University of California, San Diego
Akos Rona-Tas, Project Director
Alya Guseva, Project Co-Director
Tamara Perkins, Project Co-Director
Official Project Title:
Comparative Research Project on Credit Card Markets in Transition Countries
Akos Rona-Tas, University of California, San Diego
The Research Focus:
There are two contrasting methods of deciding on creditworthiness.
-
Lenders can gather diagnostic information about each applicant to render case-specific decisions in the form of individual judgment
-
or they can collect standardized, survey information about a large set of applicants and use statistical models that hand down decisions specific to a category of cases.
While reasoned and often reasonable, the first method is irreducibly subjective, and the ultimate question becomes whether or not on the basis of incomplete evidence the decision maker trusts the applicant.
In the second process, the decision is based purely on mechanized rational calculation. The reasoning in the first relies on causal arguments, in the second it turns on correlation.
These are, of course, two conceptual poles, and most practices are somewhere in between mixing elements of the two to varying degrees.
Credit Cards
Credit card markets became possible with a radical shift from judgment to calculation, which in turn was facilitated by improved information sharing among banks and the fast development of new statistical techniques that could take advantage of standardized information.
Credit scoring made consumer lending in general, and credit card in particular, economically viable by cutting the marginal cost of credit decisions.
Standardization and quantification also made the decision process more transparent and easier to control, and, in the US, allowed for the securitization of credit card receivables.
Our earlier research in post-socialist countries showed, that issuers are actually employing a variety of methods to judge card applicants. A few rely primarily on scoring models, while others use judgment or a mixture of the two.
Judgment is often based on information collected through face-to-face interviews, personal recommendations, social networks of bank employees or the banking history of the individual client with the issuing bank. Judgment draws on the personal experience of loan officers (which is in short supply in developing countries) and follows flexible and not-too-complex rules of thumb.
In many cases, issuers resist statistical calculation even when they are well aware of scoring methods and its potential advantages. They counter that effective scoring systems require valid data that are properly coded and classified (i.e. standardized), and observations that are numerous enough to allow reliable estimation.
One key hindrance is the weakness of information sharing among banks in the region. Moreover, good scoring also requires social and economic stability that allow extrapolation from past to future behavior in a mechanistic fashion. They point to the changes and crises these economies experienced in the past decade.
Without an effective scoring system, the size of the credit card market remains limited. As they try to expand, to lessen their vulnerability issuers compensate for the absence of reliable calculation by increasing control over clients and by reducing the credit granting function of cards. Hence most cards in those countries are debit or ATM cards.
In all of these countries there is a large, unrealized potential for consumer credit. Poland, Hungary and the Czech Republic are joining the European Union in 2004, Russia and Bulgaria have been on a growth trajectory for the last 4 and 5 years, respectively, and China and Vietnam have sustained one of the most impressive rates of economic expansion in the last decade.
National Science Foundation Grant Proposal
Log In for Members