The behavior of credit card interest rates during the decline in other interest-rate markets.

Evans, Stephen T.

INTRODUCTION

Although some consider 1949 to be the approximate beginning of the modern credit-card era (Brooker, 2004), the genesis of this powerful social phenomenon certainly goes back at least to the early 1900s. In 1914, for example, Western Union was using metal cards to allow its best customers to defer payments, and the cards became known as "Metal Money" (ETI, 2003). Prior to World War II it was fairly common for "department stores, communication companies, travel and delivery companies, and oil companies" (ETI, 2003) to issue credit cards to encourage and facilitate the purchases of their own products. But generally the cards were for "local purchases" at a single location and rarely used for the products of other companies.

During World War II the use of credit cards was prohibited, but the practice re-emerged following the war as customers discovered the "beauties of enjoying now and paying later." In approximately 1949 the Diner's Club card was instituted and allowed "travel and entertainment people" to use one card to charge purchases from many retailers. The success of this "one card for many businesses" encouraged others, and in 1951 the Franklin National Bank in New York issued the "Charge It" card (ETI, 2003). This was really the period of time when the idea was established of financial intermediaries handling credit card processes for other organizations.

MODERN ERA OF CREDIT CARDS

In the 1960s, credit-card organizations began to emerge that issued licensing agreements to other financial intermediaries. The Bank of America, for example, created the BankAmericard and licensed its use to other banks throughout the country. Later to be known as Visa International, the BankAmericard approach was a new dimension in the development of the credit-card industry and became the precursor of companies like Master Card and Discover. The 1970s saw a significant movement toward the internationalization of credit cards, and the 1980s were characterized by the development of ATM machines and other electronic devices designed to facilitate monetary transfers.

The use and effectiveness of credit cards was enhanced in the 1990s by the personal computer revolution--especially the emergence of the Internet. And now well into the 21st Century, the Internet and the emerging cashless society are maturing as major companions of the credit-card industry. Whether used for big-ticket items like computers or smaller items like books and videos, over 90 percent of Internet sales are facilitated by credit-cards (ETI, 2003). At the present time the five leaders in the credit-card industry are Visa International, MasterCard, American Express, Discover, and Diner's Club (ETI, 2003), but many overseas credit-card providers are making inroads in this profitable industry including Euro Card.

EVOLUTION OF THINKING

While the credit-card has evolved institutionally and organizationally, it has also evolved philosophically and strategically. Considered initially as a provider of convenience, it soon evolved as a facilitator of sales revenues with the sales themselves being the means to enhance profits. During this period it was usually expected that the card balances be paid off each month, and to encourage that practice an interest charge was instituted almost as a negative penalty to enforce compliance. But it didn't take long for the companies to discover and appreciate the profitability of the interest charges themselves, and the industry generally shifted from unpaid balances as a negative to a thinking that was at least neutral on balances being carried forward.

From this era of neutrality toward unpaid balances evolved a more aggressive strategy that even encouraged the carry-over of balances, and payments became a main source of income for the credit-card providers. As a further income source, the methods used to calculate the interest charges were "enhanced" and resulted in actual interest charges being greater than advertised or theoretical interest rates. One article, for example, quoted the Vice-Chairman of a bank who indicated that the credit-card business was among the most profitable business segments of the bank and made a return on equity of 34 percent (Zweig, 1997).

As the industry has moved into the 21st century, high interest rates on credit-card balances have continued to provide major profits for financial intermediaries, but probably the greatest evolution that has taken place more recently relates to the charging of fees. For many credit-card providers, the fees have become a strategic focus for generating earnings, and the dollar amount, types, and number of fees, have been extraordinary in growth. And all of the above has evolved into what some describe cynically as an industry where powerful profit machines are created to hook as many customers as possible and extract as much money from them as possible.

IMPACT ON THE ECONOMY AND CULTURE

Whether one thinks of the credit-card industry in such cynical terms or with a more positive orientation that it has been good for the consumer-portion of the economy, there is no question that the credit-card phenomenon has consistently grown faster than the economy as a whole and has had a powerful impact on it. In the mid-1980s approximately 75 million Americans had credit cards (Ausubel, 1991) or about 30 percent of the population. By 2002, "well over one billion cards (were) in circulation (with) the average household (having) about a dozen credit cards" (CFA, 2003).

"Since 1997 credit card issuers have nearly doubled the amount of credit they offer to consumers, to more than $3 trillion-about $30,000 per household. Revolving debt, which is almost entirely card debt, increased from $554 billion to $730 billion between 1997 and 2002" (CFA, 2003). And in 2003, the credit-card industry reached two significant milestones. For the first time in history, Americans bought more with cards than with cash (Brooker, 2004). Also for the first time, the industry came close to generating $2 trillion in transactions in one year (Brooker, 2004). Considering that GDP for the entire nation is a little more than $10 trillion, credit cards are now a key part in nearly 20 percent of the economy.

