'Some Doubts About NGDP Targeting'

When Steven Williamson isn't suffering from Krugman Derangement Syndrome, a frequent ailment, he can ask good questions:

Some Doubts About NGDP Targeting

In one part of the post, he says:

If we were to judge past monetary policy performance by variability in NGDP, that performance would appear to be poor. What's that tell you? It will be a cold day in hell when the Fed adopts NGDP targeting. Just as the Fed likes the Taylor rule, as it confirms the Fed's belief in the wisdom of its own actions, the Fed will not buy into a policy rule that makes its previous actions look stupid.

I'll be interested to hear the responses to his questions (assuming he has more luck than David Andolfatto in getting advocates of NGDP targeting to repond). I'm trying to be open minded, but it is not yet clear to me that NGDP targeting is the optimal policy rule in the sticky wage and price models used to evaluate monetary policy. (I have doubts about the models as well, which is why I'd like to better understand the classes of models for which NGDP targeting is optimal, and the classes for which it's not. I have a pretty good idea of how to answer that question for the Taylor rule, but am less certain for NGDP targeting. It could be that NGDP targeting is relatively robust to model uncertainty, i.e. it does well in many classes of models even though it may not be optimal within any particular class -- that would work in its favor -- but that is not yet clear either.)

Update: Scott Sumner responds. Update: Nick Rowe responds to David Andolfatto.