Journal of the Operational Research Society

ISSN: 0160-5682 (Print) 1476-9360 (Online) Journal homepage: https://www.tandfonline.com/loi/tjor20

Online-offline competitive pricing with reference

price effect

Ningning Wang, Ting Zhang, Xiaowei Zhu & Peimiao Li

To cite this article: Ningning Wang, Ting Zhang, Xiaowei Zhu & Peimiao Li (2020): Online-offline
competitive pricing with reference price effect, Journal of the Operational Research Society, DOI:
10.1080/01605682.2019.1696154

To link to this article: https://doi.org/10.1080/01605682.2019.1696154

View supplementary material

Published online: 05 Feb 2020.

Submit your article to this journal

Article views: 23

View related articles

View Crossmark data

Full Terms & Conditions of access and use can be found at

https://www.tandfonline.com/action/journalInformation?journalCode=tjor20
JOURNAL OF THE OPERATIONAL RESEARCH SOCIETY
https://doi.org/10.1080/01605682.2019.1696154

ORIGINAL ARTICLE

Online-offline competitive pricing with reference price effect

Ningning Wanga, Ting Zhangb, Xiaowei Zhuc and Peimiao Lib
a
School of Business, Anhui University, Hefei, China; bSchool of Management, Shanghai University, Shanghai, China; cWest Chester
University of Pennsylvania College of Business and Public Affairs, West Chester, PA, USA

ABSTRACT ARTICLE HISTORY

Nowadays, consumers can easily compare the prices of the online and offline channels Received 31 July 2018
before making purchase decisions, which might arouse the reference price effect among Accepted 17 November 2019
consumers. Considering the reference price effect, how should online and offline retailers set
KEYWORDS
prices? To answer it, we incorporate the reference price effect into a Hotelling model to for-
e-commerce; pricing; game
mulate the online-offline price competition. We find that, when the reference price effect is theory; behavioural OR;
low (high), the offline (online) retailer monopolizes the market; when the reference price reference price effect
effect is moderate, the retailers co-exist. Interestingly, a first mover disadvantage and a
second mover advantage exist in terms of market share, but they do not necessarily exist in
terms of profit. However, both the online and offline retailers can be better off by negotiat-
ing on the game sequence. Furthermore, we find that each retailer prefers its own price to
be regarded as the reference price. A lower reference price, although benefiting the e-retail-
ing in the short run, might compromise the product image and hurt both retailers in the
long run.

1. Introduction the consumer perceives a loss (Zhang, Chiang, &

Liang, 2014).
With the availability of both the online and offline
Common wisdom suggests that the price com-
channels, consumers can easily compare the prices
parison behaviour and reference price effect will
across different channels before making purchase
aggravate the market competition and harm
decisions. A recent survey by Market Track
retailers. Amazon has granted a patent to prevent
reported that 80% of the respondents would do
in-store shoppers from online price checking2. If a
price comparisons before purchasing1. Moreover,
customer is using the store Wifi to check the
some websites (e.g. PriceGrabber.com and Google
competitor’s price online, the store can suggest a
Shopping) provide the price information of differ- complementary item or even block content out-
ent channels, which further facilitates consumer right. Nevertheless, it is impossible to prevent con-
price comparison. sumers from price comparison with current
The price comparison behaviour might arouse a available technologies. Undoubtedly, the reference
cognitive bias – the reference price effect – among price effect will affect the online-offline
consumers. Consumers use historical prices, adver- competition.
tised prices, suggested retail prices or competitor Considering the reference price effect, our paper
prices to form “reference prices” (Bell & Bucklin, studies the pricing strategies of the online and off-
1999; Mazumdar, Raj, & Sinha, 2005; Putler, 1992; line retailers in a competitive setting. We consider
Rajendran & Tellis, 1994). Reference prices arise as two scenarios: our basic model considers the off-
price expectations against which consumers evaluate line price as the reference price, while our
products in their purchase scenarios (Wang, Zhang, extended model considers the online price as the
Fan, & Zhu, 2018). Consumers might evaluate the reference price.3 The basic model can be seen for
price of a product by comparing with the price some product categories (e.g. consumer packaged
which is presumably charged by other retailers in goods). Offline retailers always advertise their pri-
the same trade area (Ahmetoglu, Furnham, & ces on TV and send out advertisement to consum-
Fagan, 2014). If the actual price is lower than the ers’ house. The extended model is applied to some
reference price, the consumer perceives a gain; and products categories (e.g. consumer electronics,
if the actual price is higher than the reference price, home furnishing, and office goods) where

CONTACT Ting Zhang [email protected] School of Management, Shanghai University, Shanghai, China
Supplemental data for this article can be accessed here.
ß Operational Research Society 2020
2 N. WANG ET AL.

