Eco 403: Industrial Organization Economics, Fall 2012

Eco 403: Industrial Organization Economics, Fall 2012
Dr. Abdel-Hameed H. Nawar

Price Discrimination
Price discrimination means products with identical costs
are sold in different markets at different prices.
A firm price discriminates to increase its profits mainly
by capturing as much of the consumer surplus ().
Example.
- Senior citizen or student discounts.
- Identical products sold with different packaging
We know that:
() : is the maximum a consumer is willing to pay for
each unit of the product.
(): is the minimum a producer is willing to accept
for each unit of the product.
Consumers in one segment (A) of the inverse demand

curve are willing to pay more than the simple
monopoly price, P , but the firm is not asking
king them to
do so. This results in consumer surplus CS.
Consumers in the other segment (B) of the inverse
demand curve are willing to pay more than its cost the
monopolist to produce one additional unit, MC
MC of the
product, but the firm excludes them from. T
This
his results
in DWL.
The potential for capturing these left amounts, and
as additional profit for the firm
firm, requiress creative
thinking about pricing strategies.
Recall, the (non-discriminating) monopoly problem we

studied earlier:
..
max ()
=

)
(!
! -
()
"#$%& '()(*+(
"#$%& .#/$
F.O.C:
0=
() +
() +
..

=
=
23()
()
2
!
4
23()
()
2 3
5
() ()
78(9)9
79 8
() ()
() :1 <= <> ()

8
!
?@ ()
The A from producing and selling more in the market at a

uniform price is the price less the decline due
..
23()
2
.
A() = () :1 <= <> = ( )

8
The goal of price discrimination is to reduce the negative

impact of
23()
2
. In fact, :1 <= <> equals the ratio of

8
marginal cost to price which is identical to different

consumers.
Conditions for price discrimination

To be successful, price discrimination requires:
Market power (simply monopoly)
Identifiability of who to charge the higher price, i.e.
the demand curve in the markets in which the product
is sold must have different price elasticities
Separability of the markets in which the product is
sold, i.e the ability to prevent re-sales
There are several reasons why resale may be difficult:

1. Services
2. Warranties
3. Adulteration
4. Transactions (storage, search, transportation, tax,..)
5. Contractual remedies
6. Vertical integration (upstream firm is a firm that
supplies the inputs in the production process.
Downstream firm is a firm that produces the final
good.
7. Government intervention.
4
In practice, it turns out that the various types of price

discrimination (PD) that can be implemented depend
substantially on the information available to the seller,
namely detailed information about the consumer and about
the volume of consumers purchases.
Three types of price discrimination

1. First-degree price discrimination (FDPD)
A (also known as perfect, ultimate or one-to-one) price
discriminator knows precisely and timely each individual
consumer`s marginal willingness to pay,[1] i.e. the seller
can identify where each consumer lies on the demand
curve.
If each consumer buys only one unit, then as long as the
monopolist can prevent resale, the firm sells to any
consumer who will at least as much as the firm`s marginal
cost, i.e. the perfect price discriminator sells units of the
product and charges each consumer the maximum
willingness to pay until quantity Q units where
..
p = MC(Q )
[1] The market segments are the individuals.
(i.e. the firm continues to sell units of the product until the
marginal consumer whose willingness to pay p is just the
marginal cost MC(Q )).[2]
Production Efficiency: the price on the last purchase
equal . Thus, perfect price discrimination entails
no efficiency loss.
EF is maximized under perfect competition but
eliminated under perfect discrimination. Thus, is
completely appropriated as and this raises debate
regarding income distribution issue.
FDPD can be implemented using a two-part tariff
mechanism (See below). However, a firm usually does not
have enough information to identify each consumer willing
to pay and therefore to perfectly price discriminate.
2. Second degree price discrimination (SDPD)

