Saving in Cycles

Abstract

Low personal savings rates are an important social issue in the United States. We propose and test one particular method to get people to save more money that is based on the cyclical time orientation. In contrast to conventional, popular methods that encourage individuals to ignore past mistakes, focus on the future, and set goals to save money, our proposed method frames the savings task in cyclical terms, emphasizing the present. Across the studies, individuals who used our proposed cyclical savings method, compared with individuals who used a linear savings method, provided an average of 74% higher savings estimates and saved an average of 78% more money. We also found that the cyclical savings method was more efficacious because it increased implementation planning and lowered future optimism regarding saving money.

Keywords

cyclical time orientation personal savings implementation plan future optimism financial decision making motivation

The U.S. personal-savings rate has hovered below 5% for much of the past decade (Federal Reserve Bank of St. Louis, 2012), significantly below what is required to retire comfortably (Skinner, 2007). Although participation in employers’ automatic-payroll-deduction plans is an effective way for employees to increase their savings (Thaler & Benartzi, 2004), only 38% of eligible employees participate in such plans (Helman, Copeland, & VanDerhei, 2012). Getting people to save more money is an important challenge for policymakers and social scientists.

In this research, we proposed and tested one particular method to get people to save more money. In contrast to popular savings methods that encourage individuals to ignore past mistakes, focus on the future, and set goals to save money (Adams & Rau, 2011; Ramsey, 2007; Ülkümen & Cheema, 2011), our proposed method frames the savings task in cyclical terms, focusing on the present. It emphasizes the diagnostic significance of current savings behavior over its causal significance in motivating the person.

The difference between our proposed method and conventional methods can be traced to psychological, religious, and cultural perspectives on time orientation that distinguish between cyclical and linear models of time. Whereas a linear model of time characterizes life events in past, present, and future terms, a cyclical model sees them as a series of recurring experiences (e.g., Caillois & McKeon, 1963; Müller & Giesbrecht, 2006). Under the linear model of time, the individual’s focus is on the forward flow of time and on an improvement from the current state (Müller & Giesbrecht, 2006). In contrast, under the cyclical model of time, the future is seen as determined by past and present events, and the diagnostic significance of current actions provides motivation.

The conventional wisdom among personal-finance experts is that approaches to saving money that draw on the linear model are more effective.¹ They motivate individuals by rejecting their past failure to save, orient them toward the future, and encourage them to strive to reach savings goals. However, we think that people will save more when encouraged to view time in a cyclical fashion, which emphasizes recurrent behavioral patterns, than when encouraged to view time in a linear fashion, which emphasizes the time remaining to achieve goals. We believe this for two reasons. First, our proposed cyclical method should lead to an increase in concrete and detailed implementation planning (Liberman & Trope, 1998), whereas the linear approach should lead people to frame the savings task in more abstract terms (Trope & Liberman, 2003), which reduces implementation planning (Gollwitzer & Sheeran, 2006).

Second, individuals who use these models should have different degrees of future optimism about saving money. Because they believe that the future will be better than the present, linear-method savers should make more optimistic assessments about their future saving. This prediction is supported by prior research that has shown that individuals make overoptimistic predictions regarding future behavior (Weinstein, 1980) and their ability to achieve future goals (Zauberman & Lynch, 2005). The upshot is that individuals who use the linear approach to saving should be more likely to defer saving money (Tam & Dholakia, 2011), with the expectation of having more money and an increased ability to save and control spending in the future. Because the cyclical model does not endorse a progressively improving life pattern, the decision maker should be less optimistic and less likely to defer saving money. Given that Americans tend to be linear in their time orientation (Briley, 2009), we believe that teaching them the cyclical savings approach offers a potentially useful and effective method to help them save more money.

We tested our predictions in three studies. In Study 1, we examined the validity of the hypothesized effects of the linear and cyclical models of time by assessing the effects of a general (non-savings-specific) manipulation based on these two time orientations on savings estimates of participants. In Study 2, we developed and tested savings-specific instructions based on the cyclical and linear models and measured both savings estimates and actual savings of participants. Finally, in Study 3, we examined the roles of two potential mediators of the effect of models of time on savings behavior: the individual’s degree of implementation planning and his or her future optimism about saving money on savings estimates in the near and distant future. Overall, our findings show that getting people to think of saving money in cyclical terms significantly increases the amount of money they predict they will save and the amount of money they actually save.

