Argument preview: No April Fools – Even after thirty years, the reach of the federal bank fraud statute is hard to predict
on Mar 30, 2014 at 9:01 pm
Tuesday’s arguments in Loughrin v. United States will present a persistent and important — but highly technical — circuit split about the details of permissible federal bank fraud prosecutions. Although the “plain language” of the relevant statute seems clear, the outcome is not. Indeed, the surprising thing is that even thirty years after the bank fraud statute was enacted, the extent of its reach is still very much in dispute. The initial questions from “plain language” adherents like Justice Antonin Scalia are likely to signal the outcome – but even then, the precise contours of future jury instructions will depend on the precision of the opinion writer and the willingness of agreeing Justices to pay attention in the flood of end-of-Term opinion work that remains this year.
The statute, the facts, the rulings and the Question Presented
18 U.S.C. § 1344, enacted in 1984 to correct perceived “gaps” in federal bank fraud prosecutions, is broad and seems straightforward. Congress imposed a sentence of up to 30 years in prison for anyone who:
“[K]nowingly 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.”
The text and structure of this statute appear to create two possible offenses, both dependent on “knowingly execut[ing] or attempt[ing] to execute a scheme.” Clause (1) punishes a “scheme to defraud a financial institution.” Clause (2) punishes a “scheme to obtain [by false representations] any of the moneys … owned by, or under the custody or control of,” such an institution, even if the institution itself was not the victim or target of the falsity.
Kevin Loughrin was convicted under clause (2). His was a simple scheme: he would steal checks from people’s mailboxes, alter the payee to “Target” (or some other similar retailer), forge the account holder’s signature if necessary, and then go buy merchandise at that retailer with the altered checks (often then “returning” that merchandise for cash). After doing this a number of times, Loughrin was unsurprisingly identified on security camera footage by Target’s “loss prevention” staff, and arrested on his next visit. (Why the case was prosecuted by federal, rather than state and local officials, is – as is so often the case — a mystery unanswered by the record. But just as in Bond v. United States, argued earlier in the Term, the fateful decision to “go federal” forces legal issues that otherwise would have laid dormant.)
At Loughrin’s jury trial, the district court opined that the government had failed to prove a “risk of loss,” and finding that Tenth Circuit precedent required such a risk under Clause (1), the court restricted the prosecution to Clause (2). The district court then rejected Loughrin’s requests to instruct the jury that Clause (2) requires an “intent to defraud a financial institution” or, alternatively, an intent to cause “financial loss to a financial institution.” Loughrin wanted to argue that he had intended to defraud (if anyone) only Target, rather than the bank on which the checks were drawn; and that in fact there was no “risk of loss” to the bank (only to Target). But, given the court’s jury instructions, the prosecutor was able to argue that Loughrin needed to have only an “intent to defraud” someone – “he did not have to intend to defraud the bank.” And who ultimately might bear the loss was irrelevant, so long as money “in the custody of” the bank was at risk.
On appeal, the Tenth Circuit concluded that Section 1344 creates “separate offenses” in its two Clauses, and that Clause (2) does not require intent to defraud a bank, but merely “an intent to defraud someone” in order to “obtain money from a bank.” Risk of loss to the bank is not required, merely a risk of obtaining such money. The circuits are divided on these points, and the court granted Loughrin’s petition for certiorari to review “whether the Government must prove that the defendant intended to defraud a bank and expose it to risk in every [Section 1344(2)] prosecution.” (Some Justices may ask whether all the arguments are “fairly included” within this question, but that won’t be the main event.)
Arguments flowing from the text versus the reach of a criminal statute
Pretty obviously, what the government has going for it is the plain language of Clause (2), which doesn’t require an intent to defraud “a bank” specifically. Indeed, Clause (2)’s failure to name the particular target of the fraud would seem to be what distinguishes it from Clause (1). And Clause (2) doesn’t appear to require any “risk of loss to the bank,” as opposed to a scheme to obtain money under the bank’s “custody or control.” Thus, the government argues that Clause (2)’s requirements are met perfectly when someone knowingly “executes … a scheme … to obtain [such] moneys” – here, the money in the bank accounts of the checks’ true owners – “by means of false … representations” – here, the altered payee names and forged signatures.
How then, you might ask, could the courts of appeals “split” over the meaning of Clause (2) and insert non-textual requirements like “intent to defraud a bank” and “risk of loss”?
