Microeconomics MECN 430 Spring 2016

Download Report

Transcript Microeconomics MECN 430 Spring 2016

session nine
dynamic pricing (II)
industry consolidation (ad companies) ………….1
industry consolidation (newspapers) ………….8
spring
2016
microeconomi
the analytics of
cs
constrained optimal
microeconomics
lecture 9
dynamic pricing (II)
the analytics of constrained optimal
decisions
AD COMPANIES
industry consolidation
Consolidation: Omnicom and Publicis
► Omnicom Group Inc. and Publicis Groupe SA said Sunday , July 28th, 2013, they will merge to create a $35.1 billion advertising
giant, overtaking current market leader WPP PLC in the industry's biggest deal ever. U.S.-based Omnicom and France's Publicis, the
second and third biggest ad companies respectively by revenue, said they will create a new entity called the Publicis Omnicom Group
in a merger of equals.
Revenue for 2012
(billions)
WPP PLC
Omnicom Group
Publicis Groupe
Interpublic Group
Proforma Dentsu & Aegis
Havas
Western
Europe
$5.80
$3.59
$2.38
$1.44
$0.96
$1.12
North
America
$5.65
$7.30
$4.12
$3.78
$0.77
$0.73
Rest of
World
$4.97
$3.32
$1.89
$1.78
$4.68
$0.41
Total
$16.42
$14.21
$8.39
$7.00
$6.41
$2.26
Market
Share
30.02%
25.98%
15.34%
12.80%
11.72%
4.13%
► We will use this case as a motivation for an analysis of consolidation in an industry in which firms compete in capacities. To keep
the setup as simple as possible let’s assume:
● there are three identical firms that compete in the market, index them as i = 1,2,3
● quantity for firm i is therefore qi, i = 1,2,3
● marginal cost (in cents) for each firm is MC = 200
● market demand is P = 800 – Q, where Q is the total quantity supplied by the firm active in the market
► Step 1: we derive the market equilibrium with three firms as described above
► Step 2: we assume firm 2 and firm 3 merge and compute the equilibrium with the resulting two firms in the market
► Step 3: under what conditions is the merged firm better off compared to the un-consolidated situation?
 2016 Kellogg School of Management
lecture 9
page | 1
microeconomics
lecture 9
dynamic pricing (II)
the analytics of constrained optimal
decisions
industry consolidation
MERGER ANALYSIS
► Step 1: Market outcome with three firms, pre-consolidation
● From the perspective of firm 1:
- the residual demand is
P = (800 – q2 – q3) – q1 with marginal revenue MR1 = (800 – q2 – q3) – 2q1
- profit maximization for residual demand requires MR1 = MC thus (800 – q2 – q3) – 2q1 = 200
- the solution is immediate as q1 = 300 – 0.5(q2 + q3)
[equation 1]
● From the perspective of firm 2:
- the residual demand is
P = (800 – q1 – q3) – q2 with marginal revenue MR2 = (800 – q1 – q3) – 2q2
- profit maximization for residual demand requires MR2 = MC thus (800 – q1 – q3) – 2q2 = 200
- the solution is immediate as q2 = 300 – 0.5(q1 + q3)
[equation 2]
● From the perspective of firm 3:
- the residual demand is
P = (800 – q1 – q2) – q3 with marginal revenue MR3 = (800 – q1 – q2) – 2q3
- profit maximization for residual demand requires MR2 = MC thus (800 – q1 – q2) – 2q3 = 200
- the solution is immediate as q3 = 300 – 0.5(q1 + q2)
[equation 3]
We have to solve this system of three equations with three unknowns (q1, q2 and q3). In this particular case, when the firms are
perfectly identical we get a symmetric system which implies that q1 = q2 = q3. Say q* is this common value, then
q* = 300 – 0.5(q* + q*) with solution q* = 150, i.e. q1 = q2 = q3 = 150
 2016 Kellogg School of Management
lecture 9
page | 2
microeconomics
lecture 9
dynamic pricing (II)
the analytics of constrained optimal
decisions
industry consolidation
MERGER ANALYSIS
► Step 1: Market outcome with three firms, pre-consolidation
● From the perspective of firm 1 the best response (reaction function) is given by q1 = 300 – 0.5(q2 + q3)
● The solution is q1 = 150 and q2 = q3 = 150, this
is point (O) in the diagram
q1
● Price in the market is given by
P* = 800 – (q1 + q2 + q1) = 350
● Profit for each firm is the same and equal to
300
firm 1’s best response to
firm 2 and firm 3 actions
q1 = 300 – 0.5(q2 + q3)
Π* = (P* – MC)q* = (350 – 200)·150 = 22,500
The pre-merger cumulative profit for firm 2 and firm 3 is
thus
Π2+3* = 45,000
pre-merger
150
(O)
300
 2016 Kellogg School of Management
lecture 9
600
q2 + q3
page | 3
microeconomics
lecture 9
dynamic pricing (II)
the analytics of constrained optimal
decisions
industry consolidation
MERGER ANALYSIS
► Step 2: Market outcome with two firms, post-consolidation
Here we assume that firm 2 and firm 3 merge into firm (2,3) with a resulting marginal cost of MC2,3 = 200. There are two firms in the
market now. Let q1 the quantity produced by firm1 and q2,3 the quantity produced by the consolidated firm (2,3).
