ArticlePDF Available

Abstract and Figures

Although money-back guarantees (MBGs) have a long tradition in marketing and retailing practice, a deeper understanding of how consumers value this instrument is still lacking. The results of two experimental studies show that in addition to cognitive effects, MBGs evoke a positive emotional response, thereby increasing consumers’ purchase intentions and willingness to pay a price premium. Moreover, MBGs positively affect consumers’ responses for search and experience goods, although for experience goods, MBGs should be designed with stricter return conditions as compared to MBGs for search goods. The results should help retail managers understand the consumer impact of MBGs, as well as assist them in pricing guaranteed items and designing effective MBGs according to the type of product.
Content may be subject to copyright.
Journal of Retailing 87 (4, 2011) 462–478
Understanding Money-Back Guarantees: Cognitive, Affective,
and Behavioral Outcomes
Thomas Suwelack a,, Jens Hogreveb,1, Wayne D. Hoyerc,2
aInstitute of Marketing, University of Muenster, Am Stadtgraben 13 - 15, 48143 Muenster, Germany
bCatholic University of Eichstaett-Ingolstadt, Ingolstadt School of Management, Chair of Service Management, Auf der Schanz 49, 85049 Ingolstadt, Germany
cDepartment of Marketing- B6700, McCombs School of Business, The University of Texas at Austin, Austin, TX 78733, United States
Although money-back guarantees (MBGs) have a long tradition in marketing and retailing practice, a deeper understanding of how consumers
value this instrument is still lacking. The results of two experimental studies show that in addition to cognitive effects, MBGs evoke a positive
emotional response, thereby increasing consumers’ purchase intentions and willingness to pay a price premium. Moreover, MBGs positively affect
consumers’ responses for search and experience goods, although for experience goods, MBGs should be designed with stricter return conditions
as compared to MBGs for search goods. The results should help retail managers understand the consumer impact of MBGs, as well as assist them
in pricing guaranteed items and designing effective MBGs according to the type of product.
© 2011 New York University. Published by Elsevier Inc. All rights reserved.
Keywords: Money-back guarantee; Risk perceptions; Emotions; Willingness to pay a price premium; Search and experience goods
By offering a money-back guarantee (MBG), a seller
promises that any customer who is not satisfied with a pur-
chase can return the item within a certain period and receive
a full refund (Davis, Gerstner, & Hagerty 1995). In response to
intense competitive forces in business environments, especially
during the recent recession, MBGs have been widely imple-
mented by retailers and manufacturers as a promotional tool
to gain consumers’ attention and positively influence their pur-
chase decisions (Sullivan 2009). Thus, firms are increasingly
taking advantage of MBGs, even though empirical investigations
about MBG outcomes on consumer behavior remain insufficient
(d’Astous and Guèvremont 2008).
Several authors argue that MBGs serve as extrinsic cues
of quality (e.g., Moorthy & Srinivasan 1995; Shieh 1996),
reduce consumers’ perceived risk (Grewal et al. 2003; Heiman,
McWilliams, & Zilberman 2001; Lei, de Ruyter, & Wetzels
Corresponding author. Tel.: +49 251 83 25 027.
E-mail addresses:
(T. Suwelack), (J. Hogreve), (W.D. Hoyer).
1Tel.: +49 841 937 1861; fax: +49 841 937 2976.
2Tel.: +1 512 471 1128.
2008), increase consumer satisfaction (e.g., McCollough &
Gremler 2004), and enhance purchase intentions (e.g., Davis
et al. 1995; Wood 2001). Yet despite these various effects ana-
lyzed in previous research, important gaps in literature remain.
In response, this article offers four contributions to retailing
and marketing literature. First, we extend previous research on
MBGs that has focused on cognitive variables such as quality
or risk perceptions, by examining their impact on affective out-
comes. We therefore respond to Chandon, Wansink, and Laurent
(2000) claim that promotions offer both utilitarian and hedo-
nic benefits. Furthermore, investigating emotions is important
because cognitive models only provide partial explanations for
responses to promotions (Erevelles 1998), and affective influ-
ences appear more important than previously believed (e.g.,
Bagozzi, Gopinath, & Nyer 1999), especially for retailing
(Puccinelli et al. 2009). To begin exploring the role of emotions
in an MBG context, we restrict our investigation to one negative
and one positive emotion that could both mediate MBG effects
on important consumer outcomes. Specifically, the focus is on
anticipated regret and liking, two emotions that strongly affect
consumer behavior (Kaltcheva & Weitz 2006; Simonson 1992).
Second, MBGs might increase profits by increasing con-
sumers’ willingness to pay a price premium (WTP), an effect
that has been discussed only theoretically (Fruchter & Gerstner
1999), despite calls for empirical investigations into both this
0022-4359/$ – see front matter © 2011 New York University. Published by Elsevier Inc. All rights reserved.
T. Suwelack et al. / Journal of Retailing 87 (4, 2011) 462–478 463
Model of the Outcomes of MBGs
Type of product
(search vs.
experience good)
Performance risk
Financial risk
Anticipated regret
Tes t du rat ion
Return conditions
Willingness to pay more
Purchase intention
Only in study 2
Study 1 and 2
Fig. 1. Model of the outcomes of MBGs.
specific relationship (Hogreve & Gremler 2009) and the vari-
ables that affect WTP in general (Chaudhuri & Ligas 2009).
We investigate MBG effects on consumers’ WTP, as mediated
through risk perceptions and emotions based on equity the-
ory. From an academic perspective, we gain new insights into
whether and how MBGs affect consumers’ WTP. From a man-
agerial perspective, the greater understanding of the relationship
between the presence of an MBG and WTP helps managers price
guaranteed items to achieve higher margins.
Third, we note a lack of research on the effectiveness of dif-
ferent MBG designs (Hogreve & Gremler 2009), which vary in
terms of the conditions for returning a product (i.e., what cus-
tomers must do to get their money back) and duration (i.e., the
period during which an item can be returned). Although both
design elements are critical for consumers’ decision making
(Heiman et al. 2002), research has not examined the behavioral
outcomes of variations in MBG designs. To reduce customer
cheating, some researchers have proposed implementing higher
“hurdles” in the form of more restrictive guarantee designs (e.g.,
Chu, Gerstner, & Hess 1998), as marketing practice often does
(Heiman et al. 2001). According to Anderson, Hansen, and
Simester (2009) restrictive MBG designs can reduce the $100
billion product return costs retailers and manufacturers face each
year by offering return policies. However, restrictively designed
return policies also might have negative effects on consumer
evaluations, leading to less demand (Anderson et al. 2009).
The literature on low price guarantees3(e.g., Kukar-Kinney,
Walters, & MacKenzie 2007; Kukar-Kinney, Xia, & Monroe
2007) reveals that the credibility of a guarantee suffers when it
is restrictive. Considering the strong impact of a signal’s cred-
ibility on consumer behavior (Ho, Ganesan, & Oppewal 2011;
Kukar-Kinney & Walters 2003), we suggest that MBG credibil-
ity – or the extent to which the message the MBG transmits is
believed – should be examined as a key mediator of MBG design
effects on consumer behavior.
Fourth, empirical research on MBGs has not investigated
the type of products for which they are most effective. This
issue is important because the impact of promotional tools on
consumer behavior varies across product types (Huang, Lurie,
3A low price guarantee is a tool similar to an MBG, though it only addresses
consumers’ price dissatisfaction.
& Mitra 2009). We categorize products according to Nelson’s
(1974) search versus experience goods classification, which has
a long tradition in marketing literature. Research has stressed the
importance of extrinsic cues (e.g., MBG) to lower prepurchase
uncertainty, especially in the presence of experience qualities
(Huang et al. 2009). In contrast, in the case of search goods,
consumers might value an MBG less. By analyzing consumers’
cognitions and emotions, we investigate whether it makes sense
to offer MBGs for search goods. With these results, managers
can make better decisions about whether to provide an MBG in
a specific product category and how to design MBGs according
to the level of risk involved.
Overview of our experimental studies. To address the identi-
fied research gaps, we conducted two experimental studies (see
Fig. 1). In Study 1, we test the general outcomes of an MBG and
examine its effects on consumers’ cognitions, affect, and inten-
tions. In particular, we investigate whether risk perceptions (i.e.,
performance and financial risk) and emotions (i.e., anticipated
regret and liking) mediate the MBG effects on consumers’ WTP
and purchase intentions. Furthermore, we analyze the type of
product (i.e., search vs. experience good) as a key moderator of
these consumer behavior outcomes. Then in Study 2, we inves-
tigate whether variations in MBG terms affect these responses
differently and examine MBG credibility as an important medi-
ator variable.
Study 1: General impact of a money-back guarantee
Impact on risk perceptions
The concept of risk plays an important role in understanding
consumer decision making (Mitchell 1999). Because consumers
sometimes want to avoid making mistakes, rather than just max-
imizing their utility, perceived risk often prevents them from
purchasing (Mitchell 1999). Therefore, it is essential for a seller
to offer information that reduces their perceived risk to a level
that is acceptable to the consumer (Bettman 1973). We focus on
performance and financial risk because research has documented
their high relevance for consumer decision making (Agarwal &
Teas 2001). Moreover, performance and financial risk effectively
explain variance in overall risk (Kaplan, Szybillo, & Jacoby
464 T. Suwelack et al. / Journal of Retailing 87 (4, 2011) 462–478
Performance risk arises when a consumer believes that a pur-
chased item may perform below expectations (Horton 1976).
According to Shimp and Bearden (1982, p. 39), “higher per-
ceived quality may serve to mitigate the risk that accompanies
the uncertainty of whether a ... product will satisfactorily
perform its intended function.” Prior research has indicated
that MBGs can increase quality perceptions (e.g., Moorthy &
Srinivasan 1995), according to the following reasoning: When
a seller provides an MBG, it promises to pay back the amount
of the purchase price if the consumer is not satisfied with the
product. Thus, the offer implies a self-imposed penalty for bad
performance, meaning that it would be economically unwise
for sellers of low-quality products to provide an MBG, because
they would incur higher return costs (Kirmani & Rao 2000;
Mitra & Fay 2010). Therefore, offering an MBG communi-
cates high product quality and thereby reduces performance risk
Financial risk refers to the consumer’s potential financial loss
associated with the purchase, including the possibility that the
product must be repaired or replaced (Horton 1976). Because
product repairs or replacements depend on product quality, the
argument is similar to that of performance risk: An MBG that
signals product quality also reduces financial risk. Furthermore,
MBGs ensure that the customer will not suffer financial losses if
she or he is dissatisfied with the product or finds the same item
for a lower price postpurchase at another retailer. Thus MBGs
should reduce financial risk perceptions.
Impact on emotions
Researchers and marketers appear increasingly interested in
studying the role of emotions in consumer behavior, includ-
ing the context of promotions (Chandon et al. 2000). Emotions
are frequently classified into independent negative and pos-
itive dimensions, which consumers may feel simultaneously
(Ramanathan & Williams 2007).
We propose that the negative emotion of regret, which occurs
when a consumer realizes that she or he would be better off hav-
ing made a different purchase decision, has major relevance in an
MBG context, because an MBG can eliminate negative purchase
consequences during the trial period. Research shows that regret
can be anticipated before purchase and plays an important role in
consumer decision making (e.g., Shih & Schau 2011; Zeelenberg
& Pieters 1999). Specifically, to avoid regret, consumers rely
on diverse informational cues. For example, Simonson (1992)
finds that consumers tend to favor well-known and higher-priced
brands when they anticipate regret. Moreover, Tsiros and Mittal
(2000) find that regret is lower in the case of reversible com-
pared with irreversible outcomes. Considering the informational
role of an MBG in prepurchase situations and the opportunity it
4In our empirical studies we focus on risk perceptions and not on quality
perceptions (compare Fig. 1), first, to keep our model parsimonious, and, second,
because the reduction of risk perceptions is assumed to be a central outcome
of an MBG having important implications for consumer behavior (e.g., Grewal
et al. 2003; Heiman et al. 2001).
offers to reverse a purchase, we propose that an MBG reduces
the anticipation of regret.
Furthermore, we postulate that an MBG can evoke the
positive emotion of liking, an affective variable frequently stud-
ied in sales promotion research (e.g., Lichtenstein, Burton, &
Netemeyer 1997; Naylor, Raghunathan, & Ramanathan 2006).
