Pricing Strategy in Multi-Channel Retailing and Fairness Perception: an Examination of Boundary Conditions

This article aims to examine the boundary conditions that influence the relationship between pricing strategy in multi-channel retailing and fairness perception, since past research have been founding controversial results concerning this subject. In experiments 1 and 2, we show that differential pricing is perceived as fairer for products in comparison to services. In experiment 3, we show that, when the price difference is justified by an explanation based on costs, it is perceived as fairer in comparison to an explanation based on channel benefits. These studies help to elucidate the controversial relationship between price strategy and perceived fairness, addressing boundary conditions that had not been tested before. We suggest that product retailers should consider differential pricing strategy, since many benefits have been reported in the literature. However, service managers should be careful about using this strategy, because fairness perception influences returning intentions. Additionally, whenever possible, the price difference should be justified by an explanation based on costs.

On the one hand, authors (Montoya-Weiss et al., 2003;Wolk & Ebling, 2010) argue that customers perceive the use of different prices for distinct channels (defined in this study as differential pricing) as fairer than the use of the same price for distinct channels (defined in this study as uniform pricing).On the other hand, other authors (Choi & Mattila, 2009) claim that customers perceive uniform pricing as fairer than differential pricing.
Despite the relation between differential pricing and fairness perception being addressed in the literature, Haws and Bearden (2006) state that issues related to the context remain unexploited.Considering that, the goal of this research is to explore boundary conditions that affect the relationship between pricing strategy (differential vs. uniform pricing) in multi-channel retailing and fairness perception.
Specifically, this study explores two boundary conditions that can affect the relationship between pricing strategy and fairness perception: (1) type of product (product vs. service) and (2) type of explanation for the price difference (based on costs vs. based on channel benefits).Through three experimental studies, we identify in this paper for what type of product customers perceive differential pricing to be fair.In addition, we show what type of explanation for the difference in prices most increases customers' fairness perceptions.

Pricing strategy and fairness perception
The differential pricing is defined by Carroll and Coates (1999) as a strategy in which companies have identical products with different prices, based on factors such as consumer and product localization, for example.Considering multi-channel retailing, differential pricing occurs when companies charge different prices for an identical product or service in different channels, letting the customer decide the best price-channel combination, what is called self-selection (Khan & Jain, 2005;Lii & Sy, 2009;Mussa & Rosen, 1978).While authors (Brynjolfsson & Smith, 2000;Cheng, Chen, & Chang, 2008) show evidence that adopting differential pricing is a viable strategy, many companies refrain from adopting this strategy fearing that consumers will perceive differential pricing as unfair (Wolk & Ebling, 2010).In fact, Flores and Sun (2014) show that, in general, several companies use the same price in online and offline channels.Considering that, fairness perception is an important element when considering which pricing strategy a company should adopt in multi-channel retailing.
Through experiments, Choi and Mattila (2009) demonstrate that uniform pricing was perceived as fairer than differential pricing for a hotel.Results also indicate the existence of consumer bias, since when the differential pricing was favorable to consumers they evaluated differential and uniform pricing offers as equally fair.However, this study may have some limitations: the authors present differential and uniform pricing in the same scenario to participants (within-factor design).
Conversely, Huang et al. (2005) show that uniform pricing in multi-channel retailing was perceived as less fair than differential pricing.Participants had to evaluate price fairness of a hotel booking through fax and through the Internet.They considered that a higher price should be charged in the fax channel than in the internet.In summary, there is a controversial relationship between pricing strategy (differential vs. uniform) and fairness perception.Therefore, we present two boundary conditions that will help to elucidate this controversy: type of product and type of explanation.

