ECON 312

Microeconomics II

OLIGOPOLY

Lecturer: Dr. Kwame Agyire-Tettey, Department of Economics

Contact Information: [email protected]
 Lecture’s objective

The objective of this lecture is to introduce

Students to Strategic Interactions under the theory
of Oligopoly. Specifically the lecture will examine
the following:
Ø Introduction and concepts of oligopoly models
Ø Quantity adjusting models: Cournot Model
Ø Reaction functions/curves
Ø Price adjusting models: Bertrand Model
 Reading List

l Perloff, J. M. (2023). Microeconomics, 9th edition. Pearson

Education Canada. [Page 448-489]
l Varian, H. R. (2016). Intermediate Microeconomics with
Calculus: A Modern Approach: Ninth Edition International
Student Edition. WW Norton & Company.
l Perloff, J. (2013). Microeconomics with Calculus, Global
Edition. Pearson Education UK. [Page 514-562]
l Varian, H. R. (2014). Intermediate microeconomics: a
modern approach: ninth international student edition. WW
Norton & Company. [Page 516-539]
l Nicholson, W., & Snyder, C.M. (2014). Intermediate
microeconomics and its application. Cengage Learning.
[Page 374-410]
©2005 Pearson Education, Inc. 3
 Oligopoly – Characteristics

l The term ‘oligopoly’ is derived from two Greek words:

oligi (meaning ‘few’) and polein (meaning ‘sellers’)

l An oligopoly is defined as a market structure with a few

firms and barriers to entry. There is often high level of
competition between firms, as each firm makes
decisions on prices, quantities, and advertising to
maximise profits.

l Automobile industries, soft drinks companies like Coca-

Cola and Pepsi, and oil companies.
©2005 Pearson Education, Inc. 4
 Oligopoly – Characteristics
l In an oligopoly, a market structure of immense significance,
we observe a scenario where a mere handful of firms hold a
substantial market share, exerting a significant influence on
the market dynamics.

l What distinguishes it from other market structures is the

existence of entry barriers, which can be high initial
investment costs or complex regulatory requirements.

l Due to these barriers, competition among firms is intense,

and each firm strategises on pricing, production quantity, and
advertising to maximise its profits.
©2005 Pearson Education, Inc. 5
 Oligopoly – Characteristics

mFew firms

mIntensive competition

mInterdependence

mEither homogeneous or differentiated

products

mDifficult entry
©2005 Pearson Education, Inc. 6
Few firms and intensive competition
l Fewer nature of firms operating in the market
creates intensive competition, with each firm
desiring to control a larger share of the market.
l This leads to massive advertising expenditure,
sales and promotion etc.
l Due the fewer numbers and competitive nature,
firms keep a close watch on activities of rival
firms.
l Firms have a set of aggressive and defensive
market strategies in response to rival’s actions.
©2005 Pearson Education, Inc. 7
 Characteristics- Mutually interdependent

l A key characteristic of oligopolies is that a firm’s

action(s) potentially affect market outcomes, thus
each firm’s choices are dependent on the choices
of other firm(s). They are interdependent.

l Hence, the decision a firm takes into account the

expected reaction of rival(s)

l Oligopoly firms are thus mutually interdependent

l Oligopolies can be collusive or non-collusive

©2005 Pearson Education, Inc. 8
 Characteristics - interdependency

l The importance of interdependence is that it leads

to strategic behaviour.

l Strategic behaviour occurs when “what is best

for A depends on action(s) of B, and what is
best for B is also dependent on action(s) of
A”.

l Oligopolistic behaviour includes both ruthless

competition and cooperation and leads to a
great deal of uncertainty in the behaviour of firms
©2005 Pearson Education, Inc. 9
 Oligopoly – Characteristics

Barriers to entry
m Scale economies

m Patents

m Technology

m Brand name recognition/ loyalty

m Strategic action by firms

©2005 Pearson Education, Inc. 10

 Oligopoly
lExamples
m Airline Industry m Cellular Phone Services
m Textile industry m Computer Operating Sys.
m Beer industry m Aluminium and Steel
m Pharmaceuticals m Oil and Gas industry
m Computer & Software m Auto Industry
Industry

©2005 Pearson Education, Inc.