According to Roseman’s (1991) appraisal theory, liking occurs
when the subject can gain a personal advantage from someone
else (e.g., seller). For example, consumers value greater flexi-
bility when making choices (Wood 2001). An MBG provides
greater flexibility in a choice task because it offers the option to
postpone the final purchase decision. In other words, irreversibil-
ity of the purchase is not determined at the time of purchase but
at the end of the MBG test duration. Thus, an MBG creates an
advantage in the purchase context and in line with Roseman
(1991), may therefore evoke the positive emotion of liking.
Impact on consumers’ willingness to pay a price premium
We assume that an MBG increases consumers’ WTP by
affecting their risk perceptions and emotions. To explain this
indirect effect, we apply equity theory which helps us understand
how informational cues lead to perceptions of fairness. Accord-
ing to equity theory, fairness perceptions arise when exchange
partners believe that the ratios of what they invest or sacrifice
relative to what they receive (i.e., their cost–benefit ratios) are
equal across partners (Adams 1965; Oliver & Swan 1989). In
the case of positive or negative inequity (i.e., the ratio for one
partner is higher or lower than the other’s), both partners will
be motivated to change the exchange parameters to reestablish
equity (Homburg, Koschate, & Hoyer 2005).
Because an MBG should reduce the consumer’s risk percep-
tions, the cost side of the consumer’s cost–benefit ratio decreases
(i.e., as anticipated future costs, risk enters the cost side of
the ratio; Zeithaml 1988). Furthermore, the benefit side of the
ratio increases because the MBG evokes a positive emotional
response. Thus, assuming the purchase price of the product
remains constant, the consumer’s cost–benefit ratio improves
compared with a situation with no MBG. In contrast, the seller
incurs costs when offering an MBG (e.g., restocking, deprecia-
tion of returned items). Although a seller strives for benefits from
offering the MBG such as additional sales, the total effect on its
cost–benefit ratio is unclear (Anderson et al. 2009). Thus the
seller might incur negative inequity while the consumer incurs
positive inequity.
By demanding a price premium for the MBG, the seller can
reestablish transaction equity. A higher price should seem fair
up to the level that equalizes the cost–benefit ratio of the seller
and the consumer. Because an MBG reduces risk perceptions
and generates more positive emotional responses, the consumer
might be willing to accept higher prices when the seller pro-
vides an MBG. In turn, we predict an indirect effect of an MBG,
through risk perceptions and emotions, on WTP and hypothe-
H1a.By reducing (i) performance risk and (ii) financial risk,
an MBG increases WTP.
T. Suwelack et al. / Journal of Retailing 87 (4, 2011) 462–478 465
H1b.By reducing (i) anticipated regret and increasing (ii) lik-
ing, an MBG increases WTP.
Impact on consumers’ purchase intentions
We also propose that an MBG increases consumers’ pur-
chase intentions by affecting their risk perceptions and emotions.
First, prior research has shown that risk perceptions negatively
influence behavioral intentions (e.g., Bearden & Shimp 1982).
More specifically, the belief that an MBG results in lower per-
formance and financial risk should induce the consumer to form
a more favorable attitude toward buying the guaranteed item,
which leads to a stronger purchase intention (Fishbein & Ajzen
Second, according to Schwarz and Clore (1983), people
base their intentions toward an object on the self-referencing
heuristic: How-do-I-feel-about-it? Negative emotions lead to an
unfavorable evaluation, positive emotions lead to a favorable
evaluation, and they consistently influence intentions toward
the object (Pham 1998). Applying the How-do-I-feel-about-it?
heuristic here, we suggest that an MBG may avoid an unfavor-
able evaluation of the product offer by reducing the negative
emotion of anticipated regret and induce a favorable evaluation
by evoking the positive emotion of liking. In so doing, the MBG
should increase purchase intentions. Thus we hypothesize:
H2a.By reducing (i) performance risk and (ii) financial risk,
an MBG increases purchase intentions.
H2b.By reducing (i) anticipated regret and increasing (ii) lik-
ing, an MBG increases purchase intentions.
Moderating effects of type of product
We assume that an MBG has a stronger effect for experience
goods compared with search goods in terms of risk reduc-
tion. Because the evaluation of experience goods is inherently
more subjective, less diagnostic, and more difficult than that of
search goods (Hoch & Deighton 1989), consumer risk is nec-
essarily greater for experience goods. According to Ostrom and
Iacobucci (1998), the impact of a guarantee on consumer evalu-
ations might be stronger when the risk level increases. An MBG
enables the consumer to test the product and verify its char-
acteristics which is especially valuable for experience goods.
Thus risk perceptions can be significantly reduced for experi-
ence goods. In contrast, a search good’s prepurchase uncertainty
is relatively lower because consumers can evaluate these prod-
ucts before purchase. Thus risk reduction is only minimal for
search goods.
Furthermore, we assume that an MBG has a stronger effect
for experience goods in terms of the emotional response. With
an MBG, consumers can test the product and “update” their
information. This option has great importance in the case of
experience goods and makes the consumer feel better about a
purchase decision. That is, the consumer should anticipate less
regret and like the offer more when the seller provides an MBG
for goods whose quality is hard to predict. We derive:
Table 1
Search and experience goods classifications.
Good Mean (experience–search good)a
Pretest Study 1 Study 2
Vitamin pillsb1.03 .32 .42
Washing machineb.83 1.08 .47
Printerb.59 .99 .54
Television set .37
Printer paper .33
Detergent .29
Insect repellent .28
Electric toothbrush .19
Digital camerab.18 1.14 .11
Coffee machine .12
Car navigation system .18
Mobile phone .27
Face cream .39
MP3 player .46
Probiotic yogurt .49
Anti-dandruff shampoob.51 .67 .94
Mattressb.71 2.18 2.89
Ski jacket .74
Running shoes .99
Skier 1.05
Office chairb1.22 2.00 2.15
Ski bootsb1.23 1.88 2.01
aHigher values represent more experience attributes.
bProducts included in Study 1 and Study 2.
H3a.The effects of an MBG on (i) performance risk and (ii)
financial risk are stronger for experience goods than for search
H3b.The effects of an MBG on (i) anticipated regret and (ii)
liking are stronger for experience goods than for search goods.
Based on H3a,b and the mediations specified in H1(an MBG
affects WTP via risk perceptions and emotions) and H2(an MBG
affects purchase intentions via risk perceptions and emotions),
we further suggest the following hypotheses:
H3c.The indirect effect of an MBG on WTP is stronger for
experience goods than for search goods.
H3d.The indirect effect of an MBG on purchase intentions is
stronger for experience goods than for search goods.
Pretest of products. We conducted a pretest of 22 products
to select those perceived as either search goods or experience
goods (see Table 1). A representative sample of 160 German
consumers took part in the pretest.
Each participant rated five 7-point Likert scale items for
each product: Three to assess experience qualities and two to
assess search qualities. We adapted the items from Weathers,
Sharma, and Wood (2007; see Appendix A). We randomly
allocated products across participants using an online ques-
tionnaire. To avoid response fatigue, each participant evaluated
only three products. We averaged the results to create mea-
sures of experience and search qualities. Next, we computed the
466 T. Suwelack et al. / Journal of Retailing 87 (4, 2011) 462–478
Fig. 2. Example of a stimulus in Study 1 (Translated from German).
difference between these measures. Positive values in Table 1
indicate that a good contains mainly experience qualities, and
negative values represent a good that contains mainly search
We selected only products with highly positive or highly
negative mean differences, which fit our requirements for het-
erogeneous price levels and product categories. Accordingly,
we chose a mattress, an office chair, an anti-dandruff shampoo,
and ski boots as experience goods and vitamin pills, a washing
machine, a digital camera, and a printer as search goods. The
price levels of these products were not equal within product cat-
egories but were approximately equal across product categories,
because for each experience good, we chose a search good that
was equivalent in terms of price. The experimental studies thus
feature four search and four experience goods whose pretest
means differed significantly (Mexperience = 1.22, Msearch =.82;
p< .01, two-tailed).
Sample and experimental design. A large market research
firm specializing in online panels provided the sample for Study
1. The firm possesses representative national databases that con-
sist of consumers who have agreed to participate in surveys
in exchange for cash incentives. The final sample in Study 1
consists of 200 consumers and is representative of the German
market in terms of gender, age, and income.
To test H1–H3, we used a 2 (MBG present vs. absent) ×2
(search vs. experience good) between-subjects factorial design.
Within an online survey, participants were randomly assigned
to one of the four treatment conditions, each represented by a
section of a firm’s advertising brochure that contained one of the
eight products.5We manipulated the MBG-present condition by
inserting the following statement (see also Fig. 2): “Money-Back
5Wepooled the results of the eight products across the two product categories,
as we show in the section of the manipulation checks. We thus could work with
a binary product category variable.
Guarantee: Test our [name of search or experience good] for 30
days. If you are not satisfied with your purchase we give you
your money back-guaranteed!” In addition, to make the scenario
more realistic, an “MBG logo” appeared. Half the respondents
saw the same advertisement but without an MBG. Furthermore,
we provided product information for the search and experience
goods. This information described the product in the specific
category, chosen by screening vast print and online media. We
used fictitious brand names to remove any confounds due to
prior brand perceptions. However, we provided price informa-
tion so that participants could indicate their purchase intentions
and WTP. We chose an average price level for each product
to minimize price-oriented quality evaluations. All additional
information was constant across the four treatments. To ensure
that the additional information given in the advertisement did
not affect the results of the final estimation, we controlled for
perceived information content (i.e., information to be transmit-
ted by the advertisement beyond the MBG). We also controlled
for participants’ knowledge in the specific product category (see
Appendix A).
Measures and manipulation checks. To measure the con-
structs of interest, participants responded to a series of
multi-item Likert scales. We measured performance and finan-
cial risk perceptions using three items each, based on Shimp and
Bearden’s (1982) scale. We measured anticipated regret with
three items adapted from Tsiros and Mittal (2000) and Shih and
Schau (2011). For liking, we selected three items from a scale
by Tripp, Jensen, and Carlson (1994). We assessed WTP with
three items from Zeithaml, Berry, and Parasuraman (1996) and
Fuchs, Prandelli, and Schreier (2010). Purchase intentions was
measured using the scale that Grewal, Monroe, and Krishnan
(1998) developed. Table 2 lists all items and displays the psy-
chometric properties, means, and standard deviations of the
Each construct demonstrated good psychometric properties
in terms of composite reliability (CR) and average variance
extracted (AVE); no CR was lower than .8, and no AVE was
lower than .7 (Bagozzi & Yi 1988) which indicates the sufficient
reliability and convergent validity of the construct operational-
izations (though one item of the WTP factor needed to be
dropped due to very low factor loadings). The goodness-of-
fit indices for the measurement model yielded good results
(χ2= 177.62, df = 104, confirmatory fit index [CFI] = .98; root
mean square error of approximation [RMSEA] = .06; square
root mean residual [SRMR] = .03). We assessed discriminant
validity in two ways. First, we evaluated discriminant validity
using Fornell and Larcker’s (1981) procedure. As Table 3 shows,
the square root of the AVE for every factor was always greater
than the highest correlation between two constructs, indicat-
ing discriminant validity. Second, each correlation between two
variables was constrained to equal 1, and then we examined any
differences in the chi-square values between the constrained and
unconstrained models (Anderson & Gerbing 1988). Significant
test statistics (p< .05) confirm discriminant validity. We also
analyzed whether the measurement model was robust to com-
mon method variance using a single-method factor approach
(Podsakoff et al. 2003). We compared a single-factor solution
T. Suwelack et al. / Journal of Retailing 87 (4, 2011) 462–478 467
Table 2
Construct descriptions and final measures.
