Imperfectly Competitive Markets:

Monopolistic Competition and Oligopoly

-
Dr. Sanjay K. Singh
Indian Institute of Management Lucknow
Monopolistic Competitors in the Short Run

(a) Firm Makes a Profit

Price
MC
ATC

Price
ATC
Profit Demand
MR

0 Profit-
Quantity
maximizing quantity
Monopolistic Competitors in the Short Run

(b) Firm Makes Losses ATC

MC
Price
Losses

ATC
Price

Demand
MR

0 Loss- Quantity
minimizing
quantity
A Monopolistic Competitor in the Long Run

Price
MC
ATC

P=ATC

Demand
MR
0
Qπ-max Quantity
Monopolistic versus Perfect Competition

Perfect competition Monopolistic competition

Price Price
MC MC
LRAC ATC
PM
PC D
PC

MR D
0 QC Quantity 0 QM QC Quantity
 The example What is the
equilibrium for this
The Pay-Off Matrix game?
The left-hand American
number is the
pay-off to
Delta Morning Evening

Morning (15, 15) (30, 70)

Delta The right-hand
number is the
Evening (70, 30) (35, 35)
pay-off to
American
 If American The If American
example
chooses an evening
chooses a morning
The Pay-Off Matrix
departure,
The morning Deltadeparture Delta The morning departure
departure,
willis choose will also choose
a dominated is also a dominated
eveningfor Delta
strategy evening strategy for American
American
Both airlines
choose an
Morning Evening
evening
departure
Morning (15, 15) (30, 70)
Delta
Evening (70, 30) (35, 35)
 IfThe
Deltaexample
choosesThe Pay-Off Matrix However, a
a morning
departure, American morning departure
But if Delta
American haswill no
choose is still a dominated
chooses an evening
dominated strategy American
evening strategy for Delta
departure, American
American knows
willand
this choose
so Morning Evening
choosesmorning
a morning
departure
Morning (18, 12) (30, 70)
Delta
Evening (70,30)
(70, 30) (42, 28)
 NASH EQUILIBRIUM REVISITED
The Beach Location Game

Beach Location Game

You (Y) and a competitor (C) plan to sell soft drinks on a beach.
If sunbathers are spread evenly across the beach and will walk to the closest vendor, the two of you
will locate next to each other at the center of the beach. This is the only Nash equilibrium.
If your competitor located at point A, you would want to move until you were just to the left, where you
could capture three-fourths of all sales.
But your competitor would then want to move back to the center, and you would do the same.
 The example
If Delta prices high
If both price highand The Pay-Off low
American Matrix
then both get 30then American gets
passengers.
If Delta prices
Profit
low
all 180 passengers.American
and
perAmerican
passengerhigh
isProfit
If both
perprice low
passenger
then$300
Delta gets they each get 90
is $20
all 180 passengers. passengers.PH = $500 PL = $220
Profit per passenger
Profit per passenger
is $20
P = $500 is($9000,$9000)
$20 ($0, $3600)
H

