Press "Enter" to skip to content

Analysing price discrimination in the e-commerce industry and its

Analysing price discrimination in the e-commerce industry and its

impact on consumers’ preferences

Haven't found the right essay?
Get an expert to write you the one you need
GET YOUR PAPER NOW

Keshigeyan Chandrasegaran 1002327

02.208 Microeconomics

Singapore University of Technology and Design

A note to Prof. Zsombor

The topic presented in this article, as you

mentioned before, is quite an understudied

topic due to the practical difficulties

associated with collecting concrete

evidence. Though some organizations have

done similar research, they do not publish

raw data or any quantitative insights from

data.

I was able to find only one reliable resource

(From Media and Communication

department, LSE) that has limited

quantitative insights and I have utilized

those results in validating my hypothesis. I

used a lot of material outside class since we

did not discuss much about dynamic pricing

in class. The article is appropriately cited as

well.

Also I did not include random graphs (no

offense to those who do) to increase the

content size and only included graphics that

are very essential. I kept the article size in

terms of word count also to be minimal in

order to convey the context in a concrete

manner. (I also understand the practical

difficulties of reading ~50 articles per course

in 4 days). Happy reading!

Introduction

B2C e-commerce markets give e-commerce

businesses a large spectrum of

opportunities for expressing soft power by

constructing informational power. This is

primarily done through e-personalization, by

understanding consumers’ willingness to

pay, thereby driving them to pay

discriminatory prices for homogenous

products and services. In other words”,

e-commerce markets stand out to be a

fascinating locus for studying variation in

consumers’ preferences to price

discrimination stimuli.

Let us take a step back and try to

understand Internet in the context of

e-commerce. For sellers, e-commerce has

given the power to construct reality in

real-time and on the other side, consumers

are empowered with the ability to interact

with personalized, reconstructed reality in

real-time.

With retail e-commerce sales continuing to

rise and expected to hit USD 4.9 billion by

2021, we could potentially place ourselves

in two extremes in interpreting these figures.

One can interpret sellers’’ power of

constructing reality in the Internet as a

particular expression of soft power over the

consumers leading to exponential increase

in revenues. On the other extreme, one can

interpret these figures as an example of

enhanced consumer power over time.

Though e-commerce is expected to be

crucial in creating competitive and

sustainable environments, according to

Daripa and Kapur (2001)​[2]​, some online

firms can acquire excessive power, which

may lead to lower market concentration.

This is relatable if you carefully observe the

new e-commerce startup scenes. Startups

with great idea and implementation get

acquired by big players such as Amazon

and Alibaba for whooping costs and the

remaining ones simply fail. This results in

increase of market power of particular

e-commerce producers resulting in

decrease of choice in the market for

consumers since only the major sellers

survive. (Do note that from the perspective

of new sellers, getting acquired could be a

great exit point. Since the entry barrier for

e-commerce industries are quite relaxed”,

getting acquired could be a planned exit

strategy for new e-commerce startups.)

This can intrinsically motivate surviving

sellers to deploy monopolistic strategies to

target consumers with different prices

thereby resulting in price discrimination.

Power of constructing information by sellers

through e-personalization is the prime

reason for price discrimination. Some

interesting instances of price discrimination

using informational power are as follows:

1. Orbitz, a Spanish travel agency

targeted Mac users with higher

prices compared to Windows

users.​[1]

2. Online educational platform Udemy

targeted users with IP addresses

tied to educational institute with

costlier prices for popular courses

compared to users with domestic IP

addresses.

The information asymmetry might result in

seller power overtaking consumer power.

Therefore, on the one hand, informational

power combined with soft power, the sellers

could reach a state where they exercise

monopoly that could reduce the economic

advantages enjoyed by the consumers in

the e-commerce space. But on the other

hand, consumers can exercise dominance

over the seller by stopping purchase of

products and services from online platforms

employing discriminatory price strategies .

In other words, this means that ‘perfect’

discriminative state could be distorted by

individuals’ awakening ​[1]​. Therefore the

attitude of consumers is a key factor that

determines the balance between seller and

consumer power.

