Printer Friendly

Helping consumers choose a credit card.

This article focuses on consumers' use of information to choose credit cards. The objectives are to describe a Canadian governmental information program on credit card costs and present the results of a study designed to test the effectiveness of specific modifications in the information presentation format. For many people, the credit card has become a practical and natural way of purchasing consumer goods and services. Since the late 1970s the growth in use of credit cards in the United States and Canada has been substantial. In 1977 in Canada, there were 8.18 million bank ards (Visa and MasterCard) in circulation representing $3.61 billion in sales transactions. In 1988, the number of bank cards possessed by Canadian consumers had risen to 19.40 million with $30.33 billion in sales (Turner 1989). nearly ten percent of all consumer transactions are made with bank cards. These figures attest to the importance of credit card usage as a consumption phenomenon.

The utilization of credit cards incurs potential financial costs. Financial institutions may charge customers a fixed fee per annum or per transaction and interest on monthly balances not paid in full. The interest rate applied in these cases is usually much higher than the bank rate. For example, the Visa-bank rate spread in Canada at the end of September 1989, was 7.52 percentage points (Turner 1989). Consumers may end up paying more than necessary because they do not understand how interest charges are computed by card issuers. Information about fees, interest rates, and methods of calculating interest charges is often incomplete and hard to understand (Edwards 1982). Policy-makers see the lack of informaton about credit card costs as a threat to market competitiveness and to consumer interests. The Fair Credit and Charge Card Disclosure Act of 1988 passed by the U.S. Congress represents governmental intervention aimed at providing consumers cost information on credit cards at the time they are solicited by card issuers. The Act requires the disclosure of the annual percentage rate, fees, grace period, and balance calculation method associated with any credit card.

In Canada, there is no specific legislation on credit cards, although a recent report of the Standing Committee on Consumer and Corporate Affairsn and Government Operations recommends that the method of interest calculation be regulated and that a limit on credit cards' interest rates be imposed (Turner 1989). In 1987, however, the Canadian government showed its concern for credit card issues by creating a program to provide consumers with regularly updated information on credit card costs. Three times a year, the Department of Consumer and Corporate Affairs Canada (CCAC) makes available to the general public the most recent data on interest rates, grace periods, and moment at which interest charges begin for 16 bank cards (e.g., Bank of Montreal MasterCard), 12 retail cards (e.g., Sears' card), and seven charge cards like American Express. In this article, an evaluation of CCAC's consumer information program as it pertains to bank credit cards (only) is presented. An empirical study designed to examine the effects of modifications in the information presentation format is reported.


Since its inception in 1987, the CCAC information program on credit card costs has pursued the objective of informing consumers on a continuous basis about the costs involved in using credit cards so that they are able to make informed choices among the credit cards available to them (Consumer and Corporate Affairs Canada 1990). The CCAC report presents comparison information on credit cards in a matrix array (Table 1).


One presumed advantage of the CCAC information table is that it reduces the costs (time, money, frustration, etc.) that consumers must assume in order to get comparison data to make informed and eventually better choices. To evaluate the effectiveness of the information program however, several other factors must be considered. As a first step, one must determine if the information reaches credit card users. Although reliable data on this issue are lacking, there are reasons to believe that consumer exposure may be an important problem. Upon publication, the credit card information is sent to CCAC regional offices in Canada which make the information available to the public. It is also sent to the media and journalists interested in consumer issues who are free to decide whether they will publish it. Unless consumers visit a CCAC office or directly ask for the information, it is doubtful that they are exposed to the information or even aware of its existence.

Though consumer exposure to the information represents a potentially serious problem for the CCAC program, other questions related to the interpretation and integration of the information must also be addressed (Peter and Olson 1990). Do consumers pay attention to the information? Do they understand the various pieces of information in the CCAC table? Can they combine the pieces of information to form evaluative judgments and make appropriate choices.?


