Nasdaq’s approach to trade surveillance trains its sights on traders in addition to events. Because if individual performance seems too good to be true, maybe it is.
Traditional trade surveillance is based on the generation of alerts that are triggered when a series of specific events breach designated rules. They are focused on particular markets, instruments or contracts. Nasdaq, however, has a wholly new approach, which compliments traditional rules-based surveillance.
The firm’s new module, called Lens, is part of its flagship SMARTS Trade Surveillance product, and was launched in January of this year after beta-testing in the second half of 2017 with three major global investment banks and one high-frequency trading firm. It now has seven clients, with another three in the pipeline, and is used in both the first and second lines of defence.
Spotting their tells
Underpinning Lens is a novel ‘follow the money’ approach to trade surveillance. In other words, if the end result of a period of trading seems too good to be true and lies beyond the expected trading context, even if no rules have been breached, Lens will provide a wealth of detailed data about the trader in question.
“We decided to start focusing on the individual,” says Michael O’Brien, global head of product management for Nasdaq’s risk and surveillance solutions. “We look at how they normally trade, and their attributes, and we track where that behaviour changes in a significant way. Certain types of behaviours align themselves with particular types of abuse like insider trading or manipulation.”
Lens is not designed to replace event-based surveillance, but to sit alongside it and offer a further and different line of enquiry for risk control professionals. The problem with event-based surveillance is that while it is very good at spotting key occurrences usually found in market abuse, good actors are also skilled at finding ways of achieving their nefarious ends by different means.
They will come up with methodologies that won’t necessarily be picked up. For example, they might be trading in such small amounts that it slips under the radar of traditional surveillance.
Nasdaq stresses that there is nothing wrong or outrageous about dealers making money – that’s what they are there to do after all – but Lens does shed light when that outcome is allied to certain types of behaviour.
As Saker Asllan, principal product manager for Nasdaq’s risk and surveillance solutions, explains: “For example, with insider dealing on the basis of positive news, you will observe traders that will just buy, never sell. They’re very directional, and they rarely cancel orders. Their orders are aggressive, crossing the spread, and performance is obviously very good. So you track this behaviour alongside their historical behaviour and see if it is unusual.”
Using Lens effectively requires a new way of working for banks. Traditional event-based surveillance will generate alerts that clearly need to be examined. There are no alerts with this module – instead, tailored visualizations are created based on a vast wealth of information to easily identify outliers, which can then be further analyzed.
For example, control professionals at one firm using Lens have decided to spend the last half hour of every day reviewing the Lens outliers looking for red flags. As O’Brien says, this is not a job to give to a junior analyst. “You need someone experienced, someone who knows the market and knows what looks unusual. Someone who also has a sense of curiosity so that if they see something that looks a bit odd they will follow the breadcrumbs,” he says.
But the proof of the pudding is in the eating. Another existing customer reports that it has launched one suspicious transactions report solely as a result of trading identified within the Lens module, which wasn’t identified by traditional alerts and which, in turn, led to a report to the regulator. This is what Nasdaq wants to achieve.
The rich data that Lens collates surrounding an individual trader, including details like average trade size, usual market sectors, when they usually trade, when they enter the order book, and whether they trade at the best bid or ask, generates a 360-degree portrait. In it is contained all trades across all markets across all asset classes.
Thus, if traditional event-based alerts are generated, a compliance officer can turn to this portrait immediately, without spending hours collating additional information. In this way, Lens can operate hand in hand with existing surveillance methodology.
A glimpse into the future
At the moment, Lens is sold as a subscription-based package in addition to SMARTS Surveillance, but it can be fine-tuned to suit customer preferences. It is also not set in stone. Its development is ongoing, and if, for example, feedback from clients suggests that in an overwhelming majority of cases key types of behaviour suggest malfeasance of some kind, then appropriate alerts will be built into the package.
SMARTS Surveillance, Nasdaq’s award-winning market surveillance product, is used 146 market participants, 50 marketplaces and 17 regulatory authorities for day-to-day trade monitoring obligations.
Lens represents a new approach to identifying trading risk and so has not yet been formally advocated by regulators, but O’Brien says they are coming around to the new ways of working and the different discovery-based approach to monitoring that it embodies. “No-one yet foresees the day where rule/logic-based alerts will disappear. There are defined risks that dealers have to look for. Regulators want to see that, but they are getting more open to seeing that this is not the only way,” he says.