CFTC Continues to (Wrongfully) Target Small Overseas “Spoofers”

Financial crime specialists Rahman Ravelli details the Commodity Futures Trading Commission’s current hardline approach to allegations of spoofing. Despite the recent failure of two defendants to have a spoofing indictment dismissed, he believes there are strong grounds for challenging CFTC allegations.

Introduction

On September 29, 2020, the Commodity Futures Trading Commission (CFTC) announced a record-setting $920 million settlement with JP Morgan for alleged spoofing in U.S. precious metals and Treasury futures markets. That same day, it filed an anti-spoofing enforcement action against decidedly smaller targets: Roman Banoczay, Jr. and Sr., a father-son trading duo based in Bratislava, Slovakia.¹ While the JP Morgan settlement shows the continued reluctance of large financial institutions to challenge spoofing allegations in court, the Banoczays and other small-scale traders have significantly greater incentives to put up a fight.  

For one, unlike in the JP Morgan case, the action against the Banoczays seeks a broad injunction that would permanently preclude them (and anyone working for them) from ever again trading futures on U.S. markets.² In other words, the CFTC is effectively trying to put the Banoczays and their firm—Bazur Spol. S.R.O.—out of business for good. To the extent the Banoczays wish to continue participating in U.S. futures markets—where the vast majority of futures trading takes place globally—simply accepting the allegations or letting the case lapse into a default judgment is not a viable option.  

Second, on top of the injunction the CFTC is seeking a large financial penalty. While the specific amount regulators want is unspecified in the complaint, it will no doubt be significant: The CFTC asked the court to order disgorgement of at least $332,000, a penalty of up to $1 million or more, plus costs and fees related to the litigation.³ Whereas the penalty against JP Morgan will hardly put a dent in its overall revenue (which last year totaled more than $115 billion),⁴ a million+ dollar judgment against a small firm like Bazur will be crippling, if not fatal. And further, there is no guarantee that a financial resolution with the CFTC will be the end the matter (as Michael Coscia discovered in 2014 when he was indicted on spoofing charges after settling with the CFTC in connection with the same conduct a year earlier).⁵

Third, the Banoczays can confront the CFTC’s case without having to travel to the U.S. In contrast to criminal spoofing cases, which generally require a defendant to physically submit to U.S. jurisdiction before mounting a defense,⁶ civil actions are much more easily opposed while the defendants remain in the (relative) safety of their home countries. Even though there is no indication that criminal charges have been, or will be, filed against the Banoczays, given the Department of Justice’s recent spate of spoofing cases against individual traders, that possibility cannot be completely discounted. Thus, despite the existence of a U.S./Slovakia extradition treaty, prudence would advise against voluntary travel to the U.S. until the likelihood of a parallel criminal case is further clarified.  

A Defense Blueprint 

Per the above, the cost/benefit analysis for individuals like the Banoczays who are facing spoofing charges is different from that of a big bank like JP Morgan. Since the reasons such individuals might choose to contest the CFTC’s allegations seem clear, the next consideration is what defense strategy is most likely to succeed. As in most civil cases, such a strategy should generally begin (and hopefully end) with a carefully crafted motion to dismiss. But defendants in both civil and criminal spoofing cases have been, to date, universally unsuccessful in dismissing the allegations against them.⁷

So how might the Banoczays’ succeed where others failed? One approach could be attacking the Commodity Exchange Act’s anti-spoofing provision (ASP)—which prohibits “bidding or offering with the intent to cancel the bid or offer before execution”—as unconstitutionally vague.⁸

Concededly, this tactic has been tried before. Just last May, for example, Judge Lee of the Northern District of Illinois denied a motion to dismiss a spoofing indictment on vagueness grounds filed by two former precious metals traders in United States v. Bases.⁹ Judge Lee’s reasoning appeared to be underpinned in large part by the Seventh Circuit’s decision in the Coscia case.¹⁰ There, the court highlighted the importance of the ASP’s specific intent requirement: “when the government must prove intent and knowledge, these requirements . . . do much to destroy any force in the argument that application of the [statute] would be so unfair that it must be held invalid.”¹¹ The court then went on to conclude that the ASP was constitutional as applied to Coscia’s conduct: “The text of the [ASP] requires that an individual place orders with the ‘intent to cancel the bid or offer before execution.’ This phrase imposes clear restrictions on whom a prosecutor can charge with spoofing; prosecutors can charge only a person whom they believe a jury will find possessed the requisite specific intent to cancel orders at the time they were placed.”¹²