Credit Cards

The work of mathews and Slocum (1969; 1970) is relevant to the present discussion. These researchers distinguish between "convenience" and "Installment" users. The former utilize the credit card mainly as a mode of payment and typically pay the balance in full upon receiving the account statement. Installment users on the other hand use the card principally as a mode of financing and elect to pay interest charges on the unpaid balance. The distinction is important because it conditions any judgment about the appropriateness of credit cards (Chisholm 1987; Durand 1986). Thus, for installment users it is preferable to have a low interest card with interest charges that are calculated from the date of the statement. The best credit card for convenience users would be one with no fixed fees and a long grace period. Clearly, knowledge of one with no fixed fees and a long grace period. Clearly, knowledge of one's type of utilization can simplify the process of choosing an appropriate credit card.

Note that this view of what is an appropriate credit card strictly relates to economic factors. This is not to say that consumers should only select a credit card on the basis of such factors. There may be perfectly valid noneconomic reasons for choosing a credit card, such as the perceived convenience of using a card offered by the bank with which the consumer does business or the desire to diversity one's credit sources. The focus on economic factors in this study is dictated by the context of the inquiry, namely to examine an information program whose objective is to help consumers identify the lowest cost credit cards. Other issues that policy-makers must consider in their efforts to help consumers are addressed in the discussion section.

Consumer Information processing

There is a large consensus in the consumer research community that simply providing consumers information is insufficient; consumers must also be able to process the information (Bettman 1975; Muller 1977; Russo 1977; Russo et al. 1986). However, "processability" is a relative concept; it depends on consumers' knowledge and motivation. Therefore, one useful way of looking at the adequateness of consumer information programs is cost-benefit analysis (Russo et al. 1986). According to this view, the more benefits associated with information processing exceed the costs, the higher the likelihood that consumers will process and use the information. The corollary to this general principle is that increasing benefits or decreasing costs should enhance information use.

In order to assess consumers' ability to process the information contained in the CCAC table, a preliminary study was conducted. A convenience sample of 68 individuals divided about equally into installment and convenience credit card users was shown an information matrix similar to that in Table 1 and asked to pick the card they thought best for their needs. Subjects were free to take as much time as needed to make thir choice. The results show that 26 consumers (38 percent) chose a card adapted to their particular utilization type. Moreover, the majority of participants (84 percent) reported that the informaton was sufficient to make a good decision. From this small-scale study it is concluded that although the CCAC table allows consumers to make informed choices, it does not necessarily lead them to appropriate choices, that is choices that are economically preferable given their usage type (installment versus convenience). Moreover, assuming that the subjects were motivated to accomplish their task seriously, the results suggest that improvements in choice would be more likely to come about by reducing the processing efforts of consumers than by convincing them of the benefits of considering the information on credit card costs to make a better choice.

Some authors argue that reducing processing efforts may be more effective than enhancing benefits in changing consumer behavior. In the area of consumer

nutrition, for instance, Russo et al. (1986) examine the findings of several studies of information programs designed to influence consumers to buy more nutritious food. They conclude that effort-reduction strategies (e.g., providing nutritional summary ratings) are in general more successful than benefit-enhancement strategies (e.g., promotion). In their study, presenting information on the level of added sugar in breakfast cereals has a significant impact on purchase behavior. According to Russo et al., ". . . these effects are directly attributable to a reduction in the effort needed to compare brands of breakfast cereal on added sugar" (1986, 65). It is not clear that these findings are generalizable to the domain of financial services. One can argue that selecting a credit card is more involved than choosing a brand of breakfast cereal. The results of the preliminary study suggest that there is room for improvement in increasing the processability of the CCAC information table.


The preceding discussion leads to two conclusions that guide this study. First, it appears that the economic appropriateness of a credit card depends on whether the consumer is a conenience or installment user. Second, effort-reduction strategies in information presentation appear to be effective in leading consumers to process and use information, especially when involvement is not high.

Alternative Information Formats

Russo et al. (1986) argue that summary ratings represent an effective way of reducing processing efforts. Based on their findings, two alternative information presentation formats were developed in order to facilitate processing of the credit card information. Figure 1 presents the summary format. In the upper part of Figure 1, the consumer is first categorized as an installment or convenience user and directed toward specific rating symbols. For convenience users, the best credit cards are those with no fees and longer grace periods. In Figure 1, these cards are labeled "**" and "*". For installment users, the best credit cards are those that offer the lowest interest rates with interest charges calculated from the date of the statement and the longest grace periods. In Figure 1, these cards are labeled "+ + +", "+ +", and "+". Cards that are not good options for either installment or convenience users do not get a rating.
 Bank of Montreal **
 Canada Trust +
 Canada Trust Supercharge *
 CS CO-OP **
 National Bank **
 National Trust **
 Bank of Nova Scotia +
 Canadian Imperial Bank of commerce
 Central Guaranty **
 Centre Desjardins +
 Guaranty Trust **
 Laurentian Bank + + +
 Royal Bank
 Toronto Dominion Bank + +
 Vancouver City Savings