CONCERNS ABOUT THE CREDIT-CARD REVOLUTION

The fact that 20 percent of the economy is now facilitated by credit cards is obviously a positive in many respects. But there is also great concern because credit use has grown fastest among debtors with the lowest incomes. "In the early and mid-1990s, Americans with incomes below the poverty level nearly doubled their credit card usage, and those in the $10,000-$25,000 income bracket came in a close second ..." (CFA, 2003). "By 2000, about one-third of lower income families spent more than 40% of their income on debt repayment, compared to 20% of moderate income households and 14% of middle income families. (Likewise), riskier borrowers typically carry a higher debt burden, pay more interest, and suffer more defaults" (CFA, 2003). "About sixty percent of cardholders carry credit card debt from month to month. And the average credit-card debt for households that carry a balance is more than $10,000" (CFA, 2003).

"Consumer debt has more than doubled in the past ten years to a record level of $1.98 trillion ... (and) represents primarily credit card-debt and car loans and does not include home mortgages ... By itself, credit-card debt--nearly all accruing interest at double digit rates-stands at $735 billion despite the fact that many families have refinanced their homes in recent years in an attempt to deal with mounting credit card debts and fees (Nader, 2004). "Skyrocketing credit card balances is a significant factor in the increasing number of bankruptcies. Consumer bankruptcies have exceeded one million a year since 1996, reaching 1.54 million in 2002 and topping 1.25 million in the first nine months of 2003" (Nader, 2004).

Much of the credit-card burden has "been fueled by deceptive credit card operations, predatory lending and other 'easy credit' schemes (and is) causing a dark cloud across the national economy . The numbers paint a sad picture of low, moderate and middle income citizens caught in impossible burdens of debt plus mounting fees and late charges" (Nader, 2004). The characteristically high credit-card interest rates is a cause for concern, but an additional dimension of the problem is the fact that the credit-card rates do not noticeably respond when interest rates decline in other financial markets. In fact, this "stickiness" of interest rates has been the focus of much research in the fields of economics and finance.

LITERATURE SEARCH

With reference to credit-card interest rates, under classic economic theory, if and when credit-card profits become excessive, other credit-card providers will be attracted to this specialized segment of the financial markets, and the influx of new participants and resulting competitive forces will tend to pull down the rates and "normalize" the financial rewards. However, for a variety of reasons that continue to be explored, these basic economic forces do not seem to apply as well to the credit-card industry.

In a well-cited journal article in the American Economic Review, some solid research was presented to partially explain the unusual interest-rate behavior. Interestingly, the regulatory climate imposed by the government had been largely removed more than a decade earlier, and no solid evidence could be found of any significant barriers to entry, constrained inputs, significant sunk costs, or collusion that might be at work to thwart the workings of the free market (Ausubel, 1991).

With the industry appearing to have a paradigm of near-perfect competition, it was a paradox that the interest rates demonstrated such unresponsiveness to interest-rate changes in other financial markets (Ausubel, 1991). A possible explanation was found in some evidence that many consumers apparently make credit-card choices without taking into account the very high probability that they will eventually need to pay interest on their outstanding balances. In any case, there was no question in the research about the fact that credit-card interest rates do demonstrate a "stickiness" or inelasticity in their behavior relative to other financial markets (Ausubel, 1991).

In a later journal article in the American Economic Review, similar evidence was given on the "stickiness" of credit-card interest rates. It also provided conclusive evidence of abnormally high profits for many issuers and suggested that the cardholders do not conform to the behavioral assumptions of perfect competition because (1) cardholders face search costs, (2) cardholders face switch costs, and/or (3) firms would face an adverse-selection problem if they were to reduce their interest rates unilaterally. The paper further concluded, and significantly, that credit-card providers do not cut rates because consumers are not responsive to rate cuts (Calem and Mester, 1995).

In a journal article in the Journal of Political Economy, further research was presented to explain why cardholders live with higher credit-card interest rates. They found that credit cards are useful as a low-cost method of financing transactions and arranging short-term loans. Despite large interest rates, rational consumers frequently pay interest on outstanding credit-card balances rather than pay the transaction costs associated with arranging loans from banks or other financial institutions. A rational consumer may also pay interest on credit-card debts to avoid some of the costs associated with holding precautionary money balances (Brito and Hartley, 1995).

