Antecedents of Spending and Saving Habits and Use of Credit Cards in the Midwestern United States

ABSTRACT - A telephone survey was conducted in seven Midwestern States in the United States to assess household spending and saving habits. Household tendencies to save or invest, use credit cards, and pay bills in full were related to sociodemographic factors such as age, education, marital status, employment status, and rural vs urban living environment, and to perceptions of personal and national economic well-being. Results were related to differing economic attitudes among different demographic groups and it was suggested that the predictor variables employed here may be useful for explaining regional and national differences.

Citation:

Irwin P. Levin and Shu-Fang Kao (1993) ,"Antecedents of Spending and Saving Habits and Use of Credit Cards in the Midwestern United States", in E - European Advances in Consumer Research Volume 1, eds. W. Fred Van Raaij and Gary J. Bamossy, Provo, UT : Association for Consumer Research, Pages: 459-466.

European Advances in Consumer Research Volume 1, 1993      Pages 459-466

ANTECEDENTS OF SPENDING AND SAVING HABITS AND USE OF CREDIT CARDS IN THE MIDWESTERN UNITED STATES

Irwin P. Levin, University of Iowa, Iowa City, U.S.A.

Shu-Fang Kao, University of Iowa, Iowa City, U.S.A.

[The authors would like to thank Arthur H. Miller, Director of the University of Iowa Social Science Institute and Mary Losch, Program Director for the Heartland Poll, 1990, for permission to contribute items to the Poll and to use the complete dataset.]

ABSTRACT -

A telephone survey was conducted in seven Midwestern States in the United States to assess household spending and saving habits. Household tendencies to save or invest, use credit cards, and pay bills in full were related to sociodemographic factors such as age, education, marital status, employment status, and rural vs urban living environment, and to perceptions of personal and national economic well-being. Results were related to differing economic attitudes among different demographic groups and it was suggested that the predictor variables employed here may be useful for explaining regional and national differences.

The study reported in this paper has two primary goals: to analyze factors that underlie individual and regional differences in spending and saving and to provide baseline data on spending and saving habits of persons in a particular geographic area for possible comparisons with other areas.

The need to understand the influence of individual characteristics and cultural factors on the relative value of spending vs. saving has been emphasized by researchers studying "the psychology of money," including preferences for immediate vs. delayed gratification and attitudes towards being in debt. Maital and his associates have stressed the importance of time preferences on economic choice by arguing that the question of whether to consume now, or save and invest in order to consume more later, is fundamental to individuals, families, corporations, and entire nations (Maital, Maital, and Pollak 1986). Factors such as education and health status are thought to differentiate between an emphasis on current as opposed to future gratification.

Katona and his associates suggest that attitudinal differences underlie spending and saving habits and that different cultures may foster different attitudes towards being in debt and saving (Katona 1975; Katona, Strumpel, and Zahn 1971). VanVeldhoven (1985) and Poiesz and Levin (1986) suggest that individual behavioral differences and cultural differences in the psychology of money may exist. For example, Poiesz and Levin conducted a survey in which American college students were more apt than Dutch students to associate money and wealth with personal factors such as status while Dutch students were more apt than American students to associate money with global issues such as injustice and world problems. The present study will help provide a conceptual and methodological foundation for further research in these areas.

Data for this study were provided by a large telephone survey of adult residents in seven Midwestern states. Questions included perceived family economic status, backward comparisons and forward projections of economic well-being, use of credit cards, and saving and investment habits. The inclusion of judgments of the future and perceptions of economic status is representative of the increased use of psychological measures in studies of economic behavior (e.g., Warneryd 1989). In fact, some of our measures of consumer sentiment are similar to those shown by van Raaij and Gianotten (1990) to improve the explanatory power of models of consumer spending and saving.

