ArticlePDF Available

The Benefits of Emergency Reserves: Greater Preference and Persistence for Goals having Slack with a Cost

June 2017
Journal of Marketing Research 54(3)

DOI:10.1509/jmr.15.0231

Authors:

University of Pennsylvania

Marketers of programs that are designed to help consumers reach goals face dual challenges of making the program attractive enough to encourage consumer signup while still motivating consumers to reach desirable goals and thus stay satisfied with the program. The authors offer a possible solution to this challenge: the emergency reserve, or slack with a cost. They demonstrate how an explicitly defined emergency reserve not only is preferred over other options for goal-related programs but can also lead to increased persistence. Study 1 demonstrates that consumers prefer programs with emergency reserves to programs that do not have them, and Study 2 further clarifies that consumers’ preference for an emergency reserve depends on the presence of a superordinate goal. Study 3 reveals that consumers prefer goals with emergency reserves because they perceive them to have both higher attainability and value than other goals. Study 4 demonstrates that reserves can lead to increased goal persistence in a ...

Content uploaded by Marissa A. Sharif

Content may be subject to copyright.

MARISSA A. SHARIF and SUZANNE B. SHU*

Marketers of programs that are designed to help consumers reach goals

face dual challenges of making the program attractive enough to encourage

consumer signup while still motivating consumers to reach desirable goals

and thus stay satisﬁed with the program. The authors offer a possible

solution to this challenge: the emergency reserve, or slack with a cost. They

demonstrate how an explicitly deﬁned emergency reserve not only is

preferred over other options for goal-related programs but can also lead

to increased persistence. Study 1 demonstrates that consumers prefer

programs with emergency reserves to programs that do not have them, and

Study 2 further clariﬁes that consumers’preference for an emergency

reserve depends on the presence of a superordinate goal. Study 3 reveals

that consumers prefer goals with emergency reserves because they perceive

them to have both higher attainability and value than other goals. Study 4

demonstrates that reserves can lead to increased goal persistence in a

realistic task that involves persistence over time. Finally, Studies 5 and 6

reveal that consumers persist more with reserve goals because they want to

avoid using the “emergency”reserve.

Keywords: emergency reserve, slack, goals, motivation, preference

Online Supplement : http://dx.doi.org/10.1509/15.0231

The Beneﬁts of Emergency Reserves:

Greater Preference and Persistence for

Goals That Have Slack with a Cost

Students cheer when they learn that they have extra credit

available; dieters love having a possible “cheat day”in their

diet; consumers greatly appreciate having a buffer in case they

exceed their minutes in their mobile plans. People seem to love

ﬂexibility, speciﬁcally a “just-in-case”type of ﬂexibility. From

the marketer’s perspective, offering this “just-in-case”type of

ﬂexibility may encourage consumers to sign up initially for

various programs, such as dieting programs, personal training

programs, or phone plans. However, not only do companies

want to attract consumers to sign up, they also want to ensure

that consumers perform well and meet desirable goals (e.g.,

lose more weight, exercise more, use fewer minutes) so that

they will stay satisﬁed with the program and continue to use it.

How can marketers offer consumers some kind of “just-in-

case”ﬂexibility while still encouraging them to reach their

original, difﬁcult goal? In this article, we offer a solution to this

challenge based on the notion of ﬂexibility with a cost: the

emergency reserve. We deﬁne an emergency reserve as pre-

deﬁned slack around a goal that can be used if needed but at a

small cost. For example, a reserve can be 500 extra emergency

calories available in a diet for the week, anemergency skipday

for the gym, or anemergency buffer range in a phone program.

Small costs associated with these reserves might be purely

psychological costs (consumers try to not use the emergency

reserve unless absolutely necessary), opportunity costs (if

consumers use it today, they cannot use it tomorrow),or future

costs (if consumers use the reserve today, they might have to

do something tomorrow to make up for it).

In this article, to test how emergency reserves affect both

preference and persistence, we examine three types of goals:

reserve goals, hard reference point goals (hard goals), and easy

reference point goals (easy goals). Reserve goals have a difﬁcult

*Marisa A. Sharif is a doctoral candidate in marketing (e-mail: marissa.

sharif.1@anderson.ucla.edu). Suzanne Shu is Associate Professor of Marketing

(e-mail: suzanne.shu@anderson.ucla.edu). Anderson School of Management, Uni-

versity of California, Los Angeles. This article is based on part of the ﬁrst author's

dissertation. Coeditor: Rebecca Ratner; Associate Editor: Anirban Mukhopadhyay.

ISSN: 0022-2437 (print) Vol. LIV (June 2017), 495–509

1547-7193 (electronic) DOI: 10.1509/jmr.15.0231495

reference point plus an additional emergency reserve (e.g.,

a goal of going to the gym seven days a week plus two

“emergency skip”days). Hard goals have the same difﬁcult

reference point but without the additional emergency reserve

(e.g., a goal of going to the gym seven days a week). Easy

goals have an easier reference point, with the additional

emergency reserve already incorporated into the goal with no

additional cost (e.g., a goal of going to the gym ﬁve days a

week). We propose that including an explicitly deﬁned emer-

gency reserve within a goal will (1) be preferred to both hard

and easy goals when the consumer has a superordinate goal

(e.g., lose weight) because of both the higher expectancy and

value of the reserve goal and (2) result in greater goal per-

sistence than hard and easy goals as a result of a resistance to

break into the emergency reserve.

In this research, we investigate the inﬂuence of emergency

reserves on preference and persistence in six studies. The ﬁrst

three studies explore when and why people prefer emergency

reserves, and the last three studies investigate how and why

emergency reserves inﬂuence goal persistence. We ﬁrst pro-

pose the theoretical basis for why consumers prefer reserve

goals and then discuss why these goals better enable consumers

to persist.

PREFERENCE FOR RESERVES

According to expectancy value theories, a consumer’s

motivation to reach a goal depends on both the expectancy of

reaching the goal and the value of the goal (Atkinson 1957;

Tolman 1955). However, because there are distinct phases

within goal pursuit, consumers are motivated differently de-

pending on how much progress they have made toward their

goal (Gollwitzer 1990; Heckhausen 1977; Heckhausen and

Gollwitzer 1987). In the initial stages of goal pursuit, con-

sumers primarily focus on the attainability of the goal rather

than the value of the goal (Zhang and Huang 2010). Both the

sociocognitive model (Bandura 1997) and goal-setting theory

(Locke and Latham 1990) suggest that consumers’willingness

to adopt a goal is largely inﬂuenced by how attainable the goal

seems. Consistent with this notion, consumers are more mo-

tivated with high perceived velocity (vs. low perceived ve-

locity) toward reaching a goal in initial goal pursuit (Huang

and Zhang 2011). In addition, because consumers perceive

goals that can be achieved through multiple means as more

attainable (Kruglanski, Pierro, and Sheveland 2011), they are

more motivated when there are multiple means to achieve a

goal (Huang and Zhang 2013) or when there is more variety

among these means (Etkin and Ratner 2012). Therefore, this

stream of research suggests that making a goal more attainable

(or making it seem more attainable) motivates consumers in

the initial stages of goal pursuit.

In line with this theorizing, before adopting a goal (or

signing up for a program), consumers’preferences for reserve

goals compared with hard or easy goals may depend on the

attainability of these goals. Consumers may initially prefer

reserve goals to hard goals because they are more attainable

than hard goals. Reserve goals have the same difﬁcult refer-

ence point (e.g., going to the gym seven days a week) as hard

goals but provide additional ﬂexibility (e.g., two emergency

skip days), making them more attainable than hard goals.

However, consumers may not initially prefer reserve goals to

easy goals because they are equally attainable as easy goals

(e.g., consumers with both goals have technically not violated

their goal if they go to the gym only ﬁve days a week). In some

cases, reserve goals may even be perceived as less attainable

than easy goals because they focus on a more difﬁcult ref-

erence point than easy goals (e.g., going to the gym seven

days a week vs. going to the gym ﬁve days a week). In general,

these perceptions of attainability may determine preferences

between hard, easy, and reserve goals, with hard goals viewed

as the least attainable.

However, we expect preferences for these goals to be af-

fected by more than just attainability if there is a salient su-

perordinate goal. We hypothesize that the value of a goal (in

addition to the attainability of the goal) becomes relevant in

goal preference if there is a salient superordinate goal. Prior

research has demonstrated that people can perform better

in tasks with more difﬁcult reference points/goals (Heath,

Larrick, and Wu 1999; Locke and Latham 1990). When there

is a superordinate goal, more difﬁcult goals may be perceived

as more attractive (valuable) because they increase the chance

of obtaining the superordinate goal. For example, if a person

achieves a more difﬁcult goal of going to the gym seven days a

week instead of an easier goal of going to the gym ﬁve days a

week, (s)he will also be more likely to achieve the superor-

dinate goal of becoming more ﬁtorlosingweight.

In line with this theorizing, how does the existence of a

superordinate goal affect preference for reserve goals over

other goals? A reserve goal focuses consumers on a more dif-

ﬁcult reference point than an easy goal, increasing the chance

of performing better at the superordinate goal and thus also

increasing the attractiveness (value) of the goal. However, the

attainability of both goals will remain the same; people with a

reserve goal(e.g., going tothe gym seven daysa week plus two

“emergency skip”days) and people with an easy goal (e.g.,

ﬁve days a week) can both meet their goals by going to the gym

ﬁve days a week. Because the goals are equally attainable but

the reserve goal has a higher value, we expect that consumers

will prefer reserve goals toeasy goals if there is a superordinate

goal.

We hypothesize that consumers will still prefer reserve goals

to hard goals when there is a superordinate goal. Both reserve

and hard goals have more difﬁcult reference points (e.g., going

to the gym seven days a week) and thus are equally valuable.

However, the reserve goal has added ﬂexibility (e.g., two

emergency skip days) that makes it more attainable. Because

the goals have equal value but the reserve goal is more at-

tainable, we expect that consumers will prefer reserve goals to

hard goals.

: Consumers prefer reserve goals to both easy goals and hard goals

if there is a superordinate goal. Consumers prefer reserve goals

to hard goals, but not to easy goals, if there is no superordinate

goal.

PERSISTENCE WITH RESERVES

We propose that not only will consumers predictably prefer

reserve goals to easy and hard goals when there is a super-

ordinate goal, but they will persist more with reserve goals than

easy goals, hard goals, and other goals with ﬂexibility. Al-

though we posit that the higher expectancy and value of the

reserve drives preference for the reserve goal, we expect that

when actually pursuing a reserve goal, consumers will persist

more because of an additional element of the reserve: the cost

of using it.

496 JOURNAL OF MARKETING RESEARCH, JUNE 2017

Unlike other types of slack,1the emergency reserve is

unique because there is (1) a small cost of using the available

ﬂexibility and (2) a predeﬁned ﬁnite amount available. It is

designed to be used in “just-in-case”scenarios. We predict that

this small cost of using the emergency reserve causes con-

sumers to resist using it, leading to more persistence than those

with other goals.

Costs of the Reserve

Costs of using the reserve may include psychological, op-

portunity, and/or future costs. We examine cases in which the

reserve has only a psychological cost, only an opportunity

cost, or both. Although we suggest that the emergency re-

serve could be effective with other costs (e.g., future costs),

by examining only psychological and opportunity costs, we

do not affect the external incentives for achieving the goal.

