“Backtesting”

Very soon to be five year anniversary of this prediction:

May 13 (2009) (Bloomberg) — The Federal Reserve may soon need to raise interest rates, said John Taylor, the former Treasury official who devised the “Taylor Rule,” a formula for rate- setting based on the outlook for inflation and growth.

“My calculation implies we may not have as much time before the Fed has to remove excess reserves and raise the rate,” Taylor, a Treasury undersecretary under President George W. Bush from 2001 to 2005, said yesterday at an Atlanta Fed conference in Jekyll Island, Georgia.

And also five years after this “prediction” by  Scott Sumner:

Suppose you have a crystal ball, and are given one peak at the future, say May 2011.  But you are only allowed to look at one variable—and it’s not the Dow, it’s the fed funds rate.  Now suppose I tell you the following, it will be one of these two numbers:

a.  0.25%

b.  3.75%

If I looked into the ball and saw 0.25% fed funds rates in 2011, I would have a sickening feeling—like I’d been punched in the solar plexus.  Krugman would be right, we’d be another Japan.

In contrast a 3.75% fed funds rate would put a big smile on my face, as it would indicate nominal GDP growth had bounced back strongly.  It would have been a V-shaped recovery.

In my view, promising year after year of near zero rates is like promising year after year of sub-par nominal growth.  The central bank should adopt a policy that is expected to produce a quick recovery from recession, not years more of economic misery.  A policy that is successful will result in much higher nominal interest rate in the future.

Five years on the “winning prediction” is obvious! Yes, nominal growth remained sub-par all the way through.