In this paper we will provide frequency distributions and descriptive statistics for these measures, and the results of regression analyses relating the measures to demographic characteristics such as age, gender, race, education, occupation, and income. In addition, we will show how spending and saving habits differ as a function of household characteristics (e.g., married couples vs. non-married partners, whether or not spouse working) and geographic unit (urban vs. rural). We will develop a conceptual framework for understanding the observed relationships between socio-demographic factors, economic attitudes, and spending and saving habits. We conclude by discussing how the methods used in this study may be applied to the analyses of cross-cultural differences in spending and savings habits.

METHOD

The survey, known as the "Heartland Poll, 1990" was conducted by the University of Iowa Social Science Institute. Approximately 300 individuals from each of the following midwestern states were interviewed by telephone during the period of October 8, 1990 - January 30, 1991: Illinois, Iowa, Minnesota, Missouri, Nebraska, South Dakota and Wisconsin. A stratified random sampling procedure was used where samples were stratified to all counties in proportion to each county's share of telephone households in the survey area. The resulting data were then weighted according to 1990 census figures to adjust for differences in population across the seven states.

The Poll contains about 250 questions supplied by different researchers, dealing with economic, political, and social service issues. The total sample of 2127 interviews represents a 42% response rate. The mean interview length was 23 minutes.

There were two forms of the questionnaire. The questions of main interest in the present study were thus answered by about half the respondents. The following is a listing of the key questions:

1. Personal financial situation, retrospective: "We would like to know how people are getting along financially these days. Would you say that you (and your family living there) are better off, about the same, or worse off financially than you were a year ago?"

2. Personal financial situation, prospective: "Now looking ahead, do you think that a year from now you (and your family living there) will be better off, about the same, or worse off financially than you are now?" (Some respondents added the category "it depends".)

3. National economy, retrospective: "Now thinking about the country as a whole, would you say that over the past year, the nation's economy has gotten better, stayed the same, or gotten worse?"

4. National economy, prospective: "Would you say that during the next year, the nation's economy will get better, stay the same, or get worse?" (Some respondents said "it depends".)

5. Family economic status: "Please rate your family's current degree of economic well-being on a scale of 1 to 100 where 1 = worst possible and 100 = best possible."

6. Savings or investments: "Does your household routinely put money into a savings account or an investment program?"

7. Number of credit cards used: "How many credit cards do you regularly use?"

8. Monthly balance: "On a typical month, do you pay the balance in full or do you carry the charges over to the next month?"

Questions 5-8 were used as the primary dependent variables for this study. Questions 1-4 were treated as independent variables, along with demographic variables such as age, gender, race, marital status, children, occupation, income, whether spouse is employed and where they live (urban vs. rural).

TABLE 1

FREQUENCY DISTRIBUTION OF KEY ITEMS

RESULTS AND DISCUSSION

Table 1 gives the distribution of responses on the key items. It can be seen that the modal response is "the same" when respondents compared their personal financial situation now with a year ago and when predicting what will happen in the coming year. Predictions for the coming year were somewhat more optimistic than retrospective comparisons. The same is true for judgments of the national economy, but here the modal response is "worse." A full three-quarters of the respondents felt that the country as a whole is worse off now than a year ago and over half feel that this tendency will continue for the coming year.

The distribution of ratings of family economic status tended to be bimodal with one mode in the 41-50 category and another in the 71-80 category. The mean rating of 60 is about one-half standard deviation above the midpoint of the scale.

About 70% of the respondents indicated that their households routinely put money into savings or investments, and about 60% said that they typically pay their bills in full rather than carrying charges over. About one-third of the respondents did not use any credit cards on a regular basis, but the number of credit cards used ranged up to 20, with a median of 1 and a mean of 1.66. [These numbers are lower than those sometimes reported for "number of credit cards owned", but recall that we asked, "how many credit cards do you regularly use?".]

Table 2 shows the pairwise interrelationships between the four dependent variables: family economic status, savings or investments, # of credit cards used, and monthly balance. Most pairwise relationships were statistically significant and, as indicated in the table, are in the logical direction. For example, rated family economic status is directly related to the number of credit cards used and to the tendency to save or invest and to pay bills in full. Families who save or invest are apt to use more credit cards but to pay in full. The relationship between # of credit cards used and the tendency to pay in full was not statistically significant.