To impose a psychological cost for the reserves in our

studies, we simply label the reserve an “emergency”reserve,

inducing people to avoid using the reserve unless absolutely

necessary. Prior research has suggested that people are sen-

sitive to similar labels and the psychological costs for vio-

lating them. More speciﬁcally, research in mental accounting/

budgeting has indicated that people use their resources dif-

ferently depending on how they are labeled (Heath and Soll

1996; Henderson and Peterson 1992; Kahneman and Tversky

1984; Thaler 1980, 1985). Most relevant to the “emergency

reserve,”people have “rainy day”funds, in which they set

aside money in case ofan emergency. They restrict themselves

to use that money only if absolutely necessary (Shefrin and

Thaler 1988; Thaler 1990). In this case, the cost of using the

funds is purely psychological as a result of the “rainy day”

labeling. In addition, as a more tangible example of salient

psychological costs, poor residents in India who were given

part of their salary in a separate, specially designated sealed

envelope were highly resistant to opening it and spending it

relative to those who receivedthe salary all at once (Soman and

Cheema 2011). Thus, drawing on this literature, by labeling the

reserve as “emergency-only use,”we induce a psychological

cost of breaking into the emergency reserve. We hypothesize

that consumers will try to resist using their reserve unless they

absolutely need it, avoiding the psychological cost associated

with using the emergency reserve.

In addition, to mimic real-life goal environments, we ex-

amine goal pursuit “over time”or in situations in which there

are multiple instances to apply the emergency reserve. In these

scenarios,the framing of the emergency reserve goal (e.g., goal

of going to the gym seven days a week with two “emergency”

skip days) makes the limited amount of ﬂexibility (e.g., only

two “emergency”skip days) very salient to consumers. Con-

sumers are more sensitive to their opportunity costs when their

resources are constrained (Spiller 2011) and also more resistant

to use resources with smaller budgets (Heath and Soll 1996;

Krishnamurthy and Prokopec 2010; Morewedge, Holtzman,

and Epley 2007; Shefrin and Thaler 1988; Stilley, Inman, and

Wakeﬁeld 2010). Thus, because the reserve goal is framed

with a limited number of emergency reserves available (e.g.,

two emergency skip days), it is very salient to consumers that if

they use their emergency reserves earlier, they will not be able

to use them later.As a result, because of the salient opportunity

cost of using their emergency reserves earlier, consumers may

resist using their emergency reserves, waiting for a time when

they might need to use them more.

We thus hypothesize that consumers will resist using their

emergency reserve because of the (1) psychological cost

(induced by the “emergency”labeling) and (2) opportunity

cost of using their emergency reserve too early, leading to more

persistence than those with other goals. We predict that the

underlying process for why consumers persist more with reserve

goals compared with other goal types is the same: resistance to

use the reserve because of the cost. Next, we formally describe

our hypotheses and explain why consumers with reserve goals

will be more likely to persist than those with easygoals,hard

goals, and other goals with ﬂexibility.

Reserve Versus Easy Goals

In addition to the cost of using the reserve, reserve goals

focus on a more difﬁcult reference point (e.g., goal of going to

the gym seven days a week plus two “emergency skip”days)

than easy goals (e.g., goal of going to the gym ﬁve days a

week). Prior research has suggested that people perform better

with more difﬁcult goals (Locke et al. 1981). Using prospect

theory as a framework (Kahneman and Tversky 1979), Heath,

Larrick, and Wu (1999) demonstrate that people expect to

work much harder if they have not yet reached their goal (in

the losses domain) than if they have already succeeded at their

goal (in the gains domain).

Thus, after succeeding (or being able to succeed later) at

reaching this easier reference point (e.g., going to the gym ﬁve

days a week), we expect those with easy goals to be less likely

to persist. However, we expect consumers with reserve goals

to try to avoid using their reserve and thus strive for a more

difﬁcult reference point. Therefore, we expect consumers with

reserve goals to be more likely to persist than those with easy

goals because of the following dual inﬂuence: (1) easy-goal

consumers will be less likely to persist because they have an

easier reference point and (2) reserve-goal consumers will be

more likely to persist as a result of a resistance to use the

emergency reserve.

: Consumers with reserve goals are more likely to persist than

those with easy goals.

Reserve Versus Flexible Goals

To ensure that the beneﬁcial effects of the reserve are not

due to simply being a ﬂexible goal, we also examine how

reserve goals compare with other ﬂexible goals. Prior research

in both the mental accounting and goal literature has found that

ﬂexible goals and ﬂexible mental accounts generally do not

result in better performance and can even result in worse per-

formance (compared with less ﬂexible goals and accounts).

Consumers exploit malleability and ambiguity in their mental

accounting rules to justify indulging in temptations, thus re-

ducing self-control performance (Ainslie 2001; Cheema and

Soman 2006). They perform worse with goals that have

multiple means of achieving them, once they are past the initial

stages of goal pursuit (Huang and Zhang 2013), and consumers

with high–low goals (e.g., a goal toscore between two and four

points) are more likely to pursue their goal again but are not

more likely to perform better than those with single goals

(e.g., a goal to score two points or a goal to score four points)

1Because of their cost, emergency reserves differ from Zauberman and

Lynch’s (2005) deﬁnition of slack (the perceived future surplus of a given

resource).

The Beneﬁts of Emergency Reserves 497

(Scott and Nowlis 2013). Furthermore, consumers with goals

with backup plans for the superordinate goal perform worse on

their primary goal than those without backup plans (Shin and

Milkman 2016). Finally, goal speciﬁcity has different effects

on consumers depending on the construal level. Nonspeciﬁc

goals under high construal lead to a lower perception of im-

portance and, in turn, lower success rates compared with speciﬁc

goals; however, goals under low construal lead to a reduced

sense of difﬁculty and, thus, higher success rates ( ¨

Ulk¨umen and

Cheema 2011).

To verify that goals with emergency reserves act dif-

ferently from other ﬂexible goals, we compare the effects of

reserve goals with the effects of other goals with ﬂexibility

(what we term “range goals”). We examine two types of range

goals: range-easy goals, which focus on the easier reference

point (e.g., “Your goal is to go to the gym ﬁve days a week;

however, you should aim to go to the gym seven days a

week”), and range-hard goals, which focus on the more dif-

ﬁcult reference point (e.g., “Your goal is to go to the gym seven

days a week; however, it is okay if you go to the gym ﬁve

days a week”). We hypothesize that consumers with range

goals will be less likely than those with reserve goals to reach

the more difﬁcult reference point of this ﬂexibility region

because there is no cost (psychological or otherwise) associ-

ated with using the ﬂexibility inherent in those goals.

: Consumers with reserve goals are more likely to persist than

those with range goals.

Reserve Versus Hard Goals

Reserve goals and hard goals both focus on the same dif-

ﬁcult reference point (e.g., go to the gym seven days a week).

However, reserve goals have some ﬂexibility, with a cost,

available (e.g., two emergency skip days). We hypothesize that

consumers with hard goals will make an effort to reach their

goal, but those with reserve goals will make an even greater

effort than those with hard goals to avoid using their reserve

because of its associated cost. As we have mentioned, the

psychological cost of the reserve is ampliﬁed by terming it an

“emergency”reserve, and people are sensitive to such label-

ing (Heath and Soll 1996; Henderson and Peterson 1992;

Kahneman and Tversky 1984; Shefrin and Thaler 1988; Thaler

1980, 1985, 1990). Therefore, in mere goals (i.e., there is no

external incentive for reaching the goal), the cost of using the

reserve may be perceived as greater than the cost of violating

the hard goal, leading to more persistence with reserve goals

than hard goals. While participants with hard goals will try

their best to reach their goal, those with reserve goals will try to

avoid using their “emergency”reserve unless they absolutely

need it.

: Consumers with reserve goals are more likely to persist than

those with hard goals.

This article demonstrates that consumers prefer reserve

goals to hard and easy goals and also persist more with reserve

goals than with other goals. Studies 1–3 test initial preferences

for emergency reserves. Studies 1a and 1b demonstrate that

consumers prefer reserve goals to hard and easy goals in a

weight loss program, and Study 2 further clariﬁes this pref-

erence by showing that reserve goals are preferred to both easy

and hard goals when they are part of a superordinate goal.

Study 3 reveals that consumers prefer reserve goals to easy and

hard goals because reserve goals have a higher perceived value

and attainability than many other types of goals. The ﬁnal

three studies examine consumers’persistence with reserve

goals compared with those with easy, hard, and range goals

in tasks with superordinate goals. Study 4 demonstrates the

beneﬁcial effects of an emergency reserve in a realistic work

situation that requires persistence in a task over multiple

days. Finally, Studies 5 and 6 reveal that consumers with

reserve goals persist more because they try to resist using

their emergency reserve.

STUDIES 1A AND 1B: WEIGHT LOSS STUDIES

In Studies 1a and 1b, we explore consumers’preference

foranoptionwithanemergencyreserveinadomainwitha

superordinate goal: a hypothetical point-based weight loss

program. We expected that participants would prefer the reserve

option to the easy and hard options because there is a salient

superordinate goal within this domain (i.e., to lose weight).

Method

One hundred Amazon Mechanical Turk (MTurk) par-

ticipants completed each study (Study 1a: M

age

=34.33

years, age range: 19–65 years; 56 men; Study 1b: M

age

32.91 years, age range: 18–62 years; 12 men). Participants

were told to imagine that they wanted to lose weight and that

they were considering three different point-based weight

loss programs. Participants were asked to indicate which

program they would be most likely to sign up for. The easy

option offered 32 points per day, and the hard option offered

30 points per day. In Study 1a, the reserve option offered 30

points per day with 2 optional emergency points per day.

The cost of using the reserve in this scenario was purely

psychological by labeling the points as “emergency”(for

details, see Web Appendix A). In Study 1b, the reserve

option offered 30 points per day plus 14 optional emergency

points per week. This increases the ﬂexibility of the reserve

relative to Study 1a, but it also generates an opportunity

cost, in addition to a psychological cost, for early use of the

optional points.

Results

For both studies, we conducted a multinomial logistic re-

gression with the reserve option as the reference group. In

Study 1a, participants were marginally signiﬁcantly more

likely to choose the reserve option (46%) over the easy option

(29%; p= .052) and signiﬁcantly more likely to choose the

reserve option over the hard option (25%; p= .014). In Study

1b, participants were signiﬁcantly more likely to choose the

reserve option (55%) over the easy option (22%; p<.001) and

the hard option (23%; p<.001).

Discussion

In support of H

, we found in Studies 1a and 1b that

participants preferred reserve goals to hard goals and easy

goals when there was a superordinate goal (e.g., to lose

weight). In Study 1a, the only cost of using the reserve was

psychological; in other words, simply labeling some of the

points for emergency use was enough to affect preferences. In

Study 1b, participants also preferred the reserve if it had an

additional cost, an opportunity cost, such that use of the

emergency points today reduces the opportunity to use them

tomorrow.

498 JOURNAL OF MARKETING RESEARCH, JUNE 2017

In Study 2, we aim to explore the boundaries of pref-

erence for the reserve goal by testing how it is affected by the

salience of a superordinate goal. We use a scenario about

taking a class and manipulate the presence of a superor-

dinate goal (a career exam). We predict that participants will

prefer the reserve option to the easy option only if the class is

needed for a larger career exam (superordinate goal) but not

if the material would not be needed again in the future (no

superordinate goal).

STUDY 2: CLASS STUDY

In this study, we explore consumers’preference for a

reserveoptioncomparedwithaneasyoptionandahard

option within an education scenario in the presence and

absence of a superordinate goal. Participants indicated their

preference on a Likert scale between only two of the options

(either reserve vs. hard or reserve vs. easy).2We also begin

to explore why consumers may prefer the reserve option to

easy and hard options.

Method

Two hundred paid MTurk participants (M

age

= 31.07 years,

age range: 18–59 years; 126 men) took part in this study.

