UNITED STATES DISTRICT COURT EASTERN DISTRICT OF NEW YORK
June 14, 2011
UNITED STATES OF AMERICA,
SANDRA HATFIELD AND DAVID H. BROOKS, DEFENDANTS.
The opinion of the court was delivered by: Seybert, District Judge:
MEMORANDUM AND ORDER
Pending before the Court is the Government's motion to enter a preliminary order of forfeiture against Defendants David
H. Brooks and Sandra Hatfield. For the following reasons, the Court GRANTS this motion IN PART and DENIES this motion IN PART. However, the Court does not issue any specific forfeiture award at this time. Instead, it RESERVES JUDGMENT IN PART and hereby requests supplemental calculations and briefing consistent with this Order.
In October 2007, the Government charged Mr. Brooks and Ms. Hatfield with numerous crimes, including securities fraud, mail fraud, and wire fraud. In connection with this prosecution, the Court restrained numerous bank accounts as potentially subject to criminal forfeiture. The Court also issued seizure warrants directed at various pieces of personal property, including three automobiles, 6,007,099 shares of Point Blank Solutions, Inc., gold watches, designer pens, and a jewel-encrusted American flag belt buckle.
On September 14, 2010, the jury convicted Mr. Brooks on counts 1-11 and 15-17 of the Indictment, including all counts that alleged securities fraud, mail fraud, wire fraud, and/or conspiracy to commit those crimes. The jury convicted Ms. Hatfield on counts 1-3 and 12-16 of the Indictment, but acquitted her on counts 4-5. In so doing, the jury convicted Ms. Hatfield on all counts that alleged securities fraud, and conspiracy to commit securities, mail and wire fraud. But it acquitted her on the direct mail and wire fraud counts.
The parties agreed to waive their rights to a jury trial on forfeiture. So, in November 2010, the Court conducted protracted non-jury forfeiture proceedings. It then permitted the parties to submit lengthy post-hearing briefs, with reply submissions not coming in until April 14, 2011.
DISCUSSION I. Standard of Review
The Government seeks forfeiture under 18 U.S.C. § 981(a)(1)(C), which authorizes forfeiture of "[a]ny property . . . which constitutes or is derived from proceeds traceable to" many kinds of offenses, including securities fraud. 18 U.S.C. § 981(a)(1)(C). "Because criminal forfeiture is viewed as part of the sentencing process, the government need prove facts supporting forfeiture only by a preponderance of the evidence." United States v. Gaskin, 364 F.3d 438, 461 (2d Cir. 2004) (internal citations omitted).
"The calculation of forfeiture amounts is not an exact science." United States v. Treacy, 639 F.3d 32, 48 (2d Cir. 2011). Thus, "[t]he court need not establish the loss with precision but rather need only make a reasonable estimate of the loss, given the available information." Id. (internal citations and quotations omitted). And, consequently, the Court "is permitted to use general points of reference as a starting point for calculating the losses or gains from fraudulent transactions and may make reasonable extrapolations from the evidence established by a preponderance of the evidence at the sentencing proceeding." Id.
II. The Court's April 21, 2010 Order
The Government first argues that the Defendants must forfeit every penny of revenue stemming from their insider stock sales, and that the Court's April 21, 2010 decision erred in holding otherwise. In that decision, the Court held that only the "difference between the stock's inflated value, and what it would have sold for absent the fraud" is subject to forfeiture.
2010 WL 1685826 at *3, 2010 U.S. Dist. LEXIS 39618 at *10. For the following reasons, the Court's April 21, 2010 Order stands and the Government's argument is rejected.
A. The Government's Previous Reconsideration Motion
On June 30, 2010, the Government sought reconsideration of the April 21 Order. On August 3, 2010, the Court denied the Government's motion by Electronic Order. The Court promised to explain its reasoning in a written opinion, which it intended to complete shortly after the August 3 Electronic Order. Regretfully, the Court's demanding docket (including protracted proceedings in this case) prevented it from finishing this work. Nevertheless, to keep its promise, the Court explains its decision now.
i. Standard of Review
Although the federal and local rules of criminal procedure do not specifically provide for motions for reconsideration, courts in this Circuit have applied Local Civil Rule 6.3 in criminal cases. See, e.g., United States v. Aleynikov, 10-CR-96, 2011 WL 939754, at *10 (S.D.N.Y. Mar. 16, 2011); United States v. Yannotti, 457 F. Supp. 2d 385, 388-89 (S.D.N.Y. 2006).
Under this standard, a motion for reconsideration will generally be denied unless "the moving party can point to controlling decisions or data that the court overlooked - matters, in other words, that might reasonably be expected to alter the conclusion reached by the court." In re BDC 556 LLP, 330 F.3d 111, 123 (2d Cir. 2003) (citation omitted); Aleynikov, 2011 WL 939754 at *10. "Local Rule 6.3 is narrowly construed and strictly applied so as to avoid repetitive arguments on issues that have been considered fully by the court." Yannotti, 457 F. Supp. 2d at 389 (internal citations and quotations omitted).
ii. The Government's Statutory Argument
The Government's reconsideration motion principally argued that the Court "overlooked a number of cases that implicitly" favor its position. See Docket No. 1144 at 3. But, overwhelmingly, the Government cites to district courts, other circuits, and Second Circuit summary orders. See generally id. at 3-4. Thus, even if the Court agreed with the Government that these cases support its argument (and it does not), these authorities cannot justify reconsideration. For, as noted above, the moving party must point to "controlling decisions," to obtain reconsideration. In re BDC 556 LLP, 330 F.3d at 123. Decisions of other district court judges and sister circuits are not "controlling" authority. Similarly, the Second Circuit's own rules declare that its summary orders "do not have precedential effect." Second Circuit Local Rule 32.1.1(a).
The Government does cite a few published Second Circuit opinions. But these cases are either inapposite, or actually strengthen the Court's own reasoning. For the sake of completeness, the Court discusses each of these cited authorities below:
a. United States v. Awad, 598 F.3d 76 (2d Cir. 2010), applies forfeiture law in a narcotics case. It does not address how to calculate forfeiture in the context of insider stock sales. And it stands only for the non-controversial proposition that a defendant is subject to forfeiture "irrespective of his assets at the time of sentencing." Id. at 79. Nothing in the Court's April 21 Order held otherwise. b. United States v. Zvi, 168 F.3d 49, 56 (2d Cir. 1999), is equally inapposite. Indeed, it references forfeiture only insofar as it reverses a forfeiture judgment because the underlying money laundering counts were time barred. And, though the Government cites it for how it defines "proceeds" under the wire fraud statute, that definition actually supports the Court's April 21 ruling. Specifically, the Second Circuit concluded that monies obtained through gold sales that preceded the filing of a false insurance claim were the scheme's "proceeds." The Second Circuit reasoned that, because the defendants used the insurance money to "pay creditors and consignors," the "pre-robbery sale of gold represented the defendants' take from the scheme." Id. In effect then, the Second Circuit defined "proceeds" as the scheme's fraudulent profits. So to here: Defendants' "take" from their underlying securities frauds was the amount that these frauds inflated DHB's stock price -- not DHB's entire market value.
c. United States v. Lizza Industries, Inc., 775 F.2d 492, 498 (2d Cir. 1985), also augments the Court's April 21 holding. In this case, the Second Circuit held that RICO forfeiture depends upon "gross, rather than net profits." In so holding, the Second Circuit recognized that this method of calculation "leaves open a possibility that defendants will be forfeiting profits that they would have made outside of their criminal activities." Id. But, the Second Circuit held, "This should not cause us to scuttle the method of computing forfeiture" because forfeiture "is a punitive, not a restitutive, measure" and "does not require the prosecution to prove or the trial court to resolve complex computations, so as to ensure that a convicted racketeer is not deprived of a single farthing more than his criminal acts produced." Id.
Contrary to the Government's thinking, Lizza Industries does not endorse seizing a convicted Defendant's property willy-nilly, without proving that the Defendant obtained this property "as the result of the commission of the offense." 18 U.S.C. § 981(a)(2)(A). Instead, it views calculating a defendant's true ill-gotten gains as an ideal, but recognizes that this ideal might not always be obtainable or practical. 775 F.2d at 498 ("[c]onced[ing]" that its holding "leaves open a possibility" of the defendant forfeiting more than he illicitly earned). And it instructs district courts, when faced with imprecise measurements, to risk error in the Government's favor, reflecting forfeiture's punitive nature. Id. (no need to "ensure that a convicted racketeer is not deprived of a single farthing more than his criminal acts produced").
Consistent with Lizza Industries, the Court will not worry if this Order over-calculates Defendants' forfeiture liability by farthings. But the Court sees nothing in Lizza Industries to justify awarding the Government every farthing from the Defendants' stock sales, especially if most of that cash reflected the stock's bona fide value, not fraudulent appreciation "result[ing]" from "the commission of the offense."
