Opinion
19-CR-602 (RA)
09-12-2023
UNITED STATES OF AMERICA, v. MICHAEL HILD, Defendant.
OPINION & ORDER
RONNIE ABRAMS, UNITED STATES DISTRICT JUDGE
Defendant Michael Hild was convicted following a fourteen-day jury trial of committing securities fraud, wire fraud, and bank fraud, as well as conspiring to do so. The evidence at trial established that Hild and his co-conspirators at Live Well Financial (“Live Well”) engaged in a multi-year scheme to fraudulently inflate the value of a portfolio of Home Equity Conversion Mortgage “interest only” bonds (“HECM IO bonds”) used as collateral to secure cash loans. Although the loan amounts were nominally based on market prices for the HECM IO bonds provided by a third party, the evidence demonstrated that Live Well had directly supplied valuations to that third party, unbeknownst to its lenders, basing them on its own novel internal pricing methodology rather than on what the bonds could readily be sold for in the market. As a result, Live Well was able to purchase the bonds at one price, provide the third party its own inflated valuations, and then use those inflated bond values as collateral to take out loans worth significantly more than the price for which the bonds could be sold, resulting in a substantial cash windfall for Live Well.
Following his conviction, Hild filed motions pursuant to Federal Rules of Criminal Procedure 29 and 33 for, respectively, a judgment of acquittal and for a new trial. The Court denied those motions in a 64-page opinion and order issued in December 2022, see United States v. Hild, __ F.Supp.3d __, 2022 WL 17484992 (S.D.N.Y. Dec. 7, 2022), and sentenced Hild to 44 months' imprisonment, see Dkt. 147. Hild's appeal is now pending before the Second Circuit, see Hild v. United States, No. 23-6136 (2d Cir.), and this Court has granted him bail pending the resolution of that appeal, see Dkt. 154.
At his sentencing, the Court ordered Hild to pay restitution to the victim lending institutions, see Dkt. 148, but reserved judgment on the amount of restitution to be paid pending further justification from the Government. See Dolan v. United States, 560 U.S. 605, 611 (2010) (holding, where a district court has ordered restitution within the 90-day period set pursuant to 18 U.S.C. § 3664(d)(5), but has not set an amount, that it retains jurisdiction to order the amount at a later date). The Government has now filed a proposed order of restitution, along with supporting materials related to the harm suffered by the victim lenders. See Dkts. 161, 187. In critical part, the documents supporting the proposed restitution amount include information related to millions of dollars in so-called “coupon payments”-that is, annual interest payments-received by the victim lenders for holding the HECM IO bonds which were offered as collateral by Live Well.
Hild's latest motion seizes upon the Government's recent restitution submissions to argue that these materials-which show significant coupon payments to the lenders-constitute “newly discovered evidence” under Rule 33, and thus warrant Hild a new trial. In the alternative, he requests additional discovery related to coupon payments to the lenders, so that an appropriate restitution calculation can be made. Finally, Hild argues that the Supreme Court's recent decision in Ciminelli v. United States, 143 S.Ct. 1121 (2023), which invalidated the Second Circuit's “right to control” theory of wire fraud, id. at 1125-28, renders jury instructions with respect to Counts Two and Four plainly erroneous, further justifying that he be granted a new trial.
For the reasons that follow, and given the particularly stringent standard attendant to Rule 33, Hild's motion for a new trial on the basis of purported newly discovered evidence is denied. With the consent of the parties, the Court will refer the calculation of the restitution amount to be paid to Magistrate Judge Parker, who may also make a determination as to whether the Defense should obtain additional discovery related to the harm suffered by the victim lending institutions. See 18 U.S.C. § 3664(d)(6) (providing that a district court “may refer any issue arising in connection with a proposed order of restitution to a magistrate judge or special master for proposed findings of fact and recommendations as to disposition”). Lastly, the Court defers the issue of whether Ciminelli renders the fraud jury instructions in this case to be plainly erroneous to the Second Circuit, before which Hild's appeal remains pending.
BACKGROUND
The Court assumes the parties' familiarity with the underlying facts and procedural history of this action, about which the Court has previously written extensively. See United States v. Hild, __ F.Supp.3d __, 2022 WL 17484992, at *2-8 (S.D.N.Y. Dec. 7, 2022). The following is taken from the Court's prior opinion, and summarizes those facts relevant to Hild's present motion.
I. Evidence Presented at Trial
A. Founding of Live Well and Reverse Mortgage Servicing
Live Well was established in 2005, and, at all relevant times, Hild was its CEO, Tr. 1247, 1867, and largest shareholder, Tr. 1904-05; see also GX321. Hild founded Live Well to pursue a business opportunity in the burgeoning reverse mortgage space. Tr. 1785-88. Reverse mortgages, or Home Equity Conversion Mortgages (“HECMs”), are a financial product designed to provide liquidity to senior homeowners whose net worth is primarily tied up in their home equity. Tr. 5758, 578-79, 1788. HECMs permit individuals to receive monthly cash income by using their home equity as collateral. Tr. 579. For many years, Live Well operated as a traditional and reverse mortgage broker and servicer: it reviewed applications from borrowers, approved loans, and then serviced those loans. Tr. 946, 1790-91, 1814.
Beginning in approximately 2011, Live Well began securitizing reverse mortgages into bonds called HECM mortgage-backed securities (“HMBSs”). Tr. 947. After it began to sell these securities, it “[g]rew dramatically.” Tr. 1806. Packaging the mortgages as bonds allowed Live Well to generate revenue more quickly by selling pools of similar reverse mortgages to investors in bulk, rather than one-by-one. Tr. 947-49. Key among these bonds was a subcategory of HECM IO bonds, which are made up of derivatives of HECMs that include only the interest portion of the loan rather than the entire reverse mortgage. Tr. 59, 60, 950-52; see also Tr. 1814. The HECM IO bonds are particularly attractive to investors because holders of those bonds receive regular interest payments. Tr. 1818.