1. 1 Pigou’s Theory and Degrees of

Discrimination

Pigou (1922) distinguished different degrees

of price discrimination depending on the

amount of information regarding consumers’

preferences that is available to the seller.

In the case of first degree price

discrimination, the seller observes the

actual valuation of each customer, and

provided that individual pricing is feasible”,

he could ask each customer for their

individual reservation price. Though

individual pricing is not something

commonly occuring in reality, this forms the

theoretical ground to a strategy that can

yield competitive outcomes. In other words”,

first degree discrimination is equivalent to

charging different consumers different

prices that are ‘equal to their

demand/reservation price’.​[3]

In the case of second degree price

discrimination, the seller only knows the

distribution of consumers’ valuation rather

than individual valuations.

Other essay:   Technology has always been a key factor in urban development and for many reasons.

Finally, there are instances where the seller

only knows minimal information – just a

signal about consumers’ valuation-. Here”,

sellers will tend to separate consumers into

clusters/ segments based on their expected

willingness to pay. This phenomenon is

referred to as third degree discrimination.

According to Pigou’s theory, price

discrimination implies two main conditions

[3]​.

1. No unit of the commodity sold in one

market can be transferred to another

market.

2. No unit of demand, proper to one

market, can be transferred to

another market.

In the context of e-commerce, we can notice

that the condition of non-transferability

(Condition 1) does not hold because of

markets available for selling ‘used’ goods

(Carousel, ebay, Lazada). This narrows

down our analysis of online price

discrimination to be mostly associated with

the following cases:

1. Non-transferable services such as

rental car booking, hotel

reservations etc

2. General retail on inexpensive goods

such as inexpensive books, food

and beverages, computer and

mobile parts etc which are not

appropriate for reselling on online

auctions.

The research of Erramilli (2013)​[6]​ also

showed that price discrimination is

progressively eliminated as products

become more expensive. ​[1]

1.2 E-Personalization

Electronic personalization is the key feature

that allows sellers to construct informational

power about their customers which makes it

possible to for them to discriminate.

Information on IP addresses (signifying

geometric bounds), search and browsing

history (using cookies), and purchase

history (stored in persistent disks in hosting

databases) are some key metrics that allow

sellers to construct informational power

about its consumers.

I would like to give a hardcore example of

e-personalization efforts. Think about the

recommendation engine building challenge

launched by Netflix in 2007 with a cash

prize of USD 1 million. That is how much

e-personalization mattered to Netflix back in

2007 itself!

Theoretically, e-personalization means that

firms have concrete information on

customers’ willingness to pay and are able

to exercise first-degree discrimination. But

this is not practically feasible due to the

following reasons”,

1. It can explicitly show seller power

>>> consumer power.

2. Computationally not feasible to

implement when the number of

customers become very large.

(Everyone will have different prices

on the platform)

3. Can result in potential black markets

for some specific goods. (Though we

deal with general inexpensive retail

goods for our analysis in this article)

As for second-degree discrimination, it does

not directly cause intrusion and exploitation

into the personal lives since it does not rely

on private characteristics of individual

consumers when fixing prices. A common

example for this is price differences in

airline fares based on the proximity to the

flying dates.

What becomes an interesting strategy for

the sellers’ is using private characteristics of

consumers constructed through

e-personalization, and thereby clustering

them into different groups and exercising

third order pricing discrimination. This

essentially creates an interesting utopian

illusion for consumers which will be

discussed later.

1.3 Price Sensitivity

Varian (2003)​[5]​ states that price sensitivity

is a major reason behind discriminatory

pricing. In other words, it corresponds to an

underlying formulation that individuals in

poorer countries/ geometric bounds do not

have the bandwidth to pay higher prices.

This can be easily observed if you simply

compare the prices of a particular model of

Sony camera in duty-free shops in the

airports of Singapore, Sri Lanka and

Germany (Also note that these prices have

nothing to do with taxes or transportation

cost).

On a general note, according to Sirvanci

(2011)​[7]​, price knowledge and lower

incomes are key indicators of price

sensitivity. Analysing these indicators, they

fall back to the issues of benevolence trust

between sellers and consumers.

