The current issue and full text archive of this journal is available on Emerald Insight at:

www.emeraldinsight.com/2040-7122.htm

JRIM
13,4 The value of reputation in
electronic marketplaces
A moderating role of customer experience
578 Ying Yang and Xinyu Sun
The Key Lab of the Ministry of Education for Process Control and Efficiency
Received 26 November 2018 Engineering, Xi’an Jiaotong University School of Management,
Revised 11 July 2019
Accepted 15 July 2019
Xi’an, China, and
Jiayin Wang
Information System, Xi’an Jiaotong University School of Management,
Xi’an, China

Abstract
Purpose – The purpose of this study is to examine the role of customer experience moderating the
relationship between reputation (online consumer reviews) and price premium.
Design/methodology/approach – This paper collected half-year period transaction of Nokia 5230XM and
Kingston SD card from Taobao.com, the largest e-commerce platform in China. This paper combined theoretical
analysis and empirical analysis together. Two-stage regression and logistic regression analysis was applied in
this empirical analysis. The sensitivity analyses (robustness check) were also conducted in this paper.
Findings – Customer experience negatively moderates reputation price premium; thus, the positive effect of
the reputation system is weaker for the experienced customer than for the naïve customer. Customers with
more experience are less likely to pay the price premium and rely on a reputation system.
Practical implications – The results help sellers to strategize in the online marketplace. Sellers that wish
to compete in the e-market must understand the type of customers they are addressing and differentiate the
way they treat customers based on the level of customer experience.
Originality/value – This research contributes to the reputation management and customer behavior
literature by identifying the effects of customer experience on the relationship between the reputation system
and price premium. The results address the conflicts found in previous studies by extending the explanation
of the negative reputation price premium.
Keywords Customer experience, Electronic commerce, Online consumer behaviour
Paper type Research paper

1. Introduction
When purchasing online, customers cannot inspect their desired products. Instead, they
commonly seek quality information to make their purchasing decisions. Online
marketplaces employ the reputation system to address the information asymmetry problem.
As an important proxy of product quality, reputation influences customer willingness to pay
(Melnik and Alm, 2002; Resnick et al., 2006). High-quality sellers are encouraged to stay in

Journal of Research in Interactive

This work was supported by the Natural Science Foundation of Shaanxi [2017JM7009], the Soft
Marketing Science Research Program of Shannxi [2018KRZ005], the National Science Foundation of China
Vol. 13 No. 4, 2019
pp. 578-601 [31701150], the Fundamental Research Funds for the Central Universities [CXTD2017003], and MOE
© Emerald Publishing Limited Project of Humanities and Social Sciences [19YJE630002], the National Natural Science Foundation of
2040-7122
DOI 10.1108/JRIM-11-2018-0151 China [71971165], China.
the market because their reputation can be signaled and rewarded. The reputation system Role of
creates price premiums, differentiates sellers, and helps online marketplaces to prevent a customer
market failure (Pavlou and Dimoka, 2006).
Previous studies have attempted to identify the average effect of reputation on price (Lewis,
experience
2011; Neto et al., 2016). Most of these studies have assumed that customers are homogenous
and symmetrical, and such studies rarely consider the customer side. However, the impact of
the reputation system is critically shaped by the behavioral response to them. Research
indicates that customer experiences can affect purchase behavior (Roth and Ockenfels, 2002; 579
Becerra and Korgaonkar, 2011; Lin and Lekhawipat, 2014; Mosteller et al., 2014). Thus,
customers with different levels of experience may conceive the value of reputation differently.
The focus of this paper is to understand how customers with different levels of online shopping
experiences react to the information tracked by the reputation system.
In the online marketplace, customers use online information to make their purchase
decisions, which evokes perceptions of uncertainty. Information retrieval plays a critical role
in determining customer behavior (Puccinelli et al., 2009). Customers with more experience
have more confidence in online information (O. Pappas et al., 2014). Thus, customers with
more experience seem more likely to be influenced by the primary reputation information.
Conversely, customers with more experience may find the reputation system less credible.
Customers with more experience are more familiar with online marketplace rules and regulations
(Jøsang et al., 2007), and they are more likely to be exposed to other related information. On the
Taobao platform, for example, additional guarantees (e.g. seven-day return and accurate
description) are provided to signal that sellers are trustworthy, which also helps build customer
trust in a seller’s credibility and influence purchase decisions (Lewis, 2011; Mavlanova et al., 2016).
Also, sellers could treat reputation as a marketing tool to take advantage of it (Mayzlin, 2006; Fan
et al., 2016) and manipulate their reputation by giving conditional feedback rebates (Cabral and Li,
2015). As a result, a customer with more experience may discount reputation and predict product
quality more by other signals. In this case, reputation influences an experienced customer less.
The reputation mechanism enables us to quantify the experience level of customers based
on their personal purchasing history on taobao.com. Taobao.com is the largest e-commerce
platform in China, similar to the Amazon marketplace, allowing third-party sellers to sell new
or used products at a fixed price. Reputation is measured by the overall rating of the seller
(Rating), calculated by eBay as the number of positive feedbacks minus the number of negative
ones left by unique users (Melnik and Alm, 2002). Similarly, after the transaction ends in
Taobao.com, the customer and seller both have the opportunity to give feedback on each other
either positive (1), neutral (0) or negative (
1). Seller rating and consumer rating are both
calculated as the sum of their feedback scores. As sellers have an incentive to get positive
feedback from customers, they generally leave a positive (þ1) feedback for the customer
(Cabral and Li, 2015). Thus, the rating of customers is a good indicator of their experience level
because it could capture the number of transactions a customer has done in the past.
Experiences of customers are captured by their transaction history in the marketplace. Also,
the seller rating (the sum of feedback scores) and the detailed text comments automatically
displayed on the Web page of each item sold by the seller are viewed by each potential
customer. Moreover, previous studies widely applied seller rating (Bruce et al., 2004) and
positive feedback comments (Ye et al., 2013) to capture seller’s reputation. Therefore, we value
the seller’s reputation in two ways: their seller rating and positive feedback comments.
In this paper, we propose that customer online shopping experiences affect their reliance
on reputation, and this is an important factor governing the efficacy of reputation. First,
we set up a theoretical model. After laying out our theoretical perspective, we use data from
the Taobao.com consumer-to-consumer market, empirically clarifying the effect of customer
JRIM experience levels on the relationship between seller reputation and price. We find that the
13,4 relationship between reputation and price premium is positive, but customer experience
negatively moderates the relationship between them. As customer experience grows,
customers establish their way of identifying trustable sellers and qualified products, relying
less on a reputation system. The growth of customer experience weakens the incentive effect
of the reputation system because customers with more experience are less likely to pay a
580 price premium to the high-reputation seller.
This paper is organized as follows: Section 2 reviews related literature. Section 3 presents
the establishment of a theoretical model and derives the hypotheses. We introduce data
collection and description in Section 4. Then, we present an empirical analysis in Section 5.
Section 6 presents the implications and discussions. We conclude the study in Section 7.

2. Literature review
2.1 Reputation mechanism
The online marketplace, which involves virtual marketing rather than face-to-face trade,
exchanges asymmetric information. This process can lead to adverse selection, moral
hazards, market inefficiency, or even market failure (Akerlof, 1978). As a main signal
display mechanism in the online marketplace, the reputation mechanism helps the seller
gain customer trust (Mui et al., 2002; Jøsang et al., 2007; Tadelis, 2016) and reduce the degree
of adverse selection in the online marketplace (Tadelis, 1999; Lewis, 2011; Chen et al., 2015).
Major online marketplace platforms, such as eBay, Amazon and Taobao, have all introduced
reputation systems. Removing the reputation mechanism would increase the market share of an
inferior product, decrease the market price, the market size, and customer surplus (Cabral and
Hortacsu, 2010). A low reputation means the seller is more likely to exit the market (Hui et al.,
2016). As a signal of quality (Cabral and Hortacsu, 2010), the reputation mechanism proved
helpful to the online marketplace by alleviating but not eliminating the “lemons problem,” which
is caused by information asymmetry (Chen et al., 2015). Moreover, the determinants of online
trust are different across customers (Bart et al., 2005). Resnick et al. (2000) point out several
limitations related to the reputation system. For example, feedbacks are biased because not all
customers leave feedback. In practice, the customer negotiates with the seller first before
resorting to negative feedback and sellers collaborate, thus artificially improving their reputation.
As a powerful quality signaling, reputation system influence consumer’s perceived
quality in the online market (Dawar and Parker, 1994). Caruana and Ewing (2010) indicate
that a good retailer reputation enables customers to infer high quality. A good (bad)
reputation does not only have a direct effect on quality perception but also has an indirect
effect through an increase (decrease) in the diagnosticity of the low-scope cues (Purohit and
Srivastava, 2001). Perceived value in turn significantly affects consumer intentions of
repurchase and positive recommendations (Gera, 2011). Reputation had a significant
positive effect on perceived risk, moreover, perceived risk had a significant impact on
purchase intention (Kim and Lennon, 2013).

