. 285, 293–94, 305, 483, 680–81. In addition to being personal friends of Lumiere, Vandersnow and Brook worked at firms that did not usually trade the type of securities the Credit Fund held, and thus could not realistically be counted on for objective price quotes. See Tr. 314–15, 351; GX 1224. Later, at Lumiere's suggestion, Thorell located a third corrupt broker, Matthew O'Callaghan of Odeon Capital Management, another "bucket shop" broker whose services they occasionally used as well. See Tr. 306–09. The brokers allowed Lumiere and his co-conspirators to dictate the prices the brokers attested they were willing to pay. See, e.g., Tr. 489–91; GX 1209, 1227. In return for the brokers providing the sham prices, Lumiere steered business their way. See, e.g., Tr. 285–86, 294. All of this, needless to say, was concealed from everyone but the coconspirators. See Tr. 699.
To implement this scheme, Plaford and Lumiere, toward the end of each month, would devise inflated prices for poorly performing securities. See Tr. 668, 674–75; see also Tr. 684–91; GX 1208, 1212, 1216, 1228–30. Plaford would send Thorell a spreadsheet of the prices, and Thorell, at Plaford's direction, would then create a new email, attach that spreadsheet, and send it to Operations to begin the month-end NAV calculation process. See Tr. 252–57, 677–78. Meanwhile, Lumiere would contact Vandersnow and/or Brook (or, later on, O'Callaghan) and dictate the chosen prices; Vandersnow and Brook (sometimes while they were still on the phone with Lumiere) would promptly send Lumiere "verifications" of the demanded prices, without in fact independently verifying them. See, e.g., Tr. 286, 354–58, 361–62, 489–94, 679–80; GX 1209, 1227. To give a veneer of legitimacy, these exchanges were sometimes glossied-up. For example, Plaford might tell Lumiere that he thought a certain price was accurate, but that he was open to other views. See, e.g., GX 1230. Lumiere, in turn, might relay the prices to the brokers under the guise of "educating" them about Visium's perspective. See, e.g., GX 1227.
All of this, was simply window dressing intended to help conceal the fraud. Over the course of two years and hundreds of securities, none of the brokers ever disagreed with Lumiere's prices, nor was there any attempt to supply the brokers with information about the securities, with most of which they were unfamiliar. See Tr. 500–02, 675. Once the brokers dutifully supplied the dictated quotes, their emails would be sent to Operations as the necessary back-up documentation for Plaford's and Lumiere's hand-picked securities prices. See Tr. 276, 289. Thus did Lumiere and his co-conspirators grossly inflate the NAV. In addition, using broker quotes as back-up had the further advantage of preventing these securities, which were often thinly traded, from being classified as illiquid Level 3 assets. See Tr. 693.
The details of the scheme varied slightly at times. For example, on one occasion, rather than rely on a sham broker quote as back-up, Lumiere himself purchased a security held by the Credit Fund at a significantly above-market price, and then used that "real-world" transaction as the back-up for the month-end pricing—a practice known as "painting the tape." See Tr. 638–39, 708–15. Sometimes Lumiere conveyed the bogus prices to the brokers by messengering them flash drives that contained only a spreadsheet of securities prices. See Tr. 364–66, 506–07, 682. And for a few months, acting at Lumiere's direction, it was Thorell, rather than Lumiere, who obtained the bogus broker quotes. See Tr. 358–62, GX 1234–A.
Eventually, however, suspicion arose, and in April 2013, Lumiere was fired for poor performance. See Tr. 658, 679. Shortly thereafter, the remaining investment team was removed from the pricing process, and the scheme collapsed. See, e.g., Tr. 758–60.
The evidence of the foregoing scheme introduced at trial was little short of overwhelming. It included a mass of documents (electronic and otherwise); audio recordings of the conspirators in action; and credible testimony from three of the conspirators (Plaford, Thorell, and Vandersnow). The only real issue for trial was whether, as Lumiere maintained, he had a good-faith belief in the securities prices he obtained to boost the NAV.
Under Rule 33(a), a district court "may vacate any judgment and grant a new trial if the interest of justice so requires." Fed. R. Crim. P. 33(a). "The test is whether it would be a manifest injustice to let the guilty verdict stand." United States v. Guang, 511 F.3d 110, 119 (2d Cir. 2007) (internal quotation marks omitted). In other words, "[t]he grant of a Rule 33 motion requires a real concern that an innocent person may have been convicted." United States v. Parkes, 497 F.3d 220, 232 (2d Cir. 2007) (internal quotation marks omitted). Lumiere contends that a combination of supposed evidentiary errors, putative prosecutorial misconduct, and allegedly flawed jury instructions—nearly all of which he did not object to at trial—resulted in an unfair trial. In actuality, however, there were no such errors, let alone the kind of egregious errors that would justify overturning the jury's verdict.
Beginning with the evidentiary issues, Lumiere finds fault in a number of rulings made in connection with the testimony of Jason Thorell. Because Lumiere uniformly failed to object on the grounds he now advances, his arguments are reviewed under the "plain error" standard. See Fed. R. Crim. P. 52(b) ("A plain error that affects substantial rights may be considered even though it was not brought to the court's attention."). But, in fact, none of the rulings was erroneous, much less an "obvious" error that not only "affected the outcome of the district court proceedings," but also "seriously affect[ed] the fairness, integrity, or public reputation of judicial proceedings." See United States v. Olano, 507 U.S. 725, 734, 736, 113 S.Ct. 1770, 123 L.Ed.2d 508 (1993).