Participants were asked to imagine that they were required to

take a class. Participants were randomly assigned to the

career exam condition and the no–career exam condition. In

the career exam condition, participants were told that the

required class was a preparatory class for a separate larger

exam they had to take for their future career (a salient su-

perordinate goal). In the no–career exam condition, partic-

ipants were told that the class material would not be used in

their career and that they would not be required to take a

future related class (no superordinate goal). Therefore, in the

career exam condition, studying hard for the class would

increase students’likelihood of passing the class and also

help them better prepare for their career (i.e., have a better

chance of obtaining their superordinate goal). In the no–

career exam condition, studying hard for the class would

only increase their chances of passing the class.

Participants were told that two teachers taught the same

exact class in a similar fashion and that they used the exact

same ﬁnal exam at the end of the class. One of the teachers

offered an emergency reserve for the test (i.e., the reserve

option): students would have to receive a score of 20 (out of

25) on the exam to pass the class, but they also could earn 5

“emergency”extra credit points if they received below a 20.

The easy-option teacher required students to receive a score

of 15 (out of 25) on the exam, and the hard-option teacher

required a score of 20 (out of 25) to pass the class, without any

extra credit available.

Participants were randomly assigned to read about and

compare the reserve option and the easy option or the reserve

option and the hard option. They were asked, “Which teacher

would you be more likely to choose?”on a Likert scale with

1=“Deﬁnitely more likely to choose Teacher A,”4=“Equally

likely to choose Teacher A or Teacher B,”and 7 = “Deﬁnitely

more likely to choose Teacher B.”Teacher A was either the

hard option or the easy option and Teacher B was the reserve

option (for details, see Web Appendix B).

After indicating their preference, participants were asked

with which teacher they would study harder, would have

learned more after the ﬁnal, and would be more likely to pass

the class on a similar seven-point Likert scale. Participants

then completed the brief self-control measure (Tangney,

Baumeister, and Boone 2004).

Results

Because half of the participants indicated a preference

between the reserve and easy option and the other half in-

dicated a preference between the reserve and hard option, we

analyze these data separately (as combining them would not

be meaningful; see Figure 1). We used independent sam-

ple t-tests and one-sample t-tests versus the midpoint of 4

(being indifferent between the teachers) to analyze the data.

Preference for reserve goals to easy goals. Our superor-

dinate goal salience manipulation (career exam vs. no career

exam) had a signiﬁcant effect on participant’s preferences for

the reserve option versus the easy option. In support of H

participants were signiﬁcantly more likely to prefer the reserve

option to the easy option in the career exam condition than in

the no–career exam condition (M

=5.20vs.M

NCE

=3.89;

t(100) = −3.04, p= .003). Within the career exam condition,

participants were signiﬁcantly more likely to prefer the reserve

option compared with preferring both options equally

reserve

=5.20vs.M

both

=4.0;t(48)=4.24,p<.001).

However, within the no career exam condition, participants

were not signiﬁcantly more likely to prefer the reserve option

to preferring both options equally (M

reserve

=3.88vs.M

both

4.0; t(52) = −.35, p=.73).

Preference for reserve goals over hard goals. There was no

signiﬁcant effect of our superordinate goal salience manipu-

lation on participants’preference between the hard option and

the reserve option (M

=6.29vs.M

NCE

= 6.14; t(96) = −.58,

p= .57). Within both the career exam condition and the

no–career exam condition, participants were signiﬁcantly more

Figure 1

PREFERENCE FOR THE RESERVE VERSUS EASY AND HARD

GOALS, SPLIT BY CONDITION

Versus Easy Versus Hard

Preference

Superordinate goal No superordinate goal

Notes: For preference, 7 = “reserve goal,”and 1 = “other goal.”

2One could argue from Studies 1a and 1b that participants were simply

choosing at random between the reserve and nonreserve goals (easy and

hard). We test this alternative explanation in Study 2 by using a seven-point

Likert scale.

The Beneﬁts of Emergency Reserves 499

likelytopreferthereserveoptioncomparedwithpreferringboth

options equally (career exam condition: M

reserve

=6.29vs.

both

= 4.0; t(48) = 16.34, p<.001; no–career exam condition:

reserve

=6.14vs.M

both

= 4.0; t(48) = 10.50, p<.001).

Motivation (easy vs. reserve). There was no signiﬁcant

effect of our superordinate goal salience manipulation on par-

ticipants’intuition about their likelihood of passing the class

and their motivation to study between the easy option and the

reserve option. In both conditions, participants thought

they would study harder with the reserve option (career

exam condition: M

reserve

= 5.25 vs. M

both

= 4.0; t(48) = 5.20,

p<.001; no–career exam condition: M

reserve

= 4.91 vs.

both

= 4.0; t(52) = 3.36, p= .001) and would learn more with

the reserve option (career exam condition: M

reserve

= 5.23 vs.

both

= 4.0; t(48) = 6.14, p<.001; no–career exam condition:

reserve

= 4.79 vs. M

both

= 4.0; t(52) = 3.79, p<.001).

Furthermore, participants thought they would have similar

chances of passing with the reserve option and easy option in

both conditions.

Motivation (hard vs. reserve). There was no signiﬁcant

effect of our superordinate goal salience manipulation on

participants’intuition about their likelihood of passing/

learning from the class and their motivation to study be-

tween the hard option and the reserve option. In both con-

ditions, participants thought they would study similarly as

much and learn a similar amount with the reserve option and

the hard option but would be signiﬁcantly more likely to pass

the class with the reserve option than the hard option (career

exam condition: M

reserve

=5.57vs.M

both

= 4.0; t(48) = 6.63,

p<.001; no–career exam condition: M

reserve

=5.94vs.M

both

4.0; t(48) = 10.73, p<.001).

Self-control. We used ordinary least squares regression to

examine the effect of individual levels of self-control on

choice. We found that participants with high self-control were

overall more likely to choose the reserve option over the easy

option (b=.77,p= .014). Self-control did not signiﬁcantly

affect participants’likelihood of choosing the reserve option

over the hard option (b=.10,p= .66). We also did not ﬁnd a

signiﬁcant two-way interaction between individual-level self-

control and the superordinate goal manipulation (easy vs.

reserve: career exam vs. no career exam × self-control in-

teraction: b=.43,p= .49; hard vs. reserve: career exam vs. no

career exam × self-control interaction: b=.10,p=.76).

Discussion

Study 2 demonstrated that consumers prefer the reserve

option to the easy option only if they have a superordinate

goal and prefer the reserve option to the hard option in-

dependent of having a superordinate goal (H

). This study

thus reveals the importance of a contextual factor, the sa-

lience of superordinate goal, on preferences for emergency

reserves.

Study 2 also enabled us to test the importance of indi-

vidual factors, such as self-control, in predicting preference

for reserves. We found that participants with higher self-

control were more likely to prefer the reserve option to the

easy option overall. Participants may consider the emer-

gency reserve a safe self-control precommitment strategy

(Milkman, Minson, and Volpp 2013; Wertenbroch 1998);

those with high self-control are more likely to want to

precommit themselves to studying and, thus, to learning

more from the course. Consequently, both situational and

individual factors are important for predicting preference for

reserves.

Finally, Study 2 revealed why reserve options might be

preferred to easy and hard options. Participants thought they

would be more motivated to study harder and would learn

more with the reserve option than the easy option, suggesting

that reserve goals are perceived to have greater value than

easy goals. They also thought they would be equally likely to

pass the class with both teachers, which indicates that they

perceived the attainability of both goals to be equal.

Participants seemed to prefer the reserve option to the hard

option for a different reason. Participants thought they would

be more likely to pass the class with the reserve option than

with the hard option, suggesting that the reserve option is

perceived as more attainable. In addition, they thought they

would study equally as hard with both options, suggesting

that both goals are equally as valuable. In the next study, we

aim to replicate these effects and also examine consumers’

preference for other ﬂexible goals within a domain with a

salient superordinate goal.

STUDY 3: WORD SEARCH PREFERENCE STUDY

In Study 3, we further examine consumers’preference

for the reserve goal and intuitions about the attainability

(likelihood of succeeding at their goal) and value (likelihood

of performing better on the superordinate goal) of reserve

goals, hard goals, and easy goals when there is a superor-

dinate goal. In addition, we introduce two range goals. One

range goal focuses on the more difﬁcult, aspirational ref-

erence point; we refer to this range goal as range-hard. The

other range goal focuses on the easier, more attainable

reference point; we refer to this range goal as range-easy.

However, both goals are equally attainable as easy goals,

similar to reserve goals.

Method

Two hundred paid MTurk participants (M

age

= 35.43 years,

age range: 18–70 years; 79 men) completed this survey.

Participants were asked to imagine that they were com-

pletingtrainingforadifﬁcult word search test and that they

would receive a bonus if they performed very well on this

test. They were asked to complete a series of training word

searches in preparation for the test. The more word searches

they practiced, the more likely it was that they would do

better on the difﬁcult word search test at the end. Therefore,

the superordinate goal in this scenario is to perform well on

the word search test. Only participants who successfully

completed their training goal could try the difﬁcult word

search test. Participants were then able to choose which

training goal they would prefer.

For each training word search they completed, they were

told they would score one point in this hypothetical game.

Participants were randomly assigned to choose between the

reserve goal (“Goal to score 3 points. However, you also

have one emergency point that you can apply if you fail one

word search test”) and one of the other goals: easy (“Goal to

score 2 points”), hard (“Goal to score 3 points”), range-easy

(“Goal to score 2 points. However, you should aim to score 3

points. You will be able to try the word search test if you

score 2 points”), or range-hard (“Goal to score 3 points, but

it’s okay if you score 2 points. You will be able to try the

word search test if you score 2 points”).

500 JOURNAL OF MARKETING RESEARCH, JUNE 2017

Afterward, they were asked to choose with which goal

they thought they would be more likely to qualify for the

word search test (i.e., expectancy) and with which goal they

thought they would be more likely to perform better on the

word search test (i.e., value). Participants saw graphical

representations and descriptions of each goal when making

their choices. Range-hard, reserve, and hard goals all focus

on the more difﬁcult goal of completing all three training

word searches, reinforced by both text and a graphical

representation. However, the range-hard and reserve goals

offer ﬂexibility, making them more attainable than the hard

goal. Range-easy and easy goals focus on the easier goal of

completing two word searches, also reinforced by text and

graphical representation (for visual representations of all

goals, see Web Appendix C).

Results

Easy versus reserve. Participants were signiﬁcantly more

likely to choose a reserve goal over an easy goal (66.67%

reserve

vs. 33.34%

easy

(1,N=48)=5.33,

3p= .021. Participants

thought they would be equally likely to qualify to try the word

search test with both goals (52.1%

reserve

vs. 47.9%

easy

)

Lastly,

participants thought they would perform signiﬁcantly better

with the reserve goal than the easy goal on the word search test

(64.6%

reserve

vs. 35.4%

easy

(1, N = 48) = 4.08, p=.043).

Hard versus reserve. Participants were signiﬁcantly more

likely to choosea reserve goal over a hard goal (72.5%

reserve

vs.

26.5%

hard

(1, N = 49) = 10.80, p= .001). Participants

thought they would be signiﬁcantly more likely to qualify

to try the word search test with the reserve goal than the

hard goal (68.1%

reserve

vs. 31.9%

hard;

(1,N=47)=6.15,

p=.013)

Lastly, participants thought they would perform

approximately equally well on the word search test with the

reserve goal and the hard goal (56.3%

reserve

vs. 43.8%

hard

Range-easy versus reserve. Participants were equally likely

to choose the range-easy and easy goals (51%

reserve

vs.