Constructs Item descriptionsaM(SD)bFactor loadingsbCRbAV E b
Performance risk 1. Considering possible problems with the product’s performance, how
much risk would be involved with buying this product? (“very low
risk/very high risk”)
3.42 (1.32)
3.57 (1.35)
.939/.936 .932/.932 .821/.822
2. How sure are you about the product’s ability to perform?
(“very sure/not sure at all”)
3. I am very confident that the product will perform as expected. .837/.849
Financial risk 1. Given the potential financial expenses, how much risk would be involved
buying this product? (“very low risk/very high risk”)
3.25 (1.72)
2.99 (1.56)
.888/.863 .941/.919 .841/.791
2. How financially risky do you feel it would be to choose the product?
(“not risky at all/very risky”)
3. I think the purchase of the product is financially risky. .901/.857
Anticipated regret 1. Using this offer could be a decision that I will later regret. 3.39 (1.51) .949/.947 .959/.966 .887/.905
2. I might later feel regret accepting this offer. 3.33 (1.66) .940/.972
3. It is quite possible that I later regret the decision to accept this offer. .937/.934
Liking 1. I like this offer. 4.23 (1.48) .904/.891 .965/.959 .902/.885
2. Accepting this offer would give me pleasure. 3.97 (1.54) .971/.965
3. I would enjoy this offer. .973/.965
WTP 1. For this offer I would be willing to pay a higher price. 4.05 (1.48) .821/.870 .881/.910 .788/.834
2. I would still buy the product even if the seller increases the price. 2.66 (1.43) .950/.955
Purchase intention 1. If I were going to buy a , the probability of buying this one is likely to
be ...(“very low/very high”)
4.06 (1.57)
3.84 (1.61)
.872/.882 .943/.954 .846/.873
2. The probability that I would consider buying this is ...(“very
low/very high”)
3. The likelihood that I would purchase this is ...(“very low/very high”) .960/.958
MBG credibility 1. The MBG is believable. –/.939 –/.965 –/.933
2. The MBG is credible. 4.43 (1.68) –/.992
aScales not specifically defined used “totally disagree/totally agree” as anchors.
bFirst (pair of) values refer to Study 1, and second (pair of) values refer to Study 2.
in which all manifest variables could be represented by a single
factor with a model in which the manifest variables were rep-
resented by each corresponding construct. The chi-square test
statistic revealed a significantly lower fit for the single-factor
solution, indicating that the measures were robust to common
method variance (χ2= 1611.57; df = 15).
We assessed the MBG manipulation by asking respon-
dents to indicate whether the advertisement included an MBG
(Biswas, Dutta, & Pullig 2006; see Appendix A). Overall 93
percent of the participants responded to the manipulation check
correctly.6Using the same items as in the pretest, participants
also evaluated the search and experience qualities of the eight
products (Table 1). Because one product (digital camera) failed
the check for search qualities, we excluded it from further
To assess whether it is appropriate to aggregate the results
across goods within the search and experience goods categories,
we first conducted a MANOVA and follow-up ANVOVAs for
each category, with product and MBG as independent variables
and risk perceptions, emotions, WTP, and purchase intentions as
dependent variables. In neither category do we find significant
6The respondents answered as follows: MBG-condition: correct = 83;
false = 9; No-MBG-condition: correct = 96; false = 12.
interactions between product and MBG on the multivariate level
(search: F= .728, p= .88; experience: F= 1.280, p= .11) and
the univariate levels (the F-statistic for each dependent variable
is nonsignificant; p> .10). Second, we also controlled whether
products might differ in other aspects important for the purchase
decision. Pooled within each category, search and experience
goods did not differ significantly in respondents’ knowledge
in the specific product category (p= .273, two-tailed), prod-
uct involvement (p=.103, two-tailed), or utilitarian perception
(p= .715, two-tailed). However, they differed in their hedonic
perception (p= .073, two-tailed), which was not intended, and
their inherent risk (p= .002, two-tailed), with experience goods
scoring higher on this criterion, which was intended. Including
hedonic perception (and product involvement, as significance
for this variable is ambiguous) in the path model did not show
any significant effects on the dependent variables, supporting
the pooling of the data across products to create search and
experience goods categories.
Procedure. Before answering the questionnaire, participants
provided demographic information. Then they were exposed to
the treatment conditions (i.e., an advertisement with one of the
eight products including or excluding MBG information), rated
the dependent variables (i.e., performance and financial risk
perceptions, regret and liking, WTP, and purchase intention),
completed manipulation checks, and responded to controls.
468 T. Suwelack et al. / Journal of Retailing 87 (4, 2011) 462–478
Table 3
Correlations of main constructs (Study 1 and Study 2).
1 MBG credibility
2 Performance risk .906
.541 .906
3 Financial risk .047 .917
.258 .127 .889
4 Regret .073 .109 .941
.372 .199 .094 .951
5 Liking .097 .027 .069 .949
.532 .301 .107 .195 .940
6 WTP .288 .190 .029 .192 .887
.388 .319 .052 .162 .437 .913
7 Purchase intention .308 .207 .154 .654 .503 .919
.509 .487 .170 .328 .614 .675 .934
8 Guarantee .238 .190 .169 .200 .091 .247
9 Search/Experience (SE) .210 .230 .228 .164 .225 .347 .003
.092 .068 .010 .011 .072 .156 .174 –
10 Guarantee ×SE .063 .182 .063 .002 .015 .002 .478 .378
11 Test duration (TD) –––––––––
.128 .184 .063 .116 .192 .207 .156 – .079 –
12 Return conditions (RC) –––––––––
.539 .298 .128 .202 .296 .242 .323 – .063 – .390 –
13 TD ×SE –––––––––
.128 .085 .023 .026 .085 .166 .173 – .928 – .232 .105 –
14 RC ×SE –––––––––––––
.095 .068 .012 .016 .071 .149 .168 – .906 – .124 .198 .886 –
Notes. Values of Study 1 in italics; diagonal elements root mean square of AVE.
We used structural equation modeling to test our hypotheses.
The analysis was conducted by employing the ML-estimator
as implemented in MPLUS 5.2. The overall fit of the model
is acceptable (χ2= 460.63, df = 185; CFI = .92; RMSEA = .09;
SRMR = .160).
The values in Table 4 show that, as expected, an MBG
reduces performance risk, financial risk, and anticipated regret
and increases liking to significant degrees. Using the boot-
strapping procedure as a mediation test (Preacher & Hayes
2008), we find significant indirect effects of the MBG on WTP
and purchase intentions (see Table 4), emphasizing the gen-
eral importance of the suggested mediators for the analysis of
MBG outcomes. The total indirect effect on WTP is positive and
significant (.064; critical ratio [Z] = 2.64). Specifically, we find
significant scores through performance risk (.023; Z= 1.29) and
liking (.040; Z= 1.74). All other indirect paths are nonsignifi-
cant. These results support H1ai and H1bii.
The total indirect effect of the guarantee on purchase inten-
tions is also significant (.176; Z= 3.53). The results in Table 4
and the bootstrapping results indicate significant indirect effects
through performance (.030; Z= 1.68) and financial risk percep-
tions (.012; Z= 1.28) as well as through liking (.131; Z= 3.28).
Thus, H2ai,H
2aii, and H2bii are supported. Because the indirect
paths from MBG to WTP and purchase intentions through liking
explain a larger portion of the variance than all other paths, our
data highlight the central role of positive emotions in the MBG
context. In contrast, among the suggested mediators anticipated
regret is of least importance and the hypotheses regarding this
variable, namely H1bi and H2bi, need to be rejected. Compar-
ing the indirect effects of the guarantee on WTP with those on
purchase intentions we identify that the paths on purchase inten-
tions in sum explain a greater share of the variance compared to
the paths on WTP (.176 vs. .064).
For H3a,b, we proposed that the effects of the guarantee on
the mediators would be stronger for experience goods compared
with search goods. As we detect significant negative effects of
the interaction between the MBG and the search and experience
goods measure on financial risk (β=.257, t=3.34) and regret
(β=.110, t=1.62), H3aii and H3bi are supported. This means
that an MBG more strongly reduces financial risk perceptions
and anticipated regret for experience goods relative to search
goods. Considering the intentional outcomes (H3c,d), the indirect
paths through financial risk shows a significant Z-score for both
WTP (.035; Z= 1.47) and purchase intentions (.030; Z= 1.69),
signifying a mediation of the interaction terms’ impact through
financial risk. The values reveal a greater indirect effect of an
MBG on WTP and purchase intentions for experience goods
compared with search goods, supporting H3c and H3d. This effect
is mainly driven by the stronger reduction of financial risk for
experience goods.
T. Suwelack et al. / Journal of Retailing 87 (4, 2011) 462–478 469
Table 4
Results of the direct and indirect effects.
Model 1aModel 2bModel 3cModel 4dHypothesis
Path Coefficient T-value Coefficient T-value Coefficient T-value Coefficient T-value
MBG performance risk .175 2.72*** .169 2.65***
MBG financial risk .103 1.68** .103 1.68**
MBG regret .115 1.82*** .116 1.85**
MBG liking .231 3.72*** .224 3.60***
MBG WTP .138 2.06** .007 .110 .009 .142
MBG purchase intentions (PI) .295 4.88*** .040 .558 .062 1.29*
MBG ×SE performance risk .081 1.16 .075 1.07 H3ai:no
MBG ×SE financial risk .257 3.34*** .169 2.64*** H3aii: yes
MBG ×SE regret .110 1.62*.109 1.59*H3bi: yes
MBG ×SE liking .057 .803 .050 .737 H3bii:no
MBG ×SE WTP .083 1.29*.057 1.28*.039 .545
MBG ×SE PI .141 2.78*** .028 .576 .028 .545
Performance risk WTP .146 1.88** .137 1.60*
Financial risk WTP .138 1.42*.135 1.84**
Regret WTP .126 1.66** .110 1.55*
Liking WTP .185 1.97** .178 2.14**
Performance risk PI .160 2.41*** .180 2.74***
Financial risk PI .116 2.35** .118 2.74***
Regret PI .012 .225 .027 .502
Liking PI .550 8.88*** .584 11.43***
Indirect effects Coefficient Z-score Hypothesis
MBG performance risk WTP .023 1.29*H1ai: yes
MBG financial risk WTP .014 1.06 H1aii :no
MBG regret WTP .013 1.07 H1bi:no
MBG liking WTP .040 1.74** H1bii: yes
MBG WTP .064 2.64***
MBG performance risk PI .030 1.68** H2ai: yes
MBG financial risk PI .012 1.28*H2aii : yes
MBG regret PI .003 .347 H2bi:no
MBG liking PI .131 3.28** H2bii: yes
MBG PI .176 3.53***
MBG ×SE WTP .035 1.47*H3c: yes
MBG ×SE PI .030 1.69** H3d: yes
aPredictor relates to ultimate endogenous variable.
bPredictor relates to mediators.
cMediators relate to endogenous variables controlling for predictor variable.
dDirect path become insignificant in the presence of the mediators.
*p< .1 (one-tailed).
** p< .05 (one-tailed).
*** p< .01 (one-tailed).
Discussion of Study 1
The goal of Study 1 was to shed light on basic consumer
reactions to MBG offers. Our model of the effects of MBGs on
consumer evaluations is largely supported: MBGs not only cre-
ate cognitive effects by reducing consumers’ risk perceptions
but also evoke positive and reduce negative emotions. Con-
sumers’ WTP and purchase intentions are positively affected
by liking but not influenced by regret (as we discuss in more
detail in sections “General discussion” and “Limitations and
further research”). Our results hold for both experience and
search goods, though the effect is even stronger for experience
goods. Accordingly, our study makes three important contribu-
tions. First, we show that responses to MBGs can be emotional.
Second, these emotions provide an important explanation for
the surprising effectiveness of MBGs in influencing consumer
behavior toward search goods. Third, we find a positive effect
of MBGs on WTP and purchase intentions; this critical result
shows that MBGs may increase revenues by both increasing
the number of units sold (as indicated by purchase intentions)
and enhancing the prices a seller may demand (as indicated by
consumers’ WTP).
Study 2: Impact of money-back guarantee design
Impact on money-back guarantee credibility
In marketing practice, there is great variability in MBG
designs because of the specification of different MBG terms
(Posselt, Gerstner, & Radic 2008). Some retailers specify no
470 T. Suwelack et al. / Journal of Retailing 87 (4, 2011) 462–478
or only a few restrictions on product returns. For example,
JCPenney offers a “no questions asked” 400-day MBG on its
digital camera batteries (Camera Batteries 2010). In contrast, establishes several return conditions that customers
must fulfill to get money back after returning an item (e.g.,
ask for authorization) (Blinds 2010). The reasons for imposing
return restrictions are twofold. First, restrictions may increase
consumer effort and therefore help reduce product returns by dis-
satisfied customers (Davis, Hagerty, & Gerstner 1998). Second,
increasing consumer effort restrictions may reduce unethical
returns or consumer cheating (Wirtz & Kum 2004). In both cases,
the seller incurs return costs. Therefore, many retailers specify
restrictive MBG terms that limit the duration and/or impose strict
return conditions (Heiman et al. 2001).