B. Expansion Into Purchasing HMBSs & the Stifel Transaction
In 2014, Live Well expanded beyond securitization of reverse mortgages into the purchase of HMBSs. Tr. 58-59, 951-52, 1817. Hild stated that the goal of purchasing these bonds and holding them as investments was to diversify the company's revenue and reduce its susceptibility to cyclical changes in the mortgage space. Tr. 1814.
At Hild's direction, the company acquired a portfolio of fifteen HECM IO bonds worth $55 million from Stifel Financial, a small investment bank that had previously held and traded in HMBSs (the “Stifel Transaction”). Tr. 57, 72-73, 951-53. In addition to the bond portfolio, Live Well also hired three Stifel employees-Darren Stumberger, Ernie Calabrese, and Dan Foster- who had managed the portfolio at Stifel, to continue their work at Live Well. Tr. 969. Stumberger, described by Hild as the “reverse mortgage bond guru,” Tr. 1815-17, was a particularly important hire due to his expertise in the area, see id.; see also Tr. 73-74, 969.
Live Well did not have the cash on hand to buy the bonds outright, and it thus financed the acquisition using a combination of cash, “warehouse loans” it had access to through its mortgage business, and loans collateralized by the underlying bonds, the latter of which were referred to as “repo financing.” Tr. 76-77, 731-32, 952. The repo financing agreements were loans structured as repurchase agreements in which Live Well sold the bonds to lenders and agreed to buy them back at a specific price after a short period had passed, typically thirty to sixty days. Tr. 76-77, 261; see also Tr. 1826. At the end of the period, lenders would generally “roll” the loan forward, Tr. 674-75, but they could alternatively end the lending agreement and demand repayment of the loan amount, Tr. 572-75. In the case that Live Well was unable to repay the loan at the end of the term, the lenders would keep the collateral (the HECM IO bonds), which they could either hold or sell to repay the defaulted loan. Tr. 76-77.
Typically, the loan amount was determined by discounting the value of the underlying bond by 10% to 30%. Tr. 86. This discount, often called a “haircut,” ensured that the lenders remained sufficiently collateralized if the bonds decreased in value, and priced in the risk of a lender having to sell the bonds. Tr. 81, 588-89; see also Tr. 396, 573, 674-77. Different repo lenders extended loans with different haircut amounts, but Live Well preferred to borrow from lenders with lower haircuts, thereby maximizing the amount of cash it was loaned. Tr. 1069, 1162, 1461.
As the prices of the collateral fluctuated up or down, either party could request that the loan amount be adjusted accordingly. Tr. 85-87. If the value of the collateral decreased, lenders could require partial repayment of the loan amount via margin call, Tr. 85-86, and if the value of the bonds increased, Live Well could request to borrow more via “reverse margin call,” Tr. 86-87.
With one exception, the loan agreements between Live Well and the lenders required that the prices for the bonds be set by an independent third party. Tr. 78-80, 575-76, 598-600. To take one example, Live Well's contract with Mirae Asset Securities (USA) Inc. (“Mirae”) required that the amount of the loan be adjusted based on “the aggregate Market Value of all Purchased Securities.” GX603 at 3; Tr. 599-600. “Market Value” was defined as:
One lender, Nomura, had its own trading desk that set the loan amount “based on the market.” Tr. 79.
the price for such Securities on [a given] date obtained from (i) Interactive Data Corporation (“IDC”) or (ii) if no quotation is available from IDC, then a generally recognized source agreed to by the parties or the most recent closing bid quotation from such a source, plus accrued Income to the extent not included therein.GX603 at 11; see also Tr. 598-601. As lenders typically lacked the expertise or experience required to value the bonds themselves, see Tr. 575-76, they thus relied on a third party- Interactive Data Corporation (“IDC”)-to provide prices, id.; see also Tr. 81-82. Stumberger testified that, based upon conversations with the lenders, it was his understanding that they required third-party prices “because [they] wanted independent, unbiased view[s] on valuations,” and did not want the loan amount determined by the borrower or lender. Tr. 81-83.
At the time of the Stifel Transaction in 2014, however, IDC was itself unable to value the bonds, and thus relied on “broker quotes” to provide pricing. Tr. 173. Also at that time, Live Well was providing estimates as to what the bonds could be sold for in the market as broker quotes to IDC. Tr. 83. In other words: IDC published prices provided by Live Well, which were in turn used in repo agreements to calculate the size of the loans Live Well could secure from most of its lenders.
In January of 2015, IDC began using its own independent pricing model to value the bonds. Tr. 85; GX104. Hild and the co-conspirators, however, determined that IDC's model was deficient. Tr. 281-83. Among other issues, IDC was “running the portfolio evaluations daily,” and doing so caused a daily decline in the bond prices. Tr. 85-87. The overall downward pressure on the bond prices in turn led to margin calls from the lenders as they became under-collateralized. Id. The corresponding need to constantly repay the loans caused strain at Live Well, so Stumberger and his team were directed by Hild to “rectify this problem,” Tr. 87-88. Ultimately, IDC ceased modeling the bonds independently and returned to publishing Live Well's broker quotes. Tr. 8889. Under that system, Live Well “suppl[ied] prices daily to IDC,” which IDC used “verbatim.” GX104. Live Well and IDC agreed that IDC would “not model the[] bonds until [Live Well was] interested in them doing so.” Tr. 89, GX104. Thus, by February of 2015, Live Well had resumed providing broker quotes to IDC. Tr. 93-94.
C. Shift to Scenario 14 Pricing
Live Well developed its own internal models to project the value of the bonds later in 2015. Tr. 100-05, 990-93; GX113. Evidence adduced at trial showed that Hild and other Live Well employees believed that the market generally undervalued reverse mortgage bonds. As Stumberger, the so-called “reverse mortgage bond guru,” explained, the yield requirement of the bonds “should theoretically . . . drift down gradually” over time, but “it wasn't working like that in the market.” Tr. 111-12; see also GX402. He described the prevailing methodology as “improper” and “not sound,” Tr. 112, and acknowledged the difference between the theoretical value of the bonds and “what happens in reality,” GX402.
