SUSTAINABLE STRATEGIC ANALYSIS

Lecture 3

Luu Duc Toan Huynh

[Link]
Semester 2023/24
Reading
You should read at least two sources from each list below

Structure

LT3.t1: Games

LT3.t2: Nash equilibrium

The Economy (Unit 4), Industrial Organization (Chapter 4), Competitive Solutions (Chapter 3)
LT3.t3: Prisoner's Dilemma

LT3.t4: Social Preferences

LT3.t5: Coordination Industrial Organization (Chapter 7), The Economy (Unit 11.4), Competitive Solutions (Chapter 2)

LT3.t6: Cournot
Champagne industry: Financial Times
Game theory
Key concepts

Social and Strategic Interactions

Structure • Social interaction: A situation involving more than one
person/party, where one’s actions affect both their own
LT3.t1: Games
and other people’s outcomes.
LT3.t2: Nash equilibrium
• Strategic interaction: A social interaction where people
LT3.t3: Prisoner's are aware of the ways that their actions affect others.
Dilemma
• Strategy: Action(s) that people can take when engaging
LT3.t4: Social Preferences in a social interaction.
LT3.t5: Coordination From The Economy

LT3.t6: Cournot
Game theory
Key concepts

Bala
▪ Two farmers decide which crop to specialize in. Rice Cassava
▪ The interact only once (one-shot game).
3 2

Rice
A game describes a social interaction:
1 2

Anil
1. Players – who is involved in the interaction

Cassava
2. Feasible strategies – actions each player can take 4 1
3. Information – what each player knows when choosing their action 4 3
4. Payoffs – outcomes for every possible combination of actions

From The Economy

Optimal decision-making
Introduction

Structure
• Best response: Strategy that yields the highest payoff,
LT3.t1: Games
given the other player’s strategy.
LT3.t2: Nash equilibrium
• Dominant strategy: A best response to all possible
LT3.t3: Prisoner's Dilemma
strategies of the other player (does not always exist!).
LT3.t4: Social Preferences
• Dominant strategy equilibrium: An outcome of a game
LT3.t5: Coordination
in which everyone plays their dominant strategy.
LT3.t6: Cournot
From The Economy
Optimal decision-making
Crop choice example

• Best response: If Bala grows rice, Anil’s best

Bala
response is to grow cassava. If Bala grows
cassava, Anil’s best response is to grow Rice Cassava
cassava.
3 2

Rice
• Dominant strategy: Anil’s dominant strategy
is to grow cassava. Bala’s dominant strategy 1 2

Anil
is to grow rice.

Cassava
4 1
• Dominant strategy equilibrium: When Anil
and Bala each play their dominant strategy, 4 3
the outcome is (Cassava, Rice).
From The Economy
Optimal decision-making
Nash equilibrium

Bala
Nash equilibrium: A set of strategies (one per
player), such that each player’s strategy is the Rice Cassava
best response to the strategies chosen by
everyone else. 1 2

Rice
In a Nash equilibrium, no player has an incentive 0 2

Anil
to deviate unilaterally.

Cassava
4 0
Note: There may be more than one Nash
equilibrium in a game. 4 1

From The Economy

Optimal decision-making
Resolving social dilemmas

Structure Prisoner’s Dilemma

A game with a dominant strategy equilibrium, in which
LT3.t1: Games
playing the dominant strategy yields lower individual and
LT3.t2: Nash equilibrium total payoffs compared to other strategies
LT3.t3: Prisoner's Dilemma
Socially optimal outcome is not achieved
LT3.t4: Social Preferences
Pesticide example: Both farmers choose to use the more
LT3.t5: Coordination
harmful pest control (T).
LT3.t6: Cournot
From The Economy
Optimal decision-making
Resolving social dilemmas

Bala
Structure
IPC Terminator
LT3.t1: Games

LT3.t2: Nash equilibrium 3 4

IPC
LT3.t3: Prisoner's Dilemma 3 1

Anil
LT3.t4: Social Preferences

Terminator
1 2
LT3.t5: Coordination

LT3.t6: Cournot
4 2
From The Economy
Optimal decision-making
Why did we predict this outcome?

1. Players only care about their own payoffs.

Introduce social preferences

2. Nobody could make players pay for the consequences of their actions on others.

Introduce repeated games, social norms, and peer punishment

3. Players could not coordinate their actions beforehand

Change the rules of the game (institutions and policies)

From The Economy
Optimal decision-making
Indifference curves

A E F G H D

Hours of free time 15 16 17 18 19 20

100
Final grade 84 75 67 60 54 50
A
84
Marginal rate of substitution between grade and free time (MRS) 9 8 7 6 4 75 E

H
54 D
50
Choices depend on preferences

Final grade
Indifference curves show all combinations of goods that give the
same utility (satisfaction)

The marginal rate of substitution is the slope of the indifference

curve, and represents the tradeoffs that an individual faces 15 16 19 20 24
0
Hours of free time per day

