(What Determines) What's in Your Wallet?

How do credit card companies make their lending decisions? Or more precisely, how do they decide who to target with which “pre-approved” offer or other solicitation? This is one of the major “known unknowns”. I'd like to give some intentionally provocative musings and hopefully start a discussion in the comments.

On a basic level we know the answer—lots of econometric analysis of consumer data, probably starting with Fair Isaac FICO scores or some equivalent. But credit scores are surely only the start. The card industry sent out some 6 billion solicitations in 2005. Solicitations are expensive—I’d guess that it costs at least 50 cents, if not over a dollar to send each individual solicitation when postage, paper, printing, data sorting and handling are added up. In spite of the myriad solicitations sent to minors and pets, the card industry is assuredly more sophisticated then sending out costly mass mailing barrages based solely on credit scores. Indeed, the solicitations to minors and pets indicate that credit scores are often not necessary for card solicitations. So what determines whether or not I get an offer for a premium Capital One Visa Signature Card or for a subprime First Premier Bank MasterCard offer?

My sense is that card issuers (and maybe the networks) undertake incredibly sophisticated analyses of consumer data probing for particular profit points (or put another way, for particular consumer vulnerabilities). The card industry is capable of gather more consumer data than virtually anyone else because it is involved in so many consumer transactions. It makes good business sense to gather up as much consumer data as possible and then analyze it for ways in which to maximize profit.

Perhaps the fact that someone purchased diapers and an infant card seat or that they regularly buy Goya products or kitty litter or red wine over the Internet says something about that person as a credit risk and causes that person to get certain solicitations, but not others. On the surface there's nothing wrong with such a system. It appears to cater to consumer preferences. If I like sailing, maybe I'd like a card that gives me rewards points toward a new yacht. But there's a potential flip-side: if I purchase items with certain class or other group-status implication, say pick-up trucks, perhaps I'll be targeted with more expensive card solicitations. Or perhaps my application for certain cards will be declined. Econometrics can easily bleed over into discriminatory practices like red-lining, as the history of the insurance and housing-lending have shown.

Unfortunately, there’s really no way of knowing what factors determine which solicitations I receive and which cards I am able to get or what credit limits. Does my race or my gender or family status or religion affect which credit cards are marketed to me and which I am able to obtain? I hope not, but I am confident that the fact that I make purchases at toy stores affects the solicitations I receive, just as my professional affiliations (e.g. ABA) do. Maybe my ZIP code, a potential proxy for race, is a factor. We simply don't know the factors going into credit card marketing and lending decisions, but it's a question worth asking: what determines what’s in your wallet?

Comments

[[email protected]]

Being a married female who uses her credit card sparingly, I can attest to the fact that I do not receive offers to increase the allowable balance on my card as does my husband. Yes, one's postal code does in fact send racial and socio-economic bias signals and don't be naive about it, as well as the type of merchandise one purchases and from where the merchandise is purchased. We've all come a long way from the blatant discrimination previously practiced, but let's face it, we still have a long way to go. Most of those credit card offers are really good for nothing other than shredded char-base for the barbecue, anyway!

[Adam Levitin]

My post aimed to provoke, but I want to clarify that I’m not alleging Equal Credit Opportunity Act violations. Instead, my point was to emphasize the lack of transparency in credit card lending and how we should be uncomfortable with that given the card industry’s track record with consumers elsewhere. ECOA violations are really hard to prove because individual consumers never see the big picture.

I should also clarify that although card issuers do not get detailed transaction breakdowns as part of processing transactions, they can easily obtain all the consumer data they want. All sorts of consumer data is available for purchase both through brokerages and directly from merchants. Sometimes card issuers' co-branding relationships involve an interchange rebate in exchange for consumer data. Whether this information is merely used in the solicitation process and not the underwriting process is not clear, but there are certainly cards that target minorities in their solicitations. For example, the Third Circuit heard a reverse confusion trademark case over the term “Freedom Card. See Freedom Card, Inc. v. JP Morgan Chase & Co., 432 F.3d 463 (3d Cir. 2005). One product was marketed to affluent Wall Street Journal readers. The other product used Queen Latifah as its spokesperson to market to subprime African-Americans. The direct solicitation model of the credit card industry makes it different from other financial services that wait for consumers to contact them. Even if race is not a factor in granting credit, the same effect is accomplished when a credit product is marketed solely to a particular racial demographic.