0% found this document useful (0 votes)
4 views8 pages

Problem Set 4 Short Solutions

The document presents a problem set on advanced microeconomics focusing on imperfect competition, covering various models such as Cournot, Bertrand, and Stackelberg. It includes mathematical formulations for outputs, prices, and profits under different scenarios, including asymmetric costs and collusion strategies. Additionally, it discusses concepts like entry deterrence and consumer surplus in competitive markets.

Uploaded by

bernx2125
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
4 views8 pages

Problem Set 4 Short Solutions

The document presents a problem set on advanced microeconomics focusing on imperfect competition, covering various models such as Cournot, Bertrand, and Stackelberg. It includes mathematical formulations for outputs, prices, and profits under different scenarios, including asymmetric costs and collusion strategies. Additionally, it discusses concepts like entry deterrence and consumer surplus in competitive markets.

Uploaded by

bernx2125
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd

Problem Set #4 - Short Solutions

Advanced Microeconomics - Imperfect Competition

P1. Cournot with asymmetric costs


For interior outputs:

𝑎 − 2𝑐1 + 𝑐2 𝑎 − 2𝑐2 + 𝑐1
𝑞1∗ = , 𝑞2∗ =
3 3

𝑎 + 𝑐 1 + 𝑐2
𝑃∗ = , 𝜋𝑖 = (𝑞𝑖∗ )2
3
If 𝑐1 < 𝑐2 < 𝑎 and 2𝑐2 > 𝑎 + 𝑐1 , firm 2 is inactive:

𝑎 − 𝑐1 𝑎 + 𝑐1
𝑞1∗ = , 𝑞2∗ = 0, 𝑃∗ =
2 2

(𝑎 − 𝑐1 )2
𝜋1 = , 𝜋2 = 0
4

P2. Classical models


Given 𝑃 = 15 − 𝑄, 𝑐 = 3:

Model 𝑞1 𝑞2 𝑄 𝑃 𝜋1 𝜋2 Total
Cartel 3 3 6 9 18 18 36
Cournot 4 4 8 7 16 16 32
Bertrand 6 6 12 3 0 0 0
Stackelberg 6 3 9 6 18 9 27

P3. Sequential game with one leader and two followers


Let 𝐴 = 𝑎 − 𝑐. The followers’ reaction is:

𝐴 − 𝑞1
𝑞2 = 𝑞 3 =
3
The leader maximizes 𝜋1 = 𝑞1 (𝐴 − 𝑞1 )/3, so:

𝐴 𝐴 𝐴
𝑞1∗ = , 𝑞2∗ = 𝑞3∗ = , 𝑃∗ = 𝑐 +
2 6 6

𝐴2 𝐴2
𝜋1 = , 𝜋2 = 𝜋3 =
12 36
Classical Cournot with three firms gives 𝑞𝑖 = 𝐴/4, 𝜋𝑖 = 𝐴2 /16. The leader gains; followers lose.

1
P4. Competitive fringe and price leader
Demand: 𝑄 = 70000 − 2000𝑃 . Each small firm has 𝑀 𝐶 = 𝑞 + 5, so 𝑞 = 𝑃 − 5.
a) Aggregate supply:

𝑄𝑆 = 1000(𝑃 − 5)

Equilibrium:

1000(𝑃 − 5) = 70000 − 2000𝑃 ⇒ 𝑃 = 25, 𝑄 = 20000

b) With a leader at 𝑀 𝐶 = 15, residual demand is:

𝑄𝐿
𝑄𝐿 = 75000 − 3000𝑃 ⇒ 𝑃 = 25 −
3000

Set 𝑀 𝑅 = 𝑀 𝐶:

2𝑄𝐿
25 − = 15 ⇒ 𝑄𝐿 = 15000, 𝑃 = 20
3000

Fringe output: 𝑄𝐹 = 1000(20 − 5) = 15000. Total output: 𝑄 = 30000.


c) Consumer surplus increases because price falls from 25 to 20. If the new firm behaved competitively, price would
fall further to 15, so consumer surplus would be even larger.

