The Taylor principle plays a central role in the analysis of inflation anchoring in New Keynesian dynamic
general equilibrium models. Ensuring determinacy of equilibria, the Taylor principle attributes
to the central bank the power of anchoring inflation. A by-product of this approach is that the Taylor
principle sharply reduces inertia in inflation, as it allows to achieve steady state equilibrium with a
stroke of a pen. In this paper, we empirically analyse the relationship between inflation inertia and
the Taylor rule, focusing on a sample covering the US starting from the 1970s. Two main results of
our empirical analysis stand out. First, we find that inflation persistence increased over time across a
number of inflation indexes. Second, we find evidence of parameter instability in the monetary policy
rule followed by the Federal Reserve Bank across a variety of econometric methods. Building on
recent extensions of the standard New Keynesian Dynamic Stochastic General Equilibrium model in
which the central bank sets its policy rate on a liquid asset, we propose a parsimonious model that
sheds light on our empirical findings. The model, which does not require the Taylor principle for
determinacy, reproduces the behaviour of the standard 3-equation New Keynesian Dynamic Stochastic
General Equilibrium model (NKSGE) as well as the increased persistence in simulated inflation
retrieved in the data.