2.2 Price premium

The price premium indicates an above-average price level, which does not mean that sellers
can get a higher price than another sales channel (Shapiro, 1983). The price premium is a
means of assuring performance collaboration; the non-performing seller loses the benefits
(Klein and Leffler, 1981). Product uncertainty and seller uncertainty both affect price
premiums than seller uncertainty (Dimoka et al., 2012). Specifically, product quality could
influence price premium in three ways:
 (1) customer judgments of product quality; Role of
(2) seller influence on product quality; and customer
(3) market influences on product quality (Rao and Monroe, 1996). experience
The price premium that sellers can command over the competition is a measure of their
reputation (Ghose et al., 2007). The relationship between reputation and price premium has
been widely studied. Several researchers have argued that the relationship between 581
reputation and price premium is negative. For example, Liu et al. (2012) pointed out that
sellers with high reputations could set a relatively low price to compete with others. Fan
et al. (2016) found a negative relationship between reputation and revenue because of
reputation management. However, most of the previous studies found that the relationship
between reputation and price is positive. Resnick et al. (2006) summarized previous
empirical research related to reputation in the eBay auction market and stated that
reputation price premiums (where the customer is willing to pay more to the seller with a
high reputation) are consistently positive in their results. Quality-sensitive buyers are
willing to pay subcontractors a price premium to prevent quality debasement, but buyers
are not willing to pay price premiums to suppliers with a quality reputation (Biong, 2013).
Kwak et al. (2015) analyzed how the brand influences the perceived unfairness of price and
the perceived fairness of the price. Seller reputation is a causal contributor to customer
behavior; it increases customer trust and willingness to pay (Cabral and Hortacsu, 2010;
Chang et al., 2013). There is no consistent conclusion as to whether the reputation price
premium is negative or positive in the previous studies. Table I summarizes the previous
literature that discussed the relationship between the reputation system and price.

2.3 Experts vs novices

Customer with different expertise, usually categorized as experts and novices (Park, 1980; Raju
et al., 1995; Ju, 2015). As a combination of product knowledge and experience (Yale and Gilly,
1995), customer expertise influences the way they collect and organize information (Lee, 2014). In
customer purchase decisions, customer experience contains two parts, customer engagement and
customer expertise; moreover, customer engagement exerts a positive moderating effect, while
customer expertise is found to have a negative moderating effect (Cheung et al., 2014).
Previous literature widely discussed the difference between expert and novices in the use
of information, but the findings in the literature are inconsistent. Babutsidze (2012)
summarized the previous research on the relationship between information sources and the
difference among customer in decision-making. Take the early literature as an example,
Bettman and Park (1980) found that novices and experts use less external information,
while customers with intermediate levels of expertise use more. While, other researchers
indicate that experienced customers will be using less external information than novices
(Chi et al., 1981). Raju et al. (1995) pointed out that novices use less information in making a
decision because of their lower ability of processing information.
Moreover, customers use different message-processing strategies depending on their
level of expertise (Boon and Park and Kim, 2008; Park and Kim, 2008; Hong and Sternthal,
2010; Ju, 2015). The type of reviews on purchase intention is stronger for experts than for
novices, while the effect of the number of reviews on purchase intention is stronger for
novices than experts (Boon and Park and Kim, 2008). The main difference between experts
and novices customers is the way they use obtainable information from market
communication process, even the information content stays the same (Ju, 2015). Moreover,
when information is limited, Ju (2015) found that novice customers are more likely to prefer
variety than expert customers.
JRIM Citation Market Items Type Remarks
13,4
Resnick et al. eBay 200 items Auction There is a reputation price
(2006) premium, as for vintage postcards,
which is 8.1% on average
Hui et al. (2016) eBay Multiple categories Auction and Adding buyer protection reduces
Posted price the premium for the reputation
582 badge and increases marketplace
efficiency
Houser and eBay Pentium III 500 Auction Seller reputation affects price
Wooders (2006) processors significantly
Melnik and Alm eBay $ 5 gold coin in 1999 Auction Reputation has a significantly
(2002) positive but small effect on price
Cabral and eBay IBM Thinkpad T23 PIII, Auction Negative feedback decreases sales
Hortacsu (2010) Collectible Coins and rate, especially the first one
Teddy Bear
Zhang (2006) eBay 15G and 20 G Apple iPod Auction Final price and sale probability are
MP3 Players correlated with seller’s selling
reputation but uncorrelated with
seller’s buying reputation
Chen et al. (2008) Amazon Books Posted price The relationship between rating
score and sales is positive, and
higher quality reviews impact
consumers more
Amblee and Bui Amazon Books in PDF Posted price Book ratings by readers do not
(2007) predict sales well; book sales are
related to some reviews, author’s
reputation, and complementary
goods’ reputation
Roth and Amazon Computers and Antiques Auction The amount of late bidding is
Ockenfels (2002) and eBay different between Amazon and
eBay because of different auction
rules. The amount of late bidding
decreases with experience
You et al. (2011) Taobao Multiple categories Posted price Seller reputations can be
manipulated; new buyers are more
likely to get involved
Ye et al. (2013) eBay and Cups Posted price Seller reputation has a positive
Taobao influence on price on eBay but
sales volume on Taobao.com
because of market structure
Table I. Liu et al. (2012) BizRate Eight categories Posted price A high-reputation seller charges a
Summary of primary lower price than a low-reputation
research literature seller

Several studies indicate that experts and novices also difference in the way they evaluate
the quality of the product (Raju et al., 1995). Srivastava and Mitra (1998) explored
whether prior knowledge moderates customers’ evaluation of product quality. They
found that, for experts, a better warranty leads to perceptions of higher quality,
regardless of firm reputation. While novices tend to perceive a better warranty as a signal
of higher quality only when the firm is reputable but not when its reputation is low. They
also indicate that reputation is the more salient cue. Previous studies demonstrated that
expert can make better predictions compare to novices, based on external noisy
information (Shirai and Meyer, 1997). Similarly, Cindy (2014) indicated customers who Role of
are experts can be trusted to give an accurate reflection of brand quality. In the wine customer
market, D’Alessandro and Pecotich (2013) found that novices experienced difficulty in
evaluating quality, while experts did use physical quality and price evaluations.
experience
The use of price as an indicator of product quality also differs between experts and
novices. Specifically, novice or unfamiliar customers are more likely to use price as an
indicator of quality than experts or familiar customers (Rao and Monroe, 1988). Hamilton
and Chernev (2013) described the key drivers of price image formation and their 583
consequences for consumer behavior. Moreover, the use of intrinsic cues for product quality
assessments become stronger as subject product familiarity increases. Mattila and Wirtz
(2001) design an experiment and get a similar result, that expert customers reliance on price
signals less, while, novices are more likely to rely on high price in making quality inferences.
Ye et al. (2014) empirically investigated the influence of price on customers’ perceptions of
service quality.

2.4 Customer online shopping experience

Previous studies indicate that knowledge in the area of customer experiences in online
shopping environments remains emergent (Martin et al., 2015; Rose et al., 2012; Trevinal and
Stenger, 2014). Except for Hui et al. (2016), studies have noted that customers with different
experiences may conceive a value difference in an eBay seller’s reputation through an
analysis of the seller’s customer protection program. Kukar-Kinney et al. (2012) indicated
that a customer’s extensive shopping experience influences knowledge of store prices. The
effect of initial trust on a customer’s purchasing intention is the difference between an
experienced and inexperienced customer (Stouthuysen et al., 2018). Zou et al. (2011) indicated
that customers with low knowledge are more likely to search and believe online review.
Cheema and Papatla (2010) found that experienced internet users have lower interest in
online information sources. Roth and Ockenfels (2002) proved that customer experience
influences their bidding behavior. However, the previous research literature on reputation
price premiums seldom considers the effects of customer characteristics. To bridge this gap,
this paper focuses on the role of the customer experience in moderating the relationship
between reputation and price premium, and it enriches studies of the reputation system and
price premium in the online marketplace.