An assumption regarding yield requirements, among others, was incorporated into a new methodology for valuing the bonds called Scenario 14. Because the Scenario 14 assumptions differed from factors “in the market,” the prices of the bonds derived using Scenario 14 were typically higher than those for which they could be sold. Tr. 113-14. Indeed, the jury heard testimony from multiple co-conspirators that they knew the Scenario 14 prices exceeded what the bonds were worth in the market. Tr. 158, 993-97, 1014.
Nevertheless, at Hild's direction, Tr. 91, 93, 103, Live Well began to submit Scenario 14 prices to IDC in September of 2015, Tr. 137, 150, 322-23, 410, 998-99. The shift to Scenario 14 prices in turn resulted in an $11 million increase in the value of Live Well's portfolio. Tr. 11922; GX402 at 29. But because such a substantial increase in a single day would “raise alarm bells all over the place,” in Hild's own words, GX402 at 29, he directed his employees to implement the price increase gradually using a “glide path,” id., so as to avoid triggering any “red flags,” Tr. 121-22; see also Tr. 1004-11; GX402; GX403; GX404.
Given that most of its repo agreements relied on IDC, the change to Scenario 14 pricing allowed Live Well to enter loan agreements where the loan amount would exceed the purchase price of the bond, which resulted in an immediate windfall for the company. Tr. 128 (Stumberger testifying Live Well could “purchase new bonds in the market and apply the pricing, the [S]cenario 14 pricing, submit that to IDC, and because the lenders were relying on IDC pricing, they would be lending more money than needed to purchase the bond at the market.”); see also Tr. 129-34. As Eric Rohr, the company's Chief Financial Officer, explained, “every time we bought bonds, we would have money left, extra money coming back to our pocket.” Tr. 1002. Hild referred to this process as a “little bit of a self-generating money machine.” Id. All told, by the end of 2015, the shift to Scenario 14 valuations had resulted in an increase in the value of the company's portfolio of more than $47 million. Tr. 1084.
The witnesses' testimony was consistent that the Scenario 14 assumptions model was created as an “academic exercise” to value the bonds. Tr. 995. The jury heard competing narratives, however, concerning the intent behind submitting the Scenario 14 prices to IDC. According to Hild, these prices were the company's best effort to value the bonds, and thus submitting those values to IDC was appropriate. See, e.g., Tr. 1874-76. Several co-conspirators, however, testified that they knew that providing the Scenario 14 prices to IDC and then borrowing against those prices was “wrong,” Tr. 307-09, 1002, 1539, and that they believed that if the lenders had known the IDC prices were based on Live Well's internal investment “thesis,” rather than the prices the bond could be readily sold for in the market, they would not have entered the agreements, Tr. 136, 309-12, 996-1009. The lenders expressed this same view directly. Tr. 605-10.
Stumberger went further, testifying that Hild directed the submission of Scenario 14 prices not as part of a good faith effort to value the bonds, but in order to increase Live Well's ability to borrow cash from lenders:
Q. How did you feel about the decision to submit these [Scenario 14] prices to IDC?
A. I was concerned. I was nervous.
Q. You said this was done at whose direction?
A. Michael [Hild]'s.
Q. Based on your conversations with Mr. Hild, did you have an understanding about why he wanted these higher prices to be submitted to IDC?
A. Yes.
Q. What was your understanding?
A. That it would create a situation, because of the higher valuations, that Live Well could borrow more money via [reverse] margin call.Tr. 115-16. Live Well continued submitting prices to IDC, and, from 2015 onward, IDC published those prices “verbatim.” Tr. 1014; 1254-55.
D. 2017 Liquidity Events
In January 2017, one lender, Wedbush, asked to speak with a Live Well dealer for more information about how the bonds were being valued. Tr. 166-67. On a recorded call, Hild and other co-conspirators discussed how to respond to the request. See GX407. There were several risks posed by Wedbush discussing prices with a dealer, in that Wedbush could learn the IDC rates were above market rates and/or learn that the prices were broker quotes that came from Live Well. Tr. 166-77. Hild and others discussed their options: Live Well could screen a dealer in advance, to see if they would come up with a valuation matching the Scenario 14 prices on IDC, or they could find a “slimy” dealer willing to represent the IDC prices were correct. Tr. 169-75, 1147, GX407. Eventually, Hild questioned whether they could persuade Wedbush that the rates were not improper, given that Live Well was uniquely qualified to evaluate the bonds. Tr. 136-42. Stumberger responded that Wedbush and other dealers were not concerned with the “intrinsic” value of the bonds, but rather on what they could be sold for in the market:
HILD: . . . the reason we're in this space is because we're one of the few people that know how to value these securities, right? Would you not disagree with the statement, Darren? That most of the folks that are out there don't know how to value these securities including the big guys?
STUMBERGER: Well, the dealers-the dealers make the market. They offer bonds a certain way. They finance them a certain way. So, whether they-they're doing it right or not is -
HILD: Well, that's my point-
STUMBERGER: -I don't think Scott [at Wedbush] cares about that. Scott cares about what is the market. And it's gonna be defined-the market is going to be defined as something that is an 8 or 9% yield to 50 HPC on the- on the-on the sell side. On the-on the ask. The bid is behind-is a higher yield. That's what he's gonna get from, we know, the guys who are pedaling the paper. May not be right, or proper, or sound, but that's what's gonna be defined as the market. That's just realistic.
HILD: There's no debating that.GX407; see also Tr. 136-42.Hild and his co-conspirators also discussed whether to disclose their methodology to the market, which would hurt Live Well “in some respects” because, even if the others in the market agreed, it would increase the purchase price of the bonds. Tr. 1155-56. Another option they considered was to mark down the portfolio to market rates. Tr. 1156-57. Hild considered this option “obviously a non-starter,” as it would have resulted in a $50 million loss on the company's income statement and triggered a substantial margin call. Tr. 1157.
