the flexibility of the oligopoly game by mike reynolds email: [email protected] dee 2011 -...

27
The Flexibility of the Oligopoly Game By Mike Reynolds Email: [email protected] DEE 2011 - Slide 1

Upload: susanna-wheeler

Post on 28-Dec-2015

214 views

Category:

Documents


0 download

TRANSCRIPT

The Flexibility of the Oligopoly Game

By Mike ReynoldsEmail: [email protected]

DEE 2011 - Slide 1

Session Outline

• Introduction• Example Game• A Closer Look• Extending the Game

DEE 2011 - Slide 2

Introduction

• Meister (1999)– Describes an oligopoly game

• Students split into teams/firms– Make decisions on quantity– Complicated set up

• Simpler version of game

DEE 2011 - Slide 3

Game Set-Up

• Class Room arranged• Each team/firm is given an envelope

containing instructions– See PDF

DEE 2011 - Slide 4

Example Game

• Sixth-form visit– Year 12

• 8 teams• 50 minute session• Outcomes will not necessarily look like the

theory– This is fine

DEE 2011 - Slide 5

Round 1 - Monopoly

• Teams hand in output decisions– Team 1: 2000 olives– Team 2: 2500 olives – Team 3: 3000 olives– Team 4: 900 olives– Team 5: 3000 olives– Team 6: 3500 olives– Team 7: 4000 olives– Team 8: 3125 olives

DEE 2011 - Slide 6

Round 1 - Profits

• Students then see profits of all firms– Team 1: £4300 (from 2000 olives)– Team 2: £3862.50 (from 2500 olives) – Team 3: £2800 (from 3000 olives)– Team 4: £3062.50 (from 900 olives)– Team 5: £2000 (from 3000 olives)– Team 6: £1112.50 (from 3500 olives)– Team 7: £-1200 (from 4000 olives)– Team 8: £2438.72 (from 3125 olives)

DEE 2011 - Slide 7

Round 2 – (Cournot) Competition

• All teams in same market  Team 1 Team 2 Team 3 Team 4 Team 5 Team 6 Team 7 Team 8 Q P