18 U.S.C. § 981(a)(2)(A). iii. The Government's Eighth Amendment Argument
The Government also argues that the Court's April 21 opinion erred by conducting a "premature" Eighth Amendment analysis. The Government is wrong. The Court considered the Eighth Amendment as part of its statutory interpretation efforts. See 2010 WL 1685826 at *4; 2010 U.S. Dist. LEXIS 39618 at *11-12. This was not "premature"; it was required. For, when faced with dueling statutory interpretations, courts should construe a statute to avoid serious constitutional problems. Id. (citing U.S. v. Colavito, 19 F.3d 69, 71 (2d Cir. 1994)); Edward J. DeBartolo Corp. v. Florida Gulf Coast Bldg. and Const. Trades Council, 485 U.S. 568, 575, 108 S. Ct. 1392, 99 L. Ed. 2d 645 (1988) ("where an otherwise acceptable construction of a statute would raise serious constitutional problems, the Court will construe the statute to avoid such problems unless such construction is plainly contrary to the intent of Congress").
In any event, the Court's Eighth Amendment analysis was not dispositive. The Court found the Government's proposed interpretation "untenable" not just on constitutional grounds, but also as a matter of "statutory construction." 2010 WL 1685826 at *4; 2010 U.S. Dist. LEXIS 39618 at *13. Specifically, the Court noted that the Government's interpretation ignored § 981(a)(2)(A)'s language that limits forfeiture to property acquired "as the result of the commission of the offense."
Thus, the Government failed to meet the high burden necessary to support reconsideration. So the Court denied this motion.
B. United States v. Treacy
Setting aside whether reconsideration was appropriate last August, the Court must still consider whether intervening Second Circuit authority has impacted its analysis. In this regard, both parties point to the Second Circuit's recent decision in Treacy, referenced earlier in this opinion.
In Treacy, the Second Circuit addressed forfeiture in a securities fraud case arising from stock option backdating. The Government sought forfeiture under 18 U.S.C. § 981(a)(1)(C), the same statute it charges in the Indictment here. Compare Treacy, 639 F.3d at 40-41 with April 21 Order, 2010 WL 1685826 at *2; 2010 U.S. Dist. LEXIS 39618 at *6-7. But unlike here, in Treacy the Government did not claim that the defendant's securities fraud required him to forfeit 100% of the cash he generated when he exercised and then sold backdated options. Instead, the Government sought only the net profits the defendant earned because of the fraudulent conduct. The parties dispute, then, concerned the appropriate "measurement date" for valuing the stock options the defendant received. See id. at 41. The Second Circuit held that the district court committed "error" when it chose one measurement date, and remanded "for recalculation of the appropriate forfeiture." Id. at 44.
The Government argues, curiously, that Treacy does not "stand for the proposition that the Government may forfeit only the net profits, not the gross proceeds, of insider trading." Docket No. 1435 at 4. But, in this regard, the Government argues only that "for reasons wholly unrelated to this case, the government [in Treacy] took an extremely conservative approach to the forfeiture calculation" and thus did not ask the district court or the Second Circuit to interpret § 981(a)(1)(C) as requiring the forfeiture of a securities fraud's "gross proceeds." Id. at 5.
The Government's position is, frankly, bizarre. A statute does not mean one thing as to one defendant, and an entirely different thing with respect to a less favored defendant. Central to Due Process is that courts must interpret the same statute "consistently," regardless of what the context is, or who is on trial. Leocal v. Ashcroft, 543 U.S. 1, 12 n. 8, 125 S. Ct. 377, 160 L. Ed. 2d 271 (2004). Indeed, as Judge Richard Posner has opined, "The rule of law means that judges" must "decide cases 'without respect of persons,' that is, without considering the social status, attractiveness, etc. of the parties or their lawyers."*fn1 Here, the Government argued before the Second Circuit that § 981(a)(1)(C) means one thing, and the Second Circuit adopted and applied that interpretation in a published order that binds this Court. The Government may not, necessarily, be estopped from challenging that interpretation in another case before the Second Circuit. See generally Nat'l Fed'n of Fed. Employees, Local 1309 v. Dep't of Int., 526 U.S. 86, 108, 119 S. Ct. 1003, 143 L. Ed. 2d 171 (1999) (administrative agencies not estopped from changing their interpretations of a statute, but affording revised interpretations less deference). But, for now, and in this lower Court, Treacy precludes the kind of indiscriminate seizure the Government desires.
Applying Treacy, the Court concludes that it commands the same standard the Court adopted in its April 21 Order. The only "difference" in Treacy's approach reflects the disparate facts between that case and this one: fraudulent price deflation at acquisition v. fraudulent price inflation at sale. Treacy, as an options backdating case, required the district court to properly calculate what the options' strike price would have been, absent the fraud. This case, on the other hand, arises from illegal insider sales. Thus, it requires the Court to calculate what Defendants' shares would have sold for, absent the fraud. Under both metrics, the Court must distinguish property that the Defendants fraudulently acquired from property that the Defendants would have acquired anyway, legitimately.
Thus, the Court finds that Treacy does not change its April 21 analysis.*fn2 So, as before, the Court finds that only the "difference between the stock's inflated value, and what it would have sold for absent the fraud" is subject to forfeiture. 2010 WL 1685826 at *3; 2010 U.S. Dist. LEXIS 39618 at *10.*fn3
III. What Should Be Included In Forfeiture?
A. Mr. Brooks' Liability for Inflated Inventory
The Indictment*fn4 alleged that DHB overvalued its
Interceptor Vest Inventory during the 2003 and 2004 fiscal years, spanning the calendar years 2003-2005. Indictment ¶ ¶ 25-27. The Indictment alleged that the overvalued inventory inflated earnings from $19.6 million to $26.2 million in 2003, and from $34.4 million to $48.2 million in 2004. Id. at ¶ 27. The Indictment similarly alleged that the overvaluation inflated gross profit margin from 21% to 28% in both fiscal years. Id. at ¶ 27. And the Indictment further alleged that a DHB employee (identified at trial as Travis Brooks) "repeatedly informed" Ms. Hatfield in November 2004 that the inventory was overvalued, but that Ms. Hatfield refused to take action. Id. at ¶ 25.
After considering the Defendants' Rule 29(a) motions, the Court agreed that the Government had sufficiently evidenced that DHB materially inflated the value of its Interceptor Vest Inventory. See 724 F. Supp. 2d at 327-328. However, the Court noted that the Government's opposition to Mr. Brooks' motion "point[ed] to no testimony indicating that Mr. Brooks had knowledge of the allegedly overvalued inventory before his 2004 trades," though substantial evidence linked Ms. Hatfield to this fraud. Id. at 328. Thus, after providing the Government with an opportunity to show cause, the Court acquitted Mr. Brooks "of the insider trading charges to the extent that those charges are predicated on DHB's allegedly overvalued inventory." 2010 WL 2838525 at *2; 2010 U.S. Dist. LEXIS 72297 at *5. The Court's Order did not, however, extend to the Indictment's conspiracy to commit securities fraud count.
On October 18, 2010, in connection with the upcoming forfeiture trial, the Court denied a motion from Mr. Brooks that, among other things, argued that he could not be ordered to forfeit assets obtained through the Interceptor Vest Inventory Scheme. See 2010 WL 4177159; 2010 U.S. Dist. LEXIS 111084. Although denying Mr. Brooks' motion, the Court expressed "concerns" regarding the Government's theory that it could use the Interceptor Vest Inventory Scheme to support forfeiture without proving, by a preponderance of the evidence, that Mr. Brooks knew about this scheme before he traded. 2010 WL 4177159 at *7; 2010 U.S. Dist. LEXIS 111084 at *20-21. Notwithstanding this Order, the Government adduced no evidence at the forfeiture trial that reflects Mr. Brooks' knowledge of the scheme.
The parties disagree about the scope of the Court's prior orders, which the Court acknowledges could have been clearer. Mr. Brooks argues that the Court's prior orders preclude a forfeiture awarded predicated, in part, on the Interceptor Vest Inventory scheme. The Government, conversely, alleges that Mr. Brooks has incurred forfeiture liability for the Interceptor Vest Inventory scheme because he entered into a conspiracy to commit securities fraud, and thus is liable for his co-conspirator's acts.