The jury heard similar testimony, that the lenders ultimately cared only about the price for which the bonds could be sold in the market, from one lender directly. See Tr. 600-01.
In addition to requesting additional quotes, Wedbush also stated it was “no longer comfortable extending credit” with respect to a certain type of bond (MACRs), and that it would reduce Live Well's overall credit line. Tr. 1157. These events created “potential liquidity problems” for the company. Tr. 1157. No other lenders would finance the MACRs, and because they could not be sold at the rate Live Well was “carrying them,” Live Well would need almost $22 million in cash to avoid defaulting on its debts. Tr. 1158-63; GX151. Even after Stumberger and Foster “optimized” as best they could, Live Well was left with an effective deficit of $21.8 million, a situation Hild described as a “catastrophe.” Tr. 1164; GX151.
Rohr testified that “at some point” it became clear to the Live Well team that the bonds could not be reshuffled and new financing was unavailable such that the “only way out of this situation was somehow getting the value of [Live Well's] collateral increased and getting more credit, and therefore, being able to borrow more against the existing collateral to kind of help facilitate some of these moves and pending liquidity hits.” Tr. 1166-67. Rohr testified that Hild instructed them “to raise prices in IDC,” although Hild acknowledged that “the trading desk is not going to like it.” Tr. 1167. No market rationale was supplied for this proposed increase, and Rohr testified the prices were no longer consistent even with the inflated valuations calculated under Scenario 14. Tr. 1167-68. Rohr described feeling “sick to [his] stomach” at this point, as the prices being submitted were “beyond even our own investment thesis . . . way beyond that.” Tr. 1168.
Stumberger and others eventually modeled potential scenarios that would resolve the liquidity crisis, including one option, “Scenario 4,” which they ultimately agreed to submit. Tr. 1169. Rohr testified that the “purpose” of Scenario 4 was “to resolve the pending liquidity crisis” at Live Well. Tr. 1169-70. Like Scenario 14, the prices were gradually increased over the course of several weeks to avoid setting off “red flags.” Tr. 1169-75; GX408.
In February of 2017, Wedbush indicated that it may not renew its loans when they expired in March, which created another substantial threat: because Wedbush offered the most favorable rates, it would “create a really negative liquidity situation,” in Rohr's words, if Wedbush stopped financing the loans. Tr. 1183-84. Rohr testified that Hild continued to direct price increases at that point in an effort to avoid “potentially business-ending, company-ending situations,” but that Hild's position was ultimately untenable, as there was “no way out” of the situation. Tr. 1185. Live Well continued to “adjust pricing in IDC” in response to further liquidity events in March 2017, Tr. 1198-1200, increasing the value of the portfolio by another $21 million, Tr. 1202-04.
After the liquidity crisis eased, these prices were gradually returned to the prices derived using Scenario 14. Tr. 1196-97, 1205. Soon after, in September of 2017, Live Well received a subpoena from the Securities and Exchange Commission (“SEC”), which led to a two-year investigation. Tr. 1205.
E. Financial Statements
In addition to the evidence regarding the IDC prices and loans, the Government also presented evidence regarding Live Well's financial statements. Because the bond portfolio was Live Well's largest asset, the application of Scenario 14 prices to that asset resulted in a valuation of the bonds in Live Well's financial statements that was higher than the market rate. Tr. 108189, 1601-11. Lenders testified that they relied, in part, on the financial statements in making decisions about lending to Live Well. Tr. 591, 594-98, 682, 685-93.
Rohr testified that, as CFO, he was involved in preparing the company's financial statements, Tr. 1080, and was directed by Hild to value the portfolio using IDC prices to avoid any discrepancies, Tr. 1083. It was Rohr's view that the characterization of the bonds in the financial statements was not true, both because the prices were above what the bonds could readily be sold for, and because the prices did not come from an independent third party. Tr. 1087-89.
In 2019, Rohr was replaced as CFO by Glen Haddock, Tr. 1598, who worked at Live Well but was not involved in the fraud, Tr. 1601-02. Haddock eventually learned that Live Well was providing IDC prices, Tr. 1608-09, and after evaluating the portfolio on his own, expressed his view to Hild that the bonds were overvalued, Tr. 1610-11. Ultimately, he refused to sign the company's 2019 audited financial statements, Tr. 1616-17, and shortly thereafter Live Well announced it would cease operations and unwind, see Tr. 1617-19.
II. Procedural History
Hild was charged in a five-count indictment with conspiracy to commit securities fraud, 18 U.S.C. § 371; conspiracy to commit wire and brank fraud, 18 U.S.C. § 1349; securities fraud, 15 U.S.C. §§ 78j and 78ff; 17 C.F.R. § 240.10b-5; wire fraud, 18 U.S.C. §§ 1343 and 2; and bank fraud, 18 U.S.C. §§ 1344 and 2. Hild maintained his innocence and proceeded to trial, which lasted fourteen days. He was convicted by the jury on all five counts.
Following trial, Hild moved for a judgment of acquittal pursuant to Federal Rule of Criminal Procedure 29, and, in the alternative, for a new trial pursuant to Federal Rule of Criminal Procedure 33. Following oral argument-and a plethora of supplemental submissions-the Court issued a 64-page opinion and order denying the motions in their entirety. See United States v. Hild, __ F.Supp.3d __, 2022 WL 17484992 (S.D.N.Y. Dec. 7, 2022). Hild was sentenced on January 27, 2023 to 44 months' imprisonment. He has since appealed to the Second Circuit. This Court granted bail pending appeal after concluding that his appeal met the factors for release pursuant to 18 U.S.C. § 3143. Dkt. 154. Although the Court underscored its continued assurance as to the merits of its opinion denying Hild's post-trial motions, it concluded that his appeal “presents a substantial question” that was “so integral to the merits of the conviction” that any possible contrary holding by the Second Circuit would “likely require reversal of the conviction or a new trial.” See id. at 2-3 (citing United States v. Randell, 761 F.2d 122, 124-25 (2d Cir. 1985)).
