The Prisoner's Dilemma

Why Rational People Make Irrational Choices

Two suspects are arrested and held in separate rooms. Each is offered the same deal: betray the other and go free, or stay silent. If both stay silent, both get a light sentence. If both betray, both get a heavy sentence. If one betrays and the other stays silent, the betrayer walks and the silent one gets the maximum penalty. The rational choice for each individual is to betray. The rational outcome for both individuals together is to cooperate. This gap between individual rationality and collective rationality is the core tension in economics, politics, and finance. It explains bank runs, market panics, arms races, price wars, and environmental destruction. Every system humans build to manage money is, at its foundation, an attempt to solve this problem.

The Payoff Matrix

Game theory represents decisions as matrices. Each player chooses a strategy. The intersection of their choices determines the outcome. In the Prisoner's Dilemma, the payoffs are structured so that mutual cooperation produces the best collective outcome, but each player has an individual incentive to defect regardless of what the other does. This structure, where the dominant individual strategy leads to a suboptimal collective outcome, appears across finance constantly.

• Both cooperate (stay silent): Both get 1 year. Total damage: 2 years.
• Both defect (betray): Both get 3 years. Total damage: 6 years.
• One defects, one cooperates: Defector goes free, cooperator gets 5 years. Total damage: 5 years.
• Nash Equilibrium: Both defect. Neither player can improve their outcome by changing strategy alone.

Bank Runs Are Prisoner's Dilemmas

A bank is solvent. It has enough assets to cover all deposits, but not all at once (fractional reserve). If all depositors cooperate and leave their money in the bank, everyone is fine. But if you suspect others might withdraw, your rational move is to withdraw first, before the bank runs out of cash. If everyone thinks this way simultaneously, the bank fails, even though it was solvent. The 1907 panic, the Great Depression bank failures, Northern Rock in 2007, Silicon Valley Bank in 2023: all the same game. Individual rationality (protect yourself) produces collective catastrophe (destroy the bank). Every solution humans have invented for this, FDIC insurance, central bank lending, deposit guarantees, is an attempt to change the payoff matrix so that the rational individual choice becomes "stay."

Repeated Games Change Everything

The Prisoner's Dilemma gets interesting when it repeats. In a one-shot game, defection is always rational. But when players interact repeatedly with no known end point, cooperation becomes viable because defection has future consequences. Robert Axelrod ran a famous tournament in 1980 where game theorists submitted strategies for repeated Prisoner's Dilemma. The winner was the simplest: Tit-for-Tat. Start by cooperating. Then mirror whatever the other player did last round. Cooperate if they cooperated, defect if they defected. The lesson: in repeated interactions, reputation matters. Trust can be built incrementally. Punishment for defection must be swift and proportional. Forgiveness after punishment must be possible. Every functioning financial market operates on this principle. Cheat once, lose access. The threat of exclusion maintains cooperation.

• Tit-for-Tat: Simple, transparent, retaliatory but forgiving. Won Axelrod's tournament twice.
• Always Defect: Wins in the short run, loses access to cooperative partners over time.
• Always Cooperate: Gets exploited by defectors. Naive trust is not a viable strategy.
• Implication: The best strategy depends on whether you'll interact again. One-shot deals invite defection.

Where This Shows Up in Your Financial Life

Market panics: everyone sells simultaneously because they fear others will sell first. The selling causes the crash they feared. Tragedy of the commons: overfishing, overbuilding, overleveraging. Each actor captures individual gain while distributing cost across the group. Race to the bottom: companies cut corners on quality, banks lower lending standards, workers accept lower wages. Each decision is rational in isolation, destructive in aggregate. The 2008 financial crisis was a massive multi-player Prisoner's Dilemma. Every mortgage originator had an incentive to approve bad loans (they sold the risk immediately). Every bank had an incentive to package and sell those loans (fees today, consequences later). Every rating agency had an incentive to rate them highly (paid by the issuer). Individually rational. Collectively catastrophic.

The Prisoner's Dilemma reveals why rational actors produce irrational outcomes when incentives are misaligned. Bank runs, market panics, and financial crises are all instances of this structure. Solutions require changing the payoff matrix (insurance, regulation, reputation systems) or enabling repeated interaction where defection has consequences. Understanding this framework is prerequisite for understanding why trustless verification systems were invented.