Quantitybased price discrimination: typically quantity
discounts where the revenue is non-linear in the quantity
purchased. This is referred to as a menu of price plans.
Consumers separate themselves voluntarily into different
groupsi.e. price discrimination is based on SELFSELECTION.
[2] Recall that the supply function of the firm is the MC above the average variable cost and is generated
by profit maximizing behavior while the demand function is MU and is generated by utility maximizing
behavior.
Example: Google Checkout

Checkout.
The transaction processing pricess are charged based on
sales volume. Thus, ddifferent prices are charged by block of
service. Such block
lock pricing requires metering of the service
provided.
Why would consumers willingly select themselves into a

group that includes those who are willing to pay more
rather than those who are willing to pay less for the
good/service!? As we shall see, there are various reasons
why this could be a best strategy for a consumer (and the
seller has to make it so).
3. Third degree price discrimination (TDPD)

A monopolist can determine whether a consumer
belongs to one group or another
another.
Consumer
onsumer types are aggregated into different markets
each with differ
different willingness to pay for the product
and thus different aggregate price elasticity of
demand.
Segmentation can be based on some specific
information about such observable characteristics
characteristic as
age (young/old
(young/old), geographical location (rural/urban),
(rural/urban)
gender (male/female)
(male/female), income (rich/poor), etc. An
observable characteristic is called an index or an
indicator.
Based
ased on the observed characteristics of the consumer
groups and thus on each groups elasticity of demand,
demand
the seller can discriminate by charging different
prices.
Example.
Student discounts; as long as the monopolist can prevent
resale between the two groups, it is profitable to thirddegree price discriminate, i.e. to set different prices for
each group:
maxGHIJ (Q K ) =
*KM5 pK (Q K )Q K
= p(Q5 )Q5 +
TC( *K Q K )
pO (Q O )Q O
+ p* (Q * )Q * TC(Q5 + Q O + + Q * )
Example.
Assume that the monopolist faces two independent
markets, i = 1,2. The firm has zero fixed cost, TC(Q) =
cQ and MC = TC`(Q) = c. The monopolist profits in this
context are given by:
maxUV (X ) = X X (X ) TC(Q K )
= P5 (Q5 )Q5 + PO (Q O )Q O TC(Q5 + Q O )
where X (X ), the inverse demand function for Y, is the

price unit sold that the monopolist must charge to market Y.
Note that K (X ) does not depend on Z , [ Y. Using the
same method as for the non-discriminating monopolist,
1
MR5 = P5 (Q5 ) :1 > = MC
5
9
MR O = PO (Q O ) :1
1
> = MC
O
That is the monopolist sets the marginal revenue from sales

to group Y equals marginal cost. Since marginal cost is
constant, c, the monopolist sets: A5 = AO :
5 (5 ) :1
1
1
(
)
> = O O :1 > = `
_5
_O
Three sources of inefficiency in TDPD:

1. Monopoly inefficiency: > > 0
2. Consumption inefficiency: because different group of
consumer pay different prices, each consumers marginal
willingness to pay is not the same.
A firm has three units to sell.

Maximum Willingness to Pay 1st unit
Group 1
13
Group 2
10
2nd unit
9
8
3rd unit
6
5
If the firm sells 2 units to Group 1 and 1 unit to Group 2.

..
-A = 5 5 + O O = 9 2 + 10 1 = 28
Now, suppose that the firm sells 2 units to Group 2 and 1

unit to Group 1.
..
-A = 5 5 + O O = 13 1 + 8 2 = 29
10
If re-sale is possible, group 2 can sell the second unit to

group 1 at $8.5 and both groups will be better off.
3. Resource inefficiency such as travel cost, waiting cost
and search cost, etc.
11
Two-part pricing (tariff) [3]

Whenever the market can be segmented based on
individuals or groups based on precise and timely
knowledge of the consumers demand curve, the optimal
strategy is to charge an upfront lump-sum fee, -, such as
admission fee, and a per unit price, , such as = . In
this context -deducts partially or entirely from consumers
surplus from buying = g() units of output. Such a price
scheme is called a two-part tariff.
Example: monthly subscription fee + usage fee
A single two-part tariff for identical consumers

