Monetary policy and large crises in a financial accelerator agent-based model.

Highlights • Monetary policy and large-scale crises in an Agent based model. • Sharp contraction of short term interest rate can cause large recessions. • ZLB as an effective tool to avoid "double dip" recessions in the short run. • Pattern search and counter-factual simulations for policy analysis. Abstract An accommodating monetary policy followed by a sudden increase of the short term interest rate often leads to a bubble burst and to an economic slowdown. Through the implementation of an Agent Based Model with a financial accelerator mechanism we are able to study the relationship between monetary policy and large-scale crisis events. A two-step computational approach is proposed which performs (i) a pattern search of "double dip" episodes and (ii) counter-factual simulations implementing unconventional monetary policy. The main results can be summarized as follow: a) sudden and sharp increases of the policy rate can generate recessions; b) after a crisis, returning too soon and too quickly to a normal monetary policy regime can generate a "double dip" recession, while c) keeping the short term interest rate anchored to the zero lower bound in the short run can successfully avoid a further slowdown. [ABSTRACT FROM AUTHOR]