TABLE 2

RELATIONSHIP BETWEEN DEPENDENT VARIABLES

Table 3 gives the direction of effect on each of the dependent variables of a variety of socio-economic and demographic measures obtained in the survey, including differences between the seven states in the survey. In addition, the items concerning retrospective and prospective ratings of personal financial situation and national economic condition were related to the dependent measures. The direction of each of these relationships is shown in Table 4.

Step-wise regression analyses were conducted on the four dependent variables, using all the factors listed in Tables 3 and 4 as independent variables, as well as "states" (differences between the seven states in the region sampled). These analyses are summarized in Table 5, which lists the order of importance of all the statistically significant variables. Some of the relationships noted in Tables 3 and 4 lose statistical significance when primary variables such as income are factored out. The R² values for these analyses are only in the range of .2-.3, probably because the economic behaviors recorded here are subject to unmeasured situational factors such as health care needs, temporary household needs, and changes in occupational status. Table 6 summarizes the results (excluding "states") by indicating the direction of the significant relationships and by organizing factors into logical clusters.

Many of the results in Tables 3-6 are logical and intuitive. For example, it is not surprising that income is a dominant factor and that higher income households rate themselves higher on family economic status, are more apt to save or invest, use more credit cards, and are more apt to pay in full, and that those who live in houses score higher on each measure than those who live in apartments. Persons in professional occupations score higher than persons in other occupations.

Other results are more surprising. Respondents with part-time jobs gave higher ratings of family economic status than respondents with full-time jobs C possibly because respondents with part-time jobs were significantly more apt than respondents with full-time jobs to have spouses who work full time -but housewives also gave high ratings. Respondents who were married were more apt to indicate that they routinely saved or invested, as did those with children, and those with working spouses. Also, those who think they are better off now than last year and those who think they will be better off a year from now are more apt to save or invest, but are not more apt to pay bills in full.

TABLE 3

EFFECTS OF SOCIO-ECONOMIC AND DEMOGRAPHIC VARIABLES

Those who think they are better off now than a year ago are also apt to use more credit cards, but, notably, those who think the national economy has gotten worse are apt to use more cards. Females used more credit cards than males, and those respondents who were married and those who had children were also apt to use more cards. Males were more apt than females to pay their bills in full, older respondents more apt than younger, those with children less apt than those without children, part-time workers more apt than fulltime workers, and highly educated more apt than less educated. Interestingly, those who think they will be better off financially a year from now are less apt to pay in full.

The effect of type of living environment was statistically significant in the regression analyses for three out of the four dependent variables. The pattern here is that residents of suburban and rural areas and small cities rate themselves higher on family economic status and are more apt to save or invest than residents of big cities or small towns. However, big city residents used more credit cards than those in small cities or rural areas. Suburbanites used the highest number of credit cards. These results were related to observed differences between the seven states in the survey region. For example, the State of Illinois was at or near the top on family economic status, % who save, and number of credit cards used, but was at the bottom on % who pay in full. However, when measures such as income and rural/urban environment were factored out (Illinois has a large suburban population), State was significant in only two of the regression analyses (Table 5). Its highest importance ranking was on monthly balance, a measure not affected by rural/urban distinctions. This suggests that different distributions of units of varying population density may account for some regional differences in saving and spending habits.

TABLE 4

EFFECT OF ECONOMIC ATTITUDES

TABLE 5

RANK-ORDER OF FACTORS IN STEPWISE REGRESSION ANALYSES

TABLE 6

SUMMARY OF RESULTS

CONCLUSIONS AND INTERPRETATIONS

Saving and spending habits, as reflected in the dependent measures selected for this study, were impacted by both demographic factors and economic attitudes. The R² values for the regression analyses in Table 5 were increased by almost 100% when the attitudinal variables were added, compared to when only the socio-demographic were included. Figure 1 represents our conceptualization of the important relationships observed in this study. Because we want to focus on attitudes towards saving and spending, "perceived family economic status" is treated as an antecedent socio-demographic variable in this conceptualization.

