Dynamic Pricing and Consumer Fairness

Perceptions

Downloaded from https://academic.oup.com/jcr/article/33/3/304/1891884 by Ludwig-Maximilians-Universitat Munchen user on 07 April 2022

KELLY L. HAWS
WILLIAM O. BEARDEN*

Dynamic pricing practices by sellers in response to segment and individual-level

differences have been made more feasible as internet buyer behavior increases.
While benefits from these pricing practices can accrue to sellers and buyers, the
potential for (un)fairness perceptions to mitigate these advantages is important. In
an effort to investigate these issues, this article reports the results of three studies
that examine the effects of seller-, consumer-, time-, and auction-based price dif-
ferences on perceived price fairness and purchase satisfaction. The findings un-
derscore the potential negative effects associated with price differences from dy-
namic pricing practices.

W ith nearly constant attention on the prices of gas,

pharmaceuticals, and other products, perceived price
fairness is a phenomenon deserving attention from consumer
Kahneman 1974) and fairness heuristic theory (Van den Bos
et al. 1997) are used to develop a general framework re-
garding consumer perceived price fairness judgments. Hy-
researchers. Toward that objective, several researchers have potheses based on this review are then tested in a series of
recently addressed price fairness (e.g., Bolton, Warlop, and three studies. Finally, ideas for future research are proposed.
Alba 2003; Campbell 1999; Grewal, Hardesty, and Iyer
2004; Vaidyanathan and Aggarwal 2003; Xia, Monroe, and
Cox 2004). Questions remain regarding the contexts that THEORETICAL FRAMEWORK AND
affect whether or not consumers perceive prices as fair. An- HYPOTHESES
other issue that is increasingly being investigated is dynamic
pricing. Dynamic pricing, often referred to in economic
terms as individual-level price discrimination, has become
Pricing Concepts and Relevant Heuristics
much more common with the increased prevalence of In-
In keeping with the current conceptualization in the lit-
ternet marketing. It is important to note that little is known
erature (Xia et al. 2004), price fairness refers to a perceived
about the impact of dynamic pricing strategies on consumer
fairness judgment by a buyer of a seller’s prices. Although
perceptions of price fairness.
fairness is a difficult concept to define, the perception of
The present article seeks to address the circumstances that
price fairness is part of a broader judgment of the overall
affect judgments about price fairness in varying dynamic
merits of a deal. Transaction value has been found to have
pricing environments based on consumer, time, and price-
a direct impact on overall acquisition value (Grewal, Mon-
setter differences. First, price fairness and dynamic pricing
roe, and Krishnan 1998), thereby emphasizing the impor-
are discussed within the context of the present research.
tance of perceived fairness. Dual entitlement suggests that,
Next, comparability (Xia and Monroe 2004) and its rela-
in an economic transaction, the buyer is entitled to a fair
tionship to the representativeness heuristic (Tversky and
price and the seller is entitled to a fair profit. However, if
a conflict exists, the seller’s entitlement takes precedence
*Kelly L. Haws is a doctoral candidate in marketing and William O. (Vaidyanathan and Aggarwal 2003).
Bearden is the Bank of America Chair in Marketing at the Moore School The approach taken by Bolton et al. (2003) required study
of Business, University of South Carolina, Columbia, SC 29208 (kelly
_haws@moore.sc.edu and bbearden@moore.sc.edu). The authors acknowl-
participants to consider the various costs involved with pro-
edge the helpful input of the associate editors and reviewers. In addition, ducing and selling products. Across several studies, Bolton
the authors thank Tim Silk and Cait Poynor for their comments on previous et al. (2003) found that profit estimates were both high and
drafts. inflexible and that consumers often underestimate the impact
of inflation. Additionally, consumers did not consider many
Dawn Iacobucci served as editor and Eugene Anderson served as associate
editor for this article. costs, and not all cost increases were seen as equally fair.
Moreover, motives, company reputation, and previous cus-
Electronically published October 9, 2006
tomer satisfaction have all been shown to affect consumers’
304

䉷 2006 by JOURNAL OF CONSUMER RESEARCH, Inc. ● Vol. 33 ● December 2006

All rights reserved. 0093-5301/2006/3303-0002$10.00
DYNAMIC PRICING AND FAIRNESS PERCEPTIONS 305

perceptions of the fairness of price increases (Campbell FIGURE 1

1999; Homburg, Hoyer, and Koschate 2005).
TRANSACTION CHARACTERISTICS AND CONSUMER
Although the concept of dynamic pricing has existed for JUDGMENTS OF PRICE FAIRNESS
some time, it has recently reemerged as a particularly viable
strategy due to advancing technology and the increasing
prevalence of Internet retailing (Jayaraman and Baker 2003;

