Article | May 26, 2021
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 | May 14, 2021
Digital banking is not new – major banks began to offer internet banking services in the mid-1990s. However, the traditional banking industry is facing significant pressure from rapidly shifting consumer expectations, changing regulations and increasing competition from digital-native disruptors. Younger Gen Z customers are more apt to use alternative transaction methods such as mobile wallets or P2P payments (e.g., PayPal or the Dutch payment app Tikkie), and businesses are beginning to favor real-time digital payments to improve efficiency, reduce cost and better manage their cash flow. Moreover the ongoing global health crisis is accelerating the movement toward real-time contactless digital payments. Fifty-six countries are now live with real-time payments, and six countries more than doubled their volume of real-time payments in the past year. [i] Due to a joint implementation of the major banks led by the Dutch Payments Association (Betaalvereniging Nederland), the Netherlands is a European leader in terms of the adoption of real-time payments.
In the midst of this fast-changing landscape, new business models are arising as digital-natives, FinTechs and incumbent banks partner to offer new banking and payment services in the cloud. One example is Dutch Cobase – a subsidiary of ING Group that bundles business accounts – which recently signed a cooperation agreement with the Nordic bank Nordea and the French Crédit Agricole. Amsterdam-based banking platform Five Degrees supplies its technology to banks such as ABN Amro, Van Lanschot and Knab, among others. Collaboration like this is spurring further innovation as these digital ecosystems expand, attracting new participants. But successfully delivering these new digital services requires the direct and secure, low-latency, reliable exchange of data between partners that interconnection can provide.
BaaS needs FinTechs AND banks
FinTechs born in the cloud have the IT infrastructure, skills and agility to deliver digital banking and payment services on-demand. They can also offer these BaaS capabilities to any brand who wants to embed financial services in their customer experience. Sometimes referred to as “embedded finance,” BaaS enables businesses to create new products and services along the customer journey as the diagram below illustrates.
However, FinTechs typically lack the assets and regulatory license to fulfill financial transactions, and that’s where banks come in. To ensure that deposits and money transfers stay safe, banks are heavily regulated and often insured up to a certain dollar amount for each depositor. This combined with a longer history with customers means that banks have an advantage when it comes to perceptions of how safe and secure a financial transaction will be. As a result, there are a few collaboration paths that FinTechs and banks generally pursue to bring BaaS services to the market:
The FinTech buys a bank that already has a license such as Jiko purchasing Mid Central National Bank in the U.S. or Raisin GmbH buying MHG-Bank AG in Germany.
The FinTech partners with a bank to borrow their license such as Chime partnering with Stride Bank, N.A. and The Bancorp Bank.
The FinTech acquires its own license (a lengthy process that could take up to three years) such as Railsbank in the U.K. or Varo Money in the U.S.
The bank partners with a FinTech to launch BaaS services such as Deutsche Bank partnering with Traxpay to integrate supply chain financing technologies and solutions within its own offerings.
Regulations are shaping the partnering model
The regulatory environment may also impact the partnering model. For example, open banking laws in the European Union and the U.K. require banks to open APIs to third-party developers, making it easier for FinTechs to gain access to bank data. Regulations like these are helping to reduce uncertainty for startups and accelerate innovation in the European banking system. Challenger banks such as U.K.-based Revolut have also benefitted from special licenses that allow them to directly accept deposits, process payments or lend.
In the U.S., the Durbin Amendment is accelerating partnerships between small-medium banks and FinTechs in a different way. The Amendment, which has been in effect since 2011, aimed to lower prices for consumers by reducing the fees that retail stores pay to banks when customers use debit cards. In reality, banks just responded by increasing the fees that consumers pay to make up the lost revenue. However, the Durbin Amendment exempts financial institutions with less than $10 billion, making them ideal partnering candidates for FinTechs.
How BaaS actually works
A hybrid digital architecture for BaaS with a mix of on-premises, colocation and public/private cloud elements. In this example, the bank is the license holder partnering with the FinTech BaaS provider to deliver embedded financial services to a Brand (such as a retailer or transportation business). The bank has also partnered with other FinTechs for real-time and cross-border payments, although it handles any card transactions in-house. Interconnection will be critical for ensuring secure, low latency data flows between the partners and digital infrastructure across the regions where the BaaS is offered.
Partnerships like these are steadily growing into ecosystems of digital exchange around financial services that include clouds, networks, banks, FinTechs, payment rails, fraud detection and other service providers. By placing their digital infrastructure close to these ecosystems, leveraging an interconnection approach, banks and FinTechs alike can maximize their competitive advantage. Interconnection provides a more scalable, reliable, secure approach to moving data between members of the value chain than the public internet. With an interconnection strategy, banks and FinTechs can deploy a digital core, extend across edge locations and enhance their capabilities through digital exchange to create new BaaS markets for any brand.