Lumiere first argues that a portion of Thorell's testimony interpreting a recording was improperly admitted as lay opinion testimony. Under Rule 701, lay witnesses may testify to an opinion only if it is "rationally based on the witness's perception" and is "helpful to clearly understanding the witness's testimony or to determining a fact in issue." See Fed. R. Evid. 701(a), (b).
At trial, the Government played portions of a recording Thorell made of a conversation he had with Lumiere in January 2014. In one excerpt played for the jury, the following exchange took place:
STEFAN LUMIERE: Yeah, I think, I think, the way we explain it from my perspective to investors is I don't want to blow everything they've done out. For example, Chris has been mismarking shit since the beginning.
JASON THORELL: Um-hum.
STEFAN LUMIERE: Right? Originally in 2011, he was marking them too low so he could save ear—like retaining earnings for a later date—
JASON THORELL: Yeah.
STEFAN LUMIERE: —and smoothing them out.
JASON THORELL: Yeah.
STEFAN LUMIERE: So he's been doing that since the beginning. He's been mismarking shit egregiously—outside context of what was right from the beginning. But the things I'd like to focus on are stuff like ATI—
JASON THORELL: Um-hum.
STEFAN LUMIERE: —and Chinamed—
JASON THORELL: (indiscernible) Look, look. I, I—
STEFAN LUMIERE: —which were just egregious. And I'd like to use that as a reason for my deciding to depart them.
GX 1223–A. On direct examination, Thorell testified that "Chris" referred to Plaford, their former boss. See Tr. 237. The following exchange then occurred:
Q: When you spoke to Mr. Lumiere and he used the word "Chris did this," what did you understand him to mean?
A: I understood him to mean that, in the context of this transcript, that he's speaking for Chris, but also himself.
See Tr. 238.
Although he raised no objection at the time, Lumiere now claims that this testimony was inadmissible because it was not rationally based on Thorell's personal knowledge and because "Chris" was not a code that the jury needed cracked.
But, in fact, this testimony was properly admitted. As a participant in the conversation, Thorell plainly had first-hand knowledge of the conversation itself. See United States v. Urlacher 979 F.2d 935, 939 (2d Cir. 1992) ("[T]he first prong requires that witness meet the familiar requirement of first-hand knowledge or observation." (internal quotation marks omitted)). Just as importantly, Thorell had first-hand knowledge of the conspiracy being discussed. Prior to explaining what he thought Lumiere meant by "Chris," Thorell testified that he had participated in the mismarking scheme with Plaford and Lumiere, that he had been sufficiently distressed by the scheme to report it to Visium management, and that, when his concerns were not allayed, he reported the fraud to the SEC. See Tr. 223–25; see also Tr. 373–85, 388–89. He later explained the mechanics of the fraud in detail. See, e.g., Tr. 274–76, 364–66. In other words, Thorell not only had firsthand knowledge of what was actually said, but firsthand knowledge of the underlying scheme sufficient to interpret the statements of his former partner-in-crime. See United States v. Yannotti, 541 F.3d 112, 125 (2d Cir. 2008) ("DiDonato's testimony was rationally based on his own perception because it derived from his direct participation in the loansharking activities of the charged enterprise ....").
Thorell's testimony was also helpful to the jury within the meaning of Rule 701. The conversation Thorell was called upon to interpret took place in January 2014, after the conclusion of the years-long securities mismarking scheme. The recording thus contained the kind of shorthand and oblique references that only a person familiar with the background would readily grasp. Moreover, Thorell, as the most junior member of the conspiracy, had taken direction from both Plaford (who had Thorell send the override spreadsheet to Operations in Thorell's own name to mask Plaford's involvement) and Lumiere (who, inter alia, supplied the necessary back-up documentation for the overrides in the form of sham broker quotes). See Tr. 252– 57, 276, 289. It is against that backdrop that Thorell testified that he understood Lumiere's references to "Chris" to mean both Plaford and Lumiere himself. Such testimony was plainly "helpful to clearly understanding the witness's testimony or to determining a fact in issue." See Fed. R. Evid. 701 ; Yannotti, 541 F.3d at 126 (upholding admission of "interpretative testimony because the language on the tape was punctuated with ambiguous references to events that are clear only to the conversants" (alterations, ellipsis, and internal quotation marks omitted)). This testimony was therefore properly admitted under Rule 701.
Lumiere's cases are all inapposite. For example, in United States v. Grinage, the Second Circuit found that it was error to admit lay opinion testimony from a case agent who came by his knowledge second hand, rather than by participating in the crime and in the conversation he interpreted, as Thorell did. See 390 F.3d 746, 750–51 (2d Cir. 2004). And in United States v. Garcia, the Second Circuit found that the district court had improperly admitted lay opinion testimony where the witness testified that a superficially legitimate conversation was in fact conducted in secret code without supplying any basis to believe that the other participant knew they were speaking in code. See 291 F.3d 127, 140–42 (2d Cir. 2002). But Thorell's testimony is not about secret codes; it is simply a case where a conspirator explained "ambiguous references to events that are clear only to the conversants," as has been repeatedly endorsed by the Second Circuit. See, e.g., Urlacher, 979 F.2d at 939 (alterations omitted); United States v. Aiello, 864 F.2d 257, 265 (2d Cir. 1988).