49%

range-easy

). In addition, they thought they would be equally

likely to qualify for the word search test with both goals

(51%

reserve

vs. 49%

range-easy

) and thought both goals would

help them perform equally well (56%

reserve

vs. 44%

range-easy

Range-hard versus reserve. Participants were signiﬁcantly

more likely to choose a range-hard goal over a reserve goal

(34%

reserve

vs. 66%

range-hard

(1,N=47)=4.79,p=.029).

Participants thought they would be equally likely to qualify for

the word search test (45.1%

reserve

vs. 54.9%

range-hard

)and

would perform equally well (48.1%

reserve

vs. 51.9%

range-hard

)

under both goals.

Discussion

This study replicates the effects of Studies 1 and 2, such that

participants were more likely to choose a reserve goal over

both an easy goal and a hard goal in the presence of a su-

perordinate goal. In addition, replicating Study 2, participants

seem to prefer reserve goals over easy goals because they

perceive reserve goals as having a higher value (greater

likelihood of performing well on the superordinate goal) than

easy goals, but with the same attainability as easy goals

(equally likely to qualify for the ﬁnal word search). Con-

versely, participants seem to prefer reserve goals over hard

goals because they perceive reserve goals as being more

attainable than hard goals, but with the same value as hard

goals (see Figure 2).

Furthermore, the results of this study suggest that con-

sumers overall prefer ﬂexible goals. Participants had no pref-

erence between reserve and range-easy goals, perceiving

them to have similar value and attainability. Notably, par-

ticipants preferred range-hard goals to reserve goals but

thought they had similar value and attainability. Consumers

may prefer range-hard goals to avoid the psychological cost

associated with the reserve. However, we demonstrate that

this cost they may be trying to avoid ends up being a key

motivating factor in goal persistence.

Studies 1–3 demonstrated that participants prefer reserve

goals to other goals. In the remaining three studies, we examine

how participants persist with these different types of goals.

STUDY 4: DAILY CHALLENGE STUDY

In Study 4, we use an incentive-compatible task with real

behavior and real consequences to examine goal persistence.

This task requires persistence over multiple days because it

asks that participants set aside time each day to complete an

annoying task to reach their goals, much like daily gym at-

tendance or other long-term goals. In addition to comparing

participants’persistence under hard and easy goals with par-

ticipants’persistence under reserve goals, we begin by ex-

ploring how the persistence of participants with one type of

ﬂexible goal, range-easy, compares with the persistence of

those with reserve goals. While our prediction is that it is the

cost associated with the emergency reserve that encourages

persistence, it may simply be that introducing ﬂexibility is

enough to encourage more persistence; the range-easy goal

thus provides an early test of this alternate explanation. In

Studies 5 and 6, we examine the persistence of participants

with other types of ﬂexible goals. Drawing on H

–H

,we

predicted that participants with reserve goals would persist

more than those with other goal types, such as easy, range-

easy, and hard goals.

Method

Two hundred twenty-six paid participants (M

age

=22.12

years, age range: 18–53 years; 60 men) from a large public

university in the southwestern United States participated in this

study. Every morning for seven days, a different set of 35

CAPTCHAs was posted online for participants to complete.

Participants were randomly assigned to one of four conditions

in which we manipulated the goal for the number of days they

should complete the CAPTCHAs (easy, range-easy, reserve,

and hard).

Procedure

Participants in this study were on their university’swinter

break. A different set of 35 CAPTCHAS was posted every

morning and then removed at midnight. Participants received

$1 for every day that they completed each set of 35

CAPTCHAs. Participants were informed that there would be

no penalty for not completing the task daily.

Participants were randomly assigned to one of four con-

ditions, in which the goal of the task was manipulated. In

addition to receiving $1 per day for completing the task,

3All chi-square tests compare the number of participants who chose the

reserve option versus the other option (easy, hard, range-hard, or range-easy,

depending on condition) with what would be expected by chance (50% vs.

50% of observations).

The Beneﬁts of Emergency Reserves 501

participants received an additional $5 bonus if they com-

pleted their goal. In the easy condition, participants’goal was

to complete the task ﬁve days a week. In the range-easy

condition, participants’goal was to complete the task ﬁve

days a week, but they were also told that they “should aim to

complete the task every day of the week.”In the reserve

condition, participants’goal was to complete the task every

day (seven days) of the week. They were also told, “In case

you need it, up to two days will be excused,”and that they

would still receive their bonus if they missed up to two days

(but they would not receive the $1 per day payment for the

days they missed). In the hard condition, participants’goal

was to complete the task every day of the week (seven days).

In this study, there is both a psychological and an op-

portunity cost associated with using the reserve. If a par-

ticipant decides to skip the task on Monday and Tuesday,

(s)he cannot skip the task on Wednesday and still make the

goal (thus, there is an opportunity cost for early use of the

emergency reserve). In addition, there is a psychological cost

in the framing of the emergency reserve because participants

aretoldtouseitonly“in case they need it.”We predicted that

participants with reserve goals would be more likely to

complete their goal than participants with easy, hard, and

range-easy goals.

Results

Goal attainment. Reserve, range-easy, and easy goal

participants all received their bonus if they completed the

task ﬁve days a week. Hard-goal participants received their

bonus if they completed the task seven days a week. We used

logistic regression predicting successful goal attainment

(1 = successfully attained their goal, 0 = otherwise) from

three dummy variables representing the conditions (easy,

range-easy, and hard), with the reserve condition as the

reference group.4

A test of the full model against a constant-only model was

statistically signiﬁcant, indicating that our predictors as a set

(indicators of different goal types) had a signiﬁcanteffecton

goal attainment (c

(4, N = 226) = 14.76, p<.001). Reserve

participants (52.5%) were signiﬁcantly more likely to re-

ceive their bonus than easy participants (25.9%; b=−1.15,

(1) = 8.07, p= .005), range-easy participants (33.9%;

b=−.77, c

(1) = 4.00, p= .046), and hard participants

(21.1%; b=−.1.42, c

(1) = 11.68, p= .001). Because hard-

goal participants did not receive their bonus unless they

completed the task seven days a week (more than all of the

other conditions), we ran another logistic regression pre-

dicting achievement of the easier reference point goal of ﬁve

days a week from the same three dummy variables repre-

senting each condition. Reserve participants were still sig-

niﬁcantly more likely to complete this lower threshold than

hard participants (52.5%

reserve

vs. 31.6%

hard

;b=−.88,

(1) = 5.13, p= .023; see Figure 3).

Preference. On the last day of the task, all of the conditions

were described, and participants (84 total) were asked to in-

dicate which condition they would prefer if they had to

complete the task again. Using a multinomial logistic re-

gression, we found that participants chose the reserve option

Figure 2

PROPORTION OF PARTICIPANTS WHO PREFERRED THE RESERVE GOAL, CONSIDERED IT TO BE MORE ATTAINABLE, AND

CONSIDERED IT TO HAVE MORE VALUE VERSUS THE OTHER GOAL OPTION

Versus Easy Versus Range-Easy Versus Range-Hard Versus Hard

Proportion of Participants Who

Chose the Reserve Goal

Preference Attainability Value

4The effects reported next are the betas and p-values for each of these

dummy variables from one regression model (not individual pairwise

comparisons).

502 JOURNAL OF MARKETING RESEARCH, JUNE 2017

signiﬁcantly more than each of the other options (62%

reserve

vs. 14%

easy

, 14%

hard

, and 10%

range-easy

;ps<.001).

Discussion

This study provides evidence that consumers with emer-

gency reserves persist more than those with other goals (easy,

range-easy, and hard). Reserve participants were more likely

to receive their bonus and reach the easier goal compared

with participants in all other conditions.

We also found that participants prefer the reserve con-

dition to all other conditions after actually experiencing the

task. There is a superordinate goal in this study to make as

much money as possible because participants could make $1

per day in all conditions. Therefore, this result is consistent

with the preferences expressed in Studies 1–3 and with the

predictions in H

Although this study provides evidence for the basic effect

(i.e., consumers with reserve goals persist more thanthose with

other goals), it does not demonstrate why consumers with

reserve goals persist more. According to expectancy-value

theories, reserve goals may provide more motivation by fo-

cusing consumers on the value (i.e., the more difﬁcult goal;

e.g., complete the task for seven days) while maintaining the

expectancy (by providing a reserve); under this explanation,

the cost of the reserve is an unnecessary part of the process.

To explore whether the cost of the reserve is a critical compo-

nent of the process, we next examine participants’persistence

with a range-hard goal, which similarly focuses on a more

difﬁcult, aspirational goal while keeping the same attain-

ability as an easy goal. We expect that if the cost of the reserve

is an unnecessary part of the motivational process, partici-

pants with range-hard goals will persist equally as much as

those with reserve goals. However, if participants with re-

serve goals persist more, we can conclude that there is an

additional motivational beneﬁt of the cost of the reserve

beyond what expectancy-value theories would predict.

In addition, we examine participants’persistence with a

psychological cost-free reserve by terming it a “bonus”reserve

rather than an “emergency”reserve. Although “bonus”re-

serves may still have an opportunity cost associated with using

them, the psychological cost of the reserve is reduced.

Therefore, if cost is an important motivational component of

the reserve, participants with “emergency reserves”should

persist more than those with “bonus reserves”as a result of

the increased cost (emergency reserve: psychological cost +

opportunity cost vs. bonus reserve: opportunity cost).

STUDY 5: SPOT-THE-DIFFERENCE STUDY

In Study 5, participants were asked to complete three spot-

the-difference “training”gamestohelpthempreparefora

more difﬁcult spot-the-difference test at the end of the survey.

Participants were randomly assigned to one of six training

goals to beat a particular number of the three spot-the-difference

training games: easy, range-easy, range-hard, emergency re-

serve, bonus reserve, or hard.

The ﬁrst game was very easy (leading all participants to

score one point), but the second game in the training set was

very difﬁcult. The number of differences participants found

before giving up on this difﬁcult game was our dependent

variable of interest and measure of persistence. We expected

that participants with an emergency reserve would ﬁnd more

differences before giving up than those with other types of

goals, as a result of the cost of reserve. As in Study 4, in this

study, there were two costs of using their emergency reserve:

a psychological cost (i.e., labeling it “emergency”)andan

opportunity cost (i.e., if participants use their emergency re-

serve on the second game, they cannot use it on the third

game).

Procedure

Six hundred one paid MTurk participants (M

age

=35.71

years, age range: 18–78 years; 211 men) completed this study.

The experiment began by explaining to participants that

theywouldbecompletingtrainingforadifﬁcult spot-the-

difference test at the end of the study. If they performed

well on this test, they could qualify for a potential survey in

the future. To train for this test, they would be asked to

complete a series of training spot-the-difference games. The

more spot-the-difference games they practiced, the more

likely it is that they would do better on the test. Thus,

performing well on this difﬁcult spot-the-difference test was

the superordinate goal. Only participants who successfully

completed the training could try to take the difﬁcult spot-the-

difference test and thus be eligible for a potential survey in

the future. Before participants were assigned their goal, they

received instructions and completed a practice spot-the-

difference task.

Participants were then randomly assigned to one of the

training goal conditions (easy, range-easy, range-hard, hard,

emergency reserve, or bonus reserve), in which we manipu-

lated the goal for the number of spot-the-difference games they

should complete (out of three). They were told that they would

receive one point for every spot-the-difference game that they

won. In the easy condition, participants’goal was to score two

points. In the range-easy condition, participants were told,

“Your goal is to score 2 points. However, you should aim to

score 3 points. You will be able to try the spot-the-difference

test if you score 2 points.”In the range-hard condition,

Figure 3

PROPORTION OF PARTICIPANTS WHO REACHED THE FIVE-DAY

GOAL IN EACH CONDITION

Easy Ran

e-Easy Reserve Hard

Proportion Who Completed

Five-Day Goal

5Only those participants who completed the seventh day completed the

preference measure. Thus, our preference results from this study may be not

representative of the full sample of participants.

