A Developmental Study of Family Financial Management Practices

Donald H. Granbois, Indiana University
Dennis L. Rosen, University of Kansas
Franklin Acito, Indiana University
ABSTRACT - Family financial management and control practices have received little attention. Yet they are increasingly important, due to changes in the financial environment such as deregulation of the banking industry and the increasing number of dual income families. This paper reports on a pilot study relating demographic variables, measures of sex role, and locus of control to five variables thought to be indicators of the family's financial management and control tendencies. An overall financial management scale is constructed and its relationship to the independent variables is discussed.
[ to cite ]:
Donald H. Granbois, Dennis L. Rosen, and Franklin Acito (1986) ,"A Developmental Study of Family Financial Management Practices", in NA - Advances in Consumer Research Volume 13, eds. Richard J. Lutz, Provo, UT : Association for Consumer Research, Pages: 170-174.

Advances in Consumer Research Volume 13, 1986      Pages 170-174

A DEVELOPMENTAL STUDY OF FAMILY FINANCIAL MANAGEMENT PRACTICES

Donald H. Granbois, Indiana University

Dennis L. Rosen, University of Kansas

Franklin Acito, Indiana University

ABSTRACT -

Family financial management and control practices have received little attention. Yet they are increasingly important, due to changes in the financial environment such as deregulation of the banking industry and the increasing number of dual income families. This paper reports on a pilot study relating demographic variables, measures of sex role, and locus of control to five variables thought to be indicators of the family's financial management and control tendencies. An overall financial management scale is constructed and its relationship to the independent variables is discussed.

INTRODUCTION

Although seldom explicitly acknowledged in consumer research studies, normative assumptions about consumers' pre-purchase behaviors are strongly emphasized in the consumer economics and family management literatures (Deacon and Firebaugh 1975; Maynes 1976). Writers in these disciplines tend not to conduct empirical studies of how consumers actually do behave. However, research in marketing has provided data on numerous measures of what one researcher has labeled "good consumership" (Hill 1963). Reviews of these studies document wide variability in pre-purchase behavior, and various interpretations and tentative profiles of "good consumers" have been offered (Granbois 1977; Granbois and Braden 1976; McNeil 1974; Newman 1977; Olshavsky and Granbois 1979).

A second normative theme in family management and consumer economics -- that consumers "should" devise and utilize appropriate financial planning and control procedures regarding spending and saving -- has not been subjected to much empirical testing (Beutler and Sahlberg 1980; Ferber 1973). Ferber (1973) reports few studies and fewer generalizations about how families handle financial tasks and decisions regarding budgets, bill payment, or saving.

The underlying rationale for prescribing specific financial practices seems to be that families engaging in such practices are expected to be more effective in achieving goals and in getting the greatest possible utility from their income. One writer refers to these behaviors as "... family competences which differentiate families who make the most of their resources in getting where they wish to go from families who lag in this respect" (Hill 1963, p. 425). A clear statement of the predicted link between financial management and goal achievement is provided by Deacon and Firebaugh (1975), who note:

The expected outcome of financial management is to meet demands (whether long-term goals, short-term goals, or events) by effective use of the resources available to each managerial unit. A major output is the achieved level of living and the related sense of fulfillment of goals. (p. 287)

Recent changes in the financial environment make this process more important. Consumers today are faced with many more savings options as the line blurs between savings and investments. Taking advantage of these options generally requires better management than was true for the traditional passbook savings account because restrictions or penalties may be applied to deposits and withdrawals. The dual incomes of many families may also require better financial management to take advantage of increased opportunities. Differences among families in the extent and pattern of financial planning and control practices may be useful for defining segments that will respond differently to financial institutions' strategies.

An ultimate theoretical justification for the empirical study of consumers' financial management practices appears to be the need to verify that "good managers" will make more effective use of resources in meeting goals. Presumably, both objective measures of level of living, household net worth, etc. and subjective measures of perceived achievement and satisfaction should be affected by consumers' financial management practices. Before such predictions can be tested, though, it is necessary to identify the nature of consumers' financial management practices, devise measures for determining performance, and determine characteristics and situations associated with their use.

This paper reports on an exploratory study of five aspects of the financial management practices reported by married couples. The topics were suggested by a literature review and by our earlier developmental research involving unstructured depth interviews with married couples. Our findings on husband-wife role structures with respect to these financial management behaviors have been reported elsewhere (Rosen and Granbois 1983). Our emphasis here is on the extent to which our pilot study respondents used the management practices investigated, the testing of tentative hypotheses about family characteristics that may predict which families use the practices, and the development of a composite scale measuring financial management behavior with the family.

