PS420/520: International Organization

I. Intro

II. Problem Structures: Thinking about Incentives in Game Theory terms

A. This is another way to help you think about the stuff we discussed in previous class

B. Games Notation:

1. CELLS contain 2 Payoffs, i.e., the value of a particular outcome to that player, with payoffs being potentially different for each player - right hand number is always Col's payoff and left hand is always Row's payoff

2. Columns and rows contain Strategies of one player

3. Rs-s is the payoff to player R (for "Row") of the outcome produced when Row plays S and Col also plays S

4. Cns-s (right side of bottom left cell) is the payoff to player C (for "Col") of the outcome produced when Row plays not-S but Col plays S

5. Numbers arranged so that higher the payoff number, the better or more preferred that is

6. Games are simplified as two players but can usually have multiple players

C. Harmony - no conflict situation where independent decisions lead to outcomes preferred by all parties. Preferences are independent or mutually beneficial.

1. No institution needed to resolve conflict because there is no conflict

D. Deadlock – most basic and unrepentant form of conflict. No middle ground. Interests are diametrically opposed. Zero sum conflicts.

1. Both sides have an unambiguously preferred strategy which involves harming the other side.

2. Although both sides preferred OUTCOME (the cell) is the other side stopping from engaging in the conflict while they continue (i.e., kill them without being killed in exchange), if they can't have that, they prefer continuing the conflict over halting it

3. Although Deadlock can become another game over time, and then may be subject to cooperation, so long as preferences are truly Deadlock preferences, there is no resolution

4. Examples: ancient hatreds, ethnic violence, conflicts over values (e.g., whaling?)

E. Positive externalities plagued by incapacity – hard because incentives on one side to remedy may not be sufficient to generate solution to the problem because of incapacity.

1. Problem is one of incapacity not incentives

2. All states better off if they cooperate in addressing the problem but some states simply don't have the capacity

3. States that do have the capacity must have sufficient interest to pay for others to contribute - capable have to "foot the bill" for the incapable

4. Examples: health care, particularly AIDS – all have incentives to prevent but nobody wants to pay.

F. Upstream/downstream problem - similar to suasion

1. Upstream states receive no direct benefits from cooperation and probably incurs costs to cooperate. Willing to cooperate only if receive sidepayments that outweigh these costs

2. Very strong incentives for upstream states to renege on their commitments after agreement is reached - problems of obsolescing bargains

3. Example: French reduction of pollution of the Rhine river, Canadian control of dams on the Columbia; nuclear weapons programs of weak (as "upstream") states

G. Coordination (Battle of the Sexes) – easy because all states have incentives to agree and once states agree on how to cooperate, there are no incentives to defect. Example: English as common language for airline pilots and air traffic controllers; international ship navigation rules (sail on the right side of a sea-lane); assignment of domain names to avoid conflicts

1. Two possible equilibriums, one of which benefits one side and one of which benefits the other but where both prefer cooperation to non-cooperation

2. Examples: satellite slot allocations, navigation rights, language for air traffic controllers, URLs for websites

3. This is what Stein calls "Dilemma of common aversion and divergent interests" or Battle of the Sexes in Krasner

4. Krasner makes point that how one resolves these problems depends on whether power is symmetric among actors or unbalanced.

H. Suasion problems – really hard because some states have no incentives to agree and even if they do agree, they still have incentives and ability to defect.

1. Both strong and weak states benefit from cooperation but strong state is willing to provide all of the benefits without help but prefers that weaker states contribute

2. Weaker states have "a strong incentive to free-ride, knowing that public goods… will nevertheless be provided" [Martin, 1992 #1964].

3. Example: burden sharing with respect to security in Europe during Cold War (or, arguably, burden sharing for Iraq war)

I. Collaboration (Prisoners' Dilemma) – hard because although all states have incentives to agree, even after states agree on how to cooperate, there are still strong incentives to defect. Standard Prisoners’ Dilemma (PD) or Tragedy of the Commons.

1. There are costs to both (or all) players of cooperating and the value of cooperating depends on all others contributing, so it makes little sense to contribute if nobody else does (since the goal won't be accomplished, but it also makes little sense to contribute if everybody else does (since the goal will be accomplished even if you don't contribute and you can save the cost of compliance)

2. Both (or all) sides in the problem prefer mutual cooperation to mutual cheating but both prefer even more their own unilateral cheating

3. Game assumes symmetry in the positions of the actors - both are assumed as polluters, which may not always be the case

4. Examples: arms races, trade wars, OPEC price setting, many environmental problems particularly overharvest of a resource (e.g., fish, fur seals)

5. Known as Tragedy of the Commons in environmental issues

J. Assurance problems - All states want to collaborate and believe they will be better off if they do, but they lack perfect information and so may fail to coordinate their behavior by accident and because of excessive lack of trust.

1. Problem is one of imperfect information, not one of incentives

2. War games and pre-announcing them. Each country wants to make sure that other state knows it is only doing a war game and not actually attacking.

K. Capacity, knowledge and norms problems are not well-suited for placing in game theoretic terms but doesn't mean they aren't important. Just can't fit them readily into these sorts of intellectual matrices.