Benevolence is the ability of a firm to hold

consumer interests ahead of its own

self-interest and indicates sincere concern

for the welfare of the customers [2].

According to a research done by Garbarino

and Lee (2003)​[9]​, being targeted by both

lower and higher prices leads to a reduction

in the mean benevolence trust. Therefore

price knowledge directly affects the level of

trust on a firm.

So far I have given an overview to how

online sellers have obtained a significant

level of power (seller power) that represents

a type of monopoly/ oligopoly power over its

consumers. But to analyze the balance

between seller power and consumer power”,

we should make efforts to understand and

scrutinize consumer attitudes and strategies

in the event of online price discrimination”,

since consumers have the intrinsic ability to

turn around seller strategies. ​[1]

1.4 Understanding consumers’

preferences in the event of experiencing

online price discrimination

Careful and transparent implementation of

dynamic pricing strategies that do not

strongly influence consumers’ behaviour

has been a topic discussed by many

economists (Kannan & Kopalle, 2001, 79)​[8]​.

Nevertheless, the above strategies are

mostly applicable to second-degree

discrimination where sellers know/ can

approximate the distribution of consumers’

valuations.

But in the case of third-order discrimination”,

to understand the strategies implemented

by firms, we must understand consumers’

perception of price discrimination first. I will

present three key arguments.

My first argument is that for homogenous

products and services sold online , ‘the

accompanying product bundle is

heterogenous’ (Daripa and Kapur , 2001)​[2]​.

The ‘accompanying product bundle’ here

refers to customer-specific

recommendations, personalized search

capabilities and dashboard personalization

capabilities. The idea of ‘heterogeneous

product bundle’ seems to be a reasonable

Other essay:   Global strategic management transcript

rationale and may incline the sellers to

employ third-order discrimination strategies.

My second argument relies on trust. It is

indeed proven my many research papers

that higher consumer trust towards an

online seller will generate more favourable

attitudes towards shopping at that store. In

other words, a prior level of trust should be

regarded as a buffer against the effects of

price discrimination. (Gabarino and

Maxwell, 2010)​[10]​. So my second argument

is that individuals are ready to be targeted

at higher prices by trusted e-commerce

websites and online price discrimination

therefore, not result in diminished trust. ​[1]

This argument directly represents soft

monopoly power of sellers and reflects

consumers’ incapability to buy responsibility

(to express consumer power) because of

market failures (Varian, 2003)​[5]​.

Specifically, consumers lack expert power

because of”,

1. Information asymmetries that may

arise due to sellers’ ability to

exercise monopoly. ​[1]

2. Lack of sanction power where

consumers have zero/ minute ability

to reward or punish brands. ​[1]

3. Lack of legitimate power where

consumers have zero/ minute ability

to influence corporate policies due to

seller power >>> consumer power. ​[1]

Hence, under this situation, sellers will

objectify consumers as passive objects

because of their power dominance.

My third argument relies on the fact that

consumers have strong knowledge both in

terms of product and the underlying

technology. In this case, even minute

changes in personalization strategies or

privacy strategies will have a significant

effect on the purchasing decision of

consumers. Recent research papers show

that price discrimination strategies arouses

strong opposition from the public because it

corresponds to infringement of consumers’

privacy. Beyond, invasion of privacy rights”,

according to Turow (2003)​[4]​, consumers do

not appreciate differential treatment in

comparison with other consumers even if it

is beneficial to the particular concerned

consumer ​[1]​. With electronic privacy laws

getting tight, invasion of privacy rights is

directly correlated to diminishing trust.

Hence firms should be ‘extra’ careful when

designing their pricing strategies in order to

not lose any consumers. With entry barriers

being minimal for establishing new

e-commerce startups as I stated in the

introduction (Shopee, HonestBee as you

can clearly see), maintaining goodwill and

trust of customers is of prominent

importance to e-commerce sellers. Hence

consumers are able to exercise sanction

power on sellers where they can exit using

services of sellers who practice price

discrimination and reward ‘authentic’ sellers.

However in the case of legitimate power”,

both consumers and sellers have equal

power thereby establishing an overall

balanced power state.