Lumiere also argues that Thorell's testimony that he took "Chris" to mean both Plaford and Lumiere was inadmissible because its probative value was substantially outweighed by the danger of unfair prejudice. See Fed. R. Evid. 403. This kind of objection is pre-eminently the kind that should be raised at trial so that any supposed prejudice can be evaluated at the time; but here no objection was made. But this was for good reason, for the objection now raised is meritless. To be sure, Thorell's testimony was highly damaging to Lumiere's case; but it was hardly unfair for Thorell, an admitted participant in the years-long fraudulent scheme, to explain that he understood Lumiere's comments as an admission of Lumiere's own guilt. This is not the kind of extraneous prejudice that Rule 403 is directed at.
Lumiere's other evidentiary attack on Thorell's testimony is that his out-of-court statements admitted to show his participation in the conspiracy should have been excluded as hearsay because the Government failed to show that Thorell and Lumiere were, in fact, co-conspirators. See Fed. R. Evid. 801(d)(2)(E) (excluding from the definition of hearsay a "statement ... offered against an opposing party" that "was made by the party's coconspirator during and in furtherance of the conspiracy"). Evidence may be admitted under Rule 801(d)(2)(E) only if a court finds, by a preponderance of the evidence, that the defendant and the declarant joined a conspiracy, and the challenged out-of-court statements may themselves be considered in making this determination. See Bourjaily v. United States, 483 U.S. 171, 175–76, 178–79, 107 S.Ct. 2775, 97 L.Ed.2d 144 (1987). Lumiere claims that because Thorell resisted the notion that he had been involved in a "criminal enterprise" until he was so informed by counsel after the fact, see Tr. 425, he never joined the conspiracy.
This argument is all but frivolous. Here, there was abundant of evidence of a conspiracy involving Thorell and Lumiere, which easily satisfies the preponderance standard. Thorell admitted to having participated in the conspiracy with Lumiere and explained the conspiracy in great detail. See Tr. 223–24. He testified, for example, that Lumiere provided the back-up documentation for Thorell to send to Operations each month-end, a key step in manipulating the NAV. See Tr. 274–76. Thorell also testified that Lumiere directed him to find a third corrupt broker to supply sham price quotes, prompting Thorell to enlist Matthew O'Callaghan of Odeon Capital Management. See Tr. 308. And Thorell further testified that on several occasions, Thorell, acting at Lumiere's direction, solicited the sham quotes from these brokers himself. See Tr. 358–60. Plaford corroborated much of Thorell's testimony. See, e.g., Tr. 632–34, 677–78. Weighed against all this evidence, the cherry-picked testimony that Thorell "wouldn't use the phrase ‘criminal enterprise’ " to describe how he initially viewed his involvement in the mismarking scheme does not move the scales. See Tr. 424–25.
Turning to the next group of hypothesized evidentiary errors, the defendant argues that the Court erred in excluding the remainder (or at least selected excerpts) of the Thorell–Lumiere recording in which the defendant stated that "Chris has been mismarking shit since the beginning," see GX 1223–A, which Thorell interpreted to mean that Plaford and Lumiere alike had been engaged in the mismarking scheme. See Tr. 238. The Government introduced a few minutes of the recording, which amounted to a few pages in transcript form; the recording in its entirety has a runtime of multiple hours and the full transcript spans well over 100 pages. Lumiere argues that the balance of the recording, which supposedly contains evidence of Lumiere's good faith, ought to have been admitted as well, both under the "rule of completeness" and because the Government "opened the door."
Under the completeness doctrine, "[i]f a party introduces all or part of a writing or recorded statement, an adverse party may require the introduction, at that time, of any other part—or any other writing or recorded statement—that in fairness ought to be considered at the same time." Fed. R. Evid. 106. This rule requires admission of so much of a statement or recording as necessary "to explain the admitted portion, to place it in context, or to avoid misleading the trier of fact ... or to ensure a fair and impartial understanding of the admitted portion." United States v. Marin, 669 F.2d 73, 84 (2d Cir. 1982) (internal quotation marks omitted). "The completeness doctrine does not, however, require introduction of portions of a statement that are neither explanatory of nor relevant to the admitted passages." Id. In particular, courts have discretion to exclude a criminal defendant's post-hoc justifications for his criminal actions. See United States v. Rutigliano, 614 Fed.Appx. 542, 545 (2d Cir. 2015) (summary order) ("The exclusion of the disputed statements as post-hoc explanations for [defendant's] conduct was ... a permissible exercise of discretion.").
Lumiere tries two tacks under Rule 106. First, he argues that the Court erred in excluding portions of the Thorell–Lumiere recording that his counsel actually offered at trial, which, he contends, provide necessary context for Lumiere's remarks because they demonstrate that Lumiere was unaware that Plaford had been mismarking securities until much later in time. In particular, at trial, Lumiere sought to introduce portions of the recording in which he suggested that Plaford initially had semi-cogent explanations for the unusual levels at which he priced at least some of the securities held by Visium, which Lumiere only later came to doubt. See Tr. 266–74; Proposed Ex. 1224–T and 1225–T, at 5–6 ("And then when it came to [ATI], everything kind of made sense. A lot of people didn't know where the levels were.... The fact that we were in a control position kind of like made more sense. And we kind of backed into the price based on estimates of EBITDA and value for the company because we knew who we, who had bidding for sale so we knew more than RBC did about it. So that kind of was okay."). Lumiere claims on this recording that he realized Plaford's explanation was "bogus" when Plaford, citing Visium's written pricing policies, refused Lumiere's demand to reduce their ATI valuation after the company dramatically reduced its estimated 2013 EBITDA. See id. at 6–8. Without this "context," Lumiere argues, the jury was misled.