In an article in The Review of Economics and Statistics, there were similar findings to suggest that there are two distinct types of credit-card providers. Some use variable rates that move in lock-step fashion with other interest markets while other providers adopt a policy of fixed-rate charges for as long as three to five years. Consistent with this finding, the article presents a convincing argument that explains that markets of high rate-cost margins and aggressive non-price competition for new customers can exist concurrently (Stango, 2000).

Two recent studies also provide helpful data on the relationship between credit-card interest rates and other general market rates. In a journal article in the International Journal of Management, evidence is presented showing that the federal funds rate affects the credit card rate with a relatively small elasticity of 0.173 but the prime rate and cost of funds do not affect credit card rates (Yu Hsing, 2003).

As a final research note, the Consumer Federation of America presented some research that recognized the increasing interest-rate gap between the benchmark rate set by the Federal Reserve and interest rates charged by credit-card providers. The analysis identified a 4.75-percent drop in Federal Reserve interest rates in 2001 but only a 1.35-percent decline in the rates charged by credit-card providers (CFA, 2003). It is this research data that this study focused on, because a preliminary review of new data could identify no such drop in credit-card interest rates. An analysis seemed to be in order to find the discrepancy in conclusions.

PRELIMINARY OBSERVATIONS

While acknowledging the validity of the economic relationships and theories that apply to the credit-card industry and the power of the competitive forces at work in these financial markets, there nevertheless is much evidence that exceptions exist in some "pockets" of the industry. If one judges from the number of credit-card solicitations that the average American home receives each year and the nature of these advertisements, then two seemingly contradictory conclusions might be (1) that there is intense competition within the industry and (2) that profitability remains abnormally high.

The literature that has been described has done much to explain why some of these exceptions seem to exist, especially in relationship to the "stickiness" of interest rates. But the research has not yet adequately explained (1) the trend of the "introductory interest rates" that are granted to new cardholders for a limited period of time and (2) the fact that there is a difference in characteristics between various credit-card providers. It has been observed, for example, that local depository institutions with well-established customer relations provide different credit-card terms than large, aggressive credit-card mail advertisers that have less to worry about in customer relationships when they solicit new cardholders from a distance.

As to these aggressive national solicitors, the literature has not yet fully addressed the level or direction of the "ongoing interest rates" that have been charged to cardholders after the introductory rates have expired. It has been especially interesting during the last three years when other rates have declined to 45-year lows. Although the report by CFA indicated a 1.35 percent decline in credit-card rates during 2001 (CFA, 2003), the difference between that finding and this study showing no such decline might be explained by the possibility that this study focused exclusively on the mailed solicitations.

In a general review of over 1,000 unsolicited credit-card offers over the past seven years, the impression has been left that companies (1) that aggressively market their credit cards through the mail (2) at long distances from their locations (3) to potential customers who have no established community relationships with the lenders (4) tend to be more competitive in the short-term enticements, (5) are less responsive to longer-term interest rates after the introductory offers expire, and (6) largely attract many customers who can least afford to service the debts after the arrangements are established.

HYPOTHESES TO BE TESTED

Having recognized some differences between traditional financial intermediaries and aggressive credit-card solicitors, it is the purpose of this paper to focus primarily on #4 and #5 of the previous paragraph. In other words, the analysis is to look for evidence among "aggressive marketers" that introductory offers are becoming more competitive to "hook" the customers and, secondly, that financial commitments after the introductory period are not as responsive to market forces, especially those that relate to the ongoing interest rates.

Specifically, this study tests for the following possibilities: The first hypothesis is that "introductory interest rates" that are advertised in mailed credit-card solicitations have been declining. The second hypothesis is that these "introductory interest rates" have actually declined at a faster rate in recent years than other well-publicized rates that have declined to 45-year lows during the same time period. The third hypothesis is that the so-called "ongoing interest rates" charged by these credit-card solicitors after the introductory period have shown little if any response to declining interest rates in other financial markets. The fourth and last hypothesis is that these post-introductory-period rates or "ongoing interest rates" have actually been increasing while other rates in the financial markets have been declining.

METHODOLOGY

Although the main thrust of this study has been the period of declining interest rates from January 2001 through December 2003, there is value in first establishing (1) the more "normal" patterns of "general interest rates" over a longer period of time and (2) the more "normal" patterns of "credit-card interest rates" charged by the more traditional credit-card providers. Consequently, the well-known federal funds rate and prime rate used in the banking industry have been identified over an eleven-year period from January 1993 through December 2003. Also obtained were the interest rates charged by a representative regional bank over the same 11-year time period to show a "typical credit-card interest-rate pattern for a less-aggressive credit-card provider.

