Oligopoly and Games

Oligopoly and Games:
Strategic Behaviour
Oligopoly & Games: Strategic Behaviour December 07, 2019
2
Frame for today’s discussion
• Focus – A practical (ad hoc) way rather than a rigorous way
• Systems approach – Setting up a ‘reasonable’ boundary!
• Rational player vs. behavioral player (= rational + biases) setup
• Typical sources of bias – Reference dependence, loss aversion,

framing, …
3
The art of strategy
• Life as continuous stream of decisions
• Subtle place for – customs, ethics, etiquette and strategy
• Some strategic decisions may not confirm to ‘moral’ concerns
• Better to be a good strategist than a bad one!
• Ten stories!
4
Story #1: Pick a number
• Microsoft CEO Steven Ballmer (2000-14) – I pick, you guess
• Reward – 100, 80, 60, 40, 20.
• Clues – High/Low; Discuss my strategy after 4th
• Play!
5
Story #2: Winning by loosing
• Reality competition show Survivor 1st season
• Final -1 round: Richard (39), Rudy (72), Kelly (23)
• Prize USD 1mn, Winner picks opp. in final, Contestants vote
• Pole & immunity idol challenge
• 3 outcome scenarios
6
Story #3: Hot hand
• Sachin Tendulkar or Virat Kohli – Hot hand ‘run machine’
• Psychologists – Long flips can have a spell of continuous heads
• Experiments in Hockey/Basketball and several other games
• Hot hand ~ warm game
• Team streaks, Maradona’s pass in 1986 (2 goals), Ice hockey

(points)
7
Story #4: Lead or not
• 1983 America (Sailboat) Cup final: In 4/7 US-Australia had 3-1
• Australia changed sail to right, US kept left. Australia won 4-3.
• If you lead, the surest way to win is ‘monkey see, monkey do’
• Application – Stock market forecasters, Dell vs. Apple & Sun, P&G
vs. Kimberly-Clark
8
Story #5: Intransigence
• Martin Luther vs. Popes & the Council (led to Protestant reforms)
• Charles de Gaulle’s negotiation with Roosevelt & Churchill

(France’s way in Euro Econ Comm, England out of EEC – 1963,
‘68).
• Ferdinand de Lesseps –Nile sand (Suez), tropical malaria (Panama)
• Two aspects – Intertemporal concerns, degree of intransigence

9
Story #6: Empty fridge
• Cindy/Laurie/Ray-wife & ABC Primetime – 15 pounds in 2 months
• Cortés (Spain – troop of 600) vs. Aztecs (Mexico) – Scuttled ships
• Xenophon (Athens) vs. Sparta – Back against ravine
• Cindy/Laurie/Ray-wife & ABC Primetime – 17/15+13/22 pounds
• Too many options are not good always!

10
Story #7: Buffett’s dilemma
• Warren Buffett, promoting campaign finance – individual contri.
(1K to 5K), others (nothing)
• Great! But, will not pass – Senators’ fund-raising gives job security.
• What to do (Buffett) – Opp. pay 1 bn to supporters if opp. succeeds.
• Prisoners’ dilemma – Bell cat’s neck, Khrushchev on Stalin, Hijack

11
Story #8: Mix plans
• Takashi Hashiyama (Pres., electronics co) – Sell Co.’s art ($18 mn)
• Sotheby & Christie offered auction @ 12 %. Hash suggested R-P-S
• C (scissor) & S (paper) – C’s local experts vs. S’s random frame
• Applications – Formula based tax audit, Repeated games
• Biggest looser – Hash

12
Story #9: Even bet
• When two sides trade, each expect making money. One is wrong!
• Poker – Others to fold when you bluff, no high-pot victories
– Others to raise when you are genuine, caught bluffing
• Winning an auction – winner’s curse

13
Story #10: Dangerous Game knowledge
• Barry Nalebuff (Yale) – in Jerusalem, took a cab to hotel. Driver
promised lower fare, no meter
• Upon reaching – 2500 shekels ($2.75), Barry bargained for 2200
• Locked, Returned and re-returned for 2200. Driver’s narrative
• There is always a larger game!