3. Model and hypotheses

3.1 Model
Sellers offer similar products in the online marketplace. Let Q be the product value (i.e. the
fundamental truth of the product’s quality). Each product has an inherent value, which
we denote as Q [ [0,1]. The larger the value of Q, the higher the inherent value. Let # be the
customer’s marginal willingness to pay for a product. P is the price of the product. The
utility function can be expressed as U(P, #) = #Q – P.
A customer estimates the product quality by referring to the quality information of a
seller, that is, seller reputation R [ [0,1] and product price P. A seller with a high-value
reputation R indicates the seller is reputable and has a low transaction risk; thus, let 1 – R
represent customer risk in a transaction with the seller. What’s more, customers with
different experience u have different knowledge and, therefore, different ability to deal with
the information (Hui et al., 2016). Let gðu Þ denote that the customer’s perception of signaling
risks depends on the seller’s reputation. Customers with more experience could deal with
other quality signals better and are less sensitive to the risk indicated by the seller’s
reputation; in this case, g’(u ) < 0. However, if customers with more experience realize the
JRIM limitation of reputation and perceive more risk from the reputation system, then g’(u ) > 0.
13,4 Let f(u ) denote that the customer’s marginal willingness to pay for a product reflects
customer experience u . If a customer with more experience is more likely to search for the
lowest price, then f 0 (u ) < 0. However, if the experienced customer is more likely to perceive
price as a quality signal, then f 0 (u ) > 0. We denote the estimated quality as Qe [ [0,1],
expressed as Q – g(u )(1 – R). The utility that a customer with experience level u will
584 purchase from the seller is:


UðR; P; u Þ ¼ fðu Þ Q
gðu Þð1
RÞ
P
We impose a non-negativity constraint on price P. We assume the customer will purchase
the product if his or her utility is positive:


Uð R; P; u Þ ¼ fðu Þ Q
gðu Þð1
RÞ
P > 0
We get the minimum requirement for the quality of products sold by sellers:
P
Q>Q¼ þ gðu Þð1
RÞ
f ðu Þ
Therefore, we have a seller demand:
P
D¼1
Q ¼1

g ð u Þ ð 1
RÞ
f ðu Þ
We describe seller revenue as:
1 2

p ¼
P þ 1
gðu Þð1
RÞ P
f ðu Þ
The seller’s decision problem is to find an optimal price to maximize profit, which is @@Pp ¼ 0.
Then, we have:


1
gðu Þð1
RÞ fðu Þ
P¼
2
The maximum revenue of a seller is:

2
1
gðu Þð1
RÞ fðu Þ
p max ¼
2

P1. The reputation price premium is positive if customer willingness to pay for the
product is positive and perception risk depends on reputation.
Proof.

@P 1
¼ fðu Þgðu Þ
@R 2
@P
We denote @R to represent the relationship between reputation and price. It is positive where
f(u ) and g(u ) are both positive. The Taobao-like e-commerce platform typically uses a
reputation system to indicate seller trustworthiness, depending on the seller’s previous
dealings. A higher-reputation seller implies trustworthiness. Considering the potential Role of
transaction risk to customers of a low-reputation seller in the asymmetric information customer
markets, trusted sellers (sellers with high reputations) could charge the customer more. experience
P2. Customer experience negatively moderates the relationship between reputation and
price if f 0 (u ) < 0 and g 0 (u ) < 0.
Proof. 585
@2P 0 0
¼ f ðu Þgðu Þ þ fðu Þg ðu Þ < 0
@R @ u

The moderating role of customer experience between reputation and price are denoted as
@2 P
@R @ u . If f’(u ) < 0 (i.e. the relationship between customer experience and the price is negative)
and g’(u ) < 0 (i.e. customer experience relates to the perception of signaling risks
negatively), then customer experience decreases the reputation price premium effect. First,
with an increase in experience, customers are more likely to search for the lowest price.
Second, with an increase in experience, customers are less likely to estimate the quality,
depending only on seller reputation. For example, taobao.com provides some additional
guarantees for sellers to signal their creditability. A customer with greater experience
estimates the quality comprehensively, not merely on seller reputation. On the other hand, a
less-experienced customer is more likely to estimate quality solely on the seller’s reputation.

3.2 Hypotheses
We build a research model to identify the effects of customer experience on the relationship
between reputation and price in the online marketplace as depicted in Figure 1. Based on the
propositions proposed in the theoretical model, we build two hypotheses to clarify the role of
customer experience.
We focus on search products in our study. Search products are easy to identify by
product name and code in the online marketplace. Also, search products are less uncertain
because the product characteristics are easy to observe. Customer experience is expected to
influence the relationship between reputation and price premium when customer reliance on
the reputation system is related to customer experience. An underlying assumption behind
the research model is that the degree of reliance depends on customer experience. Also,

Figure 1.
Research model
JRIM transaction characteristics may affect price premium; thus, we include them as control
13,4 variables in our empirical analysis.
The reputation system aggregates feedbacks about seller transaction histories. Ideally,
the reputation system helps the customer to identify whether a seller is trustable or not.
Trustworthy sellers providing good products and services receive positive feedback,
eventually enhancing seller reputations. In this paper, we present price by two measures, the
586 relative price, and the price premium. The relative higher price or price premium is needed
to compensate the seller for reducing transaction risks. As proposed in P1, a higher-
reputation seller would enjoy a higher price because a good reputation is more likely to
increase a customer’s trust in the seller. Hence, we propose our first hypothesis below.

H1. Reputation price premium is positive. A higher reputation seller (e.g., seller rating,
the number of positive feedback comments) could charge a higher price.
The relationship between reputation and price is modified by the customer experience
negatively, as proposed in P2. Customers with wider experience are more likely to have
encountered negative experiences (Cheema and Papatla, 2010). For example, customers
negotiate with the seller before leaving negative feedbacks. Thus, customers with more
experience are more likely to find that seller reputations are not perfectly credible. What’s
more, customer experience in the online marketplace increases knowledge of the market
rules and regulations (Hui et al., 2016). Experienced customers are more likely to be exposed
to other quality signals, such as guarantees. Also, customers with broader experience are
more likely to have an accurate internal reference price and less likely to pay a price
premium. As experience grows, customers become more capable of distinguishing sellers by
multiple signals; they depend on seller reputations less and become more capable of
searching for the lowest price of a qualified product. Therefore, we propose the following
H2.

H2. Customer experience negatively moderates reputation price premium. An increase

in customer experience decreases their willing to pay the reputation price premium.

4. Data and description

4.1. Data collection
We collected the data in this study from Taobao.com, which consists of the entire
transaction records of the Kingston TF/Micro SD (2 G) and Nokia 5320XM products during
half-year period. Specifically, 16,842 transactions of the Kingston TF/Micro SD (2 G)
purchased from 1,737 sellers, and 57, 997 transactions of the Nokia 5320XM purchased from
2,317 sellers. Considering the wide range of items sold on taobao.com, we focus our analysis
on the Kingston TF/Micro SD (2 G) and the Nokia 5230XM for the following two reasons.
First, they are standard electronics products, which we can assume are homogeneous.
Second, we could test our hypothesis in different product markets. The intrinsic value of the
product is different: the reference price of a new Nokia 5230XM is US$160; a new Kingston is
US$6.
Atypical product Web page offered on taobao.com, usually show information that
includes price, postage, seller reputation, sales volume, guarantees and payment. On taobao.
com, guarantees consist of accurate descriptions, seven-day returns, 30-day repairs,
lightning deliveries and donations. Guarantees are often a voluntary process in which
sellers choose to participate. The information on the Web page can affect purchase decisions
(Wells et al., 2011). We show some summary statistics of the transactions of the Nokia Role of
5230XM and Kingston TF/Micro SD (2 G) products in Table II. customer
experience
4.2 Key variables
Our analysis uses two reputation measures. One is the numerical seller rating, a number
between 0 and 20, which indicates the performance of a seller since opening. The second
reputation measure is the number of positive feedback comments, which is based on the
number of positive feedbacks that a seller has received in the last month. Since seller
587
reputation is shown to potential customers in a different form on the Web page, we can use
different measures to capture reputation information. The variables contained in the
empirical analysis are shown in Table III. On taobao.com, the actual transaction price
consists of posted price, discount, and postage. For a better understanding of customer
behavior in the posted price marketplace, we calculate the transaction price as:

Transaction price ¼ Posted price
Discounts þ Postage

5. Empirical analysis
Our theoretical model provides an analytical framework to predict customer purchasing
behavior in response to hidden sellers and products, given a customer’s different experience
level. In this section, we empirically test our hypotheses using the data of the Nokia 5230XM
collected on Taobao.com, the biggest online market in China. We focus on three questions:
(1) How sellers adjust prices as their reputations change (reputation price premium)
on average;
(2) How willingness to pay a certain price changes with adjustments in customer
experience (experience price discount) on average; and
(3) How the relationship between reputation and price changes with adjustments in
customer experience (moderator effect).