Type of product
According to Wolk and Ebling (2010), some product characteristics affect the decision of adopting a differential pricing strategy.Products that are adequate to resale would be more appropriate to a differential pricing strategy because customers are able to buy this product online and resell it on another channel (Wolk & Ebling, 2010).The physical characteristics of the product are also important, since the offline and online channels differ in their ability to provide information about product attributes (Wolk & Ebling, 2010).Huang et al. (2009) compared products with experience and search qualities in offline and online channels.The authors showed that consumers differ in their ability to evaluate the products with experience and search qualities in the offline channel, since it is easier to evaluate products with search qualities before buying them.However, this difference in evaluation disappeared in the online channel.They claim that the online channel allows consumers to evaluate a product with experience qualities in the same way that they evaluate a product with search qualities (Huang et al., 2009).
Connecting these results with the differential pricing literature, Wolk and Ebling (2010) state that when a product presents the same performance in both channels (in this case, the ability to evaluate a product), differential pricing would be less adequate than uniform pricing.On the other hand, when a product performs better in one channel than another (for instance, products with search qualities in the offline channel), differential pricing is more adequate than uniform pricing.
Other authors (Bolton & Alba, 2006;Martín-Ruiz & Rondán-Cataluña, 2008) argue that unfairness perception related to pricing may be higher for products than for services, because it is easier to evaluate the quality of products (most with search qualities) than to evaluate the quality of services (most with experience or credence qualities).Although these authors do not investi-gate fairness perception in the differential pricing context, these studies strengthen the argument that differential pricing would be more adequate for products than for services, because the tangibility and search qualities of products make them perform better in the offline channel than in the online channel.Services, on the other hand, perform in the same way in both channels because they are intangible and have experience qualities, which means, one can only evaluate the quality of services after consumption.
In line with this, Huang et al. (2005) showed that differential pricing was perceived as fairer than uniform pricing for a hotel.Participants said that they attribute this result to costs, which means, they believe that booking in online channel costs less to the provider than booking by fax.The authors argue that other retailers can have higher cost savings than a hotel.An online book retailer saves costs from rent and employees, for example.On the other hand, for a hotel, the cost difference is smaller, because the service will be the same no matter in which channel the booking is made.Thus, differential pricing can be more adequate to products than to services because the cost savings are higher for retailers for this type of offering.
Based on these arguments, we propose that differential pricing is more adequate for products than services because: 1) products can be bought in the online channel and be re-sold, unlike services; 2) products are more tangible and have more search qualities than services, which are more intangible and have more experience and credence qualities than products, making products easier to be evaluated in the offline channel; and 3) the costs savings in online channels are higher for products than for services.Thus, we hypothesize that: H1: Differential pricing for products is perceived as fairer than differential pricing for services.

Type of explanation
Consumers base their judgments on the knowledge or beliefs they have (Cox, 2011).A price above average (compared to the price of past purchases or competitors' price, for example) could be considered unfair, whereas a price below average could signalize low quality (Dodds, Monroe, & Grewal, 1991;Kerin, Jain, & Howard, 1992).Campbell (1999aCampbell ( , 1999b) ) claims that unfairness perception is related to motives inferred by consumers Pricing Strategy in Multi-Channel Retailing and Fairness Perception to a price increase.His studies demonstrate that a price increase is perceived as unfair when customers infer that the company had a negative motive to it.Thus, customers judge a price increase as unfair if it is supposed to boost profit, take advantage, or exploit the market for example.
Conversely, consumers perceive a price increase as fair if the company has a good reason, such as to cover monetary costs, maintain profits, improve employees' welfare, or do a good deed (Campbell, 1999a(Campbell, , 1999b)).Ferguson (2014) states that when companies are transparent and reveal information about price rises during turbulent economic times, it decreases price unfairness perceptions.
Hence, fairness perception can be influenced by the customer's inferred motive to the price difference in a multi-channel retailing context.If the differential pricing is justified by costs, this difference can be perceived as fairer than if it was motivated for other reasons.For instance, a company that uses differential pricing and gives an explanation based on the benefits of buying offline or gives no explanation at all would be perceived as having a more unfair pricing than a company that gives an explanation based on the higher costs of an offline channel.When a company gives an explanation for the differential pricing based on the benefits of a channel, consumers can interpret it as an opportunistic behavior to take advantage of them, making them perceive this price difference as unfair (Frey & Pommerehne, 1993).Consequently, we propose that an explanation for the price difference is a boundary condition of the relationship between pricing strategy and fairness perception, in a way that an explanation based on costs will be perceived as fairer than an explanation based on benefits or the absence of explanation.
H2: An explanation for a price difference based on costs is perceived as fairer than (a) an explanation based on benefits or (b) an absence of explanation.

Study 1
Study 1 investigates which prices consumers perceive as fair for products and services in offline and online channels.The objective of this study is to test H1, which states that differential pricing for products is perceived as fairer than differential pricing for services.