The answer flows from the Supreme Court’s longstanding apparent hostility to the broad federal mail fraud statute, and repeatedly expressed concerns that statutory interpretation ought not “federalize” broad areas of criminal law without clear congressional statements. As Loughrin’s reply brief artfully notes, the Tenth Circuit’s broad reading would “reach every case in which a defendant purchases groceries or gasoline with a check the defendant knows will bounce,” and “turn the bank fraud statute into a federal ‘bad check’ law.”
Why mail fraud precedents are relevant
The parties agree that the 1984 bank fraud statute was modeled on the much earlier federal mail fraud statute. Like Section 1344, that statute (18 U.S.C. § 1341) has two clauses (albeit not marked off by subsection numbers): it criminalizes schemes “to defraud or for obtaining money or property by means of false pretenses” (when the mails or, now, interstate delivery services are used). When the bank fraud statute was enacted in 1984, the courts of appeals had unanimously interpreted these two clauses to mean slightly different offenses. But three years later, in McNally v. United States, the Court ruled that Section 1341’s two clauses describe basically the same, single offense: frauds to obtain money or property. That holding was undoubtedly non-textual (and many, including Justice Stevens in dissent, thought it was flat wrong). But it was reaffirmed thirteen years later in Cleveland v. United States. And in 2010 in Skilling v. United States (the Enron case), the Court again narrowed the mail fraud statute non-textually (and three dissenting Justices, led by Justice Scalia, went further to say that the statute’s “honest services” language was unconstitutionally vague).
The government argues that the two-clause structure of Section 1344 should be governed not just by its plain language but also by the judicial reading given to the similar two-clause mail fraud statute in 1984, to which Congress referred when enacting Section 1344. By contrast, Loughrin argues that McNally’s later “one offense” reading should control.
The Court’s repeated concerns about broad criminal statutes.
Loughrin’s argument is bolstered by two concerns repeatedly stressed by the Court in mail fraud (and other criminal) cases: the judiciary ought not presume that Congress intended to broadly federalize large areas of criminal law without a clear statement; and when there is truly doubt about the meaning of a criminal statute, the “rule of lenity” directs that the construction that more narrowly imposes criminal liability be adopted. The Solicitor General responds that there is no ambiguity in the text here, and that Congress in fact wanted to “reach a wide range of fraudulent activity” (quoting from the Senate Report) and protect assets in the “custody” of federal financial institutions, even if the ultimate “loss” might be borne by the account-holder.
Interestingly, the Solicitor General himself gives some credence to the “overbroad federalization” concern here, by suggesting that the Court could “alleviate concerns about Clause (2)’s breadth” by reading the statute as requiring false or fraudulent means that “naturally or foreseeably or inherently” are directed toward a bank and would have “the potential to influence” it. While this reading would be consistent with some cases, and arguably flow from the statute’s textual focus on property “under the custody or control of” a bank, Loughrin notes that “the Government shows no sense of irony” in arguing for its own non-textual limiting principle.
In the end, this case is so technical and full of nuance that only the most dedicated legal technicians are likely to be awake at the end of sixty minutes of argument. Justice Ginsburg wrote the Court’s opinions in both Cleveland and Skilling, so her first questions at argument may show a majority’s attitude. Justice Scalia, of course, has expressed views that might be said to pull in opposite directions: adherence to “plain language” statutory text on one hand, but a dislike of vague over-federalization of crime on the other. (Bond will likely provide strong guidance on this last concern.) Despite the clarity with which Congress’s plain language in Clause (2) seems to speak, it does indeed allow applications far afield from actual risks to federal institutions. Thus it would not be surprising to see the Court interpret the statute to narrow its potential reach.
But the reality remains that once you decide to leave the statute’s “plain language,” it is far from clear how exactly to fashion the precise limiting principle (or principles) – and from whence specifically to draw them. The circuits are not split just on one side or the other of Section 1344, but in multiple directions and each with their own nuances. (Judge Gerald Lynch of the Second Circuit wrote at length about this in his 2012 concurrence in United States v. Nkansah.) Now the Court will either have to allow the plain language to operate in all its (possibly intended) breadth – or determine how precisely to limit its reach. (And no matter what the result, jury instructions in federal bank fraud prosecutions will have to be rewritten.) The Court faced the same problem in Skilling, and the majority’s answer was not satisfying to Justices Scalia, Kennedy, and Thomas in dissent. Perhaps the addition of Justice Kagan will make some difference here – or perhaps not.
[Disclosure: Kevin Russell of Goldstein & Russell, P.C., whose attorneys contribute to this blog in various capacities, is among the counsel to the petitioners on this case. However, the author of this post is neither affiliated with the firm nor otherwise involved in the case.]