5.1. Empirical model

To investigate the moderating influence of customer experience in the relationship between
reputation and price in the actual transaction, we specify standard cross-sectional models in
the following form:

Nokia 5230XM Kingston SD card

AVG SD MIN MAX AVG SD MIN MAX

Price 1139.450 93.809 880 1396 34.653 4.469 20 44

Experience level 4.997 1.870 1 12 5.470 1.877 1 12
Seller rating 13.9 3.069 1 18 12.38 3.597 1 18
Positive comment 0.998 0.009 0 1 0.993 0.009 0 1
Credit card 0.582 0.493 0 1 0.519 0.500 0 1
Donation 0.483 0.500 0 1 0.073 0.260 0 1
Acc_description 0.932 0.251 0 1 0.862 0.344 0 1
Seven-day return 0.853 0.354 0 1 0.735 0.441 0 1
30-day repair 0.488 0.5 0 1 0.323 0.468 0 1 Table II.
Free shipping 0.025 0.155 0 1 0.110 0.313 0 1 Summary statistics
JRIM Variables Description
13,4
Price premium P_REL Represents standardized transaction price, which consists of
(PRICE) posted price, discount, and postage
P_PRE 1 if the price is higher than the average price among all
transactions in the half-year period
Seller reputation REP_RAT Sum of feedbacks received by the seller to date
588 (REPUTATION) POS_NUM Number of positive feedbacks received by the seller
Customer experience EXPE Number of purchases completed by the customer to date
Seller characteristics SALES_PRE Cumulative sales of the seller
CREDITCARD Indicates whether the seller accepts credit card payments
DONATION Indicates whether the seller provides donation guarantees.
Donates part of the payment to charitable organizations
ACC_DEC Indicates whether the seller provides an accurate description
guarantee or not. Descriptions and photos are guaranteed to
match the real product
RETURN Indicates whether the seller provides a return guarantee:
without reason, the customer can return the product within a
week after receipt
REPAIR Indicates whether seller provides a 30-day repair guarantee:
maintenance service for free within 30 days after receipt
DELIVERY Indicates whether the seller provides a lightning delivery
guarantee: delivers the product as soon as the order is placed
Table III. Transaction FREESHIP Indicates that the transaction includes free shipping
Main variables used characteristics DISCOUNTS Indicates that the transaction enjoys a discount price
in the analysis WEEK Indicates the week period when the transaction occurs

Yi;t ¼ Xi;t b þ « i;t

where Y represents the endogenous variables, such as the transaction price; X represents a
vector of independent variables, including seller reputation, customer experience, seller
characteristics, and transaction characteristics; b is a vector of estimated parameters; and «
is the error term associated with each transaction i placed at time t.
In detail, we carried out a moderated regression analysis, using two-stage regression, to
assess the effect of the moderator variable. We regress the price with the reputation,
customer experience, and control variables in the first stage; then, we add the cross term
(REPUTATION  EXPE) in the second stage. We present the reputation of the seller by two
measures: rating score and positive feedbacks. Accordingly, we specify our empirical
models for testing our hypothesis as follows:

PRICEit ¼ b 0 þ b 1 REPUTATIONi;t þ b 2 REPUTATIONi;t  EXPEi;t þ b 3 EXPEi;t

þ b 4 SALES_PREi;t
1 þ b 5 CREDITCARDi;t þ b 6 DONATIONi;t
þ b 7 ACC_DECi;t þ b 8 RETURN i;t þ b 9 REPAIRi;t þ b 10 DELIVERY i;t
þ b 11 FREESHIP i;t þ b 12 DISCOUNT i;t þ e i;t

where i denotes the transaction and t denotes the time at which the transaction is placed.
We are interested in the reputation variables, and we have specifications that consider
the score-based variable (REP_RAT), the comment-based variable (POS_NUM), and the
combination of both. We include variables that refer to the Web page on taobao.com, Role of
which influence customer purchasing behavior. To be specific, we account for seller-level customer
heterogeneity by including relevant factors, such as guarantees (accurate descriptions,
experience
seven-day returns, 30-day repairs, lightning deliveries, and donations), credit card
payments, and previous sales. We implicitly control for the transaction-level
heterogeneity in our models, which consist of free shipping and discounts. In addition, we
include a dummy for the week period. We also include seller fixed effect to control for the 589
other unobservable characteristics of the seller.
If the coefficient of the reputation variables is positive and significant, then H1 (the
transaction price increases with seller reputation) will be supported. The expected effect of
customer experience on price is negative. On the one hand, the higher the experience levels
of customers, the more likely they can search for the lowest price in the market. On the other
hand, customers with wide experience are more capable of distinguishing sellers by multiple
signals, depending less on seller reputations.
We do not suppose that reputation and experience are independent of each other, and our
models account for any potential correlation among variables. The correlations among the
variables are relatively small in our sample. Table IV shows the correlations among the
variables in the transactions of Nokia and Kingston. Despite the significant correlations
among these variables, there was no evidence of multicollinearity with all variable-inflation
factors well below the threshold of 10 (Hair et al., 2006). We also verified that the normality
assumption and the heteroscedasticity are satisfied.
We observe that the transaction price has a positive correlation with seller reputation and
the number of positive comments. Higher-reputation sellers are more likely to enjoy higher

1 2 3 4 5 6 7 8 9 10

Nokia 5230XM
1 PRICE 1
2 EXPE
0.023* 1
3 REP_RAT 0.046*
0.05* 1
4 REP_POS 0.014*
0.059* 0.074* 1
5 CREDITCARD 0.098*
0.018*
0.2718
0.006 1
6 DONATION 0.017*
0.047* 0.642* 0.06*
0.065* 1
7 ACC_DEC 0.055*
0.031* 0.109* 0.057* 0.053* 0.109* 1
8 RETURN 0.014*
0.058* 0.241* 0.04* 0.063* 0.227* 0.65* 1
9 REPAIR
0.013*
0.019* 0.676* 0.085*
0.264* 0.516* 0.263* 0.351* 1
10 DELIVERY 0.099* 0.005
0.132* 0.002 0.095*
0.114* 0.043* 0.029*
0.103* 1
Kingston SD card
1 PRICE 1
2 EXPE
0.097*
3 REP_RAT 0.146*
0.001
4 REP_POS 0.128*
0.025* 0.056*
5 CREDITCARD 0.126* 0.045* 0.505* 0.066*
6 DONATION 0.054*
0.010 0.067*
0.066* 0.161*
7 ACC_DEC 0.065* 0 0.427* 0.045* 0.289* 0.092*
8 RETURN 0.037*
0.027* 0.214* 0.075* 0.208* 0.073* 0.666*
9 REPAIR 0.128*
0.025* 0.398* 0.187* 0.299* 0.096* 0.276* 0.366*
10 DELIVERY
0.029*
0.003
0.045* 0.013 0 0.021* 0.140* 0.183* 0.027* Table IV.
Pairwise correlations
Note: *Significant at the 1% level among variables
JRIM prices, which provide the initial support for H1. Also, customer experience shows a negative
13,4 relationship with transaction price, seller reputation, and positive feedbacks. The correlation
among the variables verifies the conditions f 0 (u ) < 0 and g 0 (u ) < 0 in P2 and provide the
initial support for H2.

590 5.2. Results

We estimate the coefficients by using a fixed-effects estimator (least squares dummy
variable), which are the estimates shown in Table V. Columns 1 and 2 use the seller rating,
columns 3 and 4 use the number of positive comments, and columns 5 and 6 combine those
two sources of reputation information. For each group, the first column provides the results
of stage 1 regression, and the second column adds interaction between reputation and
experience. For example, column 1 shows the results for the coefficients of the seller rating
and experience. The results show that price does increase with reputation. Therefore, H1 is
validated. For an additional reputation point, price increases by 6.6 per cent of the standard
deviation. The coefficient of experience shows that price decreases with customer
experience. Column 2 adds the interaction between reputation and customer experience. The
cross-product term (REPUTATION  EXPE) is significant, proving the moderating effect
on the relationship between reputation and price. The sign for the cross-product term
(REPUTATION  EXPE) coefficient is negative, indicating that the reputation price
premium slope moves in a negative direction as customer experience increases. This result
supports H2. Using positive comments as reputation information (Columns 3 and 4), we
have qualitatively similar results. When combining the two sources of reputational
information, the results continue to hold although customer experience not significant
(probably because of the high correlation with the cross-product terms).
Figure 2 further shows that the highest price is reached when a less-experience customer
meets a high-reputation seller (Y = 0.1) or more positive-comments seller (Y = 0.04). As
predicted, an experienced customer enjoys a lower price even though trading with a high-
reputation seller (Y = 0.04) or more positive-comments seller (Y = –0.01). The relationship
between positive comments and price could even be negative in the group of more-
experience customers because of the moderating effect of customer experience. Sellers with
more positive comments cannot enjoy price premiums when trading with experienced
customers. Sellers with high positive comments offer relatively low prices, because of
reputation management (Fan et al., 2016) and competition with other sellers (Liu et al., 2012).
This result indicates that customer experience could be an explanation for the conflicting
results in reputation price premiums studies. Customers with more experience are mainly
affected by reputation management and competition. Previous studies underrate the
negative-reputation price premium in consideration of customer heterogeneity.
Overall, these results show support for our hypothesis on the relationships among
transaction price, seller reputation and customer experience. We identify the moderator
effect of customer experience on the relationship between seller reputation and price.

