forthcoming ijrm volume 32 #1 (2015)
TRANSCRIPT
Elsevier Editorial System(tm) for International Journal of Research in Marketing Manuscript Draft Manuscript Number: IJRM-D-13-00177R2 Title: The effects of promotional frames of sales packages on perceived price increases and repurchase intentions Article Type: Full Length Article Corresponding Author: Dr. Hsuan-Yi Chou, Ph. D. Corresponding Author's Institution: National Sun Yat-sen University First Author: Hsin-Hsien Liu, Ph. D. Order of Authors: Hsin-Hsien Liu, Ph. D.; Hsuan-Yi Chou, Ph. D. Abstract: This article explores how framing a promotional package (i.e., presenting the promotion as a bundle versus as a free gift) influences consumers' price assignments to the individual items in the package. It examines the potential influences of framing on consumers' perceptions of price increases and repurchase intentions after the promotion expires. The findings show that when a package contains two different products, consumers in the free gift (bundle) condition assign a higher price to the focal (supplementary) product, perceive a smaller price increase, and exhibit higher repurchase intentions toward the focal (supplementary) product after the promotion ends. If the promotional package contains two identical products, the free gift promotion generates higher perceived price increases and lower repurchase intentions than a price bundle, through similar price assignment mechanisms. An incentive-compatible experimental design finds that a free gift promotion lowers consumers' willingness to pay for the target product compared with a price bundle promotion. The findings of this research have significant implications for both framing research and marketing practice.
Forthc
oming
IJRM V
olume 3
2 #1 (
2015
)
The effects of promotional frames of sales packages on perceived price
increases and repurchase intentions
Hsin-Hsien Liu
Department of Asia-Pacific Industrial and Business Management
National University of Kaohsiung
700, Kaohsiung University Rd., Nanzih District, Kaohsiung City, Taiwan, R.O.C.
Tel: 886-7-5919796
Fax: 886-7-5919430
Email: [email protected]
Hsuan-Yi Chou (corresponding author)
Institute of Communications Management
National Sun Yat-sen University
70, Lien-hai Rd., Kaohsiung City, Taiwan, R.O.C.
Tel: 886-922995278
Fax: 886-7-5254969
Email: [email protected]
==========================================================
ARTICLE INFO Article history: First received in June 12, 2013 and was under review for 5½ months. Area Editor: Zeynep Gurhan-Canli
============================================================
* This work was supported by the National Science Council, Taiwan [NSC
101-2410-H-390-001-].
*Title Page _FINAL
Forthc
oming
IJRM V
olume 3
2 #1 (
2015
)
1
The effects of promotional frames of sales packages on perceived price
increases and repurchase intentions
ABSTRACT
This article explores how framing a promotional package (i.e., presenting the promotion as a
bundle versus as a free gift) influences consumers’ price assignments to the individual items
in the package. It examines the potential influences of framing on consumers’ perceptions of
price increases and repurchase intentions after the promotion expires. The findings show that
when a package contains two different products, consumers in the free gift (bundle) condition
assign a higher price to the focal (supplementary) product, perceive a smaller price increase,
and exhibit higher repurchase intentions toward the focal (supplementary) product after the
promotion ends. If the promotional package contains two identical products, the free gift
promotion generates higher perceived price increases and lower repurchase intentions than a
price bundle, through similar price assignment mechanisms. An incentive-compatible
experimental design finds that a free gift promotion lowers consumers’ willingness to pay for
the target product compared with a price bundle promotion. The findings of this research have
significant implications for both framing research and marketing practice.
Keywords: Sales frame, Bundle, Free gift, Mental accounting, Reference price
*FINAL APPROVEDClick here to view linked References
Forthc
oming
IJRM V
olume 3
2 #1 (
2015
)
2
1. Introduction
By combining two or more products and selling them at one price (Guiltinan, 1987;
Popkowski Leszczyc & Häubl, 2010), a bundle can reduce consumers’ time, transaction, and
monetary costs (Yadav & Monroe, 1993). Marketing promotions often feature such bundles;
for example, convenience stores in Taiwan frequently offer promotions like “buy a sandwich
and a drink for only NT$X.” McDonald’s consumers similarly appreciate value meals, which
include a hamburger, fries, and a drink for a fixed price. Other common promotions feature
free gifts (Kamins, Folkes, & Fedorikhin, 2009; Raghubir, 2004), such as the free toy
McDonald’s sometimes includes with the purchase of a Happy Meal. In some cases, the free
gift promotion represents an alternative version of a bundle promotion. For example, the “buy
shampoo and conditioner for $X” is economically equivalent to “buy shampoo for $X and
receive conditioner for free.”
Marketing managers need to know which type of promotion is more likely to increase
short-term sales, even if the cost is the same. Then they must calculate whether the different
frames influence future sales of the items in the package, after the promotion has expired. To
answer these problems, marketing managers first need to discern which promotional frame is
more attractive to consumers, then they must determine which frame produces fewer negative
responses when the promotional campaign ends and consumers potentially perceive a price
increase (compared with the previous promotional price). These consumer perceptions of
Forthc
oming
IJRM V
olume 3
2 #1 (
2015
)
3
price increases are extremely important because they can affect their repurchase intentions.
Previous studies explore many theoretical and empirical aspects of free gifts (Laran &
Tsiros, 2013; Nunes & Park, 2003), as well as of bundles (Johnson, Herrmann, & Bauer, 1999;
Lee, Tsai, & Wu, 2011; Popkowski Leszczyc & Häubl, 2010; Yadav, 1994; Yadav & Monroe,
1993). Several studies contrast the effects of promotional type with the effects of simple price
discounts (Chandran & Morwitz, 2006; Chou & Lien, 2012; Harlam, Krishna, Lehmann, &
Mela, 1995; Raghubir, 2004). Kamins et al. (2009) even compare the short-term attractiveness
of bundles versus free gift promotions, and they conclude that the relative attractiveness does
not differ. However, despite an increasing number of studies that explore the long-term effects
of promotions (e.g., Jedidi, Mela, & Gupta, 1999; Mela, Gupta, & Lehmann, 1997; Palmeira
& Srivastava, 2013; Tsiros & Hardesty, 2010), no research compares the long-term effects of
bundles versus free gift promotions, which is a critical issue for many marketing managers.
Therefore, this study investigates the relative effects of framing (i.e., bundle vs. free gift) on
consumers’ post-promotional responses to a focal product (i.e., the item with the highest
reservation price or highest perceived value) and a supplementary product (i.e., the one with a
lower reservation price or lower perceived value) (Kamins et al., 2009; Raghubir, 2004; Yadav,
1994).
To develop a more comprehensive theory of the long-term effects framing has on
individual items in a package, we integrate mental accounting theory, which suggests that
Forthc
oming
IJRM V
olume 3
2 #1 (
2015
)
4
different semantic characteristics influence the way consumers assign prices to individual
items (Soman & Cheema, 2001; Soman & Gourville, 2001). We infer that the framing of the
promotional packages influences price assignments towards the focal and supplementary
items. Building on research that examines reference pricing (e.g., Mayhew & Winer, 1992;
Winer, 1986), we further propose that the framing of the package influences consumers’
perceptions of price increases and repurchase intentions for the individual items after the
promotion expires.
We also test the framing effects of price bundles and free gift promotions in which the
two products are identical. Sinha and Smith (2000) suggest that consumers prefer “buy one,
get one free” (i.e., free gift promotion) rather than “buy two, get 50% off” (i.e., price bundle),
because in the latter condition consumers believe they must buy two items to get the discount,
whereas the former condition promises them a discount when buying just one item. However,
we do not know how such frames influence the future sales of the product subsequent to the
promotion’s expiration. By considering how consumers assign prices to the first and second
items of a package, we propose that the use of free gift or price bundle promotions have
different influences on perceived price increases and future purchase intentions.
To test our propositions, we develop and run a series of systematic experiments. In the
pilot study and Experiments 1 and 2, we use different designs to compare the effects of
framing on two different products contained within a single promotional package.
Forthc
oming
IJRM V
olume 3
2 #1 (
2015
)
5
Experiments 3 and 4 introduce different samples and experimental scenarios to examine the
framing effects on promotional packages that contain two identical products. The findings of
all five studies contribute significantly to our understanding of the long-term effects of
framing promotional packages. Furthermore, the results can help marketers select more
suitable frames and encourage the careful optimization of cross-branding and cooperative
promotional designs.
2. Theoretical background, hypotheses and conceptual framework
2.1. Mental accounting, sales frame, and price assignment
Consumers use cognitive rules to organize, evaluate, and record their financial activities
(Thaler, 1985, 1999). They keep track of their financial activities and record various costs and
benefits in specific mental accounts, then they use accounting rules, explicitly and implicitly,
during their decision making process. Such mental accounting often involves assigning
activities to specific accounts, though people can categorize a single event in various ways.