For the credit-card interest analysis specifically, over 1,000 credit-card solicitations have been saved over approximately seven years (from about 1997 through early 2004). The available data was less substantial for the years 1999 and 2000, but sufficient flyers were available during the periods under study so the evaluation could be accomplished. For the most recent year (2003), over 200 new advertising flyers were acquired and preserved for the study.

For the analysis, two basic approaches were initially contemplated. The first was to do a head-to-head comparison of advertising flyers received over time from single companies to determine whether these companies raised or lowered their interest rates, especially during the three-year period of greatest interest. While helpful in many regards, this procedure did not prove to be statistically significant because of the sparse number of data points that were statistically valid. The reason related to how many participants even qualified as national marketers.

This situation was referred to as early as 1991 when only 20 of the 4,000 credit-card providers were considered to be aggressive marketers (Ausubel, 1991). There are still, perhaps, no more than 30 firms that aggressively solicit customers on a national scale, and these were the target of the current study. While mail solicitations were obtained from most of these companies, much of the data had to be eliminated for various technical reasons, and the remaining data only related to eight of the companies. Four of these raised interest rates and the other four lowered rates over the test period, so the data proved inconclusive.

That left the study with what was considered anyway to be the more significant statistical sampling procedure. The main analysis was the random selection of 120 flyers from 30 credit-card providers over the three-year period of interest from January 2001 through December 2003. An additional 24 flyers were evaluated for calendar year 1998 to extend the time horizon somewhat. This will be described in the analysis section. The aggregate information from the 144 flyers (an average of three flyers per month) was used to establish average, monthly interest-rate data on both (1) the "introductory interest rates" offered by these companies and (2) the "ongoing interest rates" that were applied after the introductory rates expired.

While the evaluation was considered statistically significant in determining the general direction of interest rates over time, the analysis proved to be more difficult than first envisioned because even a single company can be involved in a variety of offers to different demographic targets. There were situations, for example, where two mailings were received at about the same time by the same person but with different rate offers. One explanation seemed to be the fact that address information is regularly purchased from various organizations and the interest-rate offers are tailored to what are considered to be the demographic characteristics of that organization. Because of these data challenges, some degree of judgment needed to be applied in data extraction, but the resulting data was randomly derived and considered statistically significant.

RESEARCHING THE INTEREST-RATE ENVIRONMENT

In starting the analysis it seemed crucial to begin by establishing the general pattern of interest rates in the financial markets over time, and that did prove helpful both for its orientation value and for identifying the most fertile period of time for analysis. The most representative data seemed to be the discount rate (that the Federal Reserve charges banks), the federal funds rate (that banks charge each other), and the prime rate (that banks charge their best customers). The discount rate, however, recently went through a change in assumptions and formula, so only the federal funds rate and prime rate were used for the analysis. In the following visual (Exhibit 1), these two rates can be seen over the 11-year period from January 1993 through December 2003:

[GRAPHIC OMITTED]

For this study, the two most significant observations from Exhibit 1 are (1) the strong relationship and lockstep pattern of the two rates and (2) the dramatic drop in interest rates that began in about January 2001 (data point #97). This interest-rate drop has continued for most of the last three years and consequently, this period became the target period for determining if interest rates contained in unsolicited credit-card offers have been affected much by changes in the general pattern of interest rates.

Before analyzing the interest rates associated with unsolicited credit-card offers, it is important to note that the interest rates charged by many and probably most traditional banking organizations do follow the interest-rate patterns of the financial markets. To illustrate this, Zions First National Bank of Salt Lake City, Utah was selected as a regional bank of intermediate size, and its credit-card charges associated with Master Card were obtained over the same 11-year period. The three rates are graphed together in the following visual (Exhibit 2):

[GRAPHIC OMITTED]

As shown in Exhibit 2, the interest-rate charges by Zions First National Bank on its Master Card (that has consistently been advertised as nine points over prime) has been strictly adhered to over the eleven-year period. Of course Zions Bank and other traditional banks in local communities with strong ties to existing customers have significant motivation to stay in line with current interest rates.

But staying within "reasonable limits" may not be as much a motivation for financial intermediaries that aggressively solicit customers from out of their local service areas. A preliminary review of the hundreds of credit-card solicitations on file suggest that these soliciting institutions generally use a combination of low "introductory rates" to entice customers to sign up for the credit cards and then high "ongoing rates" to generate high profits afterwards. Analyses also suggest that the introductory rates used to "hook" customers have been getting more competitive over time (dropping) while the ongoing rates following the introductory period have been remaining steady, perhaps even rising somewhat during the decline in interest rates in other markets.

