Oligopoly Models

Chapter 9 (skip discussion on

Sweezy Oligopoly)

1
Oligopoly
Definition-
A market structure in which there
are only a few firms each of
which is large relative to the total
industry (results in strategic
interaction)

2
Warning
Due to the complexity involved in
analyzing oligopolies and the
differences across
industries/markets, there is no
single model that is relevant to
all oligopolies.

3
Types of Oligopoly
1. Cournot Oligopoly
2. Stackelberg Oligopoly
3. Bertrand Oligopoly

4
Cournot Oligopoly
1. Few firms in market serving many
customers.
2. Firms produce either differentiated
or homogeneous products.
3. Each firm believes rivals will hold
their output constant if it changes its
output.
4. Barriers to entry exist.

5
Numerical Example of Cournot Oligopoly

Two Firms: Firm 1 and Firm 2

Firms produce a homogenous
product
Market Demand is P=100-Q
Q=Q1+Q2 where Q1 is Firm 1s
output and Q2 is Firm 2s output
Each firm has constant marginal
cost of 20 and zero fixed costs.

6
What if the firms perfectly collude?
What total output should they produce?
100
Q=40. Cant have more
90
profits than what a
80 monopolist would.
70
60
50
D
40
30
20 MC
10
Q
0
0 10 20 30 40 50 60 70 80 90 100

MR
7
Suppose firms collude where both firms
produce an output of 20 (i.e., Q1=Q2=20)
100 Firm 1s Profits = 60*20-20*20=800
90
80
Firm 2s Profits = 60*20-20*20=800
70
60
50
D
40
30
20 MC=AVC=ATC
10
Q
0
0 10 20 30 40 50 60 70 80 90 100

8
Why might you expect that the firms will not be able to
collude in this manner?
If Firm 1 thinks Firm 2 will produce 20, then Firm 1 can
increase his profits to 900 if produce 30.
100
Firm 1s Profits = 50*30-20*30=900
90
80
Firm 2s Profits = 50*20-20*20=600
70
60
50
D
40
30
20 MC=AVC=ATC
10
Q
0
0 10 20 30 40 50 60 70 80 90 100

9
What output would Firm 1 produce if Firm 1
expected Firm 2 to produce an output of 0?
Q1=40
100
Firm 1s Profits = 60*40-20*40=1600
90
80
70 D1
60
50
D
40
30
20 MC=AVC=ATC
10
Q
0
0 10 20 30 40 50 60 70 80 90 100

MR1
10
What output would Firm 1 produce if Firm 1
expected Firm 2 to produce an output of 20?

100
Q1=30
90
80
70
60
50
D
40 D1
30
20 MC=AVC=ATC
10
Q
0
0 10 20 30 40 50 60 70 80 90 100
0
Q2
10 20 30 40 50 60 70 80 Firm 1s
Output
11
MR1
What output would Firm 1 produce if Firm 1
expected Firm 2 to produce an output of 40?

100
Q1=20
90
80
70
60
50
D D1
40
30
20 MC=AVC=ATC
10
Q
0
0 10 20 30 40 50 60 70 80 90 100
0 10 20 30 40 50 60 Firm 1s
Q2 MR1 Output
12
What output would Firm 1 produce if Firm 1
expected Firm 2 to produce an output of 80?

100
Q1=0
90
80
70
60
50
D
40
30
=AVC=ATC
20
D1 MC
10
Q
0
0 10 20 30 40 50 60 70 80 90 100
0 10 20 Firm 1s
Q2 MR1 Output
13
Graphing the Reaction Function (or Best
Response Function) of Firm 1
90 Q2
80 Note: The x-axis depicts the
70 quantity produced by Firm 1
r1(Q2)
60 and the y-axis depicts the
50 quantity produced by Firm 2.
40
30
20
10
0
Q1

100
10

20

30

40

50

60

70

80

90
0

14
Reaction Functions of Firm 1 and Firm 2

90 Q2
80
70
r1(Q2)
60
50
40
30
20 r2(Q1)
10
0
Q1

100
10

20

30

40

50

60

70

80

90
0

15
Cournot Equilibrium
A situation in which neither
firm has an incentive to
change its output given the
other firms output.

16
 Cournot Equilibrium:
Q1=26.67 and Q2=26.67
90 Q2
80
70
r1(Q2)
60
50
40
30
26.67
20 r2(Q1)
10
0
Q1

100
10

20

30

40

50

60

70

80

90
0

26.67
17
Profits from Cournot Equilibrium:
Q1=26.67 and Q2=26.67 so Q=Q1+Q2=53.3
100
90 Firm 1 Profits=46.66*26.67-20*26.67= 713
80
70 Firm 2 Profits=46.66*26.67-20*26.67= 713
60
50
46.6640 D
30
20 MC =AVC=ATC
10
Q
0
0 10 20 30 40 50 60 70 80 90 100

53.33
18
Cournot Equilibrium compared
to Perfect Collusion
Cournot Equilibrium
Q1=26.67 , Firm 1 Profits = 713
Q2=26.67 , Firm 2 Profits = 713

Perfect Collusion
Q1=20 , Firm 1 Profits = 800
Q2=20 , Firm 2 Profits = 800

19
Why cant achieve Perfect Collusion with
Q1=20 and Q2=20? Both Firms have
incentive to cheat!!
90 Q2
80 Note that we are assuming
70
r1(Q2) the firms interact just once.
60
50
40
30
26.67
20 r2(Q1)
10
0
26.67 Q1