consumers can easily check online price and hence traditional retailers and manufacturers’ direct chan-
use online prices as the reference point.4 In each nel (Cattani, Gilland, Heese, & Swaminathan,
model, we derive the equilibrium under three dif- 2006; Chiang, Chhajed, & Hess, 2003; Fruchter &
ferent game structures. (i) The offline retailer led Tapiero, 2005; Liu & Zhang, 2006; Mukhopadhyay,
Stackelberg (SA) game, in which the offline retailer Zhu, & Yue, 2008). Chiang et al. (2003) construct
sets its retail price firstly and the online retailer a pricing game between an independent retailer
sets its retail price secondly. (ii) The online and a manufacturer with a direct channel, and
retailer led Stackelberg (SB) game, in which the find that direct marketing can benefit both the
online retailer sets its retail price firstly and the manufacturer and the retailer by mitigating the
offline retailer sets its retail price secondly. (iii) double marginalization effect. Fruchter and
The simultaneous (N) game, in which the offline Tapiero (2005) find that under risk-neutral pricing,
retailer and the online retailer set their own retail the manufacturer sets the same price for the
prices simultaneously and independently. For online and offline channels; and, both the manu-
example, Walmart entered the China market in facturer and consumers are better off with the
1996 and soon became a retail giant. In 2013, introduction of an online store. Liu and Zhang
Alibaba launched an online retailer, Tmall super- (2006) find that although the retailer is worse off
market, which competes with Walmart. In the owing to his own or upstream personalized pricing
early development of e-commerce, Walmart had a mechanism, it may still be optimal for the retailer
solid foundation among consumers and acted as a to implement personalized pricing to deter the
leader, while Tmall supermarket passively adjusts manufacturer from direct entry or from personal-
the prices as a follower. The SA game can be ized pricing. Cattani et al. (2006) find the specific
used to model this scenario. Later on, with the equal-pricing strategy that optimizes profits for the
rapid growth of e-commerce, Tmall supermarket manufacturer is preferred by the retailer and
has gained significant market power. This scenario consumers. Mukhopadhyay et al. (2008) study how
can be modelled by the simultaneous game. In to eliminate or reduce the channel competition
some industries, e.g. books and flight tickets, the between the retailer’s physical store and the
physical retailers are overwhelmed by the competi- manufacturer’s direct selling channel by allowing
tion from the online retailing. We use the SB the retailer to offer some value-added services to
model to capture this scenario where the online the end customers at the store. Zhang, Ge, Gou,
store proactively sets the price first and the offline and Chen (2018) study the online-offline competi-
retailer passively reacts to the online tion under showrooming behavior and sunk cost
retailer’s strategy. effect. The authors find that the showrooming
Our core research questions (RQs) are stated as behavior will aggravate the pricing competition
follows. RQ1: How should the online and offline while the sunk cost effect may mitigate the pricing
retailers set the prices under the reference price effect? competition.
RQ 2: In the presence of reference price effect, what Our paper is particularly related to the papers
are impacts of the different pricing sequences? RQ3: which investigate the effect of power structures on
How do the different reference prices affect the the online-offline competitive pricing by modeling
online-offline competitive pricing and different game sequences. The firms’ market power
retailers’ profits? will decide the firm’s move sequences in the price
The rest of this paper is organized as follows. game, and thus have different implications for price
Section 2 reviews related literature. In Section 3, we decision and profits (Kadiyali, Chintagunta, &
introduce the model setting. Section 4 and Section 5 Vilcassim, 2000; Pan, Lai, Leung, & Xiao, 2010).
analyze the online and offline pricing games with Yao and Liu (2005) consider that the manufacturer
two different reference prices. Concluding remarks adds an online channel to compete with the offline
are summarized in Section 6. All proofs are retailer, and investigate the impacts in both the
provided in the Online Supplement. Bertrand and the Stackelberg models. Chen, Wang,
and Jiang (2016) investigate the pricing competi-
tion of a supplier and a retailer in various games,
2. Literature review
including Supplier-Stackelberg, Retailer-Stackelberg,
Our paper is related to two streams of research: the and Nash game.
online-offline competition and the reference Second, our paper is related to the literature of
price effect. reference price effect. Previous literature has recog-
First, our paper follows the analytical modeling nized that the reference price effect will influence
literature on the online-offline competition. Most the optimal pricing strategy. Fibich, Gavious, and
of these papers study the competition among the Lowengart (2003) study the effect of asymmetric
 JOURNAL OF THE OPERATIONAL RESEARCH SOCIETY 3

reference price on the optimal pricing strategies. impacts of the reference price effect on retailers’
Fibich, Gavious, and Lowengart (2005) consider pricing strategies and performances under various
asymmetric reference price effects, and identify the game structures.
time dependence of the short and long-term effects
on consumer behavior. Anderson, Rasmussen, and
3. Model development
MacDonald (2005) study the competitive pricing
between two firms with different market power, Consider an offline retailer (denoted by ‘A’) with an
considering that the demand of a firm is a function online retailer (denoted by ‘B’) compete to sell the
of his price, his competitor’s price and a reference same product to end consumers. Retailer i sells the
price. Popescu and Wu (2007) study the dynamic product at price pi with unit operational cost ci,
pricing of a monopolist firm with demand affected where i ¼ A or B. Consumers perceive valuation v
by both the current and past prices, and find that on the product when purchasing it in store, but
managers will systematically price too low and lose they perceive valuation hv when purchasing it
revenue if they ignore the long-term implications. online (Chiang et al., 2003). Parameter h represents
Geng, Wu, and Li (2010) study the pricing strategy the acceptance of online channel and 0 < h < 1: A
and the promotion frequency with the reference higher (lower) h implies a higher (lower) acceptance
price effect, and find that the periodic promotion for the online channel. There are several reasons
can increase the profits of both the manufacturer why an online consumer receives a lower valuation
and the supply chain. Zhang, Gou, Zhang, and (h < 1). First, the online consumer cannot touch,
Liang (2014) explore the impact of reference price feel, and try on the product when shopping online
effects on the skimming price strategy in a two- (Dzyabura, Jagabathula, & Muller, 2019). Online
period selling seasons. Wu and Wu (2016) study the shoppers might find they do not like the product
optimal pricing strategy under demand uncertainty after purchases and return it. Even if many online
caused by the competition and reference price channels offer full refund, the hassle cost of product
effects. They propose that firms can utilize the return is inevitable to consumers. This reduces the
information from the big data era to reduce such consumer expectation on the product valuation.
uncertainty when making pricing decisions. Hu, Second, the sales associates at the stores can offer
Chen, and Hu (2016) study a dynamic pricing prob- professional illustrations and teach consumers how
lem of a firm facing reference price effects at an to use the products. Finally, offline retailers have
aggregate demand level, where demand is more sen- more experiences on after-sale services and cus-
sitive to gains than losses. The authors show that tomer relationship management. Notably, the value
the optimal pricing strategies may not admit any of h varies across the product categories (Levin,
simple characterizations and the resulting reference Levin, & Heath, 2003). For some product categories
price/price dynamics can be very complicated. (e.g. shoes, clothing, and cosmetic), consumers face
Wang et al. (2018) recognize that multiple ways for a considerably high risk of post-purchase product
consumers to form a reference price based on the misfit if purchasing online. For other product cate-
information about the offer set and product prices gories (e.g. computer software, books, and flight
at the point of purchase-for example, the lowest tickets), consumers can easily evaluate the product
price, the highest price, the assortment variety, the values from digital data. The former categories have
weighted sum of prices and the weighted average lower h than the latter.
price. The results show that, reference prices should Different consumers have different channel
be taken into account in model estimation and preferences. Following Balakrishnan, Sundaresan,
operations management, and ignoring reference pri- and Zhang (2014), our model incorporates the het-
ces may lead to substantial losses. Chen, Zha, erogeneity among consumers in terms of their rela-
Alwan, and Zhang (2019) investigate two-stage tive costs for purchasing in-store versus purchasing
dynamic pricing strategy in the presence of strategic online. The consumers are uniformly distributed
consumers who consider the first-stage price as the along a Hotelling line over a unit interval ½0, 1,
reference price. The authors find that the seller’s with the mass of the consumers normalized to 1.
profit will increase (decrease) with the reference The offline retailer and the online retailer are
price effect if consumer strategic behaviour is located at x ¼ 0 and x ¼ 1, respectively. A con-
low (high). sumer who is located at x has to pay a shopping
However, no prior research has investigated the cost tx to purchase in-store and a shopping cost
online-offline competition under the reference price tð1  xÞ to purchase online, where t represents the
effect. To fill the gap, this paper incorporate the ref- unit shopping. The shopping cost includes all of the
erence price effect into the online-offline competi- time, money, and effort the consumer has to occur
tive pricing model, and attempts to reveal the to search and buy the product from a certain
4 N. WANG ET AL.

