Correspondent banking is a vital system that helps to connect smaller and underdeveloped communities to the world’s financial ecosystem. In the last decade or so, the need for correspondent banking has boomed across the globe. The global pandemic further kicked this increase into overdrive, with many people needing to transfer money to loved ones abroad. In fact, according to Juniper Research, the cross-border payment industry will facilitate over $35 trillion in payments by next year. However, despite the volume of correspondent banking transactions going up, the number of traditional banks participating in these types of cross-border payments has been steadily declining.
This poses the question, “If the demand for correspondent banking is increasing, how in the world are banks not able to meet the demand?” There are a couple of reasons why this is the case, with the first being the risk involved.
Simply put, most banks are not properly equipped to handle the risk of correspondent banking. These transactions can typically involve up to a dozen smaller banks spread throughout the world, making it close to impossible for banks to conduct proper KYC due diligence and understand who exactly is processing and receiving the money. To make matters even more complicated, each individual bank in the process has its own, disparate set of data that needs to be analyzed by the main bank involved. So for instance, if a correspondent banking network has 10 member banks and one million customers, 10 million unique reports need to be analyzed. This data deluge makes manual review nearly impossible without incurring massive compliance and regulatory risk.
Slow and Expensive Does Not Win the Race
Another factor that’s keeping banks out of the correspondent banking game is the fact that emerging payment service providers (PSPs) are able to process the same transactions faster and cheaper than traditional financial institutions. If you’re reading this wondering how that’s possible after I just told you the vast amount of data that needs to be analyzed, it’s because PSPs like Transferwise and PayPal aren’t required to follow the same stringent regulations and compliance requirements as banks, giving them a boost in the speed department and allowing them to price their services accordingly.
Here’s an example to highlight the processing speed and cost disparity between banks and PSPs: Say i’d like to transfer $1,000 to Estonia from Israel. If I went to my bank to conduct this transaction, it would likely cost me $40 and take a week to arrive at the recipient. On top of that, the bank would likely also require a lot of information on the transaction, including information about its recipient, the reason for the transaction, etc. On the other hand, if I took that same $1000 to Transferwise, they would charge around $10, it would only take a few days for the payment to arrive and, as an added bonus, they wouldn’t require any information about the transaction. Put two and two together, and why would anyone decide to use the slow and expensive bank over the new guys?
A(I) Solution Emerges
So, how can banks address these issues and find their way back into the correspondent banking game? The answer lies in technology adoption, specifically artificial intelligence (AI).
By incorporating an AI-based solution into its anti-money laundering (AML) review process, a financial institution can automatically process and analyze KYC data and the associated reports from each member bank, saving significant time and effort. This also keeps the regulators happy, because the financial institution is able to ensure that the funds aren’t coming from or being sent to criminals, black-listed countries, or shady shell companies.
There are even some advanced AI solutions that can mimic human decision making and ‘gut feelings.’ This isn’t as Skynet-esque as it sounds. This type of AI doesn’t look at the data points themselves, but at the relationships between the data points. This enables it to detect suspicious activity hidden within seemingly innocent transactions. By incorporating this advanced AI into a sophisticated transaction monitoring system, banks are given a bird’s-eye view of each transaction, empowering analysts to easily determine which transactions are legitimate and which are masking hidden criminal activity.
It’s easy to see why this level of transaction analysis would reduce the risk for banks looking to engage in correspondent banking, but it also helps with the speed and cost. If banks are able to analyze data quickly and automatically, payments are processed much more rapidly. Additionally, the money these banks save by integrating an advanced transaction monitoring system into their practice allows them to price correspondent banking fees accordingly. No more long waits and $40 fees? Suddenly, Transferwise has some ‘old school’ competition.
The demand for correspondent banking is clear. Banks know it’s there, but they just aren’t properly equipped to join the party. The good news is that artificial intelligence can be the ticket to accessing this lucrative line of business.