Oligopoly
15
Oligopoly – An example
• 2 chocolate stores – East (ES) & West (WS) at few minutes distance
• Advertising same brand chocolate for 500
• Scenario 1: E cuts price to 450 while the W price stays at 500
• E will attract some W customers. If the increase in sales is large

enough, E’s profits would increase.
16
Oligopoly – An example (contd.)
• Scenario 2: E cuts its advertised price to 450 while the W price
stays at 500 with a promise to match any lower price.
• Those who find W more convenient can just bring in the E ad and
get the discounted price. Then, E attracts no new customers from
its price cut. In fact, it loses revenue.
17
Market characteristics
Characteristic Competition Monopoly Oligopoly
No. of firms Many One Intermediate
Entry barrier Negligible Extreme Intermediate
Influence on price Taker Maker Intermediate
Profits Normal Abnormal Intermediate
Interdependence Not required Not applicable Low to High
Non-price factors Not required Not applicable Low to High
18
Unique features of oligopoly
• Large possibilities – Opportunity cost requires ‘strategizing’
• For what – Profits? Costs? Revenue? Market power?
• How – Quantity? Price? Following? Simultaneous?
• When – Conducive time? Foresight – Term? Attributes?
• To what – Competition? Cooperation? Mix?

19
Forms of Oligopoly
• Quantity leadership or sequential quantity setting (Stackelberg)
• Price leadership or sequential price setting
• Simultaneous quantity setting (Cournot)
• Simultaneous price setting (Bertrand)

20
Quantity leadership (Stackelberg)
• Leader – Sets the quantity it wants to supply
• Follower – Takes leader’s quantity given, then chooses quantity

that maximizes its profits
• Works in reverse – Follower has a ‘best response/reaction function’.

Leader endogenises follower’s reaction function to decide supply.
• Applications – Capacity setup, Supply chains, Marketing

21
Price leadership
• Leader – Sets price at which it wants to supply
• Follower – Takes leader’s price given, then chooses a quantity
• Works in reverse – Follower faces ‘residual demand curve’. Leader

endogenises follower’s supply to decide on its quantity supply at
the chosen price.
• Practice – Price menu, airlines

22
Simultaneous Quantity Setting (Cournot)
• Based on forecast of other firms’ quantity supply in the industry
• The forecasts need not be the equilibrium quantity choice
• In practice, we often note a ‘multi-degree’ play with forecast value

as starting point
• Application – Software industry

23
Simultaneous Price Setting (Bertrand)
• Based on forecast of other firms supply price in the industry
• But a price higher than marginal cost cannot be equilibrium. Stable

equilibrium is competitive equilibrium.
• Paradoxical: Competitive price with only two firms in the market?
• Practice: Competitive bidding

24
Collusion/Cartel
• Under collusion, firms choose the output that maximizes total
industry profits and then divide up the profits.
• Such a collusion behaves like a monopolist. There is an important

distinction though.
• There is always a temptation to cheat.

25
Basic comparison of forms of oligopoly
Form Price Quantity
Bertrand Lowest Highest
Stackelberg
Price leadership Intermediate
Cournot
Collusion Highest Lowest
26
Sustaining a collusion
• Carrots or sticks (Carrots latter – time to think!)
• If you do not cheat, i.e., maximizes joint industry profits, fine.
• But if I discover cheating, will punish by producing Cournot level.
• How to detect cheating?
• Meet or beat pricing!

Games
28
Example: Voluntary Export Restraints
• 1985-6, Japanese auto cos. ‘voluntarily’ reduce export to US.
• US opinion – victory for US trade negotiators (may create jobs)
• Estimates – Japanese cars in US were $2500 more expensive relative

to what they have been without restraint
• Allowed US producers to sell their cars at $1000 more

29
Example: Voluntary Export Restraints
• US consumers paid $10 billion more for Japanese cars than would
have otherwise.
• This extra profit allowed the Japanese auto cos. gain efficiency.
• The cost per job saved to US was about $160,000 per year.
• A $2500 tariff on imported Japanese car may have worked better!

