In response to Professor Littwin’s article, Ronald Mann compliments the article’s focus on the consumers of credit cards, rather than the lenders.  Mann cites two important contributions from the article.  The first is the surprising importance of credit cards in maintaining social status in the milieu that Littwin studies.  The second contribution is crystallizing the need for credit-card products that offer real precommitment.  Mann also praises her research on the use of credit cards at the family level, as opposed to the national or company level.  Finally, Mann defends Littwin’s decision to focus on a particular demographic group because the differentiation of credit-card products means that choices available to low-income consumers will be very different from middle- and upper-income consumers.  But Mann wants to broaden the research.  Mann asserts that a targeted project involving respondent-directed sampling should enable reliable analysis of the stability of Littwin’s conclusions over a larger area.

Pick a Card, Any Card

Ronald J. Mann*

At the heart of all serious thought about consumer financial products is the difficulty of understanding the mental processes by which consumers evaluate, compare, and use those products.  Usury proposals from scholars and policy makers depend on explicit or implicit assumptions about how interest-rate caps will affect the mix of products available in the marketplace and the choices that consumers make among them.  Legislators and lobbyists that decry a torrent of consumer bankruptcy filings rely explicitly on the claim that consumers abuse credit products.  Proposals to outlaw products like payday loans assume that those who use the products are so cognitively impaired that they would be better off with fewer options.

The weakness of existing work on the subject underscores the difficulty of the problem.  The leading theoretical models of consumer finance empha­size the importance of harsh remedies to discourage borrowers from casual defaults.  Yet the high repayment rates even among judgment-proof consum­ers demonstrate that those models miss something important about the decision of consumers to use credit products.  With respect to payment products, my own work emphasizes a tendency of consumers to spend more when they use cards than when they use cash or checks, but I cannot readily explain the psychological relation between cards and spending.

The most useful work has explored cognitive limitations that drive consumer behavior.  Bob Cooter and Ed Rubin wrote twenty years ago to explain why rational consumers would not take the trouble to understand the agreements that define their relations with the banks where they maintain deposit accounts.[1]  The intervening years have brought an increasing variety of products to the marketplace.  Consumers now choose not only among bank accounts, credit cards, pawnshops, and rent-to-own providers, but they face new products like payday loans, overdraft loans, refund-anticipation loans, and even car-title lenders.  Moreover, even vintage products like the credit card have become much more complex; only the most attentive con­sumer can keep track of the relevant charges and fees on a modern credit card.

The burgeoning variation of product design has brought forth a rich scholarly literature that explores the business models that drive those products.  Those models depend less and less on direct competition based on price and quality and more and more on the ability to balance multiple attributes.  Thus competition proceeds on one set of up-front attributes, while profits are driven by back-end attributes on which competition is marginal.  DellaVigna and Malmendier, for example, show how different fee structures make it harder (or easier) for consumers to manage their total expenditures on products like health clubs, cell phones, and credit cards.[2] Similarly, Gabaix and Laibson show how businesses can “shroud” the unattractive at­tributes of their products so that they compete only on the more favorable attributes.[3]

Against that backdrop, Angie Littwin presents her study of the role of credit cards in the lives of women in Boston-area low-income neighborhoods.  The frame of the project is elegant; she situates her study against the well-known and seemingly irreconcilable usury debate.  Most participants in that debate focus on the lending entities—whether and how those entities should be permitted to operate.  For the most part, the debate pits a laissez-faire perspective that profitable market products are their own justification against a condemnation of the immorality of grasping usurers.  The Dickensian tone so common to both sides suggests that the point of the discussion is to force a choice between Ned Cheeryble and Ralph Nickleby as the archetype of the modern-day consumer lender.

Littwin’s project, by contrast, puts the customers at the center of the story instead of the lenders.  Because she takes the individuality of the cus­tomers seriously, the insights of her work derive almost entirely from her efforts to understand the perspective of those at whom the products are directed.  On that point, her consciousness of her role as an interviewer is noteworthy.  Littwin shows great sensitivity to the possibility that her inter­action with the subjects will move their views to match her preconceptions about what the subjects are thinking.

Her success at maintaining an objective distance from her subjects leads directly to the sophisticated behavioral story Littwin develops.  She notes that the subjects would be judged poorly by conventional standards of finan­cial literacy—their understanding of the time value of money, compound interest, and the like.  At the same time, Littwin’s subjects are astonishingly self-aware of their own behavioral limitations.  Thus, the products that the subjects seek are, for the most part, products designed to respond specifically to the behavioral limitations that afflict use of credit cards.

From my perspective, the article offers two important contributions to the literature.  The first is the surprising importance of credit cards in main­taining social status in the milieu that Littwin studies.  If we regard credit cards as a primarily middle-class phenomenon, we would not expect them to be of great social significance in the low-income households involved in this project.  But Littwin’s discussion shows that cards are important both as a marker of responsibility and as a symbol of inclusion in mainstream society.  It is precisely because credit cards are central to the middle-class ethos that they are socially important to the households Littwin examines.  If we grant the broader importance of ensuring that the poor remain within reach of the mainstream of the middle-class economy, Littwin’s work suggests that credit cards fulfill a crucial function here.  The role of credit cards in establishing social status complicates debates about the propriety of paternalistic inter­ventions to protect consumers from these products.

