Article | March 11, 2020
The banking and financial industry is changing at a blinding pace. Advancements in mobile payments, digital currencies, blockchain and innovative lending strategies are causing a radical shift and have challenged nearly every established convention. These new financial technologies (known collectively as “fintech”) are giving rise to new jobs while also changing existing roles and completely displacing some positions in the industry.
Article | March 11, 2020
The enormous amounts of data accessible to banks and their high demand for forecasting make the financial industry a perfect area for machine learning (ML) to shine. In this article, we explore the current applications of machine learning in banking when it comes to risk management, define its challenges and provide a future outlook.
Credit Risk Management
For the past few decades, banks have mostly used logistic and probit regression models for credit risk assessments and internal risk management. However, all conventional models inherit the same flaw — they predict outputs based only on linear relationships between input variables.
This limitation was exposed in the catastrophic 2008 housing market crash. Although the crisis’s negative consequences have been multiplied by uncontrolled sales of credit default swaps and other complex financial instruments, the fundamental reason for failure was in the inaccurate credit risk model.
In the aftermath, with the intent to force financial institutions to provide more detailed reports, The Federal Reserve’s CCAR now requires banks to account for more than 2,000 economic attributes. Consequentially, this also led to other regulating authorities introducing new standards that improve supervisory data quality and reporting.
At the same time, with the proliferation of banking apps, social media, and digital communication overall, financial institutions now collect lavish amounts of unstructured data. If gathered and processed correctly, these new datasets can help gauge critical insights for a wide range of banking operations.
This is where machine learning comes into play. More advanced non-linear approaches to credit risk modeling including neural networks enable banks to make predictions with a previously unseen level of accuracy and granularity.
The utter superiority of machine learning over traditional credit risk modeling approaches comes at the cost of the prevailing ‘black box’ problem. While we can decide to trust ML algorithms based on statistical evidence of their feasibility, current regulatory constraints won’t allow it to happen.
However, machine learning can still be used to a great extent while being regulation-compliant. Even simple linear machine learning approaches still yield more accurate results than conventional ones. Many banks also use unsupervised machine learning methods to explore data, while using traditional classification and regression models to make predictions.
Fraud Management and Surveillance
Nowadays, the majority of banks’ fraud detection systems use rule-based approaches. This causes banks to deal with a significant number of false positives, forcing them to spend inordinate amounts of resources to distinguish meaningless behavioral deviations from real threats.
The ability of machine learning to capture subtle trends and uncover non-linear relationships allows banks to get a complete picture of a client’s activity and significantly lower the probability of false positives. For example, by integrating ML into its fraud detection model, Danske Banks managed to reduce false positives by 60%.
Similar to many other AI-based solutions in the financial space, the biggest adoption hurdles concern regulations and the unexplainability of AI systems. For example, depending on the jurisdiction, banks are often unable to provide developers with sensitive information related to past breaches. Next, the outputs of unsupervised monitoring systems sometimes can’t be explained, which makes them non-compliant.
However, financial institutions have found a way to at least partly leverage the power of ML for fraud management. A fraud prevention system’s alerts will still be triggered by rule-based models, but the integration of an ML algorithm on top of them can allow adjusting surveillance methods to a person’s behavior fluctuations. Such ML models are typically less complex and explainable, which makes them applicable in a regulatory context.
Article | March 11, 2020
The current global pandemic has changed our mindset and habits, as we are forced to revaluate the current ways we do things by thinking further outside the box. Over the last 12-18 months there has been a complete contrast of fortunes in the capital markets technology sector, with some firms flourishing, some struggling to survive, and others having to reinvent themselves to do so. Here at GMEX Group it has presented a substantial opportunity for innovation, which continues to accelerate on the back of the momentum already built.
One such opportunity centres on digital market infrastructure-enabled digital assets which, despite near-term market-driven volatility, will continue to experience increasing demand for solutions and services from institutional capital markets firms.
