Compromise is Good for Politics, Not Surveillance
February 15th, 2019
Compromise is essential in politics, but unnecessary when it comes to market surveillance. Over the last few years, technology and regulation have transformed the benefits that well-executed surveillance initiatives can bring to an organization. Although compliance teams have never been asked to accomplish more than they are now, they also have an unprecedented arsenal of tools at their disposal to accomplish it all – and in record time. Behavioral profiling, anomaly detection, rule-based models, custom models, the ability to correlate communications data with trade alerts and automated case management should now be in every compliance team’s toolbox. If your surveillance program isn’t capitalizing on these new analytics and tools, you are most certainly compromising on surveillance and putting your firm at risk.
Today’s compliance teams must be able to detect known forms of market manipulation and identify who is putting the firm at risk. However, until recently, surveillance was limited to monitoring trades for known forms of market manipulation like front running or wash trades.
Compliance analysts were like traffic cops – only enforcing a prescribed set of violations. Today’s analysts are detectives. They identify risky individuals and uncover hidden threats. Like a police detective, the analyst relies on evidence and clues to investigate suspicious behavior and uncover the hidden threats that rule-based models are designed to detect. Behavioral profiling and anomaly detection techniques, such as machine learning, are the tools that make this possible. Behavioral profiles enable analysts to monitor individuals by establishing the “normal” behavior of an individual or their peer group, while machine learning is used to identify anomalies from an individual’s normal behavior.
Just like the police force, the compliance function needs detectives using anomaly detection to identify risk and cops using rule-based models to detect known forms of market manipulation.
It’s also crucial that those rule-based models have kept pace with new regulations, which require compliance to detect a lot more than basic forms of manipulation. Today, the compliance function requires out-of-the-box models that can detect manipulation across products and markets. To help detect intent and forms of manipulation like quote stuffing, the models should analyze orders as well as executions. They also need to be robust. For example, insider trading models should be based on type of news, sentiment and price movement. Models based solely on price movement generate too many false positives. In short, using analytics from a “tick in the box” vendor can be a ticking time bomb for firms.
As every Chief Compliance Officer knows, even exceptional out-of-the-box models only cover about 80 percent of a firm’s surveillance requirements. That’s because firms have different interpretations of regulatory mandates and the typologies needed to meet them. As a result, an out-of-the-box price ramping model that’s perfect for one firm may not meet the needs of another.
Firms need exceptional out-of-the-box models, and the ability to create custom models, to cover all their surveillance requirements. Furthermore, they need custom machine learning and rule-based models that can be created by a range of people including business analysts, developers and data scientists. Don’t compromise: get a solution that provides both proven out-of-the-box models and the ability to create custom ones purpose-built for your organization’s unique situation.
Holistic Surveillance – It’s Essential
Holistic surveillance – correlating and analyzing related trade and communications data – is essential to meet regulatory requirements to detect intent and reconstruct trades. It also cuts the time analysts spend chasing false positives by providing them with the context they need to quickly determine if an alert is a false positive or a true violation.
A firm’s holistic surveillance program should cover all forms of communications data – phone calls, emails, instant messages and chats. Holistic surveillance has progressed from an aspiration to a reality over the last couple of years. Don’t compromise on surveillance and risk regulatory violations by viewing trade and communications data separately.
The process of investigating threats is just as important as detecting them. A case manager that saves analysts time by automating workflows, requests for information and escalations is the foundation of an efficient investigative process. Case managers that support holistic surveillance, by ingesting and correlating trade alerts with communication data, provide analysts with exceptional investigative capabilities. Holistic surveillance enables analysts to investigate threats faster and better by providing them with more information and unique insights. Don’t compromise on surveillance with a case manager that produces time consuming investigations and doesn’t support holistic surveillance.
There are few purchases in life where cost is no object; the markets surveillance solutions category isn’t one of them. As surveillance expands to cover an array of threats and technologies, calculating the Total Cost of Ownership (TCO) for your surveillance infrastructure is vital. A collection of point systems for market surveillance, anomaly detection and communications surveillance can dramatically increase your hardware, training, support and professional services cost. Your calculations should also include the cost of internal resources for system and data integration. An integrated compliance platform that supports multiple solutions and lines of business is the best way to keep your capabilities high and your costs low.
Surveillance is at an exciting inflection point. You can create a program that addresses more threats and better protects the firm from regulatory and reputational damage. But only if you don’t compromise on surveillance.