The Court agrees with the Government. A conspiracy's proceeds are subject to forfeiture. See United States v. Capoccia, 402 Fed. Appx. 639, 641 (2d Cir. 2010). And a defendant incurs forfeiture liability for the "proceeds of the entire conspiracy," including acts that the defendant was not personally responsible for. United States v. Sanchez, 2011 WL 1244216, at *3 (2d Cir. April 5, 2011); United States v. Stathakis, 04-CR-790, 2008 WL 413782, at *11 (E.D.N.Y. 2008). Here, Mr. Brooks was convicted of conspiring with Ms. Hatfield to commit securities fraud, including by conspiring with her to inflate DHB's gross profit margins. Thus, Mr. Brooks incurred forfeiture liability for Ms. Hatfield's acts within that conspiracy, regardless of what he personally might have known. Indeed, forfeiture is particularly appropriate in this case, because Mr. Brooks benefited when Ms. Hatfield's fraudulent overvaluations caused DHB's stock to rise.
B. The R&D Reclassification Scheme
The Indictment, prior to redaction, alleged that DHB fraudulently reclassified $22 million in cost of goods sold expenses as "research and development," thereby improperly inflating its gross profit margin. At trial, the parties agreed that DHB performed the reclassifications. But they disagreed about whether the reclassifications were fraudulent.
Defendants contended that, instead of being fraudulent, the reclassifications largely reflected their good-faith but imperfect efforts to account for R&D that DHB actually performed. In this regard, Defendants cited significant testimony, from the Government's own witnesses, establishing that: (1) DHB lacked internal controls sufficient to track R&D expenses; (2) in light of those non-existent internal controls, Ms. Hatfield instructed DHB's Chief Financial Officer, Dawn Schlegel, to reclassify a percentage of gross sales each month as R&D, with this percentage supposedly reflecting Ms. Hatfield's estimate of R&D that DHB actually performed (later a consistent 3% each month); and (3) DHB did a significant amount of R&D, which it did not otherwise account for in its financial statements because of these weak internal controls. In light of this evidence, the Court concluded that, although the Government had sufficiently evidenced a few fraudulent "acts," they did not provide the Court with sufficient evidence of the "larger fraudulent scheme" alleged in the Indictment. See 2010 WL 2710616, at *3; see also 724 F. Supp. at 326-27. Consequently, the Court struck the paragraph that quantified the fraudulent reclassifications as $22 million, concluding that the Government had not adduced evidence showing that this entire amount was fraudulent.
In its October 18th Order, the Court noted that Professor Harris' proposed testimony "incorporates assumptions regarding the proper R&D amount." 2010 WL 4177159 at *6; 2010 U.S. Dist. LEXIS 111084 at *19-20. The Court then instructed that "During the forfeiture phase, the Court will require the Government to submit evidence that supports those assumptions.
Absent evidence, the Court will not infer that the R&D reclassifications, in their entirety, reflected R&D that was never performed." Id.
The Government has now done so. First, it has highlighted Gov. Ex. 4587, a document introduced at trial but not brought to the Court's attention during either the Rule 29(a) briefing process, or in response to the Court's Order to Show Cause. See Gov. Rep. Br. at 11; 2010 WL 4177159 at *6, n. 8; 2010 U.S. Dist. LEXIS 111084 at *18-19 n. 8. Considering this document together with the other evidence before it, the Court agrees that Ms. Hatfield used the R&D reclassifications to "back-into" a desired gross profit margin. Id.
Second, and more importantly, the Government has now introduced DHB's restated financials into evidence. See Gov. Ex. 5129. These restated financials set forth that DHB overstated R&D expenses by $10.4 million in 2003, and by $8.4 million in 2004. As such, the restated amounts exceed the reclassifications listed in the Indictment, which Professor Harris used to calculate DHB's "true" gross profit margin. (See Tr. 145, 148-49, 230-31). The Court agrees with the Government that DHB's own restated financials serve as persuasive evidence regarding what the Company actually spent on R&D. And, with those restated financials, Professor Harris' calculations actually underestimate the degree that the R&D errors inflated DHB's stock price.
Defendants do not seriously dispute that DHB's restated financials evidence what the Company actually spent on R&D. But, Defendants contend, because DHB's restatement accounted for both "intentional and unintentional" R&D inflation, the restatement did not quantify Defendants' fraud. (Brooks Reply Br. at 9). Thus, Defendants contend, the Government did not truly establish how much Defendants' fraud inflated R&D.
The Court is not persuaded. As to R&D, Defendants engaged in two separate and distinct frauds. First, the record reflects that the Defendants fraudulently reclassified some costs of goods sold expenses as R&D for the purposes of inflating DHB's gross profit margins. To the extent that these reclassifications inflated DHB's share price, they are clearly subject to forfeiture.
Second, although not charged in the Indictment, the trial evidence demonstrated that the Defendants: (1) knew that DHB possessed woefully inadequate internal controls to track R&D expenses; (2) in the absence of those controls, decided to account for R&D by reclassifying cost of goods expenses as R&D; (3) failed to publicly disclose either those inadequate internal controls, or their crude substitute mechanism for recording R&D; and (4) traded while possessing this material non-disclosed information. See 2010 WL 2816326, at *2 n. 2; 2010 WL 4177159, *2; see also October 18 Order, 2010 WL 4177159 at *6; 2010 U.S. Dist. LEXIS 111084 at *18 (holding that Government can obtain forfeiture of proceeds from "uncharged" schemes). And, by enabling DHB's "negligent" reclassification policy, this fraudulent scheme led directly to any "unintentional" R&D overstatement in late-2003 and 2004. See generally 18 U.S.C. § 981(a)(2)(A) (defining proceeds as "property of any kind obtained directly or indirectly, as the result of the commission of the offense") (emphasis supplied). Indeed, it is inconceivable that DHB could have continued, in late-2004, to recklessly "account" for R&D if it had publicly announced, more than a year before, that: "The Company has barely any internal controls, much less a well-functioning accounting system. So, rather than disclose R&D expenses as are actually incurred, the Company instead reports whatever number its chief executives feel like, based on their subjective 'guess' as to how much R&D the Company performed." Such a disclosure would, assuredly, have caused pressures (i.e., lawsuits, SEC investigations, proxy challenges) that would, in turn, have forced DHB to adopt better business practices.
In any event, DHB's restatement recorded actual 2003 R&D expenses of $0.4 million, indicating that DHB originally overstated these expenses by $10.4 million, and recorded actual 2004 R&D expenses of $1.3 million, indicating that it originally overstated these expenses by $8.4 million. But Professor Harris based his analysis on the substantially lower reclassifications identified in the Indictment: $7 million in 2003 (suggesting actual R&D expenses of $3.8 million) and $6 million in 2003 (suggesting actual R&D expenses of $2.7 million). Thus, Professor Harris' calculations effectively built-in an assumption that the Defendants "negligently," but not fraudulently, miscalculated R&D by 950% in 2003 and by 208% in 2004. And Professor Harris did not consider these "negligent" misstatements in his forfeiture calculations. Given forfeiture's preponderance of the evidence standard, the Court can certainly infer that Defendants' "negligence" did not exceed these absurdly high sums.
IV. Stock Sale Forfeiture & Professor Harris' Testimony
With the issues posed by its April 21 and October 18 orders resolved, the Court can now turn to examining Professor Harris' testimony. But, before it does so, it must first decide the threshold question of this testimony's admissibility.
Defendants contend that Professor Harris' testimony was so flawed or speculative that it fails to comply with either the Federal Rules of Evidence, or the expert testimony standards established in Daubert v. Merrel Dow Pharm., Inc., 509 U.S. 579 (1993). In this regard, Defendants note that, among other things, Professor Harris: (i) made certain adjustments and assumptions based on his "gut feeling" (Tr. 50); (ii) admitted to making pretty basic mistakes, such as by correcting for an accounting fraud that did not occur until long after Defendants sold their stock (Tr. 130-31); and (iii) used his "judgment" to weigh the different studies he performed (Tr. 165). Thus, Defendants argue, the Court should exclude Professor Harris' testimony in its entirety.
The Government disagrees. It contends that neither the Federal Rules of Evidence nor Daubert apply to forfeiture hearings. Thus, the Government argues, Professor Harris' testimony need only have "sufficient indicia of reliability to support its probable accuracy." (Gov. Repl. Br. at 8) (internal quotation omitted).
The Court agrees. "[T]he Federal Rules of Evidence, on which Daubert is based, do not apply at sentencing." United States v. Seay, 553 F.3d 732, 741-42 (4th Cir. 2009); see also United States v. Fields, 483 F.3d 313, 341-42 (5th Cir. 2007).*fn5
"[C]riminal forfeiture is viewed as part of the sentencing process." Treacy, 639 F.3d at 48. And, during sentencing, the Court may consider "any relevant evidence having sufficient indicia of reliability." United States v. Brown, 52 F.3d 415, 425 (2d Cir. 1995).
