"Mail Order Companies Who Offer Preapproved Credit Lines"
March 1, 1988 eCONOMIC COMMeNTaRY Federal Reserve Bank of Cleveland ---- The Bank Credit-Card Boom: Some Explanations and Consequences - by Paul R Watro I n financial activities, there is a trade- In view of the expense, the risk, and off between risk and return. The the higher rate of charge-offs, are What is behind the surge in credit- higher the risk, the greater the banks acting irrationally when they try card lending and the sharp rise in required return. to increase their credit-card lending credit-card charge-off rates at banks? business? The answer depends on the Factors related to risk-taking help This fundamental principle helps trade-off between risk and return. explain why some banks have higher explain why credit-card interest rates charge-off rates than others. are still in the 17-18percent range, In this article, we discuss factors even though rates on most other behind the surge in credit-card lend- kinds of bank loans have fallen signif- ing, and identify factors related to icantly since the early 1980s. risk-taking that help explain why some banks have higher charge-off to 20 percent in 1982. This growth Credit-card loans have higher interest rates than others. has occurred even though lenders rates because they generally lack col- charge higher rates on credit-card lateral and involve more risk. If such • Consumer Demand loans than on other consumer loans, loans are not repaid, banks often must Consumer spending has propelled such as auto loans. One reason for charge them off and suffer a loss. economic growth, particularly in the the higher rates is that credit-card early years of the current economic In the last three years, the rate of such credit has no collateraJ.2 If the card- expansion, which began in November charge-offs has more than doubled, holder defaults, the credit-card issuer 1982.1 Rising income levels and indicating a decline in credit quality. is without recourse against the mer- improved wealth positions have con- In spite of this, however, strong con- chandise purchased with the card. tributed to increased consumer sumer demand and attractive profit spending. Consumers have also been The growing popularity of credit margins have led to greater credit- borrowing more. For instance, con- cards may be attributed to many fac- card lending by banks. sumer installment debt as a percen- tors. From a user standpoint, accep- This increased lending comes at a time tage of disposable personal income tance, convenience, safety, and flexi- when fewer potential customers are rose from 14 percent in 1982 to over bility have encouraged consumers to without credit cards and when banks 18 percent in the last year or two. make greater use of charge cards. are using market segmentation to com- Credit-card transactions provide users Credit cards have been the fastest- bat the more intense competition, with a convenient way to maintain growing form of consumer credit. In which has made growth in credit· card records for tax and other purposes the last five years, credit-card balances operations more expensive. and a way to minimize the risk and have more than doubled and now financial cost of carrying large cash account for nearly 25 percent of all consumer installment debt, compared ISSN 0428·1276 balances. Credit cards might even be superior to checks or cash for some transactions, such as those in foreign countries and those over the tele- phone and through the mail. credit-card interest rates enable banks to buffer the expected higher credit losses associated with lending to risk- ier customers. Credit-card accounts may be more Credit-card performance was also quite poor during the developing stages of bank credit-card systems." In addition to operating problems, banks under- estimated the burden of controlling - CHART 1 NET PRETAX PROFIT MARGINS ON VARIOUS lYPES OF BANK CREDIT Percent credit type outstanding • Credit Quality One widely used measure of loan quality is net charge-offs.w These are loans judged to be uncollectable in a given year, minus any recoveries of • Banks' Role valuable than their direct dollar credit losses. In an effort to grow and to gain market acceptance, banks 6~----------------------------------~ loans previously charged off. Banks generally write off unsecured loans, Spurred by growing consumer return if they provide banks with use- turned to mass mailing of unsolicited such as credit-card receivables, faster demand, high returns, and declining ful marketing and credit information. credit cards during the late 1960s.This than well-secured loans like home commercial lending profits, many For instance, banks could judge the marketing strategy led to large-scale mortgage loans. banks have placed greater emphasis credirworthiness of cardholders for credit and fraud losses, that, in turn, on consumer lending, especially on larger loans based on their payment Chart 2 shows that bank credit-card led to legislation prohibiting the unsol- credit-card lending. As a percentage history. Some banks, such as Citibank, quality has deteriorated. Net charge- icited distribution of credit cards.' 