6. Robustness analysis
We perform robustness checks to verify the validity of our results. First, we consider an
alternative definition of a price premium to show the effect of reputation and experience.
Then, we perform an analysis of another product category market. We give an extensive
discussion of robustness checks to provide confidence in our findings.
 Relative price
Seller rating Positive comments Both measures
Stage1 Stage2 Stage1 Stage2 Stage1 Stage2

EXPE
0.022*** (0.004 )
0.012** (0.005 )
0.022*** (0.004 )
0.012** (0.006 )
0.022*** (0.004 )
0.078 (0.006 )
REP_RAT 0.066*** (0.009 ) 0.071*** (0.010 ) 0.067*** (0.009 ) 0.041*** (0.009 )
REP_POS 0.013*** (0.004 ) 0.013*** (0.004 ) 0.013*** (0.004 ) 0.020*** (0.004 )
RAT *EXP
0.016*** (0.006 )
0.038*** (0.006 )
POS*EXP
0.015*** (0.006 )
0.002*** (0.005 )
SALES_PRE
0.162*** (0.009 )
0.163*** (0.009 )
0.120*** (0.006 )
0.120*** (0.006 )
0.164*** (0.009 )
0.159*** (0.009 )
CREDITCARD 0.122*** (0.009 ) 0.125*** (0.009 ) 0.126*** (0.009 ) 0.126*** (0.009 ) 0.121*** (0.009 ) 0.134*** (0.009 )
DONATION 0.141*** (0.011 ) 0.143*** (0.011 ) 0.158*** (0.011 ) 0.158*** (0.011 ) 0.141*** (0.011 ) 0.121*** (0.011 )
ACC_DEC 0.289*** (0.022 ) 0.290*** (0.022 ) 0.277*** (0.022 ) 0.276*** (0.022 ) 0.285*** (0.022 ) 0.225*** (0.022 )
RETURN
0.083*** (0.016 )
0.083*** (0.016 )
0.080*** (0.016 )
0.081*** (0.016 )
0.081*** (0.016 )
0.038*** (0.016 )
REPAIR 0.098*** (0.012 ) 0.101*** (0.012 ) 0.114*** (0.012 ) 0.115*** (0.012 ) 0.096*** (0.012 ) 0.050*** (0.012 )
DELIVERY 0.384*** (0.027 ) 0.382*** (0.027 ) 0.381*** (0.027 ) 0.381*** (0.027 ) 0.384*** (0.027 ) 0.489*** (0.027 )
FREESHIP
0.175*** (0.009 )
0.174*** (0.009 )
0.186*** (0.009 )
0.186*** (0.009 )
0.175*** (0.009 )
0.188*** (0.009 )
DISCOUNTS
0.153*** (0.037 )
0.147*** (0.037 )
0.145*** (0.037 )
0.141*** (0.037 )
0.154*** (0.037 )
0.145*** (0.037 )
CONSTANT
0.551*** (0.018 )
0.553*** (0.018 )
0.555*** (0.018 )
0.554*** (0.018 )
0.547*** (0.018 )
0.428*** (0.017 )
Seller Fixed H H H H H H
No. of Obs. 57,997 57,997 57,997 57,997 57,997 57,997

Notes: ***, ** and * indicate significance at the 1, 5; and 10% levels, respectively

effect of customers’
The moderating
Table V.
customer

experience
591
experience
Role of
JRIM
13,4

592
Figure 2.
The moderating
effect of consumer’s
experience of seller
rating (upper),
positive comments
(below) with relative
price as an
independent variable

6.1 Definition of a price premium

We start by showing that the results are not sensitive to the definition of price and our
normalization method for the price. The relative price is a continuous variable, normalized
by price. In this section, we define the price premium as a transaction price that is higher
than the average price. Changing this definition to format price premium as 0 and 1 has little
to no effect on results.
We applied logistic regression to estimate the probability of price premium. We include
similar variables as before; the variables set are represented as X = (REPUTATIONi,t,
EXPEi,t, . . .). We assume the conditional probability P(Y = 1|X) = p as the probability of the
price premium. The logistic regression is expressed as:
1
PðY ¼ 1 jXÞ ¼ p ðXÞ ¼
1 þ e
gðXÞ

gðxÞ ¼ b 0 þ b 1 REPUTATIONi;t þ b 2 REPUTATIONi;t  EXPEi;t þ b 3 EXPEi;t

þ b 4 SALES_PREi;t
1 þ b 5 CREDITCARDi;t þ b 6 DONATIONi;t
þ b 7 ACC_DECi;t þ b 8 RETURNi;t þ b 9 REPAIRi;t þ b 10 DELIVERYi;t
þ b 11 FREESHIPi; j;t þ b 12 DISCOUNTSi; j;t þ e

Obviously, p (X) [ [0,1]. Let b = ( b 1, b 2, . . ., b 12). We apply a maximum likelihood method

to estimate the probability of the price premium, Y = 1. If we assume y [ [0,1] and m
observations {y1, y2, . . ., ym}, the likelihood function of these observations is:
Y
m Y
m
y
1
yi
LðbÞ ¼ pðyi Þ ¼ pð X Þ i 1
pð X Þ
i¼1 i¼1
Then, we have: Role of
m h 
X   i customer
lnL ¼ yi gð X Þ
ln 1 þ egð X Þ experience
i¼1

We estimate the b 0, b 1, b 2, . . ., b 12 to maximum lnL, using the Newton-Raphson method

for achieving a solution in the iterative algorithm. Table VI presents the logistic regression 593
results to test the effect of customer experience and seller reputation on price premium. The
regression results indicate that customer experience reduces the effects of a seller’s rating
( b = –0.021) and positive comments ( b = –0.02) on price premium. Additionally, the
relationship between seller rating (positive comments) and the price premium is positive,
which is consistent with our earliest results.
Based on the results of Table VI, we illustrate the direction of the moderating effect
of customer experience in Figure 3. A high-reputation seller with more positive
comments enjoys a positive price premium. Thus, H1 is verified. As for the more-
experience customer, the relationship between price premium and seller rating (positive
comments) at a lower level slowly increases as the seller rating increases. In contrast,
for the less-experience customer, the relationship starts at a higher level and increases
at a faster rate as the seller rating (positive comments) increases. Thus, less-experience
customers value seller rating (the sum of feedback scores) or positive comments more
compared to customers with more experience. Customer experience moderates
reputation price premium negatively.

6.2. Robustness check on the product category

Previous studies have indicated that product categories might affect customer behavior. For
example, product types influence customer acceptance of online shopping (Lian and Lin,
2008), and product popularity affects the relationship between online reviews and sales (Zhu
and Zhang, 2010). In consideration of product differentiation, our results may not be
consistent. Thus, we introduce another product category, the Kingston SD card, a low-value
and less-popular product. To check the sensitivity of the moderator effect on the product
categories, we repeat this analysis for the Kingston SD card market.
Results are presented in Table VII. As in the previous analysis of the Nokia 5230XM, we
use relative price and price premium to measure price, analyzed by moderated regression
and logistic regression separately. Similarly, we use the seller rating, positive comments,
and the combination of these two to measure seller reputation. The signs of the regression
coefficient of REP_RAT and REP_POS are positive, indicating that the Kington SD card
seller also enjoys a reputation price premium. H1 is confirmed. The results indicate that the
number of seller positive comments have a positive and significant effect on the relative
price ( b = 0.155) and the probability of a price premium ( b = 0.407). However, the relative
price is not influenced significantly by the seller rating ( b = 0.011), although there is a
significant and positive effect on the probability of a price premium ( b = 0.204). The
coefficient of the cross-product term RAT*EXP (POS*EXP) is negative and significant,
indicating an adverse moderating effect of customer experience on the relationship between
seller rating (or positive comments) and price in the Kingston SD card market, which is
consistent with our analysis results on the Nokia 5230XM market. The moderating effect of
customer experience on the relationship between reputation and price is negative for both
high-value products and low-value products.
 13,4