Downloaded from https://academic.oup.com/jcr/article/33/3/304/1891884 by Ludwig-Maximilians-Universitat Munchen user on 07 April 2022

Kannan and Kopalle 2001). Dynamic pricing is defined here
as a strategy in which prices vary over time, consumers,
and/or circumstances. Elmaghraby and Keskinocak (2003)
distinguish between two dynamic pricing models: price-
posted mechanisms and price-discovery mechanisms. With
price-posted mechanisms, frequent price changes are offered
as “take-it-or-leave-it” prices, that is, the company is still
in charge of setting the price. With price-discovery mech-
anisms, such as eBay, Priceline, or similar negotiated ap-
proaches, consumers have input into setting the final price.
Consistent with the recommendations of Bolton et al.
(2003), the present research investigates the effects of var-
ious dynamic pricing contexts and considers additional
transaction characteristics. The proposed framework de-
picted in figure 1 incorporates and extends those of Bolton
et al. (2003) and Xia et al. (2004) and includes representative
transaction characteristics as antecedents and price-level dif-
ferences as a moderator of price fairness judgments. The
framework also depicts fairness as a predictor of the effects
of transaction characteristics on purchase satisfaction. establish a norm that would be considered in a represen-
The representativeness heuristic and fairness heuristic the- tativeness comparison. In the present research, different
ories are relevant to understanding how consumers make seller is used primarily as a comparison condition, with the
price fairness judgments. Briefly, fairness heuristic theory, focus remaining on consumer, time, and price-setter differ-
based heavily on procedural justice and relational authority ences.
theories, provides a psychological perspective regarding
how and why people form judgments of fairness (Van den Other Consumers. Consumers often have information
Bos et al. 1997). Fairness heuristic theory suggests that about the prices paid by others, especially with the easy
fairness has the capacity to address the fundamental threat accessibility of information-sharing mechanisms both on-
of exploitation that individuals face in a wide variety of line (e.g., chat rooms, blogs, and message boards) and off-
relationships (Lind 2002). Regarding the current research, line (e.g., word of mouth) (Cox 2001). Fairness heuristic
an important aspect of fairness heuristic theory is the attempt research suggests that individuals pay more attention to in-
to examine conditions under which fairness is important formation about others’ outcomes than to other pertinent
(Van den Bos et al. 1997). information (e.g., procedural fairness) when making fairness
judgments (Van den Bos et al. 1997). Moreover, social com-
Representative Transaction Characteristics parisons affect price fairness judgments (Cox 2001). The
concept of equity, which underlies fairness, suggests that
A commonly utilized heuristic, studied in detail by Tver- generally consumers should pay the same price for the same
sky and Kahneman (1974), is representativeness. A con- product (Darke and Dahl 2003). Without explanation, con-
sumer using the representativeness heuristic evaluates the sumers will perceive a price as unfair when it differs from
probability of an uncertain event by its degree of similarity the price paid by other consumers. Even when consumers
to essential properties known to exist in that category (Tver- know that differential prices are based on a buyer’s iden-
sky and Kahneman 1974). Within the current research, a tification (i.e., as a new customer), they regard consumer-
transaction may be deemed fair or unfair based on the com- based price differences as unfair (Grewal et al. 2004). Xia
parability (and therefore representativeness) of essential and Monroe (2004) propose that comparisons with other
transaction characteristics including the product, terms of consumers will have a larger effect on perceived price fair-
sale, time, and seller (Xia et al. 2004). The characteristics ness than comparisons with other sellers or self-references.
considered in the present studies include different consum- The following hypothesis relates to the effects of consumer
ers, times, sellers, and price setters. Previous research has differences:
shown that, although consumers typically expect a particular
retailer to have consistent prices within short time periods, H1: Higher prices paid relative to other consumers will
price promotions are common (Blattberg, Briesch, and Fox trigger stronger negative fairness judgments than
1995; Grewal and Marmorstein 1994), and these price offers seller, time, and/or price-setter differences.
306 JOURNAL OF CONSUMER RESEARCH

Price Setter. The price setter is important because of H3: Consumers will perceive temporally proximal price
the increased use of dynamic pricing and specifically price- differences as more unfair than temporally distant
discovery mechanisms. Price-setter transaction characteris- price differences; however, this effect will dissi-
tics may well affect fairness judgments since attributions of pate over time.
fairness are likely to depend on the level of control the
consumer has regarding price determination (Weiner 1985).
STUDY 1: COMPARING REACTIONS

