SF2972 GAME THEORY

Lecture 1: Introduction and simple

examples

Jörgen Weibull

22 January 2015
1 What is game theory?

A mathematically formalized theory of strategic interaction between

– countries at war and peace, in federations and international negotiations

– political candidates and parties competing for power

– firms in markets, owners and managers, employers and trade-unions

– members of communities with a common pool of resources

– family members and generations who care about each other’s well-being

– animals within the same species, from different species, plants, cells

– agents in networks: computers, cell phones, vehicles in traffic systems

2 A brief history of game theory

• Emile Borel (1920s): Small two-player zero-sum games

• John von Neumann (1928): Two player zero-sum games of arbitrary

size. Maximin Theorem

• von Neumann & Morgenstern (1944): Games and Economic Behavior,

The Expected Utility Hypothesis

• John Nash (1950s): Non-cooperative equilibrium [Film: “A Beautiful

Mind”]

• John Harsanyi (1960s-1980s): Incomplete information

• Reinhard Selten (1970s-today): Rationality as the limit of bounded
rationality

• John Maynard Smith (1970s-1990s): Evolutionary stability of behaviors

in large populations

• Robert Aumann (1959-today), Thomas Schelling (1960-1990s), and

many current game theorists
3 Two interpretations

• In 1950 John Nash (then Ph D student at Princeton Math Dept)

invented normal-form games and an equilibrium concept for these,
Nash equilibrium.

— A normal-form game specifies a strategy set and a payoff function

for each player, where each such function’s domain is the set of all
strategy profiles.

— A Nash equilibrium is a strategy profile such that no player can

increase his or her payoff by a unilateral deviation. Equivalently: a
Nash equilibrium is a strategy profile that is a best reply to itself.
 John Nash

(born 1928, PhD 1950)

Nash suggested two interpretations:

1. The rationalistic (or epistemic) interpretation [the by far most well-

known and the foundation of modern economics]

2. The ”mass action” (or evolutionary) interpretation [discovered in 1994

and more in line with sociology, biology and computer science]
3.1 The rationalistic interpretation

1. The players have never interacted before and will never interact in the
future

2. The players are rational [Nowadays usually interpreted in the sense of

Savage, The Foundations of Statistics (1954)]

3. Each player knows the game in question (knows all players’ strategy
sets and payoff functions)
• This does not imply that they will play a Nash equilibrium (For example,
player i may believe that player j does not know i’s payoffs)

• One nowadays assumes common knowledge (in the sense of Lewis,

Convention: A Philosophical Study (1969), and as formalized in Au-
mann, Agreeing to Disagree (1976)] of the game and of all players’
rationality
 A B
Coordination game: A 2, 2 0, 0
B 0, 0 1, 1

H T
Zero-sum game: H 1, −1 −1, 1
T −1, 1 1, −1

L C R
T 7, 0 2, 5 0, 7
Game with a unique NE:
M 5, 2 3, 3 5, 2
B 0, 7 2, 5 7, 0
3.2 The mass-action interpretation

1. For each player role in the game: a large population of identical indi-
viduals

2. The game is recurrently played, in time periods t = 0, 1, 2, 3, ... by

randomly drawn individuals, one from each player population

3. Individuals learn from experience (own and/or others) to avoid subop-

timal actions

• A mixed strategy for a player role is a statistical distribution over the

actions available in that role
• If all individuals avoid suboptimal actions, and the population distrib-
ution of action profiles is stationary, then it constitutes a Nash equi-
librium

• Reconsider the above examples in this interpretation!

4 More examples

4.1 Hawk-dove game

• Start-up business with two partners, or a pair of students who are to

write a joint thesis

• To work or shirk?
W S
W 3, 3 0, 4
S 4, 0 −1, −1

• Nash equilibrium? Your prediction under each prediction?

4.2 Prisoners’ dilemma

• Two fishing companies, fishing in the same North Sea area

• Each company can either fish modestly, a, or aggressively, b. The

profits are
a b
a 3, 3 1, 4
b 4, 1 2, 2

• If each company strives to maximize its profit, (b, b) will result (ir-
respective of what they believe about each other’s goal function or
rationality)

• Would monopoly be better? An agreement on fishing quota?

• The First Welfare Theorem does not hold: Competition leads to over-
exploitation

• What if the interaction is repeated over time? One hundred periods?

Infinitely many periods?
4.3 An even worse welfare failure

• Traffic (or network) equilibrium under congestion. Infinitely many play-

ers, a continuum, and nevertheless equilibrium may results in the worst
possible outcome

— All wish to travel from location A to location B. There are two

roads from A to B.

— Road 1 takes one hour, independent of traffic. Road 2 takes longer

the more traffic there is. If everybody takes road 2, it takes 1 hour.

— Each individual wishes to minimize his or her travel time. They

make their choices simultaneously and leave A at the same time.

— Define this as a game. What will they do in Nash equilibrium? Is

there any worse outcome, in terms of total travel time?
• Suppose the travel time on road 2 is x when the population share x
takes this road (for any given x ∈ [0, 1]).

— In Nash equilibrium: x = 1.

— What x-value would minimize total travel time?

T (x) = x · x + (1 − x) · 1
4.4 Cournot market competition

• n firms competing in a homogeneous product market

Stage 1: Simultaneous choice of output levels q1, q2, ..., qn ≥ 0. Aggregate

supply Q = q1 + ...qn

Stage 2: Market clearing price p, determined by D (p) = Q, where D (p)

is demand at price p
 100
D(p), Q
90

80

70

60

50

40

30

20

10

0
0 10 20 30 40 50 60 70 80 90 100
p
• How represent this as a game? Players? Strategy sets? Payoff func-
tions?

• Suppose you are the manager of firm i, that all firms have the same
production costs, and it is common knowledge among the managers
that they all are rational profit maximizers and know the game

• What level of output, qi, would you choose if you were the manager
of firm i?

• Consider the special case when n = 2, demand is linear, constant unit

cost, c, of production!
5 Strategic interactions over time

In many, if not most, real-life interactions there is a time element; players

may move after each other, acquire information etc. Then a strategy is a
complex object, a ”rule” that specifies what to do in any contingency that
may arise. Hence, strategy sets may be very big.
5.1 Example: Sequential Cournot duopoly

• Suppose that the previous Cournot duopoly example takes place se-
quentially:

— Stage 1: Firm 1 chooses q1

— Stage 2: Firm 2 observes q1 and then chooses q2

— Stage 3: The market clears, D (p) = q1 + q2

• What are now the strategy sets of the two players? [Player 2 may now
condition her quantity choice on 1’s quantity choice]
• If you were manager of firm 1, how would you reason and what would
you do?

• If you were manager of firm 2, how would you reason and what would
you do?

• Will firm 1 (2) earn a higher or equal profit than in the case of simul-
taneous moves? Or the same profit as before?

• Such sequential Cournot duopoly games are called Stackelberg leader-

ship games
6 Next lecture

• We will make precise the notions of normal-form games and Nash

equilibrium, and we will establish a general existence theorem for such
equilibria, and consider examples

• In that lecture, you are supposed to know the following properties of

sets in Euclidean spaces: finite, open, closed, bounded, compact and
convex. And you need to know what is a continuous function

• Please read ahead in the text-book!