Assume that all the consumers have the same willingness
to pay. Then, the consumer surplus (CS) will be entirely
appropriated as producer surplus (PS) as in the perfect price
discrimination. Therefore, total exchange value equals the
total use value to the consumer.
Example: Consider a linear inverse demand function

= 40 0.1, and = 10 with no fixed cost. The nondiscriminating monopolist produces ? =
j6k
Ol
m65
O(.5)
[3] Based on Chapter 5 in Shy Oz (2010) How to Price, Cambridge University Press, pp. 151-178.
12
150 , the price ? = 40 0.1(150) = 25 and ? =

? ? `? = 25(150) 10(150) = 2250.
Let p be where () = . Thus, 40 0.1 = 10, gives

p = 300 . The upfront lump-sum fee - = p = =
q .
5
- = O (r ` )p
5
= O (40 10)(300) = 4500

Thus, using a two-part tariff generates twice as much the
profit of simple monopoly pricing.
s single two-part tariff
Assume that the consumers have t different

willingness to pay.
The firm charges the same p but different lump-sum
fees.
Hence, it is an alternative to perfect price
discrimination.
Unknown individual consumers type
13
Suppose that there are two types of consumers and a firm

that knows each type
types demand curve but does not know
individual consumers type
type.
Consumer
Single uniform two

two-part tariff -. Since -5 u -O all types of
consumers will buy. Note that
Type 1 consumers gain most of the CS from buying the
first few units.
Type 2 consumers gain most of the CS from buying a
large amount.
An applied solution is
is:
-5 u -O and 5 > O
Thus, thee firm designs its pricing structure (menu) subject
to a SELF-SELECT
SELECT constraint: a restriction such that a
14
consumer in any group participates in the market and does

not prefer the other groups two-part tariff schedule.
A) Incentive compatibility: both consumer types reveal

their true type by selecting the right tariff.
5 (-5 , 5 ; 5 ) 5 (-O , O ; O )
O (-O , O ; O ) O (-5 , 5 ; 5 )
B)
Participation: Both consumer types prefer buying to

not buying.
5 (-5 , 5 ; 5 ) 0
O (-O , O ; O ) 0
Example: Assume only two types of customers.

5 = 2.25 0.255 ,
or
O = 10 2O ,
or
Marginal cost ` = 1
15
5 (5 ) = 9 45
O (O ) = 5 0.5O
The following table summarize

summarizes some key results
xy, y = z, {
|z
EFz
}~z
1
2
8
0
2
1.75
6.125
1.75
3
1.50
4.5
3
4
1.25
3.125
3.75
5
1.00
2
4
6
0.75
1.125
3.75
7
0.50
0.5
3
8
0.25
0.125
1.75
|{
EF{
}~{
z , {
with STP
8.0
16
0
8
16
6.0
9.0
6
6.125
20
4.0
4.0
8
4
19
2.0
1.0
6
1
11.75
0.0
0.0
0
0
4
0.0
0.0
0
0
3.75
0.0
0.0
0
0
3
0.0
0.0
0
0
1.75
Single two-part (STP) tariff:

- 8, 1
= 2T + P5 (Q5 Q5 1 PO Q O Q O , TCQ5 1 QO
2 8 1 1 2 1 1 8 , 10
16
- 6.125, 2 a
16
2T 1 P5 Q5 Q5 1 PO Q O Q O , TCQ5 1 Q O

2 6.125 1 2 1.75 1 2 6 , 7.75
{
By offering STP the firm cannot fully extract both

Types consumer surplus.
Interestingly in some cases, as in this example, the firm
may abandon marginal cost pricing in the tariff scheme.
Multiple two-part (MTP) tariff (menu of price plans):

