445 F.3d 634

BECKER, Circuit Judge,
concurring and dissenting.
I join fully in Parts II, III.C, III.E, III.F, and V of the majority opinion. I join in Part III.D to the extent that it holds that the willful blindness instruction was justified as to Defendant Leahy and constituted harmless error as to Defendants Smith and Dantone, although I will note my reservations regarding the use of that instruction as to Dantone. I join in Part IV except to the extent that it conflicts with my analysis of the elements of bank fraud, set forth in detail below. I do not join in Parts III.A and B or in the judgment because I believe that the majority’s resolution of the bank fraud and moral uprightness jury charge issues is incorrect. In particular, I think the majority’s discussion of the elements of bank fraud is inconsistent with this Court’s decision in United States v. Thomas, 315 F.3d 190 (3d Cir.2002), a fair reading of which compels the conclusion that the jury instructions used here were erroneous.
I.
The majority finds that the jury instructions used in this case were consistent with our decision in United States v. Thomas, 315 F.3d 190 (2002), which interpreted the federal bank fraud statute. I disagree.
A.
The bank fraud statute reads:
Whoever knowingly executes, or attempts to execute, a scheme or artifice—
(1) to defraud a financial institution; or
(2) to obtain any of the moneys, funds, credits, assets, securities, or other property owned by, or under the custody or control of, a financial institution, by means of false or fraudulent pretenses, representations, or promises;
shall be fined not more than $ 1,000,000 or imprisoned not more than 30 years, or both.
18 U.S.C. § 1344. In Thomas, we held that subsection (2) of the statute cannot serve as an independent basis for conviction. Thus, a defendant cannot be convicted of bank fraud solely because he uses false pretenses to obtain money or property that is under the bank’s custody but does not belong to the bank; rather, he must also act to “defraud the bank,” thus subjecting him to liability under subsection (1). While we acknowledged that this conjunctive reading was in tension with the plain disjunctive language of the statute, we found that it was necessary to effectuate to the statute’s purpose “ ‘to protect the federal government’s interest as an insurer of financial institutions.’ ” 315 F.3d at 200 (quoting United States v. Davis, 989 F.2d 244, 247 (7th Cir.1993)). As Thomas explained, if a defendant takes money that is under the bank’s custody but does not belong to the bank, and in so doing does not subject the bank to any risk of losing its own funds, he has not threatened this interest and therefore has not committed bank fraud.
The District Court in this case, acting before our decision in Thomas was issued, instructed the jury that either subsection of the bank fraud statute could serve as a basis for conviction. It stated:
The bank fraud law provides that whoever knowingly executes or attempts to execute a scheme or artifice, one, to defraud a federally chartered or insured financial institution, or two, to obtain any of the moneys, funds, credits, assets, security or other property owned by or under the control or custody of a financial institution by means of false or fraudulent pretenses, representations or promises, shall be guilty of the crime of bank fraud....
Members of the jury, the first element is that the government must prove beyond a reasonable doubt that there was a scheme or artifice to defraud a financial institution, or a scheme or artifice to obtain any of the money owned by or under the custody or control of a financial institution by means of false or fraudulent pretenses, representations or promises.
(emphasis added). Because the District Court instructed the jury that it could convict under either subsection of the bank fraud statute, it committed error under Thomas. ,
The majority does not dispute that the above language, standing alone, is inconsistent with Thomas. But it argues that the District Court’s error is saved by a later portion of the charge: “The intent element of bank fraud is an intent to deceive the bank in order to obtain from it money or other property.” Maj. Op. at 642. This instruction, the majority argues, precluded the jury from convicting solely on the basis of subsection (2).
I respectfully disagree. First, the intent instruction was an isolated one, preceded and succeeded by the disjunctive language quoted above. Indeed, when the jury later requested that the District Court repeat the elements of bank fraud, the Court again instructed the jury that it could convict under either subsection of the statute and failed to repeat the sentence setting forth the intent standard. Rather, it simply instructed the jury that it could convict if it found that the defendants acted “with the intent to defraud,” not, as we required in Thomas, “with the intent to defraud the bank.” See 315 F.3d at 197.