This argument has no merit. The Government introduced limited portions of this recording to show what Plaford and Lumiere did: beginning in 2011, they mismarked securities held by the Credit Fund by having budget brokers parrot back made-up prices that Lumiere gave them. The admitted portions of the recording thus did no more than introduce simple historical facts about the mechanics and scope of the mismarking scheme. By contrast, the portions of the recording that Lumiere argues were improperly excluded are post-hoc attempts by Lumiere to minimize his own culpability by explaining that at least some facets of the scheme seemed superficially legitimate. In other words, the admitted portions were a simple admission to participating in the mismarking scheme; the excluded portions were Lumiere's improbable attempts to claim, long after the fact, that he had done so innocently.
Take, for example, Lumiere's statement that Plaford's explanation of the ATI marks "kind of made sense." Set aside the fact that a reasonable jury might well take that as an admission that Plaford's explanation also "kind of" made no sense, and, by extension, that none of Plaford's other explanations—assuming he gave any—made any sense at all. Even if a credulous jury might agree that that statement shows Lumiere was acting in good faith, it neither explains nor has relevance to the bare historical facts the Government introduced: that together, the conspirators mismarked securities. Instead, that statement, like all the others Lumiere sought to introduce, "offer[ed] context only insofar as they represent[ed] [Lumiere's] self-serving attempts to shoehorn after-the-fact justifications for his actions into his descriptions of his actions. This is not the type of material envisioned by Rule 106." See United States v. Lesniewski, No. 11–cr–1091 (VM), 2013 WL 3776235, at *5 (S.D.N.Y. July 12, 2013), aff'd sub nom. Rutigliano, 614 Fed.Appx. at 545. The additional portions of the Thorell–Lumiere recording thus do not fall within the rule of completeness, and, "as offered by him, are inadmissible hearsay." See United States v. Jackson, 180 F.3d 55, 73 (2d Cir. 1999). Alternatively, Lumiere argues that the entire recording should have been admitted under Rule 106, i.e., not merely the few extra minutes his trial counsel sought to introduce but the entirety that he never sought to introduce. He claims that the recording as a whole contains further compelling evidence of his good faith, including, for example, statements to the effect that he was considering reporting the misconduct to the SEC under a "whistleblower" provision, and that he did not believe he had been engaged in wrongdoing because pricing Visium's "book" was Plaford's responsibility. See GX 1202, Ex. A. to Declaration of Jonathan Halpern, ECF No. 93, at 30, 36, 52–53, 117. But in addition to the dispositive failing that the entire recording was never offered, this argument fails for the same reasons that Lumiere's attempt to admit the more limited portions of the recording fails. To the extent that statements in the transcript are even arguably relevant to Lumiere's conduct—and a review of the 131–transcript demonstrates that nearly all of it is palpably irrelevant—they are post-hoc explanations of the conduct described in the admitted portion, and hence are inadmissible hearsay as offered by Lumiere.
Lumiere also argues that the Court erred in excluding these statements in part because they were not "reasonably close" to the portions that the Government offered, see Tr. 269. He cites a decision by another judge of this district that states that "mere proximity is a poor metric in deciding ‘fairness' under [Rule 106 ]." See Lesniewski, 2013 WL 3776235, at *4. But quite aside from the fact that Lesniewski is not binding on this Court, which adheres to its view that proximity is relevant, Lesniewski itself held that a defendant cannot have his self-serving and after-the-fact justifications admitted as "context" for what he actually did, see id. at *5, so this case provides no support whatsoever for Lumiere's ultimate position. Moreover, Lesniewski made that proximity-based point in the course of rejecting the defendant's argument that close proximity is sufficient to admit a statement; it says nothing about whether, all else being equal, one statement's remoteness from another is relevant to whether it provides necessary context.
Lumiere's cases are not to the contrary. In each, the erroneously excluded portion of the statement or document was needed to avoid misleading the jury about the evidence before it. See Phoenix Assocs. III v. Stone, 60 F.3d 95 (2d Cir. 1995) (error to exclude work papers that showed how values appearing in financial statement were calculated, where the financial statement appeared to exclude certain assets that were in reality accounted for); United States v. Walker, 652 F.2d 708 (7th Cir. 1981) (error to exclude defendant's prior testimony about his refusing a bribe and the reasons for his attendance at a meeting where a bribe was offered, where admitted portions of the testimony contained fragments of both ideas).
Taking another crack at the same recording, Lumiere also argues that it was also necessary to admit the full recording because the Government "opened the door" when it elicited the supposedly improper testimony from Thorell that he understood "Chris" to mean both Plaford and Lumiere. See Tr. 238. "The concept of ‘opening the door,’ or ‘curative admissibility,’ gives the trial court discretion to permit a party to introduce otherwise inadmissible evidence on an issue (a) when the opposing party has introduced inadmissible evidence on the same issue, and (b) when it is needed to rebut a false impression that may have resulted from the opposing party's evidence." United States v. Rea, 958 F.2d 1206, 1225 (2d Cir. 1992). Lumiere contends that the balance of the Thorell–Lumiere recording would demonstrate that when Lumiere said "Chris," he meant only Plaford, and would thereby correct the supposedly false impression that Thorell's testimony created.