Article | April 4, 2020
Banks have an Essential Role to Play as Systemic Stabilizers.
COVID 19 has created disruptive economic fallout within human society across all religions/races/geographies/countries/continents. The path ahead is hence a dangerous one, driven by epidemiological uncertainty.
While this situation is occurring for the first time in human history, this has also been an eye-opener to have a more comprehensive look at the way we operate. The footfalls of branch banking have decreased to a large extent, and banks have urged customers to use digital channels.
Table of Contents
• COVID 19- An Accelerant to Digital Transformation
• How to Increase Digital Banking Adoption
- Start with a comprehensive plan
- Keep employees morale up to get back on solid ground
- Leverage digital and traditional channels
- Enable Seamless Customer Experience
COVID 19- An Accelerant to Digital Transformation
The coronavirus (COVID-19) outbreak is indirectly promoting digital transformation, as many service providers, including banks, have significantly decreased or even closed their offline services, asking their customers to adopt the digital banking for any assistance. This trend is resulting in speeding up digital transformation. But, the question here is, are customers ready for this drastic change?
Yes, slowly yet firmly, Banks are witnessing customers picking up services that have distinguished digital capabilities. The mechanics of customer interaction has quickly shifted from physical handshakes to virtual communication over personal & extensive digital channels.
Learn more: https://capital.report/blogs/6-digital-banking-best-practices-during-the-covid-19-outbreak/8287
Consultant McKinsey & Co, in an article titled Leadership in the time of coronavirus: COVID-19 response and implications for the banks, reports that Banks have already taken a series of actions in reaction to the spread of COVID-19. Common steps we’ve seen include establishing a central task force, curtailing travel, suspending large-scale gatherings, segregating teams, making arrangements for teleworking, and refreshing external-vendor-interaction policies. It also highlights some of the fundamental changes banking organizations are going to need to make in the way they do business.
Workplace dynamics and talent management, already evolving in a digitizing world, maybe durably changed after an extended period of remote working,” it says. “Likewise, customer routines and expectations may also shift further in meaningful proportions, both in terms of digital adaptation and the expectation for proactive communication and care.
How to Increase Digital Banking Adoption
Digital is not a destination but it's a journey with more and more innovations and discoveries happening across multiple industries. But with COVID 19, it is a forced change for human society to adapt to a digital and contactless mechanism for business transactions between individuals or companies across the border (B2B, B2C, B2G, G2G, etc). The countries which are adopting this methodology are more likely to succeed in the new digital paradigm post-Covid 19.
Below are some simple steps you can use to increase customer engagement with digital banking:
Start with a comprehensive plan
To increase awareness of your digital banking platform, it is always essential, to begin with, a first cut action plan. In the current crisis, there are immediate actions banks can take to help retail and small business customers, support the use of digital channels so that customers can bank from home.
In the United States, many banks struggle to increase digital adoption among their customers; for example, nearly half of banking customers either never use their mobile app or do so infrequently.
According to McKinsey& Company, In the United States, the most satisfied customers use digital multiple times per week, the second-most satisfied customers do not use digital at all. The least satisfied banking customers are those who use digital tools infrequently, less than once per month. This is because customers go through a learning curve as they adopt digital tools, and most banks under-support their customers in the adoption journey. In the current environment, banks should redouble their efforts to smooth customers’ transition to digital. Here’s how:
• Easy-to-find and clear communication
• segment-specific campaigns
• remote coaching and advice,
• And unified experiences across each journey, such as written and video explanations for how to accomplish specific digital tasks, along with ways to try them out.
Banking services that involve branch interaction, digital tools can still play an important role by providing information on adjusted hours, essential services, reduced staff numbers, heightened safety precautions, social-distancing measures, and digitally-enabled queuing.
Keep employees morale up to get back on solid ground
One of the first things that can go for a toss in such challenging times is the morale of employees. It is natural for employees to feel frustrated. The uncertainty of the company's future, the fear of recession, and uncertainty around jobs can damage motivation, productivity, and can create a lot of fear among employees.
This entails bank managers to set precise directions for remote teams that are pursuing common goals. Bank managers should also take this opportunity to delegate and empower their employees for decision making. Both intrinsic and extrinsic motivation through rewards, clear spans of control, and meaningful appreciation can go a long way in supporting employee morale.
A vital portion of the employees is working from home facing operational challenges, such as internet-bandwidth issues, network connectivity, technology glitches, and childcare priorities. Following practices could help these banks perform a more effective distributed-work environment:
According to McKinsey &Company, following practices could help these banks perform a more effective distributed-work environment:
• Enable technology setup and infrastructure for remote work.
• Supporting remote-work technology and infrastructure
• Assisting employees with home-office setup
• Ensuring adequate VPN bandwidth
• Providing remote application access
• Adopting a suite of digital tools that facilitate effective communication, and decision making, such as videoconferencing, file sharing, real-time communication, coediting, and task management, and
• Ensuring that agents have the necessary tools and resources to handle calls from home while maintaining customer-data-confidentiality standards.
