How High-Frequency Trading Firms Can Rig the Game

The U.S. Securities and Exchange Commission
The Securities and Exchange Commission logo adorns an office door at the SEC headquarters in Washington, June 24, 2011. Jonathan Ernst/Reuters

"Let's make sure we don't kill the golden goose," a manager warned cohorts at New York high-frequency trading firm Athena Capital Research, which had developed a rapid-fire, complex algorithm code-named "Gravy" to fraudulently manipulate the closing prices of tens of thousands of big-name U.S. stocks, such as eBay (EBAY) and Northern Trust Corp. (NTRS).

The internal email followed an automated alert received by the multimillion-dollar fund from the Nasdaq stock market informing it that "suspicious orders or quotes that are potentially intended to manipulate the opening or closing price will be reported immediately."

Thursday the U.S. stock market watchdog the Securities and Exchange Commission stated that's exactly what was happening, slamming Athena with a $1 million penalty, in what it said was its first high-frequency trading manipulation case.

"When high-frequency traders cross the line and engage in fraud, we will pursue them as we do with anyone who manipulates the markets," said SEC Chair Mary Jo White, who told Congress last spring the agency would try to be vigilant in pursuing offenders in high-frequency trading arenas, which have become increasingly challenging as they approach the speed of light.

Athena Capital Research, which specializes in quantitative trading strategies with offices in New York and Hong Kong, was not immediately available to comment.

The fraudulent and manipulative trades targeting stocks with market capitalizations of billions of dollars, according to the SEC, were achieved "using high-powered computers, complex algorithms and rapid-fire trades" to manipulate the "closing prices of tens of thousands of stocks during the final seconds of almost every trading day" from June to December 2009.

According to the SEC, this hurt ordinary Americans and other investors whose mutual funds and retirement portfolios base their net asset valuations on prices at the end of the trading day.

"This is our first case in this area, and there's no question these cases take time," said Andrew J. Ceresney, director of the SEC's Division of Enforcement, explaining why the case was concluded four years after the offense took place during a conference call late Thursday.

While Ceresney would not say whether a whistle-blower assisted in the apprehension of the high-frequency trading firm, he added that the case required the exhaustive and lengthy analysis of "the actual algorithmic code," which, according to another email circulated at Athena, allowed the firm to engage in "punching the stock."

During the conference call, the SEC declined to disclose the estimated profit reaped by Athena from the Gravy algorithm. However, a close read of the order against the firm shows the daily "gravy" could come to several thousand dollars a day—which means, from June through December, the total could have easily reached the size of the penalty paid by Athena Capital Research to the SEC, which said Thursday it sought to "maximize" the fine.

A breakdown of Athena Capital Research's profits from Gravy, according to one internal email released by the SEC, stated: "PM Gravy made 5.3K, trading on 33 [stock] symbols, biggest dollar move NTRS $.12 (.15 percent), percentage move PCBC $.06 (.41 percent)."

An Athena manager (who was out of the office on vacation) replied, "Looks like we have some Mach chips….going to Vegas tonight."

No individuals were charged, according to the SEC, which didn't name the managers behind the emails it cited in a release Thursday. "We determined it was appropriate to bring charges against the firm instead of individuals," Ceresney said, adding that the SEC is pursuing other high-frequency cases based on the agency's "increased use of quantitative data and analytical tools identifying patterns of trading."

The case brought against Athena gave a rare glimpse into how bad actors in high-frequency trading shops can artificially inflate or depress prices in order to turn an enhanced profit.

In the case of Gravy, the manipulative scheme zeroed in on order imbalances at the end of the trading day, when orders to buy or sell shares exceed demand to fill those orders and Nasdaq holds an auction to fill them at the best price. Athena would accumulate shares and dominate the market during the last few seconds of the trading day to manipulate prices, the SEC said, accounting for 70 percent of total Nasdaq trading volume of the affected stocks.

"Make sure we always do our gravy with enough size," a manager told Athena's staff after analyzing trading in which Gravy accumulated only approximately 25 percent of its targets, and, thus, had no price impact on the stock.

Over time, Athena's algorithmic strategies became more focused on ensuring the firm, although small, was the dominant one in the market—and sometimes the only one—trading desirable stock imbalances at the end of each day.

Through aggressive pricing, Athena "refined Gravy to reduce risk and increase profit, eventually resulting in Athena's imbalance-on-close orders being at least partially filled more than 98 percent of the time," the SEC said. In addition, Athena's high-frequency trading scheme enabled its orders to be executed at more favorable prices.

"These enhancements paved the way for Athena to push the price in the direction Athena wanted," said Ceresney, noting that, at least for a time, the firm was able to use "sophisticated, fraudulent trading strategies" to create "a virtual monopoly over favored stocks."

The SEC declined to give a full list of the stocks affected.