BFS Solutions APAC
The accelerated volumes of digital transactions and changing customer behavior post-pandemic have put the financial services industry under pressure to increase focus on risk and compliance. Furthermore, emerging risks have compounded issues regarding anti-money laundering (AML) compliance. With criminals using more unique ways of laundering money, efforts to combat money laundering need to be increased. Also, due to frequent changing regulatory expectations, banks find it challenging to keep their AML compliance programs effective, guarding against these increasingly complex money laundering activities. In a research carried out by Fenergo, financial institutions incurred fines totaling $10.4 billion in 20201 for data breaches and compliance failures related to AML, know your customer (KYC), and markets in financial instruments directive (MiFID) violations. The current approach adopted by financial institutions, despite being extensive and expensive, is focused on regulatory compliance and not identifying and intercepting financial crime. As a result, financial organizations are subjected to paying huge fines.
According to Deloitte2, as much as two to five percent of the global GDP, or $2 trillion, is laundered yearly. The authorities intercept less than one percent of the amount laundered, which points to a disconnect between financial institutions (facilitators of such financial transactions) and enforcement agencies (who work based on intelligence gathered from various agencies, including banks and financial institutions) to detect and intercept financial crime.
However, in the recent past, financial institutions, enforcement agencies, and regulators have been trying to evolve mechanisms at global, regional, and local levels. This approach offers better collaboration and sharing of financial information. Doing so also allows us to detect and intercepts a financial crime early and intelligently – even before it proliferates within the system. The collaborative sharing of money laundering/terror financing) (ML/TF) information and cases (COSMIC) initiative from the Monetary Authority of Singapore (MAS) is a case in point. COSMIC will allow financial institutions to query and alert each other on potential illicit behaviors and allow MAS, law enforcement agencies, and financial institutions to act early to disrupt criminal networks.
Robust information technology (IT) systems have always been critical for AML compliance. Even the banks have increased their investments in automated systems for transaction monitoring and sanction screening over the years. However, these investments may not have yielded the desired result. Throughout the end to end transaction journey, many disparate manual processes, functions, data sources, and applications come into play. This complexity, which is inflexible, creates challenges for organizations that result in delays, errors, and poor customer experience. Thus, leading to poor compliance, increased costs, loss of credibility, and poor customer experience
In multiple research papers, the three most significant pain points that the financial industry currently faces in the KYC/AML area are:
From a compliance perspective, banks and financial services organizations often lack visibility, controls, and governance. As a result, these institutions have limited audibility regarding AML and KYC obligations. The main reason for this limitation is the inability to consolidate all the disparate data necessary for KYC, AML, fraud and financial crime risk assessments. One can’t deny that consolidating disparate data is complex and tedious as it requires multiple interactions with scattered data sources. Even in today’s digitally driven economy, many financial institutions and internal or third-party data are scattered across siloed systems requiring manual intervention. The poor management of such data often leads to inconsistencies, errors, and missed steps.
The effectiveness of any AML transaction monitoring tool depends on the quality of data that gets fed into the tool. Unfortunately, banks face two main challenges on that front. The first is feeding the data in an inappropriate format into the transaction monitoring system that does not meet AML/KYC compliance requirements. The second issue arises once data is sourced from multiple systems. As the data travels through an internal process, it leads to possible changes in the data composition by the time it reaches the transaction monitoring system. Apart from data, many financial institutions are not satisfied with the efficacy of the fuzzy logic algorithms implemented by some of the transaction monitoring systems. It results in many false positives, leading to cost and operational overheads.
The manual nature of risk assessment is interwoven and part of all the other pain points listed above. None of the organizations have implemented any degree of automation, which is a huge challenge for financial institutions in this space. Everything from horizon scanning to documenting relevant threats, risks, and controls to working with risk calculations and presenting relevant results is done manually. Financial services firms now know that the staffing levels required to triage and service up to a 90 percent plus false positives detection rate are unsustainable. Only recently have AI and machine learning (ML) been deployed within the banks for financial crime detection, leading to fewer false positives. Even as detection systems improve and produce fewer false positives, part of the investigation process is still largely manual. But that’s not all. The situation worsens, considering the effects of disparate detection systems with differing levels of automation within the case management workflows. This leads to inconsistent user experience for employees responsible for these outcomes.
These challenges not only weigh down financial institutions in terms of the higher cost of operations, non-compliance, low employee morale, and ineffective financial crime assessment and detection; but also in terms of regulatory fines, loss of reputation, attracting talent, and investment in future initiatives. So the question remains, how can banks combat these financial crimes?
In the last few years, financial institutions have invested heavily in enhanced transaction monitoring systems by taking advantage of capabilities from fintech that specialize in AI and ML. Financial institutions use the best-of-breed approach that marries investments in legacy systems with newer, AI-based technologies. However, banks can go a step further and improve the accuracy and cost-efficiency of the investigation process. They can achieve this by leveraging the detection output from multiple transaction monitoring systems and combining it with a unified workflow and case management system that is specifically designed and built to cater to the needs of the financial crime investigation process.
Virtusa has designed and built a low code framework for financial crime investigation. Low code platforms allow organizations to accelerate the delivery and implementation of applications and adapt and evolve to meet changing business requirements. In addition, the visual processes used to build low code applications serve as a common language, enabling business subject matter experts and project deployment experts to collaborate more easily. Prototyping and building applications can be made much faster. Data can be automatically fetched from multiple systems and in whatever format it exists and brought into the platform, allowing users and AI to run processes to complete tasks.
Virtusa financial crime investigation framework has several benefits for financial institutions:
At Virtusa, we provide a low code platform built explicitly for financial crime investigation. Our platform offers a valuable alternative to custom-built applications and commercial off-the-shelf software solutions. Thus, giving financial organizations a platform to build applications using visual process models and drag-and-drop functionality. It speeds up the pace of building applications because the framework comes with various application building blocks ready to use, such as AI, robotic process automation (RPA), a rules engine, case management, workflow, and data integration. Another key advantage is that financial organizations do not require to put all their data into a new proprietary data model. The data can be sourced in any format in existing applications and brought into the platform, allowing users and AI to run processes to complete the required tasks. It integrates new technologies and innovative solutions, like AI, into the end to end AML and KYC processes. Virtusa framework will help banks reduce the cost of operations by automating their complex and interdependent AML/KYC/Financial Crime investigation processes and enable better compliance and enhanced preparedness for new initiatives in the Financial crime space by regulators in the future.
Emma Woollacott. “Fines against banks for data breaches and non-compliance more than double in 2020 | December 31, 2020 | https://portswigger.net/daily-swig/fines-against-banks-for-data-breaches-and-noncompliance-more-than-doubled-in-2020#:~:text=Fines%20levied%20against%20financial%20firms,with%20penalties%20totalling%20%2410.4%20billion
Anti-Money Laundering Preparedness Survey Report 2020 | https://www2.deloitte.com/content/dam/Deloitte/in/Documents/finance/Forensic/in-forensic-AML-Survey-report-2020-noexp.pdf
Vice President, BFS Solutions APAC
Saby Dsouza is heading the Banking & Financial Services (BFS) solutions practice for Virtusa in the APAC region. He has over 26 years of IT consulting/solutions experience with major global banks covering consumer banking, wholesale banking, and risk management & payments.
Subscribe to keep up-to-date with recent industry developments including industry insights and innovative solution capabilities
An in-depth perspective on the significance, challenges, and models for data monetization for financial and strategic business benefits
By embracing low code development, organizations are gaining significant benefits.