Professor Harris' testimony meets that threshold. His curriculum vitae reflects impeccable credentials and a distinguished economics career, which included a two-year stint as the SEC's Chief Economist, and several published, peer-reviewed articles theorizing methods for modeling or calculating stock and bond prices. Given this background, the Court credits his testimony that the four stock valuation analyses he conducted (a forward-looking event study, a backward-looking event study, a discounted cash flow analysis, and a P/E ratio analysis) are all reliable methods for valuing a stock in the event of a fraud, even though each method has its own strengths and weaknesses. Moreover, each of those studies depends, largely, on historical events and data that are either not in dispute, or have been sufficiently proved by the Government (e.g., stock price movements on particularly important days, quantity of inflated inventory and misclassified R&D). It follows then that, even though Professor Harris' analyses are far from perfect (and the Court does not entirely accept his findings), they are at least admissible. See generally United States v. Williams, 506 F.3d 151, 160 (2d Cir. 2007) (listing indicia of reliability identified in FED. R. EVID. 702).*fn6 Their imperfections go not to their admissibility, but to their weight and credibility.
B. Loss Causation
Defendants spend significant time arguing that
Professor Harris' testimony fails to prove loss causation. Thus, Defendants contend, the Government has not met its burden under Dura Pharmaceuticals v. Broudo, 544 U.S. 336, 125 S. Ct. 1627, 161 L. Ed. 2d 577 (2005). Defendants are mistaken. This Court conducted a forfeiture trial, not a restitution hearing. Thus, as this Court previously held, the proper inquiry focuses on quantifying the Defendants' "ill-gotten gains," not calculating investor losses. See 2010 WL 1685826 at *2-3; 2010 U.S. Dist. LEXIS 39618 at *10-11. And Dura Pharmaceuticals does not help answer that question.*fn7
C. Overview of Findings
Professor Harris conducted four separate analyses to estimate Defendants' ill-gotten gains: a forward-looking event study, a backward-looking event study, a discounted cash flow analysis, and a price-to-earnings ("P/E") method. According to Professor Harris, Mr. Brooks' ill-gotten gains amount to $107,778,252 under the forward-looking event study, $138,608,415 under the backward-looking event study, $85,018,786 under the discounted cash flow method, and $93,219,210 under the P/E method. Ms. Hatfield's ill-gotten gains total $3,149,614 under the forward-looking event study, $3,945,478 under the backward-looking event study, $2,525,572 under the discounted cash flow method, and $2,653,475 under the P/E method.
Professor Harris then averaged these estimates by assigning different weights to each study, based on his subjective opinion as to each study's worth and reliability. He assigned the discounted cash flow method a 35% weight, the P/E method 30%, the forward-looking event study 20%, and the backward-looking event study 15%. Based on this weighted average, he opined that Mr. Brooks owed $95,818,284 in forfeiture based on his stock sales, while Ms. Hatfield owed $2,775,459 based on her stock sales.
The Court considers each study, in turn.
D. The Backward-Looking Event Study
The parties spend the most time and effort discussing backward-looking event study method. Indeed, both Professor Harris and Mr. Brooks' expert, Professor Jamie Zender, prepared such studies. So the Court considers this method first.
As Professor Harris explained, a backward-looking event study examines "the actual price response observed in the marketplace when the information about the fraud was fully disclosed," and then uses this reaction to assess what the stock price would have been on a particular date, if the fraud had never occurred. (Tr. 35). Professor Harris and Professor Zender each performed this analysis, but applied different assumptions, leading to drastically different results. For the following reasons, the Court chooses to ignore both these studies.
i. Constant Percentage v. Constant Dollar Method Professor Harris estimated ill-gotten gains using a constant percentage method. In other words, he assumed that, if a corrective disclosure caused DHB's stock price to decline by 50%, the revealed fraud had previously inflated the price by 50%. Professor Zender, on the other hand, used a constant dollar method. Thus, if a corrective disclosure caused DHB's price to decline by $1, he assumed that the revealed fraud had previously inflated its stock price by $1. The constant percentage method resulted in a much higher ill-gotten gains estimate than the constant dollar method.
The Court finds that both the constant percentage and constant dollar approaches have some merit, and some downsides. In this regard, the Court notes that -- as Professor Harris testified -- a fraud can harm a stock price in two ways. First, the fraud can cause investors to misjudge a firm's quality today, in much the same way that an orchard owner can lie about the quality of his apples. (Tr. 31-32). Second, the fraud can cause investors to overestimate a company's future prospects, because investors will extrapolate certain false information well into the future. In the orchard example, this would be akin to misjudging the tree's productivity. This kind of analysis sheds considerable light on the constant percent versus constant dollar enigma.
To use an example the parties supplied, assume that an oil company has stock trading at $100 a share, after falsely representing that it owns two wells, when it actually owns one. Such a misrepresentation would not just deceive investors as to the Company's current health (i.e., halving its assets and, presumably, current earnings); it would also equally deceive investors as to the Company's future performance -- because the Company will actually produce and sell half as much oil as investors expect. So, in actuality, the company is a $50 stock, representing 50% of its inflated value. Then suppose that, over the next year, an economic slowdown causes the price of oil to fall sharply, resulting in the company's stock falling to $50. If that happened, the fraud would then amount to $25 a share. But that $25 would still represent 50% of the Company's market value, suggesting that the fraud should be appropriately analyzed under a "constant percentage" method. Professor Zender, in fact, agreed that the constant percentage method is appropriate for such a fraud. (Tr. 710).
Conversely, it is possible that a fraud might affect a company's current health, but have little or no impact on future expected earnings. Say, for example, that, instead of falsely reporting owning two wells, the company instead falsely reports that its art collection, displayed in the corporate offices, is worth $100 million, when the collection actually consists entirely of worthless forgeries. Such a fraud would inflate the company's current assets, and thus its market value, by about $100 million. But it's highly unlikely that investors would expect the inflated art collection to generate higher future earnings. So, most likely, the fraud is "worth" $100 million (plus a reputational penalty), regardless of when the fraud is revealed, or the company's market value when it's revealed (unless, of course, the company deteriorates so much that it's no longer worth even $100 million at the time of disclosure). Such a fraud would, effectively, be a constant dollar fraud. See Tr. 56 (use constant dollar method to measure "one-off" fraud stemming from a "single event").
The fraud at issue here, however, concerned neither the misrepresentation of a perpetual profit-generating asset
(i.e., an extra oil well), nor the misrepresentation of an essentially useless asset (i.e., an art collection). Instead, it concerned DHB's economic health and profitability. Thus, neither of these two hypotheticals accurately models the fraud. Instead, the Court must assess the fraud with the understanding that, in this kind of fraud, the passage of time changed the value of the information misrepresented.
To that end, the Court begins its analysis by noting that DHB's investors lacked clairvoyance. So, as Professor Harris testified, they estimated DHB's future profits based on information then-available. Among other things, they used DHB's inflated gross profit margins, extrapolating those margins forward to help estimate future earnings. (Tr. 40-41). It follows then that, when made in 2003 and 2004, Defendants' false statements not only misrepresented DHB's current health; they also inflated expectations for DHB's performance in 2005, 2006, and so on. But, by the time the fraud was finally revealed, investors had much more information regarding the Company's actual performance in 2005 and 2006 (nearly all of it bad). This new information meant that DHB's gross profit margins in 2003 and 2004 were much less important, in 2006, for assessing either the Company's then-current health or its future earnings. So the fraud's "worth," when finally revealed, was much lower than its initial importance when propagated. Accordingly, the Court believes that the market's reaction to the fraud's revelation says little about how much overvaluation it caused when Defendants' traded. (see generally Tr. 58-60) (noting that the backward-looking event study would be "far, far more reliable" if the fraud's disclosure "occurred immediately after the insider trading," instead of years later).
For this reason, neither the constant percentage nor the constant dollar method can remotely approximate the fraud's value. This is especially so because, akin to the extra oil well hypothetical (or the pharmaceutical company hypothetical discussed in footnote 7), intervening events between the fraud's propagation and its revelation drastically reduced DHB's market value. (Tr. 61-62) (the constant dollar method would "grossly underestimate ill-gotten gains because of the decrease in the stock price due to other factors"). Indeed, when finally revealed, DHB may not have even been worth the fraud's original value.
Both experts concede as much. With respect to the constant dollar method, Professor Zender implicitly acknowledged that the constant dollar method probably underestimated inflation. (Tr. 708-712, 729-30, 755-56). On the other hand, with respect to the constant percentage method, Professor Harris agreed that this method would "overestimate the ill-gotten gains." (Tr. 63). And, even more worryingly, Professor Harris commented that "we don't know the extent" that either method misjudges the ill-gotten gains. (Tr. 63).