Market Infrastructure Evolution
GMEX Group started as a FinTech company over 9 years ago and focused on supplying technology to traditional exchanges and post trade operators based on a partnership-driven approach. Over the last few years, as a growth-stage company, we have focused on both digital market infrastructure solutions (including issuance, exchange trading, clearing, settlement and digital custody for digital assets), as well as continuing with traditional market infrastructure enablement. Our hybrid market infrastructure approach has enabled us to deliver technologically advanced, institutional grade, future-proofed solutions that take advantage of the inherently positive characteristics of both traditional and digital market infrastructure.
In today’s environment, exchange matching engines, digital trading platforms and post trade systems need to embrace a hybrid ecosystem approach. Bridging the gap between traditional and digital capital markets, whilst effectively mapping to evolving regulatory frameworks, is essential. This requires an approach which encompasses traditional and digital assets, digital currencies, security tokens and digital securitisation of traditional assets including derivatives and commodities. The increasing regulatory requirements for digital asset infrastructure and the resultant demand for solutions that are fit for purpose has played into our core strengths.
We’ve worked to provide a complete hybrid market infrastructure product suite called GMEX Fusion, which is ideally suited for regulated exchanges, trading venues, custodians and banks focused on both traditional and digital assets of all kinds. Our solution set has been designed to support the latest technology and business challenges that are impacting the way traditional exchanges are looking to operate as they look to embrace digital transformation. GMEX Fusion also addresses the demands from the cryptocurrency exchanges, digital asset trading venues, Non Fungible Token (NFT) marketplaces and emerging markets looking to start-up or enhance their exchange ecosystem and support digital assets. GMEX is working with many of these entities across multiple jurisdictions as our footprint is very much global, with clients and partners all over the world.
The Fourth Industrial Revolution (4IR) is driving technological innovation in many spheres, and with it comes the need to move from analogue to digital - and embrace Exchange 4.0. The industry-changing network will see exchanges, trading venues, post trade operators, custodians, and other services interconnect more seamlessly, with the ability to swap services and assets across jurisdictions and across different types of users. This transformational solution will necessitate digital exchange trading systems, order matching engines and post trade platforms to transition from the legacy solutions that have been around for decades.
We are now moving past the second and third generation of blockchain in financial services towards Exchange 4.0 at an accelerated pace.
As an industry, we’re in a state of flux which has merely been exacerbated by the crisis. If we look at FinTech firms now, I would argue it’s the most exciting time ever because so many new technologies are emerging. With blockchain on the one hand, and AI, Internet of Things (IoT) and quantum computing on the other. From being nascent, many of these are now starting to grow as well as integrate. With all this technology around, the opportunity for innovation is immense. But that’s counter-balanced by the inertia of existing legacy platforms, processes and mindsets.
We know how the smartphone revolutionised the way we communicate, online and in every other fashion. Even 10 years ago, we couldn’t have envisioned where we are now and the extent to which it’s developed.
We are now in the same place in financial markets. We don’t necessarily see it and despite the innovation there are many silos which don’t talk to each other effectively. There is strong client demand for the full spectrum of digital and hybrid services. However interoperability and time to market remain a challenge, with traditional and multiple types of blockchain-enabled digital market infrastructure being severely fragmented.
The team at GMEX group firmly believe that digital market infrastructure and related services need to integrate with existing market infrastructure and technologies to foster interoperability. By doing so there is an opportunity to interconnect the whole capital markets value chain of participants across international nodes (jurisdictions), to more easily trade, clear and settle traditional assets and digital assets and eradicate the age-old exchange silos.
The immense opportunities
As unfortunate as the current crisis is, it will end and immense opportunities will follow once normality resumes. There is expected to be exponential growth in digital assets over the next five years, with a continued uptick in institutional demand. This is not only a huge opportunity for FinTech firms, but also a big opportunity for financial markets firms and those that provide financial services.
To GMEX, this presents an opportunity where the right answer isn’t the traditional status quo and it isn’t the decentralised Wild West. The right answer is somewhere in between, and that presents an opportunity to create new products, new asset classes and new revenue streams.
The ability to harness hybrid market infrastructure will be essential in the capital markets sector, irrespective of whether the underlying asset class is traditional or digital. And to achieve the winning position, innovation now is key!