11
 Oligopoly – Equilibrium
l The monopolist is not concerned about the
reaction of its rival to actions it pursues mainly
because it has no rivals
l A competitive firm potentially faces many rivals,
but the firm and its rivals are price takers à also
no need to worry about rivals’ actions
l In this market, the firm in taking a strategic action
– reducing market price, increasing of output,
etc., – must first consider the reaction of rival(s)

©2005 Pearson Education, Inc. 12

 Oligopoly – Equilibrium
l The oligopolist needs to choose an appropriate
response to the rivals’ actions. Similarly, rivals must
anticipate the firm’s response and act accordingly,
depending on the interactive setting and Behaviour
patterns.
l Actions and reactions are dynamic, evolving over time.
l Game Theory, with its assumption that firms are
rational decision makers, is a powerful tool to analyse
strategic actions in such an interactive setting.

©2005 Pearson Education, Inc. 13

 Oligopoly – Equilibrium
l Defining Equilibrium
mFirms are doing the best they can and have no incentive
to change their strategy (e.g. output or price)
mAll firms assume competitors are taking rival decisions
into account.
l Nash Equilibrium à John Nash
mEach firm is doing the best it can given what its
competitors are doing à no incentive to deviate.
l For this class, we will focus on duopoly models à
Markets in which two firms compete
©2005 Pearson Education, Inc. 14
 Duopoly Models

• Three models:
– Cournot model
– Stackelberg model
– Bertrand model
Oligopoly – Cournot model

l Underlying assumptions
m All firms are identical – they have the same cost
functions and produce homogenous good
m Firm will adjust its output based on what it thinks the other
firm will produce

m Constant marginal cost, actually Cournot’s model

assumes zero marginal cost
m Each firm acts on the assumption that its rival will not
change its output and decides its profit maximizing output
level
©2005 Pearson Education, Inc. 16
 Cournot Model
l Assume firm A starts production and sales of
mineral water.
l It produces a quantity “b” and charges a price of Pa
where profit is maximized.
l Because at this price and quantity MR =MC =0.
Note also that at this profit maximizing point price
elasticity is equal to one and TR is maximized, and
with zero costs maximum TR implies maximum
profits.
l Firm B reacts to A’s output & enters the market on
an assumption that A’s output will be unchanged
©2005 Pearson Education, Inc. 17
 Cournot Model
l B produces half of what firm A produces, that is ¼
of the market demand, in order to maximize its
own profit.
l Now firm A will no long maximize its profit as the
market output has change. Firm A reenters the
market and produces half of the market which is
not supplied by firm B assuming that B will retain
its output.
l A supplies ½ of ¾ = 3/8 of the total market.
l B will also react under the same assumption.
©2005 Pearson Education, Inc. 18
 Cournot Model
l Firms in this model are said to be naïve and they
do not learn from their previous experience.

l This action and reaction pattern continues, since

the naïve behaviour of not learning the past
reaction pattern of rival exists.

l Equilibrium is reached when each firm in the

market produces 1/3 of the market demand.
©2005 Pearson Education, Inc. 19
©2005 Pearson Education, Inc. 20
 Oligopoly – Cournot model

©2005 Pearson Education, Inc. 21

 Oligopoly
l The Reaction Curve
mThe relationship between a firm’s profit-maximizing
output and the amount it thinks its competitor will
produce.
mA firm’s profit-maximizing output is a decreasing
schedule of the expected output of Firm 2.

©2005 Pearson Education, Inc. 22

 Isoprofit curves

©2005 Pearson Education, Inc. 23

 Isoprofit
l An isoprofit curve for firm A is the locus of points
defined by different levels of output of A and its
rival B, which yield to A the same level of profit
mIsoprofit curves for substitute commodities are
concave to the axes along which we measure the
output of the rival firms. For example, an isoprofit
curve of firm A is concave to the horizontal axis QA.
This shape shows how A can react to B’s output
decisions so as to retain a given level of profit. For
example, consider the isoprofit curve ΠA1 in

©2005 Pearson Education, Inc. 24

 Isoprofit

l Similarly, an isoprofit curve for firm B is

the locus of points of different levels of
output of the two competitors which yield
to B the same level of profit

©2005 Pearson Education, Inc. 25

 Reaction Curves and Cournot
Equilibrium – constant MC

©2005 Pearson Education, Inc. 26

 Reaction Curves and Cournot
Equilibrium
QA

100 x Firm B’s reaction curve shows how much it

will produce as a function of how much
it thinks Firm A will produce.

Firm B’s Reaction

75 Curve Q*B(QA)

x
Firm A’s reaction curve shows how much it
50 will produce as a function of how much
it thinks Firm B will produce. The x’s
correspond to the previous model.
Firm A’s Reaction
Curve Q*A(QB)
25

x x
25 50 75 100 QB
©2005 Pearson Education, Inc. 27
 Reaction Curves and Cournot
Equilibrium
Q1

100 x Firm B’s Reaction In Cournot equilibrium, each firm

Curve Q*B(QA) correctly assumes how much its
competitors will produce and
thereby maximize its own profits.