P5. Bertrand collusion with trigger strategies


a) With two firms:

𝜋𝑀 /2 1
≥ 𝜋𝑀 ⇒ 𝛿 ≥
1−𝛿 2

b) With 𝑛 firms:

𝜋𝑀 /𝑛 1
≥ 𝜋𝑀 ⇒ 𝛿 ≥ 1 −
1−𝛿 𝑛

More firms make collusion harder.

P6. Medical device duopoly


Demand: 𝑄 = 5000 − 100𝑃 , so 𝑃 = 50 − 𝑄/100. Marginal cost: 𝑀 𝐶 = 10.
a) Cournot:

4000 8000 70
𝑞1 = 𝑞2 = , 𝑄= , 𝑃 =
3 3 3

160000
𝜋𝑖 = ≈ 17777.78
9

b) Collusion threshold under Cournot:

160000
𝜋𝐶 = 20000, 𝜋𝐷 = 22500, 𝜋𝑁 =
9

2
𝜋𝐷 − 𝜋𝐶 9
𝛿≥ 𝐷 𝑁
= ≈ 0.529
𝜋 −𝜋 17

c) Under Bertrand the threshold is 1/2. Cournot collusion is slightly harder to sustain.

P7. Differentiated toothpaste


Demand: 𝑞𝑖 = 10 − 𝑝𝑖 + 𝑝𝑗 /2, 𝑐 = 4.
a) Simultaneous prices:

𝑝𝑗
𝑝𝑖 = 7 +
4

28 16 256
𝑝1 = 𝑝 2 = , 𝑞1 = 𝑞2 = , 𝜋1 = 𝜋2 =
3 3 9

b) Firm 1 leads. Firm 2 reaction:

𝑝1
𝑝2 = 7 +
4

68 66 38
𝑝1 = , 𝑝2 = , 𝑞1 = 5, 𝑞2 =
7 7 7

200 1444
𝜋1 = , 𝜋2 =
7 49

c) Firm 1 uses a puppy dog strategy: it sets a high price to soften price competition.

P8. Stackelberg and entry deterrence


Demand: 𝑃 = 120 − 𝑄, 𝑐 = 0.
a) Stackelberg:

𝑞1 = 60, 𝑞2 = 30, 𝑃 = 30

𝜋1 = 1800, 𝜋2 = 900

b) If 𝑘2 = 50, deterrence requires:

(120 − 𝑞1 )2 √
≤ 50 ⇒ 𝑞1 ≥ 120 − 10 2
4

Deterrence profit is about 1497, below 1800. Firm 1 accommodates entry.


c) If 𝑘2 = 100, deterrence requires 𝑞1 ≥ 100. Profit is 100(20) = 2000, above 1800. Firm 1 deters entry.

3
P9. Leader and 𝑛 followers
Demand: 𝑃 = 100 − 𝑄, costs zero.
a) Followers’ output:

100 − 𝑞𝐿
𝑞𝑖 =
𝑛+1

Leader maximizes:

100 − 𝑞𝐿
𝜋𝐿 = 𝑞𝐿
𝑛+1

Thus:

∗ 50 50
𝑞𝐿 = 50, 𝑞𝑖∗ = , 𝑃∗ =
𝑛+1 𝑛+1

2500 2500
𝜋𝐿 = , 𝜋𝑖 =
𝑛+1 (𝑛 + 1)2

b) With entry cost 𝑘 = 64:

(100 − 𝑞𝐿 )2 𝐷
≤ 64 ⇒ 𝑞𝐿 = 100 − 8(𝑛 + 1)
(𝑛 + 1)2

c) For 𝑛 ≥ 6, 𝑞𝐿 = 50 already deters entry. For 𝑛 < 6, the leader must increase output. The leader wants zero entry
when feasible.

