Pandemic Analytics in Banking and Financial Services Industry: Gauging the Economic Impact of COVID-19

Towards mid-April 2020, the Bank of America reported that “its quarterly profits dropped by 45 per cent from a year ago,” as a result of the coronavirus pandemic, and “the amount of money BofA set aside for loan losses nearly quintupled from a year ago, from USD 1.01 billion to USD 4.76 billion.” [1] While many governments and businesses prepared themselves for risk events, especially after the recession in 2008, the sheer magnitude of the ongoing COVID-19 pandemic’s effect on the world economy was unforeseen.

How then is the Banking & Financial Services industry coping with this situation and what role is data science and analytics playing in navigating through challenges that the market is posed with? How are risk managers stepping up to the challenge and working towards managing risk? What are the major challenges that lie in their path as they strive to achieve this endeavor? These are the questions the article delves into – read on to know more.

Credit Risk managers dealing with the economic slowdown

Many managers are setting up as many defensive strategies as possible to deal with the economic slowdown. As noted by Naeem Siddiqi, Senior Advisor, Risk Research and Quantitative Solutions at SAS,[2] on the economic turbulence that lays ahead in the aftermath of the virus outbreak,

“Currently, many risk managers are adjusting cutoffs and policy rules to control credit. Many are using historical default rates combined with expert-based safety factors as guidelines. For example, default rates experienced during the 2008/9 credit crisis can function as a benchmark, but only as a broad indicator given differences in the current situation. While 2008 brought a liquidity crisis, we did not witness massive unemployment as we do today.”

As a result, quantum of pre-approved loans are decreasing, restrictions have been placed on going over credit limit, early defaulters are being tracked more proactively, and in some countries the government is asking people to defer payment on loans and there are people taking advantage of the same. While setting up these defense strategies, Risk Managers are taking quick decisions while striving to help the financial institutions stay afloat.

Governments are trying hard to reboot the economy post the lockdown. With central banks playing the enabler of the policy we could witness downward revision of interest rates across the board to encourage banks to issue credit. This policy coupled with several other initiatives by the government like subsidies on interest rates will help SMEs access cheap credit and resume operations.

Streamline and catalyze traditional modelling set-ups

In an era of economic slowdown, existing models come under severe scrutiny and a fresh look is warranted. New development does not have enough latitude due to limited data and other resources. In such scenarios the modelling processes need to be streamlined with pre-trained contextual solutions, reusing components and modules that are relevant. Alternatively, models need to be adjusted so that banks can prepare for the new reality and the future. We further estimate a significant reduction in the shelf life of the risk models and could lead to a 6-9 month redevelopment cycle as opposed to 12-18 month currently in practice. Banks need to identify bottlenecks in existing operational processes, capture as much data as possible, and reduce time to deployment. Banks that do so, will have a massive advantage over the competition. With new models and a reduced development cycle, bBanks acting now will be in a better position to assess and service customers.

The new models have an added responsibility of being robust and flexible within limits to fit into the oscillatory nature of the industry in the aftermath of the pandemic. We can no longer rely as heavily on global indicators alone, for example – states & businesses opening earlier than other will demonstrate a vastly different behavior. And existing models might not effectively reflect the impact of the economic slowdown as it is a result of a health crisis – a butterfly effect that would not have been anticipated by the models. While methods cannot be straight jacketed now, Banks will have to begin experimentation now to arrive at a structure that works for them.

There is no certainty as to how long the pandemic will be around and will leave a long lastinglong-lasting imprint of the way banks operate. In either case, we strongly recommend leveraging data to paint realistic scenarios and map how the market is transforming

How can AI be leveraged in this market condition?

Many banks will be leveraging AI to move from a traditional hands-on approach to a more digital approach where information and insights can be consumed to quantify risk and reap long-term growth and development benefits. Here are some of the steps financial institutions can take in this regard:

1. Use data to strategize: Proactively identify collect data, derive insights, and leverage them to strategize upcoming policies, talk to the customers to understand their needs, and cross verify the information

2. Tap into unstructured data: Mine data from newspaper articles, blogs and other publicly available resources that provide updates on market movement and customer responses. Perform streaming analytics and text processing on this data to identify companies that might be in trouble or are facing a liquidity crunch or are on a positive uptick.

3. Analyze Customer behavior: Trends predict a significant reduction in discretionary spends, this scenario will continue until there is decisive treatment for the disease. Encouraging customers to spend has tomust be reassessed as the traditional rewards and cashbacks might not have the same impact.

4. Fraud detection: The flipside of many online transactions as people engage in e-commerce and as bank employees work from home, is that there is an increase in the likelihood of fraudulent transaction. As an IDC blog article notes, “Some regulators, such as the Monetary Authority of Singapore, have already mandated split team arrangements. This means that different teams split so that if an infection cluster appears in one, the business unit can still carry on operating.[3]” Apart from this, organizations can also leverage AI to detect fraud patterns and sanitize transactions.

5. Catalyze digital transformation: Today, many firms are struggling with making sense of structured data, but with every passing day, the emphasis on digital transformation increases and failing to keep up with the same might prove catastrophic. Working towards digital transformation is necessary if one wishes to steer ahead of competitors and work towards a technologically-sound future.

6. Real-time insights: Everyday there is a new update about the number of coronavirus cases, the lockdowns being eased or re-introduced. All these factors are impacting the global financial market; real-time analytics is the need of the hour. However, this is easier said than done. Banks take time to get the data needed for analysis, and then take decisions based on the insights derived. Therefore, creating changes in organizational structure to streamline this process is vital for survival.

7. Identify most likely scenarios: Some banks are referring to trends from the 2008 recession timeframe and leveraging stress testing, scenario analysis, etc., to forecast the scenario a year from now to drive decision-making. It is a health crisis – so understandably, many firms do not know what to do. For now, collective decisions by leveraging the resources and combining it with AI and analytics to make sense of the same, can help us get through these trying times.

In this crisis, there is no rulebook one can refer to and readymade solutions one can deploy. While AI can help to skim through complex data and derive insights, human judgement can identify data issues and increase the efficiency of the model. Firms are now leveraging their learnings from the 2008 recession for the ongoing economic slowdown, and this too shall enable us to be better prepared for future risk events.