Through correspondent banking relationships, firms can access financial services in different jurisdictions. These banks provide international payment services, which is critically important as it facilitates international trade, financial inclusion and prevents payment flows from moving underground.
However, correspondent banking is in steep decline. Rather than being viewed through a “value” lens, it’s seen primarily as a high-risk proposition due to profitability concerns, the liability associated with a correspondent account, and a lack of clarity on the degree of granularity Know Your Customer (KYC) profiles need to be to meet the regulatory requirements for AML programs.
According to the International Monetary Fund (IMF), many financial institutions have exited whole markets, geographies or regions due to perceived financial and reputational risk. Customers in these areas will either be unable to affect correspondent banking activities or have a more limited number of options for these services. The risk-reward assessment that correspondent banks conduct, is now skewed to the risk aversion side of the equation, with the result being a less secure and integrated world financial system.
“The risk-reward assessment that correspondent banks conduct, is now skewed to the risk aversion side of the equation, with the result being a less secure and integrated world financial system.”
However, the Deutsche Bank Trust Company has a different view. CEO Susan Skerritt was quoted in The Banker saying that the correspondent banking industry is a proven success story: “The correspondent banking industry has a proven record of continuous improvement – automation/STP processing, reachability, and scalability. We process thousands of payments every second.”
So how does one de-risk correspondent banking in order to reverse this steep decline?
Banks can accomplish this through the implementation of a customer risk decision platform. These platforms deliver essential capabilities to make rapid, accurate and insightful decisions on their correspondent banking relationships.
- Access to a wide range of highly curated and open source risk-related data
The only way to access a range of information like this is with Robotic Process Automation (RPA). State-of-the-art implementations use bots to crawl the internet and gather legally accessible and curated data sets that are immediately utilized by firms. Enterprises can identify those sources, which could be anywhere from a dozen to thousands, that they deem essential to meet their data needs. This data is used for uncovering new or unknown risks when the entity is unknown to ensure there are no red flags. This data is also part of the raw material for entity development or supplementation.
- Entity construction through a combination of firmographics and demographics
This base data should be enhanced by RPA tools to assemble and sort content as well as utilizing machine learning technology to merge additional information outside of the core entity.
- Machine learning based decisions
Machine learning is essential for data categorization and helps determine how to map data to business decisions and provides professionals with the needed data. Additionally, the clustering capabilities that machine learning provide enables grouping of events that refer to the same risk entity. This can dramatically reduce the number of false positives. Since multiple risk events can reference that single entity, it gives analysts the power to disposition them en masse.
Machine Learning also makes entity development intelligent. Meaning, as data is gathered, these algorithms can distinguish between what data is important for entity construction and maintenance and which not. Analysts can leverage the learned results of past investigations to rapidly and accurately conclude a current investigation.
- On-demand information visualization to serve the needs of a broad user population
Other benefits include:
- Network diagrams of risky entities to identify bad actors before they cross the line. Ensuring correspondent banks have a complete view of their high-risk groups which drives tailored monitoring frequencies based on risk level
- Automated Ultimate Beneficial Owner identification to take the legwork out of identifying unknown risks
- Emerging market data orchestration to increase identification of risk vectors, as well as auto-population of entity data from third-party sources. This enables differentiation between high-risk and low-risk jurisdictions, thus removing the manual work associated with data scarcity, particularly in emerging markets
- Automated investigation triggering, operations and due diligence for associated entities in a single source.
A customer risk decision platform that provides these capabilities must also reside in the cloud. This is essential to deliver the reach, scalability, and security required for effective customer risk decision support.
The cloud-based system must:
- Provide an API-driven open architecture for integration of data, in-house or third-party systems
- Isolate and secure client data in either secure single-tenant or multi-tenant environments
- Provide for a choice of deployment options
- Enable adaptive resourcing and automation to deliver a scalable solution
- Have a data lake capability to access and organize risk data, metrics, proprietary and open source content.
These capabilities, taken together, enable financial institutions to move beyond country or market level risk vis-a-vis a detailed understanding of risks at an entity level. The more accurate and insightful a view of entity risk, the easier it is for an institution to make an informed decision about a potential correspondent banking relationship.
Banks that utilize customer risk decision management solutions make insightful relationship assessments because content, entity, and policy are unified and automated across the risk decision lifecycle. This will result in significant upscaling of their business without fear of regulatory repercussion or reputational damage.