PayThink

Coronavirus stresses traditional payment-related credit models

Register now

There was a time when a high credit score provided lenders with an increased level of confidence that the creditor will repay future debt, but given the growing number of newly unemployed and failed businesses in the post-COVID-19 environment, credit scores may no longer be indicative of the consumer’s ability to make payments or fulfill their debt obligations.

In this first phase of the post-pandemic era, the expectation is that financial institutions will pursue basic remediation efforts to make sure credit scores are still relevant by adjusting their algorithms and lending parameters. Some may take additional measures by lowering tiers for acceptable scores, extending existing grace periods, and waiving fees and late charges. Even simple measures — like not reporting credit scores to the credit companies — may help in the interim.

But in today’s environment, this may not be enough. Old, inflexible models for determining creditworthiness may hinder viable customers from getting loans, putting loan portfolios at risk and jeopardizing revenue potential.

There is no substitute for the lender getting to know the customer and making the lending decisions on that basis. A more layered approach to Know Your Customer is needed to develop risk assessments on customer creditworthiness.

New credit models for risk assessment will be based on transaction history and behavioral profiles, supporting more data analysis on the client and their banking interactions as a whole.

There is much to be learned just by examining existing information available through internal sources. Financial institutions should start by taking a heuristic look at their existing relationship with the customer, segmenting products into digestible pieces — leases, loans, equity lines and even DDA accounts.

Lenders should also examine existing data points like risk history and BIN transaction files to assess customer creditworthiness. Transaction history provides insights on customer patterns including typical time of day for purchasing, payment types used, channel preferences, geocodes and preferred merchants. Lenders should be examining this data to detect transaction activity and/or spending habits that are outside the norm. Google, Apple and Facebook have perfected the art as they understand the movement and activity of the customer and thus have more keen insight into the customer’s behavior.

How frequently does the customer change phones? How many channel access points do they use? Are they current with software version updates? These client channel interactions can provide valuable behavioral clues about customers. Apple is a perfect example of a company that understands and executes against lots of criteria based on the customer’s use of the phone.

Future models for assessing creditworthiness will use enriched data to augment customer credit scores and KYC. Fintechs have introduced a different approach to customer risk management. They focus on customer information from a data value perspective based on what the client is doing. For a 30-year loan, they might look at transaction history and the risk score on payment history or other indicators; for a short-term loan, more immediate factors like the client’s history of paying monthly bills might weigh more heavily.

This is the first of two parts. Maria Arminio is President and CEO of Avenue B Consulting; and Bo Berg is founder of Hygge Consulting Corp.

For reprint and licensing requests for this article, click here.