75

Cournot
x Equilibrium
50

Firm A’s Reaction

25 Curve Q*A(QB)

x x
25 50 75 100 Q2
©2005 Pearson Education, Inc. 28
 Cournot Equilibrium

l Each firm's reaction curve indicates the

quantity the firm should produce given
the output of its competitor.

l Equilibrium in the Cournot model, in

which each firm correctly assumes how
much its competitor will produce and sets
its own production level accordingly.

©2005 Pearson Education, Inc. 29

 Cournot Equilibrium

l Cournot equilibrium is an example of a

Nash equilibrium (Cournot-Nash
Equilibrium)

l Mathematical derivation

©2005 Pearson Education, Inc. 30

©2005 Pearson Education, Inc. 31
Cournot Model Mathematical Analysis
l The linear Cournot model
m Model
l Homogeneous product market with n firms
l Firm i sets quantity qi
l Total output: q = q1 + q2 + ... + qn
l Market price given by P(q) = a - bq
l Linear cost functions: Ci(qi) = ci qi
l Notation: q-i = q - qi

mResidual demand
P(qi ,q-i ) = (a - bq-i ) - bqi
º di (q-i )
©2005 Pearson Education, Inc.
Cournot Model Mathematical Analysis
lThe linear Cournot model (cont’d)
mFirm’s problem
l Cournot conjecture: rivals don’t modify their quantity
l Firm i acts as a monopolist
on its residual demand: max qi di (q-i )qi - ci qi
l FOC: a - ci - 2bqi - bq-i = 0
l Best-response function: qi (q-i ) = 2b (a - ci - bq-i )
1

mNash equilibrium in the duopoly case

l Assume: c1 £ c2 and c2 £ (a + c1 ) / 2
l Then, q1* = 1
(a - 2c1 + c2 ) and q2* = 1
(a - 2c2 + c1 )
3b 3b

q1* ³ q2* Þ p 1* ³ p 2*
©2005 Pearson Education, Inc.
Cournot Model Mathematical Analysis
l The linear Cournot model (cont’d)
mDuopoly

• Lesson: In the linear Cournot model with homogeneous

products, a firm’s equilibrium profits increases when
the firm
©2005 Pearson becomes
Education, Inc. relatively more efficient than its rivals.
Cournot Model Mathematical Analysis

l Symmetric Cournot oligopoly

mAssume ci = c for all i = 1n
mThen
a - c p *
(n) - c a - c
q (n) =
*
® L(n) = =
b(n + 1) *
p (n) a + nc

l If n­ ® individual quantity ¯, total quantity ­, market

price ¯, mark-up ¯
l If n® µ, then mark-up ® 0

• Lesson: The (symmetric linear) Cournot model

converges to perfect competition as the
number
©2005 of firms
Pearson Education, Inc. increases.
Example: of Cournot Model Curve

l An Example of the Cournot Equilibrium

mTwo firms face linear market demand curve
mWe can compare competitive equilibrium and the
equilibrium resulting from collusion
mMarket demand is P = 30 - Q
mQ is total production of both firms:
Q = Q1 + Q2
mBoth firms have MC1 = MC2 = 0

©2005 Pearson Education, Inc. 36

 Oligopoly Example

l Firm 1’s Reaction Curve à derived from

firm 1’ profit max problem à MR=MC

P1 = TR1 = PQ1 = (30 - Q )Q1

= 30Q1 - (Q1 + Q2 )Q1
= 30Q1 - Q12 - Q2Q1

©2005 Pearson Education, Inc. 37

 Oligopoly Example

l An Example of the Cournot Equilibrium

¶TR1
MR1 = = 30 - 2Q1 - Q2 and MC1 = 0
¶Q1
MR1 = 0 = MC1 Þ 30 - 2Q1 - Q2 = 0
Firm 1's Reaction Curve
Q1 = 15 - 1 2 Q2

Similarly, firm 2's Reaction Curve can be derived

Q2 = 15 - 1 2 Q1
©2005 Pearson Education, Inc. 38
 Oligopoly Example
l An Example of the Cournot Equilibrium
Cournot Equilibrium: Q1 = Q2
1æ 1 ö 1
15 - ç15 - Q1 ÷ = 10 ® Q1 = 2.5 ® Q1 = 10
2è 2 ø 4
1
Q2 = 15 - Q1 = 10 Þ Q = Q1 + Q2 = 20 Þ P = 30 - Q = 10
2

