Credit card lending is a pretty fair game; you probably could not identify a good example of red-lining or discriminatory lending. Part is due to the 1974 Equal Credit Opportunity Act of 1974 and also the practicality of credit card lending. Credit card lending does not care where you live, what your values are or lifestyle is, or much anything more than your credit score, payment history, and ability to repay.
Volume and numbers drive the credit card business. If you slot into a certain set of underwriting criterion, you get a pre-defined line of credit.
An opinion piece in today’s Credit Union Journal talks to using alternative data at credit unions. We disagree for a variety of reasons. The claims are:
- But its benefits extend far beyond risk mitigation, with advanced alternative credit scoring models providing the basis for extending credit to roughly 6 million consumers who were previously unable to attain a credit score.
- The Consumer Financial Protection Bureau (CFPB) estimates that 45 million Americans struggle to secure a loan because of insufficient credit history.
- The bureau also found that one in ten adults have zero credit history with any of the “big three” credit bureaus. Alternative data can help them create digital footprints — and is now seen as a chief driver in enabling financial inclusion — that financial service providers can capture and analyze to reach consumers with commercially viable services.
That is all fine and dandy. However, credit cards, whether issued by large banks, middle tier community banks, or credit unions, must run on a business model. On the top end, you need to process applications. Underwriting must follow a rigorous, albeit fast, decision process on creditworthiness to set appropriate interest rates.
These decisions must consider historical trends and future economic expectations. If experience is that a 700 FICO cutoff will yield writeoffs of say 3.5%, then if you drop that threshold to 680, expect the write-off rate to surge. Tighten it up and use a 740 score, you will see lower losses and less volume.
It isn’t rocket science; it is good old fashioned lending, automated to make the volumes flow.
The article builds a case for alternative bureau data when the customer has a limited credit bureau profile:
- Arguably, the dominant source of alternative data is payment history records from utilities and telecommunications service providers.
- Consider the insights into credit risk that creditors and lenders can gain through access to a consumer’s name and address, payment due date, the actual date of the payment, the amount due and the amount paid.
- This level of data can add to a profile and create a more comprehensive picture of an individual and their ability to manage timely payments.
It is easy to suggest tampering with current industry standards that require credit files. Right now, using the conservative methods in play today, issuers can tweak underwriting to reduce risk or tolerate more risk on a controlled basis.
Start using mobile phone bills and electric bills as a surrogate for proven methods is risky. It may seem harsh, but current lending standards do require some credit history. Adding in thin file accounts by looking at how consumers pay their cell phone bills, you’d better have a rigorous collection process in force.
And, if you use this strategy at credit unions, which require low-risk customers and operate with a cap on interest rates at 18%, it might be worthwhile for you to prepare your credit union shareholders for the risk.
Overview by Brian Riley, Director, Credit Advisory Service at Mercator Advisory Group