3 of total bank loans, credit -card receiv- have apparently used credit-card cus- offs as a percentage of credit-card ables jumped from 3 percent in 1982 tomers as target groups for selling Despite a shaky track record, the balances have more than doubled, to over 5 percent by year-end 1986. insurance products and for penetrat- recent prosperity in credit-card earn- 2 jumping from 1.5 percent in 1984 to Banks now hold close to two-thirds of ing out-of-state markets. Moreover, ings has spurred new entrants and 3.2 percent in 1986.The deterioration the credit-card outstanding balances, banks typically include advertise- more intense competition in the 1 reflects many factors, some and per- Commercial up from just over one-half in 1982. ments in monthly bill statements in credit-card business. A few years ago haps the most important of which have and other an effort to sell other banking servi- Sears introduced the Discover Card, no relation to the business cycle. Technological advances, economies of ces to credit-card customers. which has not become profitable yet." scale, deregulation, and favorable mar- In our study, we examine 148 large American Express also introduced the ket conditions have encouraged lend- ers to mass-market credit on a nation- Perhaps earnings have been the underlying force behind the rapid Optima card, which offers a revolving -1 \I banking organizations to identify the level and variability of credit-card credit line with a lower interest rate wide basis. Banks issuing credit cards expansion in credit-card operations at than most bank cards." -2~--~----~--~----~--~----~--~--~ 1971 charge-off rates among individual face sizable volume requirements in banks. Chart 1 shows that credit-card 1973 1975 1977 1979 1981 1983 1985 banks. We look at banking organiza- order to achieve profitability because profit margins have improved sharply With greater competition and fewer NOTE: Based on annual data from the Federal Reserve System's Functional Cost tions as a unit rather than as individ- of high operating costs. This is why and have been better than those from consumers without credit cards, how- Analysis. ual banks because some organiza- many banks that offer credit cards other types of lending in recent years.' ever, large-scale expansion is probably tions have shifted credit-card balances participate with larger banks or organ- From 1984 through 1986, the annual becoming more expensive and more among subsidiary banks. In fact, more izations that actually issue cards and pretax net returns on bank credit-card risky. Nevertheless, mass solicitations than 20 bank holding companies determine their rates, fees, and serv- balances averaged 3.6 percent. In the with preapproved credit lines, waived operate special banks dealing primar- ice features.' Because of economies same period, banks earned 2.4 per- annual fees, and other enticements CHART 2 NET CHARGE-OFFS ON BANK CREDIT CARDS ily in credit-card accounts." of scale in credit-card operations, cent on real estate mortgages, 2.7 continue to be common among large (AS A PERCENT OF CREDIT-CARD AVERAGE BALANCES) Our sample included all bank hold- banks have an incentive to expand percent on consumer installment issuers. ing companies established before and become the low-cost issuers.' debt and 1.4 percent on commercial In an effort to build consumer loyalty, 4 1982 that have a subsidiary bank with and other loans. Bank credit-card When the cost of money fell signifi- many banks have sought to tie credit assets over $1 billion and that have returns have benefited not only from 3.2 cantly in the mid-1980s, banks gener- cards to affinity groups such as air- total credit-card receivables of more ally sought to expand credit-card lending through mass-mailing solici- the robust consumer demand, but also from the removal or relaxation of lines, hotel chains and alma maters 3 - 2.4 than $25 million. These banking during the past year or rwo. The organizations hold 80 percent of bank state usury laws in the early 1980s tations with preapproved credit. sponsoring organization typically credit-card receivables and accounted and from the large decline in funding 2~ Banks have largely relied on credit- endorses the bank's card for financial 1.5 for nearly 85 percent of credit-card costs from those years. card availability and services rather compensation based on members' charge-offs at banks in 1986. than on price in issuing cards. Over a longer period, however, acceptance and use of the card, While 1~ Credit-card charge-off rates varied credit-card profits look quite differ- aggressive marketing and liberal Credit-card issuers generally have considerably among our bank sam- ent. Credit-card issuers experienced a credit policies have promoted credit- increased credit limits, have provided severe profit squeeze in the 1979-81 card growth, some of the expansion o 1984 1985 1986 ple. Net charge-offs as a percentage of a wider range of enticements, and credit-card balances ranged from -0.6 period because of historically high came at the expense of loan quality. have offered credit to riskier groups to 8.9 percent and averaged 2.2 per- interest rates and binding usury laws. NOTE: Based on data from reports of income and condition for all banks. of consumers who were previously cent for 1986.12 Charge-off differen- Credit-card