Firms Fined for Failing to Monitor Manipulative Trading
Commentary by Glen Barrentine
Two firms, both broker-dealers under a parent company, settled FINRA charges for failing to supervise trading activities to detect and prevent potentially manipulative practices.
FINRA found that the firms relied on third-party automated surveillance systems that were not appropriately configured to capture various forms of manipulative trading activity. These included:
- Narrow Surveillance Parameters. FINRA found the automated systems were set up with overly restrictive criteria that only flagged trades executed within the same account simultaneously or within a narrow window (e.g., within ten seconds) and for the same volume and price. FINRA said that the setup failed to capture a broader range of potentially manipulative behaviors, such as trades that occurred over longer periods or those not involving immediate reversals.
- Lack of Oversight for Specific Securities. FINRA found the surveillance systems excluded certain types of securities from oversight. FINRA described an instance where one of the firms did not include over-the-counter securities in its surveillance, due to not purchasing the required data feed from a third-party vendor. FINRA also found that warrants were not surveilled due to a coding error.
- Unreviewed Alerts. FINRA found the firms failed to review approximately 155 alerts related to potentially manipulative equity trades and about 1,000 alerts for potentially manipulative options trades. FINRA said this oversight resulted in the firms failing to investigate around 700 equity trades and 125,000 options trades flagged as potentially manipulative.
Further, FINRA found that, despite red flags, the firms failed to update their surveillance parameters or review processes to ensure the detection of all potentially manipulative trading.
FINRA concluded that the firms violated FINRA Rule 2010 ("Standards of Commercial Honor and Principles of Trade") and Rule 3110 ("Supervision").
To settle the charges, the firms agreed to (i) a censure, (ii) a $3 million fine (of which $669,000 would be paid to FINRA, with the remainder distributed among other exchanges) and (iii) undertake efforts to remediate the deficiencies identified in the action and certify compliance by senior management within 180 days.
Commentary
Surveillance of potentially manipulative trading by customers is not always easy. Getting the surveillance parameters right requires judgement and can always be second guessed by FINRA.
In this enforcement matter, it appears the firms made things more difficult than they needed to. First, they failed to notice when, because of a coding error, systems stopped monitoring for wash trading in warrants. Anytime regulatory systems are changed, whether as a result of software or upgraded hardware, firms need to test the system to ensure that it continues to function properly.
Second, the firms stopped reviewing alerts for an almost four-year period. While it is not clear why this happened, the firms should have had a process in place to ensure that alerts were being reviewed on a timely basis such that it would have noticed their absence.
Third, the firms did not have a process for reviewing OTC securities for wash trading, prearranged trades, matched trades, or spoofing or layering for over a year. Again, the firms should have had a process to flag the fact that it was no longer conducting these routines.
Fourth, and finally, FINRA criticized the firms' surveillance routines as too narrow and, therefore, unreasonable. It can be difficult to get surveillance routines for manipulative trading right. Set them too narrowly and run the risk of missing actionable behavior. Set them too broadly and risk getting flooded with false positives. While this can be difficult, fortunately, the standard is one of reasonableness rather than perfection. In establishing reasonableness, firms should document the steps they took to arrive at the surveillance routines in use and why they took these steps. These steps should include running the routines with different parameters and documenting the impact that such changes have, especially on the number of false positives. Firms should repeat this process periodically, especially if there are (i) meaningful changes in the firm's customer base, (i.e., an influx of foreign customers, or trading characteristics) and (ii) if the firm is seeing increased evidence of manipulation, especially manipulation that is not being identified by the firm’s surveillance routines. Careful review and documentation would make it harder for FINRA to second guess the firm's judgement.