Abstract
This paper develops recursive methods to study optimal and time consistent policy in dynamic models. We analyze a version of Calvo's 1978 monetary model and show that its time consistent outcomes can be completely characterized as the largest fixed point of either of two operators. Recursive application of these operators provides a computing algorithm which always converges to the set of time consistent outcomes. Finally, we obtain valuable information about the nature of time consistent outcomes: It is discovered, in particular, that all such outcomes are Markovian. The methods obtained are intuitive and should be useful for many applications.Journal of Economic LiteratureClassification Numbers: E61; E52; C61.