However, both a short duration and strict return conditions
might also prompt negative consumer evaluations. This concern
is supported by prior research which has shown that a nega-
tively framed semantic structure (i.e., verbal specifications) of a
promotion negatively affects consumer behavior (Berkowitz &
Walton 1980). Moorthy and Srinivasan (1995) state that MBG
effects depend on the probability of truth telling, which is con-
sistent with effects suggested for low price guarantees (e.g.,
Kukar-Kinney, Xia, et al. 2007; Srivastava & Lurie 2004). There-
fore, MBG credibility should be a key construct that mediates
the effects of different MBG designs on consumer behavior.
Specifically, a short duration and more strict return conditions
could evoke counterarguments from the consumer and therefore
may result in lower ad credibility (Swinyard 1981). According
to the persuasion knowledge model (Friestad & Wright 1994),
consumers know that sellers attempt to persuade them and there-
fore infer different motives, depending on the specific nature of
the attempt (Jain & Posavac 2001). Consumers may be skepti-
cal of an offer and consider ulterior seller motives (Campbell
& Kirmani 2000) which prevents the ad from transmitting
the intended message or even could evoke negative consumer
responses. In the context of a restrictive MBG, consumers may
question why the seller offers an MBG but sets high hurdles to
the return. In this case, they might counterargue that the retailer
fails to live up to its promises because the restrictions protect
it from expensive product returns. Therefore consumers do not
consider the self-imposed penalty of promising money back a
credible signal by the seller (Boulding & Kirmani 1993). We
H4a.A short duration of an MBG decreases MBG credibility.
H4b.Strict return conditions stated in an MBG decrease MBG
Furthermore, the return conditions should have a stronger
effect on MBG credibility than its duration. First, inferring nega-
tive seller motives should occur to a stronger degree in the case of
strict return conditions because the consumer effort to return the
product increases which should evoke strong counterarguments
from the consumer and therefore makes the guarantee signal less
credible. Second, Ford, Smith, and Swasy (1990) postulate that
consumers are less skeptical of objective claims than subjective
claims. The MBG duration is objective information because the
period is specific and not subject to interpretation. In contrast,
MBG return conditions tend to be outlined in multiple written
paragraphs, demand interpretation, and leave room for unex-
pected seller reactions (i.e., the consumer might fear the seller
will deny a product return because she or he did not conform with
one of the many conditions), which makes them subjective infor-
mation. According to Ford et al. (1990), consumers thus should
be more skeptical when faced with strict return conditions than
when faced with a short MBG duration. We propose:
H4c.MBG return conditions exert a stronger influence on MBG
credibility than does MBG duration.
Moderating effects of type of product
Because experience goods can be evaluated only after con-
sumption, consumers are often skeptical of their advertising
claims (Franke, Huhmann, & Mothersbaugh 2004). In contrast,
for search goods, consumers might reason that “this must be true
as I can verify it without much ado” (Jain & Posavac 2001,p.
170) and therefore should be less skeptical. Thus credibility con-
cerns should be greater for experience goods (Franke et al. 2004)
and a seller needs to send strong signals to reduce information
asymmetry; otherwise, credibility problems arise. Consumers
should not view MBGs that provide only short durations and/or
strict return conditions as strong signals because these guaran-
tee restrictions evoke counter argumentation and inferences of
negative seller motives (Kukar-Kinney, Xia, et al. 2007). When
consumers attribute negative motives to the seller, the credibil-
ity of the guarantee decreases. Considering the higher degree of
consumer skepticism for claims that advertise experience goods,
we suggest that the negative impact of a short duration and strict
return conditions on MBG credibility will be more pronounced
for experience goods. That is:
H5a.A short duration of an MBG reduces MBG credibility to
a higher degree for experience goods than for search goods.
H5b.Strict return conditions stated in an MBG reduce MBG
credibility to a higher degree for experience goods than for
search goods.
Effects of money-back guarantee credibility
Consumers evaluate a product more favorably when the cred-
ibility of its advertised claims is higher (Jain & Posavac 2001;
Kukar-Kinney & Walters 2003). In contrast, they counterargue
and discount a claim with low credibility which results in its
lower acceptance (Swinyard 1981). Thus, an MBG will not be
accepted if it is not credible. As we demonstrated with regard to
H1and H2, an MBG that is accepted reduces risk perceptions,
enhances emotional responses, and increases a consumer’s WTP
and purchase intention. To achieve positive effects on consumer
risk and emotions and thus on WTP and purchase intentions,
an MBG thus needs to be credible. We propose the following
indirect effects of MBG credibility on the dependent variables:
H6a.By reducing (i) performance risk and (ii) financial risk,
MBG credibility increases consumers’ WTP.
T. Suwelack et al. / Journal of Retailing 87 (4, 2011) 462–478 471
H6b.By reducing (i) anticipated regret and increasing (ii) lik-
ing, MBG credibility increases consumers’ WTP.
H6c.By reducing (i) performance risk and (ii) financial risk,
MBG credibility increases consumers’ purchase intentions.
H6d.By reducing (i) anticipated regret and increasing (ii) lik-
ing, MBG credibility increases consumers’ purchase intentions.
Sample and experimental design. Similar to Study 1, we used
a consumer panel which is representative of the German market.
To test H4–H6, we applied a 2 (less strict vs. strict MBG return
conditions) ×2 (long vs. short MBG duration) ×2 (search vs.
experience good) between-subjects experimental design. Thus,
Study 2 contained eight experimental conditions in which 400
people participated. We employed the same products as in Study
1. However, the results of the search and experience qualities
check indicated no need to exclude products from the final
sample (see Table 1).
The MBG return conditions were manipulated with state-
ments that indicated a product return would need to be justified
by at least 20 written words (strict conditions) or that no ques-
tions would be asked (less strict conditions). Duration was
manipulated by indicating that a product could be returned
within 30 days (long duration) or five days (short duration).
We chose these manipulations as representative of those fre-
quently used in marketing practice. The setting was the same
as that for Study 1, except that all respondents of Study 2 read
a scenario that included information about an MBG offer. We
again controlled for the advertisement’s information content and
respondents’ prior knowledge in the product category.
Measures and manipulation checks. The dependent vari-
ables of Study 2 were the same as in Study 1, except for
MBG credibility. We measured MBG credibility using two
items based on Kukar-Kinney and Walters’s (2003) scale.
Again, the values for the reliability and validity measures
were adequate for all constructs (see Table 2), and we did
not detect any serious problems of discriminant validity or
common method variance (χ2= 179.23, df = 131, CFI = .99;
RMSEA = .03; SRMR = .019). We measured the manipulation
checks on a seven-point Likert scale (the scale of return condi-
tions ranged from 1 = “not strict at all” to 7 = “very strict”; the
scale of duration ranged from 1 = “very long” to 7 = “very short”;
see Appendix A) and obtained evidence of successful manip-
ulations (Mless strict = 3.02, Mstrict = 5.75; p< .01; Mlong = 4.00,
Mshort = 4.88; p< .01). We also performed a 2 ×2 analysis of
variance (ANOVA) of the statistical significance of all main
and interaction effects (for a detailed description, see Perdue
& Summers 1986). Only significant effects for the main effects
of the manipulations in their corresponding manipulation checks
were in evidence.
Procedure. Before answering the questionnaire, partici-
pants provided their demographic information. They were then
exposed to the treatment conditions, rated the dependent vari-
ables, completed the manipulation checks, and responded to
Similar to Study 1, we employed MPLUS 5.2 to calcu-
late the model results (see Table 5). We obtained good results
for the model’s overall fit (χ2= 662.15, df = 288; CFI = .96;
RMSEA =.057; SRMR =.09). We then tested for the three-way
interaction of return conditions, duration, and type of prod-
uct on MBG credibility, which was not significant (β=.086,
t=.234). Thus, the effects suggested in the hypotheses can
be interpreted. H4a and H4b predict that a short MBG dura-
tion and strict MBG return conditions reduce MBG credibility.
The path from MBG duration to MBG credibility is significant
(β=.084, t=1.64), in support of H4a. Moreover, our results
confirm H4b by showing that strict conditions negatively influ-
ence MBG credibility (β=.554, t=12.83). Considering the
relative effectiveness of the MBG design elements, path coef-
ficients show that the effect of duration on credibility is lower
than the one of return conditions, which supports H4c. We also
conducted a chi-square difference test in which we held the two
paths to be equal. This assumption yielded a significantly worse
model fit (χ2= 23.22; df = 1), in further support of H4c .
H5refers to the interaction of the two MBG design ele-
ments separately with the type of product. The results reveal
a significant interaction effect between test duration and type of
product on MBG credibility (β=.170, t=1.31). That is, the
negative effect of a short duration on MBG credibility is signif-
icantly stronger for experience goods. However, the interaction
between the type of product and return conditions is signifi-
cant with a positive coefficient (β= .445, t= 408), meaning that
strict return conditions lead to higher credibility perceptions in
the case of experience goods relative to search goods. Thus our
results contradict the effect we propose in H5b, while H5a is
Finally, we analyzed the behavioral outcomes of MBG
credibility using a bootstrapping technique to calculate the
significance of the indirect effects. First, as we expected,
greater MBG credibility reduces performance risk (β=.523,
t=13.38), financial risk (β=.277, t=5.65), and antic-
ipated regret (β=.377, t=8.32) and increases liking
(β= .494, t= 12.78) to a significant degree. Second, we find
significant indirect effects of MBG credibility on both WTP
(.216; Z= 5.63) and purchase intentions (.457; Z= 9.78). Specifi-
cally, performance (.146; Z= 4.27) and financial risk perceptions
(.021; Z= 1.45), regret (.070; Z= 2.21), and liking (.219;
Z= 4.96) all mediate the effect on purchase intentions, whereas
only performance risk (.073; Z= 2.13) and liking (.136; Z= 3.95)
mediate the effect on WTP. Thus, with the exception of H6aii and
6a to H6d are supported. In terms of the relative strength
of the MBG effect on the two intentional variables results show
its higher importance in increasing consumers’ purchase inten-
tions compared to their WTP. As in Study 1, the positive emotion
of liking plays a central role in addressing WTP and purchase
Finally, though not hypothesized, retailers are also interested
in learning more about the effects of the MBG design elements
on WTP and purchase intentions. Therefore we calculated the
total effects of the design elements. The values in Table 5 reveal
472 T. Suwelack et al. / Journal of Retailing 87 (4, 2011) 462–478
Table 5
Results of Study 2.
Path CoefficientaT-valueaHypothesis supported
Effects of MBG design elements
Test duration (TD)WTP .154 2.62**
TD PI .068 1.19
TD MBG credibility .084 1.64** H4a: yes
Return conditions (RC) WTP .144 2.58***
RC PI .249 4.69***
RC MBG credibility .554 12.8*** H4b: yes
TD ×SE WTP .135 .899
TD ×SE PI .125 .859
TD ×SE MBG credibility .170 1.31*H5a: yes
RC ×SE WTP .060 .472
RC ×SE PI .015 .118
RC ×SE MBG credibility .445 4.08*** H5b:no
Model 1bModel 2cModel 3dModel 4e
Path Coefficient T-value Coefficient T-value Coefficient T-value Coefficient T-value
Results of mediator analysis: MBG credibility
MBG credibility performance risk .428 11.0*** .523 13.3***
MBG credibility financial risk .242 5.25*** .277 5.65***
MBG credibility regret .367 7.48*** .377 8.32***
MBG credibility liking .421 10.59*** .494 12.78***
MBG credibility WTP .376 8.25*** .137 2.42** .136 2.31***
MBG credibility PI .468 11.5*** .032 .743 .021 .448
Performance risk WTP .135 2.04** .139 2.14**
Financial risk WTP .014 .232 .015 .252
Regret WTP .029 .458 .030 .499
Liking WTP .270 4.15*** .276 4.36***
Performance risk PI .258 5.15*** .280 5.31***
Financial risk PI .073 1.56*.076 1.63*
Regret PI .167 3.42*** .187 3.79***
Liking PI .407 8.31*** .443 8.81***
Indirect effects Coefficient Z-score Hypothesis supported
MBG credibility perf. risk WTP .073 2.13** H6ai: yes
MBG credibility fin. risk WTP .004 .251 H6aii:no
MBG credibility regret WTP .011 .472 H6bi:no
MBG credibility liking WTP .136 3.95** H6bii : yes
MBG credibility WTP .216 5.63***
MBG credibility perf. risk PI .147 4.27*** H6ci: yes
MBG credibility fin. risk PI .021 1.45*H6cii: yes
MBG credibility regret PI .070 2.21** H6di: yes
MBG credibility liking PI .219 4.96*** H6dii: yes
MBG credibility PI .457 9.78***
aModel with design elements as predictors of WTP, PI, and credibility.
bPredictor relates to ultimate endogenous variable.
cPredictor relates to mediators.
dMediators relate to endogenous variables controlling for predictor variable.
eDirect path become insignificant in the presence of the mediators.