Following Hild's sentencing, the Government filed a proposed order of restitution on April 24, 2023, resulting in a flurry of letters from both parties regarding the appropriateness of the Government relying on the same affidavits it had previously submitted to the Court while affirming that they were “not sufficiently detailed to determine which expenses the victims have incurred that are compensable” for restitution purposes. Dkt. 144 at 19. The Court filed an order indicating that, although it had ordered Hild to pay restitution to the victims, it would not specify the amount “without additional justification for doing to,” further setting a briefing schedule on the issue. Dkt. 164. The Government made an additional submission on May 16, 2023, Dkt. 168, and Hild thereafter filed his motion for a new trial on May 29, 2023, Dkts. 173-74.The motion was fully briefed in July, and the Court heard oral argument on August 9th. Based upon the Court's questioning at oral argument, the parties filed supplemental letters on discrete issues related to the restitution calculation on August 18th.
Although the Government initially moved to seal the records showing the amount in coupon payments received by the victim lending institutions, see Dkt. 168 at 2, it has since withdrawn this motion given the “passage of time since the events at issue and the fact that certain information has already been made public through the Government's in-court statements and other filings,” Dkt. 188. Accordingly, the Government has now filed the records on the public docket. See Dkt. 187, Ex. 1 (Flagstar materials), Ex. 2 (Mirae materials), Ex. 3 (ICBC materials), Ex. 4 (Customers materials).
LEGAL STANDARD
Federal Rule of Criminal Procedure 33 provides that, “[u]pon the defendant's motion, the court may vacate any judgment and grant a new trial if the interest of justice so requires.” Fed. R. Crim. P. 33(a). “The defendant bears the burden of proving that he is entitled to a new trial under Rule 33.” United States v. McCourty, 562 F.3d 458, 475 (2d Cir. 2009). “[T]he standard for granting such a motion is strict,” United States v. Gambino, 59 F.3d 353, 364 (2d Cir. 1995), and it should only be granted in “extraordinary circumstances,” McCourty, 562 F.3d at 475. See also United States v. Zagari, 111 F.3d 307, 322 (2d Cir. 1997); United States v. Locascio, 6 F.3d 924, 949 (2d Cir. 1993). The “ultimate test” in evaluating a Rule 33 motion is whether “letting a guilty verdict stand would be a manifest injustice.” United States v. Canova, 412 F.3d 331, 349 (2d Cir. 2005). There “must be a real concern that an innocent person may have been convicted.” Id.
Rule 37 permits a district court, while an appeal is pending, to “(1) defer considering [a Rule 33] motion; (2) deny the motion; or (3) state either that it would grant the motion if the court of appeals remands for that purpose or that the motion raises a substantial issue.” Fed. R. Crim. P. 37(a). Put differently, a “district court retains jurisdiction to deny a Rule 33 motion during the pendency of an appeal, even though it may not grant such motion unless the Court of Appeals first remands the case to the district court.” United States v. Camacho, 302 F.3d 35, 36 (2d Cir. 2002).
DISCUSSION
I. Documentation of Lender Coupon Payments is Not “Newly Discovered Evidence”
In his latest motion, Hild argues that the supplemental documents related to coupon payments that the Government obtained from lenders-and thereafter submitted to the Court in connection with the ongoing restitution calculation-constitute “newly discovered evidence” within the meaning of Rule 33. See Mot. at 18. He argues that these materials confirm that the lenders who held the HECM IO bonds at issue regularly received tens of millions of dollars in coupon payments. This fact, Hild argues in turn, undermines the Government's principal arguments at trial that the bonds were overvalued by Live Well, corroborates Hild's arguments at trial that the bonds were properly valued and the lenders fully secured, and impeaches the testimony of lender representatives who told the jury that they would not hold the bonds and any coupon payment proceeds would not cover the loans made.
Hild's current Rule 33 motion appeared initially to further argue that documentation of the lenders' coupon payments also constituted undisclosed Brady material. See Mot. at 3 (“Whether viewed as newly discovered evidence under Rule 33 or a belated disclosure pursuant to Brady v. Maryland, the result should be the same: a new trial on all counts.”) In the Government's response, however, it represented that it was “not previously in possession of this information.” Opp. at 4; see also id. (“[A]nd to the extent the Government obtained information regarding coupon payments beyond what was previously produced in discovery or elicited at trial, that information was included in the Government's court filings.”). In light of the Government's representation, Hild does not appear to continue to argue that evidence of coupon payments constituted undisclosed Brady material. Instead, he now asks merely that the Court order the Government “to disclose, pursuant to Brady, all information relating to the lenders' receipt of coupon payments, together with any other information relating to the holding and sale of the bonds and offsets for losses.” Reply at 15. That request is discussed, infra in Section II.
Although this latest argument appears compelling at first blush, for the reasons which follow, the Court cannot conclude that the documents pertaining to coupon payments received by the lenders constitute newly discovered evidence under Rule 33's stringent standard. Accordingly, Hild's latest motion for a new trial is denied.
* * *
A motion for a new trial pursuant to Rule 33 based on newly discovered evidence should be granted only when “(1) the evidence [was] newly discovered after trial; (2) facts are alleged from which the court can infer due diligence on the part of the movant to obtain the evidence; (3) the evidence is material; (4) the evidence is not merely cumulative or impeaching; and (5) the evidence would likely result in an acquittal.” United States v. Owen, 500 F.3d 83, 87-88 (2d Cir. 2007). This is a particularly “strict” standard, Gambino, 59 F.3d at 364, and a motion for a new trial is thus only to be granted in “extraordinary circumstances,” McCourty, 562 F.3d at 475. “The balance between protecting the finality of judgments and the interests of justice is inherent in the Rule 33 analysis,” United States v. Forbes, 790 F.3d 403, 408 (2d Cir. 2015), and the Circuit has “never before held that a new trial may be granted pursuant to Rule 33 on the basis of evidence that was known by the defendant prior to trial, but became newly available after trial,” Owen, 500 F.3d at 89. As noted previously, the “ultimate test” is whether “letting a guilty verdict stand would be a manifest injustice.” Canova, 412 F.3d at 349.