The Court is mindful that the Government's burden is only preponderance of the evidence, and that the Court need only make a "reasonable estimate" of the loss. Treacy, 639 F.3d at 48. But choosing either the constant percentage or the constant dollar method fails those liberal standards. The preponderance of the evidence reflects that one method overestimates the fraud, while the other understates it. And the evidence also reflects that, far from "reasonabl[y] estimat[ing]" the fraud, it's impossible to tell the extent of this over or underestimation. (Tr. 63).
ii. Intervening Events
Professor Harris' backward-looking study suffers from another problem. It made no effort to account for, and control for, events unrelated to the fraud that caused DHB's stock price to collapse. Nor does it even ignore these events. Instead, oddly, its methodology assumes that the fraud caused even assuredly non-fraud related declines. For example, DHB's stock fell 31% after the National Institute of Justice revoked the certification for protective vests containing Zylon, causing DHB to discontinue the use of this material. The parties, and Professor Harris, conceded that this price decline had nothing to do with fraud. Yet, inexplicably, Professor Harris' backward-looking study attributes 75% of the Zylon-related share drop to Defendants' fraud, a problem that Professor Harris freely acknowledges. (Tr. 243-44). This makes it highly unreliable.
iii. Statistical Significance
The Court is also concerned about Professor Harris' reliance on statistical significance. Professor Harris' backward-looking event study considered only three events, and just a handful of trading days, on the grounds that only those events were associated with "statistically significant" price changes. (Tr. 103-105.) By "statistically significant," Professor Harris meant that, following each event, DHB's stock price moved so differently from the overall market that one could say with 95% confidence that the event caused the price move. (Tr. 226-228).
The Court recognizes that the 95% confidence interval is the threshold typically used by academic economists in their work. (Tr. 227). But the Court strongly questions whether its use is appropriate in a forfeiture hearing, where the Government's burden is by a preponderance of the evidence. Indeed, when the Government cross-examined Professor Zender, it raised similar concerns. (Tr. 655). Preponderance of the evidence does not anywhere near require 95% certainty, and Professor Harris' study should have made accommodations for this lower evidentiary burden. See generally Financial Information,Inc. v. Moody's Investors Service, Inc., 81-CV-6001, 1984 WL 2119, at *3 n. 1 (S.D.N.Y. 1984) (considering testimony that applied a 51% statistical confidence level to a preponderance of the evidence determination).
Here, the use of a 95% confidence level resulted in two problems. First, it excluded nearly every day within the event window, including (presumably) several dates that had sufficient price moves to support a 50.1% to 94.9% confidence level. This would not, necessarily, be a problem -- if the excluded data carried no bias towards either side.*fn8 But the Court has no idea if that's the case, or whether the excluded information would radically alter Professor Harris' findings. And the parties did not discuss this issue.*fn9
Second, even if the excluded dates, individually, failed to fit within the 95% confidence interval, collectively they may have. In other words, while DHB's stock may have moved "insignificantly" (i.e., at less than a 95% confidence level using Professor Harris' framework) on any particular day in response to a minor event (i.e., a Newsday article), that move may have been statistically significant when viewed together with two dozen other reactions to similar minor events. And considering such days may very well have drastically altered Professor Harris' findings.
Again, the Court acknowledges the low preponderance of the evidence standard. But the parties supplied no information concerning these other dates, or what kind of effect they would have had on Professor Harris' study.
The Court does not find the backward-looking event study sufficiently informative to use it in calculating forfeiture. It so concludes for the foregoing reasons, discussed at length above. First, due to the nature of this fraud, it was worth less to the market when revealed then when propagated. Second, as Professor Harris recognized, using a constant percentage method would inflate ill-gotten gains estimates, while the constant dollar method would substantially deflate them. And neither Professor provided even a "ballpark" estimate for the degree of this over or undervaluation, much less a basis for that estimate. Third, the backward looking study does not properly take into account reasons for DHB's stock price decline that had nothing to do with fraud. Fourth, the study's reliance on a 95% confidence interval improperly excludes too many days, given the applicable preponderance of the evidence burden. And fifth, as discussed below, reliance on the backward-looking event study is unnecessary, because there are better ways to calculate forfeiture.
E. The Forward-Looking Event Study
Professor Harris, but not Professor Zender, also conducted a forward-looking study. This study analyzed the effect that sixteen separate events had on DHB's stock price, in the period leading up to Defendants' stock sales. Defendants, however, successfully impeached this study in several ways. Consequently, for the reasons discussed below, the Court also chooses to ignore this study.
i. "Gut" Based Adjustments
Professor Harris adjusted DHB's "excess return" (its stock's return, minus the market's overall return) on specific days by 50%, to account for the fraud's effects. (Tr. 49-53). Professor Harris made these adjustments based on his reasonable assumption that, if investors knew DHB's true financial picture, they would have reacted more negatively to bad events, and less enthusiastically to good events. But Professor Harris did not derive the 50% figure from any kind of objective analysis, such as a statistical study. Instead, he admitted that the 50% adjustment emerged from his "gut." (Tr. 50). Professor Harris also admitted that these adjustments followed no generally accepted method, because "to the best of my knowledge, there is no generally accepted method" as "these are studies that are not generally done." (Tr. 282).
ii. Bad Data-Based Adjustments
Professor Harris also adjusted DHB's excess returns stemming from new sales contract announcements down by 23.2%, 24.2%, 24.4%, or 24.5%, depending on the contract. These adjustments did not come from Professor Harris' "gut." But they did stem from bad data. As Professor Harris admitted on cross examination, these adjustments depended on the Indictment's round-number allegation that DHB enjoyed a 21% true gross profit margin, not the 27% to 28% that it publicly claimed. (Tr. 44, 146-148).*fn10 But, despite this summed up "round number," the Indictment's factual allegations suggest that DHB's gross profit margin during the relevant time period ranged from 21.63% to 21.86%. Id. According to Defendants (and not rebutted by the Government), this error alone inflated Professor Harris' ill-gotten gains estimate by 10%. (Brooks Br. at 28).
iii. Excluded Days
Professor Harris' forward-looking event study considered only 16 days during a 421 day period. According to Professor Zender, accounting for those 405 excluded days would reduce Professor Harris' ill-gotten gains estimate "by roughly 50%." (Tr. 547). And the Government failed to rebut this criticism.
Relatedly, by excluding those 405 days in this study, Professor Harris failed to consider factors that impacted DHB's stock price, but bore no relation to the frauds. For instance, Professor Harris expressed befuddlement concerning DHB's 27.3% jump between April 6 and April 8, 2004. He opined that, according to his model, DHB's true value did not change during this period. (Tr. 174-177). In reality, the stock apparently rose on heightened terrorism concerns, a distinct, non-fraud related event. (Id.). Similarly, Professor Harris acknowledged that his study ignored that the stock price of DHB's two principal competitors, Armor Holdings and Ceradyne, increased by 255% and 536% respectively during the study period, despite the no fraud in either company. (Tr. 179-180).*fn11
iv. PACA Inventory
Professor Harris' forward-looking event study incorporated the PACA inventory fraud into his estimate of DHB's true gross profit margin, and thus his ill-gotten gains estimates. (Tr. 154-158). But, by all accounts, the PACA inventory fraud did not begin until after Defendants completed their stock sales. Thus, the PACA inventory fraud could not have affected DHB's stock price when Defendants traded.*fn12
The problems with Professor Harris' forward-looking event study are not as severe as those with his backward-looking event study. But they are still pervasive. And, given these errors, the Court fails to see how the study informs a forfeiture analysis. This is especially so given that, as discussed below, the discounted cash flow and P/E methods have far fewer issues, and better approximate how investors would have valued DHB's stock in the absence of fraud. Consequently, the Court also elects to ignore the forward-looking event study.
F. The Discounted Cash Flow and P/E Method Studies Professor Harris' final two studies, the discounted cash flow analysis and the P/E ratio analysis, are similar in many respects. In both studies, Professor Harris seeks to calculate DHB's true stock value by taking DHB's accurate, non-fraudulent financial results, and plugging these figures into accepted stock valuation methods. In the discounted cash flow analysis, Professor Harris calculated DHB's true value as the appropriately weighted sum of its future expected cash flows, discounted to reflect market risk. In the P/E method, Professor Harris sought to determine DHB's true value by correcting its earnings per share to remove the fraud's influence, and then reducing the market-awarded P/E ratio to reflect less optimistic growth prospects. Defendants raise several objections to these methods. But, on the whole, these objections raise less concern. The Court considers each in turn.
i. Discounted Cash Flow Analysis: Reduced Short Term Grown Rate Assumption
Defendants devote the most space to challenging Professor Harris' short term growth rate assumptions. So the Court will consider this criticism first.
Professor Harris testified that, in the period before Defendants' stock sales, analysts valued DHB's stock by projecting the Company's cash flow to grow at a short-term rate of 24.17%. Professor Harris' model reduced that projection by 5 percentage points, to 19.17%, on the assumption that -- if analysts knew DHB's real financial results, they would have had less rosy expectations. Mr. Brooks contends that this was another assumption pulled from Professor Harris' "gut." (Brooks Br. 39). But Mr. Brooks is wrong.