In addition to the observed antecendents of our measures of saving and spending, we postulate several higher-order constructs to help organize and explain our data. In keeping with the model of van Raaij and Gianotten (1990), we believe that both ability and willingness variables contribute to consumer saving and credit. "Ability to pay, save and invest" mediates the linkage between various socio-demographic factors (family income, family economic status, and living in a house vs. apartment) and the dependent variables (save or invest, number of credit cards used, and whether or not the monthly credit balance is paid in full). Other factors appear to relate to saving and investment habits not through ability to save but to an attitude favoring saving. Those who are more likely to possess this attitude are people who are married, those who have children, those whose spouse is employed, those with a full-time job, and those who live in rural or suburban areas rather than in cities. Those who believe either that their own personal financial situation has improved over the past year or that it will improve in the next year are also more apt to possess a "savings attitude."

FIGURE 1

THE CONCEPTUALIZATION OF THE IMPORTANT RELATIONSHIPS OBSERVED IN THIS STUDY

Perceptions of the national economy played a lesser role than perceptions of one's personal financial situation. However, those who thought the national economy got worse (the majority of respondents) used more credit cards than those who thought it stayed the same or got better during the past year. This may represent a reaction to the uncertainty of future gratification while things are getting worse.

Somewhat counter to the "savings attitude" is the tendency to defer credit and payments and thus pay interest. Interestingly, some of the same groups who are most apt to save or invest are also apt to defer payments. These include families with children and families with both spouses working, as well as those who think they will be better off financially a year from now. Apparently the latter group feels that they will be in a better position to pay their bills in the future. The other two groups may feel a conflict between the desire to save for the future and the ability to meet current needs. Future research might focus on the economic trade-offs faced by such groups. The fact that younger respondents are more apt to defer payments probably reflects a growing trend to view being in debt as "socially acceptable."

It would be interesting to determine if the same pattern of interrelationships among measures of saving and spending and sociodemographic factors holds for other regions and other countries. Of particular interest would be differences in "national attitude" toward saving and spending that remain even after these sociodemographic variables are factored out. Our analysis comparing the seven states in the survey region suggests that many regional differences can be accounted for by traditional sociodemographic variables but that some differences remain. These remaining differences may represent the "uniqueness" of attitudes of persons in a particular population group. Comparisons of different countries using the present methods are particularly apt to yield differences that reflect prevailing attitudes towards spending and saving. Following van Raaij and Gianotten (1990) and Warneryd (1989), we suggest that cross-cultural comparisons include attitudinal measures such as judgments of future economic well-being for both the individual and the nation.

REFERENCES

Katona, George (1975). Psychological economics. New York: Elsevier.

Katona, George, B. Strumpel, and E. Zahn (1971). Aspirations and affluence. New York: McGraw-Hill.

Maital, Shlomo, Sharone Maital, and Nava Pollak (1986). Economic behavior and social learning. In Proceedings of the International Conference on Economics and Psychology: "Choice and Exchange", Ed. Shlomo Maital. Haifa, Israel: The Society for the Advancement of Behavioral Economics and the International Association for Research in Economic Psychology.

Poiesz, Theo, and Irwin P. Levin (1986). Money: A comparison of the perceptions and attitudes of Dutch and American university students. Paper presented at the Annual Colloquium of the International Association for Research in Economic Psychology, Haifa, Israel.

van Raaij, W. Fred, and Henk Gianotten (1990). Consumer confidence, expenditure, saving, and credit. Journal of Economic Psychology, 11, 269290.

van Veldhoven, Gery (1985). Economic-psychological aspects of money: An overview. In: H. BrandstStter, and E. Kirchler (Eds.), Economic Psychology. Linz, Austria: Rudolf Trauner.

WSrneryd, Karl-Erik (1989). On the psychology of saving: An essay on economic behavior. Journal of Economic Psychology, 10, 515-541.

----------------------------------------