Table 1. Summary of notations. demands of Retailers A and B, denoted by QA and QB ,

Notation Definition respectively.
pi The price of Retailer i The profit of Retailer i is:
h The acceptance of online channel
v The intrinsic product value pi ¼ ðpi  ci ÞQi : (4)
t The unit shopping cost
ci The unit operational cost of Retailer i We assume that the offline operation cost exceeds
URi The utility caused by the reference price
effect when purchasing from Retailer i the online operation cost i.e. cA >cB : Because the
Ui The utility of purchasing from Retailer i online retailer does not need to invest in physical
a The magnitude of the reference price
effect when perceiving a again space or fixtures and spends less in employing sales
b The magnitude of the reference price staff, the online retailer can enjoy a lower oper-
effect when perceiving a loss
Qji The demand of Retailer i in game j
ation cost.
pji The profit of Retailer i in game j We consider two scenarios: when the offline price
Note: i ¼ A or B, and j ¼ SA, SB or N. is the reference price, i.e. r ¼ pA ; and when the
online price is the reference price, i.e. r ¼ pB : In
each scenario, we study three pricing sequences, the
channel. Specifically, the cost of purchasing in-store offline retailer led Stackelberg (SA) game, the online
includes the opportunity cost of time, the real cost retailer led Stackelberg (SB) game, and the simultan-
of travel, and the implicit cost of inconvenience eous (N) game. We derive the pure strategy sub-
(Zhang et al., 2018); and the cost of purchasing game perfect equilibrium of the various games. All
online includes the search cost, the shipping fee, notations in the paper are summarized in Table 1.
and the physiological cost of Internet priv- We first prove Lemma 1 to simplify the subse-
acy concerns.5 quent analysis.
Consumer utility functions of buying from the
Lemma 1. In both scenarios (when r ¼ pA and when
offline and online retailers are r ¼ pB ), the online retailer has the incentive to set a
UA ¼ v  pA  tx þ URA , (1) price lower than the offline retailer.
UB ¼ hv  pB  tð1  xÞ þ URB , (2) Lemma 1 proves, if the online price is no lower
where URi (i ¼ A or B) is the utility caused by the than the offline price (i.e. pB  pA ), the online
reference price effect and given by: retailer will have no sales and zero profit. The
online retailer has an incentive to leave the regime
URi ¼ aðr  pi Þþ  bðpi  rÞþ : (3)
of pB  pA and move to the regime pB < pA : Hence,
Parameters a and b measure the magnitude of the pB  pA will not occur in the equilibrium. This
reference price. According to the prospect theory, finding is consistent with data. Empirical studies
if the actual price is lower than the reference price from Brynjolfsson and Smith (2000) and Lee and
(pi < r), the consumer perceives a utility gain; if Tan (2003) find that prices on the Internet are lower
the actual price a price higher than the reference than those in conventional outlets. So our research
price (pi >r), the consumer perceives a utility loss. work focus on the scenario of pB < pA :
In particular, aðr  pi Þþ and bðpi  rÞþ represent
the perceived utility gain and utility loss, respect-
ively, due to the effect of a reference price. Because
4. Equilibrium analysis: The offline price as
the reference price
of loss aversion, we assume a < b: That is, the
consumers are less sensitive to the gain than the loss. This section considers the scenario when the con-
Our formulation of reference price effect is consist- sumers regard the offline price as the reference
ent with K} oszegi and Rabin (2006), Shulman, price, i.e. r ¼ pA : When consumers buy offline, the
Cunha, and Saint Clair (2015) and Wang (2018). actual price equals the reference price, and we have
The consumers make purchase decisions to maxi- URA ¼ 0: Recall that Lemma 1 proves that the
mize their own utility. We assume that v is sufficiently online retail price will be lower than the offline
large (specifically, v>½ðcA  cB Þ þ 5t=ð1  hÞ) so retail price. When consumers buy online, they per-
that all consumers will buy one product. It is common ceive a price advantage, and we have URB ¼
to assume the full market coverage in Hotelling model aðpA  pB Þ: The consumer utility of buying from the
in literature, e.g. Liu and Zhang (2006) and Liu and offline and online retailers are:
Tyagi (2011). Hence, the consumer purchases from UA ¼ v  pA  tx, (5)
Retailer A if UA >UB , purchases from Retailer B if UB ¼ hv  pB  tð1  xÞ þ aðpA  pB Þ: (6)
UA < UB , and the consumer is indifferent of buying
from the two retailers if UA ¼ UB : By analyzing the We derive the demands of the offline and online
retailers as follows:
consumer purchase decisions, we can derive the
 JOURNAL OF THE OPERATIONAL RESEARCH SOCIETY 5

QA ¼
8
> ð1  hÞv þ t
>
> 0 if pA  pB þ
>
> 1þa
>
<
ð1  hÞv  ð1 þ aÞðpA  pB Þ þ t ð1  hÞv  t ð1  hÞv þ t (7)
if pB þ < pA < pB þ
>
> 2t 1þa 1þa
>
>
>
> ð1  hÞv  t
: 1 if pA  pB þ
1þa

QB ¼
8
> ð1  hÞv þ t
>
> 1 if pB  pA 
>
> 1þa
>
<
ð1  hÞv  ð1 þ aÞðpA  pB Þ þ t ð1  hÞv þ t ð1  hÞv  t (8)
1 if pA  < pB < pA  :
>
> 2t 1þa 1þa
>
>
>
> ð1  hÞv  t
: 0 if pB  pA 
1þa