The second contribution relates to the market demand for precommitment devices.  Even before Littwin wrote, it had become increasingly clear in the literature that one of the dominant strategies of credit-card lenders is to design products that make it difficult for consumers to manage the purchasing and repayment decisions that they make with re­spect to their credit cards.  What Littwin has done, however, crystallizes the need for products that offer real precommitment.  In that vein, the most at­tractive product she discusses is the hard credit limit.  Under this arrangement, an issuer that receives a request to authorize a transaction that exceeds the existing credit limit would not consider whether the customer was sufficiently creditworthy to justify extending the credit, nor would it consider the possibility that an overlimit fee was adequate to cover the risk.  Rather, the issuer would refuse to authorize the charge, holding the customer’s maximum balance at the prearranged credit limit.  The simplicity and attractiveness of that product make her proposal a powerful one.

It is also worth highlighting the way that Littwin’s research method matches the question under examination.  The academic literature contains many theoretical models and empirical examinations of credit-card markets.  Almost all of that work (including my own) relies on data sets collected for other purposes, usually at a national or company level.  There is almost no information about the use of cards at the family level except in articles ana­lyzing proprietary data sets from card issuers—a technique that has its own problems.

The most important national longitudinal surveys do collect information about financial activity at the household level, but there is far too little detail to understand how consumers choose among and use the financial products available to them.  To be sure, there is a growing recognition that face-to-face interaction with consumers is necessary to understand these markets.  Thus, the Federal Reserve recently proposed amendments to the Truth in Lending Act disclosures, which relied heavily on focus groups through which researchers could examine the reactions of consumers to different disclosures.  Similarly, Michael Barr’s recent work on Detroit low- and middle-income households uses detailed interviews to collect information about the relative attractiveness of different products.[4]  Littwin’s work goes much further, however.  Because she uses largely open-ended interviews, she is able to capture an understanding of what matters to consumers that a typi­cal survey instrument would not readily provide.  Indeed, what makes the article so readable is her skill at reproducing the voices of her interview subjects.

It is, to be sure, difficult to generalize from the experiences in this particular context.  Low-income consumers in other states choose from a different array of products (there are no lawful payday lenders in Massachusetts).  Men might approach the product choices in a different way.  Middle-income and upper-income consumers also have different experiences.  But that is more a difficulty of the subject she is examining than a flaw in her research design.

The most important change in consumer-finance products in the last fifteen years has been the increasingly fine-grained segmentation of products.  Sophisticated providers like the leading credit-card lenders develop arrays of product offerings that are finely differentiated to reflect different expecta­tions about what products will appeal to different consumers, how they will use them, and how quickly they will repay funds that they borrow.  Issuers in the “superprime” market may specialize in rewards cards that bring profits from annual fees, interchange revenues based on high-spending volumes, and occasional payments of interest on the large balances their customers carry from time to time.  Issuers in the “prime” market may specialize in affinity cards, which are likely to generate revenues from annual fees and more regular payments of interest on smaller balances.  Issuers in the “subprime” market populated by Littwin’s subjects are likely to offer cards with very low credit limits that depend heavily on late and overlimit fees coupled with rela­tively permanent interest obligations on balances that the customers often cannot repay.

The differentiation of products supports Littwin’s decision to focus her study on a particular demographic segment.  Because the product choices that confront low-income consumers are different from those that confront middle- and upper-income consumers, a study that aggregated consumers of widely varying socioeconomic levels would be far less informative than a study like Littwin’s that focuses on a group that faces a relatively homoge­nous set of products.  Still, the findings about credit-card use among her subjects were sufficiently provocative to make it important to document whether the experience of her subjects is representative of consumers in this income bracket or even of female heads of household in this income bracket.  As Littwin recognizes, her subjects are unusual in several ways that might make them more likely to hold cards than others at the same income level.  For example, those who live in public housing tend to have more stable ad­dresses than others of similar financial standing.  Also, because she collected a snowball sample starting from acquaintances of long standing, there is good reason to think that her subjects are longer term residents of the build­ing than is typical.  To me, this suggests that the next step in pursuing this agenda would be to employ some of the developing techniques for collecting representative samples of discrete but hard-to-reach populations—samples that could then be used for quantitative analysis.  Here, as I suggested above, the population of interest would be female heads of low-income households.  A targeted project involving respondent-directed sampling should enable re­liable analysis of the stability of her conclusions over a larger area.

*    Professor of Law, Columbia Law School. [return]

[1] See Robert D. Cooter & Edward L. Rubin, A Theory of Loss Allocation for Consumer Payments, 66 Texas L. Rev. 63 (1987). [return]

[2] See Stefano DellaVigna & Ulrike Malmendier, Contract Design and Self-control Theory and Evidence, 119 Q.J. Econ. 353 (2004). [return]

[3] See Xavier Gabaix & David Laibson, Shrouded Attributes, Consumer Myopia, and Information Suppression in Competitive Markets, 121 Q.J. Econ. 505 (2006). [return]

[4] See Michael Barr, Detroit Area Study on Financial Services: What, Why & How, L. Quadrangle Notes, Summer 2005, at 72. [return]

Comments (0) >>

Show/Hide comments

Write comment

Show/Hide comment form

Name

Website

Title

Comment

Add Comment