This Blog Has Moved!

My blog has moved. Check out my new blog at

Your Ad Here

Saturday, April 25, 2009

Prisoner's Dilemma Fnord

Pro-State troll economists and politicians frequently like to cite "the prisoner's dilemma" when justifying their decisions.

In the prisoner's dilemma game, you have two prisoners who are accused of a crime. The State police separate them, preventing them from communicating. The State only has enough evidence to convict them of a minor crime leading to 6 months in jail. If they "defect" and rat out their confederate, the State will drop the 6 month sentence and convict the confederate of a 5 year jail term. If you defect and your confederate cooperates, then you will be released and your confederate goes to jail for 5.5 years. If you both defect, then you both go to jail for 5 years. If both cooperate, then both go to jail for 6 months.

This scenario is unrealistic in a real-world situation. If both criminals have friends/partners, then the one who defects while his partner cooperates will have a *REAL HARD TIME* after being released. The scenario must be very artificially constructed, via State violence or a Math simulation.

The analogy works like this. Suppose you have State A and State B. If neither State subsidizes the widget industry, then you have free competition (the "mutual cooperation" scenario). If State A gives subsidies but B does not, then the manufacturers in State B are bankrupted by those in State A (the "A defects, B cooperates" scenario). If both States give subsidies, a lot of resources are squandered via these subsidies (the "both defect" scenario).

State-licensed economists and political scientists love to talk about the "prisoner's dilemma" as an analogy for economics and politics. I've thought about it, and it's all nonsense. In a free market, trust is important. If you make a trust-based economic transaction with someone and they cheat you, then you *REFUSE TO DEAL WITH THEM*. You warn your trading partners, and soon the cheater will be unable to do anything. In a free market, the "prisoner's dilemma" analogy is irrelevant, because people cannot be barred from sharing information.

Only a State can cause the "prisoner's dilemma" to be a valid model, by preventing people from sharing information and by forcing them to obey the decisions of their "leaders". For example, the average person in Israel or Gaza can't say "I don't want to participate in this war anymore." They are forced, via State violence, to obey the decisions of others.

In a true free market, protectionism doesn't work. Suppose region A has a free market, and there is neighboring State B with corporate widget monopoly X. Corporation X lobbies State B to wreck the widget industry in region A. State B subsidizes widget exports to region A. People in region A say "Cheap widgets!" They buy the subsidized imports, and widget factories in region A close. Corporation X says "HAHAHAHA!!! We have a monopoly now!" Corporation X now charges monopolistic prices in region A. Here is the key step. THE WIDGET FACTORIES IN REGION A NOW REOPEN! Corporation X and State B accomplished nothing. Region A has a free market, and there are no artificial barriers preventing people from starting a widget factory.

Only in a non-free market does it pay to use State subsidies to bankrupt your competitors. Once your competition is eliminated, State restriction of the market prevents new competitors from forming. "Abusive" business practices only make sense in the context of a non-free market. In a free market, anyone trying abusive behavior is merely squandering their wealth.

This actually happened, back when the USA had a freer market. There was a European bromide cartel. They tried to export cheap bromide to the USA, to bankrupt the USA bromide manufacturing industry. The bromide factory in the USA stopped operations. They bought the cheap bromide, repacked it and re-labeled it, and then sold bromide in Europe, undercutting the European cartel on price! The US bromide factory made a fortune, and the European monopoly lost a fortune! Of course, State regulations ban such rational economic behavior now.

"'Prisoner's Dilemma' is a valid basis for economic and political decisions" is an evil fnord. It is used to justify the actions of insiders. "Prisoner's Dilemma" is only a valid model for human behavior when artificial State restrictions prevent people from acting rationally outside the constraints of the game. "Prisoner's Dilemma" is an interesting abstract exercise, but it is an invalid model for real-world behavior in a true free market.

1 comment:

rotu said...

Prisoner's Dilemma is not at fault for the subsidy situation. The problem is the application of PD. Whoever uses this analogy as you cited it fails to see that this is actually an instance of Iterated Prisoner's Dilemma, where players can base their decisions off the other player's past decision. Under IPD, one has to deal with the ecology of other players: If it is expected that the other player will follow a course of action regardless of its consequences (e.g. always cooperate or always defect no matter what the other player has done), then it is rational to always defect. However if there is a reasonable expectation that the other player will cooperate if they see you are willing to cooperate and the other player will defect if you try to take advantage of them, then in IPD, it is a very good strategy to do the same.

Iterated Prisoner's Dilemma is, in fact, a very good model if and only if the decision is controlled by those people who are economically affected. Unfortunately, this is not usually the case.

This Blog Has Moved!

My blog has moved. Check out my new blog at