Prisoner's Dilemma: What Game Are You Playing?
Farnam Street (Shane Parrish), after Robert Axelrod · 2017
"The Prisoner's Dilemma shows that individually rational self-interest can produce a jointly worse outcome for everyone — but when the same game repeats indefinitely with an unknown end point, cooperation becomes far more likely to emerge and persist."
In the single-round version, both prisoners rationally defect (each serves 2 years) even though mutual cooperation (each serving just 1 year) would have been better for both — because neither can trust the other not to defect, and the cost of being the lone cooperator (3 years) is too high. Real-world versions of this same structure appear in oligopoly pricing (e.g., OPEC members individually tempted to overproduce oil despite a mutual agreement to restrict supply) and arms races.
The mechanism that changes everything is iteration: in a known, finite number of rounds, backward induction still favors defection on the very last round (no future round to enforce cooperation), which unravels cooperation all the way back to round one — but when the game is repeated an unknown or effectively infinite number of times, the shadow of future interactions (and future retaliation) makes cooperative strategies like tit for tat rational and stable.
What changes the incentive structure of the Prisoner's Dilemma from favoring defection to making cooperation more likely?
Read more about the topic
The explanation above is written with AI assistance. These are the originals — go to them to check it.
- The Surprising Power of the Long Game (related Farnam Street piece)Farnam Street
- Finite and Infinite Games (James Carse)Farnam Street
Tit For Tat
"In repeated (iterated) games, the most effective strategy is tit for tat: cooperate first, then simply mirror whatever your counterpart did last time — ideally with occasional forgiveness built in to escape cycles of mutual retaliation."