The Beneﬁts of Emergency Reserves 503

participants were told, “Your goal is to score 3 points.

However, it's okay if you score 2 points instead. You will be

able to try the spot-the-difference test if you score 2 points.”

In the hard condition, participants’goal was to score three

points. In the emergency reserve and bonus reserve condi-

tions, participants’goal was also to score three points.

However, participants in the emergency reserve condition

were also told, “Throughout these games, you will have one

optional ‘emergency’point available just in case you need it.

If you fail one spot the difference game, you can apply this

emergency point and receive a point for that failed game.”

Afterward, they saw a graphical representation of the emer-

gency reserve with a red “Apply Emergency Point”button.

They were informed that if they failed a game, they would

click on this button to apply their emergency point. This

description was exactly the same for the bonus reserve

condition, except the word “emergency”was replaced with

the word “bonus.”In addition, the graphical representation

of the reserve button was green rather than red and said

“Apply Bonus Point”rather than “Apply Emergency Point”

(see ﬁgure 4).”

After every spot-the-difference game, participants were

shown a graphical depiction of their progress and their goal.

To focus attention on the easier goal in the range-easy and

easy conditions and the harder goal in the range-hard,

emergency reserve, bonus reserve, and hard conditions, the

graphical representation for these goals showed a goal of

scoring two points and three points, respectively.

All participants then completed the very easy ﬁrst spot-the-

difference game. Participants were asked to ﬁnd two differ-

ences using as much time as they wanted; all participants were

informed that they had beaten the game.

For the second spot-the-difference game, participants

were told that there were between 10 and 12 differences

(there were actually only 10 differences) between the two

pictures (see Figure 5). They had as much time as they

needed, but they were informed that they could give up at any

time. The number of differences that participants found be-

fore giving up was our dependent variable of interest. After

participants gave up (or thought they found all the differ-

ences), they were informed that they either beat the game or

failed the game and thus did/did not receive a point. If they

failed this game, participants in the reserve conditions were

asked if they wanted to apply their “emergency point”or

“bonus point.”

After the second game, participants completed the third

spot-the-difference game. All participants then took the dif-

ﬁcult spot-the-difference test.6They were given two minutes to

ﬁnd as many differences as they could. After taking this test,

participants were asked to imagine that they had to complete a

similar set oftraining tasks again. They werepresented withall

of the different training goals with matching graphical de-

pictions and were asked to choose between them.

Results

Number of differences found. We used ordinary least

squares regression to predict the number of differences found

from ﬁve dummy variables representing each condition

(easy, range-easy, range-hard, hard, and bonus reserve), with

the emergency reserve condition as the reference group (see

Figure 6).7A test of the full model against a constant-only

model was marginally statistically signiﬁcant (F(5, 595) =

2.209, p= .052). Participants in the emergency reserve

condition (M = 6.98) found signiﬁcantly more differences

than those in the easy condition (M = 5.93; b=−1.05, p=

.004), the range-easy condition (M = 6.13; b=−.85, p= .020),

the range-hard condition (M = 6.21; b=−.77, p= .034), and

the hard condition (M = 5.98; b=−1.00, p= .006); emer-

gency reserve participants found marginally signiﬁcantly

more differences than those in the bonus reserve condition

e-reserve

= 6.98 vs. M

b-reserve

= 6.34; b=−.64, p= .076).

Preference. We conducted a multinomial logistic re-

gression with the emergency reserve option as the reference

group and found that participants preferred the emergency

reserve goal (14.3%) signiﬁcantly more than the hard (8.2%,

p= .002) and easy (7.8%, p= .001) goals and signiﬁcantly

less than the range-easy goal (21.1%, p= .005). Participants

had no signiﬁcant preference between the emergency reserve

goal and the range-hard goal (17.3%, p= .19). In addition,

we conducted a multinomial logistic regression with the

bonus reserve goal as the reference group and found that

participants preferred the bonus reserve goal signiﬁcantly

more than all of the other goals (31.3%

b-reserve

vs. 8.2%

hard

p<.001; 7.8%

easy

,p<.001; 21.1%

range-easy

,p= .001;

17.3%

range-hard

,p<.001; 14.3%

e-reserve

,p<.001).

Discussion

Participants with an emergency reserve goal found more

differences (i.e., tried harder) on the second spot-the-difference

game than those with all other goal types. Participants in all

conditions (except the hard condition) could try the ﬁnal spot-

the-difference test if they scored two points. All participants

scored one point on the ﬁrst game, so if participants with these

goals gave up on the second difﬁcult game, they could still

qualify to take the spot-the-difference test by completing the

third game. Despite this similarity among the goals, partici-

pants with an emergency reserve goal tried the hardest on this

second game, resisting using their emergency reserve. In ad-

dition, participants with emergency reserves persisted more on

the second game than even those with a hard goal.

In this study, we also examined the persistence of partici-

pants with a range-hard goal, a range goal that focused on the

aspirational goal, to investigate how crucial the cost of the

reserve is for increased persistence. We found that participants

with an emergency reserve goal persisted more than those

with a range-hard goal. Relatedly, we introduced a psycho-

logical cost-free bonus reserve goal. We also found that par-

ticipants persisted more with an emergency reserve goal, a

reserve with more cost (psychological + opportunity costs),

than with a psychological cost–free bonus reserve, a reserve

with only an opportunity cost. Both of these comparisons

suggest that the cost of the reserve is a crucial component of the

mechanism of the emergency reserve.

After experiencing the task, participants preferred the

emergency reserve goal to the hard goal and easy goals and

6Even participants who failed their training goal were asked to take the

spot-the-difference test. However, participants who failed their training goal

were told that they would not qualify for future surveys regardless of their

performance.

7The effects reported next are the betas and p-values for each of these

dummy variables from one regression model (not individual pairwise

comparisons).

504 JOURNAL OF MARKETING RESEARCH, JUNE 2017

preferred the bonus reserve goal the most out of all goals.

Because participants had to choose between a bonus reserve

goal and an emergency reserve goal, we expect that they may

have shied away from the greater psychological cost of the

emergency reserve goal. Nonetheless, this study provides

overall evidence that there is strong preference for having a

reserve of some kind as part of a difﬁcult goal.

In this study, we demonstrated that participants try to avoid

using their reserve on game 2 when they have an opportunity to

use it later (e.g., on game 3). In Study 6, we examine whether

participants try to avoid using their reserve even if there is no

opportunity to use it later. Thus, this ﬁnal study examines how

effective the reserve is when it has only a psychological cost

(by labeling it “emergency”) rather than when it has both a

psychological cost and an opportunity cost, as in Studies 4 and

5. Speciﬁcally, we examine how likely participants with a

reserve are to persist on a third game even after succeeding at

the easier goal (e.g., scoring two points out of three).

In addition, in Study 5, participants had to successfully

complete a training goal to take a test at the end of the survey.

In the real world, there is not always a strict qualiﬁcation for

achieving a superordinate goal. People often set goals (e.g., go

to the gym seven days a week) to help them be more likely

to reach their superordinate goal (e.g., become more ﬁt).

However, people can often still reach the superordinate goal

even if they fail the short-term goal (e.g., they can still lose

some weight if they go to the gym only six days a week

instead of seven). Therefore, in Study 6, we relax the linkage

between the goals to be more aligned with these types of real-

world scenarios.

STUDY 6: WORD SEARCH PERSISTENCE STUDY

In Study 6, participants were asked to complete three word

search “training”games to help them prepare for a more

difﬁcult word search test at the end of the survey. Participants

were randomly assigned to one of ﬁve training goals to beat a

particular number of the three word search training games:

easy, range-easy, range-hard, emergency reserve, or hard.

Participants completed two word search practice games that

they were likely to succeed at and then were asked whether

they would like to try the third word search practice game or

move on to the word search test; this is our dependent variable

and measure of persistence. We expected that participants with

reserve goals would be more likely to try the third word search

practice game than those with other goals.

Procedure

Five hundred ten participants (M

age

= 35.85 years, age

range: 18–77 years; 212 men) completed this survey on

MTurk. The experiment began by explaining to participants

that they would be completing training for a difﬁcult word

search test at the end of the study. If they performed well on

this word search test, they could be eligible for a potential

survey in the future (the superordinate goal). To train for this

word search test, they would be asked to complete a series of

training word searches. They were told that the more word

searches they practiced, the more likely it is that they would

do better on the difﬁcult word search test. Participants did not

have to successfully complete their training goal to try the

word search test. However, they were told that completing

their goal was a good indication that they were prepared for

the test. Participants received instructions and completed a

practice word search.

Participants were then randomly assigned to one of the

training goalconditions (easy, range-easy, range-hard, hard, or

reserve), in which the goal for how many word searches they

should complete (out of three) was manipulated. They were

told they would receive one point for every word search that

they beat. In the easy condition, participants’goal was to score

Figure 5

SECOND SPOT-THE-DIFFERENCE GAME

Figure 4

IMAGE FOR BONUS RESERVE VERSUS EMERGENCY RESERVE

Your Progress Your Goal

Apply

Emergency Point

Your Progress Your Goal

Apply Bonus

Point

Emergency Reserve Condition Bonus Reserve Condition

The Beneﬁts of Emergency Reserves 505

two points. In the range-easy condition, participants were

told, “Your goal is to score 2 points. However, you should

aim to score 3 points.”In the range-hard condition, participants

were told, “Your goal is to score 3 points. However, it's okay if

you score 2 points instead.”In the hard and reserve conditions,

participants’goal was to score three points. However, partic-

ipants in the reserve condition were also told, “Throughout

these games, you will have one ‘emergency’point available. If

you fail to complete one word search, you can apply this

emergency point. This emergency point does not need to be

used. It should only be used if you need it.”As in Study 5, after

every word search, participants were shown a graphical de-

piction of their progress and their goal.

All participants then completed the very easy ﬁrst word

search. Participants were asked to ﬁnd four words in three

minutes; all participants were informed that they beat the word

search. Participants next completed a slightly more difﬁcult (but

still easy) word search, in which they had to ﬁnd ﬁve words in

three and a half minutes. Most participants succeeded at both of

these timed word search practice games and thus had scored two

points by this point in the study.

After succeeding at the second word search, participants

were then given a description of the third word search. The

third word search did not have a time limit and required

participants to ﬁnd ten words. Without seeing the exact word

search, they were asked, “Do you want to try the 3rd word

search game described or would you like to move on to the

word search test?”If participants chose to move on to the

word search test, they were asked to conﬁrm that they indeed

wanted to move on and, in the reserve condition, that they

wanted to use their emergency point.

All participants then took the difﬁcult word search test.

They were given three minutes to ﬁnd as many of the ten

words listed as possible. Immediately before and after taking

this test, participants were asked to imagine that they had to

complete a similar training session again and were asked to

choose between the different training goals.

Results

Twenty-nine participants failed one of the ﬁrst two word

searches and were excluded from further analyses. We

conducted a logistic regression predicting choice toattempt the

third word search (1 = Tried the third word search, 0 = Did not

try the third word search) from four dummy variables repre-

senting each of the conditions (easy, range-easy, range-hard,

and hard), with the reserve condition serving as the reference

group. We also conducted an additional logistic regression

predicting successful completion of the third word search

(1 = Successfully completedthe third word search, 0 = Did not

successfully complete the third word search) from the same

four dummy variables.8

Try third word search. A test of the full model against a

constant-only model was statistically signiﬁcant (c

(4, N =

481) = 29.61, p<.001). We found that participants with re-

serves (86.3%) were more likely to try the third word search

compared with those in the easy condition (51.5%; b=−1.78,

(1) = 24.30, p<.001), the range-easy condition (73.4%;

b=−.83, c

(1) = 4.76, p= .029), the range-hard condition

(68%; b=−1.09, c

(1) = 8.76, p= .003), and the hard

condition (73.7%; b=−.81, c

(1) = 4.60, p= .032) (see

Figure 7).