P10. Bertrand with unequal costs


Demand: 𝑄 = 500 − 20𝑃 , 𝑐𝐴 = 10, 𝑐𝐵 = 8.
a) Bertrand equilibrium:

𝑃 = 10, 𝑞𝐴 = 0, 𝑞𝐵 = 300

𝜋𝐴 = 0, 𝜋𝐵 = (10 − 8)300 = 600

b) Collusion: only firm B produces. Monopoly with 𝑀 𝐶 = 8:

𝑃 = 16.5, 𝑄 = 170, 𝜋𝐵 = 1445

This is not a one-shot Nash equilibrium because a firm can undercut the collusive price.

P11. Netflix and HBO


Demand: 𝑃 = 402 − 2𝑄. Costs: 𝐶1 = 2𝑞1 + 15, 𝐶2 = 2𝑞2 + 30.
a) Bertrand:

𝑃 = 2, 𝑄 = 200

If split equally: 𝑞1 = 𝑞2 = 100.

4
b) Cournot:

200 400 406


𝑞1 = 𝑞2 = , 𝑄= , 𝑃 =
3 3 3

c) Stackelberg with Netflix leader:

𝑞1 = 100, 𝑞2 = 50, 𝑄 = 150, 𝑃 = 102

𝜋1 = 9985, 𝜋2 = 4970

d/e) Both firms prefer to be leader. Pure equilibrium: (𝐿𝑒𝑎𝑑𝑒𝑟, 𝐿𝑒𝑎𝑑𝑒𝑟). No non-degenerate mixed equilibrium; HBO
acts as follower with probability 0.

P12. Cournot-Bayesian demand uncertainty


Let 𝜌 = 𝑃 𝑟(ℎ𝑖𝑔ℎ). Bayesian equilibrium:

𝐻 130 − 20𝜌 𝐿 70 − 20𝜌 70 + 40𝜌


𝑞𝐴 = , 𝑞𝐴 = , 𝑞𝐵 =
3 3 3

If 𝜌 = 1/2:

𝐻 𝐿
𝑞𝐴 = 40, 𝑞𝐴 = 20, 𝑞𝐵 = 30

With symmetric information:

110
𝐻𝑖𝑔ℎ ∶ 𝑞𝐴 = 𝑞𝐵 =
3

70
𝐿𝑜𝑤 ∶ 𝑞𝐴 = 𝑞 𝐵 =
3

P13. Hotelling corridor


Given 𝐿 = 500, 𝑎 = 0, 𝑏 = 410.
a) Indifferent consumer:

𝑝𝐴 + 𝑡𝑥 = 𝑝𝐵 + 𝑡(410 − 𝑥)

𝑝𝐵 − 𝑝𝐴
𝑥 = 205 +
2𝑡

b) If prices are equal:

𝑥 = 205

c) With 𝑡 = 0.01:

𝑝𝐴 = 4.7, 𝑝𝐵 = 5.3, 𝑞𝐴 = 235, 𝑞𝐵 = 265

La Grande charges more because it has the larger market area.

5
P14. Linear city with locations A = 0.2 and B = 0.6
Transport cost: 2𝑑.
a) Indifferent consumer:

𝑝𝐵 − 𝑝𝐴
𝑥 = 0.4 +
4

𝑞𝐴 = 𝑥, 𝑞𝐵 = 1 − 𝑥

b) Equilibrium:

28 32
𝑝𝐴 = , 𝑝𝐵 =
15 15

7 8
𝑞𝐴 = , 𝑞𝐵 =
15 15

𝜋𝐴 ≈ 0.871, 𝜋𝐵 ≈ 1.138

c) These locations are not optimal in the non-cooperative location game. The class result is maximum differentiation.

P15. Entry deterrence


Demand: 𝑄 = 1000 − 4𝑃 , so 𝑃 = 250 − 𝑄/4. Costs are zero.
a) Monopoly:

𝑞1 = 500, 𝑃 = 125, 𝜋1 = 62500

b) Stackelberg accommodation:

𝑞1 = 500, 𝑞2 = 250, 𝑃 = 62.5

𝜋1 = 31250, 𝜋2 = 15625

c) Deterring entry without entry costs requires:

𝑞1 = 1000, 𝑃 = 0, 𝜋1 = 0

d) Firm 1 prefers accommodation. Consumers prefer deterrence because it gives the lowest price.