*p< .1 (one-tailed).
** p< .05 (one-tailed).
*** p< .01 (one-tailed).
a significant negative effect of both a short duration and strict
return conditions on WTP. In addition, strict return conditions
have a negative effect on purchase intentions.
General discussion
We have attempted to analyze the behavioral outcomes
of MBGs for different product categories and to identify
effective guarantee designs. In two studies, we examined the
effects of an MBG and its terms on consumer evaluations
and behavior in the context of eight search and experi-
ence goods. Moreover, we extend the literature on MBGs
by examining both cognitive and affective elements and
their impact on WTP and purchase intentions. Thus the two
studies contribute to retailing literature in several important
T. Suwelack et al. / Journal of Retailing 87 (4, 2011) 462–478 473
First, to the best of our knowledge, we are the first to show
that MBGs stimulate emotional responses in addition to cogni-
tive ones. Second, in both studies, we find that MBGs positively
affect consumers’ WTP and purchase intentions. Thus, MBGs
may increase sales and profits and can be used to gain a competi-
tive edge. Third, we investigate the effects of variations in MBG
designs and reveal that two widely used design elements, dura-
tion and return conditions, have important impacts on consumer
decision making. Fourth, empirical research has not investigated
the type of product for which MBGs will be most effective, but
we show with our representative data that they work well when
added to a search good, which constitutes a surprising effect,
considering that search goods can be effectively evaluated before
Theoretical contributions
In contrast with previous research, which has mostly adopted
an information economics approach focusing on the cognitive
outcomes of a guarantee (i.e., signaling product quality or reduc-
ing consumers’ risk perceptions), we extend knowledge in this
area by examining a largely ignored aspect, consumer emo-
tions. By finding affective outcomes of MBGs, our results nicely
extend studies that show a positive impact on emotions for other
promotional tools such as free gifts and coupons (Chandon et al.
2000; Lichtenstein, Netemeyer, & Burton 1990). Though our
investigation is restricted to the two emotions of anticipated
regret and liking we also respond to Puccinelli et al. (2009)
by detecting an additional retailer cue for positively influencing
consumer emotions (i.e., an MBG reduces anticipated regret and
increases liking). Furthermore, in both studies, we show that lik-
ing caused by an MBG strongly affects consumer behavior by
increasing consumers’ WTP and purchase intentions. Surpris-
ingly, the path coefficients also reveal that liking has a stronger
effect on important MBG outcomes than do risk perceptions, an
effect that cannot be detected if MBGs are regarded solely as risk
reducers. This finding confirms the important role of emotions
in consumer behavior, as frequently stated by researchers (e.g.,
Darke, Chattopadhyay, & Ashworth 2006; Pham 1998, 2004).
Even more supporting the latter conclusion, it is interesting
that consumer responses are positively affected when an MBG
is provided for search goods, and emotions strongly explain the
surprising effectiveness of an MBG in a search goods context.
Therefore, as expected for experience goods, our data also justify
the frequent use of MBGs to advertise search goods and highlight
the strong impact of MBGs on consumer evaluations for various
product types.
Anticipated regret had weaker or even no effects on the
dependent variables. The literature distinguishes between spon-
taneous/hedonic and more cognitive emotions, with their
different effects on consumer behavior (e.g., Ramanathan &
Williams 2007). According to Brown, Homer, and Inman (1998)
and Schwarz and Clore (1996), emotions that require a high
degree of cognitive elaboration have less impact on consumer
behavior. Liking is highly spontaneous and hedonic, but we cat-
egorize anticipated regret as a more cognitive emotion, because
for it to occur, a consumer must (cognitively) imagine future
negative purchase outcomes. Simonson (1992) shows that antic-
ipated regret is important for consumer behavior, but it likely
affects consumer behavior when the mental imagery of negative
purchase consequences gets enforced by the advertiser—which
our experimental design did not do. Simonson (1992) accounts
for the manipulation of mental imagery about negative con-
sequences, as we discuss in section “Limitations and further
With regard to the profit impact of MBGs, we confirm an
indirect link to WTP. Prior literature has noted this relation-
ship conceptually (Fruchter & Gerstner 1999); we empirically
demonstrate the effect as mediated by consumers’ risk percep-
tions and emotional responses. As explained by equity theory,
consumers are willing to “pay” the provider of an MBG to reduce
their costs by eliminating the negative consequences associated
with the purchase and providing emotional benefits. Therefore,
we also respond to Chaudhuri and Ligas’s (2009) call by pro-
viding another idea about how to influence consumers’ WTP.
As consistently demonstrated in Study 1 and 2, performance
risk and liking both affect consumers’ WTP; anticipated regret
and financial risk do not. Regarding the emotional antecedents
of WTP, we return to our preceding explanation: Liking as a
strong spontaneous and hedonic emotion has a stronger impact
on consumer behavior than does the more cognitive emotion of
anticipated regret. Financial risk pertains to two aspects in an
MBG context: First, an assurance against overpaying, because
an item can be returned after purchase if the consumer finds
a lower price for the same item at another store. However, a
consumer should not be willing to pay more for this assurance
to not overpay because the purchase would then no longer offer
the value of a lower price. Second, an MBG provides assurance
about the costs associated with product breakdown. However,
product breakdown is not likely to occur during the guarantee
duration, and these future costs often get strongly discounted, so
they may have no prepurchase effects. In contrast, performance
risk is closely related to quality perceptions which strongly affect
WTP (Steenkamp, Van Veerde, & Geyskens 2010). We therefore
suggest that WTP might be more affected by promotion-focused
goals which deal with obtaining positive outcomes (e.g., “what
quality will I get?”/“how well will the product perform?”) rather
than prevention-focused goals which deal with avoiding negative
outcomes (e.g., “how much (financial) risk is involved?”/“will
I feel regret after purchase?”) (e.g., Wang & Lee 2006; Zhu &
Meyers-Levy 2007).
We also address a previously unanswered question about
what constitutes an effective MBG design in terms of dura-
tion and return conditions. To reduce customer cheating, some
researchers have proposed higher hurdles (e.g., Chu et al. 1998),
though empirical studies offer no evidence of such unintended
outcomes (Wirtz & Kum 2004). Sellers frequently impose return
restrictions, but our results suggest that a short duration and/or
strict return conditions can be detrimental because they reduce
MBG credibility and, in turn, negatively affect cognitions, affect,
and intentions. Therefore, sellers should design less restrictive
MBGs to positively affect consumer behavior.
In contrast, for experience goods, stricter MBG return
conditions lead to significantly higher credibility perceptions
474 T. Suwelack et al. / Journal of Retailing 87 (4, 2011) 462–478
compared to search goods. This unexpected finding can be
explained by the evidence argument from service literature.
Sometimes consumers do not know why, where, when, and how
to participate in a specific process. That is, they might have only
limited process evidence that consists of process awareness and
process transparency (Fliess & Kleinaltenkamp 2004). Provid-
ing detailed information about their responsibilities for acting on
an MBG thus might increase their confidence about the return
procedure. This argument is in line with Cheema and Patrick
(2008), who state that restrictive designs of promotions may be
preferred to expansive ones because they are more precise and
therefore less ambiguous. Providing precise information should
be more important for experience goods when the consumer is
uncertain about purchase outcomes.
Managerial contributions
Our results have several important managerial implications.
MBGs reduce consumer’s risk perceptions, increase their emo-
tional response, and enhance their WTP and purchase intentions,
indicating that retailers may sell more units and demand higher
prices when offering MBGs. These results hold for various types
of goods, including search goods, which justifies the broad use
of MBGs in retailing and marketing practice and should even
stimulate further implementations. However, as financial risk
and anticipated regret are more strongly reduced and WTP and
purchase intentions are more strongly increased for experience
goods, retailers whose assortment consists of a large proportion
of experience goods may have an even greater incentive to offer
MBGs compared to retailers selling mainly search goods.
Moreover, MBG effects strongly depend on the specified
terms. When designing an MBG, managers should carefully
craft those terms to avoid negative effects on their credibil-
ity, which would result in greater risk perceptions and negative
emotional responses and thus affect behavioral outcomes. The
interaction of the design elements and type of product is impor-
tant, that is, effective MBG designs depend on the type of
product. Study 2 shows that strict return conditions result in
higher credibility perceptions for experience goods relative to
search goods. We argue this effect might reflect the greater
information stated in restrictive MBGs, which make the guar-
antee less ambiguous (Cheema & Patrick 2008). Therefore, in
the case of experience goods retailers should inform consumers
how to invoke a guarantee. Specifying the return conditions may
accomplish this goal and at the same time, decrease unintended
effects of an MBG in the form of customer cheating, in that
the return conditions provide factual and psychological return
hurdles. This specific result concerning the design of MBGs
for experience goods is interesting because conceptual research
on service recovery highlights that guarantee designs which
are easy to invoke are the most credible (Hart 1993; Hogreve
& Gremler 2009). Therefore, when stressing the informational
content of an MBG by implementing stricter return conditions,
retailers must be aware of possible negative reactions. They
might draw specific attention to the perceived fairness of the
guarantee offer as studies have found that the fairness of the
recovery procedure is important (Gelbrich & Roschk 2011; Tax,
Brown, & Chandrashekaran 1998). Accounting for perceived
procedural fairness when crafting the return conditions of an
MBG, retailers will be able to offer more effective guarantees.
With regard to duration, we find a negative effect of a short
duration on MBG credibility for search and experience goods.
Consumers perceive an MBG as adequate if it provides suffi-
cient time to evaluate the item; otherwise, credibility problems
arise. Thus, retailers should provide long MBG durations if they
want to influence purchase behavior positively. Comparing both
design elements the results unveil a significantly greater impor-
tance of the return conditions for consumer decision making
compared to the duration.
Therefore, with these results obtained from our representative
studies, managers should better understand the consumer impact
of MBGs and can make better decisions about how to design
MBGs according to the level of product risk involved and how
to price guaranteed items.
Limitations and further research
This study has several limitations that provide directions for
further research. First, to keep our model parsimonious, we
focused on two cognitive constructs (performance and financial
risk) and two affective constructs (regret and liking). However,
we did not examine other important variables, such as qual-
ity perceptions. Information economics regards the increase of
quality perceptions as a main effect of the guarantee signal
(Spence 1974), though d’Astous and Guèvremont (2008) do not
find support for this relationship in their empirical study. Thus,
further research needs to investigate the conditions in which an
MBG serves as a signal of high quality.
Second, our study provides only a first glimpse into the emo-
tional outcomes of MBGs as we only examine two of a number
of possible emotions. However, considering that we find strong
results for the positive emotion that we did study, further research
should identify other positive emotions that mediate consumer
responses to an MBG, such as emotions related to value expres-
sion (e.g., smart shopper feeling), exploration (e.g., arousal,
variety), or entertainment (e.g., amusement, aesthetic value).
Third, we note some limitations in our experimental design.
First, we did not measure actual purchase behavior, so we are
somewhat limited in our conclusions regarding the profit impact
of MBGs. The problem of measuring evaluations and intentions
toward a fictive scenario also might have contributed to the weak
effects of anticipated regret on WTP and purchase intentions.
Investigating the impact of MBGs in real shopping situations,
in which subjects must buy an item and incentives encourage
their real behavior, might reduce this problem. The second issue
regarding our experimental design might further explain the low
impact of regret on consumer intentions. In contrast with our
approach, Simonson (1992) manipulates mental imagery of the
negative consequences of a choice by telling only one group of
subjects that the superiority or inferiority of their choice, com-
pared with unchosen alternatives, would be revealed after their
choice decision. The participants in the different enforcement
of anticipated regret groups made different choice decisions,
which indicate the relevance of anticipated regret in choice tasks.
