Nash Equilibrium: Why Everyone Stays Somewhere Nobody Likes

Two petrol stations sit across the road from each other. Both sell fuel at ₹100 a litre and both do well.

One owner realises something. If he drops to ₹98, drivers will cross the road for him and he will take almost the whole street's business. So he does it.

The other owner now faces a simple choice: hold at ₹100 and watch his forecourt empty, or match. He matches. Both are now at ₹98, splitting the same customers as before, each earning less per litre.

Now the first owner thinks it through again. Cutting to ₹96 would take all the customers back. And so on down.

Eventually they reach a price where cutting further would mean selling below cost. At that point, neither owner can improve his own position by changing price alone. Raising it loses every customer. Lowering it loses money on every litre. So both stay put — earning far less than they did at ₹100, in a situation neither wanted and neither can leave by himself.

That final state is a Nash equilibrium, and the concept is one of those rare ideas that is easy to state, easy to test, and changes how you read almost every stuck situation you encounter.

The definition, and the one word that matters

A Nash equilibrium — named for the mathematician John Nash — is a state where no player can improve their own outcome by changing their own strategy, assuming everyone else keeps doing what they are doing.

That last clause is the whole thing. The test is unilateral: can you do better by moving alone? If yes, you are not in an equilibrium, because you will move. If nobody can, the situation is stable — everyone is already doing the best they can given what everyone else is doing.

Notice what the definition does not say. It says nothing about the outcome being good, fair, efficient, or desirable. It is a claim about stability only.

This is the single most important and most frequently missed point about the concept. People hear "equilibrium" and think "balance," which sounds healthy. But the petrol stations ended up in an equilibrium that made them both poorer. It held anyway, because holding is not the same as being pleasant.

Once you separate stable from good, a great many frustrating situations become legible.

Bad equilibria you have probably lived in

Everyone working late. In some workplaces the norm is to stay until eight. Almost nobody wants this. But an individual who leaves at six looks less committed than colleagues who stay, so leaving alone carries a real cost. Nobody can fix it by themselves, so it persists — even though every single person would prefer the six o'clock version. This is precisely the structure the prisoner's dilemma describes, and the exhausted equilibrium is where it lands.

Arms races. Two countries would both prefer to spend less on weapons. Either gains by outspending the other. Both spend heavily, relative security is unchanged, and the money is gone. Stable, and terrible.

Escalating credentials. When a qualification becomes common, it stops distinguishing anyone, so people acquire more. Everyone studies longer and ends up in the same relative position. No individual can opt out without falling behind.

Bad meeting culture. Everyone complains about too many meetings. Declining them individually looks disengaged. So the meetings continue.

In every case the same three features appear: nobody likes the situation, everyone is behaving sensibly given their circumstances, and unilateral virtue is punished. That third feature is why moral appeals so reliably fail here — you are asking individuals to absorb a personal cost to shift something one person cannot shift.

How bad equilibria actually get escaped

Since no individual can move, escape requires changing the situation rather than the people. There are only a few real mechanisms.

Change the payoffs. Make the undesirable choice more expensive or the desirable one cheaper. Regulation does this by removing an option outright — which is why industries that compete themselves into unprofitability sometimes ask for rules they would individually break. Everyone wants the better equilibrium; nobody can reach it alone.

Make it repeated and visible. If the same players interact many times and can see each other's behaviour, cooperation can become individually rational, because today's defection carries future costs. Reputation is this mechanism made durable.

Coordinate the move. If everyone shifts simultaneously, nobody bears the cost of moving first. This is what a binding agreement, a contract, or a union is for — it converts "I go alone and lose" into "we all go together."

Change who is playing. Sometimes a new entrant with a different cost structure or different goals breaks an equilibrium that established players could not.

There is one more thing worth knowing, because it prevents a common error. A game can have several Nash equilibria, and which one you end up in may be pure historical accident. Which side of the road a country drives on is the standard example: both are perfectly stable, neither is better, and no individual driver can switch. Once a convention settles, it is self-enforcing regardless of merit — which means "this is how it is done" sometimes carries no information at all about whether it is the better arrangement.

So the diagnostic question, whenever you find yourself in a situation everyone complains about and nobody changes: what would happen to the first person who behaved differently? If the honest answer is "they would lose," you are not looking at a failure of character. You are looking at an equilibrium, and it will need structural work — not encouragement.