Note that the demand functions are piece- In Stage 1, with the estimated best response pB in
wise continuous. Lemma 2, the offline retailer decides on its optimal
price pA : The equilibrium is concluded in
Proposition 1.
4.1. Offline retailer led stackelberg
Proposition 1. When the offline price is regarded as
(SA) game
the reference price, the equilibrium of the SA game is
In the SA game, the offline retailer sets the price pA in as follows:
Stage 1, and the online retailer set the price pB in Stage 2.
(i) If 0 < a < a1 , the offline retailer monopolizes
Lemma 2. Given the offline retailer’s price pA , the
the market by setting pA ¼ cB þ ð1 1þa
hÞv  t
;
best response of the online retailer is as follows:
(ii) If a1  a  a3 , the online and offline retailers
co-exist, and the equilibrium prices, demands
(i) If cA < pA  A, the online retailer has
and profits are given in Table 2;
no sales;
 the online retailer engages in the (iii) If a>a3 , the offline retailer has no sales, and
(ii) If A < pA < A,
the online retailer monopolizes the market;
competitive pricing and sets
pB ¼ ð1þaÞðcB þp A Þ  ð1  hÞvþt
2ð1þaÞ ; where a1 ¼ ð1 cAhÞv  5t
 1 and a3 ¼ ð1 cA hÞvþ3t  1:
 cB  cB
(iii) If pA  A, the online retailer monopolizes the
market by setting pB ¼ pA  ð1 1þa hÞvþt
; As the strength of the reference price effect (a)
becomes stronger, the online retailer’s pricing power
where A ¼ cB þ ð1 1þa
hÞv  t  ¼ cB þ ð1  hÞvþ3t :
and A increases and the offline retailer’s decreases. When
1þa
the strength of the reference price effect is weak
The implication of Lemma 2 is as follows. The offline
(0 < a < a1 ), the offline retailer monopolizes the
retailer, as the first mover, can deliberately set a low market. Conversely, when the strength of the refer-
price (cA < pA  A) to monopolize the market. Note ence price effect is strong (a>a3 ), the offline retailer
that cA < A is equivalent to cA  cB < ½ð1  hÞv  t= has no sales even if it decreases the price to the
ð1 þ aÞ: This implies that the offline retailer has the marginal cost, and the online retailer becomes a
power to monopolize the market only if it has significant monopoly. When the strength of the reference price
cost advantage or when the reference price effect is low. effect is moderate (a1  a  a3 ), the retailers are
If the offline retailer sets a intimidate price competing in the market. In this case, we find that,
(A < pA  A), both retailers coexist in the market. If the as the strength of the reference price effect increases,

offline retailer sets a price higher than the threshold A, the price and demand of offline retailer are non-
it has no sales or profit. increasing (@pA =@a < 0 and @QA =@a  0), and the
6 N. WANG ET AL.

Table 2. The SA equilibrium when r¼pA : Table 4. The Nash equilibrium when r¼pA :
Offline retailer (Leader) Online retailer (Follower) Offline retailer Online retailer
Price cA þcB ð1hÞvþ3t cA þ3cB 5tð1hÞv Price 2cA þcB ð1hÞvþ3t cA þ2cB 3tð1hÞv
þ þ þ þ
2 2ð1þaÞ 4 4ð1þaÞ 3 3ð1þaÞ 3 3ð1þaÞ
Demand ð1hÞvð1þaÞðcA cB Þþ3t ð1þaÞðcA cB Þð1hÞvþ5t Demand ð1hÞvð1þaÞðcA cB Þþ3t ð1þaÞðcA cB Þð1hÞvþ3t
8t 8t 6t 6t
Profit ½ð1þaÞðcA cB Þð1hÞv3t2 ½ð1þaÞðcA cB Þð1hÞvþ5t2 Profit ½ð1þaÞðcA cB Þð1hÞv3t2 ½ð1þaÞðcA cB Þð1hÞvþ3t2
16tð1þaÞ 32tð1þaÞ 18tð1þaÞ 18tð1þaÞ

(ii) If a2  a  a4 , the online and offline retailers

Table 3. The SB equilibrium when r ¼ pA : co-exist, and the equilibrium prices, demands
Offline retailer (Follower) Online retailer (Leader) and profits are given in Table 3;
Price 3cA þcB 5tþð1hÞv
þ
cA þcB 3tð1hÞv
þ (iii) If a>a4 , the online retailer monopolizes the
4 4ð1þaÞ 2 2ð1þaÞ
Demand ð1hÞvð1þaÞðcA cB Þþ5t ð1þaÞðcA cB Þð1hÞvþ3t
market by setting pB ¼ cA  ð1 1þa
hÞvþt
;
8t 8t
Profit ½ð1þaÞðcA cB Þð1hÞv5t2 ½ð1þaÞðcA cB Þð1hÞvþ3t2 where a2 ¼ ð1 cAhÞv  3t
 cB  1 and a4 ¼ ð1 cA hÞvþ5t
 cB  1:
32tð1þaÞ 16tð1þaÞ
The impacts of the reference price effect on the mar-
ket structure in the SB and SA games are similar.
price and demand of online retailer are non-
Moreover, we find that as the strength of the reference
decreasing (@pB =@a>0 and @QB =@a  0). This
price effect increases, both the price and demand of the
indicates that a stronger reference price effect allows
offline retailer are non-increasing (@pA =@a < 0 and
the online retailer to charge a higher price without
@QA =@a  0), and those of the online retailer are non-
decreasing the demand.
decreasing (@pB =@a>0 and @QB =@a  0). Notice that
a strong reference price effect allows the online retailer to
increase the price without decreasing the demand. This
4.2. Online retailer led stackelberg game (SB)
further highlights our result that the reference price effect
In the SB game, the online retailer sets the retail can increase the pricing power of the online retailer, and
price pB in Stage 1, the offline retailer sets the retail decreases the pricing power of the offline retailer.
price pA in Stage 2.
Lemma 3. Given the online retailer’s price, the best 4.3. Simultaneous game (N)
response price of the follower offline retailer is In the N game, the online and offline retailers inde-
as follows: pendently and simultaneously set their own prices.
Given the offline retailer’s price, the online retailer’s
(i) If cB < pB < B, the offline retailer has no sales; best response is given in Lemma 2. Given the online
(ii)  the offline retailer engages in com-
If B  pB  B, retailer’s price, the offline retailer’s best response is
petitive pricing and sets pA ¼ pB þc ð1  hÞvþt
2 þ 2ð1þaÞ ;
A
given in Lemma 3. By letting the retailers react as
(iii) If pB >B, the offline retailer monopolizes the the best responses simultaneously, we derive the
market by setting pA ¼ pB þ ð1 1þa
hÞv  t
; Nash equilibrium in Proposition 3.
Proposition 3. When the offline price is regarded as
where B ¼ cA  ð1  hÞvþt  ¼ cA þ 3t  ð1  hÞv :
and B the reference price, the Nash equilibrium is as follows:
1þa 1þa