This argument is meritless. First, and once again dispositive in itself, Lumiere never sought to admit the entire recording under this theory or any other. Further, even if he had, it is inadmissible under the curative admissibility doctrine. As the Court has already found, Thorell's testimony was properly admitted as lay opinion testimony, and "[p]roperly admitted evidence does not open the door to inadmissible evidence," such as the remainder of the recording. See Rea, 958 F.2d at 1225. At most, it simply broadens the scope of relevance. Nor does the recording as a whole rebut the supposedly false impression that "Chris" included both Plaford and Lumiere. To the contrary, the recording is repletewith references to Plaford and Lumiere jointly pulling off the mismarking scheme, that, if anything, would bolster Thorell's testimony. For example:
STEFAN LUMIERE: Well, I mean that's how, that's how we operated. So you would send us like—every Friday and month end—I'm sorry, not every Friday, every month end, Chris would call me—
JASON THORELL: Uh-huh.
STEFAN LUMIERE: —and say these are the lows that you need to, that these are the lows that they are. Some of them I knew, some of them I didn't know and he always had an explanation for why.
JASON THORELL: Right.
STEFAN LUMIERE: Oh, we had bids out. You know, Jeffrey is quoting them too low. They don't know where they are. This is where, you know, we had bids out and we weren't able to get something done in like '91, so we think they're like '93.
JASON THORELL: But that, the thing is that doesn't matter if he has an explanation because he's lying. You know that.
STEFAN LUMIERE: Yes. So that's the way he did it. Now, at a point in time—and the CMED thing, I wasn't really involved. I helped out at the beginning.
GX 1202 at 40–41 (emphasis added). Thus, even had the entire recording been offered, it would have properly been excluded.
Lumiere's last evidentiary argument is that a recording in which he discussed extorting Visium should have been excluded as inadmissible propensity evidence. The admissibility of "bad act" evidence of this nature is governed by Federal Rule of Evidence 404. "Evidence of a crime, wrong, or other act is not admissible to prove a person's character" in order to prove a defendant's propensity to commit a charged crime, but may be admissible to prove, inter alia, intent and knowledge. See Fed. R. Evid. 404(b)(1), (2). The Second Circuit "follows the ‘inclusionary’ approach, which admits all ‘other act’ evidence that does not serve the sole purpose of showing the defendant's bad character and that is neither overly prejudicial under Rule 403 nor irrelevant under Rule 402." United States v. Curley, 639 F.3d 50, 56 (2d Cir. 2011). Thus, "[p]ior bad-acts evidence must be (1) offered for a proper purpose, 2) relevant, and (3) substantially more probative than prejudicial. In addition, (4) at the defendant's request, the district court should give the jury an appropriate limiting instruction." United States v. Downing, 297 F.3d 52, 58 (2d Cir. 2002).
Here, Lumiere argues that the following recording played for the jury was inadmissible under this standard:
STEFAN LUMIERE: You got a lot of fucking shit to put him, put him out of business.
JASON THORELL: I think.
STEFAN LUMIERE: Well, first we can extort him. I'd love to be able to go to him and say listen, Jake, Chris, give us $100 million between the both of you.
JASON THORELL: Personally I'm just interested in prision time. I mean, I—
STEFAN LUMIERE: Yeah.
JASON THORELL: I, I—
STEFAN LUMIERE: Well, but I'm very confident that, like, any of of this stuff gets out, they'll be shut down.
JASON THORELL: Yeah.
STEFAN LUMIERE: They will be shut down.
JASON THORELL: I think that's very possible.
See GX 1222. On direct examination, Thorell testified that he understood Lumiere to be suggesting they blackmail Plaford and Jacob Gottlieb, the founder of Visium, based on evidence of wrongdoing that Thorell had gathered. See Tr. 374, 398–99. Lumiere argues that this recording is irrelevant to any issue in the case and "inflame[d] the passions of the jury." See Defendant Stefan Lumiere's Memorandum of Law in Support of His Motion to Vacate Judgment and Grant a New Trial ("Def. Mem."), ECF No. 90, at 19.
The extortion recording was properly admitted. The defense theory in this case, previewed in the defense counsel's opening statement and reiterated throughout his summation, was that Lumiere acted in good faith throughout the multiyear conspiracy—that he believed that the securities prices were fair and accurate and that he never had any cause to doubt those prices until shortly before his tenure at Visium came to an end. See, e.g., Tr. 20–21, 1016–17, 1030, 1045. In particular, during his opening statement, Lumiere argued that the jury should infer from the fact that Lumiere recorded himself discussing price quotes with Plaford and the brokers alike that Lumiere was acting in good faith, because the last thing a knowing fraudster would do is keep meticulous records of the fraud. See Tr. 30; see also Tr. 1017.
Against that background, the purpose and relevance of the extortion recording is plain. Multiple valid inferences could be drawn from the fact that Lumiere made recordings of the mismarking scheme. Lumiere invited the jury to infer that they showed he participated in the mismarking scheme in good faith, but the opposite inference is also possible: that he recorded the conspiracy to use as a weapon against his coconspirators because he knew what they were doing was wrong. Accordingly, the Government, both to rebut Lumiere's innocent explanation for his recordings, and to reinforce its own more sinister explanation, introduced the extortion recording, which narrowly and precisely showed just how Lumiere viewed recordings of the mismarking scheme: as fodder for blackmail. That is strong evidence that Lumiere's own recordings were made for the same purpose, thereby demonstrating his lack of good faith.
Nor was the recording unfairly prejudicial. The brief recording describes an idea for an extortion attempt that is plainly in its infancy and no further evidence of the idea was offered. The Government made only passing use of it in its own summations, again to rebut the defense's inference that Lumiere's recordings were evidence of his good faith. See Tr. 1006–07. Moreover, the extortion scheme, such as it was, did not involve some graphic threat of violence and mayhem, the sort of thing that might inflame a jury's passions; it took place instead in the arid and arcane world of distressed debt investing.