If fixed costs for both firms = 0

P1 = P 2 = 30Q1 - ( Q1 ) - Q1Q2 = 100
2

©2005 Pearson Education, Inc. 39

 Duopoly Example
Q1
The demand curve is P = 30 - Q and
30 both firms have 0 marginal cost.
Firm 2’s
Reaction Curve

Cournot Equilibrium
15

10
Firm 1’s
Reaction Curve

10 15 30 Q2

©2005 Pearson Education, Inc. 40

 Oligopoly Example
l Profit Maximization with Collusion
P = TR = PQ = (30 - Q )Q = 30Q - Q 2
¶TR
MR = = 30 - 2Q
¶Q
MR = MC = 0
30 - 2Q = 0 Þ Q = 15

If fixed costs for both firms = 0

P =30 (15 ) - (15 )
2
= 225
225
P1 = P 2 = = 112.5
©2005 Pearson Education, Inc.
2 41
 Profit Maximization w/Collusion

l Collusion Curve
mQ1 + Q2 = 15
l Shows all pairs of output Q1 and Q2 that
maximizes total profits

mQ1 = Q2 = 7.5
l Less output and higher profits than the Cournot
equilibrium

©2005 Pearson Education, Inc. 42

 Duopoly Example
Q1
30 For the firm, collusion is the best
Firm 2’s outcome followed by the Cournot
Reaction Curve Equilibrium and then the
competitive equilibrium

Competitive Equilibrium (P = MC; Profit = 0)

15 Cournot Equilibrium

10 Collusive Equilibrium

7.5 Firm 1’s

Reaction Curve
Collusion
Curve
7.5 10 15 30 Q2

©2005 Pearson Education, Inc. 43

Criticism

l Cournot's model leads to a stable equilibrium. However, his model may be criticized on several accounts:
l The behavioural pattern of firms is naive. Firms do not learn from past miscalculations of competitors’
reactions. Although the quantity produced by the competitors is at each stage assumed constant, a
quantity competition emerges which drives price down, towards the competitive level.
l The model can be extended to any number of firms. However, it is a 'closed' model, in that
entry is not allowed: the number of firms that are assumed in the first period remains the same
throughout the adjustment process.
l The model does not say how long the adjustment period will be.
l The assumption of costless production is unrealistic. However, it can be relaxed without
impairing the validity of the model. This is done in the presentation of the model, based on
the reaction-curves approach.

©2005 Pearson Education, Inc. 44

 Price Adjusting Model: Bertrand
Model

©2005 Pearson Education, Inc. 45

 Lecture’s objective

The objective of this lecture is to introduce

Students to Strategic Interactions under the theory
of Oligopoly. Specifically the lecture will examine
the following:
Ø Dynamic Quantity adjusting modelsReaction
functions/curves
Ø Price adjusting models: Bertrand Model
 Reading List

l Perloff, J. M. (2018). Microeconomics, global edition.

Pearson Education Canada. [Page 448-489]
l Varian, H. R. (2016). Intermediate Microeconomics with
Calculus: A Modern Approach: Ninth Edition International
Student Edition. WW Norton & Company.
l Perloff, J. (2013). Microeconomics with Calculus, Global
Edition. Pearson Education UK. [Page 514-562]
l Varian, H. R. (2014). Intermediate microeconomics: a
modern approach: ninth international student edition. WW
Norton & Company. [Page 516-539]
l Nicholson, W., & Snyder, C.M. (2014). Intermediate
microeconomics and its application. Cengage Learning.
[Page 374-410]
©2005 Pearson Education, Inc. 47
 Price Competition

l Competition in an oligopolistic industry

may occur with price instead of output.
l The Bertrand Model is used instead of
Cournot Model.
mOligopoly model in which each firm treats the
price of its competitors as fixed, and all firms
decide simultaneously what price to charge

©2005 Pearson Education, Inc. 48

 Price Competition – Bertrand
Model

l Assumptions
mHomogenous good
mMarket demand is P = 30 - Q where
Q = Q1 + Q2
mMC1 = MC2 = $3

l If we have Cournot competition à Q1 =

Q2 = 9 and P=$12 giving each firm a
profits of $81.
©2005 Pearson Education, Inc. 49
 Price Competition – Bertrand
Model
l Assume here that the firms compete with
price, not quantity.
l Since good is homogeneous à
consumers will buy from lowest price
seller
mIf firms charge different prices, consumers
buy from lowest priced firm only
mIf firms charge same price, consumers are
indifferent who they buy from
©2005 Pearson Education, Inc. 50
Price Competition – Bertrand Model
l The standard Bertrand model
m2 firms
l Homogeneous products
l Identical constant marginal cost: c
l Set price simultaneously to maximize profits
l Firms have no capacity constraint
mConsumers
l Firm with lower price attracts all demand, Q(p)
l At equal prices, market splits at a1 and a2=1-a1
m® Firm i faces demand