In both the SA and SB games, the leader can (i) If 0 < a < a2 , the offline retailer monopolizes
drive the follower out of market. In the SB game, the market by setting pA ¼ cB þ ð1 1þa
hÞv  t
;
the online retailer can monopolize the market by (ii) If a2  a < a3 , the online and offline retailers
setting a price below B: However, this strategy is co-exist, and the equilibrium prices, demands
infeasible only if cB  B, i.e. cA  cB  ½ð1  hÞv þ and profits are given in Table 4;
t=ð1 þ aÞ: That is, when the cost advantage of the (iii) If a  a3 , the online retailer monopolizes the
online retailer is trivial and the strength of the refer- market by setting pB ¼ cA  ð1 1þa
hÞvþt
:
ence price effect is weak, it is impossible for the
online retailer to monopolize the market. The impact of the reference price effect on the mar-
ket structure, prices and demands in the simultaneous
Proposition 2. When the offline price is regarded as game is similar to those in Stackelberg games.
the reference price, the equilibrium of the SB game is
as follows:
4.4. Analysis
(i) If 0  a < a2 , the online retailer has no sales, To gain more insights into how different pricing
and the offline retailer monopolizes the market; sequences affect the outcome, we proceed to
 JOURNAL OF THE OPERATIONAL RESEARCH SOCIETY 7

competitor’s price and adjust price accordingly; and

therefore, the second mover is better at leveraging
the reference price effect than the first mover.
Third, the retailers are less likely to co-exist
under simultaneous pricing (in the Nash game) than
under sequential pricing (in the Stackelberg games).
Figure 1. Market structures under the three game struc-
In the SA, SB and N games, the regions where both
tures. Note: a1 ¼ f½ð1  hÞv  5t=ðcA  cB Þg  1, a2 ¼ retailers co-exist are a 2 ½a1 , a3 , a 2 ½a2 , a4  and a 2
f½ð1  hÞv  3t= ðcA  cB Þg  1, a3 ¼ f½ð1  hÞv þ 3t= ½a2 , a3 , respectively. We find that a3  a1 ¼
ðcA  cB Þg 1, a4 ¼ f½ð1  hÞv þ 5t=ðcA  cB Þg  1: a4  a2 >a3  a2 , i.e. there is a smaller region for
both retailers to co-exist in the N game. This
implies that, the simultaneous pricing accelerates
compare the market structure, prices and profits monopolization.
under the three game structures.
4.4.2. Analysis of prices and profits
4.4.1. Analysis of equilibrium scenarios Now we compare the equilibrium prices and profits
We show how the equilibrium scenarios vary with when both retailers co-exist, i.e. when a2 < a < a3 :
the strength of the reference price effect under dif-
Proposition 4. By comparing the equilibrium prices in
ferent game structures in Figure 1. We have the fol-
the competing scenario, we have the following result:
lowing results.
First, the online retailer monopolizes the market ð2Þ
(i) If a2 < a  a2 , A  pA >pA ;
pSA SB N
and if
if the reference price effect is strong, the offline ð2Þ
retailer monopolizes the market if the reference a2 < a < a3 , pSB
A >pA >pA ;
SA N

ð1Þ
effect is weak, and both retailers co-exist if the refer- (ii) If a2 < a  a2 , pSA B  B >pB ;
pSB N
and if
ence price effect is moderate; and this trend holds ð1Þ
a2 < a < a3 , pB >pB >pNB :
SB SA
under all the three game structures. The online
retailer, with a lower price than the competitor’s ð1Þ
price, can constantly benefit from the reference The thresholds are given by a2 ¼ ð1 cA hÞv t
 cB  1 and
ð2Þ ð1Þ ð2Þ
price effect regardless of the game sequences. a2 ¼ ð1cA hÞvþt
cB  1, and we have a2 < a2 < a2 < a3 :
Second, in the online-offline pricing competition
with reference price effect, first mover disadvantage Proposition 4 shows the following two results.
and second mover advantage exist in terms of mar- First, the competition is fiercer under simultaneous
ket share. This result is concluded from two obser- pricing than under sequential pricing. As shown in
vations. The first observation is that retailer is less Figure 2, regardless of the strength of the reference
likely to monopolize the market when it is the first price effect, both retailers set lower prices in the
mover than when it is not. When the offline retailer simultaneous game than in the Stackelberg games.
is the first mover (in the SA game), it monopolizes Second, if the reference price effect is weak, the
the market if a < a1 ; When it is not the first mover pricing competition is fiercer in the SB game than
(in the SB and N games), it monopolizes the market in the SA game; if the reference price effect is
if a < a2 : Since a1 < a2 , the former condition is strong, the pricing competition is fiercer in the SA
more restrictive than the latter. Similarly, for the game than in the SB game. The reference price
online retailer, its monopolization region shrinks effect increases the pricing power of the online
when it becomes the first mover. The second obser- retailer and decreases the pricing power of the off-
vation is that a retailer is more likely to have posi- line retailer. We denote that when the reference
tive sales when it is the second mover than it is not. price effect is weak, the offline retailer is advanced;
When the offline retailer is the second mover (in otherwise, when the reference price effect is strong,
the SB game), it earns positive profit if a < a4 ; when the online retailer is advanced. Thereby, when the
it is not the second mover (in the SA and N games), reference price effect is weak, the online retailer is
it earns positive profit if a < a3 : Since a3 < a4 , the in an inferior position. It is better for online retailer
latter condition is more restrictive than the former. to adopt the responsive pricing, with the offline
Similarly, the online retailer is more likely to earn price announced. If not, the disadvantaged online
positive profits if it is the second mover. The retailer has the incentive to lower the price to a
explanation for the second mover advantage is as large extent, which significantly aggravates the pric-
follows. The reference price effect increases con- ing competition. When the reference price effect is
sumer sensitivities to the price and the price differ- strong, the offline retailer is disadvantaged. If the
ence. The second mover can be responsive to its online retailer announces the price first and the
8 N. WANG ET AL.