● From the perspective of firm 1:
- the residual demand is
P = (800 – q2,3 ) – q1 with marginal revenue MR1 = (800 – q2,3) – 2q1
- profit maximization for residual demand requires MR1 = MC1 thus (800 – q2,3 ) – 2q1 = 200
- the solution is immediate as q1 = 300 – 0.5q2,3
[equation 1]
● From the perspective of firm (2,3):
- the residual demand is
P = (800 – q1) – q2,3 with marginal revenue MR2,3 = (800 – q1) – 2q2,3
- profit maximization for residual demand requires MR2,3 = MC2,3 thus (800 – q1) – 2q2,3 = 200
- the solution is immediate as q2,3 = 300 – 0.5q1
[equation 2]
We have to solve this system of two equations with two unknowns (q1 and q2,3 ). Again the firms are perfectly identical thus we get a
symmetric system, which implies that q1 = q2,3. Say q** is this common value, then
q** = 300 – 0.5q** with solution q** = 200, i.e. q1 = q2,3 = 200
 2016 Kellogg School of Management
lecture 9
page | 4
microeconomics
lecture 9
dynamic pricing (II)
the analytics of constrained optimal
decisions
industry consolidation
MERGER ANALYSIS
► Step 2: Market outcome with two firms, post-consolidation
● From the perspective of firm 1 the best response (reaction function) is given by q1 = 300 – 0.5q2,3 while, from the
perspective of firm (2,3), the best response (reaction function) is given by q2,3 = 300 – 0.5q1
● The solution is q1 = 200 and q2,3 = 200, this
is point (M) in the diagram
● Price in the market is given by
q1
firm (2,3)’s best response to
firm 1’s actions
q2,3 = 300 – 0.5q1
600
P** = 800 – (q1 + q2,3) = 400
● Profit for each firm is the same and equal to
firm 1’s best response to
firm (2,3)’s actions
q1 = 300 – 0.5q2,3
300
Π** = (P** – MC)q** = (400 – 200)·200 = 40,000
post-merger
The post-merger cumulative profit for firm 2 and firm 3
is thus
Π2,3** = 40,000
200
(M)
(O)
150
200
 2016 Kellogg School of Management
pre-merger
lecture 9
300
600
q2,3
page | 5
microeconomics
lecture 9
dynamic pricing (II)
the analytics of constrained optimal
decisions
industry consolidation
MERGER ANALYSIS
► Step 2: Market outcome with two firms, post-consolidation
● Looks like the merger does not create a higher profit for the combined firms, i.e. the merger cannot make both
the shareholders of firm 2 and firm 3 better off…
Π2,3** = 40,000 < Π2+3* = 45,000
● On the other hand firm 1 is far better off…
Π1** = 40,000 > Π1* = 22,500
● Pre-merger firm 2 and firm 3 together supply a total of 300 units while post merger the supply of the consolidated
firm is 200…
● Price increases but not by enough to make for the reduction in supply, and the reason that price does not
increase is that much is that firm 1 increases its own supply (from 150 to 200) as an optimal respond to the
reduced supply from the new consolidated firm
● The consolidation would increase the profit (for firm 2 and firm 3) if firms have a limited capacity, i.e. cannot
increase its supply say above 150.
 2016 Kellogg School of Management
lecture 9
page | 6
microeconomics
lecture 9
dynamic pricing (II)
the analytics of constrained optimal
decisions
industry consolidation
MERGER ANALYSIS
► Step 3: Limited capacity, post-consolidation
● Let’s assume for the moment that all three firms have limited capacity of 150. Then the pre-merger market
outcome is unchanged. But the reaction function for firm 1 is now slightly changed as shown in the diagram.
Since it cannot offer anything above 150 the reaction function is “cut” at 150.
● The solution is q1 = 150 and q2,3 = 225, this
is point (L) in the diagram
● Price in the market is given by
q1
600
firm (2,3)’s best response to
firm 1’s actions
q2,3 = 300 – 0.5q1
300
firm 1’s best response to
firm (2,3)’s actions
q1 = min{150, 300 – 0.5q2,3}
P*** = 800 – (q1 + q2,3) = 425
● Profit for firm 1 is
Π1*** = (P*** – MC1)q1*** =
= (425 – 200)·150 = 33,500
● The post-merger profit for firm (2,3) is
Π2,3*** = (P*** – MC2,3)q2,3*** =
= (425 – 200)·225 = 50,625
200
150
► When a competitor has a limited capacity the
market outcome is altered towards an increase in
profit post-merger.