earnings were generally unable to obtain credit cards. The ces could be due to numerous factors more volatile and lower than earnings spreads berween funding costs and including differences in luck, eco- from other loans. Greater volatility nomic conditions, and risk-taking. may have reflected a combination of factors, including changes in the cost of funds, binding usury ceilings, and the higher degree of default risk in credit-card lending. - TABLE 1 BANK CHARGE-OFFDIFFERENCESON CREDIT CARDS High- Charge-off Group Low- Charge-off Group Difference We use two common tests to ascer- tain whether or not credit-card growth, specialization, volume, revenue, loan-to-asset ratios, and organizational size had any influence on charge-off rates. First, we compare Credit-card balances/ 7.9% 3.7% 4.2%a the sample extremes-those with the total loans highest and lowest charge-off rates. Change in credit-card balances/ 2.5 -1.0 3.5a The high group included those with total loans (1983 to 1985) charge-off rates greater than 4.0 per- cent and the low group included Loans/assets 63.9 60.8 3.1 those with charge-off rates less than Revenues from credit-cards/ 19.3 14.9 4.4a 1.0 percent. For each variable, average average credit-card balances values were computed for both groups and the difference was tested Credit-card balances (billions) $1.8 $0.1 $1.7b for statistical Significance. Second, to Assets (billions) 27.8 5.0 22.8b explain differences in charge-off rates for the whole sample, we use a statis- a. Denotes statistical significance at the 1 percent level. b. Denotes statistical significance at the 5 percent level. tical technique that isolates the effects NOTE: Data are as of year-end 1986, unless otherwise noted. of one variable while taking into SOURCE:Federal Financial Institutions Examination Council's Reports of Income and Condition for banks. account other factors. Our analysis, as expected, revealed that the 18 banking organizations in the high-charge-off group differed We examine credit-card growth, spe- There is a positive relationship be- Significantlyfrom 21 banking organi- cialization, volume, revenue, loan-to- tween risk and returns. Lenders charge zations in the low-charge-off group. asset ratio and organizational size as higher rates or require higher reve- The table shows that the high-charge- potential explanatoiy factors for nues for riskier loans. Accordingly, one off banks were much larger, placed interbank charge-off differences. Each would expect to find higher credit-card greater emphasis on credit-card lend- of these factors is related to risk- charge-offs at banks that generate ing, and charged higher prices. The taking in one way or another. higher revenues per dollar of credit- high-charge-off organizations expe- card balances. We also examined rienced much faster credit-card Banks that experience faster credit- loan-to-asset ratios as a measure of an growth and, on average, they held card growth might be expected to in- organization's overall attitude towards $1.8 billion in credit-card receivables, cur higher charge-offs because a trade- risk. Higher ratios are thought to which accounted for nearly 8 percent off may exist between credit growth reflect greater risk-taking since loans of their loans. Average credit-card and credit quality. Lenders that spe- are usually riskier than other assets, revenues generated by the high- cialize in or devote more resources to such as government securities. charge-off group were 4.4 percent a certain type of lending may also be higher than the low-charge-off group. more aggressive and extend riskier Product and geographical diversifica- lines of credit in those areas. Alterna- tion helps to reduce risk. The largest Although we do not know if higher tively, those banks may be better at credit-card issuers with a nationwide revenues were sufficient to offset managing risk. Specialization is mea- customer base should have more higher default costs, this finding sug- sured by the current share of loans geographically diversified portfolios gests that banks are taking credit risk held in credit-card receivables and that could lower charge-off rates. On into account at least to some degree growth is measured by the change in the other hand, the largest banking when pricing credit cards. Higher this ratio over the two previous years. organizations might choose to have lower credit standards for issuing credit cards because of potentially lower risk levels from greater product and loan diversification. - CHART 3 CREDIT-CARD REVENUE BY CHARGE-OFF GROUPS (AS A PERCENT OF CREDIT-CARD AVERAGE BAlANCES) Revenue Credit-card charge-offsvaried consider- ably among issuers. Banks with higher propensities to take risk incurred higher charge-off rates. As long as issuers receive adequate compensa- • Footnotes 1. See K.]. Kowalewski, "Is the Consumer Overextended?," Economic Commentary, Federal Reserve Bank of Cleveland, 6. Peter S. Rose, "The Promise and the Peril," The Canadian Banker & ICB Review, Volume 85, November 6, December 1978, pp. 64-65, and Joel P. tion for credit risk, however, charge- November 1986. Friedman, "Golden Goose or Ugly Duck- offs are not necessarily a problem. 