T. Suwelack et al. / Journal of Retailing 87 (4, 2011) 462–478 475
Third, our experimental design incorporated two products (ski
boots and anti-dandruff shampoo) that might be of less relevance
to some consumers. Although product involvement did not dif-
fer across the search and experience goods categories, future
research might investigate other products that are more homoge-
neous regarding this issue. Fourth, different MBG manipulations
might provide unique results. Additional research should investi-
gate whether different manipulations regarding MBG durations
or return conditions prove the robustness of our results. Other
MBG characteristics could be investigated as well. For exam-
ple, some retailers do not refund the full purchase price but
rather offer store credit (d’Astous and Guèvremont 2008). The
impact of these different compensation schemes on consumers
would be an interesting topic to study. The final issue regard-
ing our experimental design refers to the MBG presentation in
the experimental advertisements. For example, future research
could study whether additional MBG information (e.g., MBG
logo provided or not) changes the empirical results.
Fourth, our model does not examine the possible unintended
outcomes of MBGs. Some researchers have suggested that
lenient product return policies lead to abuses (e.g., McWilliams
& Gerstner 2006; Petersen & Kumar 2009). Harris (2008) finds
evidence of opportunistic consumer behavior in the form of
“retail borrowing.” Further research should determine whether
the presence of an MBG stimulates unethical behavior. It would
be particularly useful to discover which MBG terms facilitate
this behavior or how it might be repressed by the use of dif-
ferent MBG design elements. For example, Chu et al. (1998)
argue theoretically that partial refunds may help reduce con-
sumer opportunism. Empirical evidence for this notion is still
Appendix A. Manipulation checks and controls
Experience Qualities (Weathers et al. 2007)
1. It’s important for me to see this product to evaluate how well
it will perform. (“totally disagree/totally agree”)
2. It’s important for me to touch this product to evaluate how
well it will perform. (“totally disagree/totally agree”)
3. It’s important for me to test this product to evaluate how well
it will perform. (“totally disagree/totally agree”)
Search Qualities (Weathers et al. 2007)
1. I can adequately evaluate this product using only information
about the product’s attributes and features provided by the
retailer or the manufacturer. (“totally disagree/totally agree”)
2. I can evaluate the quality of this product simply by read-
ing information about the product. (“totally disagree/totally
MBG Presence (MBG manipulation check in Study 1)
1. Did the advertiser use an MBG? (“yes/no”)
MBG Return Conditions (first of two MBG manipulation
checks in Study 2)
1. I perceived return conditions stated in the money-back guar-
antee to be (“not strict at all/very strict”).
MBG Test Duration (second of two MBG manipulation
checks in Study 2)
1. I perceived the duration stated in the money-back guarantee
to be (“very long/very short”).
Inherent Product Risk (own items)
1. Buying this type of product involves high risk. (“totally dis-
agree/totally agree”)
2. Buying this type of product leaves me with high outcome
uncertainties. (“totally disagree/totally agree”)
Hedonic Versus Utilitarian Product Category Perceptions
(Chen, Kalra, & Sun 2009)
1. This product category is (“not at all hedonic/extremely hedo-
2. This product category is (“not at all utilitarian/extremely
Product Category Involvement (Flynn, Goldsmith, &
Eastman 1996)
1. In general I have a strong interest in this product category.
(“totally disagree/totally agree”)
2. This product category is very important to me. (“totally dis-
agree/totally agree”)
3. This product category matters a lot to me. (“totally dis-
agree/totally agree”)
Product Knowledge (Weathers et al. 2007)
1. Compared to people surrounding me, I am very knowledge-
able about this product. (“totally disagree/totally agree”)
2. Compared to friends, I have much experience with this prod-
uct. (“totally disagree/totally agree”)
Perceived Information Content of the Advertisement (Kalra
& Goodstein 1998)
The advertisement is (“not at all informative/very informa-
Adams, J. Stacy (1965), “Inequity in Social Exchange,” in Advances in Exper-
imental Social Psychology, Leonard Berkowitz ed. New York: Academic
Press, 267–99.
Agarwal, Sanjeev and R. Kenneth Teas (2001), “Perceived Value: Mediating
Role of Perceived Risk,Journal of Marketing Theory and Practice, 9 (4),
476 T. Suwelack et al. / Journal of Retailing 87 (4, 2011) 462–478
Anderson, Eric T., Karsten Hansen and Duncan Simester (2009), “The Option
Value of Returns: Theory and Empirical Evidence,” Marketing Science, 28
(3), 405–23.
Anderson, James C. and David W. Gerbing (1988), “Structural Equation Mod-
eling in Practice: A Review and Recommended Two-Step Approach,”
Psychological Bulletin, 103 (May), 411–23.
Bagozzi, Richard P., Mahesh Gopinath and Prashanth U. Nyer (1999), “The Role
of Emotions in Marketing,” Journal of the Academy of Marketing Science,
27 (2), 184–206.
Bagozzi, Richard P. and Youjae Yi (1988), “On the Evaluation of Structural
Equation Models,” Journal of the Academy of Marketing Science, 16 (1),
Bearden, William O. and Terence A. Shimp (1982), “The Use of Extrinsic Cues
to Facilitate Product Adoption,” Journal of Marketing Research, 19 (May),
Berkowitz, Eric N. and John R. Walton (1980), “Contextual Influences on Con-
sumer Price Responses: An Experimental Analysis,” Journal of Marketing
Research, 17 (August), 349–58.
Bettman, James R. (1973), “Perceived Risk and Its Components: A Model and
Empirical Test,Journal of Marketing Research, 10 (May), 184–90.
Biswas, Abhijit, Sujay Dutta and Chris Pullig (2006), “Low Price Guaran-
tees as Signals of Lowest Price: The Moderating Role of Perceived Price
Dispersion,” Journal of Retailing, 82 (3), 245–57.
Blinds (2010), 100% Satisfaction Guarantee ... or
Your Money Back!, (accessed December 28, 2010). Available at guar.html
Boulding, William and Amna Kirmani (1993), “A Consumer-Side Experimen-
tal Examination of Signaling Theory: Do Consumers Perceive Warranties
as Signals of Quality?,” Journal of Consumer Research, 20 (June),
Brown, Steven P., Pamela M. Homer and Jeffrey J. Inman (1998), “A Meta-
Analysis of Relationships between Ad-Evoked Feelings and Advertising
Responses,” Journal of Marketing Research, 35 (February), 114–26.
Camera Batteries (2010), JCPenney Digital Camera Battery,
(accessed December 28, 2010). Available at
Campbell, Margaret C. and Amna Kirmani (2000), “Consumers’ Use of Per-
suasion Knowledge: The Effects of Accessibility and Cognitive Capacity on
Perceptions of an Influence Agent,” The Journal of Consumer Research, 27
(June), 69–83.
Chandon, Pierre, Brian Wansink and Gilles Laurent (2000), “A Benefit Congru-
ency Framework of Sales Promotion Effectiveness,” Journal of Marketing,
64 (October), 65–81.
Chaudhuri, Arjun and Mark Ligas (2009), “Consequences of Value in Retail
Markets,” Journal of Retailing, 85 (September), 406–19.
Cheema, Amar and Vanessa M. Patrick (2008), “Anytime Versus Only:
Mind-Sets Moderate the Effect of Expansive Versus Restrictive Frames
on Promotion Evaluation,Journal of Marketing Research, 45 (August),
Chen, Tao A.O., Ajay Kalra and Baohong Sun (2009), “Why Do Consumers
Buy Extended Service Contracts?,” Journal of Consumer Research, 36
(December), 611–23.
Chu, Wujin, Eitan Gerstner and James D. Hess (1998), “Managing Dissatisfac-
tion: How to Decrease Customer Opportunism by Partial Refunds,Journal
of Service Research, 1 (November), 140–55.
d’Astous, Alain and Amélie Guèvremont (2008), “Effects of Retailer Post-
Purchase Guarantee Policies on Consumer Perceptions with the Moderating
Influence of Financial Risk and Product Complexity,Journal of Retailing
& Consumer Services, 15 (4), 306–14.
Darke, Peter R., Amitava Chattopadhyay and Lauren Ashworth (2006), “The
Importance and Functional Significance of Affective Cues in Consumer
Choice,” Journal of Consumer Research, 33 (3), 322–8.
Davis, Scott, Eitan Gerstner and Michael Hagerty (1995), “Money Back Guar-
antees in Retailing: Matching Products to Consumer Tastes,Journal of
Retailing, 71 (1), 7–22.
, Michael Hagerty and Eitan Gerstner (1998), “Return
Policies and the Optimal Level of ‘Hassle’,Journal of Economics and
Business, 50 (5), 445–60.
Erevelles, Sunil (1998), “The Role of Affect in Marketing,Journal of Business
Research, 42 (3), 199–215.
Fishbein, Martin and Icek Ajzen (1975), Belief, Attitude, Intentions and Behav-
ior: An Introduction to Theory and Research, Reading: Addison-Wesley.
Fliess, Sabine and Michael Kleinaltenkamp (2004), “Blueprinting the Service
Company: Managing Service Processes Efficiently,” Journal of Business
Research, 57 (April), 392–404.
Flynn, Leisa R., Ronald E. Goldsmith and Jacqueline K. Eastman (1996), “Opin-
ion Leaders and Opinion Seekers: Two New Measurement Scales,” Journal
of the Academy of Marketing Science, 24, 137–47 (Spring).
Ford, Gary T., Darlene B. Smith and John L. Swasy (1990), “Con-
sumer Skepticism of Advertising Claims: Testing Hypotheses from
Economics of Information,” Journal of Consumer Research, 16 (March),
Fornell, Claes and David F. Larcker (1981), “Evaluating Structural Equation
Models with Unobservable Variables and Measurement Error,” Journal of
Marketing Research, 18 (February), 39–50.
Franke, George R., Bruce A. Huhmann and David L. Mothersbaugh (2004),
“Information Content and Consumer Readership of Print Ads: A Comparison
of Search and Experience Products,” Journal of the Academy of Marketing
Science, 32 (1), 20–31.
Friestad, Marian and Peter Wright (1994), “The Persuasion Knowledge Model:
How People Cope with Persuasion Attempts,Journal of Consumer
Research, 21 (1), 1–31.
Fruchter, Gila E. and Eitan Gerstner (1999), “Selling With ‘Satisfaction Guar-
anteed’,” Journal of Service Research, 1 (May), 313–23.
Fuchs, Christoph, Emanuela Prandelli and Martin Schreier (2010), “The Psy-
chological Effects of Empowerment Strategies on Consumers’ Product
Demand,” Journal of Marketing, 74 (January), 65–79.
Gelbrich, Katja and Holger Roschk (2011), “A Meta-Analysis of Organizational
Complaint Handling and Customer Responses,”Journal of Service Research,
14 (February), 24–43.
Grewal, Dhruv, Kent B. Monroe and R. Krishnan (1998), “The Effects of
Price-Comparison Advertising on Buyers’ Perceptions of Acquisition Value,
Transaction Value, and Behavioral Intentions,” Journal of Marketing, 62
(April), 46–59.
Grewal, Dhruv, Jeanne L. Munger, Gopalkrishnan R. Iyer and Michael Levy
(2003), “The Influence of Internet-Retailing Factors on Price Expectations,”
Psychology & Marketing, 20 (6), 477–93.
Harris, Lloyd C. (2008), “Fraudulent Return Proclivity: An Empirical Analysis,
Journal of Retailing, 84 (4), 461–76.
Hart, Christopher W.L. (1993), Extraordinary Guarantees, New York: AMA-
Heiman, Amir, Bruce McWilliams, Zhao Jinhua and David Zilberman (2002),
“Valuation and Management of Money-Back Guarantee Options,” Journal
of Retailing, 78 (3), 193–205.
Heiman, Amir, Bruce McWilliams and David Zilberman (2001), “Demon-
strations and Money-Back Guarantees: Market Mechanisms to Reduce
Uncertainty,Journal of Business Research, 54 (1), 71–84.
Ho, Hillbun, Shankar Ganesan and Harmen Oppewal (2011), “The Impact of
Store-Price Signals on Consumer Search and Store Evaluation,Journal of
Retailing, 87 (June), 127–41.
Hoch, Stephen J. and John Deighton (1989), “Managing What Consumers Learn
from Experience,” Journal of Marketing, 53 (April), 1–20.