Assigning ratings to the credit cards proved to be a difficult task. Consider as a case in point the ratings for installment users. Even though the important attributes are interest rate and grace period, it is evident that fees--an otherwise unimportant attribute--may be used to discriminate between cards that have equal interest rates and grace periods (cf., Visa TD Bank versus Bank of Nova Scotia). Moreover, because interest charges depend not only on interest rate but also on the balance on which the charges are calculated, it is difficult to make any straightforward comparison between a card with a lower interest rate and one with a lower annual fee (cf. Visa Laurentian versus TD Bank). Some elements of subjectivity necessarily affected classification decisions, particularly when such tradeoffs among attributes must be made. The intent is not to minimize the issue. However, it must be kept in mind that the objective of this study is not to put forward a method of evaluating credit cards but rather to assess the impact of lowering information processing efforts on consumer choice behavior.

The other alternative information presentation format combines both the matrix information and the summary ratings. Specifically, it comprises a categorization section as in Figure 1 followed by a matrix similar to that presented in Table 1 with an additional column for the ratings.

Research Hypotheses

It is proposed that information format will influence choices made by consumers and perceived usefulness. The following research hypotheses detail the nature of this predicted influence.

H1: The availability of summary ratings, whether presented alone or in combination with information on attributes, will lead to choices of credit cards that are more economically appropriate for consumers.

H2: Consumers will judge that a combined information format is more useful than a matrix format and that the matrix format is more useful than the summary ratings format.

The first hypothesis follows from the observation that in order to choose an appropriate credit card, one must be aware of type of utilization and be able to process comparison information. From the preliminary study, it appears that the CCAC matrix information table does not permit consumers to choose credit cards that are best for them given their particular type of utilization. The alternatie information formats should be more effective because credit card usage type is explicitly considered and information processing costs are lower.

The second hypothesis follows in part from Russo et al. (1986) who find that presenting nutrition summary ratings only leads to negative consumer attitudes toward nutrition. In contrast, presenting nutrition information in a matrix format has a positive impact on attitudes. Although in the present case the dependent variable is not attitudinal but rather a measure of perceived usefulness of the information, it is believed that negative attitudes toward the ratings may form and affect the perceptions of usefulness.


A survey was conducted to test the research hypotheses. A probabilistic sample of consumer from a midsize French-Canadian city were shown one of three formats of credit card cost information and asked to choose the card that seems most appropriate for them.

Sampling and Data Collection Procedures

The city was divided into three geographic sectors and streets were drawn randomly in each sector. A total of 480 home addresses were selected by taking one address at every two residences in each street. Two interviewers visited all home addresses and asked for the residents' participation in the study. To be considered eligible, individuals had to possess at least one credit card. When several eligible persons lived in one dwelling, respondent selection was accomplished using the "birthday" method (O'Rourke and Blair 1983). That is, the person with the most recent birthday was asked to complete the questionnaire. Response rate information is presented in Table 2. In order to assess the representativeness of the sample, comparisons were made with 1986 Canadian census data. More precisely, the distributions of respondents across age and sex groups were compared to those of the city and of the province of Quebec. Because the participants in the study had to meet a condition of eligibility (possession of at least one credit card), there is an under-representation of the lower (less than 25 years) and upper (75 years or more) age strata. The distributions in other are groups, for both males and females, are fairly similar.


Response Rate Information
 Visited Households 480
 Not at home 122
 Not Eligible (no credit card) 101
 before questionnaire transmission 32
 after questionnaire transmission 6
 Unusable Questionnaires 15
 Contact Rate 46.9
 Completion Rate 90.7

Format Manipulation and Measures

The questionnaire include one of three information formats: the CCAC matrix table (Table 1), the summary format (Figure 1), or the combined format. The credit card cost information was the most recent available at the time the study was conducted (i.e., winter 1988). Assigning treatments to subjects was done by rotating formats across visited dwellings in a systematic manner.