Analysing all three arguments, the third

argument matches quite well with reality

and also naturally establishes consumer

power to be on-par with seller power. It also

hints at a state of pseudo-equilibrium

between consumer power and seller power.

In the next section, I will”,

1. Present my key hypothesis.

2. Introduce a similar research done in

this domain and adapt it inside the

article.

3. Share results from the adapted

research

4. Test my hypothesis based on

available research results

5. Formulate open-ended questions

primarily addressing the results.

6. Describe potential problems with the

research method adapted to allow

better design of user-research

studies in the future.

1.5 Hypothesis

Out of my three arguments in the previous

section, I will formulate my hypotheses

relying on my third argument stating that

consumers have strong product and

technical knowledge for the reasons

explained above. Based on this, I’m

formulating my hypothesis below

H​0​ (null hypothesis): Low income

consumers will be more sensitive to higher

prices compared to middle income

consumers who will in-turn be more

sensitive to higher prices compared to high

income consumers.

1.6 Adoption of previous similar

research results into article

To validate/ test my hypotheses, I need to

design user-research surveys to obtain real

data. Due to the tedious and

time-consuming process associated with

conducting user research, I will rely on a

similar research done by Arina Vlasova”,

London School of Economics. I will use her

results to find evidences to support/ reject

my hypotheses.

Though this user-research was done almost

4 years before, this has many overlapping

interests with this article. Also, finding

research results in this domain is very

difficult due to issues related to publishing

user-study data.

1.7 User Research Methodology of Arina

Vlasova (London School of Economics)

The user research was conducted through

an online survey that introduces consumers

to three pricing scenarios in a hotel-booking

environment (consisting of 5 online booking

websites) where they are asked to rate their

preferences based on the scenario.

Special efforts were made by Arina Vlasova

to ensure the novelty and integrity of the

user-research and I will not mention them in

detail. Refer to bibliography section to find

her thesis paper.

Definition of scenarios

1. 1st pricing scenario: Respondents

are aware of price discrimination

2. 2nd pricing scenario: Respondents

are aware that Expedia.com and

Hotels.com set lower prices.

3. 3rd pricing scenario: Respondents

are aware that Travelocity.com and

Booking.com set higher prices.

1.8 User Research Results

Ranking Scenario

1

Scenario

2

Scenario 3

1 Expedia.

com

Expedia.

com

Expedia.co

m

2 Hotels.co

m

Hotels.c

om

Hotels.com

3 Traveloci

ty.com

Traveloci

ty.com

Priceline.co

m

4 Booking.

com

Booking.

com

Travelocity.

com

5 Priceline.

com

Priceline

.com

Booking.co

m

There were 111 respondents. The raw data

is not provided but quantitative findings are

provided below”,

1. Only 8 consumers would stop buying

Other essay:   Student loan forgiveness and repayment for military and veterans

services and products online no

matter whether they are obliged to

pay higher or lower prices.

2. 46 customers would not stop buying

services and products online no

matter whether they are obliged to

pay or lower prices.

3. Participants with high and middle

incomes were more sensitive to

higher prices than low-income

respondents. 66.7% of individuals

with high income, 50.7% of

individuals with middle income and

only 33.3% of low-income

respondents would stop buying

services and products online if they

were obliged to pay higher prices.

4. 53.3% of low-income consumers

and only 33.3% of high income

consumers would not change their

preferences after being price

discriminated.

5. On aggregate, consumers punished

companies that set higher prices as

you can see in scenario 3.

1.9 Hypothesis Testing

It is quite interesting to note that we have

sufficient qualitative evidence to reject my

null hypothesis. According to the survey”,

66.7% of high income consumers exercise

sanction power over sellers compared to

50.7% of middle income consumers and

33.3% of low-income consumers.This

together with my previous analysis brings

up some interesting open ended questions.

1.10 Formulation of open-ended

questions

1. Does higher price create an illusion

of higher quality or authenticity as

you can see that only 33.3% of

low-income consumers exercise

sanction power over sellers in the

case of price discrimination (to pay

higher prices)?

2. Are low-income consumers

characterized by substandard

knowledge of online products and

the underlying technology?

