A comparison of monetary policy rules in an estimated TANK model

We compare the stabilisation properties of history-dependent and asymmetric interest rate rules, taking into account the constraint posed by the effective lower bound on nominal interest rates. Specifically, we use a medium-scale Two-Agent New Keynesian (TANK) model that was estimated on euro area data. Our simulation results suggest that under rational expectations history-dependent rules can attenuate or even undo the sizeable negative in ation bias that we observe under standard in ation targeting. They can also better stabilise in ation, but output becomes more volatile. Asymmetric rules furthermore reduce the negative in ation bias. However, the reduction in in ation volatility is less pronounced compared to history-dependent rules. We further show that an appropriate calibration of an asymmetric in ation targeting rule improves its performance along specific dimensions.