Downloaded from https://academic.oup.com/jcr/article/33/3/304/1891884 by Ludwig-Maximilians-Universitat Munchen user on 07 April 2022

Fairness heuristic theory also provides support for the ac-
ceptance of “pricing rules” through the premise that once ACROSS PRICING CONTEXTS
consumers have established fairness judgments, those judg-
ments serve as heuristics for evaluating new experiences Study 1 was designed to provide initial tests of hypotheses
(Van den Bos et al. 1997). If consumers judge a particular 1 and 2b. Study 2 directly addresses the issue of price-
pricing rule as fair, subsequent transactions using similar discovery versus price-posted mechanisms and, as such, in-
rules will also be perceived as fair. In particular, consumers vestigates hypothesis 2a. Study 3 was designed to investigate
who participate in price determination through bidding and/ hypothesis 3 relating to temporal price differences and con-
or negotiating are more likely to perceive prices as fair. This sumers’ sensitivity to various time differentials.
assertion is consistent with findings that support attribution-
theory effects associated with locus and controllability (Wei- Method and Dependent Variables
ner 1985). When consumers participate in setting the price,
the onus is more directly on themselves to ensure price A 3 (price level) # 4 (purchase situation) factorial design
acceptability and they are more likely to make internal at- with price and purchase situation manipulated between study
tributions regarding the price determination. Therefore, con- participants was employed. Data were collected from 292
sumers are less likely to perceive a price as unfair, even if undergraduate students as part of an in-class exercise in
evidence exists to the contrary, when they are involved in which individuals participated on a voluntary basis for mod-
the price-setting process. Building on these theoretical ex- est course credit. Cell sizes ranged from 22 to 26. The
planations, hypotheses related to both the overall and the scenarios involved the purchase of a DVD player at a price
relative importance of the price-setter characteristic are pro- of $100, which was based on a series of pretest results. The
posed: three price levels include a comparison price 20% higher,
20% lower, and equal to the price paid by the target con-
H2a: Consumers will perceive prices determined through sumer (i.e., 20%, as recommended by Blattberg et al. 1995).
bidding as more fair than prices set by the re- The four purchase situations reflect contexts associated with
tailer. the factors discussed above (i.e., differences associated with
seller, consumer, time, and price setter) and are included in
H2b: With price-setter differences, higher prices paid the appendix.
relative to others will trigger weaker negative Prior to collecting scaled responses, open-ended thought
fairness judgments than seller, time, and/or con- data were collected in an effort to investigate the processes
sumer differences. preceding fairness and purchase satisfaction judgments and
the extent to which consumers actually have thoughts re-
Time. As time varies, the expectation of prices remain- garding the reasons for price differences and/or the motives
ing constant should be relaxed, although consumers have of sellers in offering varying prices and the effects of these
been shown to insufficiently adjust price perceptions for thoughts on subsequent outcomes. After the thought-elici-
inflation and often look back at previous prices paid in mak- tation task, participants were again provided the purchase-
ing fairness judgments (Bolton et al. 2003). Overall, con- situation description and asked to respond to scaled depen-
sumers are accustomed to high-low strategies that vary re- dent variable measures of perceived price fairness and
tailers’ prices periodically, but they are less accustomed to purchase satisfaction. Fairness and satisfaction have been
prices that vary within very short periods of time (e.g., shown to be related but conceptually distinct constructs (Or-
within the same day). Unlike previous research, the current doñez, Connolly, and Coughlan 2000). Participants were
research investigates time differences that are both tempo- asked to evaluate perceived fairness by responding from 1
rally close and temporally distant. Price discrepancies under (strongly disagree) to 7 (strongly agree) using nine state-
close temporal proximity should be more salient and influ- ments employed by Darke and Dahl (2003) and Xia and
ential than price discrepancies spaced over time. Temporal Monroe (2004). These measures were averaged to form an
construal theory suggests that events that are temporally overall perceived price fairness measure (a p .93). Partic-
proximal are viewed in more concrete terms, while events ipants also evaluated satisfaction with their purchase using
that are temporally distant are viewed in more abstract terms the four 11-point bipolar items used by Darke and Dahl
(Liberman and Trope 1998). As such, price discrepancies (2003). The items were anchored by the following bipolar
under close temporal proximity should be more salient and adjective pairs (a p .96): dissatisfied/satisfied, unhappy/
influential than price discrepancies spaced over time. Con- happy, displeased/pleased, and disappointed/delighted. The
sumers will likely see temporally proximal time-based price correlation between the dependent variable measures was
changes as unfair. Therefore: 0.74. A series of confirmatory factor analyses was employed
DYNAMIC PRICING AND FAIRNESS PERCEPTIONS 307