DEVELOPMENT OF VARIABLES AND HYPOTHESES

Financial Management Variables

Ferber (1973) divides the domain of family economic decision making into four segments: money management, saving behavior, asset management, and spending behavior We confined our discussion of family financial management to money management and saving behavior because asset management is complex and specialized and perhaps best studied by experts in consumer finance and investments

Five primary measures of family financial management were suggested by our literature review and developmental depth interviews:

Budget: whether a budget is determined ac the beginning of a spending period;

Savings Policy: whether savings are added to on regular (periodic) basis;

Surplus Funds Policy: whether pre-planning occurs for funds left over after the payment of bills and other obligations:

Special Accounts: whether special savings accounts are used to facilitate saving for specific purposes:

Credit Transactions Analysis: whether credit card transactions are sorted to study spending Patterns .

Using each of these strategies would be consistent with a normative view of how family financial affairs might "best" be handled. However, we did not expect all five behaviors to appear with equal frequency across families. We did suspect that the five varied in importance in contributing to effective family financial control, and they were hypothesized to be ranked in importance in the order listed above.

Budgeting was viewed as most important because it includes all aspects of using household funds. While budgets tend to be prepared for each pay period, regular budget preparation has implications for long-run control of spending and saving. Thus, both short-run and long-run consequences of budgeting may be realized. Regular saving is similarly both a short-run and long-run practice, but covers disposition of only that part of the household income that is not spent. Planning for surplus funds and use of special accounts are both subsidiary practices facilitating the larger savings function. Finally, credit card sorting was visualized as an occasional aid used to track expenditures. It was thought that credit transaction sorting could occur in the absence of budgeting. Also, it was felt that planning for the use of surplus funds and having special accounts did not necessarily imply regular saving. Each of the five behaviors was therefore considered to be a possible independent component of a larger syndrome of planning and controlling the disposition of the family's income.

Independent Variables

As measures of independent variables we included several dimensions suggested in the literature on family decision making. Recent researchers and commentators on contemporary families have suggested wife's working status and motives for work, life cycle stage, and husbands' and wives' sex-role attitudes as potentially important variables for explaining differences among families. (See Rosen and Granbois 1983 for a brief review.) Since control is a central element in budgeting behavior, a personality variable, locus of control, (Rotter 1966) appeared to be especially likely to be related to family financial management. Individuals classified as "internals" on the locus-of-control variable tend to believe they are more in control of events that affect their lives while "externals" tend to believe that such events are not in their control.

Demographic variables, including education of both husband and wife, number of years married, and household income have been found in earlier studies to have some relationship to family financial management, although the exact nature of these relationships has not been clearly established (Ferber 1973). In a recent study of the extent to which families engage in three activities specified by a formal financial management system model (recording expenditures, reviewing and evaluating expenditures, and making formal spending plans), formal planning was negatively related to age of household head and life cycle stage and positively associated with number of children and education of household head (Beutler and Sahlberg 1980). We could find no support in the literature for directional hypotheses relating sex-role attitudes or wife's work status and motives to the performance of the five measures of family financial management behavior.

In summary, based on past research and judgment we expected that the five financial management behaviors would be more likely to occur among the more highly educated, those with an internal locus of control, families in the middle life cycle stages, and those with higher incomes. We also explored the relationship of sex-role attitudes and wife's work status and motives to our defined family financial management activities.

METHOD

The city directory in a Midwestern university community identified 82 couples balanced as equally as possible between families with working and nonworking wives and upper and lower income families. (See Rosen and Granbois 1983 for details and rationale.) Usable data were obtained from 76 couples. The interviewer read questions dealing with each of the five behaviors; husband and wife responded jointly. Each respondent separately completed Rotter's (1966) Locus of Control Scale and the Osmond and Martin (1975) Sex-Role Attitude Scale.

Respondents were found to be uniformly distributed across life cycle stages and across income groups, and were almost equally divided on wife's working status. As is often true in college towns, both husbands and wives were more likely to have graduated from college or to have earned an advanced degree than would be the case in other research settings. This heavy emphasis toward better-educated respondents may have tended to overstate somewhat the incidence of financial management practices, although our analysis revealed virtually no evidence of a relationship between level of education and use of these practices.