As Professor Harris testified, he derived this number from two sources. First, he explained "as recently as . . . six or nine months" before Defendants' continuing fraud culminated in their stock sales, analysts predicted only "10 to 15 percent, or 15 to 20" percent growth. (Tr. 78). Thus, this assumption stemmed, at least in part, from previous analyst estimates -- before Defendants' continuing fraud helped drive those estimates higher. And Professor Harris picked a number that was both at the high range of those estimates, and still assumed stunningly aggressive growth.
Second, as Professor Harris explained, he used the gross profit margin fraud to inform his earnings growth estimates, as used in the discounted cash flow analysis. (Tr. 150-153.) And the two figures share remarkable symmetry. For the nine months that preceded Defendants' stock sales, DHB reported a gross profit margin of 27.82%.*fn13 In fact, DHB's gross profit margin around that time was 21.86%.*fn14 So DHB's gross profit margin was inflated by 5.96 percentage points, or about 21.42%. Likewise, Professor Harris' model reduced DHB's expected short term growth rate by 5 percentage points (24.17% to 19.17%), or 20.68%.
Third, as even Professor Zender conceded, on direct, it is "possible that by overstating ending inventory we might actually expect that market expectations would diminish in terms of the growth in future cash flows." (Tr. 564). And Professor Zender went on to opine that DHB's inventory fraud suggested a "less efficient firm," and that "with a less efficient firm, one possible change in market expectations would be to say I will expect less future growth." (Tr. 564-65). Thus, at least on a theoretical level, Professor Zender agreed with Professor Harris that it was appropriate to reduce DHB's expected short-term growth.
ii. Discounted Cash Flow: Discount Rate
Defendants also criticize Professor Harris' choice of an 11.48%
discount rate. But the Court sees no error in using that rate or,
rather, any error that favors the Government. Professor Harris did not
pull this number out of thin air. Instead, as testified, DHB's actual
fraud-driven closing price implies that investors assigned an 11.48%
discount rate to investing in the Company. (Tr. 76-77).*fn15
He then plugged that same discount rate into his no-fraud
Professor Harris' selection of an 11.48% discount rate was, if anything, much too conservative. For, as Professor Harris explained, riskier investments incur higher discount rates because "the[ir] promise to deliver money in the future" is less "reliable." (Tr. 77). And, if investors knew the full extent of Defendants' fraud in late-2004 (e.g., management stealing from the Company through unauthorized compensation, management not putting in place internal controls sufficient to track R&D as performed, etc.), they probably would have considered DHB to be a much riskier investment, warranting a higher discount rate.
iii. Discounted Cash Flow Method: Event-Based Adjustments
Defendants further protest Professor Harris' use of two event-based adjustments, which he used in the forward-looking study, and then transferred to the discounted cash flow analysis. (Tr. 476-478). But Defendants neglect that, in the discounted cash flow analysis, these adjustments benefit them. For, if Professor Harris had limited himself to a calculation based on published cash flow numbers, he would not have revised DHB's true value upwards at all. Instead, he gave the Defendants significant benefit from published good news that post-dated DHB's third quarter earnings release. This was not error.
iv. PACA Inventory Fraud Issue
Defendants do make one cogent criticism of Professor Harris' discounted cash flow analysis. They contend that his analysis improperly considered the PACA inventory fraud, which did not occur until after Defendants' stock sales. And this time, the Court agrees. Indeed, Professor Harris himself acknowledged the error. (Tr. 129-131).
But this error, though significant, does not fundamentally impeach Professor Harris' methodology. And it can be easily corrected. Accordingly, rather than disregard his discounted cash flow analysis, the Court instead orders the Government and Professor Harris to provide revised calculations that apply the same methodology, but do not "correct" for a PACA fraud that had yet to occur. Professor Harris may use his reasoned judgment in deciding how best to accomplish this task (i.e., considering fraud in the fourth quarter of 2003 and thus the previous four reported quarters, extrapolating financial data from the first nine months of 2004, etc). And the Court will then use that revised analysis to order forfeiture.
v. Restated Financials Issue
With respect to the discounted cash flow analysis, one more subject deserves consideration. Professor Harris' study depended on the Indictment's allegations. But DHB's restated financials ultimately reported much bigger financial discrepancies than the Indictment alleged. Thus, whereas Professor Harris' conclusions assumed that DHB had prior year earnings of $.39 a share, its revised financials reflected negative earnings. (Gov. Br. at 43, citing GX 5129 at 94, 99). Accordingly, if Professor Harris based his study on DHB's restated financials, instead of the Indictment's claims, he almost certainly would have produced a much higher ill-gotten gains estimate. Based on this choice, the Government argues that the Court can disregard "any inaccuracy or fault in Professor Harris' analysis." (Gov. Br. at 43).
The Court largely disagrees. The Government chose to seek forfeiture based on the Indictment's allegations, not the restatement's conclusions. Quite frankly, this strategic choice surprised the Court. But the Government is stuck with its decision. And Professor Harris' discounted cash flow analysis reflected that choice. To the extent that Professor Harris admitted a significant, quantifiable, and correctible error in his analysis (i.e., the PACA fraud), the Government cannot erase that error by arguing that Professor Harris could have used a different data framework entirely (the restatement). This is particularly so because the Government does not, in fact, argue that its own witness erred in not using the restatement. So the Court can only presume that the Government continues to agree with its witness that the Indictment's numbers control.
That being said, the Government did admit the restated financials into evidence. So the Court can consider these numbers. And, in so considering, the Court finds that they increase its already considerable confidence that the discounted cash flow study surpasses the Government's preponderance of the evidence burden. This is because the restated financials further augment the Court's finding that Professor Harris used very conservative assumptions in calculating forfeiture -- calculations that, if anything, result in him understating Defendants' ill-gotten gains.
vi. The P/E Study
Professor Harris' P/E study is also helpful, but somewhat less reliable. Defendants raise two principal criticisms of this study.
First, Defendants repeat their argument that the P/E study wrongly incorporates the PACA fraud. Again, the Court agrees. Thus, just as it instructed the Government and Professor Harris to recalculate the discounted cash flow analysis to remove the PACA fraud, it likewise requires the Government and Professor Harris to recalculate the P/E study.
The second criticism concerns Professor Harris' decision to reduce DHB's original, fraud-driven P/E ratio by 30%, from 34.4 to 24.1. This critique deserves more discussion. Professor Harris admitted that "there was no statistical analysis" behind his decision to reduce DHB's P/E ratio by 30%. (Tr. 268-69). Professor Harris did, however, testify that the average 27.4 P/E ratio of DHB's competitors "informed" his analysis, and comment that his 24.1 estimate did not substantially differ from figure.*fn16 (Tr. 86-87). Professor Harris then added that he considered the gross profit margin fraud in making his assumptions about an appropriate P/E ratio. (Tr. 150-52). And he inferred that, in the absence of the PACA fraud, he probably would not have reduced the P/E ratio as severely. (Tr. 431) (testifying that, under an assumption that there was no inventory fraud, he would only have applied a 20% reduction to the P/E ratio).
The Court is not entirely comfortable with this discount. In this regard, the Court notes that P/E ratios generally reflect two distinct valuation concerns. First, as Professor Harris explained, P/E ratios depend on "expectations of future earnings growth, with high growth companies . . . commanding generally high price earnings ratios." (Tr. 84). Second, the Court recognizes that P/E ratios also "prox[y]" an investment's "riskiness," with "more secure" investments warranting higher P/E ratios. Berens v. Ludwig, 160 F.3d 1144, 1146 (7th Cir. 1998). Professor Harris' discounted cash flow study incorporated other metrics to reflect those valuation issues: the expected short-term earnings growth rate, which reflects expected growth, and the discount rate, which reflects riskiness. But, in the discounted cash flow study, Professor Harris did not reduce either of these metrics by 30%. Instead, he dropped the short term earnings growth rate by 20.68%, and left the discount rate intact. And Professor Harris did not sufficiently explain why, in his model, the P/E ratio deserved a steeper discount.
That being said, several factors tip towards the Court not rejecting this study. First, the Government's burden is only a preponderance of the evidence. And here, the Government has provided an expert opinion from a well-qualified economist, concerning what an appropriate reduction would be. And Defendants did not refute this assessment. Quite the opposite, when testifying on Mr. Brooks' behalf, Professor Zender implicitly conceded that some kind of reduction was appropriate. (Tr. 587-588) ("Previous reporting will certainly influence expectations"). And, though he commented that Professor Harris' assumptions were "subjective," Professor Zender did not suggest that the proposed 30% reduction was excessive, or propose a more appropriate reduction. (Tr. 575-76; see also 606).*fn17
Second, as even Professor Zender conceded, the P/E ratio is "the most common" valuation technique that investors use. (Tr. 599-600). Thus, unlike the event studies, the P/E ratio seeks to approximate how investors would have valued DHB at the time of the stock sales, in the absence of a fraud.