Participants and design
We adopted a single-factor between-subjects design in which we manipulated the type of product (jeans vs. movie ticket).The sample consisted of 89 undergraduate students (38% male; with a mean age of 21.3 years).We removed five cases from the sample (three outliers and two participants who did not correctly answered the manipulation check).Our final sample comprised 84 participants.

Procedure and stimulus
Participants were randomly assigned to a hypothetical scenario presented in a Qualtrics survey, in which they had to read one of the following scenarios: Jeans: Imagine that Duda is considering buying a pair of jeans to wear on everyday life and he decides to search for designs, prices and brands on the Internet.He has no preference for buying the jeans in the physical or online store.Duda decides to buy an Akme pair of jeans that he saw on the website, but he does not purchase it right away.On his way home on the same day, Duda passes in front of Akme store and sees exactly the same pair of jeans.
Movie ticket: Imagine that Duda is considering watching a movie at the theater and he decides to search on movies that are on, schedules, and online ticket prices on the Internet.He has no preference for buying the ticket online or at the box office.Duda decides to watch a particular movie that he saw in the Akme theater website, but he waits to purchase the ticket.On his way home on the same day, Duda passes in front of Akme box office and sees the price of the same movie on the window.
The description of the scenarios was based on the third study of Bolton, Warlop and Alba (2003).In order to choose the products, we conducted an exploratory online survey (N=197), where respondents had to estimate prices in offline and online channels for five products and services.We chose jeans as a proxy to products and movie ticket as a proxy to services because respondents indicated intermediate percentage differences between channels, making the experiment more conservative.In addition to that, we conducted a pretest with undergraduate students (N=77), wherein few changes were made.
The scenario was written in the third person to avoid the consumer bias that may have occur-red in the study of Choi and Mattila (2009).We did not manipulate prices in this study because we wanted participants to write the prices they considered fair in both channels.Regarding the realism of the scenario, customers evaluated it as very real (M=4.23 on a five-point Likert scale).After reading the scenario, participants answered the measures.

Measures
We adapted questions from Bolton et al. (2003) to ask the participants about fair prices to be charged for the jeans or the movie ticket in each of the channels, disregarding the value of freight: "In your opinion, what would be a fair price to be charged for the pants (vs.movie ticket) in the Akme physical store/box office (vs.Akme website)?".As control variables, we used level of familiarity with online shopping and level of online purchasing compared with the general population, all items on a five-point Likert scale.We also asked a question about realism of the scenario.To check the manipulation, respondents were questioned about the type of product/service in multiple-choice questions, with no possibility of returning to check the information.Additionally, we requested demographic questions.

Results
To evaluate whether differential pricing for products is perceived as fairer than for services, we analyzed the percentage differences of fair prices for the two channels in both scenarios.Participants in the jeans scenario considered a difference of approximately 10% higher on average in the physical store to be fair.Participants who were assigned to the movie ticket scenario considered an average difference of only 0.6%, close to a uniform pricing to be fair.
Based on Levene's test, we found that the variances are marginally different (p=0.06,rejecting the null hypothesis of equal variances), so we used the Aspin-Welch independent samples test.We found a significant difference between the jeans condition (M=9.89;SD=21.47) and the movie ticket condition (M=0.59,SD=18.22;Welch-Aspin (75.019)=2.122;p=0.037; with a 95% confidence interval (CI) of 0.57 to 18.03).None of the control variables showed significant effect.
Considering the results, Study 1 shows evidence that customers consider differential pricing for products fairer than for services, supporting H1.However, these results present some limitations concerning the generalization of results.First, we tested only one product (jeans) and one service (movie ticket), therefore, the results are not generalizable across different contexts.Second, the jeans and movie ticket are not totally comparable considering that they have different prices and can generate distinct risk perceptions for customers.Third, we used undergraduate students to test the hypotheses, also limiting the generalization of results.Despite authors defending the participation of students as research subjects in experiments (Falk & Heckman, 2009), the results would gain in external validity by using a more diverse sample of participants.Then, in order to overcome these limitations, we address these issues in Study 2.

Study 2
In order to increase the external validity of Study 1, in this study we chose six products and services to test whether the results would hold.Additionally, in this study we recruited participants from Mechanical Turk in order test H1 with a more diverse sample, thus, increasing the external validity of the results.

