VA353, Industrial Organization

1
VA353, Industrial Organization Part II
VA353, Industrial Organization Strategic
Behavior Early Models of Imperfect
Competition Formal Models of Oligopoly Game
Theory Approaches to Oligopoly Collusion Mergers
2
VA 353
Industrial Organization
Learning Objectives for Exam 2

• Become familiar with the analytical tools that
  economists use to deal with questions of
  strategic behavior in markets formal models of
  oligopoly and game theory
• Know the factors that facilitate collusive
  behavior such as bid-rigging and price fixing.
• Understand the notion of the term merger
  paradox and its implications for proposed
  mergers.
• Become familiar with the methods by which
  economists and policy makers evaluate the
  potential anti-competitive implications of
  mergers.

Rose-Hulman Institute of Technology / K.
Christ VA353, Industrial Organization
3
Early Models of Imperfect Competition

• Monopolistic Competition (Chamberlain, 1930s)

P
ATC
MR
Q
4
Early Models of Imperfect Competition

• Kinked Demand Curve (Sweezy, 1939)

This portion of the firms demand curve (for
price increases) is more elastic.
Current Equilibrium, P0, Q0
MC1
P0
MC2
This portion of the firms demand curve (for
price decreases) is less elastic.
D
MR
Q0
5
Early Models of Imperfect Competition

• Dominant Firm Model (Stigler, 1950)

MCd
6
Structural Models of Oligopoly
7
Cournot Duopoly

• Two competitors simultaneously choose their
  output, each assuming that the other behaves just
  as they do.
• The solution may be conceived of as the
  simultaneous solution of each competitors
  profit-maximizing reaction function.
• Example
• 
  and

Generalizing for n identical competitors
8
Allocative Inefficiency in Cournot Equilibrium as
a Percentage of Allocative Inefficiency Under
Monopoly
Cabral, p. 154 It does not take a large number
of identical firms for the performance of the
Cournot model to be very close to that of
perfect competition.
c 0
c a/4
c a/2
9
Bertrand Duopoly

• Two competitors simultaneously choose their
  price, each assuming that the other behaves just
  as they do.
• The solution may be conceived of as the
  simultaneous solution of each competitors
  profit-maximizing reaction function.

There are four possible price configurations 1.
p1 gt p2 gt c
This is not an
equilibrium. Firm 1 could profitably deviate by
setting p1 p2 e.

• Assumptions
• Homogenous good.
• Consumers purchase from the low-price firm.

2. p1 gt p2 c
This is not an equilibrium.
Firm 2 captures the entire market, but its
profits are zero. Firm 2 could profitably
deviate by setting p2 p1 e.
3. p1 p2 gt c
This is not an equilibrium
since either firm could profitably deviate by
setting pi pj e.
4. p1 p2 c
This is a Nash equilibrium.
Neither firm can profitably deviate and earn
greater profits even though profits are zero.
10
Stackelberg Leader-Follower

• One competitor chooses its optimal output,
  assuming that the other competitors will respond
  in a manner predicted by their reaction
  functions.
• The solution may be conceived of as the
  sequential solution of each competitors profit
  functions
• and

Step 1 Leader (firm 1) maximizes
profit, assuming follower (firm 2) responds by
staying on its reaction function
Step 2 Follower (firm 2) takes firm 1s output
as given
11
Comparing Deterministic Duopoly Outcomes
Demand Costs Numeric outcomes assume a 24, b
1, c 0, f 0
Monopoly Joint Profit Max.
Cournot
Stackelberg
Perfect Competition
6
8
12
12
6
8
6
12
12
16
18
24
12
8
6
0
72
64
72
0
72
64
36
0
144
128
108
0
12
Structural Oligopoly Outcomes Compared to
Competition
Demand and reaction functions
Firm 2s Reaction Function
Stackelberg Outcome if Firm 1 Leads
.
.
Bertrand / Competitive Outcome
.
Cournot Outcome
.
Stackelberg Outcome if Firm 2 Leads
Firm 1s Reaction Function
13
Common Models of Oligopoly
.
14
Formalizing a Game Matrix Form
Strategy sets for each player
Players
Payoff to Player A (Row Player)
Payoff to Player B (Column Player)

15
Formalizing a Game Extensive Form
Decision Nodes
50,50
In a sequential game, each decision node
defines a subgame
b1
a1
b2
80,30
A
B
A box around Player B denotes a simultaneous
decision
30,80
b1
a2
b2
75,75
16
Sequential Games / Backward Induction Subgame
Perfect NE
For each subgame, determine the optimal strategy
q
Subgame 1
2,2
b1
a1
5,0
b2
A
B
Subgame 2
0,5
b1
a2
Find the optimal strategy for the pruned tree
q
3,3
b2
17
Sequential Games / Backward Induction Subgame
Perfect NE
b1 Price War
1,1
a1 Enter
b2 Status Quo
2,2
B (Incumbant Firm)
A (Entrant)
b1 Price War
0,-1
a2 Stay Out
b2 Status Quo
0,3
18
Focal Point Equilibrium (Schelling, 1960)
In games with multiple NE, one of the NE might
stand out because of some asymmetry that is
common knowledge to the players.
NE a1, b1 and a2,b2
F
If fairness is a criteria for both players, then
a1, b1 may emerge as a focal point
equilibrium.
Thomas Schelling
B
b1
b2
a1
6,8
2,1
A
3,11
1,3
a2
19
Repeated Play / Reciprocity and Supportable
Equilibrium
Suppose that player A adopts the following
strategy
Play a2. If Player B plays b2, continue to play
a2. If Player B plays b1, then play a1 until
Player B changes.
20
Repeated Play / Reciprocity and Supportable
Equilibrium
Suppose that player A adopts the following
strategy
Play a2. If Player B plays b2, continue to play
a2. If Player B plays b1, then play a1 until
Player B changes.
Such a strategy will mean that both players
face two possible streams of payoffs
Time, t
Strategy 1 Defection
7
7
7
9
Strategy 2 Cooperation
8
8
8
8
21
Repeated Play / Reciprocity and Supportable
Equilibrium
Define
Payoff from cooperation Payoff from
defection Payoff from being defected
upon Payoff from NE (Cournot competition)
B
b1
b2
a1
,
,
A
a2
,
,
22
Repeated Play / Reciprocity and Supportable
Equilibrium
In such a game, cooperation may be sustained
if the following condition is met
23
(No Transcript)
24
Mergers and Acquisitions
Success and Failure
For a successful merger
gt

Market Value Of Combined Firm A and Firm B
Market Value Of Firm B
Market Value Of Firm A
Typically estimated as
Common Shares Outstanding X
Current Share Price
25
Mergers and Acquisitions
Rolls Hubris Hypothesis
On average, decision makers in acquiring firms
pay too much for their targets. Managerial
hubris explains why managers persist in offering
takeover bids that represent overly optimistic
valuations of the assets they are
acquiring. Richard Roll, The Hubris Hypothesis
of Corporate Takeovers, The Journal of Business
592 (April 1986) 197 216.
The Sage of Omahas Version
Many managements apparently were overexposed in
impressionable childhood years to the story in
which the imprisoned handsome prince is released
from a toad's body by a kiss from a beautiful
princess. Consequently, they are certain their
managerial kiss will do wonders for the
profitability of Company T investors can always
buy toads at the going price for toads. If
investors instead bankroll princesses who wish to
pay double for the right to kiss the toad, those
kisses had better pack some real dynamite. Weve
observed many kisses but very few miracles.
Warren Buffet
26
Mergers and Acquisitions
The Merger Paradox