Soman and Cheema (2001) and Soman and Gourville (2001) suggest that the semantic
characteristics of transactions can influence consumers’ assignments of prices to specific
items. In particular, consumers can flexibly allocate a total price across several benefits in
bundling situations because one price covers all the benefits. In a price bundle (e.g., buy two,
get 50% off), consumers might allocate the total price to one specific item and treat the other
Forthc
oming
IJRM V
olume 3
2 #1 (
2015
)
6
as free, or they might divide the price across the two items equally or unequally. Prices and
resultant benefits are unambiguous for unbundled transactions, but this connection becomes
ambiguous in price bundling settings. Thus, we infer that different promotional frames affect
consumers’ price assignments to the specific items in a package.
Chandran and Morwitz (2006) propose that consumers ignore the monetary value of a
free item, even when its price is readily available. Similarly, Raghubir (2004) argues that a
free gift paired with another focal product causes consumers to infer the free gift (i.e.,
supplementary item) has little value or that the focal product is overpriced. According to
Darke and Chung (2005), when consumers consider a free gift rather than a price discount
promotion, they are more likely to note the full price of the focal item when they make their
price–quality inferences about that focal product. In such cases, consumers might not assign
the supplementary item a price. In contrast, a bundle promotion obscures the single item price
(Raghubir, 2005), which increases the difficulty consumers have when attempting to infer the
prices of the supplementary items. As such, the regular price of a supplementary item might
have a greater impact when consumers infer its transaction value in a bundle promotion.
Consumers considering a bundle (vs. free gift) promotion should assign a higher price to the
supplementary item, whereas in a free gift (vs. bundle) promotion they likely treat the
supplementary item as complimentary and assign a higher price to the focal item. Following
this logic, we advance the following hypothesis:
Forthc
oming
IJRM V
olume 3
2 #1 (
2015
)
7
H1: Consumers assign higher prices to the focal product in a free gift promotion than
in a bundle promotion, but they assign higher prices to the supplementary product in a
bundle promotion compared to a free gift promotion.
2.2. Sales frames, reference prices, perceived price increases, and repurchase intentions
After promotions expire, the prices of the products return to their regular levels or they
might rise to even higher levels. An issue of great practical importance is to understand the
effects of the prior promotional package framing on consumers’ perceptions of the
post-promotion price increases and their repurchase intentions for the individual items (focal
and supplementary).
Previous studies (Kalyanaram & Winer, 1995; Mayhew & Winer, 1992; Rajendran &
Tellis, 1994; Winer, 1986) suggest that consumers form internal reference prices using the
price they paid for the same product in the past. Consumers also develop personal price
forecasting rules that enable them to compare a historical, deal-induced price against a normal
price. Because the prices consumers assign to the focal and supplementary items might
represent the past price paid for specific items in a package, consumers can use these as
reference prices when they consider price increases in their repurchase decision (Mayhew &
Forthc
oming
IJRM V
olume 3
2 #1 (
2015
)
8
Winer, 1992).1 Therefore, consumers considering purchasing a bundle (vs. free gift)
promotion should assign lower prices to the focal item, and those consumers considering a
free gift (vs. bundle) promotion should assign lower prices to the supplementary item. We
predict that consumers considering the bundle (free gift) promotion perceive a higher price
increase for the focal (supplementary) item after the promotion expires because their
reference price point is lower.
Classic economic theory also suggests that price is negatively correlated to demand.
Kalyanaram and Winer (1995) find a consistent, significant negative impact of reference
prices on consumer demand. Similarly, Homburg, Koschate, and Totzek (2010) propose that
price increases reduce consumer purchases. As such, consumers considering a bundle (free
gift) promotion should express lower repurchase intentions for the focal (supplementary) item
because they perceive greater price increases after the promotion expires. We provide the
following related hypothesis:
1 Many scholars suggest that people form decisions, judgments, and responses using multiple reference points
(e.g., Sullivan & Kida, 1995; Yim, Chan, & Hung, 2007). Mayhew and Winer (1992) argue that consumers use
multiple reference points, including internal (e.g., past prices paid) and external (e.g., regular prices) reference
prices to evaluate prices in their purchase decisions, which then affect the probability of purchase. Ordóñez,
Connolly, and Coughlan (2000) find that when multiple reference points are available, people compare them
separately to each referent and integrate their separate findings. Thus, even if multiple reference prices exist (e.g.,
past price paid and regular prices), each one might have a unique impact on consumers’ price perceptions for the
target product and affect their repurchase decisions. In this study, regular product prices are the same across the
bundle and free gift promotions, so they should have the same impact in each promotional frame and can be
ignored. We thus focus on the effects of one reference price (i.e., past prices paid/prices assigned to the product).
Forthc
oming
IJRM V
olume 3
2 #1 (
2015
)
9
H2: Consumers perceive higher price increases and exhibit lower repurchase
intentions toward the focal (supplementary) product after the end of a bundle (free gift)
promotion.
If these anticipated framing effects on consumer repurchase intentions result from
different price assignments and accompanying perceptions of the price increases on the focal
and supplementary products, the perceived price increases should mediate the effects of
promotional frames on repurchase intentions. We formalize this concept in the following
hypothesis:
H3: Perceived price increases mediate the effects of promotional frames (bundle vs.
free gift) on repurchase intentions for specific items in the sales package.
Fig. 1 shows our conceptual framework and the relationships between H1-H3. H1
describes how framing influences consumers’ price assignments to specific items in the
promotional package. Specially, consumers assign higher prices to the focal product and
lower prices to the supplementary product in a free gift promotion compared to a similar
bundle promotion. H2 explains how consumers’ prior price assignments influence perceived
price increases and repurchase intentions. The lower the price consumers assign to the item,
the higher the perceived price increase and the lower their repurchase intentions. The
Forthc
oming
IJRM V
olume 3
2 #1 (
2015
)
10
perceived price increases reflect the mediating roles between the promotional frames and
consumer repurchase intentions (H3). When two products in the package are identical (the
bottom section of Fig. 1), similar price assignment mechanisms influence consumers’
perceived price increases, repurchase intentions, repurchase decisions, and willingness to pay.
We discuss these issues in greater detail, including introducing H4 and H5, in later sections.
Fig. 1 near hear
We conduct one pilot study and four experiments to examine the above propositions. The
focal and supplementary items in the package are different in the pilot study and Experiments
1 and 2. In Experiments 3 and 4 we use two identical items in the package. In addition,
participants in Experiment 4 are not students, whereas the other experiments use
undergraduate and graduate student participants. In Experiment 1, the focal and
supplementary items are different brands. In the pilot study and Experiment 2, the items have
the same brand. In real world promotions, some companies use two items with the same brand
to form the promotional package (e.g., buy Dove shampoo, get Dove facial cleaner free), and
some use differently branded items (e.g., 7-Eleven offers sandwiches from one brand with
drinks of another in its promotional bundles). Participants in the pilot study encounter price
increases for the target items after the promotional period expires, but in the other studies the
Forthc
oming
IJRM V
olume 3
2 #1 (
2015
)
11
prices return to their initial prices. This design is meaningful in practice because promotions
usually feature a deadline after which consumers must buy the products at regular prices,
which can provoke negative responses. This design is conceptually consistent with
Kalyanaram and Winer’s (1995) proposition that consumers regard a return to the “normal”
price as a price increase because promotions lower their reference prices. These varying
designs enhance the generalizability and applicability of the results of this research for both
theory and practice.
3. Pilot study
We conducted a pilot study to examine H1-H2 using a within-subject experimental
design to provide an initial control for individual variance and address consumer preferences
(Wiseman & Levin, 1996). We gave two task scenarios to the participants (N=65). In the first
task, participants considered two target persons who encountered different promotions: Mr. Li
(bundle) encountered a “Buy D shampoo (700 ml, NT$200) and D facial cleaner (100 g,
NT$100) for NT$200” promotion, whereas Mr. Chen (free gift) purchased “Buy D shampoo
(700 ml, NT$200) for NT$200, get D facial cleaner (100 g, NT$100) free.” Using a projection
method, participants imagined Mr. Li’s and Mr. Chen’s situations and then answered two
questions about price assignments between the two items (Table 1, lines 2-3) and assigned
prices to the shampoo and facial cleaner for Mr. Li and Mr. Chen (Table 1, line 4). In the
Forthc
oming
IJRM V
olume 3
2 #1 (
2015
)
12
second task, participants read that the promotion had expired and both the shampoo and the
facial cleaner underwent a 20% price increase. Participants then responded to four items about
the perceived price increases and their repurchase intentions (Table 1, lines 5-8).
Table 1 near here
The results in Table 1 show that most participants believed that Mr. Chen (free gift)
would assign a higher price to the shampoo (focal product) than Mr. Li (bundle) (63.1% vs.
3.1%; χ2(2, N=65) = 35.11, p < .01), but Mr. Li (bundle) would assign a greater price to the
facial cleaner (supplementary product) than Mr. Chen (free gift) (67.7% vs. 4.6%; χ2(2, N=65)
= 39.72, p < .01). Moreover, participants indicated that Mr. Chen (free gift) would allocate a
higher price, on average, to the shampoo than would Mr. Li (bundle) (NT$186.17 vs.