Figure 2. Effects of a on the equilibrium prices (Depicted with h ¼ 0:8, t ¼ 0:7, cA ¼ 12, cB ¼ 7).

offline retailer responds accordingly, both retailers Second, although the first mover disadvantage
are allowed to set higher prices. In sum, if the and second mover advantage exist in terms of mar-
advantaged retailer proactively announces the price ket share, they do not necessarily exist in terms of
and the disadvantaged retailer adopt the responsive profit. If a retailer competes for a larger market
pricing, both retailers are allowed to charge share, it is better off under responsive pricing in
higher prices. which it is the second mover, as shown in Figure 3.
However, when a retailer is advantaged (e.g. an off-
Proposition 5. By comparing the equilibrium profits
line retailer is advantaged when the reference price
in the competing scenario, we have the follow-
effect is weak, or an online retailer is advantaged
ing result:
when the reference price effect is strong), the advan-
ð3Þ ð3Þ taged retailer can achieve a higher profit under pro-
i. If a2 < a  a2 , pSA
A  pA >pA ; and if a2 <
SB N
actively pricing in which it is the first mover
a < a3 , pA >pA >pA ;
SB SA N
(Figure 4).
ð4Þ ð4Þ
ii. If a2 < a  a2 , pSA
B  pB >pB ; and if a2 <
SB N

a < a3 , pSB
B >pB >pB :
SA N
5. Extension: The online price as the
reference price6
ð3Þ
The thresholds are given by a2 ¼ In this section, we consider that consumers use the
pffiffi pffiffi
ð1  hÞv  ð2 2  1Þt ð4Þ ð1  hÞvþð2 2  1Þt online price as the reference price (i.e. r ¼pB ). We
cA  cB  1 and a2 ¼ cA  cB  1,
ð3Þ ð4Þ
use ‘’ over a variable to denote this case. We
and we have a2 < a2 < a2 < a3 : derive the equilibrium under the SA, SB and N
Proposition 5 has the following implications. games in Propositions 6, 7 and 8, respectively.
First, when the reference price effect is weak, both Proposition 6. When the online price is regarded as
retailers achieve the highest profits if the offline the reference price, the equilibrium of the SA game is
retailer is the game leader. When the reference price as follows:
is strong, both retailers achieve the highest profits if
the online retailer is the game leader. These suggest (i) If 0 < b < a1 , the offline retailer monopolizes
that the retailers can achieve a win-win situation by
the market by setting pA ¼ cB þ ð1 1þb
hÞv  t
;
negotiating on the game sequence. As discussed in
(ii) If a1  b  a3 , the online and offline retailers
Proposition 4, when the reference price effect is
co-exist, and the equilibrium prices, demands
weak, the online retailer is disadvantaged. The off-
and profits are given in Table 5;
line retailer can announce the price first and com-
(iii) If a3 < b < 1, the offline retailer has no sales,
mit to not change it. When the reference price
and online retailer monopolizes the market.
effect is strong, the offline retailer is disadvantaged.
The online retailer can announce the price first and Proposition 7. When the online price is regarded as
commit to not change it. This can avoid fierce pric- the reference price, the equilibrium of the SB game is
ing competition and benefit both retailers. as follows:
 JOURNAL OF THE OPERATIONAL RESEARCH SOCIETY 9

Figure 3. Effects of a on the equilibrium demands (Depicted with h ¼ 0:8, t ¼ 0:7, cA ¼ 12, cB ¼ 7).

Figure 4. Effects of a on the equilibrium profits (Depicted with h ¼ 0:8, t ¼ 0:7, cA ¼ 12, cB ¼ 7).

Table 5. The SA equilibrium when r¼pB : Table 6. The SB equilibrium when r¼pB :
Offline retailer (Leader) Online retailer (Follower) Offline retailer (Follower) Online retailer (Leader)
Price cA þcB ð1hÞvþ3t cA þ3cB 5tð1hÞv Price 3cA þcB 5tþð1hÞv cA þcB 3tð1hÞv
þ þ þ þ
2 2ð1þbÞ 4 4ð1þbÞ 4 4ð1þbÞ 2 2ð1þbÞ
Demand ð1hÞvð1þbÞðcA cB Þþ3t ð1þbÞðcA cB Þð1hÞvþ5t Demand ð1hÞvð1þbÞðcA cB Þþ5t ð1þbÞðcA cB Þð1hÞvþ3t
8t 8t 8t 8t
Profit ½ð1þbÞðcA cB Þð1hÞv3t2 ½ð1þbÞðcA cB Þð1hÞvþ5t2 Profit ½ð1þbÞðcA cB Þð1hÞv5t2 ½ð1þbÞðcA cB Þð1hÞvþ3t2
16tð1þbÞ 32tð1þbÞ 32tð1þbÞ 16tð1þbÞ

(i) If 0 < b < a2 , the offline retailer has no sales, Proposition 8. When the online price is regarded as
and online retailer monopolizes the market; the reference price, the equilibrium of the simultan-
(ii) If a2  b  a4 , the online and offline retailers eous game is as follows:
co-exist, and the equilibrium prices, demands
and profits are given in Table 6; (i) If 0 < b < a2 , the offline retailer monopolizes
(iii) If a4 < b < 1, the online retailer monopolizes the market by setting pA ¼ cB þ ð1 1þb
hÞv  t
;
the market by setting pB ¼ cA  ð1 1þb
hÞvþt
:
10 N. WANG ET AL.

