The Bank of Tokyo-Mitsubishi UFJ, Ltd. agreed to pay a fine of US $600,000 to resolve charges related to the alleged spoofing activity of one its unnamed employees – identified as “Trader A” – from at least July 2009 through December 2014. According to the CFTC, on multiple occasions during the relevant time, Trader A engaged in layering activities on one side of a CME Group market, including Treasuries and Eurodollars, in order to facilitate the execution of a resting order on the other side of the market. After becoming aware of these activities, BTMU suspended Trader A; reported the trading to the CFTC; commenced an “expansive internal review;” assisted an investigation by the Commission’s Division of Enforcement; and initiated an overhaul of its system to better detect spoofing. The CFTC charged MTBU with violating the anti-spoofing provision of the Dodd-Frank Wall Street Reform and Consumer Protection Act and a provision of law that makes an employer responsible for the acts of its employees. In the CFTC’s order settling this matter, the CFTC recognized MTBU’s “cooperation” with the DOE.
My View: In a press release accompanying this matter, Jamie McDonald, the CFTC’s recently appointed DOE Director stated, “We expect market participants, through adequate supervision, to prevent this sort of misconduct before it starts.” Hopefully, this is at least a bit of hyperbole. Although it might be possible to review an algorithmic trading system's source code prior to system implementation to detect potential spoofing functionality, there is nothing in the BTMU Order that definitively suggests Trader A utilized an automated trading program. A trading firm could and should have policies and procedures that make it clear to employees that all illicit trading, including spoofing, is prohibited. A trading firm should also conduct regular training to reinforce such measures and, if it proposes to use an automated trading program, consider having an independent person within or outside the organization review the program’s source code to detect any possible layering and spoofing features as practical. In addition, firms should likely have measures reasonably designed to flag problematic trading as soon as possible after it occurs, and to address any issues promptly. However, if a trader only manually enters orders, it would be very difficult in practice for the trader’s employer to prevent rogue conduct before it happens and it should not be held to an unrealistic standard to avoid a substantial sanction. The CFTC should follow the good precedent of CME Group in a disciplinary action against Geneva Trading last October when two of its exchanges opted not to insist on a fine as a condition of settlements where the firm apparently maintained and followed good procedures to try to prevent and detect alleged spoofing by its employees, but spoofing allegedly occurred anyway. (Click here for background on this matter in the article, "Recent CME Group Disciplinary Action Suggests Liability May Not Automatically Mean Fines" in the October 23, 2016 edition of Bridging the Week.)