Article | March 11, 2020
The axing of third-party cookies by Google and the other major browser companies will require a major readjustment by financial services organisations.
The decision, coming fully into force next year, will effectively choke off the data that has enabled personalisation, optimised website interactions and driven much internet advertising. It is no comfort that the browser companies have acted because of fears about infringement of privacy and data protection legislation such as the EU’s GDPR (General Data Protection Regulation) and the CCPA (California Consumer Privacy Act) in California.
The move will affect how UK financial services organisations interact with millions of people. More than three-quarters of Britons now use online banking and 14 million use digital-only banks, expecting a slick, light-touch interaction. So it appears that just as many people go digital, financial services organisations will no longer have access to information they need for personalisation, being unable to track where customers go on the internet after they have visited a bank’s website. All that data about individuals’ habits and preferences will be unavailable.
It seems catastrophic, but in reality, it is not. Financial organisations have a new opportunity to radically improve how they interact with web visitors and customers. AI-powered behavioural analytics offer far superior, real-time capabilities, using the data from the first-party cookies on their own website domains and where available, data from customers’ transaction histories.
The result is a solution that is faster, more accurate and responsive than conventional technology relying on cookie data owned and stored by third-party organisations. Instead of relying on such data for relatively rigid profiling and personalisation, behavioural analytics enables real-time interactions based on a more dynamic picture of how an individual’s requirements are changing.
The technology analyses all the browsing characteristics including time on site, speed of movement and page views, as well as more obvious features such as interest in specific products. Historical data added to the analysis includes what customers did on previous visits and the interval between those visits, establishing patterns where possible.
The flexible advantages of behavioural analytics hubs in financial services
Segmentation allows a bank to identify customers as soon as they arrive on its site, according to whether they are a new or existing customer. Their behaviour then indicates what they want. Knowing what customers are interested in is important. Customers visit financial services websites for a host of reasons – from seeking information, to opening accounts, exploring loans and mortgage offers, making or setting up new payments. They may also want advice about investments and savings, pensions or small business finance. Almost all of these requirements involve quite complex mental processes which financial organisations can influence while consumers are on their sites.
Collecting the data is not difficult – the skill is in making it actionable in an effective way, replicating the ability of a perceptive employee to read a customer’s state of mind. Banks can do this by setting up a behavioural analytics hub to understand what a customer’s behaviour means and how it can be optimised.
Using customised parameters, the hub will, for instance, trigger a screen notification that prompts the web visitor to fill in a form requesting an appointment. In the case of existing customers, the technology can correlate health insurance offers with spending on fitness, and, in general, savings and investment recommendations can be tailored to the client’s concerns or goals as revealed by their navigation of a bank’s website or mobile app.
Banks can set up analytics to see when consumers are behaving in a way that indicates they about to leave the website, allowing them to intervene with a notification that could include an offer. This provides a positive outcome and avoids the blanket use of offers that undermines profitability.
It is a more sophisticated and personalised approach that avoids annoying pop-ups or recommendations that fail to match individual preferences. As part of a single AI-powered segmentation platform, the technology enables banks to personalise marketing content in SMS messages and emails sent to consumers (who consent), which deliver far better results through precise targeting.
Solutions for last-mile interaction in the open banking era
The single platform approach also has another major advantage. It is much easier to implement and far more efficient and streamlined compared with separate solutions for different parts of the customer journey.
The benefits of using AI-powered segmentation solutions should be part of the financial sector’s broader strategy to transform its systems for the open banking era as we approach the end of third-party cookies. For established banks, the reality for some time has been that complexity of systems has undermined their ability to deliver a high-quality last mile. This they can now address without huge disruption or investment.
The alternative is for financial services organisations to become lost on an ocean of data, losing track of customers. Behavioural analytics will bring banks new insights into customers that surpass third-party cookie data, being actionable and accurate and in real time. To provide a streamlined and profitable experience for themselves and their millions of customers, banks must now employ the latest advances in AI-powered behavioural analytics.