594
JRIM

Table VI.
Robustness check,

the price premium

logistic regression on
Price premium
Seller rating Positive comments Both measures
Stage1 Stage2 Stage1 Stage2 Stage1 Stage2

EXPE
0.040*** (0.009 )
0.027** (0.012 )
0.041*** (0.009 )
0.027** (0.012 )
0.041*** (0.009 )
0.026** (0.012 )
REP_RAT 0.113*** (0.020 ) 0.119*** (0.020 ) 0.114*** (0.020 ) 0.120*** (0.021 )
REP_POS 0.103*** (0.023 ) 0.104*** (0.023 ) 0.105*** (0.023 ) 0.107*** (0.023 )
RAT *EXP
0.021* (0.013 )
0.022 (0.030 )
POS*EXP
0.020* (0.012 )
0.002 (0.028 )
SALES_PRE
0.406*** (0.019 )
0.406*** (0.019 )
0.340*** (0.013 )
0.341*** (0.013 )
0.417*** (0.019 )
0.417*** (0.019 )
CREDITCARD 0.323*** (0.019 ) 0.326*** (0.019 ) 0.326*** (0.019 ) 0.326*** (0.019 ) 0.318*** (0.019 ) 0.321*** (0.020 )
DONATION 0.251*** (0.023 ) 0.253*** (0.023 ) 0.278*** (0.023 ) 0.278*** (0.023 ) 0.247*** (0.023 ) 0.249*** (0.024 )
ACC_DEC 0.453*** (0.046 ) 0.455*** (0.046 ) 0.417*** (0.046 ) 0.416*** (0.046 ) 0.430*** (0.046 ) 0.432*** (0.046 )
RETURN
0.045 (0.034 )
0.045 (0.034 )
0.028 (0.034 )
0.028 (0.034 )
0.030 (0.034 )
0.029 (0.034 )
REPAIR 0.289*** (0.025 ) 0.293*** (0.026 ) 0.305*** (0.025 ) 0.305*** (0.025 ) 0.272*** (0.026 ) 0.276*** (0.026 )
DELIVERY 0.630*** (0.062 ) 0.627*** (0.062 ) 0.623*** (0.062 ) 0.623*** (0.062 ) 0.628*** (0.062 ) 0.625*** (0.063 )
FREESHIP
0.353*** (0.019 )
0.352*** (0.019 )
0.375*** (0.019 )
0.375*** (0.019 )
0.356*** (0.019 )
0.355*** (0.019 )
DISCOUNTS
0.369*** (0.082 )
0.361*** (0.082 )
0.358*** (0.082 ) 0.082*** (0.082 )
0.374*** (0.082 )
0.366*** (0.082 )
CONSTANT
1.102*** (0.039 )
1.105*** (0.039 )
1.097*** (0.039 )
1.095*** (0.039 )
1.083*** (0.039 )
1.085*** (0.039 )
Seller Fixed H H H H H H
No. of Obs. 57,997 57,997 57,997 57,997 57,997 57,997

Notes: ***, ** and * indicate significance at the 1, 5 and 10% levels, respectively
 Role of
customer
experience

595
Figure 3.
The moderating
effect of consumer’s
experience of seller
rating (upper),
positive comments
(below) with a price
premium as an
independent variable

7. Discussions and conclusions

7.1 Theoretical implications
This study contributes to the literature by examining the customer characteristic influences
the value of the online reputation system and exploring the effects of seller characteristics
and transaction characteristics on price. Customers with more experience rely less on the
reputation system and value it less. Our study bridges the existing research gaps by
showing that customer heterogeneity weakens the positive effect of reputation on price
premium. Therefore, to evaluate the effectiveness of the reputation system in the online
marketplace, one must consider customer features, such as their experience level.
This research provides theoretical contributions in two primary ways. First, this study
provides a potential reconciliation of the mixed results from previous studies. For example,
Houser and Wooders (2006) found a significant positive relationship between price and
reputation, but Liu et al. (2012) indicated a negative relationship between them. Some
researchers have tried to reconcile the stark differences in results and attribute them to
marketplace differences. For example, Ye et al. (2013) pointed out that the mixed findings
could be the result of market structure. Our study suggests that data sets of different
product categories could lead to similar conclusions, but differences in customer experiences
could result differently. Our finding addresses the conflicts found in previous studies by
extending the explanation of the negative reputation price premium.
Second, this research contributes to the reputation management and customer behavior
literature by identifying the effects of customer experience on the positive relationship
between the reputation system and price premium. Previous studies seldom considered the
heterogeneity of customers in analyses of reputation value. As a marketing tool, reputation
management has become increasingly popular. Usually, online sellers could manage their
reputation to increase profits (Fan et al., 2016). While reputation management reduces
customer reliance on reputation, it decreases seller profits. Our paper highlights that
reputation management may be a double-edged sword. Additionally, our work is one of the
few studies which include purchasing experience as a variable that moderates e-customer
behavior and represents experience by the number of actual transactions in the marketplace.
 13,4

596
JRIM

Table VII.

Kingston SD card
Robustness analysis,
Relative price Price premium
Seller rating Positive comments Both measures Seller rating Positive comments Both measures

EXPE
0.049*** (0.008)
0.037*** (0.010)
0.031** (0.010)
0.122*** (0.021)
0.106** (0.029)
0.098*** (0.029)
REP_RAT 0.011 (0.022) 0.003 (0.022) 0.204*** (0.054) 0.210*** (0.056)
REP_POS 0.155*** (0.007) 0.150*** (0.007) 0.407*** (0.024) 0.393*** (0.024)
RAT *EXP
0.104*** (0.021)
0.082*** (0.022)
0.380*** (0.057)
0.352 (0.059)
POS*EXP
0.038*** (0.010)
0.030*** (0.010)
0.080*** (0.025)
0.037*** (0.026)
SALES_PRE
0.163*** (0.008)
0.192*** (0.008)
0.166*** (0.008)
0.339*** (0.023)
0.406*** (0.022)
0.358*** (0.024)
CREDITCARD 0.117*** (0.017) 0.082*** (0.017) 0.112*** (0.017) 0.250*** (0.039) 0.203*** (0.039) 0.243*** (0.039)
DONATION 0.043 (0.029) 0.053* (0.029) 0.081*** (0.029)
0.025 (0.066) 0.044 (0.065) 0.077 (0.066)
ACC_DEC 0.009 (0.029)
0.020 (0.028) 0.018 (0.029) 0.027 (0.066) 0.039 (0.066) 0.087 (0.067)
RETURN
0.162*** (0.024)
0.110*** (0.023)
0.165*** (0.024)
0.271*** (0.055)
0.220*** (0.053)
0.286*** (0.056)
REPAIR 0.168*** (0.018) 0.104*** (0.018) 0.116*** (0.018) 0.318*** (0.041) 0.185*** (0.042) 0.191*** (0.042)
DELIVERY
0.005 (0.025)
0.005 (0.025)
0.010 (0.025)
0.201*** (0.057)
0.195*** (0.057)
0.205*** (0.057)
FREESHIP
0.589*** (0.016)
0.597*** (0.016)
0.596*** (0.016)
1.192*** (0.038)
1.227*** (0.039)
1.230*** (0.039)
DISCOUNTS
0.198*** (0.043)
0.166*** (0.043)
0.170*** (0.043)
0.084 (0.099)
0.032 (0.100)
0.023 (0.100)
CONSTANT 0.452*** (0.026) 0.447*** (0.026) 0.453*** (0.026) 0.442*** (0.060) 0.413*** (0.061) 0.418*** (0.061)
Seller Fixed H H H H H H
No. of Obs 16,842 16,842 16,842 16,842 16,842 16,842

Notes: ***, ** and * indicate significance at the 1%, 5% and 10% levels, respectively
7.2 Practical implications Role of
Understanding how reputation affects customer purchasing behavior is vitally important. customer
Our findings have important implications for participants in online marketplaces because
online sellers rely on online reputation to disseminate information about their product. At
experience
the participant level, understanding how customer experience moderates the relationship
between reputation and price premium may help sellers design their online marketing
strategy more effectively in the online marketplace. For example, new websites may be
predominantly populated by the naïve customer or certain product in an early phase of the
597
product lifecycle where most customers are less experienced. Knowing this helps firms
make their optimal pricing strategy and product strategy.
Sellers that wish to compete in the e-market must understand the type of customers they
are addressing and develop their strategy according to whether they want to capture new e-
customers or retain existing ones. Our results also have important implications for the low-
reputation seller to survive in the context of “long tail” (Anderson, 2006). Besides low pricing
strategy, low-reputation sellers or new sellers could compete for the experienced customer
by other strategies, for instance, third-party certifications.
At the managerial level, our results provide the support for platforms to improve their
transaction mechanisms. As the e-commerce reputation system broadens, customers will
inevitably become more experienced with the online market. Our study suggests that, over
time, marketing managers will find the online reputation system to be decreasingly
influential and thus devote more resources to other channels to indicate trustworthiness.
Moreover, our results suggest that the benefit of high-reputation sellers decrease with
customer experience increase.

