You are being repeatedly sent reminders on the phone, email, SMS, etc. to fill-in your KYC (Know your customer) form by your bank and wondering why do you need to do that? It is a process of ‘due diligence of customer’s which is being done by banks as part of the Anti Money Laundering (AML) system. We will discuss AML and how analytics can help identify and stop illegal money from being converted to legitimate money. We will also learn how analytics helps design Anti Money laundering systems and prevents the draining of resources in our monetary systems. AML systems are very powerful applications of analytics in the financial domain that includes descriptive, predictive, and prescriptive analytics.
What is MONEY LAUNDERING?
Money laundering is the process by which money obtained from certain crimes such as extortion, insider trading, drug trafficking, and illegal gambling, etc. are transformed into apparently legitimate money i.e. money derived from legitimate sources.
Money laundering may be executed with or without the assistance of the financial sectors, but it is very unfortunate that billions of dollars of illegal money are laundered through financial institutions. The term ‘money laundering’ is combined with financial crime, used more generally to include misuse of the financial system, terrorism financing, and avoiding of international sanctions.
The nature of the services (i.e. managing, controlling, and possessing money and property belonging to others) offered by the financial institutions means that it is vulnerable to abuse by money launderers.
How does money laundering take place?
Money laundering generally takes place in three steps:
- The first step is called “placement” which involves the introduction of cash into the financial system by any means.
- The second one is “layering” which involves carrying out complex financial transactions properly to cover up the illegal sources.
- The final step “integration” requires obtaining the wealth generated from the transactions of the unlawful funds.
So what is an Anti Money Laundering (AML) system?
AML is defined as a set of procedures, laws, or regulations designed to stop the practice of transforming illegitimate money obtained from illegal sources (such as extortion, insider trading, drug trafficking, and illegal gambling, etc.) into apparently legitimate money.
Recently, the Reserve Bank of India has issued certain guidelines strictly instructing the banks to accept only some predefined “officially valid documents” which are to be obtained for identification of the customers as well as proof of address. This is also known as “KYC “or “Know Your Customer” norms.
What is the process of AML?
- Analytics is used to design Anti Money Laundering systems. KYC or verifying new customers is of critical importance. It involves due diligence investigation, initial verification of identity, comparison to the list of known entities, and risk assessment, and may involve profiling of potential client activity to aid in future monitoring.
- In the next step, assessment of activity for all customers is done with an emphasis on customers and activity with the highest risk. This helps in identifying suspicious activity that may ultimately result in the filing of a Suspicious Activity Report (SAR) and CTR.
- Assessment is generally a two-stage process whereby first, the AML transaction monitoring system is fed with transaction data from multiple internal and external data sources. It filters, compiles, and summarizes transaction data, and identifies instances of potentially suspicious behavior. Corresponding banking scenarios are also included in the model to strengthen the detection layer.
- The detection of potentially suspicious behavior is known as red-flagged or ‘alerts’. The values for the number of ‘parameters’ or ‘thresholds’ determine which activities are flagged. In this step entity link analysis is done to visualize transactional relationships to understand the source of funds and behaviors that may indicate organized rings. And, a peer group anomaly component compares an entity’s behavior to its historical behavior and the behavior of its peer groups.
- Finally, the analysts take a deep dive into these instances of potentially suspicious behavior using internal and external information sources to determine if, ultimately, a SAR should be filed.
A-priori rules, graph theory (link analysis), and predictive/data mining models, such as data mining methods to test deviations from peer groups, and predicted versus actual patterns, etc. are used in AML analytics. Highly impactful AML analytics also include time series analysis for analyzing sequential events in complex money laundering problems.
Want to learn the techniques mentioned here like predictive and data mining models, a-proper rules, and application of analytics in the financial sector, visit the website of Data Brio Academy for different training programs on offer.