Second, the intent instruction simply did not preclude the jury from convicting solely on the basis of subsection (2). As we held in Thomas, a defendant who deceives a bank in order to obtain from it money or property belonging to a third party (but in the custody of the bank) does not commit bank fraud, unless he also knowingly subjects the bank itself to a loss or risk of loss. But he could easily have been convicted under these instructions, which stated that “[t]he intent element of bank fraud is an intent to deceive the bank in order to obtain from it money or other property.”
What these instructions lack is the critical element of bank fraud identified in Thomas: namely, that the defendant must have the intent “to victimize the bank,” see 315 F.3d at 198, 200, either by taking the bank’s own funds or by putting the bank at a risk of future loss or liability. As we held in Thomas, “Congress sought to proscribe conduct that ‘victimize[d]’ banks, which suggests that the bank must be deliberately harmed before the statute is violated. We believe that, given the legislative intent, harm or loss to the bank must be contemplated by the wrongdoer to make out a crime of bank fraud.” Id. at 200. By not requiring such an intent, the instructions permitted the jury to convict under subsection (2) of the statute.
As Thomas made clear, “the intent to defraud the bank” requires more than merely “an intent to deceive the bank in order to obtain from it money or other property.” See 315 F.3d at 200 (“The Government also suggests that mere ‘deceptive conduct’ toward the bank establishes intent to defraud. We disagree.”). Indeed, Thomas herself had the “intent to deceive the bank in order to obtain from it money or other property,” but we held that she did not have the “intent to defraud the bank,” which, again, requires an intent to “victimize” the bank by exposing it to a loss or risk of loss. Thus Thomas could have been convicted under these jury instructions, even though, as we held in that case, she did not commit the crime of bank fraud. This fact alone should be sufficient to demonstrate that the jury instructions in this case were flawed.
B.
The majority makes an additional argument in order to justify the jury instructions used in this case. According to the majority, the mens rea requirement set forth in Thomas only applies to some bank fraud. Other types of bank fraud — such as that committed by the defendants in this case — are not subject to Thomas’s mens rea requirement.
The majority justifies its effort to cabin Thomas by arguing that a later decision of this Court, United States v. Khorozian, 333 F.3d 498 (3d Cir.2003), created a distinction between those cases that “involved fraud on a third party where the bank was merely an ‘unwitting instrumentality’ in the fraud” and those in which the bank was itself the “target of deception.” According to the majority, only in the former case, where the bank is merely an “unwitting instrumentality,” do we require “the additional requirement of proof of an ‘in-ten[t] to cause a bank a loss or potential liability.’ ” Maj. Op. at 645 (alteration in original). The majority argues that in cases where the bank itself was the “target of deception,” “proof of a specific intent to defraud the bank plus fraudulent conduct (e.g., misrepresentations) which creates an actual loss or a risk of loss.” Maj. Op. at 644. In such cases, according to the majority, proof of an actual intent to cause the bank a loss or risk of loss is not required.
The argument is flawed for several reasons. First, it fails on its own terms. The jury in this case was never required to find that the conduct of the defendants “exposed the bank to ... loss.” Maj. Op. at 652. Thus, even if the majority’s legal standard were correct, the jury instructions were insufficient.
Second, I do not read Khorozian in the same way as the majority. In fact, I read Khorozian as being entirely faithful to Thomas. Khorozian simply stands for the proposition that the intent to put the bank at a risk of loss is sufficient to violate the bank fraud statute, even if there was no intent to cause an actual loss. Indeed, we affirmed the jury instructions in Khorozian because they “clearly instructed the jurors that they needed to find specific intent to defraud in order to convict.” 333 F.3d at 508-09. Thus, nothing in Khorozian modified Thomas’ s core holding that, in order to be convicted of bank fraud, a defendant must act with the intent to defraud the bank.