Hogreve, Jens and Dwayne D. Gremler (2009), “Twenty Years of Service
Guarantee Research: A Synthesis,” Journal of Service Research, 11 (May),
Homburg, Christian, Nicole Koschate and Wayne D. Hoyer (2005), “Do Sat-
isfied Customers Really Pay More? A Study of the Relationship between
Customer Satisfaction and Willingness to Pay,” Journal of Marketing, 69
(April), 84–96.
Horton, Raymond L. (1976), “The Structure of Perceived Risk: Some Fur-
ther Progress,” Journal of the Academy of Marketing Science, 4 (4),
Huang, Peng, Nicholas H. Lurie and Sabyasachi Mitra (2009), “Searching for
Experience on the Web: An Empirical Examination of Consumer Behav-
ior for Search and Experience Goods,” Journal of Marketing, 73 (March),
T. Suwelack et al. / Journal of Retailing 87 (4, 2011) 462–478 477
Jain, Shailendra P. and Steven S. Posavac(2001), “Prepurchase Attribute Verifia-
bility,Source Credibility, and Persuasion,” Journalof Consumer Psychology,
11 (3), 169–80.
Kalra, Ajay and Ronald C. Goodstein (1998), “The Impact of Advertising Posi-
tioning Strategies on Consumer Price Sensitivity,” Journal of Marketing
Research, 35, 210–24. May.
Kaltcheva, Velitchka D. and Barton A. Weitz (2006), “When Should a Retailer
Create an Exciting Store Environment?,Journal of Marketing, 70 (January),
Kaplan, Leon B., George J. Szybillo and Jacob Jacoby (1974), “Components of
Perceived Risk in Product Purchase: A Cross-Validation,” Journal of Applied
Psychology, 59 (3), 287–91.
Kirmani, Amna and Akshay R. Rao (2000), “No Pain, No Gain: A Critical
Review of the Literature on Signaling Unobservable Product Quality,” Jour-
nal of Marketing, 64 (April), 66–79.
Kukar-Kinney, Monika and Rockney G. Walters (2003), “Consumer Per-
ceptions of Refund Depth and Competitive Scope in Price-Matching
Guarantees: Effects on Store Patronage,Journal of Retailing, 79 (3),
Kukar-Kinney, Monika, Rockney G. Walters and Scott B. MacKenzie (2007),
“Consumer Responses to Characteristics of Price-Matching Guarantees: The
Moderating Role of Price Consciousness,” Journal of Retailing, 83 (2),
Kukar-Kinney, Monika, Lan Xia and Kent B. Monroe (2007), “Consumers’
Perceptions of the Fairness of Price-Matching Refund Policies,” Journal of
Retailing, 83 (3), 325–37.
Lei, Jing, Ko de Ruyter and Martin Wetzels (2008), “Consumer Responses
to Vertical Service Line Extensions,” Journal of Retailing, 84, 268–80.
Lichtenstein, Donald R., Scot Burton and Richard G. Netemeyer (1997), “An
Examination of Deal Proneness Across Sales Promotion Types:A Consumer
Segmentation Perspective,Journal of Retailing, 73 (2), 283–97.
, Richard G. Netemeyer and Scot Burton (1990),
“Distinguishing Coupon Proneness from Value Consciousness: An
Acquisition-Transaction Utility,” Journal of Marketing, 54 (July),
McCollough, Michael A. and Dwayne D. Gremler (2004), “AConceptual Model
and Empirical Examination of the Effect of Service Guarantees on Post-
Purchase Consumption Evaluations,Managing Service Quality, 14 (1),
McWilliams, Bruce and Eitan Gerstner (2006), “Offering Low Price Guar-
antees to Improve Customer Retention,Journal of Retailing, 82 (2),
Mitchell, Vincent-Wayne (1999), “Consumer Perceived Risk: Conceptualisa-
tions and Models,” European Journal of Marketing, 33 (1/2), 163–95.
Mitra, Debanjan and Scott Fay (2010), “Managing Service Expectations in
Online Markets: A Signaling Theory of E-tailer Pricing and Empirical Tests,
Journal of Retailing, 86 (2), 184–99.
Moorthy, Sridhar and Kannan Srinivasan (1995), “Signaling Quality with a
Money-Back Guarantee: The Role of TransactionCosts,” Marketing Science,
14 (4), 442–66.
Naylor, Rebecca W., Rajagopal Raghunathan and Suresh Ramanathan (2006),
“Promotions Spontaneously Induce a Positive Evaluative Response,” Jour-
nal of Consumer Psychology, 16 (3), 295–305.
Nelson, Phillip (1974), “Advertising as Information,” Journal of Political Econ-
omy, 82 (4), 729–54.
Oliver,Richard L. and John E. Swan (1989), “Consumer Perceptions of Interper-
sonal Equity and Satisfaction in Transactions: A Field Survey Approach,
Journal of Marketing, 53 (April), 21–35.
Ostrom Amy, L. and Dawn Iacobucci (1998), “The Effect of Guarantees on
Consumers’ Evaluation of Services,Journal of Services Marketing, 12 (5),
Perdue, Barbara C. and John O. Summers (1986), “Checking the Success of
Manipulations in Marketing Experiments,” Journal of Marketing Research,
23 (November), 317–26.
Petersen, J. and Andrew V. Kumar (2009), “Are Product Returns a Necessary
Evil? Antecedents and Consequences,” Journal of Marketing, 73 (May),
Pham, Michel Tuan (1998), “Representativeness,Relevance, and the Use of Feel-
ings in Decision Making,” Journal of Consumer Research, 25 (September),
(2004), “The Logic of Feeling,” Journal of Consumer
Psychology, 14 (4), 360–9.
Podsakoff, Philip M., Scott B. MacKenzie, Jeong-Yeon Lee and Nathan P. Pod-
sakoff (2003), “Common Method Biases in Behavioral Research: A Critical
Review of the Literature and Recommended Remedies,Journal of Applied
Psychology, 88 (October), 879–903.
Posselt, Thorsten, Eitan Gerstner and Dubravko Radic (2008), “Rating E-Tailers’
Money-Back Guarantees,” Journal of Service Research, 10 (February),
Preacher, Kristopher J. and Andrew F. Hayes (2008), “Asymptotic and Resam-
pling Strategies for Assessing and Comparing Indirect Effects in Multiple
Mediator Models,” Behavior Research Methods, 40 (3), 879–91.
Puccinelli, Nancy M., Ronald C. Goodstein, Dhruv Grewal, Robert Price, Priya
Raghubir and David Stewart (2009), “Customer Experience Management in
Retailing: Understanding the Buying Process,” Journal of Retailing, 85 (1),
Ramanathan, Suresh and Patti Williams (2007), “Immediate and Delayed Emo-
tional Consequences of Indulgence: The Moderating Influence of Personality
Type on Mixed Emotions,” Journal of Consumer Research, 34 (August),
Roseman, Ira J. (1991), “Appraisal Determinants of Discrete Emotions,Cog-
nition and Emotion, 5 (3), 161–200.
Schwarz, N. and Gerald L. Clore (1983), “Mood, Misattribution, and Judgments
of Well-Being: Informative and Directive Functions of Affective States,”
Journal of Personality and Social Psychology, 45 (3), 513–23.
and (1996), “Feelings and Phe-
nomenal Experiences,” in Social Psychology: Handbook of Basic Principles,
Higgings Edward T. and Arie Kruglanski, eds. New York: Guilford Press
pp. 433–65.
Shieh, Shiou (1996), “Price and Money-Bank Guarantees as Signals of Prod-
uct Quality,Journal of Economics and Management Strategy, 5 (3),
Shih, Eric and Hope J. Schau (2011), “To Justify or Not to Justify: The Role
of Anticipated Regret on Consumers’ Decisions to Upgrade Technological
Innovations,Journal of Retailing,, doi:10.1016/j.jretai.2011.01.006
Shimp, Terence A. and William O. Bearden (1982), “Warranty and Other Extrin-
sic Cue Effects on Consumers’ Risk Perceptions,Journal of Consumer
Research, 9 (June), 38–46.
Simonson, Itamar (1992), “The Influence of Anticipating Regret and Responsi-
bility on Purchase Decisions,” Journal of Consumer Research, 19, 105–18.
Spence, Michael (1974), Market Signaling: Informational Transfer in Hiring
and Related Screening Processes, Cambridge: Harvard University Press.
Srivastava, Joydeep and Nicholas H. Lurie (2004), “Price-Matching Guarantees
as Signals of Low Store Prices: Survey and Experimental Evidence,Journal
of Retailing, 80 (2), 117–28.
Steenkamp, Jan-Benedikt E.M., Harald J. Van Veerde and I. Geyskens (2010),
“What Makes Consumers Willing to Pay a Price Premium for National
Brands over Private Labels?,Journal of Marketing Research, 47, 1011–24.
Sullivan, Elisabeth A. (2009), “Guaranteeing Success,Marketing News, 43 (8),
Swinyard, William R. (1981), “The Interaction Between Comparative Advertis-
ing and Copy Claim Variation,” Journal of Marketing Research, 18, 175–86.
Tax, S.S., Stephen W. Brown and Murali Chandrashekaran (1998), “Customer
Evaluations of Service Complaint Experiences: Implications for Relation-
ship Marketing,” Journal of Marketing, 62 (April), 60–76.
Tripp, Carolyn, Thomas D. Jensen and Les Carlson (1994), “The Effects
of Multiple Product Endorsements by Celebrities on Consumers’ Atti-
tudes and Intentions,” Journal of Consumer Research, 20 (March),
Tsiros, Michael and Vikas Mittal (2000), “Regret: A Model of Its Antecedents
and Consequences in Consumer Decision Making,” Journal of Consumer
Research, 26 (March), 401–17.
478 T. Suwelack et al. / Journal of Retailing 87 (4, 2011) 462–478
Wang, Jing and Angela Y. Lee (2006), “The Role of Regulatory Focus in
Preference Construction,” Journal of Marketing Research, 43 (February),
Weathers, Danny, Subhash Sharma and Stacy L. Wood (2007), “Effects of
Online Communication Practices on Consumer Perceptions of Performance
Uncertainty for Search and Experience Goods,” Journal of Retailing, 83 (4),
Wirtz, Jochen and Doreen Kum (2004), “Consumer Cheating on Service Guar-
antees,” Journal of the Academy of Marketing Science, 32, 159–75. Spring.
Wood, Stacy L. (2001), “Remote Purchase Environments: The Influence of
Return Policy Leniency on Two-Stage Decision Processes,” Journal of Mar-
keting Research, 38 (May), 157–69.
Zeelenberg, Marcel and Rik Pieters (1999), “Comparing Service Delivery to
What Might Have Been: Behavioral Responses to Regret and Disappoint-
ment,” Journal of Service Research, 2 (August), 86–97.
Zeithaml, Valarie A. (1988), “Consumer Perceptions of Price, Quality, and
Value: A Means-End Model and Synthesis of Evidence,” Journal of Mar-
keting, 52 (July), 2–22.
Zeithaml, Valarie A., Leonard L. Berry and A. Parasuraman (1996), “The Behav-
ioral Consequences of Service Quality,Journal of Marketing, 60, 31–46.
Zhu, Rui and Joan Meyers-Levy (2007), “Exploring the Cognitive Mechanism
that Underlies Regulatory Focus Effects,Journal of Consumer Research,
34 (1), 89–98.
... Nowadays, using a return policy has been proposed as a good strategy to encourage customers to purchase and boost the sales in e-commerce [4][5][6]. A return policy could have a high cost for managers. ...
Full-text available
In online purchasing, customers may return products due to dissatisfaction with the quality of the product, and receive a refund based on the return policy, which is determined by online distributors. Online distributors can offer generous policies to attract more customers, but at the cost of reducing total profits. In this paper, the effect of the pricing and quality of complementary products (products sold together with other items) in online selling under the return policy is investigated. For this purpose, a mathematical model is developed to obtain optimal values for selling price, refund amount, and quality of products. Based on analytical results, a solution algorithm is proposed to solve the numerical examples and perform sensitivity analysis. Findings reveal that, while increasing the sensitivity of demand with respect to the refund amount, the price, quality, and refund on returned products should be increased. In addition, the online distributor should increase the quality of products when customers are more sensitive to the quality of products. Among other results, the selling price is shown to be negatively affected by demand elasticity with respect to price. In this situation, the online distributor should reduce the quality level and the refund amount for returned products to avoid a sharp decline in profit. In addition, when the quality cost is high, the price and quality should be decreased and the refund amount unchanged.