Most significantly, Hild has not established that any documentation related to the lenders' coupon payments for holding the HECM IO bonds is material under the third prong, or that it would have likely resulted in his acquittal on the fraud counts under the fifth. It is well-established that actual harm is not an element of any of the offenses for which Hild was convicted. See, e.g., United States v. Abakporo, 2014 WL 2608857, at *2 (S.D.N.Y. June 9, 2014) (“As a matter of law, the actual loss amounts cannot be ‘material to the issue of guilt' and thus, cannot be the basis for granting a new trial or acquittal under Rule 33.”). It was therefore irrelevant to Hild's guilt, as a matter of law, whether any of the lenders lost any money-or even actually profited from-his fraudulent scheme. Rather, as the Court recognized in its opinion denying Hild's previous posttrial motions, the proof at trial demonstrated “that Hild deceived the lenders by negotiating loan agreements under which the valuation of Live Well's bond portfolio was based on purported market prices set by an independent third party, but which was, in reality, based on Hild's company's internal, above-market pricing assumptions.” Hild, 2022 WL 17484992, at *10 (S.D.N.Y. Dec. 7, 2022). The question for the jury's purposes, as the Government argued throughout trial, was thus simply whether Hild knowingly caused the bonds to be marked at prices that were not obtainable in the market in order to receive inflated loans from the lenders- something that was well supported by the evidence adduced at trial. See, e.g., GX402 (Hild describing the scheme as a “self-generating money machine”); Tr. 1002 (Hild discussing with coconspirators how to best avoid detection in light of an inquiry from one of the lenders); Tr. 11921 (Hild directing co-conspirator to apply the inflated Scenario 14 prices slowly, by way of a “kind of gradual glide path,” so that Live Well did not appear to have its bond portfolio increased by “11.4 million [dollars] in one day”); Tr. 1652 (describing the process by which Live Well purchased the bonds and “immediately” marked them up 20%); Tr. 207 (showing Live Well maintained records demonstrating the differences between market prices and Scenario 14 prices).
To be sure, although the Government was not required to prove actual harm, it was required to prove Hild's intent to defraud the lenders by way of contemplated harm. See United States v. Jabar, 19 F.4th 66, 81 (2d Cir. 2021); see also United States v. Dinome, 86 F.3d 277, 283 (2d Cir. 1996) (“With respect to [wire fraud], the Government need not prove that the scheme successfully defrauded the intended victim .... However, the Government must show that some actual harm or injury was contemplated by the schemer.”) (cleaned up). Hild thus argues that documentation related to sizeable coupon payments to the lenders is material because it demonstrates, as he argued at trial, that he believed the Scenario 14 pricing model to be legitimate, and that he therefore did not contemplate any harm resulting to the lenders. But contemplated harm under the fraud statute is “apparent where there exists a discrepancy ‘between benefits reasonably anticipated because of the misleading representations and the actual benefits which the defendant delivered, or intended to deliver.'” United States v. Novak, 443 F.3d 150, 159 (2d Cir. 2006) (quoting United States v. Starr, 816 F.2d 94, 98 (2d Cir. 1987)). Given the demonstrated, and amply proven, distinction between what Hild represented to be the market value of the bonds and the actual market value of those same bonds, see Hild, 2022 WL 17484992, at *8-14-and the fact that Hild knew that there was a discrepancy between the two-his intent to offer the bonds as collateral for loans at inflated market valuations was sufficient evidence of contemplated harm to establish his guilt. Proof that the lenders eventually recouped some-or, indeed, even all-of the collateral initially represented by Live Well via coupon payments is thus immaterial to the harm that Hild contemplated by engaging in the scheme, and would not more likely than not have resulted in his acquittal.
See GX407 (“STUMBERGER: I don't think Scott [at Wedbush] cares about that. Scott cares about what is the market . . . May not be right, or proper, or sound, but that's what's gotta be defined as the market. That's just realistic. HILD: There's no debating that.”); id., Tr. 169-75 (Hild and co-conspirators discussing whether they could find a “slimy” dealer willing to represent to a lender that the Scenario 14 prices were correct); see also GX416, Tr. 1040-41 (Hild directing co-conspirators to buy tranches of bonds so that lenders would not see the prices of individual bonds and discover that they did not match the prices from comparable bonds in a tranche).
So too, it is at the very least unclear whether Hild has established that he “could not with due diligence” have discovered documentation related to the coupon payments to the lenders “before or during trial.” United States v. Alessi, 638 F.2d 466, 479 (2d Cir. 1980). As the Government rightly observes, Hild could have issued Rule 17(c) subpoenas to obtain details regarding the coupon payments the lenders had received prior to (or at the time of) trial, or even could have obtained some details related to the coupon payments from Bloomberg or other providers of publicly available financial data. Indeed, Hild was certainly aware of the fact of coupon payments being received by the lenders, as demonstrated by his trial counsel's cross-examination of several witnesses regarding the payments. In one such exchange, Hild's trial counsel questioned a representative of Mirae, one of the lending institutions as follows:
Q: And so if I understand correctly, to this day Mirae's holding some of these bonds?
A: Yes.
Q: And presumably Mirae is-and has received the revenue streams generated by the bonds?
A: Yes.
Q: And Mirae would continue to receive those revenue streams generated by the bonds until they no longer generate revenue or they're sold?
A: Correct.
Q: And we have no idea how much that is going to be?
A: That is correct.
Q: And it very well could in fact be more than the amount of the loan to Live Well by Mirae?
MS. ESTES: Objection.
THE COURT: Overruled.
