COMPREHENSIVE PSYCHOLOGY

The etiology of frugal spending: a partial replication and extension1

2015, Volume 4, Article 4 ISSN 2165-2228

Ronald E. Goldsmith Florida State University

DOI: 10.2466/09.20.CP.4.4 © R. E. Goldsmith 2015 Attribution-NonCommercialNoDerivs CC-BY-NC-ND

Leisa R. Flynn University of Southern Mississippi

Received November 18, 2014 Accepted March 17, 2015 Published April 7, 2015

CITATION Goldsmith, R. E., & Flynn, L. R. (2015) The etiology of frugal spending: a partial replication and extension. Comprehensive Psychology, 4, 4.

Abstract Frugal behavior is infrequently studied in both psychology and marketing. As a consumption pattern it deserves greater attention because frugal consumer behavior is a desirable goal for both individuals and societies. Paradoxically, targeting frugal consumers might also be a profitable strategy for savvy marketers. The present study builds on previous research into the nature and motivations for frugal behavior. Data analysis from a survey of 464 U.S. student respondents shows that three individual-difference characteristics are associated with frugality. Consistent with a prior study, frugality was negatively correlated with materialism and positively correlated with consumer independence. In addition, frugality was positively correlated with self-control, suggesting that frugal consumers tend to be less materialistic, more independent, and have more selfcontrol than are their less frugal counterparts.

Frugality should be one of the most relevant and pervasive individual differences that influence consumer spending and saving. The Oxford English Dictionary defines it as “the quality of being frugal; moderate or sparing expenditure or use of provisions, goods, etc.” Positive synonyms include such laudable terms as “thriftiness, carefulness, and prudence,” while a negative sense is conveyed by “miserliness, meanness, and stinginess.” Consumer psychologists have occasionally studied frugality since the seminal paper by Lastovicka, Bettencourt, Hughner, and Kuntze (1999, p. 88) defined it as “a unidimensional consumer lifestyle trait characterized by the degree to which consumers are both restrained in acquiring and resourceful users of economic goods and services to achieve longer-term goals.” They also provided a seven-item self-report scale (F-scale) to operationalize the construct, thus enabling its empirical investigation. Despite its seeming importance, frugality is little studied. Although this lack of interest is understandable during good economic times, recent years find those concerned with the financial well-being of consumers and with promoting sustainable life-styles being drawn to the notion that increased individual frugality benefits both individuals and society as a whole (Lewis & Potter, 2011). Thus, scholarly interest in the topic seems to be increasing. The few studies since 1999 provide some insights into the motivations and consequences of frugality, but most studies concentrate on frugality's impact on spending, debt, credit card use, and saving. Fewer studies investigate motivations that lie behind frugality. It would seem that the negative consequences of not being frugal and the benefits of increased frugality are well established, but less is known about the psychology of frugality. Our paper was motivated by the lack of information about individual differences seemingly related to frugality. Thus, the purpose of the present study is to replicate some findings of a previous study (Goldsmith, Flynn, & Clark, 2014) of the etiology of frugality and to add additional insight into the possible intraindividual motivations for it.

Conceptual Background and Hypotheses A large literature concerning frugality exists outside the discipline of psychology. Most of this, however, is in the field of religion, where frugality is praised as a desirable mode

Ammons Scientific

www.AmmonsScientific.com

Address correspondence to Dr. Ronald E. Goldsmith, College of Business, Florida State University, 821 Academic Way, Tallahassee, FL 32306-1110 or e-mail ([email protected]). 1