In arguing to the contrary, Lumiere chiefly relies on a superficially similar, but readily distinguishable, case decided by the Ninth Circuit. See United States v. Hodges, 770 F.2d 1475 (9th Cir. 1985). Hodges was a fraud case in which the defendant argued that he had no knowledge of "the fraudulent nature of the scheme being carried out around him," and the district court permitted a cooperating witness to testify that five months after the conspiracy ended, the defendant demanded $5,000 or else he would include her name on a " ‘hit list’ he intended to supply to a group of his murdered brother's less savory friends." Id. at 1477–78. The Ninth reversed, finding that the district court abused its discretion in allowing this testimony, both because it was irrelevant to whether Hodges knew about the fraud at the time it was committed and because it was "highly and unfairly prejudicial," as it drew the jury's attention away from the charged offense "to a different act of a most serious nature." Id. at 1479–81.
Even if Hodges were binding on this Court (which it is not), it is plainly distinguishable. The extortion attempt in Hodges had only the most tenuous connection to the defendant's guilty knowledge; it was simply introduced to ask the jury to infer that if the defendant knew the scheme was illegal after the fact, then he must have known it was illegal while he was carrying it out. In this case, by contrast, the extortion recording directly rebutted Lumiere's argument that his own recordings were affirmative evidence of his innocence, by giving the jury a firm basis from which to infer that he made the recordings for a more sinister purpose suggesting guilty knowledge. And the danger of prejudice in Hodges illustrates why this case presents none of the same risks. The extortion evidence in this case was a brief and bloodless recording of a plan that never got past the idea stage, whereas in Hodges, it was live and lurid testimony of threats of physical violence that left little to the imagination.
In particular, the cooperating witness testified as follows:
He told me that he was coming the next day to talk to the FBI, and that if I would give him $5,000, he would paint a pretty picture of myself and my participation in the transaction to the FBI, and he further told me that there were people who were friends of his brother from New York that were not nice people; they knocked on your house in the middle of the night; they hurt you; they hurt your children, and that they had requested from him to supply them with a list of people that he felt were being uncooperative regarding some things that happened after his brother's death, and that if I would give him $5,000, he would see that my name was taken off this list.
See Hodges, 770 F.2d at 1478 n.1.
In sum, on de novo review, the Court finds no evidentiary errors, much less any calling into doubt the fairness of the trial.
Lumiere next argues that the fairness of his trial was fatally compromised by prosecutorial misconduct in the form of false statements made during rebuttal summation. To determine whether prosecutorial misconduct is so prejudicial as to require a new trial, courts examine "(1) the severity of the misconduct, (2) the measures adopted to cure the misconduct, and (3) the certainty of conviction absent the misconduct." United States v. Truman, 688 F.3d 129, 144 (2d Cir. 2012) (alterations omitted). The ultimate question is whether the misconduct "so infect[ed] the trial with unfairness as to make the resulting conviction a denial of due process." See United States v. Parkes, 497 F.3d 220, 233 (2d Cir. 2007).
Here, Lumiere faults the Government for repeatedly arguing, during rebuttal summation, that there was no evidence of Lumiere's good faith, and in particular, that in the excerpt of the Thorell–Lumiere recording that was admitted into evidence, Lumiere never stated that he only realized that Visium was mismarking securities after the fact. See, e.g., Tr. 1052 ("You know what he doesn't say there? He doesn't say, ah, I didn't realize the Credit Fund was mismarked till later."). As Lumiere sees it, this error grows out of the supposedly flawed exclusion of much of the recording from evidence. Having succeeded in keeping most of the recording out, the argument goes, the Government improperly told the jury over and again that no evidence of the defendant's good faith existed, despite knowing that the excluded portions of the recording supposedly contained such evidence.
In total, the Government made five statements along these lines during rebuttal summation. For the other four, see Tr. 1052 ("And this notion that he didn't understand till after the fact, there is no evidence of that."); Tr. 1053 ("Now, the other problem with the argument that he didn't realize until after the fact—and there is no evidence of that, by the way ...."); Tr. 1059 ("There is not one document, one model, no shred of paper in this case that shows the defendant had a good faith belief in these prices."); Tr. 1066 ("[T]here is no evidence ... that he didn't know until after the fact ....").
The argument is frivolous. To begin with, even though the Government made the allegedly offending statement five times in its rebuttal summation, no objection was lodged by the defense, either during the summation or at any time thereafter. Equally important, the questioned statements were in fact perfectly proper rebuttal to defense counsel's own argument on summation asking the jury to infer the defendant's good faith. The defense claim now is, in effect, that even if the Government's statements were fair comment on what was in evidence, they were not fair comment given what the Government knew was on the tapes that were excluded from evidence. Cf. United States v. Blueford, 312 F.3d 962 (9th Cir. 2002). But this presupposes that the Government accepts, or reasonably should have accepted, the defense's highly self-serving interpretation of what the excluded tapes evidenced. On the contrary, however, the Government's interpretation of the excluded portions of the tapes, i.e., that they were either irrelevant or, if anything, further proof of guilt, is not only a reasonable interpretation but is, in the Court's view, a much more reasonable interpretation than the one now proffered by the defense. Accordingly, the Government's comments, far from constituting misconduct, were right on target.