To be sure, Khorozian found Thomas and other cases to be “factually distinguishable because [they] involved fraud on a third party where the bank was merely an ‘unwitting instrumentality’ in the fraud rather than the ‘target of deception.’ ” See 333 F.3d at 505. The majority concludes that this statement modified the mens rea requirement for bank fraud as set forth in Thomas. I disagree. Again, the key issue in Khorozian was whether the intent to cause a risk of loss to the bank was sufficient to convict under the bank fraud statute. The above language from Khorozian simply stands for the proposition that, in cases in which a bank is that “target of deception,” it is perfectly reasonable for a jury to infer the requisite intent absent direct evidence.
It is for this reason that I join the majority’s conclusion that the evidence in this case was sufficient to support a conviction. But whether the evidence is sufficient to justify a conviction (which was the issue in the portion of Khorozian relied on by the majority) is a very different question from whether the jury instructions communicated the proper legal standard. The answer to the latter question is controlled by our decision in Thomas, and I therefore conclude that the District Court’s instructions were in error.
Furthermore, the majority’s reading of Khorozian is clearly foreclosed by Thomas. In Thomas, we held that a bank can be a “target of deception” and still not be a victim of bank fraud if the defendant does not act with the requisite mens rea. As we stated in that decision:
[United States v. Laljie, 184 F.3d 180 (2d Cir.1999) ] illustrates the kind of distinction we make between schemes which victimize banks by exposing them to liability or loss, and schemes in which banks, despite being the target of deception, are mere “unwitting instrumentalities” to the fraud.
315 F.3d at 201 (emphasis added). In the same vein, the Court also stated, Our holding that the statute is to be read conjunctively does not end this matter. We must still decide the thorny question of what is meant by the subsection (1) requirement that the defendant intends to defraud the bank.... The Government also suggests that mere “deceptive conduct” toward the bank establishes intent to defraud. We disagree. 315 F.3d at 199-200.
Again, it is clear that the Thomas Court saw the bank in that case as a “target of deception,” as the defendant deceived the bank as to the purpose of the checks she sought to cash. In fact, Thomas held that,' unless a bank is the target of the scheme, the defendant cannot be convicted of bank fraud at all. See id. at 198 (“[I]n order to prove bank fraud, a bank must be more than a mere incidental player. A defendant must have deliberately targeted his or her scheme at the banking institution.”). So while the majority today holds that a defendant can be convicted of bank fraud if either he targets his scheme at the bank or he acts with intent to cause the bank a loss or risk of loss, Thomas, on which the majority’s analysis purportedly rests, held that the defendant must both target his scheme at the bank and intend to cause the bank a loss or risk of loss.
Thus, the majority’s statement that Khorozian holds that “intent to cause risk of loss” is not required, Maj. Op. at 645, cannot be correct. This view is directly contrary to Thomas’ clear command: the defendant must intend to cause harm or loss to the bank. Thomas, 315 F.3d at 200. If the majority’s reading of Khorozian were correct, then that decision would constitute an impermissible attempt to overrule Thomas, and, under Third Circuit Internal Operating Procedure 9.1, Thomas would remain the law of this Circuit. See O. Hommel Co. v. Ferro Corp., 659 F.2d 340, 354 (3d Cir.1981) (holding that, to the extent a later decision conflicts with an earlier decision, the later decision “must be deemed without effect.”). Thus, if the majority is correct and Khorozian conflicts with Thomas, then Thomas, not Khorozian, would prevail. Either way, the jury must find intent to cause the bank a loss or risk of loss.
The majority’s reliance on United States v. Moran, 312 F.3d 480, 489 (1st Cir.2002), see Maj. Op. at 643, is also misplaced. No matter what a different Circuit has held, the Khorozian panel was bound by our prior decision in Thomas. Moreover, in Moran, the First Circuit stated that the defendant acted “with a clear motive to secure a financial windfall at the bank’s potential expense.” Id. at 491. Thus, Moran does not hold, as the majority suggests, that a defendant need only intend to make misrepresentations to the bank. See Maj. Op. at 643 (citing Khorozian, 333 F.3d at 505). The defendant in Moran did more than make misrepresentations to the bank: he acted with the intent to harm the bank by exposing it to a risk of loss.