ANALYSIS OF INTRODUCTORY INTEREST RATES

To test the first two hypotheses that relate to the "introductory rates," the advertising flyers that were randomly selected from the 36-month period did show a decline over time in the advertised introductory rates. But the individual data points that are plotted do tend to "jump all over" because of the individual strategies of the soliciting companies. It is dramatically apparent that each company has (1) a strategy for hooking the customers and (2) a strategy for making high profits afterwards. One company, for example, may have low introductory interest rates coupled with high service fees while another company may have just the opposite.

With such varied business strategies, the data points don't show a "tight" visual fit. The correlation coefficient associated with the regression line for the introductory rates only had a value of .122547, so the data would not be very good for predicting a single interest-rate value for one period, but the sum total of the 120 data points for the analysis period is statistically significant in showing the general pattern and direction of the introductory rates over the three-year period. Exhibit 3 shows (1) the average value of the data points for each month, (2) an additional "smoothing" line showing a running average of five monthly data points at a time, and (3) a straight line that shows the calculated regression line based on the data points:

[GRAPHIC OMITTED]

Although the thirty-six months of data shown in Exhibit 3 do show a slight downward trend in the advertised introductory interest rates, the numbers do not tell the whole story. Additional analysis shows that the real competitive forces in introductory rates manifested themselves before the three-year period. Although sufficient data was not available for calendar years 1999 and 2000, there was sufficient data for 1998, and it showed an average introductory interest rate of 7.017 percent for the flyers sampled for that year compared to an average introductory rate of 1.753 percent for the three-year period shown in Exhibit 3.

A look at other interest characteristics of 1998 showed the year to be compatible with the beginning of 2001. For example, the prime rate in early 2001 (February) was exactly 8.5 percent, and most of 1998 had a prime rate of 8.5 percent, so there seemed to be considerable value in splicing 1998 to the 36 months of data beginning in January 2001. As shown below in Exhibit 4, adding the introductory rates for the 12 months of 1998 to the 36 months shown in Exhibit 3 provides a more accurate picture of the lowering of introductory rates that has taken place as solicitors have become more competitive.

[GRAPHIC OMITTED]

Notice that the regression line shown in Exhibit 4 actually drops slightly below zero at the end of 2003, and that is representative of what has happened in this competitive arena. Of the random samples selected from the last nine months of 2003, only one of the nine months had a sample that showed an introductory rate greater than zero. So the first hypothesis that introductory rates have declined is statistically significant, but whether these rates have declined faster than the general interest rates depends on where the data points begin. Consequently, the second hypothesis is less significant depending on the chosen range. More will be said about the degree of decline in later pages.

ANALYSIS OF ONGOING INTEREST RATES

As to the "ongoing interest rates" charged following the introductory periods, the two hypotheses were (1) that the ongoing interest rates of these aggressive credit-card solicitors have not responded to the declining interest rates in other financial markets and (2) that the ongoing rates have actually been increasing through the general period of interest-rate decline. To test these hypotheses, a similar procedure was used as the one described for the introductory rates. For the thirty-six-month period beginning with January 2001, the 120 samples provided an average of more than three data points for each of the 36 months, and these 36 monthly averages are shown below in Exhibit 5:

[GRAPHIC OMITTED]

As with the introductory interest rates, the various business strategies have led to a pattern that "jumps all over the board." Collectively, there does seem to be a "flat" pattern over time, perhaps even a slight increase in the ongoing rates, but the profit strategies of the companies has lead to a data pattern that is not a tight fit around the mathematical regression line. With a correlation coefficient of .056764, the data could hardly be used with any confidence in predicting the ongoing interest rate for any single point in time. But the 120 data points randomly selected over the 36-month period do collectively show evidence that the ongoing rates have remained stable during the 36-month period, thus confirming Hypothesis #3 that the ongoing rates have not declined with society's other declining rates.

[GRAPHIC OMITTED]

In Exhibit 6, shown above, smoothed data (using a five-month running average) and a regression line have been added to the data on ongoing interest rates from the previous exhibit. Notice that the regression line does show a pattern that defies the declining interest rates that have been characteristic of interest rates in other financial markets during the three-year period. With a Y-intercept of 14.8047 percent and a slight positive slope of .023802 percent, there is strong statistical evidence to support Hypothesis #3 that interest rates have not declined. But the .023802 slope, although positive, does not provide strong statistical evidence to support Hypothesis #4 that ongoing interest rates are increasing as an industry pattern.

Although there is less need to splice the 1998 data in front of the 36-month data as was done with the introductory interest rates, nevertheless the additional 12 months of data is added to the data shown in Exhibits 5 and 6 to demonstrate that these credit-card solicitors have been consistent over time in keeping their ongoing rates at higher levels in spite of declining rates in other financial markets. The 48 months of data for the "ongoing rates" is shown below in Exhibit 7.