To be sure, as the defendant points out, the full recording does include some statements that a defense counsel might argue, if weakly, were evidence that Lumiere did not contemporaneously realize that he had been engaged in a fraud. But even this evidence, such as it is, is fragmented, weak, and unconvincing, and the Government would be well within its rights to deny it was evidence of good faith at all. Consider, for example, what defense counsel now claim was to be the centerpiece of Lumiere's argument that he did not realize at the time Visium was mismarking securities:
And then when it came to [ATI], everything kind of made sense. A lot of people didn't know where the levels were.... The fact that we were in a control position kind of like made more sense. And we kind of backed into the price based on estimates of EBITDA and value for the company because we knew who we, who had bidding for sale so we knew more than RBC did about it. So that kind of was okay.
See Proposed Ex. 1224–T and 1225–T, at 5–6; see also Defendant Stefan Lumiere's Reply in Further Support of His Motion to Vacate Judgment and Grant a New Trial, ECF No. 95, at 4. According to his statements on the tape, Lumiere later demanded that Plaford downgrade Visium's valuation for ATI in light of ATI's dramatically downgraded 2013 EBITDA estimates, but Plaford refused. See GX 1202 at 43–45.
But even on its face, defendant's interpretation of the statements is unconvincing. Thus, defendant states on the tape that, at the time, Plaford's explanations for his ATI pricing decisions were only "kind of ... okay" and only "kind of made sense." The Government could reasonably infer from this half-hearted endorsement of Plaford's methods that Lumiere knew all along that Plaford's explanations for the ATI marks were flimsy. Lumiere's focus on ATI alone is also misplaced, for many different securities were mismarked, and in that context, Lumiere's caveat that "when it came to [ATI], everything kind of made sense" takes on a more sinister light. The obvious inference is that even at the time, Plaford either had no explanations for his unusually high prices or, if he did, Lumiere knew that they held no water. It thus apparent that Lumiere's good-faith argument scarcely had any basis in this recording at all.
Further still, the Government's rebuttal comments cannot be evaluated in isolation, but must instead be considered in the context of the evidence as a whole. In this regard, the Government's rebuttal was a direct response to the defense summation, where Lumiere argued that the jury should infer from the tapes that Lumiere only learned of the misconduct after the fact. The defense's argument was made without any factual predicate at all, save for the absence of a statement from Lumiere that he knew of the misconduct as it was occurring. See Tr. 1017 ("[I]f a recording had Stefan Lumiere saying that he knew the securities were mismarked at the time he was involved in getting those broker quotes, the Government would have played it."). In reality, of course, this was an argument based on excluded (or not offered) portions of the tape in which, as Lumiere saw it, he made statements consistent with his having learned of the misconduct only after the fact. It was against that background that the Government argued that the record was bereft of evidence of Lumiere's good faith. That statement was, of course, literally true; the admitted recording contained no such statement. Just as important, the Government must be given some latitude to respond to the defense's argument. See United States v. Marrale, 695 F.2d 658, 667 (2d Cir. 1982) ("[A] prosecutor is ordinarily entitled to respond to the evidence, issues, and hypotheses propounded by the defense ...."). It would be perverse to find that the Government was not permitted to factually describe the admitted portion of the recording, while allowing the defense to effectively rely on excluded portions of the recording.
As the foregoing discussion makes clear, Lumiere's chief authority is readily distinguishable. In Blueford, the Ninth Circuit found that prosecutorial misconduct required a new trial where the prosecutor argued that the jury should infer from a pattern of telephone calls that the defendant was concocting an alibi before trial, even though the prosecutor was aware that the calls, which were recorded but not in evidence, contained nothing of the sort. See 312 F.3d at 968–69. Here, by contrast, even assuming arguendo that the Government was bound to conform its summation to materials outside the record, it was well within its discretion to characterize the unadmitted portion of the Thorell–Lumiere recording as containing no evidence of good faith, which is the exact inference it would have drawn (and asked the jury to draw) if those materials had come in.
Although it perhaps amounts to overkill, it also worth noting that, even if the Government supposedly overplayed its hand in repeating that there was no evidence of good faith in the tapes, the "certainty of conviction absent the misconduct" is unshakeable. See Truman, 688 F.3d at 144. Indeed, the evidence was overwhelming that Lumiere knew that mismarking securities was wrong, and knew that he was mismarking securities.
In particular, to show that Lumiere knew that mismarking securities was wrong, the Government introduced evidence relating to Lumiere's extensive professional certifications. In obtaining these credentials, he frequently encountered, and had to demonstrate his mastery of, the principles that "inaccurate market quotations" and "painting the tape" were prohibited. See Tr. 546–54; GX 809, 812. Visium's own internal policies, to which Lumiere certified he would adhere, provided that Visium's pricing practices were to be "fair, consistent, and verifiable" by independent third parties. See Tr. 55–62, 66–68; GX 906, 1120. Visium's policies further provided that while pricing overrides from the investment team were acceptable in principle, they must be done on the basis of independent judgments. See Tr. 79–87; GX 906. So it is plain that Lumiere knew that mismarking securities was wrong.
There can likewise be no doubt that Lumiere knew he was mismarking securities. Over two years, Lumiere suborned small-time brokers into providing hundreds of prices for arcane securities they never traded and did not cover. See Tr. 314–15, 351; GX 1224. Lumiere dictated prices to them with the expectation that they would be parroted back to him, and the brokers always complied, often within minutes of Lumiere's request. See Tr. 286, 354–58, 489–94, 679–80; GX 1227. They did this in exchange for Lumiere steering Visium business their way. See, e.g., Tr. 285–86, 294. Their lack of independence was thus obvious and, in light of Lumiere's professional certifications and knowledge of the compliance manual, was obviously wrong.