Third, as is the case with the discounted cash flow study, DHB's restated financials largely alleviate any fear that Professor Harris' model tips too strongly towards the Government.
Finally, the Court reminds the parties that "[t]he calculation of forfeiture amounts is not an exact science," and that the Court need only make a "reasonable estimate of the loss" using "general points of reference." Treacy, 639 F.3d at 48. And, in using these points of reference, it is not error to calculate forfeiture in a manner that "leaves open a possibility that defendants will be forfeiting profits that they would have made outside of their criminal activities." Lizza Industries, Inc., 775 F.2d at 498. Thus, even if Professor Harris' P/E model (and the earlier discussed discounted cash flow study) cannot perfectly describe an alternative history where no fraud occurred, and even if they rely on assumptions that might favor the Government to some degree (overall, they do not), it is not error for the Court to rely on them.
It follows then that the P/E study's imperfections do not preclude the Court from considering it.
G. Weighing the Studies
The Court elects to ignore the two event studies, and instead bases its forfeiture calculations on the discounted cash flow and P/E analyses.
In light of the P/E study's imperfections, the Court affords it less weight than the discounted cash flow study, just as Professor Harris himself recommends. Professor Harris initially suggested affording the P/E study a 30% weight, compared to the 35% weight he assigned the discounted cash flow study. But, because the Court has chosen to ignore the event studies, it must afford the discounted cash flow and P/E analyses more weight. Professor Harris recommended weighting them with a 6-5 ratio. The Court sees no reason to disagree with that conclusion. Consequently, the Court will afford the discounted cash flow study a 54.55% weight, and the P/E study a 45.45% weight.
H. Conclusion: Stock Sale Forfeiture
The Court will disregard Professor Harris' two event studies, and instead calculate forfeiture using his discounted cash flow and P/E studies. The Court will afford the discounted cash flow study a 54.55% weight, and the P/E study a 45.45% weight. The Court will not, however, use the calculations already provided, because they improperly incorporate the effects of the PACA fraud that post-dated the illicit securities transactions. Instead, the Court directs the Government, and Professor Harris, to submit revised calculations that omit the PACA fraud. The Court leaves to Professor Harris' discretion how to best accomplish this task (i.e., considering the fourth quarter of 2003 instead of the fourth quarter of 2004, extrapolating from nine months of 2003, etc.). These revised calculations shall be provided within fourteen (14) days, and may be accompanied by a fifteen (15) page brief explaining Professor Harris' revised work. Defendants will then have fourteen (14) days to submit any expert study and/or brief in opposition, with any such brief also not to exceed fifteen (15) pages.
V. Forfeiture: Unauthorized Compensation
The Government also seeks forfeiture of assets obtained through the unauthorized compensation scheme. With respect to Mr. Brooks, the Government seeks a $6,444,681 money judgment, plus eighty-five separate identifiable assets. With respect to Ms. Hatfield, the Government seeks a $94,000 money judgment. The Court considers each of these in turn.
A. Mr. Brooks
The jury convicted Mr. Brooks of the wire and mail fraud counts that comprised the Indictment's unauthorized compensations scheme. So the Government argues that Mr. Brooks must forfeit this scheme's proceeds, including "all of the items of personal property that were purchased and paid for using either a DHB, or DHB affiliate, credit card, or with checks drawn on accounts held in the name of DHB and/or its affiliates." (Gov. Br. at 3). The Court finds that the immense trial evidence concerning this scheme, and Robert Cappadona's tracing testimony, more than suffice to meet the Government's forfeiture evidentiary burden as to Mr. Brooks.
Mr. Brooks, indeed, barely puts up a fight with respect to these assets. He ignored the unauthorized compensation scheme in his initial brief. And, in reply, he devotes a scant few paragraphs to an argument that the Government failed to prove that DHB paid 100% of the American Express charges on a credit card issued in the Company's name. (Brooks Reply Br. at 25-26) (citing Tr. 847-48). Given this posture, the Court sees no need to reach the merits of Mr. Brooks' argument. For it is well-settled that the Court should "not consider an argument raised for the first time in a reply brief." Castro v. Holder, 597 F.3d 93, 96 n. 2 (2d Cir. 2010) (internal citations and quotations omitted); Spa 77 G L.P. v. Motiva Enterprises LLC, 09-CV-1665, 2011 WL 743581, at *15 n.12 (E.D.N.Y. Feb. 22, 2011). Consequently, the Court orders full forfeiture of Mr. Brooks' proceeds of the unauthorized compensation scheme (Items 1 through 85 listed on Appendix I), just as the Government requests.
B. Ms. Hatfield
i. Procedural Background
The Government's unauthorized compensation argument with respect to Ms. Hatfield is trickier. Originally, the Government alleged that this scheme consisted of: (i) $94,000 in payments to Ms. Hatfield's husband, Butch Hatfield, through his company, WGH Consulting; and (ii) $95,000 in payments to Ms. Hatfield's son, Keith. But, during the principal trial, the Government produced "no evidence" to suggest that the payments to Keith Hatfield were fraudulent. 2010 WL 2710616, at *4 (E.D.N.Y. July 7, 2010). And, as a result, the Court struck those allegations, although it permitted the allegations regarding the WGH Consulting payments to reach the jury. Id.
The jury then dealt the Government's case another blow. It failed to convict Ms. Hatfield of wire fraud or mail fraud, the core counts making up the unauthorized compensation scheme. Instead, it convicted her only of conspiring to commit mail and wire fraud. And, as the parties did not request a special verdict, the Court can only speculate about what this conviction stemmed from. Accordingly, the Court asked the parties for supplemental briefing dedicated to: (i) identifying evidence in the record concerning these payments; and (ii) argument regarding whether a preponderance of this evidence reflects that the payments were fraudulent proceeds. The parties responded, and, in the following section, the Court documents what this evidence reflects.
ii. The Record
In early 2001, Butch Hatfield reached out to Joseph Willoughby, a local official in Campbell County, Tennessee. (Trial Tr. 16492, 16501). Mr. Hatfield sought information about potential properties that could be used for a "needle trades" factory that would build "bulletproof vests" and "body armor products." (Id. at 16492-93). In response, Mr. Willoughby provided Mr. Hatfield with information such as "aerial photographs of industrial parks" and "contact information for agencies within our region." (Id. 16493-94). Mr. Willoughby then showed Mr. Hatfield around various properties. (Id. 16495). Mr. Willoughby also fielded inquiries from Mr. Hatfield about soliciting government grants. (Id. 16498).
Around this time, Mr. Hatfield also reached out to an old acquaintance, Benjamin Christopher Arnold. In this regard, Mr. Hatfield inquired about a property that Mr. Arnold owned. And, after DHB rented the property through Mr. Arnold's agent, Mr. Arnold showed Mr. Hatfield around it "two or three" times. (Id. 16510-13).
Around March or April 2001, Mr. Hatfield began supervising renovations to the newly leased property. This work included adding a storage facility and updating the electrical work. (Id. 16512, 16544). Altogether, the work took "three or four months." (Id. 16554). And Mr. Hatfield was on site "every day," supervising "several" workers. (Id.).
Ultimately, PACA moved into the facility and began paying rent. (Id. 16513). But even after PACA moved in, Mr. Hatfield continued to "help" out at the site. (Id. 16555). This work included tasks such as running a forklift, loading trucks, supplying gravel, and paying contractors. (Id.). One contractor testified that he dealt personally with Mr. Hatfield "ten or twelve times" after PACA moved in. (Id. at 16556).
In late 2001 or early 2002, Mr. Hatfield reached out to another local county official, Jerry Cross. Mr. Hatfield sought Mr. Cross' help in getting the city to put signs up around a rural road that led to the PACA plant. (Id. 16569).
Later on, in the middle of 2002, Mr. Hatfield had another meeting with Mr. Willoughby, which Ms. Hatfield and a retired army general also attended. (Id. 16496-97, 16549). This meeting discussed procuring government contracts and designing vests. (Id.)
And around September 2002, Mr. Hatfield reached out to Mr. Willoughby again. (Id. 16499). He sought Mr. Willoughby's assistance in assembling a list of local "dignitaries" to invite to an upcoming PACA social function. (Id.) Mr. Willoughby helped out, and later attended the party. There, he noticed Mr. Hatfield serving as a "host" and "tour guide." (Id.)