Figure 6

NUMBER OF DIFFERENCES FOUND IN THE SECOND SPOT-THE-DIFFERENCE GAME, SPLIT BY CONDITION

Easy Range-Easy Range-Hard Emergency

Reserve

Bonus Reserve Hard

Total Differences Found

8The effects reported next are the betas and p-values for each of these

dummy variables from one regression model (not individual pairwise

comparisons).

506 JOURNAL OF MARKETING RESEARCH, JUNE 2017

Score three points. A test of the full model against a

constant-only model was statistically signiﬁcant (c

(4, N =

481) = 25.74, p<.001). We found that participants with

reserves (81.1%) were more likely to score three points

(complete the third word search game) than those in the easy

condition (46.4%; b=−1.60, c

(1) = 23.21, p<.001), the

range-easy condition (77.1%; b=−.89, c

(1) = 6.84, p=

.009), the range-hard condition (63%; b=−.92, c

(1) = 7.62,

p= .006), and the hard condition (64.2%; b=−.87, c

(1) =

6.61, p= .010).

Preference. We used a multinomial regression to analyze

the preference data, with the reserve option as the refer-

ence group. Participants preferred the reserve goal signiﬁ-

cantly more than the easy goal before the word search test

(22.2%

reserve

vs. 10.1%

easy

;b=−.76, c

(1) = 19.72, p<.001)

and marginally signiﬁcantly more than the easy goal after

the word search test (19.3%

reserve

vs. 14.8%

easy

;b=−.27,

(1) = 2.93, p= .087). However, they preferred the reserve

goal less than the hard goal (22.2%

reserve

vs. 29.5%

hard

;

b= .28, c

(1) = 4.89, p= .027), and there were no differences

in preference for the reserve goal and the other range goals.

Discussion

We found that participants with reserve goals were more

likely to persist to try the third word search compared with

participants with other goals when there was only a psycho-

logical cost associated with using the reserve. Participants tried

to resist using the emergency reserve even when there was no

future beneﬁt to saving it, leading them to persist after reach-

ing the easier goal (scoring two points). Replicating Study 5,

participants with reserve goals again persisted more than those

with range-hard goals, suggesting that the psychological cost

of the reserve goal is a crucial component of its ability to

motivate. They also persisted more than those with hard goals.

After experiencing the task, we found that participants

preferred the reserve goal to the easy goal but not to the hard

goal. In Study 5, we found that participants preferred reserve

goals to both easy and hard goals. This difference may be due

to the fact that the training goals overall were much easier in

Study 6 than in Study 5. In this study, most participants

succeeded at the second word search and did not apply their

emergency reserve, whereas in Study 5, most participants

failed to complete the second spot-the-difference game,

making the emergency reserve more necessary and valuable.

Thus, preferences for the reserve goal to other goals may

depend on how difﬁcult the task is, with reserves becoming

more preferred when the task involves some level of per-

ceived difﬁculty.

GENERAL DISCUSSION

This article demonstrates that the emergency reserve is

both preferred in goal pursuit and highly motivating, leading

to more persistence than other goals. Our results suggest that

offering emergency reserves can not only encourage con-

sumers to sign up for a program initially but also help them

reach desirable goals and keep them satisﬁed with their

outcomes.

We support these claims through six studies. Study 1

demonstrated that consumers prefer weight loss programs (a

program with a superordinate goal) with emergency reserves,

and Study 2 revealed that the emergency reserve is preferred to

hard and easy options when there is a superordinate goal and is

preferred to hard options (but not easy options) when there is

no superordinate goal. Study 3 further showed that consumers

prefer reserve goals to easy and hard goals when there is a

superordinate goal because they perceive both the attainabil-

ity and value to be greater. Study 4 demonstrated that con-

sumers persist more than those with easy and hard goals in an

incentive-compatible real-life task. Finally, Studies 5 and 6

demonstrated that consumers persist more with reserve goals

compared with other goal types because they try to avoid using

the reserve. This research contributes to the literature on goals

by suggesting an innovative strategy to not only initiate goal

pursuit but also improve goal persistence. Our ﬁndings suggest

that predeﬁned ﬂexibility with a cost can actually be beneﬁcial,

rather than maladaptive, in goals and mental budgets.

We note that programs outside of the lab environment that

attempt to help people reach superordinate goals use concepts

similar to the emergency reserve. For example, Weight

Watchers, a very successful point-based weight loss program,9

gives its participants weekly “optional points”and “activity

points”that they can use any day throughout the week, which

they can earn by doing physical activity. Although the points

are available if needed, there is an opportunity cost associated

with using the optional points and a future/past cost associated

with the activity points. In a separate domain, giving children a

“bedtime pass”that allows them to leave their bedroom only

once per night has been shown to be an effective way to get

children to go to sleep at bedtime (Friman et al. 1999; Moore

et al. 2007). The success of these programs suggests that

including emergency reserves can be both a preferred and a

successful strategy to help consumers reach their goals. Our

research also suggests that labeling the optional points and

Figure 7

PROPORTION OF PARTICIPANTS WHO TRIED THE THIRD WORD

SEARCH PRACTICE GAME, SPLIT BY CONDITION

Easy Range-Easy Range-Hard Reserve Hard

Proportion Who Tried Third Word-Search Game

9According to U.S. News and World Report, in 2015 Weight Watchers was

ranked number one for the ﬁfth consecutive year for “Best Weight Loss Diet”

and ranked number one as “Easiest Diet to Follow”for the fourth consecutive

year.

The Beneﬁts of Emergency Reserves 507

bedtime pass as for “emergency use”only may further increase

their effectiveness.

Implications for Marketers

This set of studies suggests that marketers can be more

successful in recruiting consumers to sign up for their

programs by offering an emergency reserve. Programs fo-

cusing on helping consumers reach superordinate goals (e.g.,

weight loss programs) can more easily encourage consumers

to sign up for their programs by offering a reserve. However,

even companies that may not necessarily focus on super-

ordinate goals can encourage consumers to choose programs

with a reserve if they make a superordinate goal more salient.

For example, a phone company could emphasize the costs

of exceeding a consumer’s allocated data per month (a su-

perordinate goal). Making the overage costs salient should

highlight the importance of having a plan with less data plus

an emergency buffer compared with simply having more data

(with the buffer already included).

Our results also reveal that people persist more if they

have reserve goals compared with other goal types. This

increased persistence will lead to better long-term results

and, thus, more satisﬁed consumers. Furthermore, Studies 4

and 5 revealed that participants prefer reserve goals to both

easy goals and hard goals after experiencing a difﬁcult task,

suggesting that consumers are likely to be more satisﬁed with

programs with emergency reserves. Therefore, emergency

reserves have direct implications for marketers because they

are a method to encourage consumers to sign up for a pro-

gram, help them reach their long-term goals, and lead to their

higher satisfaction and return.

Further Research

It is worth noting that we explored a speciﬁc type of

persistence in this research. We focused on situations in

which participants succeeded at their goal or could still

succeed at their goal in the future (i.e., persistence before

failure). Thus, this article did not explore how goal pursuit is

affected if consumers fail at a subgoal and actually need to

use their emergency reserves. Prior research has shown that

violating a goal or subgoal can have negative consequences

(Cochran and Tesser 1996; Fishbach, Dhar, and Zhang

2006; Heath, Larrick, and Wu 1999; Polivy 1976; Soman

and Cheema 2004; Wilcox, Block, and Eisenstein 2011),

such as a deterioration of subsequent performance or complete

abandonment of the goal. In addition to helping consumers

persist to reach a more difﬁcult goal/reference point, emer-

gency reserves may have another distinct beneﬁt not explored

in this article: greater persistence after a subgoal or goal

violation. By applying emergency reserves after a subgoal

or goal violation, some negative consequences of goal and/or

subgoal violation may be reduced, leading consumers with

goals with emergency reserves to continue to persist even

after small failures.

In this article, we focused only on persistence in tasks

with imposed extrinsic superordinate goals (e.g., perform

well on the ﬁnal word search test). Research has shown that

success versus failure of a subgoal has different effects de-

pending on whether the person focuses on a superordinate

goal (Fishbach, Dhar, and Zhang 2006). Success of a sub-

goal highlights commitment rather than progress when the

superordinate goal is primed, leading people to continue to

pursue the goal, rather than shifting their attention to other

goals (Fishbach and Dhar 2005; Fishbach, Dhar, and Zhang

2006). Furthermore, consumers may have more motivation to

resist using their emergency reserves when doing so helps

them reach their superordinate goal (e.g., not using two

“emergency skip”days from the gym helps a person become

more ﬁt.) Relatedly, superordinate goals that are intrinsically,

rather than extrinsically, motivated have been shown to affect

people’s behavior in pursuing goals (Wang and Mukhopadhyay

2012). Further research should explore whether reserve goals

are as effective with an intrinsic superordinate goal or in the

absence of a superordinate goal.

Emergency reserves may differ in size and type, which

may affect how people use them. If the reserve is too small, it

may not provide as much of a needed buffer. However, if it

is too big, people may no longer mentally encode it as a

reserve and instead start incorporating it into their goal,

encoding it as an easy goal instead. In addition, rather than

being consumed all at once or not at all (as in our studies),

emergency reserves can also be continuous. For example,

in a dieting context, people could have a supply of 1,000

emergency calories. Unlike the all-or-nothing reserves, peo-

ple may choose to use some of the reserve, but not all of it, at

a given time. Exploring and understanding more about these

intricacies of emergency reserves will help us design the

most optimal emergency reserves.

Conclusion

This article provides an innovative strategy of structuring

goals to provide a sense of ﬂexibility while still maintaining

stringency through the addition of an emergency reserve. The

emergency reserve can be applied to a variety of different

goals, such as saving money in the ﬁnancial domain, studying

for a test in the education domain, or trying to lose weight in

the food/exercise domain. These are long-term goals that

people consistently have trouble achieving. By exploring

more about the mechanisms and applications of emergency

reserves, we can understand more generally about how to

help consumers initially pursue their goals as well as succeed

at many of the long-term goals that they have been struggling

to achieve.

REFERENCES

Ainslie, George (2001), Breakdown of Will. Cambridge, UK:

Cambridge University Press.

Atkinson, John W. (1957), “Motivational Determinants of Risk-

Taking Behavior,”Psychological Review, 64 (6, Part 1), 359–72.

Bandura, Albert (1997), Self-Efﬁcacy: The Exercise of Control. New

York: Freeman.

Cheema, Amar, and Dilip Soman (2006), “Malleable Mental Ac-

counting: The Effect of Flexibility on the Justiﬁcation of

Attractive Spending and Consumption Decisions,”Journal of

Consumer Psychology,16(1),33–44.

Cochran, Winona, and Abraham Tesser (1996), “The ‘What the

Hell’Effect: Some Effects of Goal Proximity and Goal Framing

on Performance,”in Striving and Feeling: Interactions Among

Goals, Affect, and Self-Regulation, Leonard L. Martin and

Abraham Tesser, eds. New York: Patent, 99–120.

Etkin, Jordan, and Rebeca K. Ratner (2012), “The Dynamic Impact

of Variety Among Means on Motivation,”Journal of Consumer

Research, 38 (6), 1076–92.

508 JOURNAL OF MARKETING RESEARCH, JUNE 2017

Fishbach, Ayelet, and Ravi Dhar (2005), “Goals as Excuses or

Guides: The Liberating Effect of Perceived Goal Progress on

Choice,”Journal of Consumer Research, 32 (3), 370–77.