7.3 Conclusions
This study attempts to investigate the influences of customer characteristics on the value of
the reputation system by using the entire transaction records of the Nokia 5230XM and
Kingston SD card in a half-year period on Taobao.com. Using theoretical and empirical
analyses, we obtained three critical findings. First, sellers with high rating and positive
comments enjoy a positive price premium, which is consistent with previous studies that
focus only on the seller side. Second, customers with more experience are less likely to pay
the price premium and rely on a reputation system. Third, customer experience negatively
moderates reputation price premium; thus, the positive effect of the reputation system is
weaker for the experienced customer than for the naïve customer. This study addresses an
important area of research that has the potential to increase our understanding of customer
behavior in the online marketplace. The results of this study suggest that customer
experience plays an essential role in the effectiveness of the reputation system in the online
marketplace.
Several directions can be explored for future research. First, some authors may examine
our analysis in other online reputation platforms to generate additional insights into
customer behavior and the reputation system. Our research context is limited to the Taobao.
com marketplace. The generalizations of our study results could be tested in other contexts.
Second, future research could apply and extend our data analysis and robustness check
methods to other reputation issues. While our analysis focuses on new search products, the
results apply to other product categories as well. Third, further research could investigate
the effect of the helpfulness of reviews. The usefulness of online consumer reviews could
be captured by a text-mining method. Therefore, additional moderators could be added to
the analysis to enrich our findings.
JRIM References
13,4 Akerlof, G.A. (1978), “The market for ‘lemons’: quality uncertainty and the market mechanism”,
Uncertainty in Economics, pp. 235-251.
Amblee, N. and Bui, T.X. (2007), “The impact of Electronic-Word-of-Mouth on digital microproducts: an
empirical investigation of amazon shorts”, ECIS, pp. 36-47.
Anderson, C. (2006), The Long Tail: Why the Future of Business Is Selling More for Less, Hyperion, CA.
598 Babutsidze, Z. (2012), “How do consumers make choices? A survey of evidence”, Journal of Economic
Surveys, Vol. 26 No. 4, pp. 752-762.
Bart, Y., Shankar, V., Sultan, F. and Urban, G.L. (2005), “Are the drivers and role of online trust the
same for all websites and consumers? A large-scale exploratory empirical study”, Journal of
Marketing, Vol. 69 No. 4, pp. 133-152.
Becerra, E.P. and Korgaonkar, P.K. (2011), “Effects of trust beliefs on consumers’ online intentions”,
European Journal of Marketing, Vol. 45 No. 6, pp. 936-962.
Bettman, J.R. and Park, C.W. (1980), “Effects of prior knowledge and experience and phase of the choice
process on consumer decision processes: a protocol analysis”, Journal of Consumer Research,
Vol. 7 No. 3, pp. 234-248.
Biong, H. (2013), “Choice of subcontractor in markets with asymmetric information: reputation and
price effects”, Journal of Business and Industrial Marketing, Vol. 28 No. 1, pp. 60-71.
Bruce, N., Haruvy, E. and Rao, R. (2004), “Seller rating, price, and default in online auctions”, Journal of
Interactive Marketing, Vol. 18 No. 4, pp. 37-50.
Cabral, L. and Hortacsu, A. (2010), “The dynamics of seller reputation: evidence from eBay”, The
Journal of Industrial Economics, Vol. 58 No. 1, pp. 54-78.
Cabral, L. and Li, L. (2015), “A dollar for your thoughts: feedback-conditional rebates on eBay”,
Management Science, Vol. 61 No. 9, pp. 2052-2063.
Caruana, A. and Ewing, M.T. (2010), “How corporate reputation, quality, and value influence online
loyalty”, Journal of Business Research, Vol. 63 Nos 9/10, pp. 1103-1110.
Chang, M.K., Cheung, W. and Tang, M. (2013), “Building trust online: interactions among trust building
mechanisms”, Information and Management, Vol. 50 No. 7, pp. 439-445.
Cheema, A. and Papatla, P. (2010), “Relative importance of online versus offline information for internet
purchases: product category and internet experience effects”, Journal of Business Research,
Vol. 63 No. 9-10, pp. 979-985.
Chen, P.Y. Dhanasobhon, S. and Smith, M.D. (2008), “All reviews are not created equal: the
disaggregate impact of reviews and reviewers at amazon.com”.
Chen, P., Chen, Y., Hu, X. and Li, S. (2015), “Can online markets attract high-quality products?”,
Economic Modelling, Vol. 51, pp. 65-71.
Cheung, C.M., Xiao, B.S. and Liu, I.L. (2014), “Do actions speak louder than voices? The signaling role of
social information cues in influencing consumer purchase decisions”, Decision Support Systems,
Vol. 65, pp. 50-58.
Chi, M.T., Feltovich, P.J. and Glaser, R. (1981), “Categorization and representation of physics problems
by experts and novices”, Cognitive Science, Vol. 5 No. 2, pp. 121-152.
Dawar, N. and Parker, P. (1994), “Marketing universals: consumers’ use of brand name, price, physical
appearance, and retailer reputation as signals of product quality”, Journal of Marketing, Vol. 58
No. 2, pp. 81-95.
Dimoka, A., Hong, Y. and Pavlou, P.A. (2012), “On product uncertainty in online markets: theory and
evidence”, MIS Quarterly, Vol. 36.
Fan, Y., Ju, J. and Xiao, M. (2016), “Reputation premium and reputation management: evidence from the
largest e-commerce platform in China”, International Journal of Industrial Organization, Vol. 46,
pp. 63-76.
Gera, R. (2011), “Modelling e-service quality and its consequences in India: an SEM approach”, Journal Role of
of Research in Interactive Marketing, Vol. 5 No. 2/3, pp. 203-225.
customer
Ghose, A., Ipeirotis, P. and Sundararajan, A. (2007), “Opinion mining using econometrics: a case study
on reputation systems”, Proceedings of the 45th Annual Meeting of the Association of
experience
Computational Linguistics, pp. 416-423.
Hair, J.F., Black, W.C., Babin, B.J., Anderson, R.E. and Tatham, R.L. (2006), Multivariate Data Analysis,
Vol. 6, Pearson London.
Hamilton, R. and Chernev, A. (2013), “Low prices are just the beginning: price image in retail
599
management”, Journal of Marketing, Vol. 77 No. 6, pp. 1-20.
Hong, J. and Sternthal, B. (2010), “The effects of consumer prior knowledge and processing strategies
on judgments”, Journal of Marketing Research, Vol. 47 No. 2, pp. 301-311.
Houser, D. and Wooders, J. (2006), “Reputation in auctions: theory, and evidence from eBay”, Journal of
Economics and Management Strategy, Vol. 15 No. 2, pp. 353-369.
Hui, X., Saeedi, M., Shen, Z. and Sundaresan, N. (2016), “Reputation and regulations: evidence from
eBay”, Management Science, Vol. 62 No. 12, pp. 3604-3616.
Jøsang, A., Ismail, R. and Boyd, C. (2007), “A survey of trust and reputation systems for online service
provision”, Decision Support Systems, Vol. 43 No. 2, pp. 618-644.
Ju, X. (2015), “Consumers’ variety seeking: from an information perspective”, International Business
Research, Vol. 8 No. 3, p. 42.
Kim, J. and Lennon, S.J. (2013), “Effects of reputation and website quality on online consumers’
emotion, perceived risk, and purchase intention: based on the stimulus-organism-
response model”, Journal of Research in Interactive Marketing, Vol. 7 No. 1, pp. 33-56.
Klein, B. and Leffler, K.B. (1981), “The role of market forces in assuring contractual performance”,
Journal of Political Economy, Vol. 89 No. 4, pp. 615-641.
Kukar-Kinney, M., Ridgway, N.M. and Monroe, K.B. (2012), “The role of price in the behavior
and purchase decisions of compulsive buyers”, Journal of Retailing, Vol. 88 No. 1,
pp. 63-71.
Kwak, H., Puzakova, M. and Rocereto, J.F. (2015), “Better not smile at the price: the differential role of