to investigate the discriminant and construct validity of the sponses in terms of the total number of fairness thoughts as
dependent variables (Gerbing and Anderson 1988). These well as negative thoughts. In both instances, the highest
procedures provided evidence of the discrimination between number of fairness thoughts and negative thoughts occurred
fairness and satisfaction and were used in all three studies for situations in which other consumers paid different prices
to support the reliability and validity of the dependent mea- in contrast to both seller- and time-based price differences.
sures. Pooled across scenario and price conditions, correlations

Downloaded from https://academic.oup.com/jcr/article/33/3/304/1891884 by Ludwig-Maximilians-Universitat Munchen user on 07 April 2022

between the number of positive and negative thoughts with
Results the perceived price fairness measure were 0.44 ( p ! .01)
and ⫺0.50 ( p ! .01), respectively. These results indicate that
Tests of Hypotheses. Examination of 3 # 4 ANOVA cognitions associated with both the dynamic price contexts
tests for each of the dependent variables revealed a consis- described in the scenarios and the price comparison differ-
tent pattern in that both main effects were significant, along ences do precede price fairness judgments as expected (Bol-
with a marginally significant interaction for perceived fair- ton et al. 2003). In addition, modest correlations between
ness (F p 2.01, p p .06) and significant interaction for sat- the number of fairness thoughts (r p ⫺0.13, p ! .05) and
isfaction (F p 2.36, p ! .05). Examination of just the low the number of deal-related thoughts (r p 0.20, p ! .01) and
($80) and same ($100) condition also revealed significant the scaled perceived price fairness variable were observed.
main effects and significant interactions for fairness (F p Overall, the cognitive response data showed that approx-
3.69, p ! .05) and satisfaction (F p 4.01, p ! .01). There- imately 10% of the thoughts recorded by participants were
fore, price level and purchase context interact to affect both related to fairness, representing 23% of the participants over-
fairness and satisfaction. The mean scores are depicted in all. Another interesting finding was that costs were only
table 1. mentioned in two out of the 880 thoughts, suggesting low
To test the effects predicted in hypotheses 1 and 2b, in- spontaneous consideration of firm costs even though costs
dependent mean comparisons were performed across the have been posited to be one of the primary determinants of
varying transaction contexts for the lower comparison price price fairness perceptions.
conditions. In this analysis, the responses for the different
consumer contexts were tested against the responses of the
remaining study participants pooled across the other three STUDY 2: EFFECTS OF PRICE-SETTING
scenarios (i.e., seller, time, and auction). These analyses re- MECHANISM
vealed that the mean scores were significantly lower ( p !
.05), as predicted for perceived price fairness and purchase A 3 (price level) # 2 (pricing mechanism) between-sub-
satisfaction. In tests of hypothesis 2b, the mean responses for jects factorial design was employed with 129 undergraduate
the auction scenario were tested against the responses of the students in a laboratory setting. A different product—boxed
remaining study participants pooled across the seller, con- sets of television program DVDs—was used as the stimulus
sumer, and time-difference conditions. These analyses re- in descriptions similar to those used in study 1. Price levels
vealed that the mean scores for the auction scenario were were identical to those used in study 1. The experimental
significantly higher ( p ! .01) for both perceived price fairness stimuli reflected that the pricing of the product involved
and purchase satisfaction. In summary, then, these results either a price-posted (asked price) or a price-discovery (bid
support the predicted relative ordering of the various trans- price) mechanism. Participants were exposed to both the
action characteristics. price they paid and an alternative price encountered at nearly
the same time. The dependent variable measures were iden-
Cognitive Responses. The cognitive-response data tical to those used in study 1, with coefficient alpha estimates
provided an opportunity to ascertain how often and under of 0.89 and 0.97 for the measures of perceived fairness and
what circumstances consumers spontaneously make judg- purchase satisfaction, respectively.
ments about the fairness of prices. Two judges working Overall, and as shown in the first two columns of table
independently coded the open-ended thoughts. Two thought- 2, the results demonstrate that consumers are more willing
category schemas were employed: (1) thoughts categorized to accept prices that they themselves played a role in setting.
as related to fairness, deal focus, or other, and (2) thoughts For the 2 # 3 ANOVA, main effects were found for both
categorized as positive, negative, or neutral. A total of 880 price (F p 11.61, p ! .01) and pricing mechanism (F p
thoughts were recorded across the 292 study 1 participants. 10.01, p ! .01). The two-way interaction was not significant
The percent agreement between the two judges was 93%, (F p 1.07, p p .35). Focusing on just the lower- and same-
and differences in coding were resolved by subsequent price conditions, both main effects are still significant, and
discussion. the two-way interaction is nonsignificant. Most interesting,
A series of one-way ANOVA tests was used to examine however, was the impact of the two pricing mechanisms at
differences in the thought content across the four dynamic specific price levels. One-way ANOVAs showed that bid
price contexts. Regarding the purchase-situation differences, prices were perceived as more fair than asked prices both
and following overall significant F-values across the two when the customer paid more or the same as the comparison
sets of four means for fairness and satisfaction, significant price: higher ($80), F p 9.74, p ! .01; same ($100), F p
pairwise contrasts ( p ! .05 ) were observed in cognitive re- 4.74, p ! .05. In contrast, when the consumer received a
308 JOURNAL OF CONSUMER RESEARCH