Participants and design
This survey adopted a single-factor design in which we manipulated the type of product (product vs. service).In addition to the jeans and movie ticket, we chose a book and a laptop as proxies for products, and a bus ticket and a tour package as proxies for services, in order to increase the external validity of the results found in Study 1.The sample consisted of 255 participants recruited through Mechanical Turk.We removed four outliers and one respondent that incorrectly answered the manipulation check.Then the final sample consisted of 250 participants (67% male; with a mean age of 30.6 years), where 140 participants were randomly assigned to the product condition (jeans: 55; laptop: 41; book: 44) and 110 participants were randomly assigned to the product condition (movies: 40; tour package: 39; bus ticket: 31).
In general, these results reinforce the findings of Study 1 by showing that a price difference for products is perceived as fairer than for services using six different products and services and with a different and more diverse sample.We also analyzed the means of each of the products and services.The patterns were supported for movies ticket (M=0.25,SD=12.23),bus ticket (M=5.04,SD=14.49),jeans (M=13.64,SD=20.62) and book (M=16.25,SD=14.32).However, there is no significant difference between laptop (M=7.03,SD=14.32) and tour package means (M=7.76,SD=12.94).A plausible explanation for this result is that products and services can be allocated in a continuum, where, at one side, a uniform pricing is perceived as fair and, at the other side, differential pricing is perceived as fair.It is possible that the ease of evaluation in online channel is a variable that influences the fairness perception in addition to the type of product.

Study 3
Study 3 tests H2, which states that an explanation for the price difference that is based on costs would increase consumers' fairness perception in comparison to an explanation based on benefits or when no explanation is given.

Participants and design
This study adopted a single-factor between-subjects design with three conditions in a differential pricing context: one that does not present an explanation for the price difference, a second one in which an explanation based on costs is given for the price difference, and a third one in which an explanation based on the benefits of the offline channel is given.The sample consisted of 143 undergraduate students.Considering that H2 was tested only in this study, we decided to recruit a sample of students prioritizing the internal validity of the experiment (cf.Campbell, 1957).After exclusion of 28 participants, who did not correctly answered the manipulation check, the final sample consisted of 115 useful cases (73% female; with a mean age of 26.5 years), where 41 participants were randomly assigned to condition 1 (no explanation), 43 to condition 2 (explanation based on costs) and 31 to condition 3 (explanation based on benefits).

Procedure and stimulus
Data collection was done by invitation via email to undergraduate students.Participants were randomly assigned to a hypothetical scenario presented in Qualtrics software, in which they had to read one of the scenarios, which were adapted from the jeans scenario in Study 1.
Participants read that they saw the price of the jeans in the Akme physical store with a 25% more expensive price than the price in the Akme website.In the no explanation condition, the scenario ended here.In the explanation based on costs condition, Duda (the third person character) asked why there is a price difference and the salesperson explained that it is because of costs, such as rent, salespeople and cleaning.In the explanation based on benefits condition, Duda asked why there is a price difference and the salesperson explained that in the physical store there are the benefits of trying on the jeans, feeling the fabric and seeing the color.The percentage of 25% was chosen to highlight the price difference between channels, so that only the effect of explanation (or its absence) on the fairness perception was considered.Participants rated the scenarios as very real (M=4.28 on a five-point Likert scale).After reading the scenario, participants answered the measures of the study.

Measures
After the stimulus was presented, subjects responded whether Akme follows policies and rules of fair pricing, on a five-point Likert scale (strongly disagree -strongly agree).This item was adapted from Choi and Mattila's (2009) study.As control variables, we measured trust in online retail in general (Sirdeshmukh, Singh, & Sabol, 2002), level of familiarity with online shopping and participants' level of online purchasing compared to the general population; all items on five--point Likert scales.The realism of the situation was also checked on a five-point Likert scale.For the manipulation check, participants had to indicate what explanation was given by the seller ("I did not read about the reasons for this difference in price setting", "the seller said that it is because of the costs", or "the seller said that it is because of the benefits that the store offers").Participants were not able to return to check the information.Finally, demographic questions were recorded.