[GRAPHIC OMITTED]

For the 48-month period, the data has a correlation coefficient of .034113, so the data would have little value for predicting an interest-rate amount for a single month, but the collective data, based on the sample size of 144, does provide evidence of the same pattern that was shown in Exhibit 6 for 36 months. Namely, the Y-intercept of 15.02687 and the slight positive slope of .011029 is further evidence to support Hypothesis #3--that ongoing interest rates charged by credit-card solicitors is inelastic and continues to remain at high levels in spite of the declining interest rates in other financial markets. As to Hypothesis #4 relating to a rise in these rates, the data is not statistically significant. There is a slight upward trend, but it is not strong enough statistically to suggest that it is part of an upward trend in that segment of the industry.

Having shown in the exhibits the clear pattern of decline in introductory rates but relative stability in the ongoing interest rates, it appears that aggressive credit-card solicitors use a dual strategy of low introductory rates to lure new customers and the higher ongoing rates to generate solid profits afterwards. In other words, there is statistically significant evidence based on the sample data that Hypotheses #1 and #3 are correct-that introductory rates are declining and ongoing rates are not declining. A summary of the data is shown below in Exhibit 8 with the regression lines for the two credit-card rates plotted with the actual data for the prime rate.

[GRAPHIC OMITTED]

Economically speaking, the data shows that competitive pressures are felt in the introductory rates but ongoing rates seem to be unaffected by competitive pressures. Stated another way, it is unlikely that lowering ongoing rates would attract a significantly greater customer base. Otherwise, the credit-card solicitors would respond by using that as a stimulus.

EVALUATING THE DECLINE IN THE INTRODUCTORY RATES

While the basic pattern and direction of the two credit-card rates are generally clear, there is still a question relating to Hypothesis #2 on whether the introductory rates have declined faster than market rates. To begin with a comparison was made over the 36-month period between the prime rate (actual data), federal funds rate (actual data), and the introductory rates (regression line). The resulting data is shown in Exhibit 9 on the following page.

Notice that the lines demonstrate highly different characteristics over time with the prime rate and federal funds rate dropping rapidly in 2001 but leveling off somewhat in 2002 and 2003. The introductory credit-card rates as a linear regression line (and also the raw data for the introductory rates) over the same 36 months dropped at a slower pace in the 2001 period but at perhaps a comparable pace during 2002 and 2003. In other words, whether one dropped more rapidly than another was greatly dependent on which months were selected for evaluation. After much analysis, the only conclusion that could be reached is that Hypothesis #2 was inconclusive based on the 36-month data.

[GRAPHIC OMITTED]

Switching to the 48-month data for the prime rate (actual data), the federal funds rate (actual data), and introductory interest rate regression line, the same observations apply. As shown below in Exhibit 10, the prime rate and federal funds rate dropped more rapidly during 2001. In the other periods, the introductory interest rates for credit cards were at least comparable in decline and perhaps even a little steeper at times, but the results depended greatly on which months were selected for evaluation. After considerable analysis, the only conclusion that could be reached was that the 48-month data also left Hypothesis #2 without sufficiently conclusive evidence.

[GRAPHIC OMITTED]

Using the 48-month regression line of the prime rate as representative of the general market rates, the regression lines of the introductory interest rates and ongoing interest rates charged by the credit-card solicitors are all plotted together in Exhibit 11 on the following page. While the slope of the ongoing interest rates is a positive .011029, the slope of the prime rate is a negative .124408 and the slope of the introductory rate is a negative .143449. It is true for the 48-month data that the introductory rate has declined a little more rapidly than the prime rate.

[GRAPHIC OMITTED]

For the 36-month period, which is cleaner data, the slope of the ongoing rate is a positive .023802 while the slope of the prime rate is a negative .119718 and the slope of the introductory rate is a negative .032317. In other words, for the 36-month data, the introductory rate is actually declining a little less rapidly than the prime rate (just the opposite of the 48-month data). The statistical characteristics of the data are shown in the following table:

Whether using 36-month data or 48-month data, the overall visual impression of Exhibit 11 is the main message. Introductory interest rates charged by the aggressive, mail solicitors of credit cards have fallen over time commensurate with the decline in other general interest rates, but the ongoing interest rates charged to credit-card customers are inelastic. Indeed, they may be described as stubbornly resilient in a period of declining interest rates.

SUMMARY AND CONCLUSIONS

The focus of this paper has been on the advertising materials that are typically sent to American homes by the so-called aggressive national marketers of credit-card services. The 144 advertising flyers from 30 of these national solicitors were analyzed and yielding the following results in relationship to the four hypotheses:

As to the first hypothesis, the research data leaves no question that introductory rates have been declining in recent years. The sample size of 144 was statistically significant and the trend of these rates can be readily seen in the exhibits that have been shown.

