Reduce Operational Risks Using Behavioral Analytics
September 8th, 2017
Conduct-related threats such as collusion, unauthorized trading, benchmark manipulation and rogue trading put firms at risk. They’re difficult to prevent because they don’t follow a predictable pattern that can be modeled. Firms also need to address both sides of the compliance coin — i.e., known forms of market abuse (such as layering and spoofing) and hidden conduct-related threats. To achieve this, many firms are adopting a risk-based approach to surveillance by analyzing patterns over time, enabling them to identify known and unknown threats as they develop. The emergence of “Big Data” and advanced analytics, such as behavioral analytics, is bringing risk-based surveillance within reach.
Behavioral analytics enables firms to identify risky entities – including traders, investment advisors and accounts – by analyzing a wide range of data and measuring deviations from normal behavior.
What’s interesting is that regulators and standard boards are generally not prescriptive about how firms mitigate risk, particularly from conduct-related threats. However, several regulatory bodies seem to favor an approach based on analyzing anomalies from normal behavior.
For example, the Bank of England’s Fair and Effective Markets Review (FEMR) suggests mitigating risk by monitoring an individual’s conduct and identifying vulnerabilities, while the Financial Industry Regulatory Authority (FINRA) advises mining trade data for deviations from normal trading patterns. The FICC Markets Standard Board (FMSB), meanwhile, endorses systems to prevent transaction-based conduct risks, while the Monetary Authority of Singapore (MAS) recommends advanced analytics to detect complex patterns for market abuse.
The Role of Data
A recent survey on behavioral analytics indicated clear differences in the types of data used for behavioral analytics, in the short term versus the long term. Respondents said that they will predominately use trade and communications data, as well as market news, in their behavioral analytics over the short term. This aligns closely with the recent trend toward the use of metadata from communications (rather than content of the communications) to most accurately identify non-compliant behavior.
In the longer term, human resources data will become increasingly significant at unearthing a range of risk, including the potential for insider trading.
Although there is a planned increase in data types used to support behavioral analysis, the majority of survey respondents – nearly 65% – agree that behavioral analytics can generate new, valuable insights from trade data (e.g., orders and executions) alone. Respondents therefore recognize the immediate value that can be derived from behavioral analytics, as well as the benefit of integrating additional data sources and metadata.
Interestingly, survey respondents revealed that they know exactly what they want to measure with analytics. Communications data, trade data and HR data are at the front of the line.
The firms responding to the survey also showed great enthusiasm for applying some form of behaviorial analytics to their trading surveillance risk programs, reflecting their strong belief that behavioral analytics can reduce operational and regulatory risk while accelerating investigations.
Internal Development vs. Third-Party Solutions
In terms of adoption of behavioral analytics, the survey made it clear that almost no one wants to go it alone. Just 15% of respondents want to build a solution themselves, while nearly 70% of respondents indicated that they want to purchase their behavioral analytics solution from a vendor or have a mix of homegrown and vendor-supplied behavioral analytics.
When selecting a partner to support a program integrating behavioral analytics, there are best practices and guidelines to consider. The technology used to support this type of approach must have the capability to (1) correlate behavioral and traditional alerts in one case manager; (2) create custom risk factors (where needed) and analyze out-of-the-box risk factors; and (3) easily incorporate trade, communications and HR data.
Firms that seek to adopt an integrated surveillance solution that combines model-based analytics and behavioral analytics will not only improve their effectiveness, but also reduce their total cost of compliance and risk.