... Studies have shown that allowing product returns reduces consumers' perception of purchase risk ( Hawes and Lumpkin, 1986 ) and increases consumers' self-reports of purchase intentions ( Van den Poel and Leunis, 1999 ). Yet another set of studies have shown that both money leniency and time leniency of a return policy can signal product or service quality and increase consumers' value perceptions ( Wang, 2009 ), cost fairness perceptions ( Bower and Maxham III, 2012 ) or willingness to pay ( Wang, 2009 ;Suwelack et al., 2011 ). Janakiraman, Syrdal, and Freling (2016) conducted a meta-analysis of the effect of return policy leniency on consumer purchase and return behavior and showed that overall leniency impacts purchases (or purchase intentions) more than returns (or return intentions). ...
The fast-paced growth of e-commerce is rapidly changing consumers’ shopping habits and shaping the future of the retail industry. While online retailing has allowed companies to overcome geographic barriers to selling and helped them achieve operational efficiencies, offline retailers have struggled to compete with online retailers, and many retailers have chosen to operate both online and offline. This paper presents a review of the literature on the interaction between e-commerce and offline retailing, highlighting empirical findings and generalizable insights, and discussing their managerial implications. Our review includes studies published in more than 50 different academic journals spanning various disciplines from the inception of the internet to present. We organize our paper around three main research questions. First, what is the relationship between online and offline retail channels including competition and complementarity between online and offline sellers as well as online and offline channels of an omnichannel retailer? Under this question we also try to understand the impact of e-commerce on market structure and what factors impact the intensity of competition /complementarity. Second, what is the impact of e-commerce on consumer behavior? We specifically investigate how e-commerce has impacted consumer search, its implications for price dispersion, and user generated content. Third, how has e-commerce impacted retailers’ key managerial decisions? The key research questions under this heading include: (i) What is the impact of big data on retailing? (ii) What is the impact of digitization on retailer outcomes? (iii) What is the impact of e-commerce on sales concentration? (iv) What is the impact of e-commerce and platforms on pricing? And (v) How should retailers manage product returns across online and offline channels? Under each section, we also develop detailed recommendations for future research which we hope will inspire continued interest in this domain.
... • Customer behavior covers topics such as whether or not serial returners make better customers (Petersen & Kumar, 2009;Ramanathan, 2011;Griffis et al., 2012;Hjort et al., 2013) and the effects of pricing policies and conditions on returns (Suwelack et al., 2011;Bower & Maxham III, 2012;Kim & Wansink, 2012;Pei et al., 2014;Janakiraman et al., 2016;Shang et al., 2017). ...
This work considers an omnichannel retailer selling a product with market demand uncertainties to customers in different regions through an e-store and multiple brick-and-mortar stores. The retailer also manages online product returns by selecting appropriate return policies. A robust omnichannel pricing and ordering optimization model is proposed with two, i.e., a full-refund and a no-refund, return policies. The online demand and the offline demand depend on the prices of the e-store and the brick-and-mortar stores, where the online demand also depends on the refund of the e-store. A linearization technique is adopted to deal with nonlinearity of the model. A data-driven robust optimization approach is used to construct uncertainty sets based on available historical data using support vector clustering to handle demand uncertainties. Furthermore, the proposed model is transformed into an approximate mixed integer linear programming model which can be solved by using commercial software. An electronics retailer in China is used as a case study to illustrate the effectiveness and practicality of the proposed model and the solution method. A comparison with the box uncertainty set reveals that the data-driven uncertainty set is less conservative and performs better by obtaining higher profit for the retailer. Furthermore, sensitive analysis results indicate that return policies and the return rate of the product affect the optimal pricing and ordering decisions and the total profit.
While consumers favor control over the price to pay, they also are deterred by the effort involved and tend to underpay in participative pricing contexts. We identify anticipated regret and deliberation as potential psychological determinants for consumers' willingness to pay (WTP). In 3 studies, we show that anticipated regret and deliberation increase the WTP for the preferred option. Our single-paper meta-analysis (includes 11 studies) demonstrates the combined effect of anticipated regret and deliberation on valuation across varying manipulations, purchase context, goods, and decision contexts. By detailing how anticipated regret and deliberate increases WTP, we offer a potential solution for underpaying and conclude by discussing directions for future research and managerial implications.
Purpose Face masks have been integrated into daily life and come to signify different meanings due to the fight against the COVID-19 pandemic. Drawing on social perception and attribution theories, this paper investigates the possible additional benefits of face mask use in the service marketplace by understanding how consumers react to the new social norm. Design/methodology/approach Four experimental studies were run across different service contexts. Study 1 examines consumers’ evaluation of a service promotion when an employee wears a face mask; Study 2 focuses on the impact of face mask usage on salesperson credibility and service satisfaction; Studies 3 and 4 investigate the consequences of not using face masks on consumers’ intention to spread positive word-of-mouth for the service provider. Findings The results revealed that the presence of a face mask in a service promotion determined a higher level of service liking, while in a service encounter, it led to a higher level of salesperson credibility, which then positively affected consumer satisfaction. Finally, the non-utilization of a face mask negatively affected consumer intention to spread positive word-of-mouth about the service provider, even when the social norm is to not wear one. Originality/value The manuscript adds to research on salespeople appearance and tries to understand consumers’ reactions toward face mask use in the services sector, as, to the best of the authors’ knowledge, few empirical studies in the service marketing literature have investigated the topic so far. Findings also provide useful insights that can further promote companies’ adoption of face masks beyond the COVID-19 emergency.
We develop a model for a supply chain with a manufacturer and a retailer facing customer returns, in which information about the rate of customer returns is obtained and held asymmetrically. The retailer implements a full refund policy and accepts returns, collecting and retaining data on the rate of returns. The manufacturer is uninformed about the returns rate. The manufacturer has four strategies as to returns rate information: not acquiring the information, acquiring the information through a side payment contract, offering a buyback policy in order to participate in the returns process, or encroaching with a direct channel. We identify the manufacturer’s optimal information strategy and discuss the retailer’s preferences. We also identify win–win outcomes for the manufacturer and the retailer. We find that a buyback policy helps the manufacturer to obtain returns rate information for free when the salvage value is the same for the manufacturer and the retailer. When the manufacturer’s salvage value is relatively low, there always exists a region where the manufacturer does not try to acquire the information. The retailer may benefit from the cases in which the manufacturer does not acquire the information, adopts a side payment contract, or institutes a buyback policy, but the retailer is always worse off from the manufacturer’s channel encroachment strategy. Both the manufacturer and the retailer may have consistent preferences as to the manufacturer’s strategy.
In this paper, we examine a manufacturer's choice of outsourcing contractor, either a competing contractor or a non-competing contractor. Both the manufacturer and the competing contractor (which also produces a product in its own brand) face customer returns. We find that the manufacturer's optimal outsourcing strategy depends strongly on two factors: the efficiency of production and sale of its brand relative to that of the competing contractor in its own brand, and the ratio of the qualities of the two brands. The competing contractor, on the other hand, always prefers to produce for the manufacturer. Interestingly, we find that when the manufacturer chooses to outsource to the non-competing contractor, both the wholesale and retail prices of the manufacturer's product decrease, while they increase if the competing contractor is chosen. In addition, the competing contractor may be chosen even if it charges a higher wholesale price than the non-competing contractor does. We find that the manufacturer and the competing contractor should offer money-back guarantees if they can efficiently recover value from any returns. We further show that when a non-competing contractor is chosen, money-back guarantees offered by the manufacturer and the competing contractor can benefit at least one firm and may even achieve a win-win situation. When the competing contractor is chosen, both the manufacturer and the competing contractor can either benefit (Pareto improvement) or suffer (prisoner's dilemma) from money-back guarantees. These results are different from those in existing studies in the literature.
This study investigates how online retailers should solve the problem of product value uncertainty through an ex ante product value delivery policy. We construct an endogenous matching probability model depending on the ex ante strategic choice among various product value delivery measures, such as displaying product information online, opening product reviews, building a virtual showroom, and building a physical showroom. This model is different from the exogenous matching probability assumption in the literature. When the matching probability is exogenously given, the retail price reduction policy is commonly used to mitigate product value uncertainty. In contrast, when retailers can determine the matching probability through the ex ante strategic choice for the product value delivery measures, increasing the retail price to rely more on ex ante value delivery may be optimal in the mitigation of a more serious product value uncertainty. In addition, we examine the interaction between the ex ante product value delivery and the ex post return policy. The results show that the adoption of the ex post mitigation policy may encourage or discourage the adoption of the ex ante mitigation policy depending on the cost input for product value delivery and the customers’ surplus loss caused by a mismatch.
With the development of e-commerce, online shopping has become increasingly common, and as a result, consumers inevitably encounter the problem of returns. Therefore, pricing, return policy and return insurance strategy have attracted considerable research attention. In this paper, we construct four models to study pricing, return policy and return insurance strategy. We show that when a product's net residual value is greater than or equal to zero, online retailers should offer a money-back guarantee (MBG) return policy; however, they do not have to offer free return insurance because the latter does not increase their market share and profit, nor does it increase consumer surplus. The optimal strategy of insurance providers is unaffected by whether the policyholder is an online retailer or a consumer and should be neutral, which helps insurance providers gain the trust of policyholders. Consumers should buy products only when online retailers offer an MBG return policy; however, consumers should not do so if the online retailer provides free return insurance.
Full-text available
Evidence from past research and insights from an exploratory investigation are combined in a conceptual model that defines and relates price, perceived quality, and perceived value. Propositions about the concepts and their relationships are presented, then supported with evidence from the literature. Discussion centers on directions for research and implications for managing price, quality, and value.
Full-text available
The authors conduct meta-analyses of relationships involving positive and negative ad-evoked feelings to determine (1) whether the effects of positive and negative feelings on advertising responses are symmetrical or asymmetrical and bipolar or bidimensional; (2) whether study design characteristics influence the strength of effects of positive and negative feelings on ad and brand attitudes; (3) whether moderator effects are differential for positive and negative feelings; and (4) whether relationships are generalizable across different study designs, product types, and media. Using the aggregated data, the authors assess competing predictions of three alternative theoretical perspectives: bipolarity, generalized asymmetry, and contingent asymmetry. Analyses indicate that positive and negative feelings have contingently asymmetrical effects on advertising responses. Study conditions related to subjects' cognitive processing sets moderated the effects of negative feelings on ad and brand attitudes but generally did not moderate the effects of positive feelings. The authors also demonstrate the utility of bootstrapping as an advantageous new way of conducting moderator analyses in meta-analysis.
The statistical tests used in the analysis of structural equation models with unobservable variables and measurement error are examined. A drawback of the commonly applied chi square test, in addition to the known problems related to sample size and power, is that it may indicate an increasing correspondence between the hypothesized model and the observed data as both the measurement properties and the relationship between constructs decline. Further, and contrary to common assertion, the risk of making a Type II error can be substantial even when the sample size is large. Moreover, the present testing methods are unable to assess a model's explanatory power. To overcome these problems, the authors develop and apply a testing system based on measures of shared variance within the structural model, measurement model, and overall model.
The study reports the results of two experiments designed to test the role of perceived quality, perceived sacrifice, and perceived risks on consumers’ perceptions of product value. The results of the experiments involving 530 undergraduate students support the hypothesized model that suggests that perceived quality and perceived sacrifice mediate the relationships between extrinsic cues-such as price, brand, store name, and country-of-origin-and perceived performance risk and financial risk. In addition, the results indicate that perceived performance risk and financial risk mediate the relationships that perceived quality and perceived sacrifice have with perceived value. Since the findings of this research suggest both performance risk and financial risk can be reduced (thus increasing perceived value) if perceived quality is high, firms have an incentive to increase consumers’ quality perceptions by means of brand name, store, country-of-origin and price.
In this article, we provide guidance for substantive researchers on the use of structural equation modeling in practice for theory testing and development. We present a comprehensive, two-step modeling approach that employs a series of nested models and sequential chi-square difference tests. We discuss the comparative advantages of this approach over a one-step approach. Considerations in specification, assessment of fit, and respecification of measurement models using confirmatory factor analysis are reviewed. As background to the two-step approach, the distinction between exploratory and confirmatory analysis, the distinction between complementary approaches for theory testing versus predictive application, and some developments in estimation methods also are discussed.