Etiology of Frugal Spending / R. E. Goldsmith & L. R. Flynn

surveyed 3710 New Zealanders and used the data from Schwartz's (1994) value inventory to show that low scores on the F-scale are related to the values of social power, authority, public image, pleasure, enjoying life, exciting life, varied life, detachment, social recognition, and true friendship. High frugality was related to choosing one's own goals, ambition, capability, honesty, politeness, obedience, responsibility, social order, national security, spiritual life, being devout, unity with nature, protecting the environment, and inner harmony. They concluded that overall low frugality seemed to be associated with values reflecting individualistic tendencies and high frugality with collectivist tendencies. Although these findings are at odds with Lastovicka, et al.’s (1999), Todd and Lawson's main conclusion supported Lastovicka, et al.’s concept of frugality as a lifestyle rather than a value because of a “lack of a clear association between frugality and any one or two value items” (p. 17). The general tendencies in their data seem to reflect the notion that: “Frugality involves restraints and conservation in use which is clearly at odds with values relating to power, stimulation and hedonism” (p. 15). Shoham and Brencic (2004) gathered data from 130 Israeli consumers using the F-scale and found that scores on the F-scale were positively correlated with price consciousness (0.41) and value consciousness (0.43), as well as with a self-report of frugal behavior (0.31; measured as “constrained consumption,” meaning actual frugal behaviors such as reusing and repurposing). This study replicated Lastovicka, et al.’s (1999) original findings, but again, the emphasis seems to be on the consequences of frugality instead of its determinants. Bove, Nagpal, and Dorsett (2009) directly addressed the issue of what determines frugality in their study of 215 Australian shoppers. Their zero-order correlations show a small correlation of 0.16 between F-scale scores and a measure of intrinsic religiosity. Their SEM analysis found scores on the F-scale were positively related to market mavenism (B = 0.37) and shopping antipathy (B = 0.18), but not related to recreational shopping or intrinsic religiosity. It can be argued, however, that market mavenism and shopping antipathy could just as well be interpreted as consequences of frugality as its antecedents. Intrinsic religiosity would seem to be the best candidate for explaining the motivational roots of frugality, as are the basic value orientations noted by Todd and Lawson (2003), because these can be conceptualized as more fundamental or higher level elements of personality and character. The marketplace characteristics of seeking value, low prices, market mavenism, disliking shopping, etc., would seem to be more akin to domain-specific behavioral patterns or outcomes rather than determinants of frugality. A recent study by Goldsmith, et al. (2014) sought to partially resolve the conceptual confusion characterizing the research on frugality and to provide evidence

of life; in the self-help and popular “stinginess” literature (e.g., Nellis & Nellis, 2010) where promises are made that one's life will improve with the adoption of a more frugal lifestyle or advice is given on how to save money, use coupons, get deals, etc.; in popular accounts of living the frugal life (e.g., Yeager, 2013); and in the cultural studies area, which has a tradition of criticizing consumer culture and encouraging an anti-consumer lifestyle (e.g., Binkley & Littler, 2011). Many academic studies document the negative consequences of not being frugal: overspending, creditcard debt, and low subjective well-being. Many studies find ties between high materialism and bad outcomes. For example, Vansteenkiste, Duries, Simons, and Soenens (2006) show that students higher in extrinsic orientation (a higher score on Kasser's 2002 materialism scale) have lower psychological well-being. There is less research on the outcomes of frugality. Corral-Verdugo, Mireles-Acosta, Tapia-Fonllem, and Fraijo-Sing (2011) report a cluster of frugal and sustainable behaviors linked to happiness. Albinsson, Wolf, and Kopf (2010) show that frugal behaviors are culturally reinforced as good and reflect “anti-hyperconsumption” in reunified former East Germany. One potential negative outcome to frugal behavior may occur when the frugal lifestyle is adopted by a large segment of the population. Too much frugality can lead to depression of demand for consumer goods and thus keep a recessionary economy out of recovery (Landes, 2011). While too little frugality has serious negative consequences for the environment, frugal behavior is linked to practicing sustainable behavior (Parkins & Craig, 2011). Few studies, however, are devoted to uncovering the factors that lead to frugality. Lastovicka, et al. (1999; U.S. students and consumers) used qualitative evidence to conclude that frugal consumers see themselves as (1) disciplined in their spending of money, (2) less impulsive in buying, (3) emphasize long-term rather than short-term gratification, (4) are resourceful in using and reusing possessions, (5) feel more independent than average, and (6) are likely to be less susceptible to interpersonal influence than other consumers are. Quantitative evidence using their frugality scale showed a negative correlation (−0.25) with compulsive buying and positive correlations with value consciousness (0.54) and with price consciousness (0.45). Note that compulsive buying is a consequence of not being frugal and value and price consciousness describe marketplace behavior, so that they might be seen as consequences as well. Lastovicka, et al. also found that frugality was negatively related both to materialism and to susceptibility to interpersonal influence, both of which are candidates for frugality's antecedents. Todd and Lawson (2003) focused on whether frugality is best conceptualized as a value orientation or as a lifestyle, as Lastovicka, et al. (1999) propose. They Comprehensive Psychology