TABLE 1

FAIRNESS AND SATISFACTION MEAN SCORES AND STANDARD DEVIATIONS: STUDY 1

Different seller Different consumer Different time Different auction

Dependent variable and
price condition Mean SD Mean SD Mean SD Mean SD

Downloaded from https://academic.oup.com/jcr/article/33/3/304/1891884 by Ludwig-Maximilians-Universitat Munchen user on 07 April 2022

Perceived price fairness:
Low ($80) 4.04 1.19 3.08 1.10 3.86 1.27 4.64 1.05
Same ($100) 5.33 1.32 5.03 1.03 5.43 1.13 5.08 .84
High ($120) 5.46 1.23 4.56 1.32 4.93 1.16 5.49 .82
Purchase satisfaction:
Low ($80) 4.38 2.56 3.33 2.24 3.55 2.00 5.75 2.17
Same ($100) 9.05 2.24 7.86 2.41 8.07 2.28 8.07 2.19
High ($120) 10.02 1.77 8.23 2.72 8.57 2.70 9.06 1.82

good deal by paying less than the comparison price of $120, F-values for purchase satisfaction were 23.30, 6.50, and
the pricing mechanism no longer affected fairness percep- 11.29. Overall, the pattern of condition mean scores was
tions (F p .32, p p .58), suggesting that a perceived price similar for both perceived fairness and purchase satisfaction.
deal overshadowed the mechanism used to set the price. To test hypothesis 3, comparisons across the three tem-
Satisfaction results showed a similar pattern, with partici- poral difference conditions were made within each price
pants in the $120 condition showing no significant differ- condition. The overall F-values for the $80, $100, and $120
ence in satisfaction based on pricing mechanism in the pres- comparison price conditions involving perceived fairness
ence of a good deal. were 6.50 ( p ! .01), 2.20 ( p p .12), and 0.21 ( p p .81),
respectively. Pairwise contrasts within the $80 price con-
STUDY 3: TEMPORAL SENSITIVITY dition revealed a monotonically increasing pattern of means
as the time difference increased. These results then support
Method hypothesis 3, as well as the premise that proximal prices
have more impact when consumers feel they are at a dis-
A 3 (price level) # 3 (time difference) # 2 (order) mixed advantage (i.e., that others paid less). Specifically, and as
design was employed. Price differences associated with the predicted in hypothesis 3, the effect dissipates as time
on-line purchase of a boxed set of television program DVDs elapses and more distal prices are involved. Like the results
were manipulated between subjects. The three-level time from study 1, if consumers encounter equal (i.e., $100) or
factor (representing prices encountered 1 hr., 1 day, or 1 better (i.e., $120) deals, perceptions of price fairness and
mo. previously) was varied within subjects. Temporal order purchase satisfaction are both higher and, in the case of
was varied (increasing vs. decreasing). No order effects were study 3, not affected by temporal differences.
observed, and subsequent analyses were pooled over the Last, two judges working independently coded the open-
two orders. One hundred fifty-five respondents from an un- ended thought data with respect to time references and at-
dergraduate subject pool participated for extra credit; cell tributions. Across the 104 relevant thoughts, agreement be-
sizes ranged from 49 to 54. The same nine-item fairness tween the two judges was 90%. Differences were resolved
and four-item purchase-satisfaction measures employed in through subsequent discussion. Analyses of these responses
studies 1 and 2 were used again as dependent variables. The revealed a number of noteworthy findings. First, respondents
average reliabilities were 0.90 and 0.79 for perceived fair- were much more likely to be opposed to short-term price
ness and purchase satisfaction, respectively. The average changes when comparison others paid less (x 2 p 11.70,
correlation between the dependent fairness and satisfaction p ! .01). In contrast, participants were more likely to men-
measures was 0.62 ( p ! .01). Finally, participants were tion that prices should drop over time when similar or higher
asked to list any thoughts they had related to the purchase comparison prices were involved (x 2 p 5.71, p p .06).
contexts and then respond to several demographic questions. Moreover, the respondents more readily provided attribu-
tions for the time-based differences when others paid less
Results (x 2 p 10.61, p ! .05). Interestingly, the attributional-related
reasons were equally distributed across individual and com-
The study 3 condition mean scores are presented for both pany-related causes, showing equal likelihood of blaming
dependent variables in the last six columns of table 2. For oneself for poorly timed purchases and blaming the company
both dependent variables, significant F-values ( p ! .01) for using unfair pricing practices.
were observed for both the price- and time-difference main
effects, as well the two 2 # 3 interactions. Briefly, the F- GENERAL DISCUSSION
values from the perceived fairness measure analysis for the
price and time main effects and the price # time interaction The studies presented offer guidance as to which dynamic
were 43.51, 4.88, and 4.81, respectively. The corresponding pricing circumstances are likely to evoke negative fairness
DYNAMIC PRICING AND FAIRNESS PERCEPTIONS 309

