October 27, 2020 by Udi Ziv, CEO of Earnix
As part of its announcement on interest rate policy, the Federal Reserve of New York stated that “sustainably achieving maximum employment and price stability depends on a stable financial system. The new interest rate policy framework will, at least in part, determine how the United States will recover from the COVID-19 economic fallout”.
Globally, financial institutions are grappling with dynamic stock market fluctuations caused by the pandemic, and it remains to be seen what the long-term impact on the stability of the United States’ financial system will be. What we do know is that the preparedness of US financial institutions to serve consumers in an era of high unemployment and low wages, and the impact of government economic assistance, both indirect and direct, on consumers, will be the key indicators.
A few weeks ago, New York Fed economists released a report assessing which borrowers are benefiting the most from debt relief and the Coronavirus Aid, Relief, and Economic Security Act (CARES Act). Interestingly, the report pointed out that “mortgage benefits are not automatic; mortgagors must actively seek out these benefits by contacting servicers and proving financial hardships”.
It is obvious what that means for consumers, but less so for banks. How do banks contribute to stabilising the economy at a time when many borrowers are experiencing or will be experiencing financial hardship? How should banks provide consumers with what they need and can afford, at the time when they need it, and in a way that does not pose a significant risk to the financial industry?
Financial institutions need predictability. Making smart decisions in the financial services industry requires assessing real-time data and using that assessment to create product offerings for consumers that reflect and proactively recognise their changing needs.
AI-powered personalisation is imperative
The COVID-19 pandemic has given rise to the need for consumer access to loans and credit, which has been challenging for banks. The problem lays in their inability to leverage the existing data to guide their decisions to lend. Consumers needs are at a peak, and without analytics and sophisticated pricing, banks are at a loss. First, banks must shift to thinking of consumers as individuals, as opposed to consumer market segments.
By tailoring a loan to a consumer’s needs and preferences and in a way that reflects their unique financial situation, banks can achieve a level of certainty in their decision to lend. This requires sophisticated use of consumer modelling which allows for a high level of product personalisation that banks have no way of doing using manual processes. Advanced modelling methods using artificial intelligence (AI) and machine learning techniques enable banks to be more agile as they adapt to changes in government loan assistance offerings, credit options and rates.
Immediate market response
Market fluctuations are dynamic and fast. Many risk-based financial institutions, such as banks and insurers, still base their prices on statistical analytical models that cannot adapt to the market at the speed that the market now requires. They have a limited basis upon which to determine the pricing of their products, which in and of itself increases risk.
Even basic consumer segment-based product personalisation no longer applies. Any insurance company incorrectly pricing a product and selling at a large volume is at risk of going out of business. Despite low interest rates, some banks constrained lending reducing consumer access to credit due to concerns about delinquency levels. Analytical systems that leverage advanced statistical methods provide a viable solution to this problem.
Systemised solutions have the ability to monitor market performance and update pricing strategies against market rates in real-time. This enables financial institutions to make minor adjustments to prices, in defined thresholds, to more precisely meet consumer needs and offer a loan that reflects changes both in the market and in a consumer’s financial situation. However, the key to accomplishing this is not to simply have solutions that monitor market performance, rather, it is in the ability to implement it as part of a financial institution’s operations in a timely manner.
Being proactive, not reactive
Deployment of AI and machine learning allows financial institutions to determine what financial product the consumer needs and prefers and do so proactively. Now more than ever, banks should recognise consumers’ financial hardship and assess their ability to afford a loan based on their new financial circumstances.
Banks can employ enterprise-wide analytical models that can detect when a borrower is experiencing financial hardship, based on late or missed payments, and proactively offer refinancing options, as opposed to letting the consumer default on a loan. That does not mean that the loan adjustment will not extend the term of the loan or have a higher interest rate. It means that recognising when a consumer is struggling to meet the terms of their loan and acting on it, can instil trust in the bank’s ability to continue to serve the consumer through thick and thin.
The COVID-19 pandemic is driving the need for financial institutions to advance their business velocity, to be more agile and dynamic in how they operate, whilst being sensitive to changing consumer needs. The use of smarter analytics can enable that and in a way that allows banks to externalise its value almost immediately.
It comes down to offering products that reflect consumers’ financial situation or risk affecting the business’ bottom line. Tailoring a product to a single consumer, as opposed to a consumer market segment, is not easy, but it is not impossible. Artificial intelligence alone is not the solution either. But the real-time deployment of AI and analytics to bring the most value to that consumer, at that particular moment in their life, is a good way to start. That way, our financial institutions can become sustainable during uncertain environments, and beyond the one we are currently in.