Monetary Policy with Imperfect Knowledge in a Small Open Economy

Incorporating adaptive learning into a small-open-economy DSGE model, we analyze how monetary policy rules should adjust when agents' information set deviates from that assumed under the rational expectations framework. We find that when agents observe current shocks but do not observe the parameters governing key macroeconomic dynamics, the resulting distortion is small and the preferred policy under rational expectations works well. However, the welfare cost of imperfect knowledge becomes quite severe when agents also have to learn about the structural shocks to the economy. Monetary policy can play a significant role in mitigating distortions associated with this form of imperfect knowledge.