Ok, in a post today Paul Krugman gets it almost perfectly:
And what this says is that price stability isn’t an adequate guideline for monetary policy. You can have stable prices and a persistently depressed economy.
The expression “price stability” has many meanings. In Greenspan´s time it was a rate of inflation “low” enough that it didn´t interfere or distort decisions. That sort of squares with this definition of the term:
“A situation in which prices in an economy don’t change much over time. Price stability would mean that an economy would not experience inflation or deflation.”
But Bernanke has made a 2% inflation target official (in the sense that that´s what the Fed is said to be targeting). So that is the new meaning of “price stability”.
Let´s take “price stability” to mean “low” (i.e. 2%) inflation. Even in this more elastic interpretation, “price stability” has really not been an adequate guideline for monetary policy. And that “fact” was, or should have been, very well known. If not, what was the use of so many conferences and papers on “Monetary Policy in a Low Inflation Environment” over the last 12 years? And Bernanke was an important contributor in those affairs.
To be fair, in 2008 Bernanke may have thought “inflation” was rising strongly, pushed by oil and commodity prices. If so, it just goes to show how schizophrenic the definition of inflation is. In textbooks it´s defined as a continuing rise in the general price level not, take note, a change in relative prices, which in the case of oil and commodities can, in addition, be just of a temporary and reversible nature. That´s why Scott Sumner is always suggesting “let´s stop talking about inflation”.
Krugman has advocated “Krugman-style inflation target”. But that´s hardly a solution and is fraught with unpleasant collateral effects. If the Fed has the power to provide a specific definition of “price stability”, it can also establish an alternative nominal target that the Fed would pursue in order to provide NOMINAL STABILITY, which is the most it can accomplish, anyway. The pay-off is that nominal stability will most likely translate into REAL STABILITY (of both real growth and employment), just like during the “Great Moderation”. Probably the best such target would be an NGDP Level Target.