Round 1

Monopoly

2000

4300

2500

3862.50

3000

2800

900

3062.50

3000

2000.00

3500

1112.50

4000

-1200

3125

2438.72- -

Round 2

Cournot

2000

-700

320

-280

200

-250

300

-275

750

-387.50

400

-300

300

-275

750

-587.505020 0

Round 3

CournotRound 4

Cournot MC*2

Round 5

Cournot Collusion 

Total Profit3600.00 3582.50 2550.00 2787.50 2412.50 812.50 -1475.00 2049.22

Round 3 – (Cournot) Competition

• Repeat of Round 2

DEE 2011 - Slide 9

  Team 1 Team 2 Team 3 Team 4 Team 5 Team 6 Team 7 Team 8 Q PRound 1

Monopoly

2000

4300

2500

3862.50

3000

2800

900

3062.50

3000

2000.00

3500

1112.50

4000

-1200

3125

2438.72- -

Round 2

Cournot

2000

-700

320

-280

200

-250

300

-275

750

-387.50

400

-300

300

-275

750

-587.505020 0

Round 3

Cournot

1000

-450

300

-275

700

-375

500

-325

500

-325

500

-325

800

-400.00

777

-394.255077 0

Round 4

Cournot MC*2

Round 5

Cournot Collusion 

Total Profit

3150.00 3307.50 2175.00 2462.50 2087.50 487.50 -1875.00 1654.97

Round 4 – (Cournot) Competition

• Marginal costs double

DEE 2011 - Slide 10

  Team 1 Team 2 Team 3 Team 4 Team 5 Team 6 Team 7 Team 8 Q PRound 1

Monopoly

2000

4300

2500

3862.50

3000

2800

900

3062.50

3000

2000.00

3500

1112.50

4000

-1200

3125

2438.72- -

Round 2

Cournot

2000

-700

320

-280

200

-250

300

-275

750

-387.50

400

-300

300

-275

750

-587.505020 0

Round 3

Cournot

1000

-450

300

-275

700

-375

500

-325

500

-325

500

-325

800

-400.00

777

-394.255077 0

Round 4

Cournot MC*2

50

-87.06

210

274.34

100

25.88

600

1155.25

200

251.75

50

-87.06

150

138.81

433

778.04 1793 2.76

Round 5

Cournot Collusion 

Total Profit

3062.94 3581.84 2200.88 3617.75 2339.25 400.44 -1736.19 2433.01

Round 5 – (Cournot) Competition

• Collusion allowed

DEE 2011 - Slide 11

  Team 1 Team 2 Team 3 Team 4 Team 5 Team 6 Team 7 Team 8 Q PRound 1

Monopoly

2000

4300

2500

3862.50

3000

2800

900

3062.50

3000

2000.00

3500

1112.50

4000

-1200

3125

2438.72- -

Round 2

Cournot

2000

-700

320

-280

200

-250

300

-275

750

-387.50

400

-300

300

-275

750

-587.505020 0

Round 3

Cournot

1000

-450

300

-275

700

-375

500

-325

500

-325

500

-325

800

-400.00

777

-394.255077 0

Round 4

Cournot MC*2

50

-87.06

210

274.34

100

25.88

600

1155.25

200

251.75

50

-87.06

150

138.81

433

778.04 1793 2.76

Round 5

Cournot Collusion 

2000

-1200

300

-350.00

500

-450.00

300

-350.00

300

-350.00

5000

-2700

133

-266.50

400

-400 8933 0

Total Profit

1862.94 3231.84 1750.88 3267.75 1989.25 -2299.56 -2002.69 2033.01

Final Results

• Team 4 win

DEE 2011 - Slide 12

  Team 1 Team 2 Team 3 Team 4 Team 5 Team 6 Team 7 Team 8 Q PRound 1

Monopoly

2000

4300

2500

3862.50

3000

2800

900

3062.50

3000

2000.00

3500

1112.50

4000

-1200

3125

2438.72- -

Round 2

Cournot

2000

-700

320

-280

200

-250

300

-275

750

-387.50

400

-300

300

-275

750

-587.505020 0

Round 3

Cournot

1000

-450

300

-275

700

-375

500

-325

500

-325

500

-325

800

-400.00

777

-394.255077 0

Round 4

Cournot MC*2

50

-87.06

210

274.34

100

25.88

600

1155.25

200

251.75

50

-87.06

150

138.81

433

778.04 1793 2.76

Round 5

Cournot Collusion 

2000

-1200

300

-350.00

500

-450.00

300

-350.00

300

-350.00

5000

-2700

133

-266.50

400

-400 8933 0

Total Profit 1862.94 3231.84 1750.88 3267.75 1989.25 -2299.56 -2002.69 2033.01

Debrief

• Need to give students a quick overview– In other groups you could go further

• Depends on ability of students– In this case, I underlined the contrast between

monopoly and competition• Monopolists made profits (and had high prices)• Whilst competition drove prices very low

– Lots of other elements that could be teased out• Collusion

DEE 2011 - Slide 13

A Closer Look

• Have an idea how the game goes• But why is the game set up this way?– Ensures the game is to be successful– Can alter elements to fit the students

DEE 2011 - Slide 14

Workings of the Game (1)

• Recall– “It is believed that in the relatively small market in

which you operate that the total demand for olives even at very low prices is unlikely to be higher than 4000 units, and that even the biggest olive lover in the market is only willing to pay a maximum of around £5 per unit.”

DEE 2011 - Slide 15

Workings of the Game (2)

• Sketching the demand curve:

– Or the equation:• P = 5 – 0.00125Q

DEE 2011 - Slide 16

Price

Quantity

5

4000

Workings of the Game (3)

• Only simple details given– Emphasise the intuitive aspect• Can appeal to all levels of student

• Possible to make it more complicated for more advanced students

• Profit maximisation– Could become welfare maximisation

DEE 2011 - Slide 17

Organisation

• Set up in advance– Need to have an estimate of the number of teams

and the abilities of the players– Spreadsheet is set up in advance• Only requires a demand structure

DEE 2011 - Slide 18

Emphasise

• Non-collusion– Only allow collusion at the end of the game• Costs for firms are the same

– Could change to non-symmetric

DEE 2011 - Slide 19

Decision Framework

• Can vary the types of competition– Start with a (relatively) simple round– Could repeat

• Pre-announced game length– Final round can bring odd results (as we saw)– Good idea to change the game at this point

• Feedback of results– Profits or…– Can provide hints to class

DEE 2011 - Slide 20

Refereeing the Game

• Set times that decisions need to be made by– Initially a longer period

• Be reactive to behaviour– Warn and then punish teams• Collusion• Late decisions

– Plan alterations to the game

DEE 2011 - Slide 21

Linking The Game

• Sixth-form conference– 5-10 minute debrief– Or an extra session talking about the results

DEE 2011 - Slide 22

Extensions

• Basic game is appropriate to all– Economic problem solving

• Can vary for the type of student or specific learning outcome– Sixth-form conferences– 1st year modules or inductions– 2nd or 3rd years on Microeconomic modules– Postgraduates

DEE 2011 - Slide 23

Extensions: Level of Information

• Details of demand– More detailed information• Students calculate solutions

– Could even start with a utility function

DEE 2011 - Slide 24

Decision Framework

• Monopoly• Cournot• Bertrand– Price becomes decision variable– Emphasis the Bertand paradox

• Stackelberg– Make one team the leader

• Bundling• Any firm problem with a demand structure

DEE 2011 - Slide 25

Other Extensions

• Can be part of a module• Focus of maximisation– Profit or collective welfare?

• Length of game– 50 minute session or over several weeks

• Assessment– Give marks based on profit/welfare (Meister, 1999)– Gives students the details of the game and ask them to

explain them• Lab-based simulation

DEE 2011 - Slide 26

References

• S. R. Beckham. 2003. “Cournot and Bertrand Games”, Journal of Economic Education 2003 34(1), pp. 27-35.

• J. Brauer. 1994. “Games Economists Play: Non-Computerised Classroom-Games in College Economics”

• S. Gächter, C. Thoni., and J-R. Tyran. 2006. “Cournot Competition and Hit-and-Run Entry and Exit in a Teaching Experiment”, Journal of Economic Education 37(4), pp. 418-430.

• J. Meister. 1999. “An In-Class Economic Game”, Journal of Economic Education 30(4): pp. 383-391.

• A. Ortmann. 2003. “Bertrand Price Undercutting: A Brief Classroom Demonstration”, Journal of Economic Education 34(1): pp. 21-26

DEE 2011 - Slide 27