While the Coscia decision remains good law in the Seventh Circuit, the fact pattern it analyzed was unique. Unlike the Banoczays and the traders in Bases—who placed their orders manually—Coscia employed a computer trading algorithm. Coscia’s use of an algorithm, which could place and cancel orders in under 5 milliseconds, was critically important to the Seventh Circuit’s determination that the ASP was constitutional: 

Mr. Coscia cannot claim that an impermissibly vague statute has resulted in arbitrary enforcement because his conduct falls well within the provision's prohibited conduct: he commissioned a program designed to pump or deflate the market through the use of large orders that were specifically designed to be cancelled if they ever risked actually being filled. . . .  [This] clearly indicate[s] an intent to cancel.¹³

But the same cannot be said of traders who do not employ algorithms. To the contrary, there are insurmountable conceptual difficulties when the ASP is applied to relatively slow, manual trading strategies.

The ASP’s Intent Element Does Not Limit Who Can Be Charged With Spoofing

Implicit in the Seventh Circuit’s analysis in Coscia is the belief that the ASP applies only to a limited subset of orders. Since the ASP is concerned with cancellations, it seems obvious that the statute targets only those orders that will not execute immediately—i.e., “resting” limit orders that do not aggress across the order book’s bid/ask spread.¹⁴ But since resting orders make up the majority of orders entered, the critical question becomes: what specific type of resting order does the ASP purport to criminalize?  

On its face, the ASP provides a seemingly straightforward answer: it only prohibits resting orders that are entered with an intent to cancel them “before execution.”¹⁵ But when this language is applied to manual traders, the parameters of what exactly it prohibits is far from clear.  

To begin with, an order cannot be cancelled after execution. Rather, once an order executes (i.e. is filled), it is instantly removed from the order book and becomes an executed trade.  While trades can be unwound in certain limited situations, it is impossible for an order to be cancelled after it is filled (because it then becomes an executed trade).¹⁶ To use a simple analogy, a marriage proposal cannot be retracted after the wedding ceremony—it is possible to unwind the marriage via divorce or annulment, but once the marriage occurs, there is no longer any proposal to “cancel.”

The ASP’s use of the phrase “before execution” is therefore meaningless. Since an order cannot, under any circumstances, be cancelled after it executes, prohibiting orders placed with an intent to cancel “before execution” is the same as a ban on orders placed with an “intent to cancel,” full stop.  

To return to the same analogy, one could imagine a statute criminalizing “making a marriage proposal with the intent to retract it.” Adding the phrase “before marriage” to this statute would have no impact on the types of proposals it makes illegal. Rather, both before and after the addition, the statute criminalizes the exact same thing: making a proposal with the intent to retract it. In the same way, the phrase “before execution” can be removed from the ASP without affecting the statute’s meaning—“bidding or offering with the intent to cancel the bid or offer before execution” is exactly equivalent to “bidding or offering with the intent to cancel.”  

So why does this matter? It matters because every resting order must necessarily be entered with an intent to cancel it if it is not filled by another trader. To see why, imagine being inside the mind of a trader about to place a resting order into the order book. With respect to how long she intends to leave it open for, she only has two possible options: either (A) she intends to leave it open indefinitely or (B) she doesn’t.  But Scenario A never happens, since futures traders operate in short timeframes that at the longest extend to the end of the trading day. This means that every resting order must fall into Scenario B (as seen in Figure 1).

Following this logic, only one of two things can happen to a resting order that is not intended to be left open indefinitely: (1) it is filled by another trader, or (2) it is cancelled at some point.  

It is important to recognize, however, that what happens to a manually placed resting order is not fully within the control of the trader who placed it. To be sure, a resting order placed close to the top of the order book and/or left open for a significant duration stands a higher chance of being filled than one placed further away and/or kept open for just a short time (all else being equal).  

But in either case, the trader placing the order cannot be certain about what will happen to it.¹⁷ This is not necessarily true of traders—like Coscia—who employ algorithms, however.  Because they can pre-program their orders to cancel before any potential counterparty has a realistic chance of filling them, such traders can ensure that their “spoof” orders are effectively un-fillable.¹⁸ Insofar as algorithmic traders are concerned, therefore, it does make conceptual sense to conclude that their orders were placed with the “intent to cancel” because they can be designed to effectively guarantee that result.  

