Wash Trading by High-Frequency Firms May Face Scrutiny

High-frequency trading firms are drawing scrutiny from U.S. regulators seeking evidence that they may be distorting market prices by conducting transactions with themselves, said two people with knowledge of the matter.

So-called wash trades, in which a party buys a contract from itself, could be executed inadvertently by firms with multiple algorithms active in the same stock or derivative, said the people, who requested anonymity because the review isn’t public. Such trades, which can alter the price of shares if they are executed above or below market rates, would be illegal if deemed intentional efforts to manipulate stocks.

The Securities and Exchange Commission and Commodity Futures Trading Commission have sharpened their focus on high- frequency and algorithmic trading since May 6, 2010, when about $862 billion was erased from stock values in 20 minutes before share prices recovered from the plunge. Regulators have expressed concern that some firms and electronic exchanges don’t have sufficient controls to prevent a range of events -- from improper trades to programming glitches -- that could roil markets even when there is no wrongdoing.

High-frequency trading, in which computer algorithms are used to buy and sell stocks in fractions of a second, accounts for more than half of equity trading volume. Getco LLC and Citadel LLC, both based in Chicago, and New York-based Virtu Financial LLC are among the biggest automated-trading firms.

Exchange operators including Nasdaq OMX Group Inc. (NDAQ) and NYSE Euronext (NYX) have started services to help firms avoid accidental wash trades.

‘Undesirable Executions’

Bats Global Markets Inc. updated a service on its two exchanges last month to help users avoid “undesirable executions against themselves,” the Lenexa, Kansas-based exchange operator told the SEC. Direct Edge Holdings LLC began a similar service on two exchanges in 2010 to prevent “the potential for (or the appearance of) ‘wash sales’ that may occur as a result of the velocity of trading in today’s high-speed marketplace,” according to a filing with the SEC.

“Regulators cannot assume that algorithms in the markets are always well-designed, tested and supervised,” CFTC Chairman Gary Gensler said at a June 20 meeting of the agency’s technical advisory committee. “To give hedgers and investors the confidence in markets that they really need and deserve, I think regulators always need to adapt.”

Broad Definition

The CFTC has been considering issuing a so-called concept release, a step prior to a formal rulemaking, which could lead to new testing, supervision and oversight requirements for high- frequency and automated trading. At a meeting of a CFTC advisory committee on June 20, representatives from Getco, NYSE Euronext and Deutsche Bank AG (DBK) suggested that regulators adopt a broad definition of high-frequency trading to limit the potential for regulatory arbitrage.

“We wanted to keep it easy to interpret and difficult to game,” Deutsche Bank’s Greg Wood said at the meeting. “We deliberately did not want to define types of high-frequency trading strategies.”

Requiring registration and audits of automated trading algorithms would be a waste of regulators’ resources because of the cost and complexity of establishing unique identifiers, a working group of the CFTC advisory committee said in a summary of its findings presented at the meeting.

“Market abuse is not fundamentally a function of the means, speed or frequency of order entry and transactions,” according to the summary. “Focus should be on specific behaviors that undermine market integrity irrespective of the means or pace of order entry.”

To contact the reporters on this story: Joshua Gallu in Washington at jgallu@bloomberg.net; Silla Brush in Washington at sbrush@bloomberg.net

To contact the editor responsible for this story: Maura Reynolds at mreynolds34@bloomberg.net

Bloomberg reserves the right to edit or remove comments but is under no obligation to do so, or to explain individual moderation decisions.

Please enable JavaScript to view the comments powered by Disqus.