Banks are squandering £2.7bn a year because of outdated anti-money laundering (AML) systems – costs that could be saved by adopting machine learning and big data technology, new calculations show.
According to FortyTwo Data, an AML technology company, financial institutions are wasting armies of staff on chasing millions of false leads – red flags that turn out to be innocent – generated every year by legacy systems that rely on stale rules and scenarios.
The firm analysed the likely impact of machine learning and big data technology using first hand intelligence on 22 of its financial services clients and partners over the last two year and the success rate of its own platform in reducing ‘false positives’.
It concludes that, on average, 55% of false positives can be eradicated by modern systems, accounting for 42% of banks’ cost on AML compliance.
FortyTwo Data refers to figures from WealthInsight, which predicts that global spending on AML compliance will hit £6.4bn billion this year. The potential savings thus equates to £2.7bn.
“That’s a colossal amount of time and money that is being wasted chasing ghosts,” says Luca Primerano, head of strategy at Fortytwo Data. “These advances in regtech represent an opportunity for companies to help reduce the ability of criminals to exploit financial networks for money laundering and terrorist financing around the globe. Machine learning has the potential not only to deliver justice but a multi-billion-pound refund for firms every year.”
As reported in GTR’s latest cover feature, regtech (regulatory technology) is gaining more attention from banks, venture capitalists and regulators, who see the huge potential for new technologies to make compliance – an area dominated by manual and mundane processes – more efficient.
This is especially the case in the trade finance space, which has historically been characterised by paper-heavy processes and the ineffective use of transaction and counterparty data in risk identification.
Big data and machine learning platforms, which allow for large data sets to be analysed to reveal patterns, trends and associations, can enable banks to monitor transactions in real time and improve the identification of unusual activity.
FortyTwo Data’s own platform, for example, uses both supervised and unsupervised machine learning techniques, Primerano tells GTR.
With supervised machine learning, the platform learns from a feedback loop with a human user. When an employee receives an alert for a transaction, they can tell the platform why the activity is not suspicious – information that the platform will later use in its identification process. “This allows the pattern recognition to continually evolve to decrease the number of false positives generated,” he says.
Unsupervised machine, on the other hand, can detect out-of-the-norm behaviour that may not be caught by knowledge-based rules or human review, as well as new types of unusual behaviour. “This is important in anti-money laundering where new money laundering techniques are invented and organisations need to detect and respond extremely quickly,” Primerano adds.
The enthusiasm for regtech is high at the moment, but whether new technologies can help financial institutions cut staff and compliance spend is still to be seen.
FortyTwo Data predicts that compliance departments adopting machine learning technology will shrink by an average of 75% by 2025. Other market players take a more modest view on the potential impact of machines in this space.
In GTR’s regtech report, Richard Davies, HSBC’s global head of propositions, commercial banking, said that it’s a “simplistic statement” to say that regtech can help cut staff and costs, emphasising that HSBC’s main focus is on making compliance processes more convenient for the customers, rather than cutting staff.
And looking at banks’ spend on compliance, nothing is, as of early 2017, indicating that it is even close to taking a downward turn.
HSBC, for example, has been adopting regtech solutions since late 2014, and yet, the bank’s total amount spent on compliance and regulation was at US$3bn in 2016, up US$0.4bn from the year before.
Marc Andrews, vice-president at IBM’s Watson Financial Services Solutions, also emphasised that regtech is more about being able to focus human attention back to revenue-generating activities, while creating better stability and security in the financial markets.
“No bank I’ve ever talked to thinks that we will get to a point where it’s completely automated machines making the decisions, but most of them are hopeful that they can get to a point where they can dramatically reduce the busy work and focus on the high-value critical business decisions,” he said.
Read the full regtech feature here.