In contrast, with respect to a manual trader’s intent (the blue box in Figure 1), it must be the case that when she first places the order, she does so with an intent to cancel it only if it is not filled by another trader (an occurrence that is outside her control). There is no other possibility.  

Judge Lee’s Flawed Opinion

This essential point was made by defense counsel in Bases in support of their motion to dismiss: “[there is no reason] why a trader cannot place an order with both the intent to cancel in the future and a willingness to trade in the meantime.”¹⁹

But Judge Lee disagreed and determined that the orders alleged in the indictment were placed “with the intent to not fill them—that is, with no ‘willingness to trade [them] in the meantime.”²⁰ To support this conclusion, Judge Lee cited the Seventh Circuit’s opinion in Coscia, which stated: “The fundamental difference is that legal trades are cancelled only following a condition subsequent to placing the order, whereas orders placed in a spoofing scheme are never intended to be filled at.”²¹

But going back to Figure 1, we see that this ignores the unique realities of manual trading. Specifically, it ignores the fact that all manual cancellations must follow “a condition subsequent to placing the order”—specifically, inaction on the part of other market participants to fill it.  To put it another way, the absence of a market reaction to an order is in itself “a condition subsequent” upon which a cancellation can (and must) depend.  

Accordingly, a manual trader’s intent when placing a resting order must be to see if it is filled, and if it is not, to cancel it (i) at some point in time or (ii) upon some other event.  And since a resting order depends on action by other traders to fill, Judge Lee’s contention that the alleged spoof orders were not “intended to be filled” makes no sense—a manual trader may intend to cancel an unfilled order following a relatively short period of time or upon an event that is likely to occur, but whether the order executes is fundamentally beyond her control.  

To use another analogy, a manual trader placing a resting order is like a fisherman dropping a hook in the water. Regardless of whether he intends to catch a fish or not, he cannot himself manufacture either result. He can leave the hook in for longer or shorter time periods, but whether a fish takes the bait is ultimately up to the fish. So unless he is willing to wait forever to get a bite, he must necessarily intend to wind up the reel at some point in time.  

In the same way, unless a trader is willing to leave a resting order open indefinitely, there will always be an intention to cancel in the trader’s mind at the time the order is placed.  

Unconstitutional Vagueness

The reason this makes the ASP unconstitutionally vague as applied to manual traders is that neither the statute itself nor the CFTC’s interpretative guidance²² provide any delineation between lawful and unlawful intentions to cancel. See Welch v. United States, 136 S. Ct. 1257, 1262 (2016) (the Constitution “prohibits the government from imposing sanctions under a criminal law so vague that it fails to give ordinary people fair notice of the conduct it punishes, or so standardless that it invites arbitrary enforcement”).

With respect to the duration an order is intended to be left open, for example, the ASP fails to give traders any indication how long they must intend to leave it open for so as to not break the law. For instance, if a trader places an order with the intent to cancel it if it remains unfilled after five minutes, has she “spoofed”? What about 10 seconds? Or 1? There is simply no way for her to know.

Indeed, the arbitrary approach taken by regulators to spoofing cases is clearly seen when the allegations against Coscia are compared with those made against the Banoczays. As to Coscia, prosecutors used evidence that his “spoofs” were generally left open for under half a second to show that he never really intended to trade them.²³ But in the Banoczay complaint, the CFTC alleges a “median cancellation” time of 6.2 seconds—nearly 12.5 times longer than Coscia’s—in support of the same point. If, in the regulators’ mind, the timeframe that characterizes “spoofs” can stretch from under 0.5 seconds to over 6, there is no reasoned basis for them not to extend it further still—to say, 12.5 times 6 seconds (or well over a minute).  

In the same way, the ASP gives zero guidance as to what events are permissible conditions on which an order cancellation may depend.²⁴ As pointed out by others, there are numerous reasons why a trader might expect to cancel an unfilled resting order. For example, an order could be placed with an intent to cancel it if the trader changes her mind, the market moves, a relevant news event occurs, she needs to leave the desk to attend a meeting, and so on.  

Regardless of whether the intention is to cancel an unfilled order after a certain period of time or upon a certain event, the critical point is that this intention must be in a manual trader’s mind at the time the order is placed (explicitly or implicitly). There is no way it could not be.

