Market Surveillance Ferrets Out Abuse
Trillium Trading has come a long way since 2010, when it was fined by the Financial Industry Regulatory Authority for using high-frequency trading tactics to create a false appearance of buy-side or sell-side pressure. The company has patented a trading surveillance system, called Surveyor, which processes and analyzes order message data from exchanges and compares it to local order flow to detect potentially manipulative events.
“Because I had been on the front lines with Finra in hearing exactly what it was about this pattern of trading that was offensive to them, I was uniquely situated to help build these new controls,” Michael Friedman, chief compliance officer at Trillium and one of the designers of Surveyor, told Markets Media.
At the time of the Finra sanctions, Friedman was a securities litigation attorney at law firm Winston & Strawn, and served as outside counsel to Trillium.
“I met with the Finra lawyers and market analysts who had put together their case against these traders, and it was pretty clear that these guys were up to no good,” he said. “We, the firm, did the right thing and paid a fine and got rid of these traders, and we promised to build new controls to make sure this kind of thing would never happen again.”
Trillium has an in-house software development team that provides trading software and technology as well as builds compliance tools. “I sat down with those guys and we put together a new type of tool that no one had before,” said Friedman.
Trillium is competing with Nasdaq OMX, and other large providers of compliance software for broker dealers, such as SunGard, Fidessa and NICE Actimize, Friedman said. “We are going to firms that are currently serviced by those big fish and saying we can beat those guys at their own game. Give us one day of your data that you’ve already checked using those other tools and we’ll see if we find things that they didn’t find. Almost without exception, we’re able to do that.”
Surveyor uses the comparison to detect “concentration events” – periods of time when a single trader is responsible for a significant percentage of the overall order book for a particular security.
Trillium has found that market manipulation is most likely to occur during concentration events because a trader has disproportionate influence over the market-wide price of a security. “We’re looking for an isolated period of time throughout the trading day when a single trader is responsible for a significant percentage of all of the bids or offers on one side of the market in one stock,” Friedman said.
Trillium’s trading conduct was deemed by Finra to be a “momentum ignition strategy” designed to improperly bait unsuspecting market participants into executing trades at illegitimately high or low prices.
“There’s a senior guy at Finra named Tom Gira [Thomas Gira, Finra’s executive vice president of market regulation] who describes momentum ignition as misrepresenting the true supply and demand in the marketplace,” Friedman said. “You do that by sending in bids and offers that arguably might be at risk to being filled, but the primary purpose of them is to deceive other people in the market as to what’s out there.”
Finra is developing its own Comprehensive Automated Risk Data System (CARDS) to collect and manage data from firms so that it can quickly identify trends and product concentrations that are harmful to investors and take swift action.
“CARDS will provide us with on-going birds eye view surveillance that complements our boots-on-the-ground exams,” said Steven Randich, Finra’s executive vice president and chief information officer, during a speech at the Sifma technology conference last month.
Featured image via viappy/Dollar Photo Club
The increase created a sudden demand for liquid assets that contributed to stress in financial markets.
Initial pricing will generate a net loss for the new exchange on each transaction.
Regulators want to aggregate data across trade repositories.
Bill Street, SSGA's head of EMEA investing, helps clients navigate low rates, negative yields and market volat...
'Big data' can make detecting network breaches simpler, if implemented correctly.