A: I don't know that.Tr. 640. Exchanges with other lender representatives on cross examination elicited much the same. See id. at 711-12 (cross of ICBC lender representative, “Q: . . . When you said before that your only option was to sell the collateral, that's not true insofar as you just said you could hold it right? A: That is an option, but I can tell you that that is not what we'd ever do in the market ....”); id. at 717 (continued cross of same ICBC lender representative, “Q: Okay. And I think we've established an option to hold the collateral and collect the revenue stream, yes? A: If that was your desire, sure. Obviously you got to get-you got to mark up that collateral. Obviously you can't just take collateral at any price. And then you can hold it as an investment.”); id. at 795 (cross of Flagstar lender representative, “Q: . . . [T]hese are bonds that generate revenue, yes? A: Correct.... Q: Okay. And we can agree that Live Well could have decided to-excuse me- Flagstar could have decided to hold the bonds? A: We could have. Q: And if it had done that, it would have received the income streams connected thereto, yes? A: Correct.”).
If anything, counsel's decision not to obtain more precise information related to the payments by way of this consistent line of cross examination evinces a strategic choice made in furtherance of Hild's defense to the jury. That is, counsel strategically elected to develop the record regarding the lenders' ability to recoup ambiguous significant sums-even potentially “more than the amount of the loan to Live Well,” See id. at 640-via coupon payments without seeking specific information regarding those payments, thereby bolstering Hild's claim that he believed the Scenario 14 pricing model to provide legitimate valuations of the bonds. As the Second Circuit has observed, “[o]ne does not ‘discover' evidence after trial that one was aware of prior to trial.” Owen, 500 F.3d at 89-90. Although the Court need not rely on the due diligence prong in denying the instant Rule 33 motion, it is therefore highly skeptical that Hild, having made the strategic decision to leave the jury with the vague impression that the coupon payments could be sizeable (without pursuing financial specifics which could have risked refuting the point), has now alleged facts “from which the court can infer due diligence on the part of the movant to obtain the evidence.” Id. at 88.
Lastly, the evidence related to the coupon payments is likely “merely cumulative or impeaching.” Id. Although Hild argues that the coupon payment information “demonstrates that the witnesses who represented the lenders gave misleading testimony about their institutions' unwillingness to hold the bonds and the lack of revenues the bonds were generating,” Mot. at 22, the law is clear that “merely . . . impeaching” evidence does not provide a basis to grant a new trial pursuant to Rule 33, Owen, 500 F.3d at 88. And it was hardly a secret at trial that coupon payments to the tune of millions of dollars were generated by the HECM IO bonds. The Government introduced financial statements reflecting the millions of dollars that Live Well was receiving on a monthly basis due to coupon payments for the same bonds it then offered to the lenders as collateral, for instance, and questioned Hild's co-conspirators about the same. See GX324 (documenting $48 million in “interest income” for FY 2018); Tr. 1083 (Rohr explaining that the “interest income” portion of Live Well's financial statements reflected “the coupon we receive on the bonds”). Moreover, as observed above, Hild's trial counsel cross-examined each of the lender witnesses, repeatedly and effectively, regarding the fact that their losses could, and likely would, be mitigated by holding the HECM IO bonds and collecting coupon payments over time. See Tr. 639-40 (cross of Maisano); id. at 711-12, 717 (cross of Levy); id. at 795 (cross of Redoutey); see also United States v. Zagari, 111 F.3d 307, 322 (2d Cir. 1997) (underscoring that the purported newly discovered evidence must be “so material and noncumulative that its admission would probably lead to an acquittal” for a Rule 33 motion to be granted) (cleaned up).
Accordingly, because Hild fails to establish that each of the five prongs for newly discovered evidence are met by the documentation regarding the lenders' coupon payments, the Rule 33 motion is denied in its entirety.
At oral argument, Hild requested that the Court hold his Rule 33 motion in abeyance until the issue of the calculation of restitution is resolved, anticipating, in effect, that any additional discovery related to lender coupon payments could “further support [the] motion and provide additional bases supporting the arguments [Hild] make[s].” Oral Arg. Tr., Aug. 9, 2023, at 5:3-11. The Court declines to do so, and instead finds prompt resolution of the motion appropriate given Hild's ongoing appeal and the likelihood that the Second Circuit would consolidate its review of this action to include any appeal of the instant motion. Moreover, for the same reasons that the Court determined that the present documentation related to lender coupon payments does not meet the standard for newly discovered evidence requiring a new trial, see supra at 17-21, the Court is highly skeptical that additional material related to those payments would satisfy Rule 33's stringent demands. Even were new materials to show that the lenders eventually profited handsomely by holding the HECM IO bonds, for instance, such evidence would not be material to Hild's fraud convictions. See, e.g., Abakporo, 2014 WL 2608857, at *2 (“As a matter of law, the actual loss amounts cannot be ‘material to the issue of guilt' and thus, cannot be the basis for granting a new trial or acquittal under Rule 33.”). In the event that Hild believes later-obtained material does constitute newly discovered evidence, he is, of course, at liberty to seek appropriate relief at that time.
II. Request for Discovery Regarding Coupon Payments for Restitution Calculation
In the alternative, Hild requests that the Court permit the Defense additional discovery related to coupon payments, so that an appropriate restitution calculation may be made. While acknowledging that the Court is only tasked with arriving at a “reasonable estimate” of actual losses suffered by the victims in ordering a restitution amount, United States v. Milstein, 481 F.3d 132, 137 (2d Cir. 2007), Hild argues that, with the materials thus far submitted by the Government, and given the repeated corrections-to the tune of millions of dollars, no less, see, e.g. Gov't Letter, Dkt. 185 at 2-3-the Court cannot be assured that any current restitution estimate would necessarily be “reasonable” in nature. Hild therefore requests that the Court broadly direct the Government to make any additional disclosures related to “the amount of coupon payments every lender has received to date.” Reply at 16. Among other things, he seeks discovery “regarding the transaction between [victim lender] ICBC and BMO, to determine whether ICBC received any additional benefits beyond the sale prices, and whether ICBC is continuing to receive benefits as the bond coupon payments continue to BMO.” Mot. at 16. He insists that, if either of those inquiries were answered in the affirmative, for instance, any such financial benefits would need to be deducted from ICBC's actual loss in order to ensure that the institution were not to obtain restitution “in excess of . . . its loss.” United States v. Gonzalez, 647 F.3d 41, 67 (2d Cir. 2011).