NT$129.71; t(1,64) = 10.27, p < .01), but Mr. Li (bundle) would assign a higher price to the
facial cleaner than would Mr. Chen (free gift) (NT$87.23 vs. NT$26.13; t(1,64) = 9.84, p
< .01). These results support H1.
Furthermore, most participants believed that Mr. Li (bundle) would perceive a higher
price increase for the shampoo (70.8% vs. 4.6%; χ2(2, N=65) = 44.89, p < .01) but would
perceive a lower price increase for the facial cleaner (3.1% vs. 81.5%; χ2(2, N=65) = 69.40, p
< .01) than Mr. Chen (free gift). Accordingly, most participants believed Mr. Li (bundle)
Forthc
oming
IJRM V
olume 3
2 #1 (
2015
)
13
would have lower repurchase intentions for the shampoo (6.2% vs. 63.1%; χ2(2, N=65) =
31.78, p < .01) but higher repurchase intentions for the facial cleaner (80.0% vs. 1.5%; χ2(2,
N=65) = 66.49, p < .01) than Mr. Chen (free gift). On the whole, these results are consistent
with H2.2
4. Experiment 1A: Framing effects on repurchase intentions
Experiment 1A uses a between-subjects design to retest the effects of sales frames (i.e.,
H1 and H2) to enhance the robustness of the pilot study’s experimental results. Additionally,
this experiment tests the mediating role of perceived price increases (i.e., H3).
4.1. Design and procedure
Experiment 1A was a 2 (frames: bundle vs. free gift) × 2 (targets of repurchase: focal vs.
supplementary) between-subjects design. The shampoo and the facial cleaner represented the
focal and supplementary items respectively. We randomly assigned the participants (N=107)
to one of four experimental cells. After an introduction, they read either a bundle or a free gift
promotion scenario, which we reproduce here:
2 These results cannot be attributed to errors in the price assignment tasks. Although 12 participants assigned
more or less than NT$200 for either Mr. Li (bundle) or Mr. Chen (free gift), on average the respondents assigned
a total price of NT$216.94 for Mr. Li (bundle), which was not significantly different from that assigned to Mr.
Chen (free gift; NT$212.31, t(1,64) = .91, p > .1). Eliminating these data did not influence the results. We
conducted similar analyses and excluded such an alternative explanation in Experiments 1A and 2. We did not
conduct relevant analyses in Experiments 1B, 3, and 4 because all participants correctly assigned the total price
between the two items in Experiment 3, whereas we did not ask participants to assign the price between the
items in Experiments 1B and 4.
Forthc
oming
IJRM V
olume 3
2 #1 (
2015
)
14
[Task 1: Frames: Bundle [Free Gift]] Imagine that Mr. Chen has run out of both his
shampoo and shower gel. He knows that the price of shampoo (700 ml) generally ranges
from NT$120 to NT$220, and the price of shower gel (1000 ml) generally ranges from
NT$140 to NT$200. Mr. Chen goes to his regular store and finds that it is conducting a
promotion: “Buy J shampoo (700 ml; NT$160) and S shower gel (1000 ml; NT$160) for
NT$160 [buy J shampoo (700 ml; NT$160) for NT$160, get S shower gel (1000 ml;
NT$160) for free].” Mr. Chen has heard that J shampoo and S shower gel perform well,
so he decides to accept the promotion.
We set the prices of the supplementary and focal products to be the same so that any
statistical significance cannot be attributed to price differences between the products.
Then participants evaluated Mr. Chen’s price assignments to the shampoo and shower gel,
similar procedure in the pilot study. After this, participants completed a post-promotion
repurchase decision task, which they read in the following scenarios:
[Task 2: Targets of repurchase: Focal [Supplementary]] After using the products for two
months, Mr. Chen has developed positive evaluations of both J shampoo and S shower
gel. He is about to exhaust his supply of shampoo [shower gel], so he intends to buy more
from the same store. When he arrives, Mr. Chen notices that the sales promotion has
expired. The shampoo and shower gel are now sold separately. The price for J shampoo
Forthc
oming
IJRM V
olume 3
2 #1 (
2015
)
15
[S shower gel] is NT$160.
Using a projection method, participants answered four repurchase intention items
(Cronbach’s α = .91): (1) “Mr. Chen is willing to repurchase the J shampoo/S shower gel”; (2)
“it is highly possible for Mr. Chen to repurchase the J shampoo/S shower gel”; (3) “under the
condition of a price increase, Mr. Chen would still consider repurchasing the J shampoo/S
shower gel”; and (4) “it is very likely that Mr. Chen will pay NT$160 for the J shampoo/S
shower gel.” We measured the perceived price increase with three items (Cronbach’s α = .73):
(1) “Mr. Chen perceives that the price of J shampoo/S shower gel has increased”; (2) “it is
highly possible that Mr. Chen will consider another shampoo/shower gel”; and (3) “it is
possible that Mr. Chen will make his final decision only after he investigates the price of other
brands of shampoos/shower gels.” The second and third items account for previous studies’
findings, which show that when consumers encounter an unfair price increase, they delay their
purchase decision and search for more information (Bell & Bucklin, 1999; Mazumdar, Raj, &
Sinha, 2005). All items used seven-point Likert scales (1 = “strongly disagree”; 7 = “strongly
agree”).3
3 To validate the items measuring repurchase intentions and perceived price increases, we performed a factor
analysis (Kaiser-Meyer-Olkin = .85 > .70; chi-square of Bartlett’s test = 429.13, p < .05) with a principal
component analysis and Varimax rotation. Two factors with eigenvalues greater than 1.0 were extracted,
accounting for 73.62% of the total variance. The four repurchase intention items loaded on factor 1, and the three
perceived price increase items loaded on factor 2, with loadings greater than .61, and no item cross-loaded above
an absolute value of .35, in support of the good construct validity of the items. We conducted similar analyses for
Experiments 1B, 2, and 3 and replicated the factor structure.
Forthc
oming
IJRM V
olume 3
2 #1 (
2015
)
16
4.2. Results and discussion
The results in Table 2 (columns 3-4) show that participants in the bundle promotion
condition assigned a lower price to the shampoo (NT$100.64 vs. NT$123.89; t(1,105) = -3.27,
p < .01), but assigned a higher price to the shower gel (NT$65.97 vs. NT$37.96; t(1,105) =
3.58, p < .01) than participants in the free gift promotion condition. These results further
support H1.
Table 2 near here
An analysis of variance (ANOVA) showed a significant two-way interaction effect
between frames and repurchase targets on perceived price increases (F(1,103) = 58.70, p < .01)
(Fig. 2, Panel A), though neither main effect was significant (ps > .1): The average perceived
price increase in the bundle promotion was higher for the shampoo (5.47 vs. 3.88, p < .01) but
lower for the shower gel (4.35 vs. 5.31, p < .01) than those in the free gift promotion.
Fig. 2 near here
Another ANOVA for repurchase intentions showed that the main effect of the targets was
significant (F(1,103) = 9.95, p < .01), though the effect of the frames was not (p > .1).
Forthc
oming
IJRM V
olume 3
2 #1 (
2015
)
17
Repurchase intentions were higher for the shampoo than for the shower gel (4.66 vs. 4.03),
which indirectly supported our manipulation of the focal (i.e., shampoo) and supplementary
(i.e., shower gel) products. That is, consumers differentiated the focal and the supplementary
products, even though their prices were the same. The focal product elicited a higher
reservation price and higher perceived value (Kamins et al., 2009; Raghubir, 2004), and
participants exhibited higher repurchase intentions for the focal product than for the
supplementary product, even when the products both returned to their regular prices.
Furthermore, the frames × targets interaction effect was statistically significant (F(1,103) =
32.32, p < .05) (Fig. 2, Panel B): Repurchase intentions for the shampoo were lower in the
bundle than in the free gift promotion (4.15 vs. 5.32, p < .05); repurchase intentions for the
shower gel were higher in the bundle than in the free gift promotion (4.66 vs. 3.55, p < .05).
Thus, the results support H2.
When we included perceived price increases as a covariate in the analysis of covariance
(ANCOVA), its effect on repurchase intentions was significant (F(1,102) = 20.97, p < .01),
but the frames × targets interaction became weaker, though its effect was still significant
(F(1,102) = 4.83, p = .03) (Table 3, columns 2). In other words, perceived price increases
partially mediated the frames × targets interaction effect on repurchase intentions, supporting
H3.4
4 We also followed Baron and Kenny’s (1986) regression procedures to test the mediation of the focal and
supplementary products respectively. We also compared the effect of frames on repurchase intention (i.e., the test
Forthc
oming
IJRM V
olume 3
2 #1 (
2015
)
18
Table 3 near here
5. Experiment 1B: Framing effects on repurchase intentions without prices assignments
Although both the pilot study and Experiment 1A supported H1-H3, one might argue that
asking the participants to first assign prices to the two items in the package and then measure
their perceived price increases and repurchase intentions is not realistic. In the real world,
consumers might not assign prices. Therefore, the previous results might be artifacts of the
experimental design. To exclude the possibility of experimental artifacts, we had the
participants in Experiment 1B report their perceived price increase and repurchase intentions
without first receiving item price assignments.