C.
Finally, this error was not harmless. See Gov’t of Virgin Islands v. Toto, 529 F.2d 278, 284 (3d Cir.1976) (holding error harmless if “it is highly probable that the error did not contribute to the judgment”). On the record, I cannot find that high probability. I acknowledge that in Neder v. United States, 527 U.S. 1, 18, 119 S.Ct. 1827, 144 L.Ed.2d 35 (1999), the Supreme Court found it was harmless error for the jury instructions to have omitted an element of the criminal offense where the “omitted element is supported by uncontroverted evidence.” That is not this case here. Indeed, at several points the government argued to the jury that the real victims of the defendants’ actions were the banks’ customers.
For these reasons, albeit reluctantly, I would set aside the convictions and remand for a new trial.
II.
The majority rightly acknowledges the dangers inherent in using the standard of “moral uprightness and fairness” to define fraud in a jury instruction. While noting the concerns trenchantly expressed in United States v. Panarella, 277 F.3d 678 (3d Cir.2002), the majority nevertheless upholds the charge in this case because “the instructions, taken as a whole, properly instructed the jury as to the proof required to establish a ‘scheme to defraud’ as well as the appropriate intent to defraud .... The jury could not have convicted the defendants merely for failing to adhere to standards of moral uprightness or fundamental honesty.” Maj. Op. at 647.
In my view, the standard of “moral uprightness” has no place in jury instructions defining fraud, as it broadens the federal fraud statute in a manner that “give[s] inadequate notice of criminality and delegate[s] to the judiciary impermissibly broad authority to delineate the contours of criminal liability.” Panarella, 277 F.3d at 698. Moreover, I am unpersuaded by the fact that Khorozian, 333 F.3d at 508-09, upheld an instruction which defined fraud as “a departure from basic honesty, fair play, and candid dealings.” Khorozian approved of this instruction after viewing the charge as a whole and determining that the instructions were clear that specific intent to defraud must be found to convict.
In affirming the District Court’s reference to moral uprightness, the majority cites to United States v. Dobson, 419 F.3d 231 (3d Cir.2005). See Maj. Op. at 648. In Dobson, a mail fraud case, the District Court instructed the jury that a scheme to defraud under 18 U.S.C. § 1341 is defined as “a departure from fundamental honesty, moral uprightedness, or fair play and candid dealings in the general light of the community.” Id at 239. We reversed the defendant’s conviction, finding that the instructions, taken as a whole, were inadequate. We stated in passing that the reference to moral uprightness was not itself objectionable, but this brief mention of moral uprightness provides virtually no support for the majority’s position because (1) the statement was pure dicta; (2) the panel was applying plain error analysis; (3) the issue was not briefed by the litigants; and (4) the panel mentioned the issue in a passing reference, without any discussion or analysis.
I disagree that the jury instructions were so innocuous in this case. We, of course, do not look to portions of the instructions in isolation, and must consider them in their totality. See United States v. Coyle, 63 F.3d 1239, 1245 (3d Cir.1995). In my view, however, the notion of “moral uprightness” — missing from the instructions used in Khorozian — was central to the definition of fraud in the jury instructions in this case, and thus I fail to see how the remainder of the instructions cures this problem, or how it could be considered harmless error under the applicable high probability standard.
. Although Thomas was not yet decided when this case went to trial, defendants nonetheless properly objected to the District Court's instruction.
. I recognize the seeming perverseness of reversing a conviction because the District Court instructed the jury by reading excerpts from the relevant statute. But Thomas clearly stands for the proposition that the bank fraud statute is not to be given its plain meaning.
. The jury asked, "Can we have the criteria for bank fraud again, please?” The District Court responded, in pertinent part:
To prove a charge of bank fraud, the defendant [sic] must establish each of the following elements beyond a reasonable doubt. First, that the defendants knowingly executed a scheme or artifice to defraud a financial institution, or a scheme or artifice to obtain any of the money owned by or under the custody or control of a financial institution by means of false or fraudulent pretenses, representations, or promises.
