The optimal inflation rate revisited.

We challenge the widely held belief that New Keynesian models cannot predict an optimal positive inflation rate. In fact we find that even for the US economy, characterized by relatively small government size, optimal trend inflation is justified by the Phelps argument that the inflation tax should be part of an optimal (distortionary) taxation scheme. This mainly happens because, unlike standard calibrations of public expenditures that focus on public consumption-to-GDP ratios, we also consider the diverse, highly distortionary effect of public transfers to households. Our prediction of the optimal inflation rate is broadly consistent with recent estimates of the Fed inflation target. We also contradict the view that the Ramsey-optimal policy should minimize inflation volatility over the business cycle and induce near-random walk dynamics of public debt in the long run. In fact optimal fiscal and monetary policies should stabilize long-run debt-to-GDP ratios in order to limit tax (and inflation) distortions in steady state. This latter result is strikingly similar to policy analyses in the aftermath of the 2008 financial crisis.