5.1 Design and procedure
The design and procedure were the same as Experiment 1A, except that we removed the
question about price assignments. Furthermore, to exclude the possibility of order effects, we
counterbalanced the order of the questions relating to the perceived price increase and
repurchase intentions. Experiment 1B (N=202) thus used a 2 (frames: bundle vs. free gift) × 2
of beta differences) when the perceived price increase was included as a predictor in the regression equation,
which follows Clogg, Petkova, and Shihadeh’s (1992) approach. The results were consistent with the ANCOVA
analyses. Similar analyses were conducted in other experiments.
Forthc
oming
IJRM V
olume 3
2 #1 (
2015
)
19
(targets of repurchase: focal vs. supplementary) × 2 (question order: perceived price increase
first vs. repurchase intention first) between-subjects design.
5.2 Results and Discussion
The results were similar to Experiment 1A. Two separate 2 (frames) ×2 (targets of
repurchase) × 2 (question order) ANOVAs showed that the frames × targets interaction effects
on perceived price increase (Cronbach’s alpha=0.70) and repurchase intention (Cronbach’s
alpha=0.90) were both significant (F(1,194) = 8.78 and 17.14, respectively, ps < .01). The
main effect of question order was marginally significant (F(1,194) = 3.23, p = .08) for the
perceived price increase but not significant for the repurchase intention (p > .1); however,
question order did not moderate other main effects or interaction effects. Therefore, we pooled
the data across two orders on Table 2 (columns 3-4). The average perceived price increase in
the bundle promotion was higher for the shampoo (5.21 vs. 4.65, p = .01), but marginally
lower for the shower gel (4.72 vs. 5.16, p = .06) than in the free gift promotion (Fig. 3, Panel
A). Compared with participants in the free gift promotion condition, participants in the bundle
condition showed lower repurchase intentions for the shampoo (4.12 vs. 4.91, p < .01) and
higher repurchase intentions for the shower gel (4.63 vs. 4.07, p = .02) (Fig. 3, Panel B).
These results again support H2.
Forthc
oming
IJRM V
olume 3
2 #1 (
2015
)
20
Fig. 3 near here
When we included perceived price increases in the ANCOVA model, its effect on
repurchase intentions was significant (F(1,193) = 89.20, p < .01), but the frames × targets
interaction became weaker, though its effect was still significant (F(1,193) = 6.35, p = .01)
(Table 3, columns 3). That is, perceived price increases partially mediated the frames × targets
interaction effect on repurchase intentions, supporting H3.
6. Experiment 2: Framing effects with lower promotional benefit levels
In the pilot study and Experiment 1, we used relatively high promotional benefits (33%
in the pilot study and 50% in Experiment 1), in line with Grewal, Marmorstein, and Sharma’s
(1996) and Hardesty and Bearden’s (2003) assertions that consumers do not process
information extensively when promotional benefits are low. Raghubir (1998) claims that
consumers typically expect price promotions in the range of 20%–40%. However, in some
real-world settings, companies may use lower promotional benefits. Therefore, to test the
robustness of our results, we lowered the promotional benefits in Experiment 2. However we
predict that the promotional frames should still influence perceived price increases and
repurchase intentions, even if the effect diminishes due to less extensive consumer
information processing.
Forthc
oming
IJRM V
olume 3
2 #1 (
2015
)
21
6.1. Design, participants, and procedure
The 2×2 design was similar to that of Experiment 1A, except that we used shampoo as
the focal item (NT$200) and hand sanitizer (NT$50) as the supplementary item, both from the
same brand, and the promotional benefit level was only 20%. To avoid the possibility that
participants would ignore the small promotional benefit, we also raised the price of the
shampoo slightly to NT$200 and used a low-priced hand sanitizer as the supplementary item.
The procedure, measures, and other details were similar to Experiment 1A. Participants
(N=224) also noted whether they had bought the shampoo and the hand sanitizer (yes/no)
themselves. Finally, they indicated whether Mr. Chen would repurchase the target item
(yes/no). This latter measure helps clarify the effects of promotional frames on consumers’
repurchase decision and expands the impact of framing beyond repurchase intentions
discussed in the previous experiments.
6.2. Results and discussion
Most participants (91.1%) had bought the shampoo; more than half (51.8%) had bought
the hand sanitizer, so the targets were generally familiar. According to the results in Table 2
(columns 5-6), participants in the bundle promotion condition assigned a lower price to the
shampoo (NT$168.18 vs. NT$178.03; F(1,222) = 10.31, p < .01), but assigned a higher price
Forthc
oming
IJRM V
olume 3
2 #1 (
2015
)
22
to the hand sanitizer (NT$31.36 vs. NT$21.20; F(1,222) = 13.89, p < .01) than did
participants in the free gift promotion condition. These results support H1. However, the
assigned prices for the shampoo significantly increased for both the bundle and free gift
promotions compared with the values in Experiment 1A. This might reflect the higher price
for the shampoo or the relatively lower total discount (NT$50) in Experiment 2.
Two ANOVAs showed the frames × targets interaction effect was marginally significant
on perceived price increase (Cronbach’s α = .71; F(1,220) = 3.13, p = .08) and statistically
significant on repurchase intentions (Cronbach’s α = .90, F(1,220) = 13.66, p < .01) (Fig. 4,
Panels A and B). Compared with the free gift promotion, the average perceived price increase
in the bundle promotion was directionally higher for shampoo (5.16 vs. 5.00, p > .1), but
marginally lower for the hand sanitizer (4.57 vs. 4.91, p = .10). Accordingly, participants
showed lower repurchase intentions for the shampoo (4.84, vs. 5.26, p < .05), but higher
intentions for the hand sanitizer (5.26 vs. 4.63, p < .01) in the bundle than in the free gift
promotion. We thus found additional support for H2.
Fig. 4 near here
Moreover, the frames had no impact on the perceived price increase, whereas the main
effect of the targets was significant (F(1,220) = 5.88, p < .05): Participants indicated a higher
Forthc
oming
IJRM V
olume 3
2 #1 (
2015
)
23
price increase for the focal than for the supplementary item (5.08 vs. 4.75). The relatively
small perceived price increase for the supplementary item likely reflected its low price
(NT$50, about US$1.7), such that participants hardly noticed when the product went back to
its regular price.
When we included perceived price increases in the ANCOVA model, its effect on
repurchase intentions was significant (F(1,219) = 42.13, p < .01), but the frames × targets
interaction became weaker, though its effect was still significant (F(1,219) = 10.42, p = .0014,
reduced from F(1,220) = 13.66, p = .0003) (Table 3, column 4). That is, perceived price
increases partially mediated the frames × targets interaction effect on repurchase intentions,
supporting H3.
Finally, compared with participants in the bundle promotion, more participants in the free
gift promotion indicated that they would repurchase the shampoo (79.7% vs. 65.5%, χ2(1; n =
117) = 2.94, p = .09), but they were less likely to repurchase the hand sanitizer (56.9% vs.
79.6%, χ2(1; n = 107) = 6.22, p < .05).
Although the Experiment 2 results support H1–H3, the frames × targets interaction effect
on the perceived price increase and repurchase intentions diminished, compared with those in
Experiment 1A (see Table 2). Following Rosenthal’s (1991) meta-analytical procedures, these
difference were statistically significant between two studies (Z = 4.86 for the perceived price
increase and 2.42 for the repurchase intentions, respectively; ps < .05). The results were as
Forthc
oming
IJRM V
olume 3
2 #1 (
2015
)
24
expected.
7. Experiment 3: Framing effects in promotional packages with two identical products
To uncover framing effects on price assignments, perceived price increases, and
repurchase intentions when the sales packages contains identical products, Experiment 3
compared a price bundle (buy two, get 50% off) with a free gift (buy one, get one free)
promotion, both of which contained the same two items. Thus, participants could not
distinguish the focal from the supplementary product. Semantic cues should lead them to
average the prices in the price bundle promotion but assign the total prices in the free gift
promotion to the first item, then perceive the second item as free (Sinha & Smith, 2000).
Krishnamurthi, Mazumdar, and Raj (1992) propose that consumers use prices they have
paid in the past as internal reference prices. Thus consumers may assign prices to items in a
promotional package. However prices encountered more recently should have greater effects
on internal reference prices than distant ones (Mazumdar et al., 2005). The second item in a
promotional package is in closer proximity to repurchase decisions, so its assigned price
should exert a greater impact on consumers’ perceptions of price increases and repurchase
intentions. Because consumers assign a lower price to the second item in the free gift
promotion compared with a price bundle promotion, consumers in the free gift promotion
condition should perceive a greater price increase and exhibit lower repurchase intentions
Forthc
oming
IJRM V
olume 3
2 #1 (
2015
)
25
toward the product after the promotions expire. We state this concept formally here:
H4: When a promotional package contains two identical products, consumers perceive
higher price increases and exhibit lower repurchase intentions following the end of a
free gift promotion compared to a price bundle promotion. Fewer consumers in the
free gift promotion condition will decide to repurchase the product.
