Glossary index › N

Nash Equilibrium

A Nash equilibrium, named after mathematician John Nash, is a situation in game theory where every participant has chosen the best strategy available to them given what everyone else is doing, so no single participant can improve their own outcome by unilaterally changing their choice. It does not mean the outcome is the best possible one for the group as a whole - it only means no individual player has an incentive to deviate on their own, which is why Nash equilibria can sometimes look inefficient from the outside while still being stable.

A simple numeric illustration: two gas stations across the street from each other are deciding whether to price gas at $3.50 or $3.30 a gallon. If both charge $3.50, both earn healthy margins; but if one drops to $3.30 while the other stays at $3.50, the discounter captures almost all the customers. Knowing this, both stations end up pricing at $3.30 even though both would have preferred the higher, more profitable price - neither can raise their own price alone without losing customers to the other, so $3.30/$3.30 is the equilibrium, even though it is worse for both than if they had somehow coordinated on $3.50/$3.50.

The concept applies across financial markets wherever competing participants react to each other's expected moves: OPEC members deciding whether to cut oil production, market makers setting bid-ask spreads against each other, or high-frequency trading firms choosing how aggressively to quote. In each case, a stable market price or behavior pattern often reflects a Nash equilibrium - a point where every participant, given everyone else's current strategy, has no immediate incentive to act differently, even if a different collective arrangement could theoretically make everyone better off.

What Nash actually proved is stronger than the definition suggests, and it is the reason the concept became foundational rather than remaining a curiosity. In a doctoral thesis of under thirty pages, he demonstrated that every finite game, with any number of players, has at least one equilibrium, provided players are permitted to mix: to choose randomly between their options with fixed probabilities rather than committing to one.

That proviso is essential, because without it many games have no equilibrium at all. In the children's game of matching pennies, whatever either player does predictably the other can beat, so no pair of fixed choices is stable; the only stable answer is for both to randomise evenly. The existence result is what allows economists to assume a solution exists before going looking for it, which is why the concept underpins so much of modern economics and why Nash shared the 1994 Nobel Memorial Prize for it. The mixed-strategy idea also explains behaviour that looks irrational from outside: a market maker varying quote sizes unpredictably, a company keeping its response to a competitor's discount deliberately hard to anticipate, an auditor choosing which accounts to inspect at random. In each case being predictable is the mistake, and randomising is the optimal choice rather than an absence of one.

What is a Nash equilibrium? A situation in which every participant has chosen the best strategy available given what everyone else is doing, so nobody can improve their own outcome by changing their choice alone.
Does that make it a good outcome? No, and this is the point that gets lost. It only means no individual has a reason to deviate on their own. The result can be plainly worse for everyone than an alternative they could have coordinated on.
Can you show that? Two petrol stations opposite each other choose between 3.50 and 3.30 a gallon. Both at 3.50 earns both healthy margins, but whichever drops to 3.30 alone takes almost all the customers. Knowing this, both price at 3.30, which is worse for both than 3.50 each.
Why do they not simply agree? Because neither can raise their price alone without losing the customers, so the arrangement they both prefer is not stable. It also happens to be illegal to agree on it, which is why competition law exists.
Where does it show up in markets? Wherever competing participants react to each other's expected moves: cartel members deciding whether to cut output, market makers setting spreads against each other, high-frequency firms choosing how aggressively to quote.
What does a stable price often represent? Exactly this. A point at which every participant, given everyone else's current behaviour, has no immediate reason to act differently, even though some other collective arrangement would leave everyone better off.
Who was Nash? John Nash, the mathematician the concept is named after. He shared the 1994 Nobel Memorial Prize in Economics for it, having produced the work in a doctoral thesis of under thirty pages in 1950.
What did he actually prove? More than the definition implies, and the extra result is what turned this from a curiosity into a foundation. He showed that any game with finitely many players and options possesses at least one such point, so long as players may randomise between their choices with fixed probabilities. Without that allowance many games have no equilibrium at all: in the children's game of matching pennies, whatever either player does predictably, the other can beat, and the only stable answer is for both to randomise fifty-fifty. The existence result is what lets economists assume a solution exists before looking for it, and the mixed-strategy idea explains real behaviour that looks irrational, such as a market maker varying quote sizes unpredictably or a company keeping its pricing response deliberately hard to anticipate.

By InsiderTicker Editorial Team · Last updated