2

2015, Volume 4, Article 4

Etiology of Frugal Spending / R. E. Goldsmith & L. R. Flynn

pirically between “frugal behavior” and a “frugal lifestyle.” To this end, we conducted this study to replicate and extend Goldsmith, et al.’s paper by testing three hypotheses. Hypothesis one, based on theoretical arguments and prior empirical findings, is that materialism is a determinant of the frugal lifestyle operationalized by the Fscale. The very description of materialism as an avid desire to own material goods seems logically opposed to living a frugal lifestyle, and both Lastovicka, et al. (1999) and Goldsmith, et al. (2014) report negative correlations between materialism and frugality (−0.26 and −0.34). Hypothesis one is a direct replication of these earlier findings. Hypothesis two proposes that consumer independence is a positive influence on frugality. The independent consumer is little influenced by the opinions of other people and general social influence, preferring instead to make his or her own decisions (Clark, 2006). In a consumer-oriented society such as the U.S., independence frees individuals from the pervading materialistic, shopping, buying atmosphere and allows them to make consumption decisions on their own. Lastovicka, et al. (1999) argued that frugal consumers felt independent based on their qualitative evidence, and Goldsmith, et al. (2014) report a positive correlation of 0.34 between frugality and consumer independence. Hypothesis two is a direct replication of this previous finding. Our third hypothesis is an extension of prior research as we seek to expand the list of potential antecedents of the frugal lifestyle. As noted above, Lastovicka, et al.’s (1999) qualitative research suggested that frugal consumers see themselves as disciplined. The concept of internally motivated frugality itself implies a consumer who disciplines his or her buying behavior in order to get the best value and to not waste money. Externally compelled consumers have little choice but to limit their spending, but the internally motivated frugal consumer, striving to live a frugal lifestyle, must practice selfcontrol in order to save rather than spend. As noted by Bove, et al. (2009), frugality is a characteristic of wealthy Americans who live modestly, save rather than spend, and practice self-discipline rather than conspicuous consumption. Moreover, Williams and Grisham (2012) show that compulsive buying is characterized by impulsivity, the antithesis of self-control. Thus, hypothesis three proposes that self-control is positively related to frugality.

from U.S. consumers to test some of the relationships found in the prior studies. They surveyed 256 undergraduate students and found that F-scale scores negatively correlated with measures of materialism, status consumption, brand engagement in self-concept (how much consumers use brands to form and express their concepts of themselves), and positively correlated with consumer independence. The relationship with materialism replicated Lastovicka, et al. (1999), but the relationships with status consumption brand engagement again seem more like consequences than antecedents of frugality. The positive correlation with consumer independence can be interpreted to represent an antecedent condition in which consumers who strive to consume independently of interpersonal influence also are more frugal in their buying. Moreover, Goldsmith, et al. (2014) make two major conceptual proposals. They first propose that there are two principal drivers of frugal behavior: external and internal. Some people practice frugal behaviors such as limiting consumption, seeking value and low prices, and the four R's (reuse, repair, repurpose, and recycle) because they have little choice owing to general economic conditions and to their personal economic circumstances. They are not really frugal by choice, but are constrained by external circumstances. Internally motivated frugal behavior is more like a personality or character trait; it is the desire to live a frugal life. It is frugality by choice. Both external and internal frugality lead to frugal behavior in the marketplace: value seeking, price sensitivity, bargain hunting, and the 4Rs. In addition, Goldsmith, et al. (2014) propose that frugal behavior is reinforced or discouraged by an individual's social and cultural environment, so that frugal behavior results from a combination of external economic forces, the surrounding social and cultural environment, and an individual's internal values and traits. Goldsmith et al.’s second major proposal is that a distinction should be made between frugal behavior on the one hand (save rather than spend, shop only sales, eat at home, etc.) and a frugal lifestyle concept comprised of activities, interests, and opinions. That is, their proposal is that Lastovicka, et al.’s (1999) concept and measure is more akin to a lifestyle concept, as those authors intended, but that this lifestyle, which is what the F-scale measures, is both a “cause” or determinant of the constellation of frugal behaviors and can result from practicing frugal behavior and adjusting one's lifestyle because one finds them to be pleasurable and rewarding. Note that popular accounts of frugal living abound with this message. This frugal lifestyle, Goldsmith, et al. (2014) argue, is primarily the product of “internal psychological factors” such as values (materialism) and traits (status seeking, independence). Thus, the cloudy pattern of empirical research can be brought more sharply into focus by carefully distinguishing both conceptually and emComprehensive Psychology

Method Questionnaire and Survey We operationalized the constructs using previously published multi-item scales. A five-point Agree/Disagree response format was used for each scale. The source and number of items in each scale are noted in Table 1. We submitted our questionnaire to the institu-

3

2015, Volume 4, Article 4

Etiology of Frugal Spending / R. E. Goldsmith & L. R. Flynn TABLE 1 The F-Scale (Lastovicka, et al., 1999)

age and income were positively correlated (r = .32, Taub = 0.17) as might be expected.