Second, that the defendants did so with the intent to defraud....
(emphasis added).
. See 315 F.3d at 195 ("As Thomas admits in her confession, her crime involved a pattern of activity intended to deceive others, including acquiring [her victim’s] trust, making deceptive misrepresentations to her, and some to the bank.”). Moreover, the goal of her scheme was to obtain money from the bank, even though the money in question belonged to her elderly victim.
. It was this language from Thomas that Khorozian relied on in observing that Thomas and other cases were "factually distinguishable because [they] involved fraud on a third party where the bank was merely an 'unwitting instrumentality' in the fraud rather than the 'target of deception.' " 333 F.3d at 505. This statement from Khorozian appears to rests on an erroneous reading of the above language from Thomas. At all events, Khorozian did not change the mens rea requirement for bank fraud, which was clearly set out in Thomas.
. For this reason, any suggestion that we can simply ignore the problematic language in Thomas as dicta is misguided.
. That Thomas held that merely causing a loss or risk of loss is not sufficient is made clear near the end of the opinion:
Moreover, even were there a colorable case for civil liability set forth here, it must also be shown that Thomas intended to victimize the bank. Even a scheme which does expose a bank to a loss must be so intended.
315 F.3d at 202.
. Indeed, in questioning our assertion that Khorozian can be read as being faithful to Thomas, Maj. Op. at 645, the majority comes close to suggesting that Khorozian did overrule Thomas.
. The majority goes so far as to claim that “it is well-established that the 'intent to defraud the bank’ element of § 1344 may be defined as ‘an intent to deceive the bank in order to obtain from it money or other properly.’ " Maj. Op. at 644 n.8. In support of this supposedly "well-established” principle, the majority does not cite a single case that is controlling in this Circuit. See id. (citing United States v. Moran, 312 F.3d 480, 489 (1st Cir.2002); United States v. Brandon, 298 F.3d 307, 311 (4th Cir.2002); United States v. Lamarre, 248 F.3d 642, 649 (7th Cir.2001); United States v. Hanson, 161 F.3d 896, 900 (5th Cir.1998)). What is “well-established” in this Circuit is our decision in Thomas, which held that “a defendant must intend to cause a bank a loss or potential liability, whether by way of 'statutory law, common law, or business practice.'” 315 F.3d at 201 (citation omitted). Given the unusual nature of our holding in Thomas — that a facially disjunctive statute is to be read in the conjunctive — it is not surprising that other courts would disagree. But the fact of their disagreement does not render Thomas any less valid. And none of the cases cited by the majority was decided by a court that reads § 1344 in the conjunctive. See United States v. Kenrick, 221 F.3d 19, 30 (1st Cir.2000) (reading § 1344 in the disjunctive); United States v. Moede, 48 F.3d 238, 241 n. 4 (7th Cir. 1995) (same); Brandon, 298 F.3d at 311 (same); Hanson, 161 F.3d at 900 (same).
. I agree with the majority that the willful blindness instruction was not erroneous as to Leahy, and that, while in error, the instruction was harmless as to Smith. See Maj. Op. at 654 n.15. My only concern with the majority's discussion of this point is that the majority concludes that the instruction — which permitted the jury to infer that the element of knowledge could be inferred based on proof that "a defendant deliberately closed his eyes” — was justified primarily by the behavior of an individual who was not a defendant. The majority approves of the charge because there was evidence that Dominic Conicelli, Sr., the sole shareholder and president of Dantone, Inc., was willfully blind to the conduct of his employees. Conicelli’s knowledge was certainly relevant to the question whether Dantone's employees committed bank fraud "within the scope of their employment” such that the corporation could also be convicted. But the jury instruction referred specifically to a "defendant’s knowledge of a fact.” The jury could reasonably have assumed that the instruction was only intended to apply to the individual defendants. Thus, to the extent it was justified based on Conicelli's conduct, the willful blindness instruction was unnecessarily vague. Nevertheless, I conclude that any error resulting from the instruction was harmless. See Maj. Op. at 654 n.15.

Passage view · GavelSight