Delta
PL = $220 ($3600, $0) ($1800, $1800)
 (PH, PH) is a Nash Nash equilibrium
equilibrium.
There are two(P HThe
, PL)Pay-Off
Nash There
Matrixis no simple
cannot be
way to (PL,between
choose PL) is a Nash
If both are pricing
equilibria to thisa version
Nash equilibrium.
highCustom and
thenofneither familiarity
wants these equilibrium.
equilibria
the game If American prices American
might lead both to
to change If both are pricing
(PL, PHprice
) cannotlowbethen Delta should
high low then neither wants
a Nash equilibrium. also price low
to change
PH = $500 PL = $220
If American prices
high then Delta should
also pricePhigh
H = $500 ($9000,$9000)
($9000, $9000) ($0, $3600)
Delta
PL = $220 ($3600, $0) ($1800, $1800)
 The Cournot model
If the output of
$ firm 1 is increased
P = (A - Bq1) - Bq2
the demand curve
The profit-maximizing A - Bq1 for firm 2 moves
choice of output by firm
to the left
2 depends upon the
output of firm 1 A - Bq’1
Marginal revenue for
Solve this Demand
firm 2 is c MC
for output MR2
MR2 = (A - Bq1) - 2Bq2
q2
MR2 = MC q*2
Quantity
A - Bq1 - 2Bq2 = c  q*2 = (A - c)/2B - q1/2
 Cournot-Nash equilibrium
q2
If firm 2 produces The
The reaction
reaction function
function
The Cournot-Nash
(A-c)/B (A-c)/B then firm for
for firm
firm 11isis
equilibrium is at
1 will choose to q*
q*11== (A-c)/2B
(A-c)/2B --qq22/2
/2
Firm 1’s reactionthe intersection
function
produce no output
Ifoffirm
the reaction
2 produces
then firmThe
functions
nothing Thereaction
reactionfunction
function
(A-c)/2B
1 will produce the for
forfirm
firm22isis
qC 2
C
monopoly output q* q*22==(A-c)/2B
(A-c)/2B--qq11/2 /2
Firm 2’s reaction function
(A-c)/2B
q1
qC1 (A-c)/2B (A-c)/B
 Cournot-Nash equilibrium
q2 q*1 = (A - c)/2B - q*2/2

(A-c)/B
q*2 = (A - c)/2B - q*1/2

Firm 1’s reaction function

 q*2 = (A - c)/2B - (A - c)/4B
+ q*2/4
 3q*2/4 = (A - c)/4B
(A-c)/2B
 q*2 = (A - c)/3B
C
(A-c)/3B
Firm 2’s reaction function  q*1 = (A - c)/3B
q1
(A-c)/2B (A-c)/B
(A-c)/3B
Cournot-Nash equilibrium: many firms
• What if there are more than two firms?
• Much the same approach.
• Say that there are N identical firms producing
identical products
• Total output Q = q1 + q2 + … This
+ qdenotes
N
output
of every firm other
• Demand is P = A - BQ = A - B(q1 + q2 + … + qN)
than firm 1
• Consider firm 1. It’s demand curve can be written:
P = A - B(q2 + … + qN) - Bq1
• Use a simplifying notation: Q-1 = q2 + q3 + … + qN
• So demand for firm 1 is P = (A - BQ-1) - Bq1
 The Cournot model: many firms
If the output of
$ the other firms
P = (A - BQ-1) - Bq1
is increased
The profit-maximizing the demand curve
choice of output by firm A - BQ-1 for firm 1 moves
1 depends upon the to the left
output of the other firms A - BQ’
-1
Marginal revenue for
firm 1 is Solve this Demand
for output c MC
MR1 = (A - BQ-1) - 2Bq
q1 1 MR1

MR1 = MC q*1
Quantity
A - BQ-1 - 2Bq1 = c  q*1 = (A - c)/2B - Q-1/2
Cournot-Nash equilibrium: many firms
q*1 = (A - c)/2B - Q-1/2
How do we solve this
 Q*-1 = (N - 1)q*1 As the number of
for
The firms q* ?
arenumber
1 identical.
firms As the
increases output of
 q*1 = (A - c)/2B - (N - 1)q*1So
/2 in equilibrium they
of eachfirms
firmincreases
falls
 (1 + (N - 1)/2)q*1 = (A - c)/2Bwillaggregate
have identical
output
As the number of
outputs
increases
 q*1(N + 1)/2 = (A - c)/2B As the number of
firms increases price
 q*1 = (A - c)/(N + 1)B firms increases profit
tends to marginal cost
of each firm falls
 Q* = N(A - c)/(N + 1)B
 P* = A - BQ* = (A + Nc)/(N + 1)
Profit of firm 1 is P*1 = (P* - c)q*1 = (A - c)2/(N + 1)2B
Cournot-Nash equilibrium: different costs

• What if the firms do not have identical costs?