At oral argument, when pressed to clarify the request being made, counsel indicated: “we want the amount of coupon payments every lender has received to date and an explanation for why these materials were withheld. And I would add to that that we would like any information about other benefits, such as settlement payments, that would also offset the restitution amounts.” Oral Arg. Tr., Aug. 9, 2023, at 33:9-15. In its latest letter response, the Government has indicated that it “has requested copies of the settlement documents from the relevant victim entities,” and that any such documents will be provided in order to assist in arriving at an accurate assessment of Hild's restitution obligation. Dkt. 188 at 3.
Given the parties' stated preference during oral argument, see Oral Arg. Tr., Aug. 9, 2023, at 3:16-19, and in the interest of resolving the issue of restitution expeditiously, by separate order, the Court will make a referral to Magistrate Judge Parker to make an appropriate calculation of the restitution to be paid to each of the lending institutions. See 18 U.S.C. § 3664(d)(6) (providing that a district court “may refer any issue arising in connection with a proposed order of restitution to a magistrate judge or special master for proposed findings of fact and recommendations as to disposition”). Any determination regarding whether additional discovery is necessary to facilitate that calculation will appropriately be before Judge Parker.
III. Whether the Fraud Instructions Given Constitute Plain Error in Light of Ciminelli is Appropriately Before the Second Circuit
Lastly, Hild argues that the Supreme Court's recent decision in Ciminelli v. United States, 143 S.Ct. 1121 (2023), further underscores “the need for a new trial.” Mot. at 26-27. In that case, the Supreme Court considered whether the Second Circuit's so-called “right-to-control theory of wire fraud”-“under which the government c[ould] establish wire fraud by showing that the defendant schemed to deprive a victim of potentially valuable economic information necessary to make discretionary economic decisions”-was a “valid basis for liability under 18 U.S.C. § 1343.” Id. at 1125-26. Reversing the Second Circuit, Ciminelli invalidated the theory, holding that the “right-to-control theory cannot form the basis for a conviction under the federal fraud statutes.” Id. at 1128. It reasoned that the “theory cannot be squared with the text of the federal fraud statutes, which are ‘limited in scope to the protection of property rights,'” id. at 1127 (quoting McNally v. United States, 483 U.S. 350, 360 (1987)), and that Second Circuit had never cited authority “that established ‘potentially valuable economic information' as a traditionally recognized property interest,” id. at 1127-28.
Hild is certainly correct that the Court instructed the jury as to the now-invalidated right-to-control theory, given then-existing Second Circuit law and without objection from either party. Specifically, it instructed, as to Count Four (wire fraud),
that the government must . . . prove that the alleged scheme contemplated depriving another of money or property .... In this regard, a person is not deprived of money or property only when someone directly takes his money or property from him. Rather, a person is also deprived of money or property when that person is provided false or fraudulent information that, if believed, would prevent him from being able to make informed decisions about what to do with his money or property. In other words, a person is deprived of money or property when he is deprived of the right to control that money or property. And he is deprived of the right to control that money and property when he receives false or fraudulent statements that affect his ability to make discretionary economic decisions about what to do with that money or property.Tr., Dkt. 70-2 at 34-35 (emphasis added). The Court's instruction with respect to Count Two (conspiracy to commit wire fraud and bank fraud) was substantially identical, and also relied upon the theory. See Tr., Dkt. 70-4 at 1. Indeed, the jury instructions from Mr. Hild's trial were cited in an amicus brief filed in Ciminelli as an example of a recent conviction the Government had secured by using the right-to-control jury instruction. See Br. of Amicus Curiae New York Council of Defense Lawyers in Supp. of Petititioner at 5a, Ciminelli v. United States, No. 21-1170 (filed Sept. 6, 2022), available at https://rb.gy/rhwdx.
The critical issue for Hild is therefore whether these jury instructions now constitute plain error in light of Ciminelli warranting vacatur of his conviction. See United States v. Marcus, 560 U.S. 258, 262 (2010) (providing that plain error review is a stringent standard, requiring an appellant to demonstrate that “(1) there is an error; (2) the error is clear or obvious, rather than subject to reasonable dispute; (3) the error affected the appellant's substantial rights, which in the ordinary case means it affected the outcome of the district court proceedings; and (4) the error seriously affects the fairness, integrity or public reputation of judicial proceedings”) (cleaned up). Central to such review is consideration of whether any error was nevertheless harmless-that is, whether it is “clear beyond a reasonable doubt that a rational jury would have found the defendant guilty absent the error.” United States v. Sheehan, 838 F.3d 109, 121 (2d Cir. 2016) (quoting Neder v. United States, 527 U.S. 1, 18 (1999)). While acknowledging that he “intends to raise this issue in the Second Circuit in his pending appeal,” Mot. at 27 n.5, and agreeing that Ciminelli “does not provide an independent basis for this Court to grant a motion under Rules 33 and 37,” Reply at 16-17, Hild nevertheless argues that, “[t]o the extent this Court provides an indicative ruling that it would grant Mr. Hild's Rule 33 motion on all counts based on newly discovered evidence, [it] may at the same time acknowledge that a new trial is required based on Ciminelli as well,” id. at 17.
The Court declines to reach the issue here. Although Ciminelli's holding appears to be at odds with certain instructions given to the jury at Hild's trial, given his ongoing appeal, whether those instructions constitute plain error, and whether any error was nevertheless harmless, are questions now appropriately before the Second Circuit.
CONCLUSION
Accordingly, for the foregoing reasons, Hild's Rule 33 motion for a new trial is denied. By separate order, the Court will refer the restitution calculation to Magistrate Judge Parker for a recommendation as to a disposition. The Court defers the applicability of Ciminelli to the jury instructions here to the ongoing appeal before the Second Circuit. The Clerk of Court is respectfully directed to terminate the motion pending at Dkt. 173.
SO ORDERED.