TABLE 2

FAIRNESS AND SATISFACTION MEAN SCORES AND STANDARD DEVIATIONS: STUDIES 2 AND 3

Study 2 Study 3
Asked price Bid price 1 hr. 1 day 1 mo.
Dependent variable and

Downloaded from https://academic.oup.com/jcr/article/33/3/304/1891884 by Ludwig-Maximilians-Universitat Munchen user on 07 April 2022

price condition Mean SD Mean SD Mean SD Mean SD Mean SD

Perceived price fairness:

Low ($80) 3.71 .84 4.49 .82 2.68 1.24 3.22 1.13 3.75 1.05
Same ($100) 4.68 1.15 5.40 1.05 4.28 1.52 4.33 1.29 4.07 1.12
High ($120) 4.88 1.02 5.07 1.14 4.72 1.76 4.78 1.08 4.73 1.18
Purchase satisfaction:
Low ($80) 3.48 1.79 4.47 2.73 2.89 3.53 3.32 1.90 4.88 2.11
Same ($100) 7.23 2.10 9.10 1.35 7.09 3.25 7.34 2.93 7.18 4.22
High ($120) 9.04 2.03 9.16 1.71 8.23 2.90 8.60 3.48 8.30 2.42

perceptions. Importantly, and compared with the other trans- and price setter, were further examined. The temporal thresh-
action characteristics included in this research (i.e., seller, old for dynamic pricing changes over time was addressed,
time, and price setter), the results of study 1 revealed through and reactions to bid versus asked prices were explicitly re-
scaled responses and thought data that differences between vealed. Caveats are warranted regarding limitations of this
consumers resulted in the greatest perceptions of unfairness research. The studies were all lab experiments involving
(in keeping with fairness heuristic theory) and the lowest student participants and purchase scenarios. Simulations in-
overall satisfaction. Study 1 also demonstrated that out- volving purchase scenarios over time in which consumers
comes associated with auctions were the least likely to be are able to learn about the dynamic pricing scenarios would
perceived as unfair. Evidence from the thought data suggests provide additional tests of the relationships found. Certainly
that consumers attribute their outcomes within auction con- the effects of dynamic pricing may be even more pro-
texts to themselves (83% vs. 17% of attributions) rather than nounced when varying prices are encountered in realistic
the seller (Weiner 1985). In addition, the cognitive-response shopping environments and under conditions of higher in-
data demonstrated the important occurrence of spontaneous volvement. Consumers may be exposed to a variety of con-
judgments about the fairness of prices, but minimal spon- texts characterized by dynamic prices. An additional study
taneous consideration of cost issues, as explanations for not reported here replicated the pattern of results found in
price differences. study 1 using a within-subjects design for the four pricing
Study 2 demonstrated that consumers have higher fairness contexts. Specifically, the consumer-based and auction-
perceptions and satisfaction across all price-level conditions based differences were perceived as the most and least un-
when they play a role in the price-setting process (i.e., price fair, respectively.
discovery) than when the prices are set by the retailer (i.e., Although the present results provide evidence regarding
price posted). However, more detailed analysis revealed that which transaction characteristics are most important to con-
this effect is diminished when the consumer receives a good sumers in making fairness judgments, other questions re-
deal. main. The thought data collected in studies 1 and 3 provided
The specific focus on timing differences in study 3 dem- evidence of various types of attributions that consumers
onstrated that consumers view price changes within very might make, but further research is needed to fully under-
short time periods as more unfair than changes over a more stand the impact of these attributions (Bolton et al. 2003).
extended time period, especially when exposed to lower In addition, the effects of uncertainty remain unresolved.
prices. After a month delay, pricing-level differences no For example, uncertainty regarding what comparison others
longer affected fairness perceptions. Thought data confirmed have paid may interact with the credibility of the source
that significantly more thoughts about the unfairness of (e.g., retailer price tier, closeness of relationships with com-
short-term price changes were expressed by those who were parison consumers) in affecting fairness judgments. Fur-
exposed to a lower price. Evidence that consumers’ per- thermore, consumer knowledge about pricing in general
ceptions of fairness can be affected both by encountering a should be examined as a potential factor explaining fairness
more favorable or a less favorable outcome was found. perceptions.
Overall, the contributions of the research were the con- Future studies can further test the transaction character-
sideration of multiple dynamic price contexts and the in- istics addressed in the present research, as well as explore
vestigation of those effects on perceived price fairness and additional contextual factors (Bolton et al. 2003; Xia and
overall purchase satisfaction. These effects were investi- Monroe 2004) and various combinations of these charac-
gated relative to one another, revealing resistance to con- teristics that are used in the marketplace. It is likely that
sumer-based price differences and the acceptance of bidding- these characteristics interact to affect consumers’ overall
based price differences. In addition, two characteristics, time evaluations of fairness and satisfaction. Future research us-
310 JOURNAL OF CONSUMER RESEARCH

ing purchase simulations would enable tests of conditions REFERENCES

when consumers encounter discrepant price information on
their own and resulting effects on fairness and satisfaction. Blattberg, Robert C., Richard Briesch, and Edward J. Fox (1995),
In such situations, consumers would have to create their “How Promotions Work,” Marketing Science, 14 (3), G122–
own representative comparison. As dynamic pricing be- G132.
comes more widespread in use, it is critical that marketers Bolton, Lisa E., Luk Warlop, and Joseph Alba (2003), “Consumer

Downloaded from https://academic.oup.com/jcr/article/33/3/304/1891884 by Ludwig-Maximilians-Universitat Munchen user on 07 April 2022

