How Artificial Intelligence Can Help Overcome Challenges in Correspondent Banking Relationships

The capacity to send and receive international payments through correspondent banking is critical to the global economy as financial institutions rely on these relationships to move their customers’ money. As a result of increased regulatory requirements and risk related to shell companies, correspondent banking relationship management has become a more challenging undertaking for financial institutions.

In some cases, organizations no longer desire to manage correspondent banking relationships linked to money services businesses (MSBs) or correspondent banks in high-risk jurisdictions over fears of potential regulatory scrutiny, investigation and exorbitant penalties. New solutions powered by artificial intelligence (AI) and advanced data analytics technologies offer much promise to financial institutions to help assist with verification of business relationships for correspondent banking customers.

According to estimates in a 2015 report by the World Bank, global remittances are expected to grow, albeit at a slow pace.1 High remittance volume brings increased regulatory pressures, risk and compliance costs that bite at the heels of financial institutions. The unfortunate outcome is that financial institutions often choose to de-risk foreign correspondent banking relationships rather than deal with the inherent risk and challenges of maintaining correspondent banking accounts.

The correspondent banking regulatory landscape changed when the Financial Crimes Enforcement Network (FinCEN) enacted “Special Due Diligence for Correspondent Accounts and Private Banking” interim regulations under Section 312 of the USA PATRIOT Act. FinCEN’s actions were in response to the 9/11 terrorist attacks and mandated that financial institutions enact specific know your customer (KYC), customer due diligence (CDD) and, in some cases, enhanced due diligence (EDD) procedures for their foreign correspondent banking relationships.

AML Tools and Processes Are Failing: New Technologies to the Rescue

Effective correspondent banking relationship risk management boils down to having the adequate tools to efficiently resolve entities that use correspondent banks. Recurring challenges for compliance and investigative teams include establishing an economic purpose and verifying complementary lines of business for correspondent bank customers.

Legacy technologies including rules-based transaction monitoring systems (TMS) attempt to detect and report on transaction details that indicate suspicious activity, but are largely ineffective. Financial institutions often find themselves filing conservative suspicious activity reports (SARs) because the necessary data is unavailable.

The reliance on TMS is presenting significant risk to financial institutions. An unacceptable number of illicit transactions, or false negatives, are not flagged by today’s TMS. It is estimated that 50 percent of financial crimes in the banking system pass through TMS unnoticed. In addition, it is understood by anti-money laundering (AML) compliance practitioners that approximately 95 percent of the alerts generated by TMS are false positives.

Unlike TMS, artificial intelligence-based systems can detect patterns of behavior, analyze the intent of those patterns and expose anomalous activities. For example, transactions that do not follow the usual frequency and directional patterns expected for a given type of account may not be flagged by a TMS, but would be identified with an effective AI solution. An AI solution can learn the baseline of normal reported payroll account activity and thus identify any irregularities in payroll transactions as potentially fictitious and worthy of further investigation.

AI Can Radically Improve Correspondent Banking

Advanced data analytics and AI technologies can help financial institutions manage correspondent banking more effectively. An important function of an AI solution is its ability to monitor customers’ relationships to other customers and entities and learn from their associated behavior. An AI-based AML solution can automate the transactional analysis of correspondent banking relationships to find anomalous behaviors and identify the end clients causing those anomalies. An AI-enhanced solution can also account for seasonality, mergers and acquisitions, randomness and other legitimate variances to find the illegitimate anomalies that are presenting significant risks to financial institutions.

Entity resolution and the investigation of entity relationships are integral to curbing the de-risking cycle. The ability to establish behavioral histories related to the volume of transactions and their amounts, expected volume of transactions and amounts, current and ongoing ultimate beneficial owner data, and any/all adverse media related to the entity in question, is the key to ensuring accurate correspondent banking risk management and know your customer’s customer.

Expanding the same data points to an entity’s related parties (customers) provides a holistic risk picture for compliance and investigative teams within a bank or covered financial institution. With data analytics and AI, time-consuming and costly correspondent bank and pseudo-customer investigations can become automated, providing investigators with the most essential data.

Advancements in data analytics and AI can enable compliance teams and AML investigators to fulfill their regulatory obligations with precision, improve SAR reporting and ultimately prevent unnecessary de-risking of correspondent banking customers.

Human Decision-Making Is Still a Key Element to AML Compliance

On the front line against money laundering are AML compliance professionals tasked with identifying suspicious activities. The average compliance professional works eight to 10 alerts each day. To make a decision on each case, the professional is required to review multiple and disparate bank systems, including KYC and customer information program databases, as well as cross-reference TMS for related flags and turning to external sources of data. The application of AI can significantly enhance the capabilities of human compliance professionals by eliminating tedious, time-consuming tasks involving massive amounts of data.

David McLaughlin, CEO, QuantaVerse, Wayne, PA, USA,

  1. “Remittances Growth to Slow Sharply in 2015, as Europe and Russia Stay Weak; Pick Up Expected Next Year,” The World Bank, April 13, 2015,

Leave a Reply