ì Q( pi ) if pi < p j
ï
Qi ( pi ) = ía iQ( pi ) if pi = p j
ï pi > p j
î Inc. 0
©2005 Pearson Education, if
 Price Competition – Bertrand
Model

l Bertrand-Nash equilibrium: firms have

incentive to cut prices below rivals
mBoth firms set P=MC à like in compt. Mkt.
mP = MC; P1 = P2 = $3
mQ = 27; Q1 & Q2 = 13.5

l Both firms earn zero profit

©2005 Pearson Education, Inc. 52

 Price Competition – Bertrand
Model – A real example

©2005 Pearson Education, Inc. 53

 Price Competition – Bertrand
Model

l Why not charge a different price

(P>MC)?
mIf charge more, sell nothing
mIf charge less, lose money on each unit sold

l The Bertrand model demonstrates the

importance of the strategic variable
mPrice versus output

©2005 Pearson Education, Inc. 54

 Bertrand Model – Criticisms

l When firms produce a homogenous

good, it is more natural to compete by
setting quantities rather than prices.
l Even if the firms do set prices and
choose the same price, what share of
total sales will go to each one?
mIt may not be equally divided à e.g. brand
name loyalty à product differentiation.

©2005 Pearson Education, Inc. 55

Price Competition – Mathematical
Derivation

l Reaction Curves can be derived:

Firm 1's profit maximizing price =
¶p 1
= 12 - 4 P1 + P2 = 0
¶P1
Firm 1's reaction curve :
1
P1 = 3 + P2
4
Likewise ® Firm 2's reaction curve
can be derived:
1
P2 = 3 + P1
4
©2005 Pearson Education, Inc. 56
 Price Competition –
Differentiated Products
l Bertrand Nash equilibrium:
1 1
P1 = 3 + P2 and P2 = 3 + P1
4 4
1æ 1 ö 15 1
P1 = 3 + ç 3 + P1 ÷ = + P1
4è 4 ø 4 16
P1 = 4 and P2 = 4

l Profit à
p 1 = 12 P1 - 2 P12 + P1 P2 - 20
p 1 = 12 & p 2 = 12

©2005 Pearson Education, Inc. 57

 Nash Equilibrium in Prices
l A digression à What if both firms collude?
mThey both decide to charge the same price that
maximized both of their profits
mFirms will charge $6 and will be better off
colluding since they will earn a profit of $16
P = P1 = P2
P =p 1 + p 2
P = (12 P1 - 2 P12 + P1 P2 - 20 ) + (12 P2 - 2 P22 + P1 P2 - 20 )
P = 24 P - 2 P 2 - 40
¶P
= 24 - 4 P = 0 Þ P = 6 Þ p 1 = p 2 = 16
¶P

©2005 Pearson Education, Inc. 58

 Nash Equilibrium in Prices
P1 Firm 2’s Reaction Curve Equilibrium at price
of $4 and profits of
Collusive Equilibrium $12

$6

$4

Firm 1’s Reaction Curve

Nash Equilibrium

$4 $6 P2
©2005 Pearson Education, Inc. 59
 Nash Equilibrium in Prices

l What if they move sequentially in A

Stackelberg fashion? àIf Firm 1 sets price
first and then firm 2 makes pricing decision
mFirm 1 would be at a distinct disadvantage by
moving first
mThe firm that moves second has an
opportunity to undercut slightly and capture a
larger market share

©2005 Pearson Education, Inc. 60

 First Mover Advantage – The Stackelberg
Quantity Competition Model

l Oligopoly model in which one firm sets its

output before other firms do.
l Assumptions
mOne firm can set output first
mMC = 0
mMarket demand is P = 30 - Q where Q is total
output
mFirm 1 sets output first, and Firm 2 then makes
an output decision seeing Firm 1 output
©2005 Pearson Education, Inc. 61
 The Algebra of the Stackelberg
Model

©2005 Pearson Education, Inc.

 The Algebra of the Stackelberg
Model

©2005 Pearson Education, Inc.

