Subgame perfect equilibrium
A plan that is a Nash equilibrium not only in the game as a whole but in every part of it that could still be reached.
Often misread asApplying subgame perfection to the outcome rather than to every node, and so checking only the path that actually gets played.
How to see it
A plan you would still follow once you actually got there
A Nash equilibrium can rest on a plan a player would abandon the moment the situation arose. Subgame perfection asks for more: the plan must be a best response not only overall but in every situation that could come up, including the ones that never do.
The checkThe off-path nodes are the entire point, because that is where plans nobody would carry out are hiding.
Try itIn entry deterrence, find the Nash equilibrium where the incumbent threatens a price war and the entrant stays out. Then ask what the incumbent would really do if entry happened. If the answer is 'accommodate', you have found the equilibrium that subgame perfection removes.
Games that show it
Know these first
Leads on to
Easily confused with
Sources
- Selten, R. (1965). Zeitschrift fur die gesamte Staatswissenschaft 121, 301-324 and 667-689. The paper that introduced subgame perfection.
- Fudenberg, D. & Tirole, J. (1991). MIT Press.