Subjects' main task was to pick a credit card that they considered best among those presented in the table. following their choice, they were asked to write down the reasons why they selected that particular card. Subjects also completed an eight-item scale was adapted the perceived usefulness of the information. The scale was adapted from an instrument initially developed to measure the perceived usefulness of advertisements (Perrien and Heppel 1985). Respondents indicated their agreement of a five-point scale with items such as "The table does not provide useful information on credit cards." In order to allow a classification of subjects into convenience and installment groups, one question concerned how the credit card bills were usually paid. The possible answers were "I pay the entire balance for the due date" (classified as convenience) or "I pay part of the balance for the due date and the rest the following months" (classified as instllment). Lastly, subjects provided standard socio-economic data.


The sample comprised 125 men and 79 women. The youngest participant was 21 and the oldest 83 years old, the sample average being 45. the mean income was 29,342 Canadian dollars with a standard deviation of 18,427. The respondents were relatively well educated; 35.8 percent held a university degree, 32.8 percent a college degree, 28.9 percent had completed high school, and 2.5 percent primary school.

Preliminary Checks

As noted in the method section, respondents wre not randomly assigned to treatments. Rather, treatments were rotated systematically across visited households. In order to verify that subjects' characteristics were similar across experimental conditions, a series of bivariate analyses (analyses of variance and cross-tabulations) were conducted using age, income, brand of credit card possessed, occupation, and education as dependent variables and format as the independent variable. None of the analyses results in statistically significant differences between the format conditions.

Credit Card Choice

Table 3 presents the choice data. To simplify, credit card choices that are best, very good, or good for a given usage type are categorized


Credit Card Choice Data
 Sample Frequency
 Format Usage Type Choice (n)
 Matrix Convenience Appropriate 31
 Matrix Convenience Other 21
 Matrix Installment Appropriate 7
 Matrix Installment Other 5
 Summary Convenience Appropriate 51
 Summary Convenience Other 7
 Summary Installment Appropriate 11
 Summary Installment Other 2
 Combined Convenience Appropriate 44
 Combined Convenience Other 13
 Combined Installment Appropriate 8
 Combined Installment Other 4

as "appropriate" and all other choices as "other." As can be seen from the table, the sample comprises a large number of convenience users (81.9 percent). This sampling result is puzzling because recent research indicates that about 50 percent of Canadians having one or more credit cards are installment users (Morrisset 1987; Wells 1987). In this study, usage type is defined from respondents' report of credit card payment habits, therefore the proportion of true installment users may be underestimated. On the other hand, one cannot exclude the possibility of a regional difference.

Overall, in the matrix information format condition, 59.6 percent of convenience users and 58.3 percent of installment users make an appropriate choice. In the summary ratings format condition, the percents are 87.9 and 84.6, respectively. In the combined format condition, they are 77.2 and 66.7 percent. It is worth noting that the percentage of appropriate choices is slightly superior for convenience users in the latter two conditions (however, as shown later the oberved differences are not statistically significant). This can be explained by the fact that seven of 15 credit cards are appropriate for convenience users as compared to five of 15 for installment users. Thus, the probability of a correct guess is somewhat higher for convenience users.

A logit analysis is conducted to test H1. The dependent variable of interest is the proportion of respondents having made and appropriate choice. The sample proportions are transformed into logits and the effects of format and usage type are estimated using the Grizzle, Starmer, and Koch (1969) (known as GSK) methodology. This approach uses a weighted least square estimation procedure to correct for heteroskedasticity. In the present case, the transformed dependent variable corresponds to the natural logarithm of the odds of making an appropriate credit card choice.

Figure 2 presents a plot of the logits in the different conditions of the design. The GSK analysis results in a nonsignificant main effect of usage type ([X.sup.2] = 0.45; 1 d.f.; p = .5046) and a statistically significant main effect of format ([X.sup.2] = 12.91; 2 d.f.; p = .0016). The residual (interaction effect) [X.sup.2] is not significant (p = .8827), which implies that the model fits the data well. Specific contrasts between format conditions support H1; the matrix format leads to worse credit card choies than the summry format (z = 1.93; p = .0262; one-tailed test) or the combined format (z = 3.54; p = .0002; one-tailed test.).

