Solutions to a prisoner’s dilemma are typically assumed to involve “coordination” in some sense. But what kinds of mechanism are appropriate examples for coordination? For an N-person prisoner’s dilemma, one form of coordination is implementing voting. Say, everyone is forced to cooperate when the majority votes “cooperate”. Nobody has a selfish interest to cooperate, but everyone has a selfish interest to vote for “cooperate”.
This is interesting because economists often see voting as irrational for decision theoretic reasons. But from the game theoretic perspective above, it appears to be rational. This is probably not a new insight, but I haven’t seen voting being portrayed as a type of solution to N-person prisoner’s dilemmas.
Note that the game-theoretic “true” prisoner’s dilemma is formulated to make coordination (both communication and outside-of-game considerations like reputation, side-payments, self-image, etc.) are ignored. All of the non-Nash “solutions” are by introducing factors into the game that change the game very significantly.
Voting (especially when defectors aren’t penalized, just forced to cooperate) is a pretty big variation, and needs to be explicitly modeled in order to determine what equilibra are rational.
This is almost unrelated to real-world voting, which has SO MANY complicating and interfering factors that simple models just don’t tell us much.
Of course for a “real” prisoners dilemma any form of coordination is ruled out from the start. But in real world instances, coordination can sometimes be introduced into systems that previously were prisoner’s dilemmas. That’s what I mean with “solving” a prisoner’s dilemma. Making the dilemma go away.
The thing I’m pointing out here is that “coordination” is a very unspecific term, and one concrete form of coordination is being able to vote for cooperation. (Example: voting on a climate change bill instead of trying to minimize your personal carbon footprint, which would make you personally significantly worse of with hardly any benefit on the whole, which is why you would defect but vote on cooperate.) I think voting is usually not appreciated as a method of coordination, only as a method of choosing the most popular policy/party, which doesn’t need to involve solving a prisoner’s dilemma.
“coordination” is a very unspecific term, and one concrete form of coordination is being able to vote for cooperation
Ah. I’d say that “voting” is pretty non-specific as well. It’s the enforcement mechanisms that bind behaviors after the votes are counted that are actually the coordination mechanisms. Voting is the easy, un-impactful part, enforcing (socially as well as legally/violently) the result is impactful.
Voting is well-known and OFTEN used as a mechanism for determining the most agreeable (or least likely to result in riots) result. It’s a key prerequisite to many coordination mechanisms. But it isn’t a complete mechanism on its own.
It’s often said that controlling the ballot is more important than controlling the vote. The pre-voting process to figure out how to coordinate the options to choose among (and the pre-pre-voting decisions for preliminary votes) matter a whole lot.
Solutions to a prisoner’s dilemma are typically assumed to involve “coordination” in some sense. But what kinds of mechanism are appropriate examples for coordination? For an N-person prisoner’s dilemma, one form of coordination is implementing voting. Say, everyone is forced to cooperate when the majority votes “cooperate”. Nobody has a selfish interest to cooperate, but everyone has a selfish interest to vote for “cooperate”.
This is interesting because economists often see voting as irrational for decision theoretic reasons. But from the game theoretic perspective above, it appears to be rational. This is probably not a new insight, but I haven’t seen voting being portrayed as a type of solution to N-person prisoner’s dilemmas.
A semi-relevant example that your take reminded me of: https://braineaser.com/brainteasers/5-pirates-puzzle/
Note that the game-theoretic “true” prisoner’s dilemma is formulated to make coordination (both communication and outside-of-game considerations like reputation, side-payments, self-image, etc.) are ignored. All of the non-Nash “solutions” are by introducing factors into the game that change the game very significantly.
Voting (especially when defectors aren’t penalized, just forced to cooperate) is a pretty big variation, and needs to be explicitly modeled in order to determine what equilibra are rational.
This is almost unrelated to real-world voting, which has SO MANY complicating and interfering factors that simple models just don’t tell us much.
Of course for a “real” prisoners dilemma any form of coordination is ruled out from the start. But in real world instances, coordination can sometimes be introduced into systems that previously were prisoner’s dilemmas. That’s what I mean with “solving” a prisoner’s dilemma. Making the dilemma go away.
The thing I’m pointing out here is that “coordination” is a very unspecific term, and one concrete form of coordination is being able to vote for cooperation. (Example: voting on a climate change bill instead of trying to minimize your personal carbon footprint, which would make you personally significantly worse of with hardly any benefit on the whole, which is why you would defect but vote on cooperate.) I think voting is usually not appreciated as a method of coordination, only as a method of choosing the most popular policy/party, which doesn’t need to involve solving a prisoner’s dilemma.
Ah. I’d say that “voting” is pretty non-specific as well. It’s the enforcement mechanisms that bind behaviors after the votes are counted that are actually the coordination mechanisms. Voting is the easy, un-impactful part, enforcing (socially as well as legally/violently) the result is impactful.
Voting is well-known and OFTEN used as a mechanism for determining the most agreeable (or least likely to result in riots) result. It’s a key prerequisite to many coordination mechanisms. But it isn’t a complete mechanism on its own.
It’s often said that controlling the ballot is more important than controlling the vote. The pre-voting process to figure out how to coordinate the options to choose among (and the pre-pre-voting decisions for preliminary votes) matter a whole lot.