This will be a simple post.
First of all, I must recognize (after a post criticizing Market Monetarists) that NGDP targeting is one of the most sensible policy options that a Central Bank must pursue.
I still doubt Central Banks can efficiently achieve it in some cases (distorting asset prices in deep recessions) and still think it must be well studied which should be the target per se. (why 5%?).
I guess, the main change in my opinion came from David Beckworth (among other Market Monetarists) whom advises that a NGDP target instead of an inflation target is capable of disentangle the effects in inflation of Demand and Supply shocks.
A Central Bank is supposed to react to Demand Shocks but not so much to Supply Shocks.
(I say not so much and not entirely because if the shock has a great amplitude and leads the economy to severe inflation/deflation, it is still the Central Bank duty to avoid Debt/Deflation problems and other high inflation problems).
If the Fed had followed a NGDP targeting policy it would recognize that the decrease in inflation in 2002 was a sign of a positive supply shock and not a negative demand shock.
It would then avoid decrease its interest rate (expansionary policy as the natural rate should have been increasing with the positive supply shock) and maybe…just maybe, avoid all the debt build-up that led to the Financial Crisis.
Speaking of natural rate, I must admit I made a mistake in my post about r* (the Wicksellian natural interest rate). This interest rate isn´t the one that equals GDP to its potential (natural level/whatever), it is the rate that equals GDP to its natural level IF we start from that position (without an output gap).