Study 1: Cyclical Versus Linear Orientation and Savings Estimates

Purpose and method

Participants were 157 employed² online panelists (80% female, 20% male; mean age = 58 years) randomly assigned to either a cyclical-orientation or a linear-orientation condition. Participants were told that we wanted their opinion regarding a method developed by life coaches, which was then described. Afterward, participants were asked to write an essay explaining ways in which they would apply the method to different aspects of their lives. Instructions in the cyclical-orientation condition were as follows:

This idea acknowledges that our lives consist of many small and large cycles, that is, events that repeat themselves, just like the four seasons return every year. In this orientation, the individual expects the future to be like the present as these cycles repeat themselves. Therefore, completing important tasks involves incorporation of these repeating cycles, and so the person tries to create routines or habits right now, such that the same actions for these important tasks are repeated in every cycle. The belief is that if you perform an action in the current cycle now, you will be more likely to perform this particular action again in the next cycle. But if you do not perform it now, you will be less likely to perform it in the next cycle.

Study participants in the linear-orientation condition were instructed as follows:

This idea acknowledges that our lives consist of separate and progressive time periods, that is, events are over once they are in the past, just like childhood, adolescence, and adulthood. In this orientation, the individual sees the future as a road that stretches forward and onward from the present. Therefore, completing important tasks involves thinking about the future, and so the person tries to choose benchmarks or goals for each time period so that discrete actions can be performed to reach towards them. The belief is that if you perform an action now, you will be in a better position for the future. But if you do not perform the action now, progress will be slowed and you will need to catch up in the future.

After completing the essay, participants indicated how much money they expected to save³ the next month and the percentage of their annual income they expected to save the next year.

Results and discussion

Participants in the cyclical-orientation condition, compared with those in the linear-orientation condition, provided marginally higher savings estimates for the next month (M_cyclical = $174.80; M_linear = $118.60), F(1, 155) = 3.53, p = .06, d = 0.31, and indicated that they would save more of their income the next year (M_cyclical = 11.8%; M_linear = 6.6%), F(1, 152) = 3.83, p = .05, d = 0.31. These data indicate that, as expected, thinking in cyclical terms rather than in linear terms led to higher savings estimates.

Study 2: Cyclical Versus Linear Savings Methods and Saving Money

Method

One hundred forty-five adult commuter students (62% female, 38% male; mean age = 26.5 years, age range = 19–53 years) participated in this two-phase study in return for course credit. In the first phase, participants were randomly assigned to one of three conditions—the cyclical-savings-method condition (n = 44), the linear-savings-method condition (n = 53), or a control condition (n = 48)—and provided with instructions concerning their specific savings task. Participants in the cyclical condition received the following instructions:

This approach acknowledges that one’s life consists of many small and large cycles, that is, events that repeat themselves. We want you to think of the personal savings task as one part of such a cyclical life. Make your savings task a routinized one: Just focus on saving the amount that you want to save now, not next month, not next year. Think about whether you saved enough money during your last paycheck cycle. If you saved as much as you wanted, continue with your persistence. If you did not save enough, make it up this time, with the current paycheck cycle.

The future will be exactly like the present: if you save money now, you will save in the next pay period. If you don’t save money during the present pay cycle, it is likely you won’t save money in the next cycle. We want you to focus on your personal savings in the present, and that is all. What’s more, at the end of the day, you will be able to look back and see how much personal savings you have achieved.

Study participants in the linear condition were instructed as follows:

This approach acknowledges that one’s life is made of separate and progressive time compartments such as the past, present, and future. We want you to think of the personal savings task as part of such a linear progress. Make your saving task a planned one: Just focus on the total amount of your savings goal for the future. Think about discrete savings tasks and do each task one at a time. Do not think about what you have or have not saved in the past. The past is already past.

The future will be a road that stretches forward and onward from the present. If you save money now, you will be in a much better position in the future, and this better future state forms the basic idea of progress. Saving money is not an action that is the end goal, but rather a means to the more important goal of attaining something which lies in the future, such as your retirement or a better life in general.