Information usefulness

Figure 3 presents a plot of the perceived information usefulness means in the different conditions of the design. Because the number of observations varies across conditions, Appelbaum and Cramer's (1974) model comparison approach to the analysis of nonorthogonal ANOVA designs is used. Neither the format x usage type interaction nor the usage type main effects are significant (F < 1 in both cases). However, there is, predicted, a statistically significant main effect of format (F = 60.21; 2, 202 d.f.; p = .0001). A priori contrasts between format conditions show that the combined format information is judged to be more useful than the matrix information (t = 2.36; 201 d.f.; p = .0096; one-tailed test). The matrix format information is judged to be more useful than the summary ratings information (t = 7.86; 201 d.f.; p = .0001; one tailed test). This patter of results corresponds to H2, namely that the format perceived as most useful by consumers would be the combined format, followed respectively by the matrix and summary ratings formats.


The Consumer and Corporate Affairs Canada information program provides consumers with current information on charges and interest calculations mechanisms of available major credit cards. The objective of this program is to influence the choice behavior of credit card users. The present study reveals that the manner of information presentation that the CCAC program employs to attain this objective is not optimal. Credit card users need more than a matrix of comparison data to make economically appropriate choices. They must be made aware of their personal credit card usage type and directed toward the consequent best options. Such a strategy is consistent with the effort-reduction approach to information provision advocated by Russo et al. (1986) and others (e.g., Bettman 1979).

Reducing the efforts needed to process information leads to better consumer decisions. The results of this study confirm that the best information provision format is one where summary ratings of credit cards are included. This type of format greatly reduces processing efforts; consumers only have to spot the "all star" cards and make a choice. Although it is most efficient in terms of choice appropriateness (Figure 2), the provision of summary ratings only may not be a commendable strategy for policy-makers. Consumers in this study judge that the information contained in the summary format is least useful. On the other hand, they think that the information transmitted through the combined format is most useful. Consequently, the combined format seems to represent an acceptable compromise between efficiency and consumer satisfaction. It is interesting to note in this regard that the difference in choice appopropriateness between the summary ratings and combined formats (using logits) does not reach statistical significance ([X.sup.2] = 3.25; 1 d.f; p = .0716).

One must note that these conclusions takes as a premise that the data are valid. As in any survey, there may be biased reporting by participants and this may influence the correctness of the judgments that are made concerning choice appropriateness. Recall that the proportion of self-reported convenience users in this study is higher than expected. This leads to the belief that some credit card installment users are not candid in answering a somewhat intrusive question. It is difficult to assess the extent to which this possible biased reporting would affect the findings. Note that if the analysis is limited to self-reported installment users only--where biased reporting seems improbable--the pattern of results is unaffected (Figures 2 and 3).

This study has focused on the information processing side of consumer information. Policy-makers should note that other issues have to be considered in developing information programs to influence consumer choice behavior in the area of credit cards. The importance of information exposure has already been discussed in this article. In addition, knowledge of credit card choice processes is crucial when assessing the impact of information provision. For instance, in this study respondents admit that the choice of their present credit card(s) was influenced by their financial institution (60.8 percent) and by the reputation and acceptance of the card in commercial outlets (50 percent). Fees are mentioned by 31.4 percent of the respondents and interest rates by three percent. These results suggest that consumers utilize information other than costs in the process of deciding which card(s) they will own. It would seem necessary to assess the weight of all types of information in credit card choice processes in order to be able to evaluate the effectiveness of a consumer information program.

Another issue has to do with the additional costs associated with credit card switching. Assuming that the CCAC program--or for that matter any information program on credit cards--is successful in making consumers realize that their current card is not the most appropriate, is that sufficient to provoke brand switching? The costs associated with applying for a new credit card are not null (filling out the form, evaluation of credit, waiting time, etc.). Moreover, interest rates, fees, and method of interest calculation are not fixed, so that a best card at some point in time may become a not so good alternative later. In sum, a consumer information program may fail for reasons other than a failure to communicate the information to the intended audience.