• Much the same analysis can be used
• Marginal costs of firm 1 are c1 and of firmSolve2 this
are c2.
for output
• Demand is P = A - B.Q = A - B(q1 + q2)
q1
• We have marginal revenue for firm 1 as before
• MR1 = (A - Bq2) - 2Bq1 A symmetric result
holds for output of
• Equate to marginal cost: (A - Bqfirm2) - 2Bq
2 1 = c1
 q*1 = (A - c1)/2B - q2/2
 q*2 = (A - c2)/2B - q1/2
 Cournot-Nash equilibrium: different costs

q2 q*1 = (A - c1)/2B - q*2/2

The equilibrium
If the marginal
(A-c1)/B output cost
of firm 2 q*
of firm 2 2 = (A - c2)/2B - q*1/2
R1 What happens
increases
falls and of q*2 = to
its reaction (Athis
- c2)/2B - (A - c1)/4B
1 equilibrium
fallsshifts to when
firmcurve
+ q*2/4
costs
the rightchange?
 3q*2/4 = (A - 2c2 + c1)/4B
(A-c2)/2B
 q*2 = (A - 2c2 + c1)/3B
R2 C
 q*1 = (A - 2c1 + c2)/3B
q1
(A-c1)/2B (A-c2)/B
 Concentration and profitability
• Assume that we have N firms with different marginal costs
• We can use the N-firm analysis with a simple change
• Recall that demand for firm 1 is P = (A - BQ-1) - Bq1
• But then demand for firm i is P = (A - BQ-i) - Bqi
• Equate this to marginal cost ci
But Q*-i + q*i = Q*
A - BQ-i - 2Bqi = ci and A - BQ* = P*
This can be reorganized to give the equilibrium condition:
A - B(Q*-i + q*i) - Bq*i - ci = 0
 P* - Bq*i - ci = 0  P* - ci = Bq*i
 Concentration and profitability
P* - ci = Bq*i
The price-cost margin
Divide by P* and multiply the right-hand
for eachside
firmbyisQ*/Q*
P* - ci BQ* q*i determined by its
= market share and
P* P* Q*
demand elasticity
But BQ*/P* = -1/ and q*i/Q* =Average
si price-cost
so: P* - ci =- si margin is
P*  determined by industry
concentration
Extending this we have (multiply by si on both side and then  )
i

P* - c =- H
P* 
Price Competition
(Bertrand Competition)
 Bertrand competition
• Dd for firm 2 can be p2
illustrated as follows: There is a
• Demand is jump at p2 = p1
discontinuous
• The discontinuity in p1
demand carries over to
profit

a – bp2 a q2
(a – bp2)/2
 Bertrand competition
Firm 2’s profit is:
p2(p1,, p2) = 0 if p2 > p1

p2(p1,, p2) = (p2 - c)(a - bp2) if p2 < p1

p2(p1,, p2) = (p2 - c)(a - bp2)/2 if p2 = p1 For whatever

reason!
Clearly this depends on p1.
Suppose first that firm 1 sets a “very high” price:
greater than the monopoly price of pM = (a +c)/2b
 What price
Bertrand competition
should firm 2
So firm 2 should just set?
With p1 >undercut
(a + c)/2b, Firm
p1 a bit 2’s profit looks like this:
Atand
p2 = p1
Firmget
2’salmost
Profit
firm
Whatall the
2 ifgets
firmhalf
1 of the The monopoly
monopoly profit
pricesmonopoly
at profit
(a + c)/2b? price

p2 < p 1

Firm
Firm22willwillonly
onlyearn
earnaa
positive
positiveprofit
profitby
bycutting
cuttingits
its
price
pricetoto(a
(a++c)/2b
c)/2bor
orless
less
p2 = p 1

p2 > p 1

c (a+c)/2b p1 Firm 2’s Price

 Bertrand competition
Now suppose that firm 1 sets a price less than (a + c)/2b
Firm 2’s profit looks like this:
Firm 2’s Profit AsWhat price
long as p1 > c,
Of course, firm 1 should
Firm firmaim
2 should 2 just
will then undercut to undercut
set now?firm 1
firm 2 and so on
p2 < p 1
Then firm 2 should also price
at c. Cutting price below cost
gains the whole market but loses
What if firm
money on1every customer
prices at c? p2 = p 1

p2 > p 1

c p1 (a+c)/2b Firm 2’s Price

 Bertrand
The best responsecompetition
function for The best response
These best response
firmfunctions
1 look like this function for
p2 firm 2
R1