 The Algebra of the Stackelberg
Model
l Since the follower reacts to the leader’s output,
the follower’s output is determined by its
reaction function
a - c2
Q2 = r2 (Q1 ) = - 0.5Q1
2b
l The Stackelberg leader uses this reaction
function to determine its profit maximizing
output level, which simplifies to
a + c2 - 2c1
Q1 =
©2005 Pearson Education, Inc.
2b
 First Mover Advantage – The Stackelberg
Quantity Competition Model

l Example
mOne firm can set output first
mMC = 0
mMarket demand is P = 30 - Q where Q is total
output
mFirm 1 sets output first and Firm 2 then
makes an output decision seeing Firm 1
output

©2005 Pearson Education, Inc. 65

 First Mover Advantage – The
Stackelberg Model
l Firm 1
mMust consider the reaction of Firm 2

l Firm 2
mTakes Firm 1’s output as fixed and therefore
determines output with the Cournot reaction
curve: Q2 = 15 - ½(Q1)
period 1 period 2

Firm 1 sets Q1, anticipating Firm 2 sets it best response

Firm 2’s best response quantity Q2. quantity Q2, against Q1.

©2005 Pearson Education, Inc. 66

 First Mover Advantage – The
Stackelberg Model
l By backward induction: Using Firm 2’s Reaction Curve for Q2:

1
l Firm 1: Q2 = 15 - Q1
2

TR1 = ( 30 - ( Q1 + Q2 ) ) Q1 Þ TR1 = 30Q1 - Q12 - Q1Q2

æ 1 ö 1 2
TR1 = 30Q1 - Q - Q1 ç 15 - Q1 ÷ = 15Q1 - Q1
1
2

è 2 ø 2
¶TR1
MR1 = = 15 - Q1
¶Q1
MR1 = MC1 Þ 15 - Q1 = 0 Þ Q1 = 15
1
Q2 = 15 (15 ) - ( ) = 7.5 Þ P = 7.5
2
15
2
©2005 Pearson Education, Inc. 67
 First Mover Advantage – The
Stackelberg Model
l Profits:
Suppose that Fixed Cost for both firms = 0
P1 = TR1 = 30Q1 - Q12 - Q1Q2
P1 = 30 (15 ) - (15 ) - (15 )( 7.5 ) = 112.5
2

P 2 = TR2 = 30Q2 - Q22 - Q1Q2

P1 = 30 ( 7.5 ) - ( 7.5 ) - (15 )( 7.5 ) = 56.25
2

©2005 Pearson Education, Inc. 68

 First Mover Advantage – The
Stackelberg Model

l Conclusion
mGoing first gives firm 1 the advantage
mFirm 1’s output is twice as large as firm 2’s
mFirm 1’s profit is twice as large as firm 2’s

l Going first allows firm 1 to produce a large

quantity. Firm 2 must take that into account
and produce less unless wants to reduce
profits for everyone

©2005 Pearson Education, Inc. 69

 First Mover Advantage – The
Stackelberg Model
l Conclusion
mGoing first gives firm 1 the advantage
mFirm 1’s output is twice as large as firm 2’s
mFirm 1’s profit is twice as large as firm 2’s

l Going first allows firm 1 to produce a

large quantity. Firm 2 must take that into
account and produce less unless wants
to reduce profits for everyone
©2005 Pearson Education, Inc. 70
 Stakelberg: Leader-Leader
l Under the Stackelberg leadership model a
duopolist gains the first mover advantage if it
becomes the leader, thus both firms strive to be
the leader.

l However, if the two firms desire to be the leader,

each assumes that the other’s behaviour is
governed by its reaction function, but neither firms
observe the the reaction function, resulting in an
unstable market situation.
©2005 Pearson Education, Inc. 71
 Stakelberg: Leader-Leader

l The situation leads to the Stackelberg disequilibrium and

Stackelberg is of the view that this disequilibrium is the most
frequent outcome. The final result of a Stackelberg
disequilibrium cannot be predicted a priori.

l If Stackelberg is correct equilibrium will not be achieved until

one has succumbed to the leadership of other or a collusive
agreement has been reached. The model shows that a
bargaining procedure and a collusive agreement becomes
advantageous to both duopolists.