understand how consumers respond to various dynamic pric- Perceptions of Price (Un)Fairness,” Journal of Consumer Re-
ing mechanisms. search, 29 (March), 474–91.
Campbell, Margaret C. (1999), “Perception of Price Unfairness:
Antecedents and Consequences of Satisfaction,” Journal of
APPENDIX Marketing Research, 36 (May), 187–99.
Cox, Jennifer L. (2001), “Can Differential Prices Be Fair?” Journal
of Product and Brand Management, 50 (5), 264–75.
STUDY 1 PRICE CONTEXT STIMULI Darke, Peter R. and Darren W. Dahl (2003), “Fairness and Dis-
counts: The Subjective Value of a Bargain,” Journal of Con-
You have been wanting a new DVD player and have sumer Psychology, 13 (3), 328–38.
decided exactly what brand and model you will buy. You Elmaghraby, Wedad and Pinar Keskinocak (2003), “Dynamic Pric-
go on-line and use your own money to purchase the DVD ing in the Presence of Inventory Considerations: Research
Overview, Current Practices, and Future Directions,” Man-
player from a well-known on-line retailer:
agement Science, 49 (10), 1287–1305.
Gerbing, David W. and James C. Anderson (1988), “An Updated
1. for a total price of $100. Later the same day, you see Paradigm for Scale Development Incorporating Unidimen-
the exact same DVD player at another well-known on- sionality and Its Assessment,” Journal of Marketing Research,
line retailer for a total price of [$80/$100/$120]. (Dif- 25 (May), 186–92.
ferent seller) Grewal, Dhruv, David Hardesty, and Gopalkrishnan Iyer (2004),
2. for a total price of $100. The next day you overhear “The Effects of Buyer Identification and Purchase Timing on
a conversation at school about another student pur- Consumers’ Perceptions of Trust, Price Fairness, and Repur-
chasing the exact same DVD player as you from the chase Intentions,” Journal of Interactive Marketing, 18 (4),
same on-line retailer the day before for a total price 87–100.
of [$80/$100/$120]. (Different customer) Grewal, Dhruv and Howard Mamorstein (1994), “Market Price
Variation, Perceived Price Variation, and Consumers’ Price
3. for a total price of $100. The next day, you go back
Search Decisions for Durable Goods,” Journal of Consumer
to the same on-line retailer to look for another product Research, 21 (December), 453–60.
and you see the DVD player you purchased the day Grewal, Dhruv, Kent B. Monroe, and R. Krishnan (1998), “The
before for a total price of [$80/$100/$120]. (Different Effects of Price-Comparison Advertising on Buyers’ Percep-
time) tions of Acquisition Value, Transaction Value, and Behavioral
4. by placing the winning bid for $100. You notice later Intentions,” Journal of Marketing, 62 (April), 46–59.
that another auction that ended around the same time Homburg, Christian, Wayne D. Hoyer, and Nicole Koschate (2005),
as yours had a winning bid of [$80/$100/$120] for the “Customers’ Reactions to Price Increases: Do Customer Sat-
same DVD player. (Different auction) isfaction and Perceived Motive Fairness Matter?” Journal of
