Correspondent banking can be vulnerable to exploitation by money launderers, terrorist financers, fraudsters, tax evaders and sanctions breachers. But this does not have to be, writes Yaron Hazan, VP – regulatory affairs at ThetaRay

Correspondent banking is the service whereby account holders from one bank in a certain jurisdiction (respondent) can benefit from the banking services of another bank in a different jurisdiction (correspondent).Correspondent banking is a critical method of connecting local and underdeveloped economies throughout the world to the global financial system. However, this doesn’t mean the process is easy.

Go deeper with GlobalData

Data Insights

The gold standard of business intelligence.

Find out more

Correspondent banking transactions involve up to a dozen banks scattered across multiple countries, making it nearly impossible for banks to conduct proper KYC due diligence and gain transparency into who is processing and receiving the money.

As a result, a number of banks have bowed out of the business, leaving cross border payments to payment service providers such as Venmo and Zelle, who face less stringent regulations than traditional banks and therefore have an easier time incorporating it into their line of business.

Correspondent banking amid a global pandemic
Globalisation, digitisation and the Covid-19 pandemic have combined to create increased reliance on cross-border payments and correspondent banking over the past year.Unfortunately, most banks simply cannot manage the risk associated with correspondent banking. There are just too many unknowns involved in each transaction.

The ability to identify and screen your customers and have full visibility into the flow and destination of a transaction is paramount, and the first priority any bank needs to take when processing payments.

GlobalData Strategic Intelligence

US Tariffs are shifting - will you react or anticipate?

Don’t let policy changes catch you off guard. Stay proactive with real-time data and expert analysis.

By GlobalData

Over the past decade, an increasing number of banks have incurred significant fines for failing to identify and remediate money laundering in cross-border transactions.

And the fines are just the icing on the cake; several banks have taken severe reputational hits due to media coverage of events like the FinCEN files leak, the Russian Laundromat, and the Danske Bank controversy.

This has led to banks of all sizes completely stepping away from the correspondent banking landscape, as they just deem the risk too great for the potential reward.

For banks large and small, this leaves a mammoth business opportunity on the table. They have a new line of business waiting to be harnessed – one that Juniper Research predicts will facilitate $35trn in payments by 2022 – if banks were able to shed light on the full transactional flow.

AI as a solution to risk reduction

Banks traditionally rely on SWIFT and KYC data to collect relevant reports and documentation for all of their traditional banking relationships.Unfortunately, when factoring the number of banks involved in a typical correspondent banking network, this vast amount of data becomes completely unmanageable.

For example, if a bank has 1 million correspondent banking customers and 20 banks in the network, a bank is looking at 20 million different reports.

To make matters worse, each bank in the network can have disparate information requirements, meaning those 20 million reports could be completely different from each other.

Sounds like an analysis nightmare. It is for humans, but an AI-based solution can automatically process and analyse the data.

The ability to access this data for cross-border payments enables banks to ensure that the funds are not coming from known criminals, unknown threats that raise suspicion, high risk geographies or questionable shell companies.

Transaction monitoring and transparency

Once banks are able to fully utilise their SWIFT and KYC data for correspondent banking through the usage of AI, they must incorporate it into a robust transaction monitoring system that provides full visibility into the lifecycle of every transaction.Traditional AI is unable to provide full transparency; for this, you need an advanced type of unsupervised machine learning that looks not at the data points, but at the relationships between the data points.

This enables it to uncover sophisticated criminal schemes hidden within transactions that appear perfectly innocent to other AI and rules-based systems.

This advanced AI helps to combine SWIFT and KYC data and other relevant data points, offering a bird’s eye view of any transaction that helps analysts quickly decipher which transactions are legitimate and which are suspicious.

This level of visibility helps to identify any notable anomalies that arise, quickly thwarting criminal schemes that may have otherwise gone unnoticed.

While having the right solution for your needs is critical, it’s also important to properly educate analyst teams on what they need to know to accurately manage the risk involved in this particular line of business, including ways to spot suspicious transactions and how to properly use transaction monitoring capabilities.

A chain is only as strong as its weakest link, therefore if a bank’s staff is not properly trained, new tech will not do much good in the long run.

Although financial institutions are often hesitant to adopt technological solutions that they fear may increase their business risk, this is an instance in which tech adoption can actually reduce the risk involved.

Advanced AI solutions enable banks to safely and easily manage risk and tap the potential of correspondent banking.