Sex-role attitude responses for the total sample showed husbands to be significantly more traditional than wives (means of 74.2 and 69.4, respectively; t = 2.82, df = 69, p ' .01), and husbands' and wives' scores to be moderately correlated (R = .63). Wives showed a significantly more external locus of control than their husbands (means of 9.21 and 7.34, respectively; t = 3.40, df = 69, p = .001); locus of control scores were only modestly correlated within families (R = .32).

RESULTS

Individual Financial Management Behaviors

As we expected, the five financial management practices studied were not employed with equal frequency by our respondents. The rank of frequency of use differed slightly from our predicted rank of importance, as shown in Table 1.

TABLE 1

INCIDENCE OF FIVE FINANCIAL MANAGEMENT PRACTICES FOR 76 FAMILIES

A higher proportion of our respondents reported regular savings patterns than has been found in earlier research, which reports a range of 15 to 40 percent (Olander and Seipel 1970). We have no comparable data from earlier studies against which we can compare the level of incidence we found for the remaining four measures.

Intercorrelations among the five measures were very low, as shown in Table 2.

TABLE 2

INTERCORRELATIONS AMONG FIVE FINANCIAL MANAGEMENT PRACTICES

The modest negative correlation between formal budgeting and planning for surplus funds suggests a reasonable pattern. Those who do not use overall budgets nonetheless have prior planning procedures for the part of their income left over after paying bills and obligations; those who plan and budget may find planning for "left overs" unnecessary if an adequate budgeting procedure is devised to start with

We found little evidence that the independent variables included in our study can help predict the incidence of our five financial management practices; one variable appeared to have an effect opposite to that hypothesized. The results of our chi-square analysis of categorical independent variables and t-test analysis of continuous variables can be summarized as follows:

(1) Family life cycle stage was significantly related to the reported use of a written or unwritten budget (x2 = 12.19, df = 5, p ' .05). Those in the earlier stages of the family life cycle made greater use of budgeting.

(2) Income was weakly related to the use of written or unwritten budgets with greater use occurring at lower income levels (x2 = 6.16, df = 3, p = .10).

(3) Conflicting results were obtained for locus of control. Wives in families that budget were more external than those in families that did not budget (t = 2.31, df = 44, p ' .05). However, husbands' in families employing special accounts were significantly more internal than were husbands in families that did not employ special accounts (t = 2.28, df = 60. p < .05).

(4) Wife's sex role was weakly related to planning for surplus funds with planners more traditional than nonplanners (t = 1.77, df = 43, p ' .10). Husbands' sex-role attitude was not related to any of the practices.

(5) Neither wife's working status nor the motivation of working wives appeared to be related to any of the five financial practices.

Composite Scale Development and Relationship to Independent Variables

We felt Justified in assuming that each of the five financial management practices contributed positively to the family's overall financial management behavior. However, since each of the individual families reported different subsets of the five practices, any effort to assess the relative financial management behavior of individual families required that a scale value be associated with each item. Scale values also could be used to confirm the a priori judgment of importance ordering of the five practices that has been previously described. The scale values for each practice were obtained from an ad hoc panel of expert judges. Included were the author of a text on consumer management, professors teaching courses on personal finance and family studies, and others with special expertise in related subjects.

Conjoint analysis was used as an assessment tool. The judges were given a set of profiles of the financial management activities and asked to rank them in order of degree of financial management evident. The budgeting factor was offered at three levels -- no budget, unwritten budget, and written budget. The other four factors were specified at two levels each. This yields 3 x 24 = 48 combinations. An orthogonal array (Addelman 1962) of 25 combinations was used to reduce the set. Several of the combinations were duplicates, so a unique set of 20 profiles was finally developed. The 20 profiles were typed on 4" x 6" cards with the order of the financial management activities systematically varied across sets of carts (that is, each set of 20 cards used a single ordering of the factors, but this order was different for each judge.)

Eight judges evaluated the financial management profiles. The judges were simply asked to rank the profiles (ties permitted) according to the degree of family financial management indicated by each. The profile rankings submitted by the eight judges were analyzed at the group level using LINMAP (Shocker and Srinivasan 1973). The single set of part-worth values that was obtained is shown in Table 3. LINMAP converted the ranked data to 190 paired comparisons for the eight judges. One element in pair is considered higher in management behavior if five or more judges ranked the element higher.

