Microeconomic Analysis

ECON 520 - SS 29-30

KFUPM - Fall-23
Dr. Muhammad Imran Chaudhry, CFA
 OPEC Game Debrief
• How is this OPEC game similar to, and different from, prisoners’
dilemma-type games:
– Similarities: Strategic interdependence in oligopolistic markets creates a
tension between self interest and collective interest.
– Differences: No dominant strategy in this OPEC game. This OPEC game is
a repeated simultaneous move-game (i.e., a dynamic game).
• When a dominant strategy does not exist, we use the definition
of Nash equilibrium to solve games.
– A set of strategies is a Nash equilibrium of a game if no player can obtain
a higher playoff by playing a different strategy, given that all players play
their Nash equilibrium strategy.
• No player has an incentive to deviate from their Nash equilibrium strategy,
and this represents the best response of a player trying to maximize his/her
payoff given their belief that rivals are trying to maximize their own payoffs.
• Every dominant strategy equilibrium is also a Nash equilibrium.
– This OPEC game has a unique Nash equilibrium, i.e., the set of strategies
that yield outcomes from which no agent has an incentive to deviate.
• Calculus is needed to compute the Nash equilibrium of this OPEC game.
 OPEC Game Debrief
• This OPEC game is a reflection of problems managers face on a
daily basis e.g., group projects, sales targets and joint ventures.
– Trying to elicit cooperation from others to accomplish tasks that are in
the collective best interest but contrary to an individuals’ self interest.
– This aspect is also common to many social dilemmas like conservation of
conservation of environmental resources (e.g., water, fish etc.), climate
change and public good contributions.
• Learning tactics to cope with this apparent dichotomy between
cooperation (collective interests) and conflict (self-interest) can
help us become better managers/problem-solvers:
– How can we elicit cooperation from others in such circumstances?
– Building trust is the key to cooperation in business. How can one build
trust in situations characterized by strategic interdependence?
• Shadow of the past: Initial actions determine path of subsequent actions,
i.e., first impressions are often the last impression (or path dependence)!
• Communication: Open communication channels are conducive for eliciting
cooperation from others, i.e., regular meetings (don’t burn the bridges).
 OPEC Game Debrief
– Taking risks: Putting oneself in a vulnerable position typically helps build
trust. Courage begets courage (i.e., need a big heart to run a big show!)
– Relational frames: Different people often view identical situations very
differently, so must determine if others share your relational frame!
• For example, one country visibly upset that counterparts reneged on a deal,
but others may have viewed deal as only tentative, and not final. Likewise,
some people played this game as a race for the highest profit, while others
played this game as a pursuit for a high profit.
– Expectations: If expectations from others are high, then untrustworthy
behavior can be very damaging, e.g., others go into their shell once you
betray their trust once. So, try carefully manage others' expectations!
• Breakdown of trust instigates a chain reaction of defect strategies.
– Endgame problems: Cooperation unlikely when the future is uncertain.
– In general, the behavior of participants was consistent with rationality:
• Though, participants do not always converge to Nash equilibrium, but many
figured out the profit-maximizing production, and cheap-talk was of no use!
– Cartels are not generally stable, even in our game with 3 players! Social
dilemmas like curbing global warming are really hard to resolve.
• Can self interested behavior solve the major challenges of the 21st century?
 OPEC Game Debrief
• Repeated prisoners’ dilemma types of games can be resolved by
rewarding cooperation and punishing defection, whereby gains
from defection in one stage leads to large losses in future plays.
• To this end, experimental studies show that tit for tat strategies
with the following properties work best in resolving the mutual
destruction observed in prisoners’ dilemma-type games:
– Clear: Opponents know how you will respond.
• If boundaries of improper behavior are not prespecified then temptation to
defect increases.
– Nice: Don’t be the first to choose “defect”.
– Provocable: Don’t let cheating by rival go unpunished.
– Forgiving: Don’t hold grudge against your rivals for too long.
• Upshot: Encourage cooperation whilst evading the possibility of
exploitation.
• Lab experiments show more cooperation is observed in the real-world then
predicted by game theory, due to other drivers of behavior (e.g., altruism).
• Lab experiments also reveal that women are more likely to cooperate than
men in prisoners’ dilemma type games.
 A Game of Chicken!
• Game Setup: Two car drivers hurtle towards edge of a cliff (or
each other), and the driver who bails out first loses.
– Economic game of chicken arise when R&D creates a new technology
that cannot be patented, i.e., both firms may benefit from the R&D of
either firm. Let's look at a concocted example:
• Suppose Apple or Nokia can spend money on R&D to develop a
touch-screen technology that both would end up being able to
use with their cell-phones, irrespective of which firm developed
the new touch-screen technology. The payoffs are as follows:
– If both do R&D, payoffs for each firm are $5 million.
– If neither does R&D, payoff for each firm is zero.
– If Nokia does R&D and Apple does not undertake R&D, then the payoff
of Nokia is $1 million and payoff of Apple is $10 million
– If Apple does R&D and Nokia does not undertake R&D, then the payoff
of Apple is $1 million and payoff of Nokia is $10 million
• What's the Nash equilibrium of this game?
