On the stability properties of optimal interest rules under learning

In recent literature on monetary policy, it has been argued that a sensible policy rule should be able to induce learnability of the fundamental equilibrium: if private agents update their beliefs over time using adaptive learning technques, they should be able to converge towards rationality. Evans and Honkapohja (2003) showed that in a New Keynesian model an expectations based rule has such a desirable property, while a fundamentals based one does not. In order to implement an expectations based rule, though, the policymaker needs to observe private sector expectations. We show that there exists an alternative rule, based only on fundamentals, that can achieve the same positive results in terms of stability of private sector?s learning dynamics. Moreover, such a rule is learnable by the policymaker, and the combined learning dynamics of the private sector and the central bank make the economy converge to the fundamental equilibrium.