With all these features, it is also critical to ensure that agents are well-equipped with the necessary tools and resources to handle calls from home while maintaining customer-data-confidentiality standards.
Leverage digital and traditional channels
The more frequently a person sees a message, and in more places, the more likely they are to engage and take action. Banks need to maximize awareness by promoting their message across channels. Promote it in their branch and online, on ATM screens and in the call center.
Banks can also find ways to cross-promote digital banking. For example, during the new account opening process, it should encourage consumers to enroll in online banking. Or, if a bank or credit union is doing a credit card promotion, they can use it as an opportunity to cross-promote its mobile app.
Enable Seamless Customer Experience
For banks, spending on customer experience was essential before the current crisis, both from a “good business” perspective and a “good bank” perspective. Now, these aspects are even more relevant. It is highly important for banks to make their genuine concern for their customers clear and to make customer interactions with the bank as easy as possible.
COVID-19 has brought customers already under health and financial stress. They will need ready access to bank products and services. It is now more important, then, to reach customers through digital channels, stay connected through innovative communication channels, meet the needs of vulnerable populations, and stabilize critical infrastructure.
• Banks should encourage more customers to use remote channels and digital products whenever possible.
• Enhancing current digital offerings, identifying key functionalities, that can be improved quickly
• Speeding up the procedure to increase limits on online transactions and simplifying password reset.
• Keeping clients involved via SMS, mobile apps, and digital media
• Minimize disagreeable surprises to customers (such as potential branch lockdowns)
• Encourage fraud-prevention measures, clarify the availability of solutions on digital channels, and
• Define preventive measures to ensure the health and safety of clients and employees in branches.
According to McKinsey & Company, some financial institutions will need to address such technology gaps in order to offer a seamless digital customer experience. This will require planning ahead by scaling infrastructure capacity and network bandwidth, stress testing and scenario planning, managing near-term patches, and identifying urgent weaknesses in architecture.
Learn more: https://capital.report/blogs/9-best-fintech-apps-to-use-while-at-home-during-the-coronavirus-lockdown/8273
All in All
Coronavirus difficulties provide an opportunity for new businesses to thrive based on a new digital reality – completely digital and contactless. Digitalization has found a new meaning and it is going to reach newer areas. The world is thinking about implementing ways to lessen the disruption caused to humanity. This is the perfect time to focus on digital transformation by realizing the necessities accelerating it.
Article | June 4, 2021
Financial institutions have, for a while now, been operating in a highly cost challenged environment. These firms will continue walking the tight rope of executing on efficiency, digital transformation and supporting the business. Post-COVID when our dear planet begins to get back to some form of normality in the months ahead, it does not necessarily assume that wallets will be loser and further budget constraints are expected to be with us for some time. As we know the Genie is out of the bottle on the whole “agile” working theory and the Cloud providers have responded in kind such as providing virtual desktops and VPN solutions. Of course not forgetting the Video calling enablement which has coined a phrase never to leave our vocabulary “sorry I was on mute”.
Cost pressures aside businesses are already reassessing the effectiveness of their technology stacks. I believe we will see an acceleration of an already giddy pace by firms to move parts of their estate and applications to the public cloud. It is not only essential from a practical basis covering the usual themes of cost, storage planning on demand compute etc but if you want to retain the best talent in technology you need to be exposing them to the likes of AWS, GCP and Azure in some form.
Data is the new oil
As to my world in data various analogies “data is the new oil” etc, but getting beyond the taglines the public cloud is shaking up the status quo. From off-the-shelf Amazon style access to data products via a web store or to throw in another term “supermarket”. Fundamentally the barrier to entry for clients to access data, storage and enormous compute resource is really down to what you can afford. Efficiencies on compute, serverless technologies pay for what you use not pay for standby is changing the paradigm in architecture. Thereby pushing boundaries in innovation, experimentation and exposing teams to AI/ML as a utility as opposed to things you read about in journals or online.
No two businesses are the same which is why certain firms are further in the journey than others. But regardless of the path financial institutions decide to go down, it does not change the fact that data needs to be delivered to the right place, at the right time, and in a preferred format. Some firms will simply want their channel partners to ship data into the cloud as an end point.
From Satellites to the Cloud
This leads me into my next comparison. I was lucky enough (or unlucky) to be there when the internet created another paradigm shift as a delivery end point for data. Prior to that I spent many years plugging firms into Satellites or Leased lines for the delivery of Market Data. As a younger man I thought those days would never end! If the internet became the end point that people used to get data into their own network, then the cloud to a certain extent is the modern day equivalent. After all, if firms want to use cloud as an end point into a physical data centre or on-premise, they can do that. Alternatively, if the firm wants to use the data exclusively within the cloud, then that is also achievable in this day and age.