Fishbach, Ayelet, Ravi Dhar, and Ying Zhang (2006), “Subgoals

as Substitutes or Complements: The Role of Goal Accessibil-

ity,”Journal of Personality and Social Psychology,91(2),

232–42.

Friman, Patrick C., Kathryn E. Hoff, Connie Schnoes, Kurt A.

Freeman, Douglas W. Woods, and Nathan Blum (1999), “The

Bedtime Pass: An Approach to Bedtime Crying and Leaving the

Room,”Archives of Pediatrics & Adolescent Medicine, 153 (10),

1027–29.

Gollwitzer, Peter M. (1990), “Action Phases and Mind-Sets,”in

Handbook of Motivation and Cognition: Foundations of Social

Behavior, Vol. 2, E. Tory Higgins and Richard M. Sorrentino,

eds. New York: Guilford Press, 53–92.

Heath, Chip, Richard P. Larrick, and George Wu (1999), “Goals as

Reference Points,”Cognitive Psychology, 38 (1), 79–109.

Heath, Chip, and Jack B. Soll (1996), “Mental Budgeting and

Consumer Decisions,”Journal of Consumer Research, 23 (1),

40–52.

Heckhausen, Heinz (1977), “Achievement Motivation and Its

Constructs: A Cognitive Model,”Motivation and Emotion, 1 (4),

283–329.

Heckhausen, Heinz, and Peter M. Gollwitzer (1987), “Thought

Contents and Cognitive Functioning in Motivational Versus

Volitional States of Mind,”Motivation and Emotion, 11 (2),

101–20.

Henderson, Pamela W., and Robert A. Peterson (1992), “Mental

Accounting and Categorization,”Organizational Behavior and

Human Decision Processes, 51 (1), 92–117.

Huang, Szu-Chi, and Ying Zhang (2011), “Motivational Conse-

quences of Perceived Velocity in Consumer Goal Pursuit,”

Journal of Marketing Research, 48 (December), 1045–56.

Huang, Szu-Chi, and Ying Zhang (2013), “All Roads Lead to Rome:

The Impact of Multiple Attainment Means on Motivation,”

Journal of Personality and Social Psychology, 104 (2), 236–48.

Kahneman, Daniel, and Amos Tversky (1979), “Prospect Theory:

An Analysis of Decisions Under Risk,”Econometrica, 47 (2),

263–91.

Kahneman, Daniel, and Amos Tversky (1984), “Choices, Values,

and Frames,”American Psychologist, 39 (4), 341–50.

Krishnamurthy, Parthasarathy, and Sonja Prokopec (2010), “Re-

sisting That Triple-Chocolate Cake: Mental Budgets and Self-

Control,”Journal of Consumer Research, 37 (1), 68–79.

Kruglanski, Arie W., Antonio Pierro, and Anna Sheveland (2011),

“How Many Roads Lead to Rome? Equiﬁnality Set-Size and

Commitment to Goals and Means,”European Journal of Social

Psychology, 41 (3), 344–52.

Locke, Edwin A., and Gary P. Latham (1990), A Theory of Goal

Setting & Task Performance. Englewood Cliffs, NJ: Prentice

Hall.

Locke, Edwin A., Karyll N. Shaw, Lise M. Saari, and Gary P.

Latham (1981), “Goal Setting and Task Performance: 1969–1980,”

Psychological Bulletin,90(1),125–52.

Milkman, Katherine L., Julia A. Minson, and Kevin G.M. Volpp

(2013), “Holding the Hunger Games Hostage at the Gym: An

Evaluation of Temptation Bundling,”Management Science,

60 (2), 283–99.

Moore, Brie A., Patrick C. Friman, Alan E. Fruzzetti, and Ken

MacAleese (2007), “Brief Report: Evaluating the Bedtime Pass

Program for Child Resistance to Bedtime—A Randomized,

Controlled Trial,”Journal of Pediatric Psychology, 32 (3),

283–87.

Morewedge, Carey K., Leif Holtzman, and Nicholas Epley (2007),

“Unﬁxed Resources: Perceived Costs, Consumption, and the

Accessible Account Effect,”Journal of Consumer Research,

34 (4), 459–67.

Polivy, Janet (1976), “Perception of Calories and Regulation of

Intake in Restrained and Unrestrained Subjects,”Addictive Be-

haviors, 1 (3), 237–43.

Scott, Maura L., and Scott M. Nowlis (2013), “The Effect of Goal

Speciﬁcity on Consumer Goal Reengagement,”Journal of Con-

sumer Research, 40 (3), 444–59.

Shefrin, Hersh M., and Richard H. Thaler (1988), “The Behavioral

Life Cycle Hypothesis,”Economic Inquiry, 26 (4), 609–44.

Shin, Jihae, and Katherine L. Milkman (2016), “How Backup Plans

Can Harm Goal Pursuit: The Unexpected Downside of Being

Prepared for Failure,”Organizational Behavior and Human

Decision Processes, 135, 1–9.

Soman, Dilip, and Amar Cheema (2004), “When Goals Are

Counterproductive: The Effects of Violation of a Behavioral Goal

on Subsequent Performance,”Journal of Consumer Research,

31 (1), 52–62.

Soman, Dilip, and Amar Cheema (2011), “Earmarking and Parti-

tioning: Increasing Saving by Low-Income Households,”Journal

of Marketing Research, 48 (Special Issue), S14–22.

Spiller, Stephen A. (2011), “Opportunity Cost Consideration,”

Journal of Consumer Research, 38 (4), 595–610.

Stilley, Karen M., J. Jeffrey Inman, and Kirk L. Wakeﬁeld (2010),

“Planning to Make Unplanned Purchases? The Role of In-Store

Slack in Budget Deviation,”Journal of Consumer Research,

37 (2), 264–78.

Tangney, June P., Roy F. Baumeister, and Angie L. Boone (2004),

“High Self-Control Predicts Good Adjustment, Less Pathology,

Better Grades, and Interpersonal Success,”Journal of Person-

ality, 72 (2), 271–324.

Thaler, Richard (1980), “Toward a Positive Theory of Consumer

Choice,”Journal of Economic Behavior & Organization, 1 (1),

39–60.

Thaler, Richard (1985), “Mental Accounting and Consumer Choice,”

Marketing Science, 4 (3), 199–214.

Thaler, Richard (1990), “Anomalies: Saving, Fungibility, and

Mental Accounts,”Journal of Economic Perspectives,4(1),

193–205.

Tolman, Edward C. (1955), “Principles of Performance,”Psycho-

logical Review, 62 (5), 315–26.

Ulk¨umen, Gulden, and Amar Cheema (2011), “Framing Goals to

Inﬂuence Personal Savings: The Role of Speciﬁcity and Construal

Level,”Journal of Marketing Research, 48 (December), 958–69.

Wang, Chen, and Anirban Mukhopadhyay (2012), “The Dynamics

of Goal Revision: A Cybernetic Multiperiod Test-Operate-Test-

Adjust-Loop (Total) Model of Self-Regulation,”Journal of

Consumer Research, 38 (5), 815–32.

Wertenbroch, Klaus (1998), “Consumption Self-Control by Ra-

tioning Purchase Quantities of Virtue and Vice,”Marketing

Science, 17 (4), 317–37.

Wilcox, Keith, Lauren G. Block, and Eric M. Eisenstein (2011),

“Leave Home Without It? The Effects of Credit Card Debt and

Available Credit on Spending,”Journal of Marketing Research,

48 (Special Issue), S78–90.

Zauberman, Gal, and John G. Lynch Jr. (2005), “Resource Slack

and Propensity to Discount Delayed Investments of Time Versus

Money,”Journal of Experimental Psychology. General, 134 (1),

23–37.

Zhang, Ying, and Szu-Chi Huang (2010), “How Endowed Versus

Earned Progress Affects Consumer Goal Commitment and

Motivation,”Journal of Consumer Research, 37 (4), 641–54.

The Beneﬁts of Emergency Reserves 509

The Interplay of Mental Budgeting, Self-Control, and Financial Behavior: Implications for Individual Financial Well-Being

Article

Full-text available

May 2024

According to behavioral finance, people are not always rational in managing their finances. The financial stability of a person is directly linked with the spending style and pattern of the individual. This research investigates how mental budgeting influences financial behaviors and altogether financial stability, with a focus on the moderating effect of self-control. To explore this, we studied 294 people to see how mental budgeting affects their financial management and well-being, and how self-control plays a role in this process. A self-administered questionnaire was adapted to gather responses of the employees working in various organizations in Lahore. Smart PLS and SPSS were utilized to analyze and measure the responses collected from respondents. This study found that people are able to better manage their finances when they use mental budgeting. Similarly, there is clear evidence that mental budgeting has an association with financial well-being and financial behavior. The path coefficient uncovers the moderating impact of self-control on financial management behavior and financial wellness. These findings suggest that teaching people about mental budgeting and self-control in financial education programs might help them make better financial decisions. This study adds to our understanding by showing how these factors work together to influence financial behavior, highlighting the need for more tailored financial advice that considers both mental budgeting and self-control. This research helps clarify how people can better manage their finances through both mental strategies and behavioral control, contributing valuable insights into effective financial management practices.

How and When Does a Used (vs. Unused) Account Affect Consumption Behavior?

Article

Full-text available

Jan 2024

How does spending from a used (vs. unused) account affect consumption behavior? An account is used when some resources of that account have been used (e.g.,

90 has been used on a gift card that originally had

100). An account is unused when no resources of that account have been used (e.g., no money has been used on a gift card that has $10). Across seven studies (N = 8,667), we find that people are more likely to spend resources from a used account than otherwise equivalent resources from an unused account. This is because people engage in within-account comparisons, comparing the remaining resources in the account with what the account originally had, leading them to value the remaining resources less in a used account. We demonstrate the robustness of the effect of a used (vs. unused) account across several domains, including gift cards, checking accounts, and credit card reward points. Further, we demonstrate a boundary condition of the effect, revealing that the proportion of the account remaining moderates the subsequent consumption. Lastly, we generalize this effect from consumption to charitable giving. The findings provide insights into how policymakers, companies, and individuals may consider leveraging the perception of an account being used or unused to curb expenses and encourage charitable giving.

Backup Plans: The Impact of Disclosure on Perceptions of Expert Competence

Article

Apr 2025
MANAGE SCI

People often consult experts to provide solutions to their problems. Oftentimes, an initial attempt to solve a problem fails, and these experts (e.g., advisors, consultants, specialists, and service providers) will resort to a backup plan. A backup plan can be presented when it is needed (i.e., after an initial failure) or in advance (i.e., before a potential failure). We examine how the timing of disclosure of a backup plan affects judgments of an expert’s competence. We show that although people expect experts to have a backup plan, they react more negatively if it is presented after (versus before) it is needed, all things equal. This “competence penalty” disappears when people are prompted to think about what an expert would do in case of failure as indeed, an alternative plan is expected. Importantly, our findings indicate that there is no competence penalty for early disclosure of a backup plan, regardless of the outcome of the initial plan: if it has failed, has succeeded, or is still ongoing. However, we do find one exception; early disclosure has a negative impact if multiple backup plans are disclosed while the initial plan is in progress. This paper was accepted by Jack Soll, behavioral economics and decision analysis. Funding: Funding for all studies was provided by the MUMA College of Business, University of South Florida. Supplemental Material: The online appendix and data files are available at https://doi.org/10.1287/mnsc.2023.02615 .