From The Economy

Optimal decision-making From The Economy

The Feasible Frontier

The feasible frontier shows the maximum 100

output that can be achieved with a given A
90 E
amount of input
80
D B

Final exam grade

70
The marginal rate of transformation (MRT) C
60
is the slope of the feasible frontier, and
F
represents the tradeoffs that an individual 50

faces 40

30 Feasible set
20

A E C F 10
Hours of free time 13 14 19 20

Final grade 84 81 57 50 … 13 14 … 19 20 24

Opportunity cost 3 7 Hours of free time per day

Optimal decision-making From The Economy

Utility-maximising choice

Feasible frontier
The utility-maximising choice is where the 90
B
D

amount of one good the individual is

Final exam grade

willing to trade off for the other good
57 IC3
E
(MRS) equals the actual tradeoff between IC3
C
the two goods (MRT). A IC2

IC1

MRS = MRT

… 13 14 … 19 20 24

Hours of free time per day

Social Preferences
Resolving social dilemmas

Prisoner’s Dilemma
Bala
Structure A game with a dominant
strategy equilibrium, in which IPC Terminator
LT3.t1: Games playing the dominant strategy
yields lower individual and
LT3.t2: Nash equilibrium total payoffs compared to 3 4

IPC
other strategies
LT3.t3: Prisoner's Dilemma 3 1

Anil
Socially optimal outcome is
LT3.t4: Social Preferences not achieved

Terminator
1 2
LT3.t5: Coordination Pesticide example: Both

LT3.t6: Cournot
farmers choose to use the 4 2
more harmful pest control (T).

From The Economy

Social Preferences 12 Feasible
payoffs frontier Anil’s indifference curves
Social preferences: Altruism 11 (when somewhat altruistic)
10 Anil’s indifference
curves (when
9

Bala’s payoff (thousands of rupees)

completely selfish)

• Social dilemmas arise when players 7

only care about their own payoffs. 6

C
5
• However, in experiments, many 4
players show altruism by choosing 3
B

the dominated strategy.

2
Feasible
• Altruistic preferences affect the 1 payoffs set
A
shape of indifference curves.
1 2 3 4 5 6 7 8 9 10 11 12

Anil’s payoff (thousands of rupees)

From The Economy

Social Preferences Bala
Resolving the Prisoners’ Dilemma

IPC Terminator
5 Anil’s indifference curves
(when completely selfish) 3 4

IPC
I,T
4
Bala’s payoff

I,I 3 1

Anil
3 Anil’s indifference curves

Terminator
(when somewhat altruistic)
1 2
2
T,T

1 T,I 4 2

1 2 3 4 5
Anil’s payoff From The Economy
The Ultimatum Game Proposer

Bargaining

A sequential game where players

choose how to divide up economic Offer Offer
(50,50) (80,20)
rents (for example, cash prize).

The proposer’s offer may be

motivated by altruism, fairness (50-50
Proposer Proposer
split), inequality aversion, social
norms, or reciprocity.

Accept Reject Accept Reject

(50,50) (0,0) (80,20) (0,0)

From The Economy

The Ultimatum Game Questions:
Example: Kenyan farmers and US students • Were the farmers really generous?

Key findings:
• Better offers, fewer rejections.
• Kenyans are more likely to reject low offers
than Americans.
• Just over 50% of Kenyan farmers rejected
the offer of the Proposer keeping 30%.
• Both groups of Responders are indifferent
between accepting and rejecting an offer of
receiving nothing.
• Kenyan farmers place higher importance on
fairness than US students

Adapted from Joseph Henrich, Richard McElreath, Abigail Barr, Jean Ensminger, ClarkBarrett, Alexander Bolyanatz, Juan Camilo Cardenas, Michael Gurven, Edwins Gwako, Natalie Henrich, Carolyn
Lesorogol, FrankMarlowe, David Tracer, and John Ziker. 2006. ‘Costly Punishment Across Human Societies’. Science 312 (5781): pp. 1767–1770.
From The Economy
The Ultimatum Game
Example: The rules of the game matter

• When there is competition,

Responders are less likely to reject
low offers.

• Introducing competition between

responders moves outcomes closer
to the self-interested outcome

From The Economy

Adapted from Figure 6 in Urs Fischbacher, Christina M. Fong, and Ernst Fehr. 2009. ‘Fairness, Errors and the Power of Competition’. Journal of Economic
Behaviour & Organization 72 (1): pp. 527–45.
The Ultimatum Game
Social preferences: Other types

• Inequality aversion: Disliking outcomes in which some individuals receive more than others

• Reciprocity: Being kind/helpful to others who are kind/helpful, and vice versa.

• We evaluate whether others have been ‘kind’ or ‘helpful’ according to social norms (common
understanding of how to act in situations when one’s actions affect others).

These motives affect outcomes in the public goods game and the ultimatum game.