H4 is shown in the bottom of Fig. 1, indicating how framing influences consumers’ price
assignments between the first and second items, and then influences their perceived price
increases and repurchase intentions/decisions.
7.1. Design, participants, and procedure
We randomly assigned the 68 participants to either a price bundle (buy two, get 50% off)
or free gift (buy one, get one free) frame condition for a new brand (not real) of toothpaste
(160 g; NT$90). Similar to our previous experiments, participants indicated their assigned
prices to the first and second tubes of toothpaste. Then participants in both conditions learned
the promotion had expired and the toothpaste price had returned to NT$90. They then
indicated their repurchase intentions (4 items, Cronbach’s α = .90), perceived price increases
(3 items, Cronbach’s α = .73), and repurchase decision (buy/not buy) for the toothpaste. The
Forthc
oming
IJRM V
olume 3
2 #1 (
2015
)
26
other details were the same as Experiments 1 and 2.
7.2. Results and discussion
As shown in Table 2 (column 7), participants in the price bundle promotion condition
assigned a lower price to the first toothpaste (NT$58.94 vs. NT$78.43; F(1,66) = 17.26, p
< .01) but a higher price to the second toothpaste (NT$ 31.06 vs. NT$11.57; F(1,66) = 17.26,
p < .01) than those in the free gift promotion condition. In addition, compared with
participants in the price bundle promotion condition, participants in the free gift promotion
condition perceived higher price increases for the toothpaste (5.21 vs. 4.42; F(1,66) = 10.25, p
< .05), and then expressed lower repurchase intentions (3.29 vs. 4.58 F(1,66) = 22.94, p < .05).
Finally, fewer participants in the free gift promotion condition indicated a repurchase decision
for the toothpaste than in the price bundle promotion condition (42.9% vs. 66.7%, χ2(1; N =
68) = 3.88, p < .05). On the whole, the above results support H4.
To test if the assigned price of the second product mediated the effects of frames on
perceived price increases, two regression analyses separately showed that the frames (price
bundle promotion = 0; free gift promotion = 1) significantly influenced the price of the second
item negatively (β = -19.49, t(1,66) = -4.15, p < .01) and the perceived price increase
positively (β = .78, t(1,66) = 3.2, p < .01). Including the price of the second item as a predictor
in the regression equation, the impact of the assigned price of the second item on the
Forthc
oming
IJRM V
olume 3
2 #1 (
2015
)
27
perceived price increase was significant (β = -.02, t(1,65) = -3.26, p < .01), but the impact of
the frames was not (β = .40, t(1,65) = 1.57, p > .1), in support of our prediction.
We also considered a potential two-stage mediation. A regression of the frames, the price
of the second item, and perceived price increase on repurchase intentions revealed a
non-significant coefficient for the price of the second item (β = .01, t(1,64) = 1.59, p > .1), a
marked reduction in the coefficient of the frames (from β = -1.29, t(1, 66)= -4.79, p < .01 to β
= -.75, t(1,64) = -2.73, p < .01, though still significant), and a significant coefficient of the
perceived price increase (β = -.43, t(1,64) = -3.31, p < .01). Therefore, the perceived price
increase mediated the effects of both frames and the price of the second item on repurchase
intentions.5
These results thus indicate that participants used the assigned price of the second item as
a reference price to evaluate their repurchase decision. Consumers might also use the assigned
price of the first item as a reference price because it represents a mirrored aspect of the second
item (i.e., NT$90 minus the assigned price of the second item). However, if consumers used
the assigned price of the first item as a reference (i.e., predictor in the regression equation),
that would imply that a higher assigned price led to higher perceived price increases (t(1,66) =
4.41, p < .05) and lower repurchase intentions (t(1,66) = -4.52, p < .05), which is logically
false.
5 To reduce complexity, we only showed how the perceived price increase mediated the effect of frames on
repurchase intentions in Table 3 (column 5).
Forthc
oming
IJRM V
olume 3
2 #1 (
2015
)
28
8. Experiment 4: Incentive-compatible design with two identical products
Similar to Experiment 3, we tested framing effects between the price bundle and the free
gift promotions in Experiment 4, but we designed a more realistic experimental scenario and
relied on a non-student sample. This experiment had four main differences. First, we gathered
information about a real brand, Essential shampoo, for the experimental scenario. Essential is
a relatively new brand with a small market share in Taiwan compared with other well-known
brands (e.g., Dove, Head & Shoulders, Lux), so our realistic design reduces the possibility of
participants’ existing brand attitude influencing the results.
Second, we used an incentive-compatible design, in which participants had to indicate
their willingness to pay (WTP) for Essential shampoo after the promotion expired. Before
starting the experiment, we informed the participants that they would have a chance to receive
a bottle of Essential shampoo or its monetary equivalent as a gift, depending on whether their
WTP was higher or lower than the sales price. Becker, DeGroot, and Marschak’s (1964)
incentive-compatible design has subjects indicate the smallest amount of money they would
accept in exchange for the opportunity to play the lottery. Many studies in the field of risk and
uncertainty have further developed and modified this procedure (e.g., Chow & Sarin, 2001;
Curley, Yates, & Abrams, 1986; Liu, 2011). In an incentive-compatible design, participants
have to indicate what is more valuable for them and they must know they have chance to
Forthc
oming
IJRM V
olume 3
2 #1 (
2015
)
29
actually acquire it. This design is effective for eliciting consumers’ real preferences and WTP
because a false preference or WTP might reduce the participants’ benefits. For example, if a
participant’s stated WTP (e.g., NT$50) is lower than her real WTP (e.g., NT$100), she might
acquire the money when the sales price (e.g., NT$75) is between the real WTP and the stated
WTP. The acquired money (NT$75) is less valuable than their real evaluation or WTP for the
Essential shampoo (NT$100). In contrast, if the participant’s stated WTP (e.g., NT$150) is
higher than his real WTP, he might acquire the Essential shampoo when the sales price (e.g.,
NT$125) is between the real WTP and the stated WTP, which will be less valuable than the
sales price.
Third, we reduced task complexity, only measuring the participants’ WTP for Essential
shampoo, repurchase decisions, and some control variables.
Fourth, the participants in Experiment 4 were all adult office workers.
If the results of Experiment 3 were due to participants focusing more on the price of the
second item than the first, and if they used this price as a reference to evaluate the transaction
value, then the participants in Experiment 4 should be willing to pay more for the target
product in the price bundle than in the free gift promotion after the promotion ends.
Participants will have assigned higher prices to the second item in the price bundle. From this
concept we form the following hypothesis:
Forthc
oming
IJRM V
olume 3
2 #1 (
2015
)
30
H5: When a sales package contains two identical products, consumers in the price
bundle condition will be more willing to pay a higher price for target product than
consumers in the free gift promotion condition after the promotion ends.
H5 is shown on the bottom right section of Fig. 1, and the inference process is the same
as H4.
8.1. Design and participants
We recruited 91 adult office workers participating in four refresher courses held at two
different colleges for this experiment. We assigned two courses (one from each college) to the
price bundle promotion condition (n=49) and the other two courses to the free gift promotion
condition (n=42). Each class consisted of 15–27 members.
Two participants were excluded because they did not complete the questionnaire. The
valid respondents (mean age = 37.6 years) had been working a mean of 13.25 years, and 23
(25.8%) had used Essential shampoo. Most participants worked in business-related
occupations (32.6%), the public sector (21.3%), or service industries (15.7%).6
6 Using the colleges, frames, and their interactions as independent variables, ANOVA results showed that neither
the college nor the interaction had any impact on participants’ WTP (F(1,85) = 1.04; F(1,85) = .06, respectively,
ps > .1). Therefore, we ignored the differences between colleges and pooled the data for the same frames across
colleges in the formal analyses.
Forthc
oming
IJRM V
olume 3
2 #1 (
2015
)
31
8.2. Procedure
At the beginning of the experiment, an experimenter informed the participants that he
would randomly select three people form each group at the end of the experiment. These
participants would win either the shampoo or the money, depending on whether the sales price
of Essential shampoo was higher or lower than their stated WTP. Next, the experimenter
explained the procedure (Table 4) carefully and displayed three bottles of Essential shampoo
alongside a bucket. The experimenter explained that there were 199 number cards (numbered
1–199) in the bucket and that one participant would be invited to draw a card from the bucket
after the experiment. The number on the card would represent the sales price for Essential
shampoo. Following the explanation, one participant was invited to check the cards to
confirm the experimenter’s claims. Because the participants could view the shampoo bottles
and the 199 cards, respondents are likely to have accepted the legitimacy of the potential
prizes.
Table 4 near here
The experimenter first ensured that all participants understood the procedure and had no
questions, then had participants write their names on the three-page questionnaire (for the
chance of winning the prize). Then the participants encountered a prompt and a request to
Forthc
oming
IJRM V
olume 3
2 #1 (
2015
)
32
describe their responses to a recent “real” promotion (though the promotions were not real).