The overall fit of the derived conjoint data was indicated by a Kendall's Tau of .83. Of the 190 paired comparisons, 165 were strictly satisfied, 17 were neither violated nor satisfied (due to ties), and 8 were violated. Overall, these results indicated a good fit of the additive conjoint model to the expert judgments. The weights shown in Table 3 were used to assign a scale value on financial management to each respondent in the study. By determining which activities a family conducted, a suitable score was derived, with higher scores representing higher degrees of financial management behavior.

Also shown in Table 3 are the relative importance of the attributes in contributing to management behavior scores. This importance is a simple function of the range of weights on each attribute. The expert judges' ordering was consistent with our hypotheses.

TABLE 3

ATTRIBUTE WEIGHTS FROM CONJOINT ANALYSIS OF EXPERT JUDGMENTS FOR FAMILY FINANCIAL MANAGEMENT ACTIVITIES

The composite score for a family was the simple sum of the part-worth weighting it received for each of the five activities. The distribution of our respondent families' scores on the composite financial management scale is shown in Figure 1.

Relationships among the couple's composite management scale values and the independent variables were sought through multiple regression analysis. No association was found between the independent variables and composite scale scores.

DISCUSSION

We were reasonably confident that our exploratory interviews and literature review had uncovered relevant and important measures of financial management practices for investigation, and our analysis of results seems to confirm this assessment. All five practices were used by a significant number of respondents, and we sensed no feelings on the part of respondents who did not use these practices that their use was inappropriate or unexpected. Not all practices were used with equal frequency, so the idea of forming a scale with importance weights assigned to each element seemed reasonable. Our expert judges found our measures to be realistic elements of family financial management and confirmed our a priori ranking of their relative importance. Inter-judge agreement was high and standard tests supported strong confidence in the importance weights devised from conjoint analysis.

FIGURE 1

DISTRIBUTION OF COMPOSITE SCALE SCORES

The set of demographic variables and sex-role attitude measures used as independent variables were strongly suggested as important determinants of family behavior by other researchers and commentators on contemporary family structure and behavior. Finally, locus of control seemed to have strong intuitive relevance and had been found useful in scattered earlier consumer research and in decision-making studies in other disciplines.

Given the methods employed, the general lack of significant findings for the relationship between the independent variables and financial management behavior is surprising. Although our sample was small and was confined to a single small city, we selected respondents in a very carefully planned and controlled fashion so that our responding couples were distributed over age, income, family life cycle, and wife work status categories. Husbands' and wives' sex-role attitude and locus of control scores differed from each other in predicted ways, and average scores were well within ranges predicted by earlier administration of the instruments as reported by their developers (Osmond and Martin 1975; Rotter 1966). The only obvious flaw in the distributions of our respondents' characteristics was the over-representation of more highly educated persons.

The small size of our sample may have limited our ability to discover significant associations that might be revealed in a larger study. We plan to include these variables in a more extensive future study. Further, there may be additional family or individual respondent characteristics that determine whether our five practices are implemented. The lack of relevant theory hampers identification of these possible determinants, and we welcome other researchers' suggestions of variables for inclusion in our next study.

We find it interesting that the strongest relationships, though inconsistent, were found for locus of control. One finding was the opposite of that predicted with greater indications of the budgeting behavior when wives had a more external orientation. This finding may indicate that internals have confidence in their ability to control finances without the help of such devices as budgets and special accounts. Externals may see such devices as ways of imposing controls, or protecting their financial status from otherwise uncontrollable external forces and influences. This view is contrary to our hypothesis concerning locus of control, but it is supported by Mueller (1966). who notes:

Budgeting is usually intended to insure that some money is available for major acquisitions or for savings. Various (unsuccessful) attempts to talk to people about their budgeting procedures have led us to the tentative conclusion that determination is a more effective control device than a plan or system developed on paper. Indeed it appears that people who resort to formal budgets may to so because their determination or self-control has not proven sufficient. (p. 33)

The significant findings for locus of control suggests that other personality variables may show a stronger relationship than standard demographics. Further research using personality and lifestyle measures may find that they prove to be more fruitful indicators of financial management behavior.

In addition to theoretical importance, an understanding of the variables most strongly related to family financial management practices should have applied relevance for financial institutions. Deregulation of the banking industry has led these institutions to provide an ever-increasing array of financial planning and management services in an effort to remain competitive in their new environment. Conventional wisdom suggests that demographic variables such as income and education should be related to the use of such financial management and savings options. However, to the extent that the five financial management practices we studied affect the family's interest in or ability to use such services, our findings suggest that conventional wisdom may be incorrect.