As a result, every resting order that is not intended to be left open indefinitely violates the plain terms of the ASP.²⁵ This then gives law enforcement free reign to capriciously draw the line between legal and illegal trading. This is the definition of an unconstitutional law.²⁶

Conclusion

Judge Lee is the only district court judge to rule that the ASP is constitutional as applied to manual traders and, in light of the above discussion, there are strong grounds to argue that his reasoning should be rejected by other district court judges and the Seventh Circuit.²⁷ In any event, should U.S. regulators continue to go after small overseas “spoofers” like the Banoczays, those targeted should not shy away from challenging them given the ASP’s inherent flaws.

1. C.F.T.C. v. Banoczay, Jr. et al., 20-cv-05777, Dkt. #1 (Sept. 29, 2020 N.D. Ill.). 
2. Id. at 29.
3. Id. at 30.
4. Indeed, although JP Morgan’s stock price dipped slightly on the day its settlement with the CFTC was announced, it has since jumped nearly 5%.  
5. See Walter Pavlo, “After Conviction on ‘Spoofing,’ Defendant Questions His Previous Counsel’s Representation,” Forbes (Nov. 20, 2019) (discussing the Coscia case).
6. See, e.g., United States v. Hayes, 118 F. Supp. 3d 620, 624-26 (S.D.N.Y. 2015) (explaining the contours of the fugitive disentitlement doctrine).
7. The one possible exception is United States v. Radley, 659 F. Supp. 2d 803 (S.D. Tex. 2009), where criminal charges involving spoofing-type trading activity were dismissed in a decision subsequently affirmed by the Fifth Circuit.  Note, however, that Radley did not implicate the ASP, since it occurred before the ASP was enacted. 
8. 7 U.S.C. § 6c(a)(5)
9. United States v. Bases, 18-cr-00048, Dkt. #301 (N.D. Ill. May 20, 2020).
10. United States v. Coscia, 866 F.3d 782 (7th Cir. 2017).
11. Id. at 794 (quoting United States v. Calimlim, 538 F.3d 706, 711 (7th Cir. 2008)).
12. Id.
13. Id. (emphasis in original). 
14. See Banoczay, 20-cv-05777, Dkt. #1 at ¶¶ 27-28 (explaining order book mechanics). 
15. 7 U.S.C. § 6c(a)(5); see also Coscia, 866 F.3d at 792 (concluding that the ASP’s parenthetical “clearly defines ‘spoofing’”).
16. See Banoczay, 20-cv-05777, Dkt. #1 at ¶ 27; Brief of Amicus Curiae Futures Industry Assoc., Bases, 18-cr-00048, Dkt. #158 at 6 (“When an order has been accepted by the exchange system and matched with another order, the trade is final; it cannot be cancelled.”). 
17. Because the order book constantly changes (often many times a second), an order placed at the current best price could be far away from the best price a split-second later.  By the same token, an order placed well away from the best price could be quickly filled by a large market order that sweeps the order book.
18. For example, prosecutors presented evidence in the Coscia trial showing that only 0.5% of his alleged “spoofs” were filled.  Coscia, 866 F.3d at 789.  
19. Bases, 18-cr-00048, Dkt. #144 at 15 (emphasis in original).  
20. Id. at 16.
21. 866 F.3d at 795.
22. Antidisruptive Practices Authority, Commodity Exchange Act Release No. 3038-AD96 (May 20, 2013).
23. Coscia, 866 F.3d at 789-90. 
24. See, e.g., Brief for Amici Curiae Jerry Markham and Ronald Filler (“Markham/Filler Brief”), Coscia v. United States, No. 17-1099 (Mar. 8, 2018) (“every [trader] intends to cancel trades before their execution for a broad range of reasons”).  
25. See Coscia, 866 F.3d at 797 (“[Coscia’s] scheme was deceitful because, at the time he placed the [spoof] orders, he intended to cancel the orders.”).
26. See, e.g., F.C.C. v. Fox Television States, Inc., 132 S. Ct. 2307, 2317 (2012). (“A fundamental principle in our legal system is that laws which regulate persons or entities must give fair notice of conduct that is forbidden or required.”).
27. Note that the constitutionality of the ASP was briefed in a civil case against a manual trader (CFTC v. Flotron, 18-cv-00158, Dkt. #27 (D. Conn.)), but the parties settled before the court had an opportunity to consider the issue.