Table 7. The Nash equilibrium when r¼pB : simultaneous pricing) and with different reference
Offline retailer Online retailer prices (when the offline price is used as the refer-
Price 2cA þcB ð1hÞvþ3t cA þ2cB 3tð1hÞv ence price and when the online price is used as the
þ þ
3 3ð1þbÞ 3 3ð1þbÞ
Demand ð1hÞvð1þbÞðcA cB Þþ3t ð1þbÞðcA cB Þð1hÞvþ3t
reference price). If the strength of the reference
6t 6t price effect is strong, the online retailer sets a low
Profit ½ð1þbÞðcA cB Þð1hÞv3t2 ½ð1þbÞðcA cB Þð1hÞvþ3t2 price to drive its competitor out of the market; con-
18tð1þbÞ 18tð1þbÞ
versely, the offline retailer does so; and if the
strength of the reference price effect is moderate,
(ii) If a2  b  a3 , the online and offline retailers both retailers adopt competitive pricing and co-exist.
co-exist, and the equilibrium prices, demands Moreover, a stronger strength of reference price
and profits are given in Table 7; effect allows the online retailer to charge a higher
(iii) If a3 < b < 1, the online retailer monopolizes price without decreasing the demand. This finding
the market by setting pB ¼ cA  ð1 1þb
hÞvþt
: could help retailers better understand how they
When consumers regard the online retail price as should react to competition and consumer reference
the reference price, the main results of Section 4 price effect.
hold. By comparing the two cases, we find that, RQ 2: In the presence of reference price effect,
when the consumers use the online price as the ref- what are impacts of the different pricing sequences?
erence price, the offline retailer is worse off Overall, we find the pricing competition is fiercer
p iA < piA , i ¼ SA, SB, N) and the online retailer is
(~ under simultaneous pricing than sequential pricing.
better off (~ p iB >piB , i ¼ SA, SB, N). If the offline Moreover, the first mover disadvantage and the
price is used as the reference price, consumers per- second mover advantage exist in terms of market
ceive monetary gains from online shopping; whereas share, but do not necessarily exist in terms of profit.
if the online price is used as the reference price, Surprisingly, when a retailer is advantaged, there
consumers perceive monetary loss from in-store exists a first mover advantage in terms of profit but
shopping. In the latter case, consumers are more a first mover disadvantage in terms of market share.
price sensitive because of loss aversion, which harms The retailer faces a trade-off. If the retailer is strug-
the high-priced offline retailer. As a result, using the gling with survival, aims at monopolization or com-
online price as the reference price causes a win-lose peting for market share, it should adopt the
situation for the low-priced online retailer and the responsive pricing. However, if the retailer aims at
high-priced offline retailer. Since either retailer pre- maximize the profit, it should adopt the proactively
fers to use its own price to be the reference price, pricing. Specifically, if the reference price effect is
they both have the incentive to launch price adver- weak, both retailers have the highest profits if the
tising frequently and reveal more on their own price offline retailer is the game leader. Otherwise, if the
information. As a remark, while a lower reference reference price effect is strong, both retailers have
price (e.g. r ¼ pB ) benefits the online retailing in the the highest profits if the online retailer is the game
short run, it might compromise the brand image of leader. This suggests that the retailers can achieve a
the product and hurt both retailers in the long run. win-win situation by negotiating on the
game sequence.
RQ3: How do the different reference prices affect
6. Conclusion the online-offline competitive pricing and retailers’
This paper builds analytical models to investigate profits? By comparing the two scenarios (using the
the online-offline competitive pricing with the ref- offline price as the reference price versus using
erence price effect. We consider two scenarios: the the online price as the reference price), we find that
offline price as the reference price and the online the former scenario is more desirable for the offline
price as the reference price. The equilibrium under retailer and the latter case is more desirable for the
three game structures are derived, including the online retailer. That is, each retailer prefers to use
offline retailer led Stackelberg game, the online its own price to be the reference price. To achieve
retailer led Stackelberg game and simultaneous that, both retailers wish the consumers exposed
game. We have answered the research questions more to their own price information rather than
as follows. their competitor’s. They both have the incentive to
RQ 1：How should the online and offline retailers reveal more on their own price information and
set the prices under the reference price effect? The launch price advertising frequently. The reference
reference price effect increases the online retailer’s price effect may have an unexpected result of
pricing power and decreases the offline retailer’s increasing the price transparency. However, we dis-
pricing power. This result holds under different courage the online retailer to overdo the price
game structures (the sequential pricing and advertising because it induces a low reference
 JOURNAL OF THE OPERATIONAL RESEARCH SOCIETY 11

price among consumers. Although a lower reference References

price benefits the e-retailing in the short run, it
Ahmetoglu, G., Furnham, A., & Fagan, P. (2014). Pricing
compromises the brand image of the product and practices: A critical review of their effects on consumer
might hurt both retailers in the long run. The man- perceptions and behaviour. Journal of Retailing and
agerial insights we get here are that by giving a low Consumer Services, 21(5), 696–707. doi:10.1016/j.jret-
reference price, the new e-retailing can use online conser.2014.04.013
Anderson, C. K., Rasmussen, H., & MacDonald, L.
channel as the way to introduce the products and (2005). Competitive pricing with dynamic asymmetric
build up the brand image. Other market strategies, price effects. International Transactions in Operational
like improving customer service, need to be consid- Research, 12(5), 509–525. doi:10.1111/j.1475-3995.2005.
ered for the long-term development. 00522.x
This paper has several limitations that point Balakrishnan, A., Sundaresan, S., & Zhang, B. (2014).
Browse-and-switch: Retail-online competition under
toward fruitful directions for future research. First, value uncertainty. Production and Operations
this paper considers consumers have homogenous Management, 23(7), 1129–1145. doi:10.1111/poms.
product valuation for tractability. In reality, con- 12165
sumers might have heterogeneous valuation on the Bell, D. R., & Bucklin, R. E. (1999). The role of internal
product. Second, we focus on two scenarios, where reference points in the category purchase decision.
Journal of Consumer Research, 26(2), 128–143. doi:10.
the online price is the reference price and where the 1086/209555
offline price is the reference price. The formulation Brynjolfsson, E., & Smith, M. D. (2000). Frictionless com-
of reference price is a complex problem and differ- merce? A comparison of Internet and conventional
ent scholars have different perspectives. Evidences retailers. Management Science, 46(4), 563–585. doi:10.
1287/mnsc.46.4.563.12061
show that consumers may regard the average price
Cattani, K., Gilland, W., Heese, H. S., & Swaminathan, J.
as the reference price. Further research may con- (2006). Boiling frogs: Pricing strategies for a manufac-
sider that. turer adding a direct channel that competes with the
traditional channel. Production and Operations
Management, 15(1), 40–56.
Notes Chen, K., Zha, Y., Alwan, L. C., & Zhang, L. (2019).
1. Renfrow, J. (2017). 80% of shoppers compare prices Dynamic pricing in the presence of reference price
effect and consumer strategic behaviour. International
online before shopping in stores. Retrieved from
Journal of Production Research, 1. doi:10.1080/
https://www.fierceretail.com/digital/80-shoppers-do-online-
00207543.2019.1598592
price-comparison-before-store-shopping
Chen, X., Wang, X., & Jiang, X. (2016). The impact of power
2. Deahl, D. (2017). Amazon granted a patent that
structure on the retail service supply chain with an O2O
prevents in-store shoppers from online price
mixed channel. Journal of the Operational Research
checking. Retrieved from https://www.theverge.com/
Society, 67(2), 294–301. doi:10.1057/jors.2015.6
2017/6/15/15812986/amazon-patent-online-price-checking Chiang, W. Y. K., Chhajed, D., & Hess, J. D. (2003).
3. The formulation of the reference price is a complex Direct marketing, indirect profits: A strategic analysis
problem. Ahmetoglu et al. (2014) conclude three ways of of dual-channel supply-chain design. Management
forming the reference price: (i) comparing the advertised Science, 49(1), 1–20. doi:10.1287/mnsc.49.1.1.12749
price with one price which the retailer formerly charged Dzyabura, D., Jagabathula, S., & Muller, E. (2019).
for the product; (ii) comparing the advertised price with Accounting for discrepancies between online and off-
one price which is presumably charged by other retailers line product evaluations. Marketing Science, 38(1),
in the same trade area; and (iii) comparing the advertised 88–106. doi:10.1287/mksc.2018.1124
price with one manufacturer’s suggested retail price. The Fibich, G., Gavious, A., & Lowengart, O. (2003). Explicit
two reference prices considered in our paper is formed solutions of optimization models and differential games
through the second way. with nonsmooth (asymmetric) reference-price effects.
4. We thank the anonymous reviewers for the Operations Research, 51(5), 721–734. doi:10.1287/opre.
insightful comment. 51.5.721.16758
5. Heterogeneous e-purchase costs have been addressed by Fibich, G., Gavious, A., & Lowengart, O. (2005). The
some previous research works, e.g., Frambach, Roest, & dynamics of price elasticity of demand in the presence
Krishnan, (2007) and Balakrishnan et al. (2014). of reference price effects. Journal of the Academy of
6. We thank the anonymous reviewer for the Marketing Science, 33(1), 66–78. doi:10.1177/
insightful comment. 0092070304267108
Frambach, R. T., Roest, H. C. A., & Krishnan, T. V.
(2007). The impact of consumer Internet experience on
Acknowledgment channel preference and usage Intentions across the dif-
ferent stages of the buying process. Journal of
This research is supported by National Natural Science
Interactive Marketing, 21(2), 26–41. doi:10.1002/dir.
Foundation of China (71901137, 71801001).
20079
Fruchter, G. E., & Tapiero, C. S. (2005). Dynamic online
ORCID and offline channel pricing for heterogeneous customers
in virtual acceptance. International Game Theory Review,
Xiaowei Zhu http://orcid.org/0000-0003-3823-1763 07(02), 137–150. doi:10.1142/S0219198905000454
12 N. WANG ET AL.