As to the second hypothesis, the question of whether the rates have been declining faster than other rates in society is considered inconclusive. The results depended primarily on the time period being evaluated. In the 36-month data, the introductory rates declined slightly less rapidly than the general interest rates used in the analysis. In the 48-month data, the introductory rates declined a little more rapidly than the general interest rates. Overall, it can be said that the introductory interest-rate declines have been at least comparable to declines in other rates.

The ongoing interest rates that were the focus of the third hypothesis have without question defied the interest-rate declines in other financial markets. This was statistically significant over both the 36-month data from 2001 through 2003 and the 48-month data. In other words, the interest rates charged by the mail advertisers of credit cards continue to resist any downward pressures. They are resilient and inelastic.

With the 144 data points that were randomly selected to test the fourth hypothesis, there was a slight increase in these rates over the time periods tested, but the results were not considered statistically significant. The interest levels remained robust and "sticky" at their recent levels, but the increases were not statistically strong enough to conclude that they were part of an upward trend in the industry.

AN ADDITIONAL OBSERVATION

The quantitative and qualitative treatment of the research information, although providing great insights, did not fully divulge an even broader phenomenon that is taking place in the credit-card industry. The motivation of credit-card providers is naturally to make as much profit as possible, and when interest charges are the primary source of income, many cardholders "escape the hook" by paying their balances in full each month. Consequently, a significant increase in fees continues to be put in place by credit-card providers. Throughout the research there has been an increasing awareness of this momentum.

To illustrate the significance of this phenomenon, the following statement is contained in an advertising flyer that has been received twice in recent months. In the mass of information that is contained in the contractual understanding of the application form, it states: "Available Credit and Cash Advance Limitations: The initial minimum credit limit will be $250 and the following fees will be billed to your first statement: Annual Fee of $48.00, Program Fee of $95.00, Account Set-Up Fee of $29.00, monthly Participation Fee of $6.00 [$72.00 annually], and an Additional Card Fee of $20.00 (if applicable). If you are assigned the minimum credit limit of $250 your initial available credit will be $72 ($52 if you choose the additional card option)."

In addition, the itemized fees above do not even include such other fees as minimum finance charge of $0.50, transaction fee for cash advances which is the greater of $5.00 or 3% of the amount of the cash advance, late payment fee of $25.00 each time the payment is late, over limit fee of $25 each month the balance exceeds the credit limit, $25.00 for returned checks, possible Copying Fees of $3.00, Internet Access Fee of $3.95, Autodraft Fee of $11.00, and Voice Response Fee of $7.00 per transaction. As to the Voice Response Fee, the cardholder who signs up better not make a later telephone call to complain, because $7.00 will be tacked on with each phone call.

Although the research project focused on the interest rates, as the rate information was being extracted for the quantitative analysis, there was increasing amazement at the type of fees, number of fees, amount of fees, and trend of these fees. After evaluating 144 of the flyers it can be confidently stated that if the fees and interest rates had been evaluated together, the upward trend in credit-card charges would not only have been statistically significant (as concluded in the third hypothesis), but dramatically upward in the slope of the regression line. There is no question that this is "fodder" for the next research project.

REFERENCES

Ausubel, L. M. (1991). The Failure of Competition in the Credit Card Market. The American Economic Review, 81(1), 50-81.

Ausubel, L.M. (1997). Credit Card Defaults, Credit Card Profits, and Bankruptcy. American Banker, LJ 249.

Balto, D. A. & B. K. Grube (2002, May). The Visa-Master Card decision and its implications for high-tech markets. The Business Lawyer, 57(3).

Berlin, M. & L. J. Mester (2004). Credit card rates and consumer search. Review of Financial Economics, 13(1), p. 179.

Bhandari, S. B (2002). Some ethical issues in computation and disclosure of interest rate and cost of credit. Journal of Business Ethics. 16(5), p. 531-535.

Bianco, C. A. & S. M. Bosco (2002). Ethical Issues in Credit Card Solicitation of College Students--The Responsibilities of Credit Card Issuers, Higher Education, and Students. Teaching Business Ethics, 6(1), p. 45.

Boyer, M. (2003, January 8). Industry Notes: Banking. The Cincinnati Enquirer.

Brito, D. L. & P. R. Hartley (1995). Consumer rationality and credit cards. The Journal of Political Economy, 103(2), p. 400-433.

Brooker, K. (2004, February 23). Just One Word: Plastic--How the rise of the credit card changed life for the FORTUNE 500-and for the rest of us. Fortune, p. 125-138.