Results
The average difference between channels of the three groups were compared by one-way ANOVA and revealed a consistent pattern, where the main effect was significant (F(2,112)=3.23;p=0.043; 95% CI of 2.52 to 3.02), as seen in Figure 2. In relation to the means, post hoc tests showed that the explanation based on costs generated the most perceived fairness (M=3.21,SD=1.26), followed by the absence of explanation (M=2.66,SD=1.48) and the explanation based on benefits (M=2.45,SD=1.29).However, the mean difference was significant only between the explanation based on costs and the explanation based on benefits (p=0.02) and the difference between the explanation based on costs and the absence of explanation condition resulted in a marginal significance (p=0.06).Thus, Study 3 shows evidence that the explanation based on costs increases consumers' fairness perception of differential pricing in comparison to an explanation based on benefits (significant) and in comparison to the absence of an explanation (marginally significant), supporting H2.

Conclusions
In Studies 1 e 2, we show that differential pricing for products is perceived as fairer than differential pricing for services, supporting H1.Respondents estimated bigger differences between offline and online channels for products than for services.Specifically, respondents estimated that a fair price difference for products is nearly 10% in the first study and nearly 12.5% in the second study.However, for services, a smaller difference is considered fair: nearly 0.6% in the first study and nearly 4% in the second study.The results of Study 2, however, also show that for the laptop (product) and tour package (service) there was no significant difference.We believe that the products and services can be allocated in a continuum from uniform pricing to differential pricing and ease of evaluation of the offering in an online channel can interfere in the relationship between pricing strategy and fairness perception.
Ease of evaluation can influence the performance of a product or a service in a given channel.For example, one can consider a bus ticket easy to evaluate in the online channel.Consider first two services.A bus ticket presents the same performance when it comes to evaluation in both channels, which means that there are no strong reasons to price it differently in distinct channels.However, when one considers buying a tour package, the customer may be more dependent on the agency's salesperson because it is a more complex purchase, which means that the consumer may be willing to pay more in an offline channel.
The same can be true for products.For instance, a laptop is a product commonly sold in online channels and all information a consumer needs is available there (such as the operational system, and other product characteristics).Consequently, a large price difference between channels can be considered unfair.However, in the case of jeans, the ease of evaluation differs in both channels, since in an online channel there is no chance to try them on, no matter how much information the company presents.According to Wolk and Ebling (2010), the performance of a product or service in channels can influence fairness perception.We suggest for future studies to consider ease of evaluation as another boundary condition of the relationship between pricing strategy in multi-channel retailing and fairness perception.
In Study 3 we tested the hypothesis that when the price difference is accompanied by an explanation based on costs, consumers would infer a good reason for the price difference and would perceive it as fairer than a price difference accompanied by an explanation based on benefits or an absence of explanation.We found that an explanation based on costs was perceived as fairer than an explanation based on benefits, but there was only a marginally significant difference in comparison to the absence of an explanation, according to previous studies about price increase (Campbell, 1999a(Campbell, , 1999b)).We believe that participants may have inferred a reason based on costs in the scenario without explanation, as it in other studies (Huang et al., 2005).We suggest for future studies to control the participants' inferred motives for a price difference by giving a neutral explanation for it.

Theoretical contributions
Since there is no consensus about this subject (Choi & Mattila, 2009;Huang et al., 2005), this article contributes to the controversial literature about pricing strategy in multi-channel retailing.We believe that one of the reasons for these inconsistent results is because authors have failed in examining the boundary conditions of the relationship between pricing strategy and fairness perception.Presenting two variables that influence this relationship -type of product and type of explanation -, our contribution to this literature is twofold.First, we present type of product as a boundary condition for the relationship between pricing strategy and fairness perception.Studies 1 and 2 showed that a price difference is perceived as fairer for products than for services.Second, we showed that an explanation based on costs is capable of increasing the fairness perceptions when there is a price difference, whereas an explanation based on the benefits of the channel reduces fairness perceptions.

Managerial implications
Concerning companies, product retailers should consider using differential pricing for different channels, considering that many benefits have been reported in the literature (Brynjolfsson & Smith, 2000;Cheng et al., 2008;Wolk & Ebling, 2010).However, service managers should be careful in implementing this strategy that could be perceived as unfair to consumers, since fairness perception influences returning and recommendation intentions (Tax et al., 1998).We also suggest that the price difference should be accompanied by an explanation based on channels costs whenever possible.

Figure 1 -
Figure 1 -Percentage difference between offline and online channel