Managerial Economics
Lec 1: Game theory and competition
Game theory: fundamentals
Agents:
- Rational preferences
Complete/perfect Information Asymmetric information
Static Strategic games Bayesian games
Dynam Extensive games, repeated Sequential (Bayesian) games
ic games
- Payoff maximizing
Main categories of games:
- Strategic
o Static
One shot
Simultaneous choice
o Complete information
All players know capabilities (available strategies) other
players
All players know consequences (payoffs) of every combination
strategies
!! does not mean certainty (probability distributions)
o Fully described by
Set of players
Set of actions (action profiles gebruiken als je een nash ofz
weergeeft)
Payoff functions (2 players: payoff matrix = normal form)
o Strategies:
X strictly dominated by Y
X heeft in alle gevallen een lagere payoff dan Y
X nooit spelen, dus verwijderen uit game
X is a strictly dominant strategy
X altijd beste optie
Altijd X spelen
o Equilibrium
In dominant strategies
Als beide een dominante strategie volgen
Alleen rationaliteit hier
Is ook een NE
Nash
1
,
Ze kunnen niet positief deviaten, elk speelt zijn beste
response op wat de andere doet
Niet per se een optimum, eerder een resting point
Niet altijd een te vinden
Soms meer dan een (players might disagree on which
is best)
Niet altijd een te vinden in pure strategies, wel altijd in mixed
strategies (matching pennies)
- Extensive /repeated games
o Dynamic
o Complete information
- Bayesian
o Static
o Asymmetric information
- Sequential
o Dynamic
o Asymmetric information
Demand, monopoly and competition
Market demand
- Price elasticity:
Δq
q dq ( p ) p
o ε ( p)= = ×
Δp dp q ( p)
p
o |ε|=1
Als de prijs dealt met x% zal de hoeveelheid
toenemen/afnemen met x%
o |ε|>1
Elastic
Increasing price reduces revenue
o |ε|<1
Inelastic
Increasing price increases revenue
Perfect competition
- Does not exist, archetype
- Consumers willing to buy any quantity, but only at price p (or lower)
price infinitely elastic
- Price taking firm
o Single-agent problem
Each firm too small to affect demand (no strategic
interactions)
Can produce q with cost C(q)
Choose quantity to maximize profit
Profit = p*q – C(q) afleiden naar q
P = C’(q) = marginale kost van q-de eenheid
IF price taking + free entry = zero profits
2
, Monopolist
- Single agent problem
o Observes downward demand slope q(p) and can set the price
o Choose quantity or price to maximize profit
Profit = p*q(p) – C(q(p))
Pm = (A+bc)/2b
In elastische deel van de vraag
Imperfect competition: Bertrand duopoly
- Assumptions: firm 1 en 2
o Homogeneous products
o Same cost function
o Each firm chooses own price simultaneously
o Consumers only buy from cheapest
o Prices equal split equally
o Only produce if units are sold (incur costs)
- Predictions: Bertrand Paradox
o 2 firms = enough to achieve competitive outcomes
Price at marginal cost
Same output produced as perfect competition
Zero profits
o Market concentration alone does not lead to market power
Beyond the paradox
- Many firms do make profits
o Increasing marginal costs (short term profits)
o Capacity constraints
Scaling up takes time and investments
Niet direct van 0 naar volledige market demand
o Search/information frictions
Consumers weten niet altijd waar het goedkoopst is
Takes time to shop around
o Product differentiation
Horizontal (like different products)
Vertical (different quality)
Advertisement
Diff erentiated products
Consumers don’t always switch to competing brands if price increases: brand
loyalty, differences quality/taste, switching costs, limited information
consumers treating products as if they are not homogeneous
market power restored
Demand differentiated products:
- Q = Ai + xp + ypi …….
- Cross price elasticity
3
Lec 1: Game theory and competition
Game theory: fundamentals
Agents:
- Rational preferences
Complete/perfect Information Asymmetric information
Static Strategic games Bayesian games
Dynam Extensive games, repeated Sequential (Bayesian) games
ic games
- Payoff maximizing
Main categories of games:
- Strategic
o Static
One shot
Simultaneous choice
o Complete information
All players know capabilities (available strategies) other
players
All players know consequences (payoffs) of every combination
strategies
!! does not mean certainty (probability distributions)
o Fully described by
Set of players
Set of actions (action profiles gebruiken als je een nash ofz
weergeeft)
Payoff functions (2 players: payoff matrix = normal form)
o Strategies:
X strictly dominated by Y
X heeft in alle gevallen een lagere payoff dan Y
X nooit spelen, dus verwijderen uit game
X is a strictly dominant strategy
X altijd beste optie
Altijd X spelen
o Equilibrium
In dominant strategies
Als beide een dominante strategie volgen
Alleen rationaliteit hier
Is ook een NE
Nash
1
,
Ze kunnen niet positief deviaten, elk speelt zijn beste
response op wat de andere doet
Niet per se een optimum, eerder een resting point
Niet altijd een te vinden
Soms meer dan een (players might disagree on which
is best)
Niet altijd een te vinden in pure strategies, wel altijd in mixed
strategies (matching pennies)
- Extensive /repeated games
o Dynamic
o Complete information
- Bayesian
o Static
o Asymmetric information
- Sequential
o Dynamic
o Asymmetric information
Demand, monopoly and competition
Market demand
- Price elasticity:
Δq
q dq ( p ) p
o ε ( p)= = ×
Δp dp q ( p)
p
o |ε|=1
Als de prijs dealt met x% zal de hoeveelheid
toenemen/afnemen met x%
o |ε|>1
Elastic
Increasing price reduces revenue
o |ε|<1
Inelastic
Increasing price increases revenue
Perfect competition
- Does not exist, archetype
- Consumers willing to buy any quantity, but only at price p (or lower)
price infinitely elastic
- Price taking firm
o Single-agent problem
Each firm too small to affect demand (no strategic
interactions)
Can produce q with cost C(q)
Choose quantity to maximize profit
Profit = p*q – C(q) afleiden naar q
P = C’(q) = marginale kost van q-de eenheid
IF price taking + free entry = zero profits
2
, Monopolist
- Single agent problem
o Observes downward demand slope q(p) and can set the price
o Choose quantity or price to maximize profit
Profit = p*q(p) – C(q(p))
Pm = (A+bc)/2b
In elastische deel van de vraag
Imperfect competition: Bertrand duopoly
- Assumptions: firm 1 en 2
o Homogeneous products
o Same cost function
o Each firm chooses own price simultaneously
o Consumers only buy from cheapest
o Prices equal split equally
o Only produce if units are sold (incur costs)
- Predictions: Bertrand Paradox
o 2 firms = enough to achieve competitive outcomes
Price at marginal cost
Same output produced as perfect competition
Zero profits
o Market concentration alone does not lead to market power
Beyond the paradox
- Many firms do make profits
o Increasing marginal costs (short term profits)
o Capacity constraints
Scaling up takes time and investments
Niet direct van 0 naar volledige market demand
o Search/information frictions
Consumers weten niet altijd waar het goedkoopst is
Takes time to shop around
o Product differentiation
Horizontal (like different products)
Vertical (different quality)
Advertisement
Diff erentiated products
Consumers don’t always switch to competing brands if price increases: brand
loyalty, differences quality/taste, switching costs, limited information
consumers treating products as if they are not homogeneous
market power restored
Demand differentiated products:
- Q = Ai + xp + ypi …….
- Cross price elasticity
3