I am willing to wait on a purchase I want so that I can save money.

Analysis and Results

There are things I resist buying today so I can save for tomorrow. I discipline myself to get the most from my money.

Preliminary Analyses

I believe in being careful in how I spend my money.

For each scale, the items were summed so that higher scores reflected a positive interpretation of the concept to form a total scale score for the analyses, with one exception. An exploratory factor analysis of the seven-item frugality scale showed that it consisted of two factors, not one as Lastovicka, et al. (1999) claimed. The two factor finding is consistent with Goldsmith, et al.’s (2014). We used only the four-item subscale that referred to careful money management because this content best fit with the theoretical substance of the present paper, frugal spending. The other subscale represents care of and reuse of goods, which we deemed unnecessary for the present study's goal. The other three scales were scored and summed as prescribed by their developers. Internal consistency was assessed by coefficient α and was acceptable for all the scales (see Table 2). The descriptive statistics for the summed scales also appear in Table 2. Skewness was less than 1.0 for all the summed scales so no transformations were deemed necessary (Osborne, 2013, p. 94). The frugality scale did have negative skewness (−0.947), and the size of the mean and median (16) compared to the range (4 – 20) showed that the sample considered itself to be more frugal than not. This result might be expected because the sample consisted of students with limited budgets. To give confidence in the findings, summed frugality scores were correlated with two criterion measures on the presumption that a self-report measure of frugality should be negatively related to measures of buying behavior. The questionnaire contained a five-item scale measuring compulsive buying derived from Edwards (1993). The items formed a unidimensional scale with coefficient α of 0.86. In addition, a single item asked “In an average month, how much money would you estimate you spend online or in stores for things other than groceries?” The nine-point response scale ranged from

Making better use of my resources makes me feel good. There are many things that are normally thrown away that are still quite useful. If you take good care of your possessions, you will definitely save money in the long run.

tional review boards of both universities. After it was approved, we loaded it into the online survey system, Qualtrics, for administration. The online questionnaire program was set to randomize the order in which each scale appeared to each participant, and the individual items forming each scale were likewise randomized for each participant. E-mail invitations were sent to undergraduate business students at two state universities in the Southeast U.S.

Sample Qualtrics returned five hundred nine completed questionnaires, but three items worded “if you read this item, do not respond to it,” used to detect blind checking or random responding (Osborne & Blanchard, 2011), detected 45 potentially suspicious cases and so the effective sample size used was 464, although selective missing data in a few cases reduced some individual analyses. The sample was disproportionally female (259 or 56%). Ages ranged from 18 to 55 with a definite positive skewness of 6.3, reflecting the student composition. The mean age was thus 21 years (SD = 3.8 years). The distribution of income measured on a nine-point scale (footnote, Table 2) also reflected the student demographic; 85% reported an income of below US$20,000. The ethnic mix reflected the distribution of these student populations: 86% white. Table 2 shows that the demographic variables were independent, with the exception that

TABLE 2 Descriptive Statistics and Correlations #items

M

SD

1

1. Frugality1

Variable

4

16.0

2.7

(.81)

2. Sex

1

3. Age

1

21

3.8

–.07

4. Income2

1

5. Materialism3

9

29.6

5 13

6. Independence

4

7. Self-control5

2

3

4

5

(.80)

6

7

.04 .04

–.05

.02

.32†

5.6

–.21†

.05

–.10*

–.01

19.7

2.9

.14†

–.03

–.01

.07

–.17†

(.81)

39.3

7.9

.33†

–.07

.06

.08

−0.34†

.18†

(.84)

Notes n = 464; coefficient α on the diagonal in parentheses; where 0 = female and 1 = male. Some Pearson coefficients are based on fewer respondents than the total owing to missing values. 1Lastovicka, et al. (1999). 2where 1 = below $20K, 2 = $20K – $29K, 3 = $30K – $39K, 4 = $40K – $49K, 5 = $50K – $59K, 6 = $60K – $69K, 7 = $70K – $79K, 8 = $80K – $89K, 9 = $90K or more. 3Richins (2004). 4Clark (2006). 5Tangney, Baumeister, and Boone (2004). *p < .05. †p < .01.

