|
||||||||||
Common wisdom has it that credit card issuers base their decisions about both new credit card applications and term changes to existing accounts mainly on credit scores. However, according to a recently released study by the Federal Reserve, FICO scores aren’t the only thing major card issuers look at.
According to the study, several large credit card issuer use what is known as behavior modeling when making decisions about existing cardholder accounts. Factors such as where cardholders shop, what items they purchase, where they live, how frequently they take out cash advances, and which mortgage lender they use may all come into play when evaluating whether to cut a cardholder’s credit limit, increase the interest rate, or even close a credit card account.
The Fed study was prompted by one of the provisions in the Credit CARD Act of 2009, which charged the Fed with looking into whether credit card issuers used cardholder profiling to make decisions about credit card terms in the three years between November 2006 and November 2009.
The Fed report was based on survey responses from 175 credit card issuers, as well as interviews with some of the largest card issuers. Among the 25 largest credit card issuers, six reported making decisions about card changes based on the types of credit card transactions the cardholder made. Five of these were among the largest card issuers, meaning that their practices affect a proportionately larger amount of cardholders; the large credit card issuers make up about 80 percent of U.S. credit card accounts.
Here are some of the features that may cause some card issuers to flag an account:
- Frequent credit card charges for everyday items, such as groceries, low-end retailers, or discount stores;
- Frequent gambling-related charges;
- Frequent cash advances, particularly for new cardholders and particularly if in conjunction with purchases of items that can easily be turned into cash, such as jewelry and electronic goods;
- Increased frequency of charges ( an indication that the cardholder may be an elevated credit risk);
- Having a mortgage with a high-risk mortgage lender;
- Frequent credit card charges with merchants within certain geographic locations (the report does not detail the characteristics of the neighborhoods).
The Fed report gives few details about how many cardholders are likely to be affected by transaction-based profiling. The one example quoted in the Fed report involves two large credit card issuers, which together had more than 53 million active accounts. In November of 2009, the card issuers reported cutting credit limits for 340,000 cardholders. Out of these, 1,900 card limits were reduced at least in part because of concerns about the cardholders’ transaction activity. If that number is representative of monthly averages, an estimated 23,000 cardholders with those two large credit card companies would be affected by transaction-based profiling.
The Fed report did not address concerns raised by Members of Congress, such as Rep. Maxine Waters of California, that some aspects of behavior modeling border on redlining, a now-banned practice used by insurance and mortgage companies in the 1960s and 1970s to target minority groups for discriminatory treatment. The Fed report stated that it was impossible to gain information about this feature, since lenders are not allowed to collect information about the race of cardholders. However, a 2008 Boston Fed study indicated that some card issuers use the zip codes on credit reports to target what they considered high-risk neighborhoods, and that card applicants living in white neighborhood had a greater chance of being approved for a credit card than someone living in a black neighborhood.
Many card issuers reported engaging less in transaction-based modeling today, because of public criticisms of the practice. In addition, card issuers emphasized that to the extent transaction-based profiling was used to evaluate the credit worthiness of cardholders, it was just one factor among many others. Frequently making late payments, for example, was quoted as a far more important factor than the above measures. In addition, changes in broad economic conditions remains one of the most significant factors leading to changes in credit card account terms or account closures.
