Non-zero net inflation in the steady state of NK DSGE

Steady state
1

Dear all,

I am currently working with New-Keynesian DSGE model (in levels) with Calvo-type price rigidities. My model features 2% (annual) inflation target (\pi=1.005), which enters a standard inflation targeting Taylor rule.

My main question: should I worry about non-zero net inflation in the steady state?

Are there any additional pitfalls (maybe for welfare measurement)? I though there was no problem, but I looked through the forum (see, example 1 or example 2) and it made me worry.

I would be happy to receive any feedback, it would be highly appreciated.

2

Most posts are about the distinction of trend inflation with and without indexing. Both cases are well established in the literature and are easily handled. But problems arise if you confused the two cases and e.g. assume no indexing while having in mind a Smets/Wouters type model with full indexing.

3

Professor, thank you for response.
To make sure I fully understand: having trend inflation while allowing firms to index their prices according to the past inflation is fully legit then (with reference to Smets/Wouter)?
Alternatively, no indexing MIGHT imply some undesired properties for the model (as in Ascari/Sbordone)?

4
  1. You can have indexing to either past or steady state inflation. That is fully legit.
  2. I would not call it “undesired” properties. Ascari/Sbordone make the point that there seems to be no indexing in practice and the “undesirable” properties may then be empirically plausible properties.
1 Like