3. Do different income groups

represent clear separated clusters

based on consumer behaviour?

4. Is group pricing based on consumer

‘income signal’ be a good strategy

for third-order price discrimination?

5. Do high and middle income

consumers value privacy more

compared to low-income

consumers?

Though I’m not able to find answers for the

above questions using the survey results”,

these are some interesting questions to

think about when designing user-research

surveys in the future.

1.11 Possible Problems with the above

user-research

Though, Arina Vlasova did not reveal much

information on the survey, I would like

highlight some possible loopholes in the

survey methodology”,

1. The survey may be biased towards a

particular range of age groups.

2. The distribution of high, low and

middle income respondents may not

be uniform.

3. Regional bias in user-research since

the respondents are mostly within a

particular state in the USA

4. Small sample size of 111. This may

not be a good representative of the

population.

Conclusion

The main thesis question for this article was

“How does price discrimination in the

e-commerce space affect the preferences of

consumers?”. I discussed three different

degrees of price discrimination according to

Pigou’s theory and discussed on the

implications of Pigou’s theory on

e-commerce industry.

Section 1.2 dealt with e-personalization

characterizing features used by sellers to

construct informational power about its

consumers. Then I focused on

understanding consumers’ preferences in

event of price discrimination using three key

arguments and formulated a key

hypothesis. Then by adapting a similar

research (done by Arina Vlasova, LSE) and

its results into the article, I found sufficient

evidence to reject my hypothesis on

consumers’ price sensitivity characterized

by their ‘income signal’.

Section 1.10 presents some open-ended

questions that emerged based on my

analysis. These questions will serve as a

good starting point when designing future

user-research experiments for analysing

price discrimination in the e-commerce

industry. Section 1.11 also presents some

possible loopholes associated with the Arina

Vlasova’s research.

Overall, though heavily constrained by

user-research limitations, writing this article

has personally given me much insights into

media economics and B2C online markets.

It also helped me to understand the pseudo

equilibrium that exists between seller power

and consumer power. I also would like to

give special credits to Arina Vlasova for her

research material contribution.

Bibliography

[1] ​Vlasova, A. (2015). Economic power of

e-retailers via price discrimination in

e-commerce: Price discrimination’s impact

on consumers’ choices and preferences and

its position in relation to consumer power.

Retrieved April 14, 2019, from

http://www.lse.ac.uk/media-and-communicat

ions/assets/documents/research/msc-disser

tations/2015/Arina-Vlasova.pdf

[2] Daripa, A., & Kapur, S. (2001). Pricing

on the Internet. Oxford Review of Economic

Policy, 17(2), 202- 216.

[3] Pigou, A. C. (1932). The Economics of

Welfare. Library of Economics and Liberty”,

URL:

http://www.econlib.org/library/NPDBooks/Pi

gou/pgEW28.html [Last consulted April 21″,

2015].

[4] Turow, J. (2003). Americans and Online

Privacy: The System is Broken. Report of

the Annenberg Public Policy Center

[5] Varian, H. R. (2003). Intermediate

microeconomics: A modern approach (6th

ed.). London: Norton.

[6] Mikians, J., Gyarmati, L., Erramilli, V., &

Laoutaris, N. (2013). Crowd- assisted

Search for Price Discrimination in E-

Commerce: First results.

[7] Sirvanci, M. B. (1993). An empirical

study of price thresholds and price

sensitivity. Journal of Applied Business

Research, 9, 43–49.

[8] Kannan, K., & Kopalle, K. (2001).

Dynamic Pricing on the Internet: Importance

and Implications for Consumer Behaviour.

International Journal of Electronic

Commerce, 5, 63-84.

[9] Garbarino, E., & Lee, O. F. (2003).

Dynamic pricing in internet retail: Effects on

consumer trust. Psychology and Marketing”,

20(6), 495-513.

[10] Garbarino, E., & Maxwell, S. (2010).

Consumer response to norm-breaking

pricing events in ecommerce. Journal of

Business Research, 63(9), 1066-1072.

Be First to Comment

Leave a Reply

Your email address will not be published.

0 Shares
Share via
Copy link

Spelling error report

The following text will be sent to our editors: