For many of us, our main contact with anti-money laundering (AML) activities is the need to prove our identity before we can open a new bank account, or buy a house. It may be annoying, but it’s just one of those things that has to be done.
But for governments, fighting money laundering is a big issue. It is a vital tool in the fight against international crime and terrorism (CTF – Counter Terrorism Financing). National and international anti-money laundering legislation now reaches into industry sectors ranging from financial institutions to international trade.
Fighting money laundering: an increasing challenge
This legislation against money laundering has become increasingly stringent. It places the responsibility for detection and avoidance of money-laundering on the shoulders of the institutions and organisations involved, and not on the police or other crime-fighting bodies. Banks and other bodies are expected to know their customers and report any suspicious activity within 60 days. Detecting bribery and corruption is now part of anti-money laundering too.
The rhetoric is about not supporting international crime, and the reputational damage is potentially huge for the institutions involved. It’s not just reputations that are suffering either: so is the bottom line. Financial settlements with regulators for non-compliance carry heavy financial penalties.
At the same time, however, international criminals are becoming increasingly sophisticated. From simply following banking rules to stay ‘under the radar’—for example, keeping transactions below a threshold limit—they have now moved to using international trade as a way of money laundering.
It is perhaps unsurprisingly, therefore, that banks, financial institutions and other responsible bodies have had to take action to get better at detecting and preventing money-laundering. Many of them have turned to analytics for help.
How analytics can help
There are a number of key ways that new and advanced analytics platforms can help in anti-money laundering activity.
Bringing together and analysing huge amounts of different types of data from multiple sources. Data useful in anti-money laundering include a bewildering array of information, from publicly-available sanctions lists, through communications with customers, and customers’ identity data to transaction and application data, which help institutions to understand customer behaviour online. In international trade, the range of documents and sources is even greater, including text, figures, and paperwork. Earlier applications, including data lakes, struggled to deal with this much information from so many sources, even though these platforms were considered quite sophisticated at the time. New platforms allow much better integration of multiple sources to enable more complex analysis.
Delivering results quickly and offering adaptability. Anti-money laundering activity, like fraud prevention, is most useful immediately. It is unlikely to be relevant three months later, because those involved will have moved on. New analytics systems can examine and analyse data in close to real time, and certainly in minutes not weeks. The analytics process can also be run iteratively, learning from experience. This enables much more rapid change when that is necessary to meet new requirements, whether these are regulatory or simply the result of adapting to new criminal modus operandi.
Reducing compliance costs. The faster processing time offered by new analytics platforms means that running and maintaining anti-money laundering systems is cheaper. The compliance is also more likely to be more effective, reducing the risk of big fines. This reduction in costs and risk makes any upfront investment in new analytics systems easier to justify. Another cost reduction enabled by analytics is the reduction of false positives to reduce the work of investigators. New tools enable transaction monitoring optimization approaches and data labs to increase the pertinence of the alerts of existing systems.
Generating new insights. The ability to handle significantly larger volumes of data also hugely increases the potential to detect anomalies and inconsistencies. And it is on these that fraud and money-laundering detection is based. Unusual transactions, odd behaviour, incoherent customer data—all these add up to potential issues that require investigation. New platforms give much more potential to see these, and generate new insights as a result as well as providing contextual information to pinpoint real issues.
Making it easy to see the answers. Current generation analytics platforms often have outstanding data visualisation tools built in. With the best will in the world, even the most tech-savvy exec may struggle to read the outputs of an analytics system without good visualisation. And not seeing the problem means not taking action to generate a solution.
Sophisticated problems require sophisticated answers One of the problems in fighting crime is that the criminals are usually one step ahead. This has frequently been the case with money laundering and its detection and prevention. Models are not perfect, but the use of new and more advanced analytics platforms has the potential to allow anti-money laundering activity to stop playing ‘catch-up’. Unsupervised models help identify new and unsuspected behaviour that had never been seen beforehand.