2. However, this does not necessarily ling?," ABA Banking journal, September 2 0- 19.3 1986, pp. 73-77. Also, the potential impact of high explain why credit-card rates have not credit-card charge-offs on the financial fallen in tandem with other loan rates. 7. The Consumer Protection Act (October 18- 17.4 condition of banks is reduced by the 1970) also limited the cardholder's legal 3. There are approximately 3,000 banks 16.9 liabiliry for unauthorized use of the card 16.2 fact that even the most aggressive and other institutions that issue credit to a maximum of S50. banks in credit-card lending still have cards. "Interest Rate Controls on Credit 16- Cards - An Economic Analysis," Lexicon, 8. "Sears Is Discovering Discover Credit 14.9 relatively small portions of their assets Card Isn't Hitting Pay Dirt," Wall Street Inc., October 1985. and loans in credit-card receivables. journal, February 10, 1988, p. 1. - 14- 4. See Christine Pavel and Paula Binkley, "Costs and Competition in Bank Credit 9. Kathleen Hawk, "Plastic Warfare," U.S. Cards," Economic Perspectives. Federal Banker, June 1987, p. 40. ~ 0 1.0-1.9 Reserve Bank of Chicago, March-April 10. Delinquencies, or past due loans, are Less than 1 2.0-2.9 3.0-3.9 4.0-8.9 1987, pp. 3-13. The authors found econo- another measure of loan qualiry, but some Net charge-offs mies of scale in credit-card operations for Paul R. Watro is an economist at the Federal of these figures are treated as confidential Reserve Bank of Cleveland. The author a sample of small- and medium-size banks. by bank regulations. NOTE: Based on 1986 data from reports of income and condition for bank sample. would like to thank john M Davis, Mark S. 5. Figures are based on data provided by 11. A list of so-called credit-card banks is Sniderrnan, andjamesB. Tbomson for belp- banks participating in the Functional Cost provided by the American Banker, January ful comments. Research assistance was pro- Analysis of the Federal Reserve System. For 4, 1988, p. 15. vided by john N. McElravey and Daniel]. a discussion of credit-card profitabiliry, see charge-offs, therefore, do not neces- Other results were also similar to the 12. The weighted average net charge-off Martin. Glenn Canner and James Fergus, "The sarily imply inferior performance or findings listed in the table." One The views stated herein are those of the Effects on Consumers and Creditors of rate was 3.4 percent for the sample. cause for immediate concern, as long important exception was that credit- author and not necessarily those of the Fed- Proposed Ceilings on Credit Card Interest 13. Credit-card charge-off rates were as lenders are being adequately com- card balances were found to be nega- eral Reseroe Bank of Cleveland or of the Rates," Staff Study 154, Board of Governors regressed on the six variables listed in the Board of Governors of the Federal Reserve table. Each of the variables except credit- pensated for credit risk. tively related to charge-off rates when of the Federal Reserve System, October System. 1987. card balances was positively and statisti- asset size was taken into account. cally related to credit-card charge-offs at The strong positive relationship This finding, although statistically least at the 5 percent level of significance. between revenues and charge-offs is Insignificant, is consistent with the The variables collectively explained 40 also depicted in chart 3. The chart percent of the variance in the credit-card view that larger portfolios are gener- shows average credit-card revenues charge-off rates across the 148 banking ally more diversified and carry less for the whole bank sample broken organizations that were examined. risk than smaller portfolios. out into five groups according to the level of credit-card charge-offs. Banks • Conclusion in the higher charge-off groups pro- Bank credit-card lending has gressively earned higher revenues, increased rapidly since the early suggesting that they charge higher 1980s.This increase has been fueled rates and fees but lend to riskier by a range of factors, including robust BULK RATE customers. consumer demand, improved tech- Federal Reserve Bank of Cleveland U.S.Postage Paid nology, removal or relaxation of state Research Department Cleveland, OH However, the loan-quality/revenue Permit No. 385 usury laws, economies of scale, cross- P.O. Box 6387 relationship could be spurious, that selling potential, substantial decline Cleveland, OH 44101 is, other factors could be causing it. in funding cost and attractive profit When we isolated the influence of margins. Higher credit-card earnings individual factors while taking into have attracted new competitors, more account other factors, such as asset aggressive marketing and more lib- size and the percentage of loans in eral credit standards that, in turn, led credit-card balances, we still found to a sharp rise in credit-card losses. that banking organizations with higher credit-card charge-offs had signifi- cantly higher credit-card revenues. Material may be reprinted provided that the source is credited. Please send copies of reprinted materials to the editor. Address Correction Requested: Please send corrected mailing label to the Federal Reserve Bank of Cleveland, Research Department, P.O. Box 6387, Cleveland, OH 44101