©2005 Pearson Education, Inc. 72

 Stakelberg: Follower-Follower

l If both firms desire to be followers, their expectations do not

materialize (since each assumes that the rival will act as a
leader), and they must revise them.

l Two behavioural patterns are possible. If each duopolist

recognises that his rival wants also to be a follower, the
Cournot equilibrium is reached. Otherwise, one of the rivals
must alter his behaviour and act as a leader before
equilibrium is attained.

l That said its not rational for a firm to desire to be a follower

knowing the outcome of surrounding the leadership to its
rival.
©2005 Pearson Education, Inc. 73
 Outcomes compared Linear demand
Consider
l Market price and aggregate output are

l The second column is increasing, the third decreasing

l Stackelberg duopoly is more efficient than Cournot duopoly
l Stackelberg duopoly has lower aggregate profits than Cournot
duopoly
©2005 Pearson Education, Inc. 74
 Price Signaling and Price
Leadership

l Price Signaling
mImplicit collusion in which a firm announces a
price increase in the hope that other firms will
follow suit

l Price Leadership
mPattern of pricing in which one firm regularly
announces price changes that other firms
then match

©2005 Pearson Education, Inc. 75

 Collusive Agreement à Cartels

l Producers in a cartel explicitly agree to

cooperate in setting prices and output.
l Typically only a subset of producers are
part of the cartel and others benefit from
the choices of the cartel
l If demand is sufficiently inelastic and
cartel is enforceable, prices may be well
above competitive levels

©2005 Pearson Education, Inc. 76

 Cartels
l Examples of l Examples of
successful cartels unsuccessful cartels
m OPEC m Copper
m Tin
m De Beers (Diamond
Cartel) m Coffee
m Tea
m Cocoa

©2005 Pearson Education, Inc. 77

 Cartels – Conditions for
Success

1. Stable cartel organization must be

formed – price and quantity settled on
and adhered to à sometimes difficult
because;
m Members have different costs,
assessments of demand and objectives
m Tempting to cheat by lowering price to
capture larger market share

©2005 Pearson Education, Inc. 78

 Cartels – Conditions for
Success

2. Potential for monopoly power

m Even if cartel can succeed, there might be
little room to raise price if faces highly
elastic demand
m If potential gains from cooperation are
large, cartel members will have more
incentive to make the cartel work

©2005 Pearson Education, Inc. 79

 Cartels

l To be successful:
mTotal demand must not be very price elastic
mEither the cartel must control nearly all of the
world’s supply or the supply of noncartel
producers must not be price elastic

©2005 Pearson Education, Inc. 80

 Cartels

©2005 Pearson Education, Inc. 81

 Cartels

©2005 Pearson Education, Inc. 82

Figure 13.1 Competition Versus Cartel
(a) Firm (b) Market

Price, p, Price,p,
$ per unit $ per unit

MC
S
em
pm pm
AC

pc pc ec

MC m MCm
Market demand

MR

qm qc q* Qm Qc
Quantity,q, Units Quantity,Q, Units
per year per year
©2005 Pearson Education, Inc.
 Maintaining cartels

to maintain cartel, firms must

l detect cheating
l punish violators
l keep its illegal behaviour hidden from
governments

©2005 Pearson Education, Inc.

 Detection and enforcement
l inspect each other's books (e.g., most-
favored nation clauses)
l governments report bids on
government contracts
l divide market by region or by customers
mercury cartel (1928-1972) allocated
U.S. to Spain and Europe to Italy
l use industry organizations to detect
cheating
l offer "low price" guarantees
©2005 Pearson Education, Inc.
 Competition Versus Collusion:
The Prisoners’ Dilemma
l Collusion with competitors will give larger
profits.
l If all agree to charge $6, each earn profit
of $16.
l Collusive agreement hard to enforce à
temptation to undercut the rival and
charge slightly below $6 à gets the
whole market.

©2005 Pearson Education, Inc. 86

 Competition Versus Collusion:
The Prisoners’ Dilemma

l Assume:

FC = $20 and VC = $0
Firm 1' s demand : Q = 12 - 2 P1 + P2
Firm 2' s demand : Q = 12 - 2 P2 + P1
Nash Equilibrium : P = $4 p = $12
Collusion : P = $6 p = $16
©2005 Pearson Education, Inc. 87
 Competition Versus Collusion:
The Prisoners’ Dilemma

l Possible Pricing Outcomes:

Firm 1 : P = $6 Firm 2 : P = $6 p = $16

P = $6 P = $4
p 2 = P2Q2 - 20
= (4)[12 - (2)(4) + 6] - 20 = $20
p 1 = P1Q1 - 20
= (6)[12 - (2)(6) + 4] - 20 = $4

©2005 Pearson Education, Inc. 88

 Payoff Matrix for Pricing Game
Firm 2
Dominant
Strategy Charge $4 Charge $6

Charge $4
$12, $12 $20, $4

Firm 1

Charge $6 $4, $20 $16, $16

Nash Eq.
($4;$4)

better outcomes
©2005 Pearson Education, Inc. 89
 Competition Versus Collusion:
The Prisoners’ Dilemma
l We can now answer the question of why firm
does not choose cooperative price.
l Cooperating means both firms charging $6
instead of $4 and earning $16 instead of $12
l Each firm always makes more money by
charging $4, no matter what its competitor does
l Unless enforceable agreement to charge $6,
will be better off charging $4