Geng, Q., Wu, C., & Li, K. (2010). Pricing and promotion Popescu, I., & Wu, Y. (2007). Dynamic pricing strategies
frequency in the presence of reference price effects in with reference effects. Operations Research, 55(3),
Supply Chains. California Journal of Operations 413–429. doi:10.1287/opre.1070.0393
Management, 8(1), 74–82. Putler, D. S. (1992). Incorporating reference price effects
Hu, Z., Chen, X., & Hu, P. (2016). Dynamic pricing with into a theory of consumer choice. Marketing Science,
gain-seeking reference price effects. Operations 11(3), 287–309. doi:10.1287/mksc.11.3.287
Research, 64(1), 150–157. doi:10.1287/opre.2015.1445 Rajendran, K. N., & Tellis, G. J. (1994). Contextual and
Kadiyali, V., Chintagunta, P., & Vilcassim, N. (2000). temporal components of reference price. Journal of
Manufacturer-retailer channel interactions and implica- Marketing, 58(1), 22–34. doi:10.2307/1252248
tions for channel power: An empirical investigation of Shulman, J. D., Cunha, J. M., & Saint Clair, J. K. (2015).
pricing in a local market. Marketing Science, 19(2), Consumer uncertainty and purchase decision reversals:
127–148. doi:10.1287/mksc.19.2.127.11805 Theory and evidence. Marketing Science, 34(4),
K}oszegi, B., & Rabin, M. (2006). A model of reference- 590–605. doi:10.1287/mksc.2015.0906
dependent preferences. Quarterly Journal of Economics, Wang, N., Zhang, T., Fan, X., & Zhu, X. (2018). Game
121(4), 1133–1165. doi:10.1162/qjec.121.4.1133 theoretic analysis for advertising in dual-channel supply
Lee, K. S., & Tan, S. J. (2003). E-retailing versus physical chains. International Journal of Production Research, 1.
retailing: A theoretical model and empirical test of con- . doi:10.1080/00207543.2018.1540894
sumer choice. Journal of Business Research, 56(11), Wang, R. (2018). When prospect theory meets consumer
choice models: Assortment and pricing management
877–885. doi:10.1016/S0148-2963(01)00274-0
with reference prices. Manufacturing & Service
Levin, A. M., Levin, I. P., & Heath, C. W. (2003). Product
Operations Management, 20(3), 583–600. doi:10.1287/
category dependent consumer preferences for online
msom.2017.0688
and offline shopping features and their influence on
Wu, L. L. B., & Wu, D. (2016). Dynamic pricing and risk
multi-channel retail alliance. Journal of Electronic
analytics under competition and stochastic reference
Commerce Research, 4(3), 85–93. price effects. IEEE Transactions on Industrial
Liu, Y., & Tyagi, R. K. (2011). The benefits of competitive Informatics, 12(3), 1282–1293. doi:10.1109/TII.2015.
upward channel decentralization. Management Science, 2507141
57(4), 741–751. doi:10.1287/mnsc.1110.1311 Yao, D. Q., & Liu, J. J. (2005). Competitive pricing of
Liu, Y., & Zhang, Z. J. (2006). Research note-the benefits mixed retail and E-tail distribution channels. Omega,
of personalized pricing in a channel. Marketing Science, 33(3), 235–247. doi:10.1016/j.omega.2004.04.007
25(1), 97–105. doi:10.1287/mksc.1050.0120 Zhang, J., Chiang, W. Y. K., & Liang, L. (2014). Strategic
Mazumdar, T., Raj, S. P., & Sinha, I. (2005). Reference price pricing with reference effects in a competitive supply
research: Review and propositions. Journal of Marketing, chain. Omega, 44, 126–135. doi:10.1016/j.omega.2013.
69(4), 84–102. doi:10.1509/jmkg.2005.69.4.84 07.002
Mukhopadhyay, K., Zhu, X., & Yue, X. (2008). Optimal Zhang, J., Gou, Q., Zhang, J., & Liang, L. (2014). Supply
contract design for mixed channels under information chain pricing decisions with price reduction during the
asymmetry. Production and Operations Management, selling season. International Journal of Production
17(6), 641–650. doi:10.3401/poms.1080.0069 Research, 52(1), 165–187. doi:10.1080/00207543.2013.
Pan, K., Lai, K. K., Leung, S. C., & Xiao, D. (2010). 831189
Revenue-sharing versus wholesale price mechanisms Zhang, T., Ge, L., Gou, Q., & Chen, L. (2018). Consumer
under different channel power structures. European showrooming, the sunk cost effect and online-offline
Journal of Operational Research, 203(2), 532–538. doi: competition. Journal of Electronic Commerce Research,
10.1016/j.ejor.2009.08.010 19(1), 55–74.