The pixel Skip to content

Reuters: Advanced Analytics Used for Tackling AML Issues Have Yet to Replace Banks’ Legacy Systems

Banks globally which have begun to use advanced analytics for the purposes of tackling anti-money laundering and financial crime challenges have not yet replaced their existing legacy systems with this new tool, said a consultant.

Financial institutions in Singapore are among the early adopters of advanced analytics in the world in using the tool for AML purposes. Adoption of advanced analytics among banks in Singapore began two years ago under the auspices of the Monetary Authority of Singapore which has been pushing for an industry-wide adoption. But progress in the use of advanced analytics has been slow and the full benefits of the tools have yet to be reaped. 

Experts have cited a number of reasons including the difficulty of replacing banks’ existing legacy systems with advanced analytics, concerns about fines being imposed if financial institutions get it wrong when using advanced analytics, consideration for the safety and soundness of the financial system, and a lack of understanding of the concerns and issues around transaction monitoring on the part of regulatory technology (regtech) and financial technology (fintech) firms.

The use of advanced analytics for AML purposes remains at an exploratory stage for many financial institutions. Most are using it as an additional tool rather than as a replacement tool, said Brandon Daniels, president of global technology markets at Exiger in New York.

“When it comes to using advanced or predictive analytics, banks are not typically using them to replace legacy transaction monitoring or screening systems. They are using advanced analytics to help prioritise the cases they review or remove irrelevant cases,” he said.

Exploratory phase; advanced analytics hard to define

Daniels also pointed out that most people are just beginning to understand the way advanced analytics work in practice. A conservative level of deployment is therefore expected until there is a common understanding of the way advanced analytics systems run and make decisions, and a way of creating an auditable reasoning for the systems’ decisions, he said.

Advanced analytics require a different mindset and involve looking at patterns, correlations and inferences that may not be obvious at first. For banks to capture the power of data analytics, it requires lateral thinking, said Kevin Nixon, founder of Nixon Global Advisory in Sydney.

“Advanced data analytics is about inferences, probability and likelihood, and looking at lateral data sets. You are putting a lot more data, not just transaction data but also personal data and location data, for example. Traditionally, you track who pays, to what account and how much. When you bring in modern analytics, you are asking the machine to scan all possible data. You are not just looking at numbers, income, expenses or trends. You are looking at patterns and indicators,” he said.

Despite that advanced analytics have been in use for the longest time, its remains difficult to define, Nixon said.

“People have been analysing data forever, but often a single or a small number of variables. If you ask 10 people what they mean by advanced analytics in 2019, you will probably get 12 different answers,” he said.

Financial institutions are adopting advanced, cutting-edge tools

The financial services industry around the world has made great strides in adopting technology in recent years, with many financial institutions using advanced and cutting edge tools for various purposes pertaining to AML, according to Daniels. This includes tools to reduce transaction monitoring alerts, conduct due diligence, and sanctions hits false positives, among others. These systems are often highly complex at the back-end though they appear simplified and familiar at the front-end, he said.

“It’s not that the entire industry is stuck in the past or slow to adopt advanced technology. It’s just that everything in financial institutions is so policy-driven. Compliance professionals are more likely to accept advanced technology if the decisions made by analytic systems conform to the process that has been approved and adopted. You can’t leave anything to the imagination,” he said.

The fact that financial institutions are procedure and policy-driven also explains why regtech and fintech firms sometimes have to come up with solutions that mirror financial institutions’ existing systems to be successful. 

“It is just part of the baby steps required to reach a more mature environment,” Daniels said.

It is also incumbent on vendors, and the regtech and fintech community to create tools in a “non-techy” way that explain exactly how advanced analytics systems make decisions, which according to Daniels, is just beginning to happen.

Lack of focus on data in the past presents challenge to financial institutions

Another hinderance to financial institutions adopting advanced analytics is the lack of consistency on data quality. This is largely because existing systems are much more simplified and rudimentary in nature requiring very little data, Daniels said. These systems are often used for sanctions screening, transaction monitoring, KYC management, and risk management.

The historic lack of focus on data quality has now become a challenge to financial institutions as they move toward adopting tools such as advanced analytics which need more data, Daniels said. Until they get their data quality up to speed, many financial institutions will not be able to make sufficient progress in adopting advanced data analytics, he said.

Many banks also fail to maintain their internal watchlists with sufficient data to take advantage of analytics in screening or transaction monitoring, the information necessary to distinguish the everyday person who are not easily found on the web from a person with potential risk.

“For everyday retail customers, [banks] haven’t collected data points that allow systems to distinguish between the billions of law-abiding citizens and the several millions [of] potential money launders. The legacy systems simply flag information for banks to review manually. But advanced data analytics, which are replicating expert decisions, demand the exact same data that a human has access to when making complex AML decisions. That kind of data isn’t typically available at scale,” Daniels said.

Game-changers: data in digital format and vast amount of data points

The vast amount of data that is now available in digital format and the significant amount of data points human generate are two game changers, both of which have facilitated the rise in importance of advanced analytics, Nixon said. 

“Advanced data analytics require a broad set of data because it involves identifying correlations and causations that may not appear obvious at first,” he said.

– Writen by Patricia Lee, chief correspondent, banking and securities regulation, Asia

This article first appeared on Thomson Reuters Regulatory Intelligence

Fighting financial crime in practice
Read more from Reuters in Exiger’s series of articles for AML professionals:

  • Collaborations Between Regulators and Industry Most Effective in Addressing Financial Crime
    How can financial institutions and regulators work together to trace the 99% of illicit funds still hidden in financial markets?
    Read article >
  • Greater Awareness of Personal Data Privacy Law Heightens Conflict with Financial Crime Regulation
    How can banks safely navigate the ‘grey zone’ that exists between how they leverage data for due diligence and how they fulfil their data privacy obligations?
    Read article >
  • 1% Illicit Funds Uncovered in Financial System Raises Question About Whether Banks Need to Do More
    Less than 1% of illicit money flows typically gets seized and frozen given today’s AML controls – so how do you trace the other 99% awash in financial markets?
    Read article >

The risk landscape is constantly changing. Hear about the latest with Exiger.