the Academy of Marketing Science, 33 (Winter), 36–49.
Jayaraman, Vaidyanathan and Tim Baker (2003), “The Internet as
an Enabler for Dynamic Pricing of Goods,” IEEE Transac-
tions and Engineering Management, 50 (4), 470–77.
STUDY 2 PRICE CONTEXT STIMULI Kannan, P. K. and Praveen K. Kopalle (2001), “Dynamic Pricing
on the Internet: Importance and Implications for Consumer
Boxed DVD sets containing entire seasons of various Behavior,” International Journal of Electronic Commerce, 5
television shows typically sell for somewhere between about (3), 63–83.
$60 and $160 for two seasons. You have been wanting to Liberman, Nira and Yaacov Trope (1998), “The Role of Feasi-
purchase a boxed set of DVDs containing the first two sea- bility and Desirability Considerations in Near and Distant
sons of your favorite television show. You go on-line and Future Decisions: A Test of Temporal Construal Theory,”
use your own money to purchase the boxed DVD set from Journal of Personality and Social Psychology, 75 (July),
a well-known on-line retailer: 5–18.
Lind, E. Allan (2002), “Fairness Judgments as Cognitions,” in
1. paying the asked price of $100. You notice later that The Justice Motive in Everyday Life, ed. Michael Ross and
day that the asked price at the retailer’s Web site is Dale T. Miller, Cambridge: Cambridge University Press,
417–31.
[$80/$100/$120] for the same boxed DVD set. (Asked Ordoñez, Lisa D., Terry Connolly, and Richard Coughlan (2000),
price) “Multiple Reference Points in Satisfaction and Fairness As-
2. by placing the winning bid for $100. You notice later sessment,” Journal of Behavioral Decision Making, 13 (3),
that another auction that ended around the same time 329–44.
as yours had a winning bid of [$80/$100/$120] for the Tversky, Amos and Daniel Kahneman (1974), “Judgments under
same boxed DVD set. (Bid price) Uncertainty: Heuristics and Biases,” Science, 185, 1124–31.
DYNAMIC PRICING AND FAIRNESS PERCEPTIONS 311

Vaidyanathan, Rajiv and Praveen Aggarwal (2003), “Who Is the Weiner, Bernard (1985), “An Attributional Theory of Achievement
Fairest of Them All? An Attributional Approach to Price Motivation and Emotion,” Psychological Review, 92, 548–73.
Fairness Perceptions,” Journal of Business Research, 56 (6), Xia, Lan, and Kent B. Monroe (2004), “Comparative References,
453–63. Trust, and Perceived Price Fairness,” working paper, Depart-
Van den Bos, Kees, E. Allan Lind, Riël Vermunt, and Henk A. ment of Business Administration, University of Illinois at
M. Wilke (1997), “How Do I Judge My Outcome When I Do Urbana-Champaign.
Not Know the Outcome of Others? The Psychology of the Xia, Lan, Kent B. Monroe, and Jennifer L. Cox (2004), “The Price

Downloaded from https://academic.oup.com/jcr/article/33/3/304/1891884 by Ludwig-Maximilians-Universitat Munchen user on 07 April 2022

Fair Process Effect,” Journal of Personality and Social Psy- Is Unfair! A Conceptual Framework of Price Fairness Per-
chology, 72 (5), 1034–46. ceptions,” Journal of Marketing, 68 (October), 1–15.