ALI Solutions

BAI Banking Strategies
August 3, 2009
By Tom Miller

Credit card lenders can use balance-control strategies to both avoid the higher-risk customers and retain valued, long-term clients.

They say that generals always fight the last war, employing tactics from earlier battles that new circumstances have rendered obsolete. Today, credit card balance-control strategies that were good enough in the pre-credit crisis world are simply not up to current demands. Among the new circumstances are a larger population of high-risk, low-value customers and a consequent climate of risk avoidance that harms or neglects low-risk, high-value customers.

In such an environment, card lenders might well wish for the gift of Janus, the god with two faces who could see in opposing directions at once, in order to keep watch over both faces of risk: the risk of not being repaid and the risk of losing good customers.

The first face of risk, the risk of not being repaid, always gets attention. Even in stable times, there are delinquent customers, including those unable to pay, others who are unwilling to pay under certain circumstances, the careless, negligent payers and the outright fraudulent customers. Exacerbating this risk today is the escalating war for every repayment dollar. Customers in financial difficulty tend to owe multiple lenders. Despite fierce competition among lenders for repayment preference, not all of them will be repaid.

Yet the other face of risk – the risk of failing to extend credit to worthy customers – also merits attention. Some people in financial trouble for the first time never expected to be there and are likely to recover their footing in the future. Such customers should be retained if possible because they represent future profits. If the financial institution is not careful, it can unwittingly diminish the loyalty of such customers with actions that tighten terms, create insecurity or cause offense.

Preemptive Risk Management Principles

Fortunately, both faces of risk can be successfully addressed with balance-control strategies. But lenders need to be able to distinguish between the two with high confidence and not settle for cursory indicators of risk and rely on old and potentially outdated risk scores. They need to be able to apply three principles of preemptive risk management equally well to both faces of risk:

Use predictive analytics preemptively. Understand the risk of each customer and in each account before the problem becomes too costly to resolve. Reach out to the customer early, when the relationship is still intact, the lender still has a variety of options to offer the customer and the customer presumably has more options and more assets.

Glean information from every single customer interaction. Each customer interaction yields information that strengthens, confirms, confounds or changes the lender’s earlier conclusion about the relative risk of that customer. Capture that information and let it drive profitable balance-control decisions that can be made live, during the customer interaction.

Consider each customer holistically. Customers either repay or they do not according to their entire financial situation, not the circumstances of one particular account. They need to be judged by their personal balance sheet or income statement, not the credit score applied at the outset of the relationship. Lenders need to ascertain those circumstances by assembling more data, accessing new data sources, running more analytics and doing it all more quickly in the interest of faster and wiser decisions.

Operating under these principles, lenders can apply a variety of balance-control strategies but not yesterday’s broad brush-versions such as cycle-based or month-end behavioral score-driven management. Today they must include an almost endless variety of acutely calibrated strategies that minimize losses on high-risk customers and maximize potential gain on high-potential customers.

Balance-Control Strategies

New account treatment has proven to be an effective balance-control strategy that can address both faces of risk. True, most lenders are not aggressively acquiring new accounts today. But new account activity does continue through multiple channels, including the Internet. That heightens the need, especially in a recessionary environment, for lenders to quickly be able to distinguish between good and bad risks in their new credit accounts.

One credit card issuer we worked with was determined to reduce charge-offs but equally determined not to jeopardize good customers. Their concern was highlighted by our research showing that bad accounts tend to use all of their credit and do so rapidly, averaging 70% or more of the available credit in the first 30 days. Balance-control strategies deployed months later, when most of the bad spend has already occurred, are of little use. But identifying risk starting on day one gives the issuer many options for constraining the bad spend.

As for the other face of risk, good customers typically spend moderately, making limited usage of available credit unless they are targeted with promotions. Identifying them as low risk early poses a potent opportunity to extend selective balance increase and incent more low-risk spending.

This particular issuer began monitoring transactions on the very first day each card was activated, immediately determining the appropriate action rather than waiting the typical three to four months for behavior scores to become effective. Early results demonstrated that new customers or those who are new to credit may simply need proactive early communication to keep them on track. A soft reminder call drove down average balances rolling into delinquency by 6%.

For another set of customers at a higher risk of charge-off, more aggressive actions were taken. In those segments, quickly revoking credit drove late stage delinquent balances down by 12% while charge-offs decreased by 14%. The issuer’s results continue to improve as they continue to refine the program.

Semi-active and reactivated cardholders represent another lucrative balance-control opportunity. In recent months, lenders have reached out widely to lower limits or close inactive revolving and card accounts. To the extent those accounts represented little or no income to the lenders, along with a lot of liability, such actions made good sense; lenders rid their books of billions of dollars in latent liability.

Again, the concern about the first face of risk was valid. In a tough economy, when a customer suddenly uses a long-inactive card, there is a reasonable concern that the customer may be in financial trouble. But what about the large segments of semi-active accounts and second-in-wallet cards? Does the customer’s recent activity indicate someone likely to use the entire credit line quickly without repayment? Or is this person simply using the card for the very purpose they acquired it – for emergency cash, which is increasingly needed in today’s economy? Or has this person chosen, for any number of innocuous reasons, to change brand preference and become a more active customer?

Finding the right answer matters; substantial profit hangs in the balance. Lenders who are not equipped to quickly arrive at the right answer feel compelled to worry more about the first face of risk and curtail activity on the account. But those who can find the answer are able to consider other risk-mitigation measures that don’t penalize high-risk customers excessively, such as providing lower prices for “good” behavior. While recent regulations restrict their ability to raise rates, lenders might offer a lower interest rate for a reactivating customer upon payment of the next bill or for a delinquent customer upon payment of a past-due bill.

Innovative lenders are now using analytics, refreshed with updated data from customer interactions, to learn more about the relative risk involved in these accounts and more about the customer as a whole. Only then can they make the right balance-control decisions.

Payment float management can also be employed for both faces of risk – to limit balances on high-risk customers and capture spend for high-revenue potential customers. For example, each time a customer pays a credit card bill with a check, the lender can use predictive analytics to refine their original risk assessment of the customer before releasing open-to-buy on the credit line and allowing the cardholder to use the card again. If the analytics reveal the customer to be high-risk, the lender may wait for the check to clear before allowing the customer to use the card again. But if the new assessment shows the customer to be lower risk, the lender may immediately release all or a portion of open-to-buy, creating goodwill and capturing an extra two or three days of spending.

Lending is a business of taking calculated risks for profit. Delinquency is a fact of life in lending that is sometimes beyond the control of risk managers. On such occasions, lenders have no levers to pull and borrowers no means to pay. But lenders can make deft use of balance-control strategies, informed by the principles of preemptive risk management, to manage down bad balances, increase good spend and balances and offset the losses with gains among good customers. By doing so, they can choose which customers to retain and cultivate and which customers to limit, wind down or cancel.

Lenders that succeed on this front will not only improve their ability to compete better for tighter payment dollars but also increase their ability to retain desirable customers on which to build their future profits.

Mr. Miller is senior vice president of ALI Solutions, a predictive analytics company based in Austin, TX. He can be reached at This e-mail address is being protected from spambots. You need JavaScript enabled to view it .