P16. Hotelling vs Salop


Locations: 𝐴 = 0.1, 𝐵 = 0.8. Transport cost: 𝑑.
a) Hotelling line:

𝑝𝐵 − 𝑝 𝐴
𝑥 = 0.45 +
2

𝑞𝐴 = 𝑥, 𝑞𝐵 = 1 − 𝑥

6
b) Salop circle:

𝑞𝐴 = 0.5 + 𝑝𝐵 − 𝑝𝐴 , 𝑞𝐵 = 0.5 + 𝑝𝐴 − 𝑝𝐵

c) Hotelling:

29 31
𝑝𝐴 = , 𝑝𝐵 =
30 30

𝑞𝐴 ≈ 0.483, 𝑞𝐵 ≈ 0.517

Salop:

𝑝𝐴 = 𝑝𝐵 = 0.5, 𝑞𝐴 = 𝑞𝐵 = 0.5, 𝜋𝐴 = 𝜋𝐵 = 0.25

Salop is symmetric because the market is circular.

P17. Salop model with 5 competitors


Perimeter = 2, 𝑛 = 5, transport cost 2𝑑.

2
𝑠= = 0.4
5

𝑝 = 2(0.4) = 0.8

𝑞𝑖 = 0.4, 𝜋𝑖 = 0.8(0.4) = 0.32

Positive profit implies expected entry.

P18. Bayesian Cournot with cost uncertainty


Demand: 𝑃 = 50 − 𝑄. Firm A has 𝑐𝐴 = 5. Firm B has 𝑐𝐵 = 7 or 𝑐𝐵 = 3, each with probability 1/2.
Bayesian equilibrium:

𝑞𝐴 = 15, 𝑞𝐵 (𝑐 = 7) = 14, 𝑞𝐵 (𝑐 = 3) = 16

With symmetric information:

47 41
𝑖𝑓 𝑐𝐵 = 7 ∶ 𝑞𝐴 = , 𝑞𝐵 =
3 3

43 49
𝑖𝑓 𝑐𝐵 = 3 ∶ 𝑞𝐴 = , 𝑞𝐵 =
3 3

7
P19. Hotelling with quadratic transport
𝐿 = 30, 𝑐 = 5, 𝑡𝑟𝑎𝑛𝑠𝑝𝑜𝑟𝑡 𝑐𝑜𝑠𝑡 = 4𝑑2

Optimal locations:

𝑎∗ = 0, 𝑏∗ = 30

Prices and quantities:

𝑝𝐴 = 𝑝𝐵 = 3605, 𝑞𝐴 = 𝑞𝐵 = 15

Profits:

𝜋𝐴 = 𝜋𝐵 = (3605 − 5)15 = 54000

Intuition: maximum differentiation softens price competition.

P20. Cournot vs cartel with 3 firms


Demand: 𝑄 = 6000 − 200𝑃 , so 𝑃 = 30 − 𝑄/200. Marginal cost: 𝑐 = 10. Number of firms: 𝑛 = 3.
Cournot:

𝑞𝑖 = 1000, 𝑄 = 3000, 𝑃 = 15, 𝜋𝑖 = 5000

Cartel:

2000
𝑄 = 2000, 𝑞𝑖 = , 𝑃 = 20
3

20000
𝜋𝑖 = ≈ 6666.67
3

Deviation profit:

𝜋𝐷 ≈ 8888.89

Trigger strategy condition:

𝜋𝐷 − 𝜋𝐶 4
𝛿≥ = ≈ 0.571
𝜋𝐷 − 𝜋𝑁 7

Since 𝛿 = 0.7, firms collude. Final outcome:

𝑃 = 20, 𝑄 = 2000, 𝑞𝑖 = 666.67, 𝜋𝑖 = 6666.67

You might also like