R2
(a + c)/2b
The Bertrand
The equilibrium
equilibrium has
isboth
with both
firms charging
firms pricing at
c marginal
c cost

p1
c (a + c)/2b
 The example

• Then, will Richards Price

change its price from $83.33

P = $60? Demand
$60

MR
$36.66

$10 MC
– since QR = 1,400 Richards is
1,400 Quantity
at capacity and does not
want to reduce price
• Same logic applies to Pepall so P = $60 is a Nash
equilibrium for this game.
 An example of product differentiation
Coke and Pepsi are similar but not identical. As a result,
the lower priced product does not win the entire market.
Econometric estimation gives:

QC = 63.42 - 3.98PC + 2.25PP

MCC = $4.96

QP = 49.52 - 5.48PP + 1.40PC

MCP = $3.96
There are at least two methods for solving this for PC and PP
 Bertrand and product differentiation

Both methods give the best response

functions:
PC = 10.44 + 0.2826PP PP
The
NoteBertrand
that these
equilibrium
are upwardis RC
PP = 6.49 + 0.1277PC
at their
sloping
These can be solved intersection
for the equilibrium $8.11 RP
prices as indicated B
$6.49
The equilibrium prices
are each greater than
marginal cost
PC
$10.44
$12.72
 Dynamic Games and
First and Second Movers
(Stackelberg Competition)
 Stackelberg equilibrium
Equate marginal revenue
This is firm 2’s
MR2 = (A - Bq1) – 2Bq2 best with marginal cost
response
function q2
MC = c But firm 1 knows Firm
Firm11knows
knowsthat that
what q
 q*2 = (A - c)/2B - q1/2 2 is going this
thisisishow
howfirmfirm22
to be will
willreact
react totofirm
firm
firm 1’s
1’s
Demand for firm 1 is: So
So firm 11can
can
output
output choice
choice
anticipate firm 2’s
P = (A - Bq2) – Bq1 From (A earlier
– c)/2B example we know
anticipate firm 2’s
reaction
P = (A - Bq*2) – Bq1 that this is the monopoly output. Thisreaction is an
P = (A - (A-c)/2) – Bqimportant
/2
1Equate
result. The Stackelberg leader
marginal revenue S
chooses the same output as a monopolist would.
(A – c)/4B
 P = (A + c)/2 – Bq1/2
Butwith
firmmarginal cost
2 is not excluded from the market R2
Solve
Marginal revenue for firm 1 is:this equation
for output q1 q1
MR1 = (A + c)/2 - Bq1 (A – c)/2B (A – c)/B

(A + c)/2 – Bq1 = c
 q*1 = (A – c)/2B  q*2 = (A – c)4B
 Stackelberg equilibrium
Aggregate output is 3(A-c)/4B
q Leadership benefits
So the equilibrium price is (A+3c)/4 2 Firm
the 1’s best
leader firm response
1 but
(A-c)/B
Firm 1’s profit is (A-c)2/8B R1 Leadership
function
harms benefits
is “like”
the follower
consumers
firm
firm
Compare 2 but
2’sthis
Firm 2’s profit is (A-c) /16B
2 Compare thiswith
with
reducestheaggregate
theCournot
Cournot
We know that the Cournot profits
equilibrium
(A-c)/2B
equilibrium
equilibrium is:
qC1 = qC2 = (A-c)/3B (A-c)/3B C
S
The Cournot price is (A+c)/3 (A-c)/4B
R2
Profit to each firm is (A-c)2/9B
q
(A-c)/3B (A-c)/2B (A-c)/ B 1
Collusion and Cartels
 An example
• Take a simple example
– two identical Cournot firms making identical products
– for each firm MC = $30
– market demand is P = 150 - Q
– Q = q 1 + q2 Price
150