Financial services represent a domain where a great deal needs to be done in terms of providing information that consumers can use to compare competitive offerings. Information availability is a fundamental component of the free market system. Governmental intervention to help consumers make informed choices in the sector of credit cards is therefore important and should be encouraged. However, such intervention will have an impact only if the information programs conceived allow consumers to improve the quality of their decisions. Consumer research can be used to help policy-makers develop better programs to inform the public about credit card costs. The issues involved are numerous and a single study cannot be expected to address them all. Research is needed to complement the findings presented here.


Appelbau, M. I. and E. M. Cramer (1974), "Some Problems in the Nonorthogonal Analysis of Variance," Psychological Bulletin, 81: 335-343.

Bettman, James R. (1975), "Issues in Designing Consumer Information Environments," Journal of Consumer Research, 2(December): 169-177.

Bettman, James R. (1979), An Information Processing Theory of Consumer Choice, Reading, MA: Addison-Wesley.

Chisholm, Patricia (1987), "B of M Leads in Fierce Credit Card War," Financial Times (October): 24.

Consumer and Corporate Affairs Canada (1990), "Government Responds to Report of the Standing Committee on Credit Card Costs," (News Release #25328 (March).

Durand, Michel (1986), "Des solution au casse-tete du credit," Le Magazine Affaires Plus, 9(November): 32-54.

Edwards, Patrick (1982), "How to Milk the Best Deal out of Your Credit Cards," Canadian Business, 5(May): 153-154.

Grizzle, J. E., C. F. Starmer, and G. C. Koch (1969), "Analysis of Categorical Data by Linear Models," Biometrics, 25 (September): 489-504.

Mathews, A. Lee and John W. Slocum (1969), "Social Class and Commercial Bank Credit Card Usage," Journal of Marketing, 33: 71-78.

Mathews, A. Lee and John W. Slocum (1970), "Social Class and Income as Indicators of Consumer Credit Behavior," Journal of Marketing, 34: 69-74.

Morrisset, Paul (1987), "L'homme qui a fait trembler les banques," L'Actualite 12 (September): 120-122.

Muller, Thomas E. (1985), "Structural Information Factors Which Stimulate the Use of Nutrition Information: A Field Experiment," Journal of Marketing Research, 22(May): 143-157.

O'Rourke, Diane and Johnny Blair (1983), "Improving Random Respondent Selection in Telephone Surveys," Journal of Marketing Research, 20(November): 428-432.

Perrien, Jean and Denyse Heppel (1985), "Consumers' Perceive Usefulness of an Advertisement," in Marketing Communication: Theory and Research, Michael Houston and Richard J. Lutz (eds.), Chicago: American Marketing Association.

Peter, J. Paul and Jerry C. Olson (1990), Consumer Behavior and Marketing Strategy, 2nd edition, Homewood, IL: Irwin.

Russo, J. Edward (1977), "The Value of Unit Price Information," Journal of Marketing Research, 14(May): 192-201.

Russo, J. Edward, Richard Staelin, Catherine A. Nolan, Gary J. Russell, and Barbara L. Metcalf (1986), "Nutrition Information in the Supermarket," Journal of Consumer Research, 13(June): 48-70.

Turner, Garth (1989), Charge It: Credit Cards and the Canadian Consumer, Report of the Standing Committee on Consumer and Corporate Affairs and Government Operations, Ottawa.

Wells, Jennifer (1987), "Credit Card Jungle," Financial Post, 81(January): 1-2.

Alain d'Astous is Professor of Marketing, Faculte d'administration, Universite de Sherbrooke, Sherbrooke, Canada; and Diane Miquelon is Products Coordinator, Bank of Montreal, Montreal, Canada.
COPYRIGHT 1991 American Council on Consumer Interests
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1991 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:D'Astous, Alain; Miquelon, Diane
Publication:Journal of Consumer Affairs
Date:Dec 22, 1991
Previous Article:The emergency food relief system: an empirical study.
Next Article:Dynamics of complaint management in the service organization.

Related Articles
Payment: differing perceptions.
Relationships among Information Search Activities When Shopping for a Credit Card.
Taking charge: attention credit card companies: when we want you to charge us hidden fees, we'll let you know.
More transparency from card firms; CONSUMER CREDIT.
Leukemia & Lymphoma Society launches affinity card.

Terms of use | Copyright © 2017 Farlex, Inc. | Feedback | For webmasters