We study macroeconomic systems with forward-looking private sector agents and a monetary authority that is trying to control the economy through the use of a linear policy feedback rule. A typical finding in the burgeoning literature in this area is that policymakers should be relatively aggressive in responding to available information about the macroeconomy (more aggressive than they appear to be in reality). A natural question to ask about this result is whether policy responses which are too aggressive might actually destabilize the economy. We use stability under recursive learning a la Evans and Honkapohja 1999a as a criterion for evaluating monetary policy rules in this context. We find that considering learning can substantially alter the evaluation of model economies in some situations. We also find that a certain type of rule is robustly associated with both determinacy and learnability. This is an active, Taylor-type rule, with only a small positive reaction to variables other than inflation.