Demand

30 MC

150 Quantity
 The incentive to collude
Profit for firm 1 is: p1 = q1(P - c)
= q1(150 - q1 - q2 - 30)
= q1(120 - q1 - q2)
To maximize, differentiate with respect to q1: Solve
Solvethis
thisfor
forqq11

p1/q1 = 120 - 2q1 - q2 = 0

q*1 = 60 - q2/2
This
Thisisisthe
thebest
bestresponse
response
The best response function for firm 2 is then:
function
functionforforfirm
firm11
q*2 = 60 - q1/2
 The Incentive to cheat

Both firms have the

incentive to cheat on
their agreement
Firm 1

Cooperate (M) Deviate (D)

This
Thisisisthe
theNash
Nash
Cooperate (M) equilibrium
equilibrium
(1800, 1800) (1350, 2025)
Firm 2

Deviate (D) (2025, 1350) (1600, 1600)

(1600, 1600)
 Finitely repeated games
• Original game but repeated twice
• Consider the strategy for firm 1
– first play: cooperate
– second play: cooperate if firm 2 cooperated in the
first play, otherwise choose to deviate

Firm 1

Cooperate (M) Deviate (D)

Cooperate (1800, 1800) (1350, 2025)

(M)
Firm 2

Deviate (D) (2025, 1350) (1600, 1600)

 Finitely repeated games
• This strategy is unsustainable
– the promise is not credible
• at end of period 1 firm 2 has a promise of cooperation from
firm 1 in period 2
• but period 2 is the last period
• dominant strategy for firm 1 in period 2 is to deviate

Firm 1

Cooperate (M) Deviate (D)

Cooperate (M) (1800, 1800) (1350, 2025)

Firm 2

Deviate (D) (2025, 1350) (1600, 1600)

 An example

• Consider the following pricing game

Firm 2

$105 $130 $160

$105 (7.31, 7.31) (8.25, 7.25) (9.38, 5.53)

Firm 1

$130 (7.25, 8.25) (8.5, 8.5) (10, 7.15)

$160 (5.53, 9.38) (7.15, 10) (9.1, 9.1)

 An example

• There are two Nash equilibria

Firm 2
Both agree that
$105 this is “bad”
$130 $160
Both agree that
$105 (7.31, 7.31)
(7.31, 7.31) (8.25, 7.25) this is “good”
(9.38, 5.53)
Both agree that
Firm 1

this is best
$130 (7.25, 8.25) (8.5, 8.5) (10, 7.15)

$160 (5.53, 9.38) (7.15, 10) (9.1, 9.1)

 An example
• Best response to a price Firm 2

of $160 is $130
• This gives profit of 10 $105 $130 $160

• So deviation in period 1
gives the profit stream $105 (7.31, 7.31) (8.25, 7.25) (9.38, 5.53)

10 + 7.31 = $17.31
• Cooperation in period 1

Firm 1
gives the profit stream $130 (7.25, 8.25) (8.5, 8.5) (10, 7.15)

9.1 + 8.5 = $17.6

• Undercutting does not pay
$160 (5.53, 9.38) (7.15, 10) (9.1, 9.1)
 Cartel stability
• This is an example of a more general result
There is always a
• Suppose that in each period value of R < 1 for which
– profits to a firm from a collusivethis are pC
equation is
agreement
– profits from deviating from the agreement satisfied are pD
– profits in the Nash equilibrium are pN
– we expect thatThispD >isispthe
This C short-run
the pN
>short-run gain
gain
from
fromcheating
cheating on
onthe
thecartel
cartel
• Cheating on the cartelThis is
Thisdoes the long-run
not pay
is the long-run loss
so long as:
loss
pD - pC from
fromcheating
cheatingon
onthe
thecartel
cartel
Rr >
pD - pN
• The cartel is stable
– if short-term gains from cheating are low relative to long-run
losses
– if cartel members value future profits (low discount rate)