As banks steer their focus from the Paycheck Protection Program back to regular small-business lending, they are struggling with a problem: how to accurately monitor a customer’s financial health as the coronavirus pandemic continues.
The data that banks traditionally rely on to vet the health of small-business customers, such as financial statements and tax filings, are less reliable in the age of COVID-19 because it can be months old and, as business conditions rapidly change, can quickly become stale.
But some banks are using technology to analyze alternative sources of data and devise new methods to assess the health of small businesses, especially as they face an influx of new customers. This could include the use of scores that look at how geography, industry and other qualities could affect a small business's likelihood of failure, ratings that look at individual businesses rather than industries as a whole and software that analyzes a loan portfolio for signs of distress.
“A lot of traditional sources of financial verification don’t address what has been happening in the last three to six months,” said Desiree Wolfe, director of product management for community banking at Webster Financial in Waterbury, Conn.
The ability to analyze a business client's health has become more urgent as many banks have suddenly taken on large numbers of unfamiliar clients through the PPP.
Garry Capers, the head of cloud solutions at Deluxe Corporation, a firm in Shoreview, Minn., that aggregates real-time data about small businesses for its bank customers, found that the PPP sparked a willingness among small businesses to shop around for new banking relationships. As a result, he says, some banks acquired clients that they didn’t target using normal procedures, and now must decide whether they want to continue the relationship or end it at PPP loan repayment.
“A number of our banking customers are asking how to think about these new relationships, identify those which are potentially profitable and sustainable, and find ways to grow the relationship,” Capers said. For those businesses that are not sustainable, he says the question becomes, “How do we manage that emerging risk within our portfolio?”
The subset of brand-new relationships that were born out of PPP is relatively small, but still presents a problem for banks that need to find new ways to monitor these customers. A small-business banking study from J.D. Power in October found that the intent to switch banks is 17% higher among small-business customers who applied for and were declined a PPP loan at their primary bank than those who were approved.
Numerated, a digital lending and sales platform for business banking, estimates that a fifth of its banks aggressively sought out new relationships from the Paycheck Protection Program.
Beyond vetting new customers, banks have also had to examine the health of their existing small-business clients, determine which ones can survive past the pandemic, and proactively mitigate risk to their portfolios, by offering payment plans, refinancing loans or discussing other needs.
Ken So, founder and CEO of Flowcast, a machine-learning platform that helps financial institutions make credit decisions, has seen banks rely less on traditional tax filing data and more on alternative sources of data, including transactions and bank account activities, to get a real-time understanding of the health of the business during the pandemic. Kabbage, OnDeck and other fintech lenders follow the same formula.
If a bank sees that a customer is at risk of missing payments, it can more proactively offer refinancing or an interest-only loan to mitigate the risk, said So, whose company built a model that predicts these incidents.
Flowcast works with three big banks and more than a dozen mid-tier financial institutions and nonbank lenders.
Capers has a similar view on the importance of real-time data.
“The concern is that the legacy data used to target prospects and perform portfolio risk management is not current, comprehensive or dynamic to adequately reflect the impact of something like COVID on small business health and survivability,” he said.
Deluxe aggregates real-time data about small businesses for its bank customers in a number of ways. For example, it analyzes the health and growth trajectory of the 4.5 million small businesses that use Deluxe’s products, such as website hosting, e-commerce tools and payroll services, to see if the business is bringing on new employees, ordering more promotional items or seeing an increase in e-commerce transactions. Deluxe also layers in information from multiple third-party data providers.
In March, Deluxe launched a COVID Index to identify which businesses are at higher or lower risks of failure, based on their geography, industry and financial attributes. It uses this data to help banks determine which businesses have a higher risk of failure and adjust marketing campaigns, such as deciding whom to target with new credit.
Wolfe of Webster has been using data from Deluxe and other partners to assess the impact of COVID-19-related economic pressures on its small businesses and their abilities to sustain themselves in the pandemic. Relationship managers use this data as part of periodic health checks, when they proactively reach out to our business clients to review their current financials, discuss upcoming or potential needs, and make recommendations.
Starting in the third quarter, the $33 billion-asset Webster resumed its regular customer servicing and marketing efforts outside of PPP lending. “We use partnerships like Deluxe to add additional filters when assessing the business’s ability to sustain itself during COVID-19,” Wolfe said. “We also use them to help us manage our portfolios and acquire new relationships.”
To help its customers adapt and survive, Webster has shared services with its clients to help manage cash flow, such as the ability to take card payments curbside or in-store through Webster Merchant Services and or process payroll through Webster’s payroll tools.
Dan O’Malley, CEO of the Boston-based Numerated, says that when deciding to extend new loans, banks are relying less on a certain debt service coverage ratio based on historical financial information. Instead, they are flagging businesses such as restaurants that are most heavily touched by the pandemic for a deeper review, and placing more weight on a business’s ability to repay a loan based on the last three months’ of cash flow rather than the previous year’s tax return.
Customers Bancorp, in Phoenixville, Pa., first
OakNorth's “COVID vulnerability rating” identifies which businesses could withstand six-week, three-month or six-month shutdowns. The platform also offers a companywide view of credit quality and industry exposure.
The $17.9 billion-asset Customers also looked at each asset class under management and considered the length of time each business needed to survive the impacts of the pandemic. As a result, Customers could take proactive measures such as encouraging borrowers to top off their interest reserves, or additional escrow accounts that fund interest shortfalls, offer niche stimulus or government assistance programs, or extend extra credit even if the borrower doesn’t need it immediately.
This automated loan-by-loan review "allowed us to be a lot more nimble and manage risk a lot better,” said Sam Sidhu, Customers' vice chairman and chief operating officer.