Individuals assigned to the control condition received no specific savings-task instructions. Participants were then told to use their assigned savings method as their personal savings approach for the next 2 weeks. All participants indicated how much money they would save during the 2-week period, answered a manipulation-check measure (“To what extent does the savings approach recommended by experts rely on thinking about life as cyclical or linear?”; 1 = life as cyclical, 7 = life as linear), and provided demographic data (gender, age, household income, and full-time vs. part-time employment). Participants were asked to specify their degree of past success at saving money; responses were made using a 7-point scale from 1, not at all successful, to 7, extremely successful. Next, we asked participants to keep track of their savings and informed them that they would be contacted in 2 weeks. In a short online survey 2 weeks later, they reported how much money they had saved and the extent to which they had applied the savings method that had been explained to them. Finally, they were thanked for participating and debriefed.

Results

Indicating the manipulation’s success, results showed that cyclical-condition participants described their savings method as cyclical rather than linear to a greater degree (M_cyclical = 2.82 vs. M_linear = 4.96), t(95) = 6.25, p < .001. As predicted, participants in the cyclical condition provided a higher savings estimate than did those in the linear condition, after we controlled for demographics and past success at saving money (M_cyclical = $223 vs. M_linear = $140), F(1, 137) = 3.99, p < .05, d = 0.40. Participants’ savings estimates in the control condition (M_control = $133) were no different from estimates in the linear condition, F(1, 137) = 0.03, p > .86, d = 0.04, and were lower than estimates in the cyclical condition, F(1, 137) = 4.58, p < .05, d = 0.42 (see Fig. 1).

Fig. 1.

Participants’ savings estimates and actual savings as a function of condition (Study 2).

At the end of the 2-week period, those using the cyclical method had saved more money than those using the linear method (M_cyclical = $216 vs. M_linear = $118), F(1, 137) = 5.17, p < .05, d = 0.44. Participants in the control group saved less money than did participants in the cyclical group (M_control = $81), F(1, 137) = 9.72, p < .01, d = 0.67, and approximately the same amount of money as participants in the linear group, F(1, 137) = 1.23, p > .26, d = 0.25. The money participants saved in both the cyclical and the linear groups was consistent with their earlier estimates (ts < 1, ps > .53), but those in the control group saved much less than they had estimated they would, t(94) = 2.65, p < .05.

Discussion

Participants who employed the cyclical savings method, compared with those who used the linear savings method, provided a 70% higher estimate of how much money they would save in the next 2 weeks and reported saving 82% more money during this period. The insignificant difference in savings estimates between the linear and control groups suggests that the linear savings method may resemble the approach that individuals would use without any intervention.

Study 3: The Mediating Effects of Implementation Planning and Future Optimism

In Study 3, we examined the mediating roles of implementation planning and future optimism on savings estimates provided under the cyclical and linear savings methods. We hypothesized that the cyclical method of saving would encourage individuals to save money—through increased planning and decreased future optimism—in the near future, whereas the linear method would lead to a deferral of saving, thereby inflating the savings estimate for the distant future but dampening the estimate provided for the near future (Shepperd, Ouellette, & Fernandez, 1996).

Method

One hundred sixty-two employed adult commuter students (67% female, 33% male; mean age = 27.3 years, age range = 19–56 years) participated in the study in return for course credit. Participants were randomly assigned to either a cyclical- (n = 81) or a linear-savings-method condition (n = 81). They read the same savings-method instructions used in Study 2 and provided estimates of how much money they would save during two time periods—the next month and a future month (the same month the next year). We asked participants to provide savings estimates for the same month 1 year apart to keep duration constant, yet avoid seasonal variations in income or expenses from influencing estimates. Finally, participants completed measures of implementation planning, future optimism, demographics, and past success at saving money.

Measures

We used four items to measure implementation planning: (a) “I feel like I know exactly what I have to do to reach the savings goal that I indicated above”; (b) “I have a detailed plan for saving the money that I indicated”; (c) “The benefits of saving the amount indicated are clear in my mind”; and (d) “I have a good idea of the things I need to do to reach this savings goal.” Responses were made using 7-point scales from 1 (strongly disagree) to 7 (strongly agree).