From The Economy

Coordination
Introduction
Bala
Structure When there is more than
one Nash equilibrium and Rice Cassava
LT3.t1: Games individuals choose
independently, the socially 1 2

Rice
LT3.t2: Nash equilibrium optimal outcome may not be
selected.
0 2
LT3.t3: Prisoner's

Anil
Dilemma Society could be “stuck” in a
suboptimal outcome since 4 0

Cassava
LT3.t4: Social Preferences there is no incentive to
unilaterally change one’s 4 1
action.
LT3.t5: Coordination

LT3.t6: Cournot From The Economy

Coordination
Example: Climate change - Outcomes depend on rules and preferences
The US

• Business as usual (BAU) is the dominant Restrict BAU

strategy for purely self-interested players
Best Good

Restrict
• An emissions treaty would produce the
socially optimal outcome Best Worst

China
• Inequality aversion and reciprocity will result Worst Worst

BAU
in two Nash equilibria here which include the
social optimum Good Worst

From The Economy

Summary
Example: Climate change - Outcomes depend on rules and preferences

1. Social interactions can be modeled as games.

Players choose best responses to others’ strategies.

2. Social dilemmas (i.e., prisoners dilemma) can be resolved by social preferences, peer
punishment, or binding agreements.
The rules of the game also matter for outcomes.

3. Multiple Nash equilibria can cause coordination problems.

Economic and political institutions can help achieve socially optimal outcomes.

From The Economy

Cournot competition
Zero-Sum Game

Structure Zero sum game Nonzero sum game

LT3.t1: Games Game where the sum of all Game where the sum of all
payoffs is constant payoffs is dependent on the
LT3.t2: Nash equilibrium actions of the players

LT3.t3: Prisoner's Dilemma Examples

Examples
▪ Chess
LT3.t4: Social Preferences ▪ Games with cooperation,
▪ Splitting a one-off joint profit e.g., standard setting
LT3.t5: Coordination ▪ Football World cup Finals ▪ Prisoner’s dilemma

LT3.t6: Cournot
Five Duopolistic Market Situations
Introduction

Simultaneous decisions Sequential decisions

(both firms decide at once) (leader decides first, then follower)

Quantity strategies
(e.g., in markets with inflexible  Cournot-Nash Game  Stackelberg Game
capacities and high inventory costs)

Price strategies
 Bertrand Game  Price-Leadership Game
(e.g., in markets with flexible capacities)

 Collusion
Formal cooperation (cartel) or informal/tacit cooperation between
companies about prices or output levels aiming for monopoly
profits.
Cournot Game Saudi Arabia
Concept (1/4)
Cartel
agreement Low High

Example: Market for crude oil 337.5 379

Low
• Players: Iran and Saudi Arabia (assumption: same
cost structure and oil reserves)
337.5 263

Iran
• Assumptions: Same oil reserves, zero costs

263 300

High
Competition / Fixed quantities

• Strategies: Oil production high (H) or low (L) 379 300

Nash
equilibrium
Cournot Game
Concept (2/4) - Competition/ Continuous quantities

Strategies: Oil production q mn barrels/day

Market price: p = 30 - 1/3 (qI + qS)
p
Profits: S = qS p = qS [30 - 1/3 (qI + qS)] 30
p = 30 - 1/3 (qI + qS)
I = qI p = qI [30 - 1/3 (qI + qS)]

Best responses: S / qS = 0 → RS(qI) = 45 - 1/2 qI qI+qS

I / qI = 0 → RI(qS) = 45 - 1/2 qS

Nash equilibrium: For RS(qI) = qS and RI(qS) = qI

qS = 45 - 1/2 [45 - 1/2 qS ] → qs = 30
qI = 45 - 1/2 [45 - 1/2 qI ] → qI = 30

Payoff per country: S,NG = I,NG = 300

Cournot Game
Concept (3/4) - Cartel/ Continuous quantities

Strategies: Oil production q mn barrels/day

Market price: p = 30 - 1/3 Q

Total payoff: CART = Q p = Q (30 – 1/3 Q)

Optimal total output: CART/ Q = 0 → Q = ?

Payoff per country: CART/2 = ?

Cournot Game

q2
𝑀𝑎𝑥 𝜋𝑖 𝑄 = 𝑎 − 𝑏 𝑞𝑖 + 𝑞𝑗 − 𝑐 𝑞𝑖
RF1
𝑎 − 𝑏𝑞𝑗 − 𝑐
𝑞𝑖 =
2𝑏

C
𝑞2𝐶

RF2

𝑞1𝐶 𝑎−𝑐 𝑎−𝑐 q1

2𝑏 𝑏
Cournot Game Cournot Cartel
Concept (4/4) - Comparison
qI (Output Iran) Output per country 30 > 22.5

90 Market price 10 < 15

Payoff per country 300 < 337.5

RS(qI)
(Reaction function SA)

45 Cournot

Cartel RI(qS)
Quantities qI, qS (Reaction function Iran)
reduced by 25%
qS (Output Saudi Arabia)
22.5 30 45 90