Recurrent hyperinflations and learning

Abstract

This paper uses a model of boundedly rational learning to account for the observations of recurrent hyperinflations in the last decade. We show that, in a standard monetary model when the full rational expectations assumption is replaced by a pseudo rational learning, the model replicates some stylized facts observed during the recurrent hyperinflations observed in some countries in the 80's much better than pre-existing models. We argue that this departure of rational expectations does not preclude falsifiability of the model and it does not violate reasonable rationality requirements.