Moreover, Lumiere tried to cover his tracks. He did not put his requests for specific securities prices in writing, preferring instead to dictate the prices by cell phone. Calls of this nature that were admitted into evidence reveal that Lumiere was doing no more than reading off prices to the brokers, contra the defense argument that he was "educating" them. See GX 1209, 1227. And when Lumiere had a particularly extensive demand, he avoided even that much of a trail, and instead messengered flash, drives containing securities prices to the brokers. See Tr. 364–66, 506–07, 682.
Finally, in another recording from April 2013 to which Lumiere has raised no challenge, Lumiere told a friend that Visium's securities pricing practices were "fucking bullshit." See GX 1231. In light of this evidence, it is plain that the Government's remarks on rebuttal that there was no evidence of Lumiere's good faith did not "so infect[ ] the trial with unfairness as to make the resulting conviction a denial of due process." See Parkes, 497 F.3d at 233.
As his last argument in support of the instant motion, Lumiere lodges two complaints relating to the jury instructions on "conscious avoidance." Lumiere first contends that there was an insufficient factual predicate for the conscious avoidance jury instruction, because, he claims, the Government's theory of the case was that the defendant had actual knowledge of the fraud, and the defense theory was that he had a good faith belief in the prices, neither of which implicates conscious avoidance. This argument is meritless. "A conscious-avoidance charge is appropriate when (a) the element of knowledge is in dispute, and (b) the evidence would permit a rational juror to conclude beyond a reasonable doubt that the defendant was aware of a high probability of the fact in dispute and consciously avoided confirming that fact." United States v. Ebbers, 458 F.3d 110, 124 (2d Cir. 2006). It is well-settled that a conscious avoidance charge is not inappropriate merely because the Government seeks to prove actual knowledge. See, e.g., United States v. Svoboda, 347 F.3d 471, 480 (2d Cir. 2003) ("[T]he same evidence that will raise an inference that the defendant had actual knowledge of the illegal conduct ordinarily will also raise the inference that the defendant was subjectively aware of a high probability of the existence of illegal conduct.").
Here, the charge was appropriate, because Lumiere's knowledge was the critical fact in dispute, and because a rational juror could find beyond a reasonable doubt that Lumiere deliberately avoided learning about the fraud based on, for example, the recordings in which Plaford listed off securities values he wanted Lumiere to obtain from brokers with no "explanation" whatsoever, see, e.g., GX 1229; the recordings in which Lumiere dictated prices to Vandersnow, see, e.g., GX 1227; and the evidence that Lumiere avoided a paper trail when requesting price quotes, preferring instead to communicate by cell phone or, when he had a particularly large request, by a flash drive he messengered to the brokers, see Tr. 364–366, 506–07, 682. See United States v. Cuti, 720 F.3d 453, 464 (2d Cir. 2013) ("This purported lack of knowledge defense, despite Tennant's deep involvement in the transactions that effectuated the fraud, all but invited the conscious avoidance charge.").
Lumiere also argues that the Court inserted the conscious avoidance instructions sua sponte, when, in fact, the Government requested such an instruction, see Government's Requests to Charge, ECF No. 48, at 32–33, to which the defendant did not object. See Tr. 960–67.
Lumiere's second attack on the jury instructions is that the use of hypothetical illustrations of conscious avoidance prejudiced the jury. The relevant portions of the instructions read as follows:
For example, if you find that the defendant simply failed to do adequate "due diligence" on the securities he was pricing, that by itself would not support an inference of deliberate disregard of possible fraud. But if, by contrast, you find that the defendant was aware of the high probability that the bond prices he gave to brokers were materially higher than the fair market prices for those bonds, but that the defendant, in order to remain ignorant of that fact, deliberately chose not to inquire further, then you may, if you wish, find that the defendant actually understood that the bond prices he gave to brokers were artificially high and did not reflect the fair market value.
See The Court's Instructions of Law to the Jury, ECF No. 61, at 14–15. Lumiere claims, for example, that these hypotheticals led the jury to believe that the knowledge element was objective, rather than subjective, and that they "imposed on the defendant a burden to ‘inquire further.’ " See Def. Mem. at 22.
These arguments are wholly without merit. The conscious avoidance charge plainly tracks the black-letter requirement that knowledge is subjective by requiring the jury to find the defendant to be "aware of the high probability" that he was dictating baseless prices; and, by its own terms, it imposes no burden on the defendant at all. What's more, jury instructions are reviewed in their entirety, not in isolation, see United States v. Shamsideen, 511 F.3d 340, 345 (2d Cir. 2008), and here, the instructions as a whole repeatedly made clear that the defendant had no burden of proof. See, e.g., Jury Instructions, at 6, 11. For the foregoing reasons, Lumiere's motion for a new trial is denied. The Clerk of Court is directed to close docket number 91.
Following the trial, Lumiere retained new counsel, who expressed an interest in "investigating" whether Lumiere's trial counsel was constitutionally ineffective. See, e.g., Memorandum of Law in Support of Motion of Defendant Stefan Lumiere to Extend the Time to File Rule 29 and 33 Motions to March 31, 2017, ECF No. 87 ; Def. Mem. at 1 n.1. The Court denied the defendant's motion for a lengthy extension of time to file post-trial motions in order to investigate ineffective assistance after the defendant failed to identify a single good-faith basis warranting such an "investigation." See Order dated Feb. 7, 2017, ECF No. 89. In the end, defendant's counsel advanced no such argument on this motion.
--------
SO ORDERED.