We assessed future optimism with the following four items: (a) “I will probably have more money to spend in the future than I have now”; (b) “In the future, my income flow will be a lot higher than it is now”; (c) “I feel optimistic that I will be able to save more in the future in comparison to what I can save now”; and (d) “I will be able to control my spending more in the future than I do now.” Responses were made using 7-point scales from 1 (strongly disagree) to 7 (strongly agree). Reliabilities for both scales were adequate, with alphas of .75 for implementation planning and .81 for future optimism, and the respective item averages were used for the analysis.

Results

A 2 (savings method: cyclical vs. linear) × 2 (time period: next month vs. future month) within-subjects analysis of variance revealed a significant main effect of savings method, F(1, 315) = 5.96,⁴ p < .05, and a significant two-way interaction between savings method and time period, F(1, 315) = 10.37, p < .01, after we controlled for demographics and past success at saving money. Participants who used the cyclical method provided a higher savings estimate for the next month than did linear-method savers (M_cyclical = $409 vs. M_linear = $256), F(1, 315) = 4.52, p < .05, d = 0.27; however, they provided a lower savings estimate for the future month (M_cyclical = $372 vs. M_linear = $530), F(1, 315) = 4.84, p < .05, d = −0.37. The savings estimates for the next month and the future month did not differ for cyclical-method participants, F(1, 315) = 0.30, p > .58, but, buoyed by optimism (see the following section), linear-method participants’ future-month estimate was much higher than their next-month estimate (M_{next month} = $256 vs. M_{future month} = $530), F(1, 315) = 16.03, p < .001 (see Fig. 2).

Fig. 2.

Participants’ savings estimates for the next month and a future month as a function of savings methods (Study 3).

Linear-method savers were more optimistic (M_cyclical = 5.08; M_linear = 5.43), F(1, 155) = 4.08, p < .05, but reported lower implementation planning (M_cyclical = 5.65; M_linear = 5.25), F(1, 155) = 4.77, p < .05, than cyclical-method savers.

The mediating effects of implementation planning and future optimism on the relationship between savings method and savings estimate

To evaluate the two mediators jointly, we followed Zhao, Lynch, and Chen’s (2010) approach and conducted a single bootstrap test of the indirect (mediated) effect. The estimated mediated effect represents the multiplicative product of the paths from the savings method to the mediator (implementation planning and future optimism) and from the mediator to the savings estimate (see Preacher & Hayes, 2008, for details).

Mediating effects on savings estimates for the next month

The direct effect of the savings method on savings estimates for the next month was significant (b = −127.42, p = .033). When we held implementation planning and future optimism constant, the cyclical (relative to the linear) savings method increased the next month’s savings estimates by $127.42.

Using a bootstrap analysis with 5,000 resamples, we found that the indirect effect for implementation planning was significant at −37.24, with a 95% confidence interval of −93.79 to −7.93. Using the linear method, relative to the cyclical method, reduced implementation planning by 0.40 units on the 7-point scale described earlier. When the savings method was held constant, a unit increase in implementation planning increased savings estimates for the following month by $94.12. For future optimism, the mean indirect effect was significant at 16.46, with a 95% confidence interval of 1.76 to 52.58. The linear method, relative to the cyclical method, increased future optimism by 0.34 units on the 7-point scale. When the savings method was held constant, a unit increase in future optimism increased savings estimates for the following month by $48.08.

From these results, we concluded that the two mediators, implementation planning and future optimism, partially mediated the effects of the savings method on savings estimates, and their effects were opposite in direction: Implementation planning complemented (worked in the same direction as) the direct effect of the cyclical (vs. linear) savings method on savings estimates for the next month, and future optimism operated in opposition to it.

Discussion

These findings provide insight into the savings decision-making process and support the explanation that the cyclical savings method encourages individuals to save money right away by producing a mind-set—through increased implementation planning and lowered future optimism—that supports this objective. In contrast, the linear savings method encourages deferral of the savings task to the future, which inflates predictions about savings in the distant future but dampens current saving. Although most people may not think about how much money they will save during specific future periods, this study demonstrated the distinction between saving now and saving in the future and the fact that these two decisions are unique and are affected by implementation planning and future optimism. People can make savings decisions at any time, and we surmise that deferral under the linear savings method may continue unabated as people’s distant future becomes the near future and, eventually, the present.