100
10

20

30

40

50

60

70

80

90
0

Perfectly Collude at Q1=Q2=20 results

Cournot: Q1=Q2=26.67 and
in profits of 800 for each firm. 20
Profits of both firms are 713.
Industry Characteristics that
Facilitate Collusion
1. Repeated Interaction
Suppose Firm 1 thinks Firm 2 wont deviate from Q2=20 if Firm 1
doesnt deviate from collusive agreement of Q1=20 and Q2=20. In
addition, Firm 1 thinks Firm 2 will produce at an output of 80 in all
future periods if Firm 1 deviates from collusive agreement of Q1=20
and Q2=20.
Firm 1s profits from not cheating
Today
800
In 1 Year
800
In 2 Years
800
In 3 Years
800
In 4 Years
800

Firm 1s profits from cheating (by producing Q1=30 Today)
Today
900
In 1 Year
0
In 2 Years
0
In 3 Years
0
In 4 Years
0

21
 Industry Characteristics that
Facilitate Collusion
2. Stable Industry
3. Few Number of Firms
4. If a firm cheats on a collusive agreement,
the probability the firm is caught is high.
5. Ability to Credibly Punish in a Severe
Manner.
6. Industry demand is growing.
7. Expectation of firms behavior is clear.
22
 My All Time Favorite example of
how expectations are formed
Coca-Cola, PepsiCo Set To Call Off Bitter Soft-Drink Price War
Staff Reporter of The Wall Street Journal

ATLANTA -- A brief but bitter pricing war within the soft-drink

industry might be drawing to a close -- all because no one
wants to be blamed for having fired the first shot.
Coca-Cola Enterprises Inc., Coca-Cola Co.'s biggest bottler, said
in a recent memorandum to executives that it will "attempt to
increase prices" after July 4 amid concern that heavy price
discounting in most of the industry is squeezing profit margins.
The memo is a response to statements made to analysts last
week by top PepsiCo Inc. executives. Pepsi, of Purchase,
N.Y., said "irrational" pricing in much of the soft-drink industry
might temporarily squeeze domestic profits, and it laid the
blame for the price cuts at Coke's door.

23
 My All Time Favorite example of
how expectations are formed
In the June 5 memo, Summerfield K. Johnston Jr. and Henry A.
Schimberg, the chief executive and the president of Coca-Cola
Enterprises, respectively, said the bottler's plan is to "succeed
based on superior marketing programs and execution rather
than the short-term approach of buying share through price
discounting."
"This is a first step to disengagement," said Andrew Conway, an
analyst in New York for Morgan Stanley & Co. "Coke and
Pepsi are out to improve profitability for the category, not
destroy it, so this would bode for a stabilization."
For all the signals of a truce, though, Coca-Cola Enterprises'
memo could just as easily be seen as throwing down the
gauntlet. Messrs. Johnston and Schimberg said in the memo
that should "the competition" view the attempt to raise prices
"as an opportunity to gain share through predatory pricing, we
will, as we have in the past, respond immediately."
24
Stackelberg Oligopoly
1. Few firms in market serving many
customers.
2. Firms produce either differentiated or
homogeneous products.
3. A single firm (the leader) chooses an
output before all other firms choose their
outputs.
4. All other firms (the followers) take as
given the output of the leader and
choose outputs that maximize profits
given the leaders output.
5. Barriers to entry exist. 25
Same numerical example as Cournot but
assume that Firm 1 is the Leader. What do
you expect to happen?
Firm 1 knows the response it will get from
90 Q2 Firm 2 (i.e. what Firm 2 will produce
80 depending on how much Firm 1
produces). Therefore, Firm 1 will select
70
r1(Q2) output to maximize profits given the
60 response function of Firm 2.
50
In this example, Firm 1 will
40
r2(Q1) produce Q1=40 so Firm 2
30 produces Q2=20. Firm 1s
20 profits are 40*40-20*40=800
10 and Firm 2s profits are
0 40*20-20*20=400.
Q1

100
10

20

30

40

50

60

70

80

90
0

FIRST MOVER ADVANTAGE!! 26

Bertrand Oligopoly
1. Few firms in market serving many
customers.
2. Firms produce a homogeneous product
at a constant marginal cost (need not
actually be the case).
3. Firms engage in price competition and
react optimally to prices charged by
competitors.
4. Consumers have perfect information and
there are no transaction costs.
5. Barriers to entry exist.
27
What if Firm 1 and Firm 2 choose price and react
optimally to price charged by other firm?
Will firms be able to collude on a price of $60?
100
Firms could not collude on
90
a price of $60 because
80 each firm would have
70 incentive to undercut other
60 firm. In the end, you would
50
D expect both firms to set a
40 price of $20 (equal to MC)
30
and have zero profits.
20 MC
10
Q
0
0 10 20 30 40 50 60 70 80 90 100

28
Some Conclusions
Level of Competition Depends on Many
Different Characteristics of the Industry not
just Number of Firms.
Level of Competition Depends on whether
the Firms Select Quantity or Price and
whether or not these Decisions are made
Sequentially.
Increased Competition is usually good for
Consumers (lower prices) and bad for
Firms (lower profits)
29