Asynchronous Choice and Markov Equilibria

We provide a theoretical foundation for the use of Markov strategies in repeated games with asynchronous moves. If admissible strategies must display finite (arbitrarily long) memory and each player incurs a “complexity cost” which depends on the memory length required by her strategy, then every Nash equilibrium must be in Markovian strategies. If, in addition, admissible strategies have uniformly bounded memory, every rationalizable strategy must be Markovian. These results are robust to considerations of perfection and also yield interesting implications for equilibrium selection in simple contexts.