General Discussion

Drawing on the culturally based distinctions between linear and cyclical time orientations (Bergadaa, 1990; Caillois & McKeon, 1963; Graham, 1981; van Geert, 2006), we developed and examined the roles of savings methods based on these orientations in affecting decision makers’ estimates of how much money they will save and their actual savings. Our findings indicate that the use of a savings method based on a cyclical time orientation that emphasizes the cyclical and routinized nature of the savings task and fosters a focus on the present leads to higher savings estimates than does the use of a savings method based on a linear time orientation, which highlights linear progress and orients the individual toward the future, as many conventional, popular methods are wont to recommend. The findings also illustrate that the diagnostic significance of one’s current savings behavior is more motivating that its causal significance.

The conceptual distinction between the cyclical and linear time orientations encompasses a number of potentially meaningful differences reflected in our savings-method instructions: The cyclical-savings-method instructions, compared with the linear-savings-method instructions, include (a) a stronger present orientation (vs. future orientation), (b) a focus on concrete implementation planning (vs. abstract implementation planning), (c) a focus on diagnostic significance (vs. causal significance), and (d) a stronger emphasis on action under recurrent conditions (vs. the time remaining to achieve goals). To examine which differences contribute to the effects we found, we conducted a study to explore whether a present-versus-future time-orientation manipulation would be sufficient to produce the effects on savings estimates we had observed in the studies reported here. We assigned 44 employed students to either a present- or a future-orientation condition and gave them instructions that were prefaced in the same way as in our three studies reported here but concerned present-oriented or future-oriented savings approaches. Results of a one-way analysis of variance revealed that participants’ savings estimates in two conditions were not different from each other, F(1, 42) = 0.18, p > .37, d = 0.13. Although this finding tentatively suggests that the present/future orientation is not the key on its own, future research should tease out which factor or (more likely) combination of factors described here are critical for explaining the observed differences between cyclical and linear primes.

Researchers recently have shown that linking one’s future and present selves, either with age-progressed renderings (Hershfield et al., 2011) or by appealing to one’s sense of social responsibility to one’s future self (Bryan & Hershfield, 2012), can increase savings. Our findings build on this line of work by suggesting that linking one’s present to one’s future in more general terms through instructions about the cyclical savings method can work in the same way. Our results are also consistent with Ülkümen and Cheema’s (2011) recent finding that when consumers do not have a specific savings goal, their focusing on how to save (i.e., by forming an implementation plan) leads to higher savings. However, a key difference between their work and ours was that Ülkümen and Cheema framed savings estimates as specific goals influencing other outcomes, such as anticipated success at saving money, whereas savings estimates constituted a key dependent variable in our research.

Not only are our findings regarding the efficacy of the cyclical savings method practically important, but they also open the door to promising research opportunities. As van Geert (2006) has noted, “Cycles abound in the cultural organization of our lives” (p. 493). In routinizing the savings task, our instructions to participants implicitly relied on the formalized paycheck cycle that is the norm in the United States; however, irregular windfalls, such as a performance bonus or an inheritance, offer significant opportunities to augment one’s savings rates over and above regular savings rates (Thaler, 1990). Methods that get people to save rather than spend more of their irregular earnings should be studied in future research, given that this practice is likely to significantly affect overall savings rates (Thaler & Benartzi, 2004).

In conclusion, the present research offers a practical yet theory-based method that can be easily implemented by personal-finance counselors and by practitioners, such as administrators of retirement plans, to help individuals in their efforts to save more money. For instance, on meeting with a retirement-plan administrator to sign up for an employer’s retirement-plan benefits, a key decision that new employees must make is what percentage of their paycheck to save. Our results clearly imply that providing instructions on the cyclical savings method is likely to increase the savings rate chosen, and as prior research has shown, such decisions can have long-lasting impacts once made (Thaler & Benartzi, 2004). A cyclical savings approach could also be used by parents and teachers in the economic socialization of children to inculcate desirable savings habits.

Footnotes

Declaration of Conflicting Interests

The authors declared that they had no conflicts of interest with respect to their authorship or the publication of this article.

Notes

References

Adams

G. A.

Rau

B. L.

(2011). Putting off tomorrow to do what you want today: Planning for retirement. American Psychologist, 66, 180–192.