Calem, P. S. & L. J. Mester (1995), Consumer Behavior and the Stickiness of Credit-Card Interest Rates. The American Economic Review, 85(5), p. 1327-1336.

Cargill, T. F. & J. Wendel (1996, Winter). Bank credit cards: Consumer Irrationality versus market forces. The Journal of Consumer Affairs. 30(2), p. 373-389.

Consumer Federation of America (CFA) (2003). Facts About Consumer Credit Card Debt and Bankruptcy. P. 4.

Cornett, M. M. & J. Musumeci (1999, Autumn). How Legislation Affects Value: The Failure of Credit Card Cap Legislation. Financial Management.

Gross, D. B. & N. S. Souleles (2002, February). Do Liquidity constraints and interest rates matter for consumer behavior? Evidence from credit card data. The Quarterly Journal of Economics. 117(1), p. 149.

Hamilton, R. & M. Khan (2001, July). Revolving Credit Card Holders: Who Are They and How Can They Be Identified? Service Industries Journal, 21(3), p. 37-48.

Kara, A., E. Kaynak & O. Kucukemiroglu (1994). Credit card development strategies for the youth market: The use of conjoint analysis. The International Journal of Bank Marketing. 12(6), 30-36.

Mantripragada, K. G & H. Benerjee (1992). Credit Card Interest Rate Caps Do Not Make Sense. Review of Business, 14(1), p. 19-24.

Mayer, R. N. & C. D. Zick (1996). The cost of "free" credit card benefits. Consumer Interests Annual, 42, p. 161-166.

Moskowitz, H., D. Cohen, B. Krieger & S. Rabino (2001). Interest and reaction time analysis of credit card offers: Managerial implications of high level research procedures. Journal of Financial Services Marketing. 6(2), p. 172-189.

Mosler, W. (2003/2004, Winter). Credit cards and interest rates: theory and institutional factors-a critical view. Journal of Post Keynesian Economics, 26(2), p. 303-307.

Nader, R. (2004, January 9). Deeper and Deeper in Debt. The Nader Page, www.nader.org.

Parloff, R. (2004, February 23). The $50 Billion Card Game. Fortune.

Reddy, A. (2003, April 16). Consumer Loans Boost Bank Profits. Washington Post, p. E01.

Scholnick, B. (2000). Regulation, competition, risk in the market for credit cards. Canadian Public Policy, 26(2), p. 171.

Sinkey, J.F., Jr. & R. Nash (1997). On Competition, Risk, and Hidden Assets in the Market for Bank Credit Cards. Journal of Banking and Finance, p. 89-112.

Stango, V. (2000, August). Competition and pricing in the credit card market. The Review of Economics and Statistics, 82(3), p. 499-508.

Stauffer, R. F. (2003/2004, Winter). Credit cards and interest rates: theory and institutional factors. Journal of Post Keynesian Economics, 26(2), p. 289-301.

Stavins, J. (1996). Can demand elasticities explain sticky credit card rates? New England Economic Review, p. 43-54.

Stavins, J. (2000, Jul/Aug). Credit card borrowing, delinquency, and personal bankruptcy. New England Economic Review, p. 15-30.

Steindl, F. G. (2000, Summer). Credit Cards, Economization of Money, and Interest Rates. Journal of Economic Education, 31(3), p. 271-279.

Toh, K. G. (1996, March). Are credit-card late fees "interest? Michigan Law Review, 94(5), p.1294.

Warnick, J. & P. Mansfield (2000). Credit card consumers: college students' knowledge and attitude. The Journal of Consumer Marketing. 17(7), p. 617.

Yu Hsing, J. G., T. Lin & D. Wallace (2003, September). Determinants of credit card rates and policy implications. International Journal of Management, 20(3), p. 395.

Zweig, P. L. (1997, October 27). The New Stars of Finance. Business Week, p. 122-125.

Stephen T. Evans, Southern Utah University

Summary of Regression and Correlation Values

 Intercept

Data for Ongoing Rates +14.80467
36 months Prime Rate +7.454508
 Federal Funds Rate +4.416365
 Introductory Rates +2.350635

Data for Ongoing Rates +15.02687
48 months Prime Rate +9.066339
 Federal Funds Rate +6.051569
 Introductory Rates +6.583245

 Slope Correlation
 Coefficient

Ongoing Rates +.023802 +.056764
Prime Rate -.119718 +.872041
Federal Funds Rate -.118332 +.869246
Introductory Rates -.032317 +.122547

Ongoing Rates +.011029 +.034113
Prime Rate -.124408 +.940051
Federal Funds Rate -.124197 +.939417
Introductory Rates -.143449 +.510362