The scenarios were as follows:
[Frames: Price Bundle [Free Gift]] Imagine that you have run out of shampoo, so you
intend to buy more at a nearby store. At the store, you notice Essential shampoo (560ML,
NT$199) is conducting a promotion, “buy two, get 50% off [buy one, get one free].”
Using seven-point Likert scales (1 = “strongly disagree”; 7 = “strongly agree”),
participants assessed the perceived attractiveness of the promotion (“I feel the promotion is
attractive”), perceived product quality (“I feel the quality of Essential shampoo is good”), and
brand likeability (“I like Essential”). These three variables could influence participants’
attitudes toward Essential, which subsequently might influence their WTP. Participants also
indicated if they had ever used Essential shampoo (yes/no).
On the final page, we presented the WTP task, as follows:
[WTP Scenario] Imagine that after one and a half months you have used up your first
bottle of Essential shampoo. After another one and a half months, you are almost out of
the second bottle, so you intend to buy more shampoo. When you go to the store where
you previously bought the Essential shampoo, you notice the promotion has ended, and
Essential shampoo is now sold only as single bottles.
Forthc
oming
IJRM V
olume 3
2 #1 (
2015
)
33
With a reminder that their WTP would determine if they would acquire the target
shampoo or money, participants responded to the item, “What is the maximum dollar amount
that you are willing to pay for a single bottle of Essential shampoo now?” Finally, participants
responded to the prompt, “If the price of Essential shampoo went back to NT$199 (without
any discount), would you buy it? (yes/no).”7 After participants completed the questionnaires,
we awarded the prizes (see steps 4 and 5 in Table 4). All other details were the same as those
in Experiment 3.
8.3. Results and discussion
With regard to the control variables, the frames had no impact on perceived attractiveness
of the promotion (5.33 vs. 5.59, F(1,87) = .51, p > .1), perceived product quality (4.15 vs.
4.24, F(1,87) = .12, p > .1), or brand likability (4.02 vs. 4.15, F(1,87) = .18, p > .1).8
However, the dichotomous product usage experience variable significantly affected the
perceived attractiveness of the promotion (6.13 vs. 5.21, F(1,87) = 5.50, p < .05), perceived
product quality (4.96 vs. 3.92, F(1,87) = 11.84, p < .01), and brand likeability (5.09 vs. 3.73,
F(1,87) = 19.75, p < .01). These three variables might influence participants’ WTP for
Essential shampoo and confound the effects of the frames, so we averaged them (Cronbach’s
7 The experimenter emphasized that participants’ answers were irrelevant to the prize contest. 8 The finding that the frames did not influence perceived attractiveness is inconsistent with Sinha and Smith’s
(2000) and Smith and Sinha’s (2000) findings that consumers prefer “buy one, get one free” to “buy two, get
50% off.” Perhaps the promotions in the current study are so attractive in both frames (5.33 and 5.59 on a
seven-point scale) that they produced a ceiling effect.
Forthc
oming
IJRM V
olume 3
2 #1 (
2015
)
34
α = .84) to form a brand attitude score, which we included as a covariate in our analyses.
An ANCOVA showed that the frames (price bundle vs. free gift) had a significant impact
on participants’ WTP (F(1,86) = 13.72, p < .01), such that participants showed higher WTP in
the price bundle than in the free gift promotion (150.40 vs. 126.51), in support of H5.
Moreover, their brand attitudes had a significant impact on WTP (F(1,86) = 11.06, p < .01).
When the promotion expired, more participants in the price bundle condition indicated they
would repurchase the shampoo than participants in the free gift promotion condition (33.3%
vs. 14.6%, χ2(1, n=89) = 4.15, p < .05), which is consistent with the results in Experiment 3.
A logistic regression with the brand attitudes as a control variable showed that both
frames (χ2(1,n=89) = 4.74, p < 0.05) and brand attitudes (χ
2(1,n=89) = 7.75, p < 0.01)
significantly influenced participants’ repurchase decision. When WTP was included as a
predictor, another logistic regression showed that the frames had no effect on repurchase
decision (χ2(1,n=89) = 1.19, p > 0.1), though WTP and brand attitudes both had an effect
(χ2(1,n=89) = 9.75 and χ
2(1,n=89) = 5.20, respectively, both ps<0.03) (Table 3, final column).
Therefore, the frames influenced participants’ repurchase decision through WTP.
9. General discussion
9.1. Conclusions
According to our results, the framing of a promotional package influences how
Forthc
oming
IJRM V
olume 3
2 #1 (
2015
)
35
consumers assign prices to the individual items in the package, and framing also affects
consumers’ perceptions of price increases and repurchase intentions after the promotion
expires. Regardless of the promotional benefit level, if two different products appear in a
single promotional package, consumers assign higher prices to the supplementary item if they
encounter a bundle promotion, but they assign higher prices to the focal item if they encounter
a free gift promotion. After the promotion expires, consumers perceive higher price increases
and have lower repurchase intentions toward the focal (supplementary) item in the bundle
(free gift) settings. However, lower promotional benefit levels weaken the effects of framing
on perceived price increases and repurchase intentions. If the package contains two identical
products, a free gift promotion causes consumers to perceive higher price increases and thus
indicate lower repurchase intentions compared to price bundle promotions. The result is due
to consumers assigning lower prices to the second product, which constitutes the comparison
price basis for consumers’ future repurchase decisions. Furthermore, consumers show lower
WTP for a product in the free gift than in the price bundle promotion; therefore, the price
bundle promotion induces more consumers to decide to repurchase the product. Finally,
perceived price increases mediate the framing effects on repurchase intentions.
9.2. Theoretical implications
Previous research on promotional frames focuses on the short-term effects (Harlam et al.,
Forthc
oming
IJRM V
olume 3
2 #1 (
2015
)
36
1995; Lowe, 2010; Raghubir, 2004). We clarify the long-term effects of framing promotional
packages on consumers’ perceived price increases and repurchase intentions after a promotion
has expired. Our experimental designs, settings (promotional benefit levels), and target items
(facial cleaner, shower gel, hand sanitizer) varied across experiments, yet the experimental
results remained consistent. Furthermore, the prices for the focal and supplementary products
were the same in Experiment 1, so the results are not attributable to price differences.
Therefore our hypotheses receive robust support.
Whereas Sinha and Smith (2000) find that consumers prefer to “buy one, get one free”
rather than “buy two, get 50% off” promotions, our study specifies how consumers assign
different prices to the first and second items under two different promotional frames. We also
explicate how the assigned price for the second item influences consumers’ perceived price
increases, repurchase intentions, and WTP after the promotion expires. This study thereby
offers a distinctive, more comprehensive view and enlarges the scope for research into
promotional framing; it also defines the conditions in which a price bundle promotional frame
is more beneficial than the free gift promotional frame. The notion of consumers’ price
assignments, as implied by mental accounting theory (Thaler, 1985, 1999), has been scarcely
explored in previous studies (Jha-Dang, 2006). This study can help scholars more fully
understand the role of mental accounting in promotional settings. The mediation analysis in
Experiment 3 also shows that consumers regard the second item in both promotional frames
Forthc
oming
IJRM V
olume 3
2 #1 (
2015
)
37
as a basis for comparison in their repurchase decisions. This finding is meaningful for further
research into anchoring and reference prices.
Raghubir (2005) explores how the sales frames with two different products influence
consumers’ WTP for the supplementary item as a standalone product. Although both our
current study and Raghubir’s discuss the effects of promotional frames, our study is different
in several ways. First, using the value discounting hypothesis and persuasion knowledge
model (Friestad & Wright, 1994), Raghubir proposes that consumers infer that the production
cost of a product is low when it is offered for free (rather than included in a bundle), which
should reduce their WTP for that product as a standalone offering. In contrast, we focus on
consumers’ price assignments viewed through mental accounting theory and semantic cues,
and we detail the effects of sales frames for supplementary and focal items in promotional
packages. This theoretical approach provides a more complete explanation of consumers’
responses toward the focal and supplementary items in a package. For example, Raghubir
(2005) cannot explain why consumers pay more for a focal product included in a free gift
promotion than in a bundle promotion; our research identifies price assignments as the
explanatory mechanism. In addition, either product can be treated as the focal or
supplementary item when firms design promotional packages, so it is important to understand
price assignments between two items in different frames. Another key difference is that we
examine the effects of framing when the promotional package contains identical products. In
Forthc
oming
IJRM V
olume 3
2 #1 (
2015
)
38
so doing, we fill the gap left in Raghubir’s (2005) research and offer a more comprehensive
view that offers broad potential applications.
9.3. Managerial implications
Promotional packages that contain different products can improve consumers’ responses
and stimulate sales, but the termination of promotional campaigns may lead to negative
effects on sales of the products due to perceived price increases. If marketers want to sell a
focal (supplementary) product, they should use free gift (bundle) promotions, which generate
higher repurchase intentions because consumers perceive lower price increases when the
promotion expires. In addition, marketing managers should carefully determine if their
products represent the focal (supplementary) item in a bundle (free gift) promotion
undertaken with strategic partners. A bundle (free gift), compared with a free gift (bundle),
promotion may later produce more negative responses to the company’s product. Finally, a
free gift promotion might appear more attractive than a price bundle promotion when
consumers make purchase decisions about a specific product (Sinha & Smith, 2000), but the
offer can have greater negative impacts on future sales after the promotion ends. Marketing
managers must take this finding into consideration when determining the most suitable
framing for their promotions.