Bergadaa

M. M.

(1990). The role of time in the action of the consumer. Journal of Consumer Research, 17, 289–302.

Briley

D. A.

(2009). Looking forward, looking back: Cultural differences and similarities in time orientation. In Wyer

R. S.

Chiu

C. Y.

Hong

(Eds.), Understanding culture: Theory, research and application (pp. 311–325). New York, NY: Psychology Press.

Bryan

C. J.

Hershfield

H. E.

(2012). You owe it to yourself: Boosting retirement saving with a responsibility-based appeal. Journal of Experimental Psychology: General, 141, 429–432.

Caillois

McKeon

(1963). Cyclical time, rectilinear time. Diogenes, 11, 1–13.

Federal Reserve Bank of St. Louis. (2012). Personal saving rate. Retrieved from http://research.stlouisfed.org/fred2/series/PSAVERT/

Gollwitzer

P. M.

Sheeran

(2006). Implementation intentions and goal achievement: A meta-analysis of effects and processes. In Zanna

M. P.

(Ed.), Advances in experimental social psychology (Vol. 38, pp. 69–119). San Diego, CA: Academic Press.

Graham

R. J.

(1981). The role of perception of time in consumer research. Journal of Consumer Research, 7, 335–342.

Helman

Copeland

VanDerhei

(2012, March). The 2012 Retirement Confidence Survey: Job insecurity, debt weigh on retirement confidence, savings (EBRI Issue Brief No. 369). Abstract retrieved from http://ssrn.com/abstract=2026003

10.

Hershfield

H. E.

Goldstein

D. G.

Sharpe

W. F.

Fox

Yeykelis

Carstensen

L. L.

Bailenson

J. N.

(2011). Increasing saving behavior through age-progressed renderings of the future self. Journal of Marketing Research, 48, S23–S37.

11.

Liberman

Trope

(1998). The role of feasibility and desirability considerations in near and distant future decisions: A test of temporal construal theory. Journal of Personality and Social Psychology, 75, 5–18.

12.

Müller

Giesbrecht

G. F.

(2006). Psychological models of time: Arrows, cycles and spirals. Culture & Psychology, 12, 221–229.

13.

Preacher

K. J.

Hayes

A. F.

(2008). Asymptotic and resampling strategies for assessing and comparing indirect effects in multiple mediator models. Behavior Research Methods, 40, 879–891.

14.

Ramsey

(2007). The total money makeover: A proven plan for financial fitness. Nashville, TN: Thomas Nelson.

15.

Shepperd

J. A.

Ouellette

J. A.

Fernandez

J. K.

(1996). Abandoning unrealistic optimism: Performance estimates and the temporal proximity of self-relevant feedback. Journal of Personality and Social Psychology, 70, 844–855.

16.

Skinner

(2007). Are you sure you’re saving enough for retirement? Journal of Economic Perspectives, 21, 59–80.

17.

Tam

Dholakia

U. M.

(2011). Delay and duration effects of time frames on personal savings estimates and behavior. Organizational Behavior and Human Decision Processes, 114, 142–152.

18.

Thaler

R. H.

(1990). Anomalies: Saving, fungibility and mental accounts. Journal of Economic Perspectives, 4, 193–205.

19.

Thaler

R. H.

Benartzi

(2004). Save more tomorrow^TM: Using behavioral economics to increase employee saving. Journal of Political Economy, 112, S164–S187.

20.

Trope

Liberman

(2003). Temporal construal. Psychological Review, 110, 403–421.

21.

Ülkümen

Cheema

(2011). Framing goals to influence personal savings: The role of specificity and construal level. Journal of Marketing Research, 48, 958–969.

22.

van Geert

(2006). Time, models and narratives: Towards understanding the dynamics of life. Culture & Psychology, 12, 487–507.

23.

Weinstein

N. D.

(1980). Unrealistic optimism about future life events. Journal of Personality and Social Psychology, 39, 806–820.

24.

Zauberman

Lynch

J. G.

(2005). Resource slack and propensity to discount delayed investments of time versus money. Journal of Experimental Psychology: General, 134, 23–37.

25.

Zhao

Lynch

J. G.

Chen

(2010). Reconsidering Baron and Kenny: Myths and truths about mediation. Journal of Consumer Research, 37, 197–206.