Type 1: -5 1, 5 5
Type 2: -O 16, O 1.
a) Incentive compatibility: both consumer types reveal
their true type by selecting the right tariff.
5 -5 , 5 ; 5 5 5 , -5
5 1,5; 1 2 , 1
1
5 -O , O ; O 5 O , -O
5 16,1; 2 8 , 16
17
,8
Thus,
5 1,5; 1 5 16,1; 2
Also,
O -O , O ; O O O , -O
O 16,1; 8 16 , 16
0
O -5 , 5 ; 5 O 5 , -5
O 1,5; 0 0.0 , 1
,1
Thus,
O 16,1; 2 O 1,5; 1
b) Participation: Both consumer types prefer buying to not
buying.
5 -5 , 5 ; 5 5 1,5; 1 1
O -O , O ; O O 16,1; 8 0
Each consumer type has non-negative surplus and thus
participates.
18
The firms profit,

T
Q5 Q5 1 T
Q O Q O , TCQ5 1 Q O

5 1 P5 !
O 1 PO !
"
"
1

1
5
!
1 1 16

1 1!
8 , 9
"
"
21
19

Eco 403: Industrial Organization Economics, Fall 2012

Uploaded by

Copyright:

Available Formats

Eco 403: Industrial Organization Economics, Fall 2012

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Eco 403: Industrial Organization Economics, Fall 2012

Uploaded by

Copyright:

Available Formats

Eco 403: Industrial Organization Economics, Fall 2012

Dr. Abdel-Hameed H. Nawar

Consumers in one segment (A) of the inverse demand

Recall, the (non-discriminating) monopoly problem we

() :1 <= <>  ()

The A from producing and selling more in the market at a

A() = () :1 <= <> =  ( )

The goal of price discrimination is to reduce the negative

 . In fact, :1 <= <> equals the ratio of

marginal cost to price which is identical to different

Conditions for price discrimination

There are several reasons why resale may be difficult:

In practice, it turns out that the various types of price

Three types of price discrimination

[1] The market segments are the individuals.

2. Second degree price discrimination (SDPD)

Example: Google Checkout

Why would consumers willingly select themselves into a

3. Third degree price discrimination (TDPD)

= P5 (Q5 )Q5 + PO (Q O )Q O TC(Q5 + Q O )

where X (X ), the inverse demand function for Y, is the

That is the monopolist sets the marginal revenue from sales

Three sources of inefficiency in TDPD:

A firm has three units to sell.

If the firm sells 2 units to Group 1 and 1 unit to Group 2.

Now, suppose that the firm sells 2 units to Group 2 and 1

If re-sale is possible, group 2 can sell the second unit to

Two-part pricing (tariff) [3]

Example: monthly subscription fee + usage fee

A single two-part tariff for identical consumers

Example: Consider a linear inverse demand function

150 , the price ? = 40 0.1(150) = 25 and ? =

Let p be where () = . Thus, 40 0.1 = 10, gives

= O (40 10)(300) = 4500

s single two-part tariff

Assume that the consumers have t different

Unknown individual consumers type

Suppose that there are two types of consumers and a firm

Single uniform two

consumer in any group participates in the market and does

A) Incentive compatibility: both consumer types reveal

O (-O , O ; O ) O (-5 , 5 ; 5 )

Participation: Both consumer types prefer buying to

Example: Assume only two types of customers.

The following table summarize

Single two-part (STP) tariff:

 2T 1 P5 Q5 Q5 1 PO Q O Q O , TCQ5 1 Q O 

By offering STP the firm cannot fully extract both

Multiple two-part (MTP) tariff (menu of price plans):

The firms profit,

You might also like

() :1 <= <> ()

The A from producing and selling more in the market at a

A() = () :1 <= <> = ( )

. In fact, :1 <= <> equals the ratio of

where X (X ), the inverse demand function for Y, is the

150 , the price ? = 40 0.1(150) = 25 and ? =

Let p be where () = . Thus, 40 0.1 = 10, gives

O (-O , O ; O ) O (-5 , 5 ; 5 )

2T 1 P5 Q5 Q5 1 PO Q O Q O , TCQ5 1 Q O