Forthc
oming
IJRM V
olume 3
2 #1 (
2015
)
39
9.4. Limitations and further research
This study used different promotional benefit levels to test the propositions; however, the
absolute monetary values of the promoted products also might be influential. For example, if
the percentage promotional benefit level from Experiment 2 (20% off) applied to more
expensive products (e.g., “buy a vacation for NT$24,000, get a free suitcase worth NT$6,000”
or “buy a vacation worth NT$24,000 and a suitcase worth NT$6,000 for just NT$24,000”),
would the framing effect be the same? A relatively higher promotional benefit amount could
strengthen the framing effect because consumers process the information more extensively.
Additional research should retest our propositions using different absolute promotional
benefit values.
In Experiments 3 and 4, participants made repurchase decisions when they had used
almost all of the previously purchased products, which is a pattern that is consistent with
consumers’ real-world behavior. Few consumers purchase a product when they still have
plenty available. However, the framing effect might be reduced or even reversed if they made
repurchase decisions right after starting to use the product for the first time (e.g., because a
really attractive promotion is about to end). In this case, they might focus more on the first
item and potentially consider its assignment price as a reference, which would be interesting
to explore in future studies.
Forthc
oming
IJRM V
olume 3
2 #1 (
2015
)
40
Finally, all the products in our experiments are utilitarian rather than hedonic. In addition,
our investigation of how frames influence consumers’ perceptions of price increases and
repurchase intentions centers solely on consumer goods. The results might vary for other
product types. For durable goods, consumers who encounter a free gift (bundle) promotion
will likely assign a lower price to the supplementary (focal) item, so they should perceive less
residual value and exhibit higher upgrade or replacement intentions toward the supplementary
(focal) product because it produces a lower sunk cost effect. Tests of these predictions might
be helpful for many industries.
References
Baron, R. M., & Kenny, D. A. (1986). The moderator-mediator variable distinction in social
psychological research: Conceptual, strategic, and statistical considerations. Journal of
Personality and Social Psychology, 51, 1173-1182.
Bell, D. R., & Bucklin, R. E. (1999). The role of internal reference price in the category
purchase decision. Journal of Consumer Research, 26, 128-143.
Chandran, S., & Morwitz, V. G. (2006). The price of “free”-dom: Consumer sensitivity to
promotions with negative contextual influences. Journal of Consumer Research, 33,
384-392.
Chou, H.-Y., & Lien, N.-H. (2012). The effects of incentive types and appeal regulatory
Forthc
oming
IJRM V
olume 3
2 #1 (
2015
)
41
framing in travel advertising. Service Industries Journal, 32, 883-897.
Chow, C. C., & Sarin, R. K. (2001). Comparative ignorance and the Ellsberg paradox. Journal
of Risk and Uncertainty, 22, 129-139.
Clogg, C. C., Petkova, E., & Shihadeh, E. S. (1992). Statistical methods for analyzing
collapsibility in regression model. Journal of Educational Statistics, 17, 51-74.
Curley, S. P., Yates, J. F., & Abrams, R. A. (1986). Psychological sources of ambiguity
avoidance. Organizational Behavior and Human Decision Processes, 38, 230-256.
Darke, P. R., & Chung, C. M. Y. (2005). Effects of pricing and promotion on consumer
perceptions: It depends on how you frame it. Journal of Retailing, 81, 35-47.
Friestad, M., & Wright, P. (1994). The persuasion knowledge model: How people cope with
persuasion attempts. Journal of Consumer Research, 21, 1-31.
Grewal, D., Marmorstein, H., & Sharma, A. (1996). Communicating price information
through semantic cues: The moderating effects of situation and discount size. Journal of
Consumer Research, 23, 148-155.
Guiltinan, J. P. (1987). The price bundling of services: A normative framework. Journal of
Marketing, 51(4), 74–85.
Hardesty, D. M., & Bearden, W. O. (2003). Consumer evaluations of different promotion
types and price presentations: The moderating role of promotional benefit level. Journal
of Retailing, 79, 17-25.
Forthc
oming
IJRM V
olume 3
2 #1 (
2015
)
42
Harlam, B. A., Krishna, A., Lehmann, D. R., & Mela, C. (1995). Impact of bundle type, price
framing and familiarity on purchase intention for the bundle. Journal of Business
Research, 33, 57-66.
Homburg, C., Koschate, N., & Totzek, D. (2010). How price increases affect future purchases:
The role of mental budgeting, income, and framing. Psychology & Marketing, 27, 36–53.
Jedidi, K., Mela, C. F., & Gupta, S. (1999). Managing advertising and promotion for long-run
profitability. Marketing Science, 18, 1-22.
Jha-Dang, P. (2006). A review of psychological research on consumer promotions and a new
perspective based on mental accounting. VISION: The Journal of Business Perspective,
10(3), 35-43.
Johnson, M. D., Herrmann, A., & Bauer, H. H. (1999). The effects of price bundling on
consumer evaluations of product offerings. International Journal of Research in
Marketing, 16, 129-142.
Kalyanaram, G., & Winer, R. S. (1995). Empirical generalizations from reference price
research. Marketing Science, 14, 161-169.
Kamins, M. A., Folkes, V. S., & Fedorikhin, A. (2009). Promotional bundles and consumers’
price judgments: When the best things in life are not free. Journal of Consumer Research,
36, 660-670.
Krishnamurthi, L., Mazumdar, T., & Raj, S. P. (1992). Asymmetric response to price in
Forthc
oming
IJRM V
olume 3
2 #1 (
2015
)
43
consumer choice and purchase quantity decisions. Journal of Consumer Research, 19,
387-400.
Laran, J., & Tsiros, M. (2013). An investigation of the effectiveness of uncertainty in
marketing promotions involving free gifts. Journal of Marketing, 77(2), 112-123.
Lee, H.-C., Tsai, D. C., & Wu, M.-D. (2011). The dilemma of mixed bundles: The effects of
price, brand, and content preference on music CD bundle choices. Journal of Retailing &
Consumer Services, 18, 455-462.
Liu, H.-H. (2011). Task formats and ambiguity aversion. Journal of Behavioral Decision
Making, 24, 315-330.
Lowe, B. (2010). Consumer perceptions of extra free product promotions and discounts: The
moderating role of perceived performance risk. Journal of Product & Brand Management,
19, 496-503.
Mayhew, G. E., & Winer, R. S. (1992). An empirical analysis of internal and external
reference prices using scanner data. Journal of Consumer Research, 19, 62-70.
Mazumdar, T., Raj, S. P., & Sinha, I. (2005). Reference price research: Review and
propositions. Journal of Marketing, 69(4), 84-102.
Mela, C. F., Gupta, S., & Lehmann, D. R. (1997). The long-run impact of promotion and
advertising on consumer brand choice. Journal of Marketing Research, 34, 248-261.
Nunes, J. C., & Park, C. W. (2003). Incommensurate resources: Not just more of the same.
Forthc
oming
IJRM V
olume 3
2 #1 (
2015
)
44
Journal of Marketing Research, 40, 26-38.
Ordóñez, L. D., Connolly, T., & Coughlan, R. (2000). Multiple reference points in satisfaction
and fairness assessment. Journal of Behavioral Decision Making, 13, 329-344.
Palmeira, M. M., & Srivastava, J. (2013). Free offer ≠ cheap product: A selective
accessibility account on the valuation of free offers. Journal of Consumer Research, 40,
644-656.
Popkowski Leszczyc, P. T. L., & Häubl, G. (2010). To bundle or not to bundle: Determinants
of the profitability of multi-item auctions. Journal of Marketing, 74(4), 110-124.
Raghubir, P. (1998). Coupon value: A signal for price. Journal of Marketing Research, 35,
316-324.
Raghubir, P. (2004). Free gift with purchase: Promoting or discounting the brand. Journal of
Consumer Psychology, 14, 181-186.
Raghubir, P. (2005). Framing a price bundle: The case of “buy/get” offers. Journal of Product
& Brand Management, 14, 123-128.
Rajendran, K. N., & Tellis, G. J. (1994). Contextual and temporal components of reference
price. Journal of Marketing, 58(1), 22-34.
Rosenthal, R. (1991). Meta-analytic procedures for social research. Newbury Park, CA: Sage.
Sinha, I., & Smith, M. F. (2000). Consumers’ perceptions of promotional framing of price.
Psychology & Marketing, 17, 257-275.
Forthc
oming
IJRM V
olume 3
2 #1 (
2015
)
45
Smith, M. F., & Sinha, I. (2000). The impact of price and extra product promotions on store
preference. International Journal of Retail & Distribution Management, 28, 83-92.
Soman, D., & Cheema, A. (2001). The effect of windfall gains on the sunk-cost effect.
Marketing Letters, 12, 51-62.