30
Features of a game
• Idea of ‘Strategizing’ goes well beyond ‘formal’ notion of markets
• Three key elements of a game – players, strategies, payoffs.
• Bloodline – Information
• Two basic types – non-cooperative, cooperative
• Early days of Games at Princeton

31
Game representation
• Two ways of representing games – Normal form, Extensive form
• Normal form – Tabular, Apt for simultaneous games
• Extensive form – Tree, Apt for sequential games
• Forms of strategies – Pure (degenerate prob.), Mixed (non-

degenerate prob.)
32
Prisoners’ dilemma
Player 2
Do not confess
Confess (Defect)
(Cooperate)
Do not confess
-1, -1 -3, 0
Player 1
(Cooperate)
Confess (Defect) 0, -3 -2, -2
Multiplying each payoff by a scalar or/and adding a constant,

retains prisoners’ dilemma
33
Solution concept – Dominant strategy
Strategy for a player that produces the highest payoff for any
strategy available to the other players.
Player 2
Do not confess
Confess (Defect)
(Cooperate)
Do not confess
-1, -1 -3, 0
Player 1
(Cooperate)
Confess (Defect) 0, -3 -2, -2
Individual incentives & overall welfare need not coincide

34
Cournot-Cartel game
Player 2
Collude
Compete (Defect)
(Cooperate)
Collude
4, 4 0, 6
Player 1
(Cooperate)
Compete (Defect) 6, 0 2, 2
Form of prisoners’ dilemma – Doubled and added six

35
Advertising Game
• Total available profits are 28, divided as 16-12. Advertising costs 8.
• If both advertise, they will split the market evenly.
• If one advertises while other does not, then advertiser captures

three-quarters. Non-advertiser the rest.
36
Advertising Game
Player 2
Not adv
Advertise (Defect)
(Cooperate)
Not adv
16, 12 7, 13
Player 1
(Cooperate)
Advertise (Defect) 13, 7 6, 6
Player 2 does not have a dominant strategy

37
Nash equilibrium
• Profile of strategies such that each player's strategy is a best
response (results in the highest available payoff) against the
equilibrium strategies of the other players.
38
Advertising Game
Player 2
Not adv
Advertise (Defect)
(Cooperate)
Not adv
16, 12 7, 13
Player 1
(Cooperate)
Advertise (Defect) 13, 7 6, 6

39
Stag hunt game
Player 2

Hunt stag Hunt hare
Hunt stag 5, 5 0, 3
Player 1
Hunt hare 3, 0 4, 4
40
Stag hunt game
• With credible commitment, hunting stag is stable and reasonable
equilibrium – highest payoff
• Hunting hare is useful as it is less risky – minimum payoff of 3 as

against 0.
• Other examples of this kind of game – selecting between two

technologies, coordinating on a meeting location.
41
Battle of sexes
Player 2

Opera Movie
Opera 3, 1 0, 0
Player 1
Movie 0, 0 1, 3
The players have an incentive to choose the same action,

but they each have a different favorite action.
42
Hawk-Dove game
Player 2

Hawk Dove
Hawk 0, 0 3, 1
Player 1
Dove 1, 3 0, 0
Anti-coordination game. E.g., Chicken, HR

43 Penalty-Kick or matching coin or zero-sum

game
Goalie

Left Right
Left -1, 1 1, -1
Kicker
Right 1, -1 -1, 1
No Nash - Use randomized or mixed strategy – 0.5

probability of kicking on either side
44
Extensive form game – Example
1
Subgame perfect Nash
L R
2 3
L R L R
3 1 0 2
3 4 0 2
3 1 0 2
45
Advertisement game
1
Adv No Adv
2 2
Adv No Adv Adv No adv

6 13 7 16
6 7 13 12
46
Penalty-Kick game
Kicker
L R
Gaolie Goalie
L R L R
-1 1 1 -1
1 -1 -1 1
Moving first is not always advantageous

