05/12/2015 03:44 pm ET | Updated May 12, 2016

Legal Confusion as to Spoofing

Five years and a week ago, the market suffered one of its most violent drops and bizarre recoveries. Just a few short weeks ago Navinder Singh Sarao was accused of having helped cause it. More granularly, the Commodities Futures Trading Commission (CFTC) and the Department of Justice (DOJ) allege that Mr. Sarao engaged in "spoofing", a form of manipulation in which a trader places and then cancels large orders, oftentimes in order to profitably manipulate prices. Many have criticized the government for misunderstanding the economics of trading. Unfortunately, confusion as to the law of spoofing persists.

Readers of spoofing stories can be forgiven if they are flummoxed by even very able journalistic accounts. Business Insider's Portia Crowe reports that spoofing was outlawed with the 2010 Dodd-Frank Act, just two paragraphs after reporting on a spoofing case from 2004. So which is it, a recent offense or an existing crime? (In Newsweek, Pulitzer prize-winning journalist Kurt Eichenwald doesn't even mention pre-Dodd-Frank cases when he asserts that Mr. Sarao ought to be acquitted for his pre-Dodd-Frank actions).

Similar questions arise as to the crime's elements. Eichenwald points out inconsistent spoofing definitions between the Sarao complaints and a 2012 complaint filed by the SEC. The New York Time's Professor Peter Henning, generally an unimpeachable source of market wisdom, asserts that high frequency trading, even though it involves many placing and cancelling thousands of orders, is "not spoofing because there is a chance the order will be filled. . . ." CFTC lawyer Cliford Histed explains things differently: "What's the intent?" Histed said. "To not trade? Then it's spoofing."" Who is right? Is it spoofing just in case you intend to cancel your order? Is it not spoofing if there is a chance someone will accept your order? And does it matter whether there was or wasn't an honest-to-goodness intent to manipulate or defraud?

Here is the key to understanding the many vaguely different stories about spoofing: Our laws prohibit spoofing a half dozen times, each time with different elements, and only one time by name. Most crucially, the law for spoofing securities -- stocks and bonds -- has remained unchanged in recent years, while spoofing of bundles of stocks and bonds (such as the S&P 500) was were singled out for especially tough treatment in 2010.

For years, the SEC brought spoofing cases by arguing that spoofing was a special type of fraud or market manipulation. Since fraud and market manipulation are illegal, the argument goes, spoofing is automatically illegal. Proceeding in this way has advantages, like allowing the SEC to pursue spoofing cases in 2004 without need for a new law. But it also requires the SEC to prove the traditional elements of fraud or manipulation, which can sometimes be quite challenging.

For bundles of stock, the approach has been different. They are regulated under the Commodity Exchange Act of 1936 (CEA) -- the same law that governs trading in pork bellies and aluminum. The Dodd-Frank Act adds to the CEA a powerful anti-spoofing rule: "It shall be unlawful for any person to engage in any trading, practice, or conduct on or subject to the rules of a registered entity that is, is of the character of, or is commonly known to the trade as, 'spoofing' (bidding or offering with the intent to cancel the bid or offer before execution)." The plain language of this statute does not require the government to show traditional elements of fraud or manipulation. It does not require, for example, that the defendant intended to trick anyone or to induce anyone into trading when they wouldn't otherwise do so.

Although the CFTC has provided interpretive guidance suggesting that it will tend to pursue only defendants who have some kind of bad intent, the CFTC is clear that it regards this anti-spoofing rule as prohibiting conduct that would not have otherwise counted as fraudulent or manipulative. In many cases, it will be possible to prove intent to cancel a trade when other misconduct could not be proven.

Mr. Sarao traded bets on stock market indices, and so is accused by the Commodities Futures Trading Commission and Department of Justice of violating an entirely different statute than the one enforced by the SEC (and a different portion of the statute than arguably covered spoofing prior to 2010). It is therefore unsurprising that the Sarao complaints define spoofing differently than the SEC and DOJ's prior and contemporaneous efforts to stop stock market spoofing. The differences between the two regimes, once they are recognized, are substantial.

All of this inconsistency and confusion is possible because we regard investments in baskets of stocks as utterly different things from the stocks themselves. We regulate forests differently than trees. This is a distinctly American approach that has often been criticized as a historical accident.

One advantage of the inconsistent treatment of similar assets is experimental data. As Michael Lewis and others have demonstrated, there is great debate right now about whether and when high frequency trading is harmful to markets. Some high frequency traders will fear that their trading strategies are more likely to be misinterpreted as illegal under this new spoofing statute than under the old fraud/manipulation law governing securities. (In Mr. Sarao's case, he was canceling 99 percent of the orders he placed. Many HFT firms have similar cancelation rates). They may take their business elsewhere.

Other traders may trade more freely if spoofing, and forms of HFT that resemble it, is aggressively prosecuted. Both sets of traders are able to vote with their feet. Because substantially similar investments are now available subject to markedly different spoofing rules, traders can take their business where they feel more comfortable. And, in any case, traders should take note that the CFTC is growing more comfortable partnering to comb data and bring cases.

John Sanders & Andrew Verstein