 2016 Kellogg School of Management
(M)
post-merger
post-merger
pre-merger
(L)
200 225 300
lecture 9
(O)
600
q2,3
page | 7
microeconomics
lecture 9
dynamic pricing (II)
the analytics of constrained optimal
decisions
INDUSTRY ANALYSIS
industry consolidation
Newsprint Industry Consolidation
► Industry facts :
31.75%
20.71%
6.83%
5.66%
5.52%
4.68%
3.93%
3.64%
3.00%
2.53%
► Consolidation:
● The economies-of-scale case:
Consolidations helps capture economies of scale …
..by reducing overhead per ton, better use of resources
This eventually translate into lower prices to customers
 2016 Kellogg School of Management
● The market power case:
Greater power over buyers, thus higher prices
Better management of capacity
Price signaling and collusion
lecture 9
page | 8
microeconomics
lecture 9
dynamic pricing (II)
the analytics of constrained optimal
decisions
INDUSTRY ANALYSIS
industry consolidation
Newsprint Industry Consolidation
► Capacity management
► The graph is suggestive
of deliberate reduction in
capacity,
resulting
in
increasing or at least
stemming the decline in
prices
► But this leaves the
following puzzle:
Why would a merged firm
shut down plants that they
were profitable to operate
pre-merger?
 2016 Kellogg School of Management
lecture 9
page | 9
microeconomics
lecture 9
dynamic pricing (II)
the analytics of constrained optimal
decisions
CAPACITY ANALYSIS
► Case I: Tight Market
industry consolidation
► Capacity management: You have three identical manufacturing units, each producing q. Current
price is p0 and you contemplate closing down one unit thus reducing your own supply to 2q. Assume all
others are always producing at maximum capacity.
close this unit
q
p
● loss of profit on closed unit at old price
● gain of profit on open units at new price
q
p
demand
demand
p1
p0
p0
profit
gain
profit
extra profit
loss
3q
 2016 Kellogg School of Management
Q0 Q
2q
lecture 9
Q1
Q0 Q
page | 10
microeconomics
lecture 9
dynamic pricing (II)
the analytics of constrained optimal
decisions
CAPACITY ANALYSIS
► Case I: Tight Market
industry consolidation
► Capacity management: in comparing the gain with loss from shutting-down capacity there are two
● Position and shape of demand curve
factors to consider:
● Competitors’ cost profiles
gain on opened units at new price
q
p
q
p
p1
“flat” demand
“steep” demand
p1
p0
gain
gain
p0
profit
extra profit
2q
 2016 Kellogg School of Management
Q1
Q0 Q
profit
extra profit
2q
lecture 9
Q1
Q0 Q
page | 11
microeconomics
lecture 9
dynamic pricing (II)
the analytics of constrained optimal
decisions
CAPACITY ANALYSIS
► Case II: Over Capacity
industry consolidation
► Capacity management: You have three identical manufacturing units, each producing q. Current
price is p0 and you contemplate closing down one unit thus reducing your own supply to 2q. Assume all
others are always producing at maximum capacity.
close this unit
q
● loss of profit on closed unit at old price
● no gain of profit on open units at new price
p
q
p
profit
extra profit
profit
demand
demand
p0
p1=p0
loss
3q
 2016 Kellogg School of Management
Q0
Q
2q
lecture 9
Q1=Q0
Q
page | 12
microeconomics
lecture 9
dynamic pricing (II)
the analytics of constrained optimal
decisions
CAPACITY ANALYSIS
► Case II: Over Capacity
industry consolidation
► Capacity management: in comparing the gain with loss from shutting-down capacity there are two
● Position and shape of demand curve
factors to consider:
● Competitors’ cost profiles
no gain on opened units at new price
q
q
p
p
profit
extra profit
profit
extra profit
“steep” demand
“flat” demand
p1=p0
p1=p0
2q
 2016 Kellogg School of Management
Q1=Q0
Q
2q
lecture 9
Q1=Q0
Q
page | 13
microeconomics
lecture 9
dynamic pricing (II)
the analytics of constrained optimal
decisions
CAPACITY ANALYSIS
industry consolidation
► Capacity management: ● orange company has capacity q
● purple company has capacity qL at low cost and qH at high cost
► If the purple firm would close the high-cost facility it would gain P
► If the orange firm acquires purple firm and would close the high-cost facility it would gain O + P
qH
close this unit
p
p
demand
demand
gain
p1
O
p0
q
 2016 Kellogg School of Management
qL
P
p0
profit orange
profit purple
qH
Q
loss
q
lecture 9
qH
qL
qH
Q
page | 14