47
Behavioural games
Four Stories
48
Story #1: Ultimatum bargaining
• Photograph in Juhu – INR 300
• Case – Divide INR 100
• Fairness concerns – Ref dependence, loss aversion, -ve reciprocity
• Experiments – $ 10/100/400 – 3.5 OECD, 1.5 Africa-Amazon-Asia
• Applications – Failed love affairs/marriages, Renegotiation in

‘nuisance’ cases, Labour strike
49
Story #2: Continental divide game
• Where to have location of ‘new media firms – near Hollywood (S)
or Silicon Valley (N) – who is central?
• Akin to – pick between 1 (Hollywood) to 14 (Silicon Valley)
• Payoff depends on the ‘median’
• Strategy – Pick high (low) if other will pick high (low). Unsure,
then pick middle.
50
MEDIAN CHOICE
1 2 3 4 5 6 7 8 9 10 11 12 13 14
1 45 49 52 55 56 55 46 -59 -88 -105 -117 -127 -135 -142
2 48 53 58 62 65 66 61 -27 -52 -67 -77 -86 -92 -98
3 48 54 60 66 70 74 72 1 -20 -32 -41 -48 -53 -58
4 43 51 58 65 71 77 80 26 8 -2 -9 -14 -19 -22
5 35 44 52 60 69 77 83 46 32 25 19 15 12 10
6 23 33 42 52 62 72 82 62 53 47 43 41 39 38
CHOICE
7 7 18 28 40 51 64 78 75 69 66 64 63 62 62
8 -13 -1 11 23 37 51 69 83 81 80 80 80 81 82
9 -37 -24 -11 3 18 35 57 88 89 91 92 94 96 98
10 -65 -51 -37 -21 -4 15 40 89 94 98 101 104 107 110
11 -97 -82 -66 -49 -31 -9 20 85 94 100 105 110 114 119
12 -133 -117 -100 -82 -61 -37 -5 78 91 99 106 112 118 123
13 -173 -156 -137 -118 -96 -69 -33 67 83 94 103 110 117 123
14 -217 -198 -179 -158 -134 -105 -65 52 72 85 95 104 112 120
51
• Equilibrium – choose 3 (low risk/return) or 12 (high risk/return).
• Continental divide game – Alaska (Arctic & Pacific Ocean divide).

Extreme sensitivity to initial condition
• Application – Pol./Office Elections

52 Story #3: Beauty contest and iterated

dominance
• Stock market – After a boom we guess what is others’ guess of it as
a speculative bubble.
• Keynes’ Beauty contest – Pick [0, 100]. Multiply average by 0.7.

person closest to this (0.7 of average) wins.
• How will you play?

53 Story #3: Beauty contest and iterated

dominance
• Think of degrees to which you go. If average 50, pick 35, then pick
~25, then pick ~17.5, …… then pick 0.
• Similar to Cournot duopoly
• Practice: People perform only a couple of iterations
• Caltech board game – Winner picked 18.1 (Most (2-3 steps), few
(exact 17.5), few (higher))
54
Story #4: Resolving prisoners’ dilemma
• Mechanism design – Carrots and sticks!
• Internal rewards – One pays the other for cooperation
• External reward – Third party rewards the two (US & EU on

conflicts in Israel & Egypt)
• Reward always after the choices are made!

55
• Mechanism design – Carrots and sticks!
• Punishment – ‘LA Confidential’ Cooperate with two others
• Baseball – American league (hit by pitchers 17% more) vs. National

league (Relationship)
56
• Axelrod’s experiment – Strategy programs for Chess (150 times)
• TIT for TAT won – clarity, niceness, provocability, forgiveness.
• Not suitable for ‘winner takes it all’ and defection

57
Prisoners’ dilemma in business – 2 cases
• Collusion is easier with ‘regular’ communication
• Case 1: GE (60), Westinghouse (30), Allied-Chalmers (10) 1950s –

Electric turbines – Bids – Lunar calendar (1-17, 18-25, 26-28)
• Case 2: Airwave spectrum auction – Firm wanting bid award – use

last 3 digits of pincode.
58
Meet or beat the price – revisited
• Reason for not allowing collusion – consumer protection – firms
decide to compete
• We saw ‘meet or beat price’ – Third degree price discrimination

