Like John Cochrane, I’m feeling “grumpy” this morning. Here’s John Taylor:
And in a new research paper, Scott Sumner suggests another way that Fed policy makers might conduct policy in which the monetary base or the interest rate is determined in an iterative manner based on nominal GDP futures without mentioning a reaction coefficient. In all these cases, the comparison with the Taylor Rule is less straightforward.
This confuses me on two levels. First, it’s a Svenssonian “target the forecast” policy, so I don’t see how a direct comparison with the (backward-looking) Taylor Rule is possible. And second, I’m confused because I did offer three very specific well spelled-out policy rules, and one discretionary rule. Here’s one example of a specific reaction function from the paper:
For instance, each $1 purchase of a long position in an NGDP futures contract might trigger a $1,000 open-market sale by the Fed. A purchase of a $1 short position would trigger a $1,000 open-market purchase by the Fed. In that case, investors would be effectively determining the size of the monetary base.
Later in the post, Taylor cites a criticism of NGDP targeting by Robert Hall:
In his paper at the recent Jackson Hole conference, Bob Hall criticized nominal GDP targeting, citing his 1994 paper with Greg Mankiw. Bob argues that “A policy of stabilizing nominal GDP growth would require contractionary policies to lower inflation when productivity growth is unusually high. Such a policy might easily trigger a spell at the zero lower bound.” No one at the conference objected to this statement, and I do not recall nominal GDP targeting being mentioned at the conference, though policy rules were mentioned quite a bit.
No one objected to that statement because I’m never invited!! Seriously, that statement seems flat out wrong, although I’m willing to change my mind if I’ve made a mistake somewhere. To begin with, Hall should say inflation is low when output growth is high, not productivity. Yes, they are correlated, but output is more accurate. Much more importantly, interest rates would probably be almost unaffected by a change in inflation, and might even move inversely to inflation under NGDP targeting. That’s because a rise in real growth would raise real interest rates by more than a fall in inflation would lower real rates.
Consider interest rates in recent decades, when inflation has been fairly stable at around 2%. I think it’s fair to say that swings in interest rates (up to 6.5%, down to 1%, up to 5.5% down to 0%), have been driven almost entirely by swings in real growth, not expected inflation. Notice that the swings in interest rates are comparable to the movement in real output over the business cycle. And recall that rates would have fallen even further in 2009 had we not hit the zero bound. If we go back to the previous recession in 2001, nominal rates fell by more than RGDP growth declined.
Thus under NGDP targeting nominal rates would probably be much more stable than under inflation targeting. No one questioned Hall at Jackson Hole? Wow, we have our work cut out for us!
On the positive note Taylor cites an excellent paper on NGDP targeting by Evan Koenig:
In all these cases, the comparison with the Taylor Rule is less straightforward. Evan Koenig’s nice All in the Family paper also delves into the relationship between nominal GDP targeting and the Taylor Rule.
And in this paper Koenig shows how NGDP targeting helps to stabilize credit markets. I should note that Koenig’s version of NGDPLT is different from the version I’ve proposed, as it relies on estimates of output gaps. But it’s still far better than current policy, and (as I’ve discussed elsewhere) more politically feasible.
I’m so grumpy that I’m not letting Cochrane off Scott-free today. Saturos asked me to comment on John Cochrane’s interpretation of the Phillips curve in recent history:
By the way, I think when the dust has settled, history will be kind to Ben Bernanke. He fits most of my job description. Inflation is stuck at 2%, the world did not melt down, and we’re all gradually coming to the realization that if $2 trillion bucks of stimulus and zero interest rates didn’t bring our economy out of the doldrums, there really is nothing more that a central bank could do. The Phillips curve has been screaming “this is supply, not demand” for a few years now.
And here’s how I responded to Saturos:
I’d say the PC has been screaming demand not supply, as have the asset markets. In 2009 we had deflation, the biggest drop in NGDP since 1938 and soaring unemployment. How is that not demand? Since then we’ve had the slowest recovery in demand in any expansion in US history (at least as far back as I know), and a slow recovery in RGDP. What’s the mystery here that needs to be explained?
I’d add that if Milton Friedman heard that University of Chicago economists were propounding “liquidity trap” theories he’d be rolling over in his grave. Especially if they were efficient markets-types and the markets were screaming that QE is effective. Not to mention Japan, where 2013:Q2 growth was just revised up to 3.6% and the yen has plunged and stocks have soared on policies that an unholy alliance on the left and right thinks are meaningless.
PS. Just to be clear, bad supply-side policies have slowed the recovery. Casey Mulligan has some good points. But the main problem has clearly been demand.
PPS. Commenter Blue Aurora has sent me information on a new film coming out soon on the Fed. This link has a trailer.