The motivating power of streaks: Increasing persistence is as easy as 1, 2, 3

Article

Mar 2025

Pathways for avoiding self‐sanction: How consumers give themselves a PASS on virtue violations

Article

Jan 2025

Stephanie C. Lin

Despite motivations to see themselves as virtuous, consumers commonly engage in behaviors that are bad for themselves or others, such as eating unhealthy food or refusing prosocial requests. I introduce the Pathways for Avoiding Self‐Sanction (PASS) model, which explains how consumers violate their standards for virtue without self‐sanction. This model posits that consumers have a subjective threshold that they must not cross lest they incur self‐sanction and outlines three main pathways through which consumers succumb to the temptation of bad behaviors without crossing this threshold: the self‐based path, the behavior‐based path, or the threshold‐based path. By drawing on shared psychological processes between self‐control and moral decision‐making, the PASS model organizes self‐sanction avoidance strategies across literature in marketing, psychology, organizational behavior, and behavioral economics, offering a comprehensive and parsimonious view of the mechanisms through which consumers engage in maladaptive behaviors that harm themselves, others, and society.

Range goals as dual reference points

Article

Sep 2024
ORGAN BEHAV HUM DEC

The Role of Conflict Representation in Abstinence Versus Moderation in Self-Control

Article

Full-text available

Feb 2024

Self-control—the prioritization of valued global goals over immediate local rewards—is typically conceptualized and studied as isolated decisions. Goal pursuit, however, generally requires people to make repeated self-control decisions across contexts. We adopt a higher order, strategic level of analysis of self-control and explore, for the first time, people’s preferences for abstinence (a pattern of choices in which one never indulges) versus moderation (a pattern of choices in which one indulges when doing so does not harm one’s goals or even helps promote the pursuit of those goals). To understand when and why people may opt for one over the other, the present work explores one psychological feature that may support these strategy preferences: the representation of self-control conflicts as inherent (i.e., choice options are mutually contradictory) versus situational (i.e., choice options compete for limited resources). We present eight studies in the main text and three in the online Supplemental Materials documenting that people associate inherent and situational conflict representations with abstinence and moderation, respectively. By documenting that strategy preferences may differ as a function of conflict representations, this work questions the assumption of abstinence as the primary indicator of self-control success, raises methodological and conceptual questions about how best to assess these strategy preferences, and calls for greater understanding of self-control as a recurrent decision-making process.

The Making Up for Failure Nudge: Framing Subgoals as Opportunities for Redemption Increases Goal Persistence

Article

Jan 2023

Cancel anytime!: How easy cancellation options enhance purchase intentions for services that require long-term commitments

Article

Nov 2023
J Retailing Consum Serv

A Psychological Account of Co-Holding: Why Some Choose to Simultaneously Save and Borrow

Article

Jan 2023

Motivational Consequences of Perceived Velocity in Consumer Goal Pursuit

Article

Full-text available

Dec 2011

The authors explore the interplay between consumers' progress levels toward attaining a goal and the perceived velocity in progressing toward the goal to determine consumers' motivation for further goal pursuit. The authors propose that when progress toward attaining a goal is low, consumers are primarily concerned about the question "Can I get there?" Thus, a high (vs. low) perceived velocity in progressing suggests greater expectations of goal attainment, resulting in greater motivation for pursuing the goal. However, when consumers have achieved sufficient progress and are approaching the end point, their attainment of the goal is relatively secured, so they become more concerned about the question "When will I get there?" and focus more on whether they are effectively reducing the remaining discrepancy so that they can attain the goal quickly. In this case, a low (vs. high) perceived velocity in progressing elicits greater motivation because it suggests that continued effort is needed to ensure a speedy attainment. Empirical evidence from lab and field experiments supports this hypothesis.

Leave Home Without It? The Effects of Credit Card Debt and Available Credit on Spending

Article

Full-text available

Nov 2011

This research examines how credit card debt affects consumer spending. In five experimental and field studies, the authors demonstrate that outstanding credit card debt increases spending for consumers with high self-control. They also show that this effect can be eliminated by increasing the available credit on the credit card. Thus, when the available credit is low, consumers with greater self-control increase spending, but when the available credit is high, they reduce spending. The results extend the literature on goal violation and self-control and offer insights into consumer decision making and consumption patterns under conditions of debt.

Breakdown of Will

Book

Mar 2001

George Ainslie

Ainslie argues that our responses to the threat of our own inconsistency determine the basic fabric of human culture. He suggests that individuals are more like populations of bargaining agents than like the hierarchical command structures envisaged by cognitive psychologists. The forces that create and constrain these populations help us understand so much that is puzzling in human action and interaction: from addictions and other self-defeating behaviors to the experience of willfulness, from pathological over-control and self-deception to subtler forms of behavior such as altruism, sadism, gambling, and the 'social construction' of belief. This book integrates approaches from experimental psychology, philosophy of mind, microeconomics, and decision science to present one of the most profound and expert accounts of human irrationality available. It will be of great interest to philosophers and an important resource for professionals and students in psychology, economics and political science.

Choices, Values, and Frames

Article

Feb 2019

Anomalies: Saving, Fungibility, and Mental Accounts

Article

Jan 1990
J ECON PERSPECT

R.H. Thaler

How backup plans can harm goal pursuit: The unexpected downside of being prepared for failure

Article

Jul 2016

When pursuing a goal, making a backup plan has many benefits, including reducing the psychological discomfort associated with uncertainty. However, we suggest that making a backup plan can also have negative effects. Specifically, we propose that the mere act of thinking through a backup plan can reduce performance on your primary goal by decreasing your desire for goal achievement. In three experimental studies, we find that individuals randomly assigned to think through a backup plan subsequently performed worse on their primary goal (Studies 1–3). We further show that this effect is mediated by study participants’ decreased desire to attain their primary goal (Study 3). This research provides a fresh perspective on plan-making, highlighting an important yet previously unexplored negative consequence of formulating plans.

Self-Efficacy: The Exercise of Control

Article

Jan 1997
J Cognit Psychother

The Bedtime Pass

Article

Oct 1999
Arch Pediatr Adolesc Med

Treatment of bedtime problems is often impeded by adverse effects. Soporific drugs can have diurnal carryover effects and rebound when they are withdrawn. Letting children sleep with parents can complicate marital relationships and delay development of independent practice of bedtime skills. Ignoring childrens' crying after bedtime can lead to increased crying and coming out from the bedroom.¹- 6 In our study, the bedtime pass reduced the frequency of bedtime problems to zero rates in the 2 primary participants without adverse effects and was rated as an acceptable intervention by parents and practicing pediatricians. The acceptability data from parents are important because parent nonacceptance often leads to treatment noncompliance and a perpetuation of bedtime problems.²,3,5 The acceptability data from pediatricians are important because their ratings of using the bedtime pass and of ignoring were equivalent, whereas parents rated ignoring as unacceptable. The difference in ratings between parents and pediatricians may be due to pediatrician reluctance to fully endorse nonempirically supported interventions. The effectiveness of bedtime procedures involving ignoring are well documented,¹,2,5,6 whereas, to our knowledge, this report is the first documentation of the effects of the pass. Systematic replications of this study may be sufficient to increase acceptance of the pass by pediatricians beyond their acceptance of ignoring.

Consumption Self-Control by Rationing Purchase Quantities of Virtue and Vice

Article

Apr 1998

Consumers' attempts to control their unwanted consumption impulses influence many everyday purchases with broad implications for marketers' pricing policies. Addressing theoreticians and practitioners alike, this paper uses multiple empirical methods to show that consumers voluntarily and strategically ration their purchase quantities of goods that are likely to be consumed on impulse and that therefore may pose self-control problems. For example, many regular smokers buy their cigarettes by the pack, although they could easily afford to buy 10-pack cartons. These smokers knowingly forgo sizable per-unit savings from quantity discounts, which they could realize if they bought cartons; by rationing their purchase quantities, they also self-impose additional transactions costs on marginal consumption, which makes excessive smoking overly difficult and costly. Such strategic self-imposition of constraints is intuitively appealing yet theoretically problematic. The marketing literature lacks operationalizations and empirical tests of such consumption self-control strategies and of their managerial implications. This paper provides experimental evidence of the operation of consumer self-control and empirically illustrates its direct implications for the pricing of consumer goods. Moreover, the paper develops a conceptual framework for the design of empirical tests of such self-imposed constraints on consumption in consumer goods markets. Within matched pairs of products, we distinguish relative “virtue” and “vice” goods whose preference ordering changes with whether consumers evaluate immediate or delayed consumption consequences. For example, ignoring long-term health effects, many smokers prefer regular (relative vice) to light (relative virtue) cigarettes, because they prefer the taste of the former. However, ignoring these short-term taste differences, the same smokers prefer light to regular cigarettes when they consider the long-term health effects of smoking. These preference orders can lead to dynamically inconsistent consumption choices by consumers whose tradeoffs between the immediate and delayed consequences of consumption depend on the time lag between purchase and consumption. This creates a potential self-control problem, because these consumers will be tempted to overconsume the vices they have in stock at home. Purchase quantity rationing helps them solve the self-control problem by limiting their stock and hence their consumption opportunities. Such rationing implies that, per purchase occasion, vice consumers will be less likely than virtue consumers to buy larger quantities in response to unit price reductions such as quantity discounts. We first test this prediction in two laboratory experiments. We then examine the external validity of the results at the retail level with a field survey of quantity discounts and with a scanner data analysis of chain-wide store-level demand across a variety of different pairs of matched vice (regular) and virtue (reduced fat, calorie, or caffeine, etc.) product categories. The analyses of these experimental, field, and scanner data provide strong convergent evidence of a characteristic crossover in demand schedules for relative vices and virtues for categories as diverse as, among others, potato chips, chocolate chip cookies, cream cheese, beer, soft drinks, ice cream and frozen yogurt, chewing gum, coffee, and beef and turkey bologna. Vice consumers' demand increases less in response to price reductions than virtue consumers' demand, although their preferences are not generally weaker for vices than for virtues. Constraints on vice purchases are self-imposed and strategic rather than driven by simple preferences. We suggest that rationing their vice inventories at the point of purchase allows consumers to limit subsequent consumption. As a result of purchase quantity rationing, however, vice buyers forgo savings from price reductions through quantity discounts, effectively paying price premiums for the opportunity to engage in self-control. Thus, purchase quantity rationing vice consumers are relatively price insensitive. From a managerial and public policy perspective, our findings should offer marketing practitioners in many consumer goods industries new opportunities to increase profits through segmentation and price discrimination based on consumer self-control. They can charge premium prices for small sizes of vices, relative to the corresponding quantity discounts for virtues. Virtue consumers, on the other hand, will buy larger amounts even when quantity discounts are relatively shallow. A key conceptual contribution of this paper lies in showing how marketing researchers can investigate a whole class of strategic self-constraining consumer behaviors empirically. Moreover, this research is the first to extend previous, theoretical work on impulse control by empirically demonstrating its broader implications for marketing decision making.

The Effect of Goal Specificity on Consumer Goal Reengagement

Article

Oct 2013

Consumers often need to decide if they want to reengage a goal, such as a goal of losing weight, a goal of saving money, or a goal of performing well on a video game. This research finds that consumers are more likely to reengage a goal when they have set a high-low range goal (e.g., lose 2–4 pounds this week) than when they have set a single number goal (e.g., lose 3 pounds this week). This effect is driven by the greater attainability and greater challenge of the high-low range goal, which then leads to a greater feeling of accomplishment. Thus, these findings suggest that in order to keep a consumer motivated over time to continue with an activity or continue using a product, that consumer should first set or be given a high-low range goal.

The Benefits of Emergency Reserves: Greater Preference and Persistence for Goals having Slack with a Cost

Abstract

Recommended publications

Data oriented program design

Human Effectiveness and Evaluation

Discovery of Linguistic Relations Using Lexical

Gender and evaluation of support programs for caregivers of older people