59
Some concluding games
• Human resource attrition – Self ‘control’, nudge and ‘talking
points’ [Carrot]
• Client relationship – Dedicated service and speed-quality signal

[Division of surplus]
60
Some concluding games
• Present vs. future relevant skills – ‘Risk’ and ‘niche’ journey –
Make/understand the game than play it [Multi-degree mobility]
• Government vs. research: Data suppression [Insider game vs.

outsider game, optimism, myopia expectations]

Oligopoly and Games

Uploaded by

Copyright:

Available Formats

Oligopoly and Games

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Oligopoly and Games

Uploaded by

Copyright:

Available Formats

What are some examples of strategic games discussed in the document?

What are some examples of strategic games discussed in the document?

How can the prisoner's dilemma be resolved according to the document?

How can the prisoner's dilemma be resolved according to the document?

Oligopoly and Games:

• Systems approach – Setting up a ‘reasonable’ boundary!

• Rational player vs. behavioral player (= rational + biases) setup

• Typical sources of bias – Reference dependence, loss aversion,

• Subtle place for – customs, ethics, etiquette and strategy

• Some strategic decisions may not confirm to ‘moral’ concerns

• Better to be a good strategist than a bad one!

• Reward – 100, 80, 60, 40, 20.

• Clues – High/Low; Discuss my strategy after 4th

• Final -1 round: Richard (39), Rudy (72), Kelly (23)

• Prize USD 1mn, Winner picks opp. in final, Contestants vote

• Pole & immunity idol challenge

• Psychologists – Long flips can have a spell of continuous heads

• Experiments in Hockey/Basketball and several other games

• Hot hand ~ warm game

• Team streaks, Maradona’s pass in 1986 (2 goals), Ice hockey

• Australia changed sail to right, US kept left. Australia won 4-3.

• Charles de Gaulle’s negotiation with Roosevelt & Churchill

• Ferdinand de Lesseps –Nile sand (Suez), tropical malaria (Panama)

• Two aspects – Intertemporal concerns, degree of intransigence

• Cortés (Spain – troop of 600) vs. Aztecs (Mexico) – Scuttled ships

• Xenophon (Athens) vs. Sparta – Back against ravine

• Cindy/Laurie/Ray-wife & ABC Primetime – 17/15+13/22 pounds

• Too many options are not good always!

• What to do (Buffett) – Opp. pay 1 bn to supporters if opp. succeeds.

• Prisoners’ dilemma – Bell cat’s neck, Khrushchev on Stalin, Hijack

• Sotheby & Christie offered auction @ 12 %. Hash suggested R-P-S

• Applications – Formula based tax audit, Repeated games

• Biggest looser – Hash

• Poker – Others to fold when you bluff, no high-pot victories

– Others to raise when you are genuine, caught bluffing

• Winning an auction – winner’s curse

• Upon reaching – 2500 shekels ($2.75), Barry bargained for 2200

• Locked, Returned and re-returned for 2200. Driver’s narrative

• There is always a larger game!

• Advertising same brand chocolate for 500

• Scenario 1: E cuts price to 450 while the W price stays at 500

• E will attract some W customers. If the increase in sales is large

• For what – Profits? Costs? Revenue? Market power?

• How – Quantity? Price? Following? Simultaneous?

• When – Conducive time? Foresight – Term? Attributes?

• To what – Competition? Cooperation? Mix?

• Price leadership or sequential price setting

• Simultaneous quantity setting (Cournot)

• Simultaneous price setting (Bertrand)

• Follower – Takes leader’s quantity given, then chooses quantity

• Works in reverse – Follower has a ‘best response/reaction function’.

• Applications – Capacity setup, Supply chains, Marketing

• Follower – Takes leader’s price given, then chooses a quantity

• Works in reverse – Follower faces ‘residual demand curve’. Leader

• Practice – Price menu, airlines

• The forecasts need not be the equilibrium quantity choice

• In practice, we often note a ‘multi-degree’ play with forecast value

• Application – Software industry

• But a price higher than marginal cost cannot be equilibrium. Stable