Mr. Hatfield apparently served a similar role at other PACA social events. He invited Mr. Cross, then the County Mayor, to at least "six or eight" such events. (Id. 16570). At each event, Mr. Cross observed Mr. Hatfield "mingling with the crowd," "assisting people who had jobs to do," and serving as a "good PR guy." (Id. 16571).
In addition to these specific tasks, Mr. Hatfield also did other assorted work for PACA during 2001 and 2002. For example, he played a role in hiring workers. (Id. 16571). In fact, Mr. Hatfield was so active in PACA that Mr. Cross mistakenly thought he "owned" the company. (Id. 16572).
Despite Mr. Hatfield's considerable work on PACA's behalf, the Government contends that DHB's payments to his company, WGH Consulting, were fraudulent. In this regard, the Government relies on testimony from DHB's Chief Financial Officer, Dawn Schlegel.
According to Ms. Schlegel, the Company offered Ms. Hatfield a raise in 2001. (Id. 5984). Ms. Hatfield, however, said "she did not want it, so [Mr. Brooks] instructed [Ms. Schlegel] to pay [Ms. Hatfield's] husband's company $5,000 a month." (Id.). Ms. Schlegel indicated that, according to Mr. Brooks, this arrangement reflected Ms. Hatfield's wishes, who supposedly "did not want a raise" because she "was paying too much in taxes." (Id. 5986). Mr. Schlegel acknowledged that Mr. Hatfield was "helpful" in doing work for PACA, but expressed her understanding that "it was not payment for his services for the most part . . . . it was payment for Sandra's services." (Id. 5989). To that end, Ms. Schlegel noted that DHB never received an invoice from Mr. Hatfield's company. (Id. at 5989-90).
In total, DHB paid WGH Consulting $35,000 in 2001, and $59,000 in 2002. The $59,000 worth of payments in 2002 included a $24,000 payment in December 2002. This balloon payment meant that, in 2002, DHB paid WGH Consulting just $1,000 less than the $60,000 related party transaction reporting requirement.
Additionally, it should be noted that, despite Ms. Schlegel's testimony regarding what Mr. Brooks told her about what Ms. Hatfield supposedly told him (double hearsay, though admissible as alleged co-conspirator statements), there is no evidence that the WGH Consulting payments served to reduce Ms. Hatfield's tax liability. On the contrary, Ms. and Mr. Hatfield declared the payments as income on their joint tax return, in addition to declaring them on WGH Consulting's business tax return. See Docket No. 1453.
At the outset, the Court notes that both sides implicitly ask the Court to draw improbable inferences. With respect to the Government, it submitted no evidence that DHB (or PACA) otherwise paid Mr. Hatfield for the considerable work he did, which included setting up meetings, serving as a "good PR guy," supervising "several" laborers "every day" for "three to four months" to get the factory in working order, and performing significant assorted work once PACA started operations. So, in asking the Court to infer that the WGH Payments amounted to fraudulently concealed compensation to Ms. Hatfield, the Government essentially asks the Court to assume that Mr. Hatfield spent untold hours working for free, including "every day" for "three to four months."
Moreover, although the Government asks the Court to infer that the payments represented fraudulently concealed compensation to Ms. Hatfield, the Government provides no motive for this alleged fraudulent concealment. There is no suggestion that DHB would not have agreed to pay Ms. Hatfield more. On the contrary, the Government's own evidence suggests that DHB's highest officers approved the payments as a "raise" to Ms. Hatfield.*fn18 Similarly, the Government's claims that the payments functioned as a tax fraud don't hold water. For the Government provided zero evidence to suggest that Ms. Hatfield, in fact, received a tax benefit from having the payments made to WGH Consulting instead of herself. And, insofar as Ms. and Mr. Hatfield declared the income on their joint return, the Court cannot fathom how such a "fraud" could have operated.
On the other hand, Ms. Hatfield also asks the Court to draw very strange inferences. Effectively, she asks the Court to assume that Mr. Hatfield performed substantial labor in exchange for deferred compensation to be paid over the eighteen months after most of his work ended, without (apparently) signing any kind of written contract or ever supplying an invoice.
That being said, and after much consideration, the Court finds that the Government has not met its preponderance of the evidence burden with respect to the WGH Consulting payments. For, despite claiming that the payments represent fraudulent proceeds, the Government has not explained what the fraud was, or who was supposedly defrauded. It wasn't DHB, because DHB's highest officers approved the payments. And it wasn't the IRS, because Ms. Hatfield reaped no tax benefits from the supposed "scheme."
The Government's strongest case is that the payments represented some kind of securities fraud, by deceiving shareholders about: (1) how much Ms. Hatfield earned; and/or (2) DHB's related party transactions. But the Government hasn't adequately supported such a case. Among other things, the Government has not identified any evidence in the record suggesting a "substantial likelihood" that any misstatement or omission concerning these modest sums ($35,000 in 2001 and $59,000 in 2002) "would have been viewed by the reasonable investor as having significantly altered the 'total mix' of information made available." Caiola v. Citibank, N.A., New York, 295 F.3d 312, 329 (2d Cir. 2002) (defining materiality under Section 10(b) and Rule 10b-5); 724 F. Supp. 2d at 325 (materiality required in the criminal securities fraud context as well).*fn19 Nor, for that matter, has the Government supplied any credible evidence that Ms. Hatfield had fraudulent intent, much less explained what this supposed "intent" was. See 724 F. Supp. 2d at 324 (fraudulent intent is an element of securities fraud under 18 U.S.C. § 1348).
Consequently, the Court DENIES the Government's motion to the extent that it seeks forfeiture of the $94,000 in WGH Consulting payments.
Forfeiture also requires the Government to trace the proceeds of Defendants' crimes to specific assets. Through Mr. Cappadona's testimony, the Government has done so. True, Mr. Brooks does contend that Mr. Cappadona's testimony fails to meet this burden. But his argument is without merit.
Specifically, Mr. Brooks contends that, because the stock sale revenue had both a legal and illicit component, the Government cannot say that any specific dollar is the proceeds of fraud. But, as the Government correctly points out, nothing precludes the Court from ordering each asset to be forfeited in the same percentage it determines the insider sales to be ill-gotten gains (i.e., if 40% of the insider sales are ill-gotten gains, then 40% of each asset is forfeitable). (Gov. Br. at 23-24). The Court sees no error in that approach, with respect to Mr. Brooks' cash and investment accounts.
The situation is more complicated with respect to the jewelry and automobile that Mr. Brooks purchased with the insider trading proceeds. Last October 20, the Court opined that "these items are both small in number and largely unique from one another." 2010 U.S. Dist. LEXIS 114493 at *5, 2010 WL 4340632 at *2. Thus, "if the evidence at the forfeiture trial shows that Mr. Brooks purchased these items using money from a commingled account, it may be 'impossible to say that the . . . items of jewelry are 'traceable to" Mr. Brooks' offenses." Id. (citing United States v. Voigt, 89 F.3d 1050, 1082 (3d Cir. 1996)).
The Court's October 20 Order invited the parties "to present focused briefing on the issue of the cars, watches and jewelry." Id. at *5-6. Understandably, the parties' forfeiture papers did not much discuss this issue, given the more important matters at stake. But the Court remains hesitant to issue a binding ruling without more guidance from the parties. So it again defers consideration of whether to order this property (Items 106 to 113 listed in Appendix I) liquidated, or instead order forfeiture of substitute property in the form of money. The parties, in the ordered supplemental briefing, may advise the Court how to proceed and cite any authority they wish.
The Government's motion for a preliminary order of forfeiture is GRANTED IN PART, DENIED IN PART, and the Court RESERVES DECISION IN PART.
The motion is GRANTED to the extent that it seeks forfeiture of assets that Mr. Brooks obtained through the unauthorized compensation scheme. It is also GRANTED insofar as the Court finds that at least a portion of both Mr. Brooks' and Ms. Hatfield's alleged insider proceeds are forfeitable. However, the Court RESERVES DECISION as to how much of these assets are forfeitable, pending revised calculations from the Government and Professor Harris, and supplemental briefing from the parties. Thus, the motion is DENIED to the extent that it seeks forfeiture of the alleged insider trading proceeds in their entirety.
The motion is also DENIED to the extent that it seeks forfeiture of the $94,000 paid to WGH Consulting.
The Government and Professor Harris are directed to supply revised calculations, consistent with this Order, within fourteen (14) days. The Government may accompany these revised calculations with a fifteen (15) page supplemental brief explaining Professor Harris' revised work and addressing the forfeiture of Items 106 to 113 listed in Appendix A. Defendants will then have fourteen (14) days to submit any expert study and/or supplemental brief in opposition, with such brief also not to exceed fifteen (15) pages. No reply papers will be accepted. No party should use this supplemental briefing to reargue or seek reconsideration of matters that this Order decided.
Joanna Seybert, U.S.D.J.
Central Islip, New York