brand anthropomorphization on perceived price fairness”, Journal of Marketing, Vol. 79 No. 4,
pp. 56-76.
Lee, S.H.M. (2014), “The role of consumers’ network positions on information-seeking behavior of
experts and novices: a power perspective”, Journal of Business Research, Vol. 67 No. 1,
pp. 2853-2859.
Lewis, G. (2011), “Asymmetric information, adverse selection and online disclosure: the case of eBay
motors”, American Economic Review, Vol. 101 No. 4, pp. 1535-1546.
Lian, J.W. and Lin, T.M. (2008), “Effects of consumer characteristics on their acceptance of online
shopping: comparisons among different product types”, Computers in Human Behavior, Vol. 24
No. 1, pp. 48-65.
Lin, C. and Lekhawipat, W. (2014), “Factors affecting online repurchase intention”, Industrial
Management and Data Systems, Vol. 114 No. 4, pp. 597-611.
Liu, Y., Feng, J. and Wei, K.K. (2012), “Negative price premium effect in online market—the impact of
competition and buyer informativeness on the pricing strategies of sellers with different
reputation levels”, Decision Support Systems, Vol. 54 No. 1, pp. 681-690.
Martin, J., Mortimer, G. and Andrews, L. (2015), “Re-examining online customer experience to include
purchase frequency and perceived risk”, Journal of Retailing and Consumer Services, Vol. 25,
pp. 81-95.
Mattila, A. and Wirtz, J. (2001), “The moderating role of expertise in consumer evaluations of credence
goods”, International Quarterly Journal of Marketing, Vol. 1 No. 4, pp. 281-292.
JRIM Mavlanova, T., Benbunan-Fich, R. and Lang, G. (2016), “The role of external and internal signals in E-
commerce”, Decision Support Systems, Vol. 87, pp. 59-68.
13,4
Mayzlin, D. (2006), “Promotional chat on the internet”, Marketing Science, Vol. 25 No. 2, pp. 155-163.
Melnik, M.I. and Alm, J. (2002), “Does a seller’s e-commerce reputation matter? Evidence from eBay
auctions”, The Journal of Industrial Economics, Vol. 50 No. 3, pp. 337-349.
Mosteller, J., Donthu, N. and Eroglu, S. (2014), “The fluent online shopping experience”, Journal of
600 Business Research, Vol. 67 No. 11, pp. 2486-2493.
Mui, L., Mohtashemi, M. and Halberstadt, A. (2002), “A computational model of trust and reputation”,
System Sciences, 2002. HICSS. Proceedings of the 35th Annual HI International Conference on,
pp. 2431-2439.
Neto, J.Q.F., Bloemhof, J. and Corbett, C. (2016), “Market prices of remanufactured, used and new
items: evidence from eBay”, International Journal of Production Economics, Vol. 171,
pp. 371-380.
O. Pappas, I., G. Pateli, A., N. Giannakos, M. and Chrissikopoulos, V. (2014), “Moderating effects of
online shopping experience on customer satisfaction and repurchase intentions”, International
Journal of Retail and Distribution Management, Vol. 42 No. 3, pp. 187-204.
Park, D.H. and Kim, S. (2008), “The effects of consumer knowledge on message processing of electronic
word-of-mouth via online consumer reviews”, Electronic Commerce Research and Applications,
Vol. 7 No. 4, pp. 399-410.
Pavlou, P.A. and Dimoka, A. (2006), “The nature and role of feedback text comments in online
marketplaces: implications for trust building, price premiums, and seller differentiation”,
Information Systems Research, Vol. 17 No. 4, pp. 392-414.
Puccinelli, N.M., Goodstein, R.C., Grewal, D., Price, R., Raghubir, P. and Stewart, D. (2009), “Customer
experience management in retailing: understanding the buying process”, Journal of Retailing,
Vol. 85 No. 1, pp. 15-30.
Purohit, D. and Srivastava, J. (2001), “Effect of manufacturer reputation, retailer reputation, and
product warranty on consumer judgments of product quality: a cue diagnosticity framework”,
Journal of Consumer Psychology, Vol. 10 No. 3, pp. 123-134.
Raju, P.S., Lonial, S.C. and Mangold, W.G. (1995), “Differential effects of subjective knowledge,
objective knowledge, and usage experience on decision making: an exploratory investigation”,
Journal of Consumer Psychology, Vol. 4 No. 2, pp. 153-180.
Rao, A.R. and Monroe, K.B. (1988), “The moderating effect of prior knowledge on cue utilization in
product evaluations”, Journal of Consumer Research, Vol. 15 No. 2, pp. 253-264.
Rao, A.R. and Monroe, K.B. (1996), “Causes and consequences of price premiums”, The Journal of
Business, Vol. 69 No. 4, pp. 511-535.
Resnick, P., Kuwabara, K., Zeckhauser, R. and Friedman, E. (2000), “Reputation systems”,
Communications of the Acm, Vol. 43 No. 12, pp. 45-48.
Resnick, P., Zeckhauser, R., Swanson, J. and Lockwood, K. (2006), “The value of reputation on eBay: a
controlled experiment”, Experimental Economics, Vol. 9 No. 2, pp. 79-101.
Rose, S., Clark, M., Samouel, P. and Hair, N. (2012), “Online customer experience in e-retailing: an
empirical model of antecedents and outcomes”, Journal of Retailing, Vol. 88 No. 2, pp. 308-322.
Roth, A.E. and Ockenfels, A. (2002), “Last-minute bidding and the rules for ending second-price
auctions: evidence from eBay and amazon auctions on the internet”, American Economic Review,
Vol. 92 No. 4, pp. 1093-1103.
Shapiro, C. (1983), “Premiums for high quality products as returns to reputations”, The Quarterly
Journal of Economics, Vol. 98 No. 4, pp. 659-679.
Shirai, M. and Meyer, R. (1997), “Learning and the cognitive algebra of price expectations”, Journal of
Consumer Psychology, Vol. 6 No. 4, pp. 365-388.
Stouthuysen, K., Teunis, I., Reusen, E. and Slabbinck, H. (2018), “Initial trust and intentions to buy: the Role of
effect of vendor-specific guarantees, customer reviews and the role of online shopping
experience$”, Electronic Commerce Research and Applications, Vol. 27, pp. 23-38. customer
Tadelis, S. (1999), “What’s in a name? Reputation as a tradeable asset”, American Economic Review, experience
Vol. 89 No. 3, pp. 548-563.
Tadelis, S. (2016), “Reputation and feedback systems in online platform markets”, Annual Review of
Economics, Vol. 8 No. 1, pp. 321-340.
Trevinal, A.M. and Stenger, T. (2014), “Toward a conceptualization of the online shopping experience”,
601
Journal of Retailing and Consumer Services, Vol. 21 No. 3, pp. 314-326.
Wells, J.D., Valacich, J.S. and Hess, T.J. (2011), “What signal are you sending? How website quality
influences perceptions of product quality and purchase intentions”, MIS Quarterly, pp. 373-396.
Yale, L.J. and Gilly, M.C. (1995), “Dyadic perceptions in personal source information search”, Journal of
Business Research, Vol. 32 No. 3, pp. 225-237.
Ye, Q., Li, H., Wang, Z. and Law, R. (2014), “The influence of hotel price on perceived service quality and
value in e-tourism: an empirical investigation based on online traveler reviews”, Journal of
Hospitality and Tourism Research, Vol. 38 No. 1, pp. 23-39.
Ye, Q., Xu, M., Kiang, M., Wu, W. and Sun, F. (2013), “In-depth analysis of the seller reputation and
price premium relationship: a comparison between eBay US and Taobao China”, Journal of
Electronic Commerce Research, Vol. 14 No. 1, p. 1.
You, W., Liu, L., Xia, M. and Lv, C. (2011), “Reputation inflation detection in a chinese C2C market”,
Electronic Commerce Research and Applications, Vol. 10 No. 5, pp. 510-519.
Zhang, J. (2006), “The roles of players and reputation: evidence from eBay online auctions”, Decision
Support Systems, Vol. 42 No. 3, pp. 1800-1818.
Zhu, F. and Zhang, X. (2010), “Impact of online consumer reviews on sales: the moderating role of
product and consumer characteristics”, Journal of Marketing, Vol. 74 No. 2, pp. 133-148.
Zou, P., Yu, B. and Hao, Y. (2011), “Does the valence of online consumer reviews matter for consumer
decision making? The moderating role of consumer expertise”, Journal of Computers, Vol. 6
No. 3, pp. 484-488.

Corresponding author
Ying Yang can be contacted at: yangyingwill@stu.xjtu.edu.cn

For instructions on how to order reprints of this article, please visit our website:
www.emeraldgrouppublishing.com/licensing/reprints.htm
Or contact us for further details: permissions@emeraldinsight.com