A Game of Chicken!
 A Game of Chicken!
• Game Analysis
– If Apple does R&D, Nokia’s best strategy is not to do R&D.
– If Apple does no R&D, Nokia’s best strategy is to do R&D.
– If Nokia does R&D, Apple’s best strategy is not to do R&D.
– If Nokia does no R&D, Apple’s best strategy is to do R&D.
• This game has two (pure strategy) Nash Equilibrium!
– The equilibrium for this R&D game of chicken is for one firm to do the
R&D i.e.,
1. Apple: R&D and Nokia: No R&D.
2. Nokia: R&D and Apple: No R&D.
• But we cannot predict which firm will do R&D, and which firm
will not do R&D
– This type of game has another mixed-strategy Nash Equilibrium!
 Sequential-Move Games-Overview
• In sequential-move games (with complete information), players
move sequentially:
– Players have complete information about previous moves of all players.
– Players have complete information about payoff functions at each move.
• Sequential-move games are analyzed using extensive-forms i.e.,
game trees, which summarize:
– (i) Number of players (ii) Sequence of moves by players (iii) Actions each
player can take at each stage of the game. (iv) Information player have
about rivals’ previous moves (v) Payoffs from all possible strategies.
• Game trees are a visual aid to solve sequential-move games.
Let’s lay out the rules for making game-trees:
– A decision node (usually a rectangle) indicates which player’s turn it is.
– Branches emanating from a decision node indicate all possible actions
available to given player with the turn.
• Two vertical lines through an action (branch) show that the player with the
turn rejects those actions.
 Sequential-Move Games-Overview
– Game trees are similar to decision trees. But, in contrast to a decision-
tree, at various branching points along the game-tree, different players
take turns making a decision.
• To predict the outcomes of a sequential-move game, we use a
stronger version of Nash equilibrium known as subgame Perfect
Nash Equilibrium
– A set of strategies is a subgame perfect Nash equilibrium if the players’
strategies are a Nash equilibrium in every subgame.
• Subgames: subsequent decisions available given previous actions.
– We solve for the (subgame perfect) Nash equilibrium of any sequential-
move game using the method of backward induction. This involves:
• Determine (1) best response by the last player to move, (2) best response
for the player who made the next-to-last move (3) repeat process until we
reach the start of the game.
• We call this method of looking forward and reasoning backwards sequential
rationality.
 Sequential-Move Games-Overview
– In simple words, we try to anticipate where your initial choice will
ultimately lead you to, and use this information to iteratively work
backwards to determine your optimal strategy.
• Example-I: Should you accept an investment proposal from
a well-dressed (aristocratic) stranger for a lucrative one-off
investment in a country with weak institutional settings?
– No under normal circumstances.
– Repeated interaction or relationships (past or future) can make the
stranger’s promise credible.
– This element of conflict and commonality of interest exists in most
business interactions:
• Coopetition: Co-operate to increase size of pie and compete to
divide up the pie.
 Sequential Move Games: Example
• Often, HHI is very high suggesting a monopoly or oligopoly
market, yet puzzlingly we observe competitive pricing in this
market e.g., airline industry?
• Game theory can help us resolve this conundrum:
– Assume new firms can enter/exit markets easily e.g., routes served
by airlines (i.e., contestable markets).
– New entrants enter markets if opportunity for economic profit exists
and stay out otherwise.
– Incumbent gets to set the price first at either a competitive (low) or
monopoly (high) level, prices once set, cannot be changed.
– Customers have no loyalty and buy from the lowest price firm.
• These settings describe contestable markets: i.e., sometimes
firms can enter and exit markets easily e.g., routes served by
airlines, so the incumbent firm faces threat from potential
entrants.
 Sequential Move Games: Example
– Assume that Agile Air is the only airline operating on a particular route.
But Wanabe Inc could also offer services on this route.
– Structure of game:
• Agile Air sets its prices first, once price is set and advertised, Agile cannot
change it. Agile can set prices at the monopoly level or competitive level.
• Subsequently, Wanabe Inc decides whether to offer services on the route
or not.
– Payoffs are as follows:
• Agile sets monopoly price and Wanabe enters (-50,90)
• Agile sets monopoly price and Wanabe stays out (100,0)
• Agile sets competitive price and Wanabe enters (-50,-10)
• Agile sets competitive price and Waanbe stays out (0,0)
– Lets see the game tree for this strategic interaction.
 Sequential Move Games: Example
• By modeling this as a sequential-entry game we can employ
a game tree to showthat:
– Incumbent firms sets prices at the competitive (low) level and new
entrant decides to stay out of the contestable markets by using
backwards induction.
– Result: Competitive pricing in uncompetitive markets!
• Related concept: Limit pricing, monopolist sets prices at the
highest possible level such that new entrant makes a loss at
that price.
– Final outcomes of any strategic interaction between incumbent and
new entrant depends on the relevant pay-offs determined by other
features of the environment e.g., the cost efficiencies, information
asymmetries and repetition.