Future Research

This study represents a first step in a larger research effort. We hope to investigate appropriate measures of effective consumer performance and satisfaction, which may then be studied as possible outcomes or consequences of applying financial management practices in handling household spending and saving. Our experience with simple satisfaction scales administered for each aspect of respondents' financial practices has been that such scales fail to reveal much variability across respondents or across the various financial practices. Further developmental work in this area is clearly needed.

lie intend to explore the possibility that income variability and frequency may be determinants of our budget and savings regularity variables. This follows the tentative suggestion by Deacon and Firebaugh (1975, p. 282 p. 282) that both budgeting and saving may be more important when household income is unstable and received less frequently. A simple measure of income stability included in the present study failed to reveal much variability across responding families, thus indicating the need for additional development of an adequate measure.

Finally, we intend to investigate role structure as determinant of financial management and control behavior. The literature does not suggest hypotheses in this area, but it seems plausible that formal budgets and savings plans will be more likely if husband and wife tend to exhibit a syncratic (joint) decision-making style.

REFERENCES

Addelman, Sidney (1962), "Orthogonal Main-Effect Plans for Asymmetrical Factoral Experiments," Technometrics, 4 (February), 21-46.

Beutler, Ivan F. and Kristine M. Sahlberg, (1980), "Spending Plans and the Family Managerial Decision Process," in Advances in Consumer Research, Vol. VIII, Ann Arbor: Association for Consumer Research, 480-85.

Deacon, Ruth E. and Francille M. Firebaugh (1975), Home Management: Context and Concepts, Boston: Houghton-Mifflin.

Ferber, Robert (1973), "Family Decision-Making and Economic Behavior: A Review," in Family Economic Behavior: Problems and Prospects, ed. Eleanor B. Sheldon, Philadelphia: J. B. Lippincott, 25-61.

Granbois, Donald H. (1977), "Consumer Shopping Behavior and Preferences," in A Synthesis of Selected Aspects of Consumer Behavior, ed. Robert Ferber, Washington: U. S. Government Printing Office for the National Science Foundation, 259-98.

Granbois, Donald H. and Patricia L. Braden (1976), "'Good Consumership' in Household Appliance Purchasing," Journal of Business Research, 4 (May), 103-16.

Hill, Reuben (1963), "Judgment and Consumership in the Management of Family Resources," Sociology and Social Research, 47 (July), 446-60.

Maynes, E. Scott (1976), Decision-Making for Consumers, New York: Macmillan

McNeil, John (1974), "Federal Programs to Measure Consumer Purchase Expectations 1946-1973: A Post-Mortem," Journal of Consumer Research, 1 (December). 1-10.

Mueller, Eva (1966), "A Look at the American Consumer," in On Knowing the Consumer, ed. Joseph W. Newman, New York, Wiley, 23-37.

Newman, Joseph (1977), "Consumer External Search: Amount and Determinants," in Consumer and Industrial Buying Behavior, eds. Arch G. Woodside, Jagdish N. Sheth and Peter D. Bennett, Amsterdam, Holland: North Holland Publishing Company.

Olander, Folke and Carl-Magnus Seipel (1970), PsYchological Approaches to the Study of Saving, Urbana: Bureau of Economics and Business Research, College of Commerce and Business Administration, University of Illinois.

Olshavsky, Richard W. and Donald H. Granbois (1979), "Consumer Decision-Making: Fact or Fiction?" Journal of Consumer Research, 6 (September), 93-100.

Osmond, Marie U. and Patricia Y. Martin (1975), "Sex and Sexism: A Comparison of Male and Female Sex-Role Attitudes," Journal of Marriage and the Family, 37 (November), 744-58.

Rosen, Dennis L. and Donald H. Granbois (1983), "Determinants of Role Structure in Family Financial Management," Journal of Consumer Research, 10 (September), 253-58.

Rotter, Julian B. (1966), "Generalized Expectations for Internal Versus External Control of Reinforcement," Psychological Monographs, (80), whole issue.

Srinivasan, V. and Allan D. Shocker (1973), "Linear Programming Techniques for Multi-Dimensional Analysis of Preferences," Psychometrika, 38 (September), 337-69.

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