AML/CFT is an ideal test field for new technologies given the risk factors and complex challenges Asian banks face, says Synpulse’s Prasanna Venkatesan.
By the end of 2018, banks connected to the 1MDB case had been fined SGD 30 million (USD 22 million) in aggregate by MAS (the Monetary Authority of Singapore) due to severe breaches of related offences, according to its inaugural Enforcement Report released last month.
These breaches highlighted the lapses in banks’ CDD (customer due diligence) measures, absence of proper monitoring of suspicious activities, and the inadequacy of the control framework that is supposed to ensure implementation of robust policies and procedures. The audit findings relating to the inadequacy of control measures in the case pushed regulators to enforce higher-standards for AML (anti-money laundering) and CFT (counter-financing of terrorism) compliance programmes.
In this article, we cover a brief introduction of the risk factors that increase money laundering risk, challenges faced by the banks in term of AML control, and the measures we suggest to help address those challenges.
Factors that increase the money laundering risk
The strong presence of cash-based industries – such as the casino industry as well as low value money-changing and remittance businesses in Southeast Asia – is a significant driver of money laundering in Singapore and Hong Kong. Cash transactions act as an impediment to the effectiveness of AML/CFT oversight and reporting systems as they generally do not leave electronic records, making them difficult to track and report.
Additionally, with more than three-quarters of funds managed in Singapore and Hong Kong coming from abroad, banks may be exposed to risks stemming from regulatory gaps or weaknesses in foreign countries. Financial institutions worldwide may not be subject to the same level of regulatory guidance, leading to an increased money laundering risk for cross-border payments and transfers. For instance, the continued tolerance of nominee-owned accounts by regulatory authorities in certain jurisdictions prevents the proper identification of beneficial ownership, which reduces transparency and hinders enforcement of KYC (know-your-customer) requirements by banks in such regulatory environments.
Finally, the introduction of new client activity and payment patterns by novel technologies increases the risk of money laundering. For example, cryptocurrencies with their inherent anonymity and de-centralised systems are creating obstacles for regulators and banks in identifying and mitigating money laundering risks.
Challenges in implementing effective AML systems
Due to the complex and varied nature of transaction information, simple rule-based AML systems struggle with large numbers of false-positive cases, leading to highly trained AML staff being forced to investigate and spend time on cases that often turn out to be irrelevant.
Many of the systems in use today are primarily rule-based, where the same assumptions are adopted for each client regardless of their financial background and intended purpose of the account. Most of these systems are not capable of learning from previously investigated transactions, leading to, again, duplicate efforts wasted on cases that are false positives. The common challenges of existing technology were explored in depth here.
The information required to investigate a potential money laundering case – such as adverse news, KYC screening results, changes in clients’ KYC information, account activities and previous investigations – are rarely aggregated on one platform, meaning that the investigation of a single case may have to be done on numerous platforms simultaneously, which slows down the process significantly and increases the risk of operational error due to procedural loopholes.
Strengthening AML systems
Banks should navigate away from simple rule-based systems to a smarter, technology-driven approach, which using fuzzy logic and machine learning to reduce false positives. Client screening can now be made more accurate using such technologies to match multiple parameters, reducing the effort required to filter out false-positive results.
Similar technologies can be applied to transaction monitoring to identify out-of-pattern transactions based on clients’ information and transaction histories. Third-party data – for example from major counterparties or AML-specific databases, such as shipment information and price validation for trade-based transactions – can be leveraged to better understand the connections between clients, and accurately determine where enhanced due diligence or further investigation could be appropriate.
The execution of investigations for AML cases should also be aggregated on to a single platform. The platform should be able to access all collected KYC information, client activity data, and relevant internal and external databases. Final output should include a summary of risk profile changes, as well as any unusual transactions, allowing for effective further investigation if necessary.
It is also important to focus on establishing a sound risk culture framework within the bank by ensuring that the tone is set clearly from the top, with the appropriate incentive systems in place to drive cultural change. Staff training on risk and compliance issues should also remain a major part in bank-wide efforts to sustain compliance. The importance of culture and conduct was discussed in our previous article here.
Banks in Asia face an increased risk of money laundering, and as such, it is time to deploy technology that is appropriate to meeting the challenge. This may also include technologies that read unstructured data and can detect the unknowns with predictive analytics.
Given the volume and complexity of AML cases, and the scale of information available to analyse and digest, AML/CFT is an ideal area for banks to consider unconventional methods and technologies.
Prasanna Venkatesan is a Singapore-based Partner and APAC Regulatory Compliance & Risk practice lead at Synpulse.