Card issuers are caught between unprecedented levels of default and delinquency and a changing regulatory landscape. Facing reduced levels of spending even from good customers, they are understandably anxious. In the absence of well-tested best practices, many are choosing to retreat into traditional defensive postures — adopting highly conservative strategies for managing risk, such as setting low initial lines for customers with lower FICO scores and aggressively repricing when there is a suspicion of possible default.
While these approaches may seem sensible, and while they have been hallowed by tradition, it's not clear whether they actually serve the bank's long-term interest or long-term profit. It could well be the case that the strategy that seems the safest is costing the bank money in lost revenue. A credit line of $2,000 for someone with a low FICO score may indeed limit the bank's exposure, but it may not generate much revenue from a customer who probably has little access to other forms of credit.
Given the opportunity, such a customer would likely use the card more than someone who had a less risky score (as defined by the bank's internal risk thresholds) and a variety of payment forms available. In other words, when you change the economics of the cardholder relationship, other variables related to card profitability — including APR and credit limits — should be taken into consideration in addition to the risk score.
Similarly, while a higher price might on paper make up for the higher costs of servicing an account if it goes into default, it is not clear that a higher APR will make it more likely that the account stays current. Indeed, since a higher APR will either mean a higher monthly payment, with a smaller portion of the balance getting paid, an increased rate may well push a performing loan into default.
Issuers are not in business to avoid risk. They are in business to make money while managing risk, and money comes from sustainable customer usage. The overriding goal therefore is to stimulate usage within the customer's capacity to handle the debt. Here are some things issuers should factor into their decision.
- Usage will depend on price, credit line and user's perceived value of the card. Because early usage is crucial in establishing a place at the top of the wallet, getting the initial terms right from the beginning is of critical importance. This will be even more the case as new regulatory guidelines further restrict banks' repricing.
- Capacity, as measured by cash flow, current debt load, and usage patterns, is a much better indicator of a customer's ability to handle debt than a standard origination risk score, and as a driver for the line assignment. To get the best possible determination of capacity, the issuer should use as many sources of information about the customer as possible. If the customer has other relationships with the bank — a checking account, for example — that can be of enormous use.
In this exercise the issuer is seeking to discover precisely how different groups of customers behave under various circumstances. That, of course, is the Holy Grail not just of managing risk, but of managing usage in general: what conditions will lead to long-range profitable behavior?The issuer, therefore, needs to become a researcher as well, conducting controlled experiments on representative segments of the customer pool to identify what is driving performance. Using a rich picture of the customers' capacity, it is possible to segment the portfolio into well-defined groups of cardholders. Building on what the issuer may already know, hypotheses can be developed about what will stimulate usage, and then different pricing and line strategies can be implemented in test runs on representative samples of particular segments.
Both out of prudence and to gain as much information as possible, the research protocol must include frequent account analysis. Beginning with origination, the account should be looked monthly over its first six months on the books. Segments that display positive and sustainable performance may be moved to less-frequent analysis. Still, in a rapidly changing economy with rapidly changing customer behavior patterns, a more frequent refreshing of the behavior score is a prudent approach to capture the severity and frequency of changes in performance. This not only allows for early intervention if there is a danger of early-stage delinquency, but more broadly it provides almost real-time results about the management strategy being used. After an analysis of the preliminary results, the hypotheses can be refined and the strategies realigned and retested.
Ultimately, these experiments may yield insights into how best to assign lines that will hit the sweet spot of optimizing usage while controlling exposure to loss, and may help define the profile of possible future defaulters.