©2005 Pearson Education, Inc. 90

 Competition Versus Collusion:
The Prisoners’ Dilemma
l An example in game theory, called the
Prisoners’ Dilemma, illustrates the
problem oligopolistic firms face.
mTwo prisoners have been accused of
collaborating in a crime.
mThey are in separate jail cells and cannot
communicate.
mEach has been asked to confess to the
crime.

©2005 Pearson Education, Inc. 91

 Payoff Matrix for Prisoners’
Dilemma
Prisoner B
Dominant
Strategy Confess Don’t confess

Confess
-5, -5 -1, -10

Prisoner A Would you choose to confess?

Don’t
confess -10, -1 -2, -2
Nash Eq.
(confess;confess)

better outcomes
©2005 Pearson Education, Inc. 92
 Prisoner’s Dilemma (PD)
• It is indeed difficult for them to cooperate!! The matrix
form representation of the game shown in the movie
clip. Female

Split Steal
(Sp) (St)

Split
(Sp) 50 , 50 0 , 100
Multiple Nash Equilibrium
Male

Steal 100 , 0 0 , 0
(St)

• Stealing is a ‘weakly dominant’ strategy.

• (Splitting; Splitting) does not constitute as a Nash-
equilibrium à unilateral incentive to deviate.
©2005 Pearson Education, Inc. 93
 Collusion
l Do the same analysis for the Cournot
competition we derived earlier!
(Homework J).
l Conclusions
1. Collusion will lead to greater profits
2. Explicit and implicit collusion is possible
à need a ‘binding’ agreement à cartel
agreement.

©2005 Pearson Education, Inc. 94

 Collusion (Price Fixing)
BA and Virgin: Flying in formation

Aug 2nd 2007

From The Economist print edition

It takes two to fix prices

FOR years British Airways (BA) described itself as “the world's favorite airline”. It no longer looks so popular in London and
Washington. On August 1st the firm was hit with a transatlantic double whammy after it was found guilty of colluding with a rival,
Virgin Atlantic, to fix prices on long-haul passenger routes. Britain's Office of Fair Trading (OFT) handed down a record fine of
£121.5m ($246m). A few hours later, America's Department of Justice (DoJ) imposed a $300m penalty of its own. The severity of the
American fine also reflected BA's role in a different international conspiracy involving Korean Air and Lufthansa.

A clearer example of illegal price-fixing than that between BA and Virgin would be hard to imagine. The two firms discussed “fuel
surcharges” at least six times between August 2004 and January 2006, during which time they rose from £5 to £60 on a return ticket.

A transatlantic bust was particularly fitting for the OFT. During Labour's period in office, it has introduced American-style, cartel-
busting sanctions on companies that prefer cozy deals with rivals to the bracing winds of competition. But despite many protracted
investigations into sectors such as banking and supermarkets that attract consumers' ire, the OFT has struggled to find the kind of
smoking-gun evidence of collusion it needed to look as terrifying as it and the government wished. That is partly the nature of the
beast. Collusion is difficult to prove: as Mr Collins observes, the tricky thing about colluders is that they do their business in secret.
Indeed, the airlines' price-fixing came to light only after Virgin's legal department alerted the authorities.

This was no selfless dedication to consumers' welfare. Virgin hoped to benefit from the “leniency policy”, which was introduced in
the 1998 Competition Act and copied from similar laws in America, granting immunity to firms that blow the whistle. Virgin was just
as complicit as BA in the price-fixing and has, presumably, benefited from it financially. Not only was the airline saving itself from
the risk of prosecution, but it was also grassing up a rival with whom it has had a bruising relationship in the past. It grates to see
one firm get away with something while another is punished, but leniency policies are, probably, a good thing. The ability to claim
immunity gives a powerful incentive for businesses to police their own industries, which ought to improve things for consumers.
After all, half a victory is better than none.
©2005 Pearson Education, Inc. 95
 Price Rigidity

l Firms have strong desire for stability

l Price rigidity – characteristic of
oligopolistic markets by which firms are
reluctant to change prices even if costs
or demands change
mFear lower prices will send wrong message
to competitors leading to price war
mHigher prices may cause competitors to raise
theirs
©2005 Pearson Education, Inc. 96