Comprehensive Psychology

4

2015, Volume 4, Article 4

Etiology of Frugal Spending / R. E. Goldsmith & L. R. Flynn

accounting for the variance in the dependent variable (Johnson & LeBreton, 2004). One measure of importance, the squared semipartial correlation (see Table 3) is the proportion of the total variance in Y that is uniquely attributable to Xi, i.e., the increase in R2 when Xi is added to the model already containing the variables in the covariate set (Hayes, 2013, p. 74). In addition, dominance analysis assesses the order in which predictor variables can be ranked in terms of their contribution to explaining R2 (Johnson & LeBreton, 2004). The general dominance weights shown in Table 3 reflect the relative importance of a predictor in all possible subsets of the regression models, and the rescaled dominance weight is the percentage in R2 explained by a variable. These analyses show that self-control plays the major role in this analysis in explaining nearly 70% of the variance in frugality in this analysis, far more than either materialism or independence explains. Joint variance is “the variance shared by a set of predictors (or independent variables) that is also shared with the criterion (or dependent variable) (Schoen, DeSimone, & James, 2011, p. 676). The importance of shared variance is that it enables researchers to better understand the relationships between correlated independent variables with a dependent variable that supplement the traditional emphasis on relative importance (i.e., beta weights, semi-partial correlations, and dominance weights). In the current study, the proportion of the explained variance in frugality (R2 = 0.127) attributable to the joint effects of materialism and consumer independence was only 1.8% and non-significant, while that between independence and self-control was 5.2%, also non-significant, but the joint effect of materialism and self-control was 20.3%. Thus, the variance in frugality in the current study was largely attributable to the influence of selfcontrol, followed by materialism, with independence playing only a minor role. Part of the variance could also be attributed to the joint effect of materialism and self-control in the sense that being both less materialistic and having more self-control simultaneously contributes to a frugal lifestyle. We can interpret this effect as a reciprocal effect (not an interaction) or a synergy between the two traits. To assess the possibility that the observed relationship between the antecedents and F-scores might be

“less than $25” to “more than $200.” The scores on the F-scale were negatively correlated with the compulsive buying scale (r = −.35) and with the spending measure (r = −.26), adding some criterion validity. Because those scoring higher on the F-scale report less compulsive buying and ordinary spending, we can thus place some confidence in the validity of the findings.

Main Analyses To test the hypotheses, we correlated summed scores on the F-scale with respondents’ reported sex, age, and income as well as the summed scores on the materialism scale, consumer independence scale, and self-control scale. These results appear in Table 2. The correlation coefficients of the three demographic variables with the scale scores showed virtually no relationships, and so these variables were no longer considered as potential control variables in the analysis. The correlations, however, do support all three hypotheses. Scores on the F-scale were negatively correlated with materialism (–.21), supporting H1 and replicating earlier findings (Lastovicka, et al., 1999; Goldsmith, et al., 2014). Likewise, the F-scale was positively correlated with consumer independence (.14), supporting H2 and replicating Goldsmith, et al. (2014). Finally, the F-scale was positively correlated (.33) with self-control, supporting H3 and adding a new insight into the psychology of frugality. To assess the combined influence of these three traits on frugality, we regressed the F-scale scores across the other three traits’ scores. These results appear in Table 3. Multicollinearity as assessed by the largest Variance Inflation Factor (1.16) appears to not influence the analysis. The regression coefficients for materialism and self-control were statistically significant (p < .05), but the coefficient for independence was not (p = .08). Possible explanations for this result could lie in the interrelationships among the independent variables. To explore further the relationships between the independent variables and the dependent variable, we performed a dominance analysis followed by an analysis of joint variance.

Follow-Up Analyses Dominance analysis assesses the relative importance of correlated predictor variables in a regression analysis in

TABLE 3 Regression of Frugality on Materialism, Independence, and Self-control (N = 455) β

Variable Constant Materialism

t

p

8.80