Soman, D., & Gourville, J. T. (2001). Transaction decoupling: How price bundling affects the
decision to consume. Journal of Marketing Research, 38, 30-34.
Sullivan, K., & Kida, T. (1995). The effect of multiple reference points and prior gains and
losses on managers’ risky decision making. Organizational Behavior and Human
Decision Processes, 64, 76-83.
Thaler, R. H. (1985). Mental accounting and consumer choice. Marketing Science, 4,
199-214.
Thaler, R. H. (1999). Mental accounting matters. Journal of Behavioral Decision Making, 12,
183-206.
Tsiros, M., & Hardesty, D. M. (2010). Ending a price promotion: Retracting it in one step or
phasing it out gradually. Journal of Marketing, 74(1), 49-64.
Winer, R. S. (1986). A reference price model of brand choice for frequently purchased
products. Journal of Consumer Research, 13, 250-256.
Wiseman, D. B., & Levin, I. P. (1996). Comparing risky decision making under conditions of
real and hypothetical consequences. Organizational Behavior and Human Decision
Forthc
oming
IJRM V
olume 3
2 #1 (
2015
)
46
Process, 66, 241-250.
Yadav, M. S. (1994). How buyers evaluate product bundles: A model of anchoring and
adjustment. Journal of Consumer Research, 21, 342-353.
Yadav, M. S., & Monroe, K. B. (1993). How buyers perceive savings in a bundle price: An
examination of a bundle’s transaction value. Journal of Marketing Research, 30, 350-358.
Yim, C. K. B., Chan, K. W., & Hung, K. (2007). Multiple reference effects in service
evaluations: Roles of alternative attractiveness and self-image congruity. Journal of
Retailing, 83, 147-157.
Forthc
oming
IJRM V
olume 3
2 #1 (
2015
)
47
Table 1
Pilot study: Relative shares
Task Question Mr. Li
(bundle)
The
same
Mr. Chen
(free gift)
Test with
[ without] “the same” dataa
First task Which one would allocate a higher
price to the shampoo?
3.1 33.8 63.1 χ2(2, N = 65) = 35.11
[χ2(1, n = 43) = 35.37]
Which one would allocate a higher
price to the facial cleaner?
67.7 27.7 4.6 χ2(2, N = 65) = 39.72
[χ2(1, n = 47) = 35.77]
How much did Mr. Chen [Mr. Li]
pay for the shampoo/facial cleaner in
this sales package?
129.71/
87.23
- 186.17/
26.13
t(1,64) = -10.27/
t(1,64) = 9.84
Second
task
Who would perceive a higher price
increase for the shampoo?
70.8 24.6 4.6 χ2(2, N = 65) = 44.89
[χ2(1, n = 49) = 37.73]
Who would perceive a higher price
increase for the facial cleaner?
3.1 15.4 81.5 χ2(2, N = 65) = 69.40
[χ2(1, n = 55) = 47.29]
Who would exhibit a greater
repurchase intention for the
shampoo?
6.2 30.8 63.1 χ2(2, N = 65) = 31.78
[χ2(1, n = 45) = 30.42]
Who would exhibit a greater
repurchase intention for the facial
cleaner?
80.0 18.5 1.5 χ2(2, N = 65) = 66.49
[χ2(1, n = 53) = 49.08]
a All chi-square and t tests were statistically significant (all ps < .01).
Forthc
oming
IJRM V
olume 3
2 #1 (
2015
)
48
Table 2
Means for assigned price, perceived price increase, repurchase intention, repurchase decision, and WTP in Experiments 1–4
Frames Dependent variables Experiment 1A (1B) Experiment 2 Experiment 3 Experiment 4
Focal
(Shampoo:
NT$160)
Supplementary
(Shower gel:
NT$160)
Focal
(Shampoo:
NT$200)
Supplementary
(Hand
sanitizer:
NT$50)
Toothpaste
(NT$ 90)
Essential
Shampoo
(NT$ 199)
Bundle
Assigned price 100.64* 65.97
* 168.18
* 31.36
* 58.94
*/31.06
*
(first/second)
-
Perceived price increase 5.47* (5.21
*) 4.35
* (4.72
﹢) 5.16 4.57
﹢ 4.42
* -
Repurchase intention 4.15* (4.12
*) 4.66
* (4.63
*) 4.84
* 5.26
* 4.58
* -
Repurchase decision - - 65.5%a*
79.6%* 66.7%
* 33.3%
*
WTP - - - - - 150.40*
Free gift
Assigned price 123.89 37.96 178.03 21.20 78.43/11.57
(first/second)
-
Perceived price increase 3.88 (4.65) 5.31 (5.16) 5.00 4.91 5.21 -
Repurchase intention 5.32 (4.91) 3.55 (4.07) 5.26 4.63 3.29 -
Repurchase decision - - 79.7% 56.9% 42.9% 14.6%
WTP - - - - - 126.51 a To be read: 65.5% participants indicate that they would repurchase the target product (i.e., repurchase share).
* (﹢) represents the mean of that cell in the bundle promotion was significantly (marginally) different from that in the free gift promotion
(p<0.05) .
Forthc
oming
IJRM V
olume 3
2 #1 (
2015
)
49
Table 3
The Mediating effect in Experiments 1-4
Experiment 1A Experiment 1B Experiment 2 Experiment 3 Experiment 4
Independent variablea→
Mediatorb
F(1,103)=58.70 F(1,194)= 8.75 F(1,220)= 3.13 F(1,66)=10.25 F(1,86)=13.72
Independent variable→
Dependent variablec
F(1,103)= 9.95 F(1,194)=14.43 F(1,220)=13.66 F(1,66)=22.94 χ2(1,n=89)=4.74
Independent variable and
Mediator → Dependent variable
FI(1,102)= 4.83
FM(1,103)= 20.97
FI (1,193)= 6.35
FM (1,193)=89.2
FI (1,219)=10.42
FM (1,219)=42.13
FI (1,65)=11.90
FM (1,65)=17.35
χ2
I (1,n=89)=1.19 (n.s.)
χ2
M (1,n=89)=9.75
Notes: a The independent variable in Experiments 1-2 refers to the Frames × Targets interaction, and refers to the Frames in Experiments 3 and 4.
Moreover, the results in Experiment 4 included the effect of the covariate. b
The mediator in Experiments 1-3 refers to the perceived price increase, and it refers to WTP in Experiment 4. c The dependent variable in Experiments 1-3 refers to repurchase intention, and it refers to participants’ repurchase decision in Experiment 4.
Forthc
oming
IJRM V
olume 3
2 #1 (
2015
)
50
Table 4
Procedure in Experiment 4
Step 1 Sign your name on the first page of the questionnaire. Then respond to the
relevant demographic items.
Step 2 Turn to the second page and assess your feelings about the promotion.
Step 3 Write down the highest price you are willing to pay for a single bottle of
Essential shampoo after the promotion has expired on the third page.
Step 4 One participant will be randomly invited to draw one card from the bucket.
The number on the card represents the sales price (e.g., X) of Essential
shampoo after the promotion ends. Three participants will be randomly
selected to win to the prize.
Step 5 a. If the number on the card is higher than the price the selected
participants are willing to pay (e.g., NT$Y), they will receive the
monetary equivalent of the sale price of Essential shampoo (i.e.,
NT$X).
b. If the number on the card is lower than or equal to the price that the
selected participants are willing to pay, they will receive one bottle of
Essential shampoo.
Forthc
oming
IJRM V
olume 3
2 #1 (
2015
)
1
Fig. 1. Conceptual framework
H1 H2 H2
H4 H4 H4 H5
Sales frames:Bundle vs. free gift
Price bundle vs. free gift
Price assignments
Perceived price increases
(H3)
Repurchase intentions/
decisions/ WTP
Time: acquiring
initial sales packages
Time: making repurchase decisionsTime: delay period
Figure_FINAL
Forthc
oming
IJRM V
olume 3
2 #1 (
2015
)
2
Fig. 2. Perceived price increase and repurchase intention in Experiment 1A
Fig. 3. Perceived price increase and repurchase intention in Experiment 1B
Focal (shampoo)
Supplementary (shower gel)
5.47 5.31 4.66 5.32
4.35 4.15
3.88 3.55
Bundle Free gift Bundle Free gift
Fig. 2A. Perceived price increse. Fig. 2B. Repurchase intention.
Focal (shampoo)
Supplementary (shower gel)
5.21 5.16 4.91
4.63
4.72 4.65
4.12
4.07
Bundle Free gift Bundle Free gift
Fig. 3A. Perceived price increse. Fig. 3B. Repurchase intention.
Forthc
oming
IJRM V
olume 3
2 #1 (
2015
)
3
Fig. 4. Perceived price increase and repurchase intention in Experiment 2
Focal (shampoo)
Supplementary (hand sanitizer)
5.16 5.00 5.26 5.26
4.57 4.91 4.84 4.63
Bundle Free gift Bundle Free gift
Fig. 4A. Perceived price increse. Fig. 4B. Repurchase intention.
Forthc
oming
IJRM V
olume 3
2 #1 (
2015
)