The debate about targeting a higher rate of growth for nominal gross domestic product (NGDP) keeps the blogosphere humming, but the discussion doesn’t mean much if Fed Chairman Ben Bernanke doesn’t embrace the idea. Don’t hold your breath. Last month he said the idea is “reckless.” That’s monetary-speak for: Don’t even think about it. But if NGDP targeting is considered a radical notion by some, including those at the pinnacle of monetary power, the empirical record suggests otherwise.
Consider how nominal and real GDP compare through the decades when measured in terms of their rolling one-year percentage changes. As the chart below shows, there’s a relationship here that isn’t terribly surprising. Higher levels of NGDP tend to be associated with higher levels of real GDP (RGDP).
This relationship is clearer in a scatterplot graph of the two sets of GDP changes. The next chart illustrates how one-year percentage changes for NGDP fare against RGDP. It’s not a perfect fit, but you’d be hard pressed to dismiss the connection as random.
For another perspective, the third chart looks at quarterly changes in the two series and the result shows that the connection is even stronger, as indicated by a slightly higher R-squared reading vs. the one-year comparison.
The message in history is that higher (lower) levels of NGDP are linked with higher (lower) levels of RGDP. On its face, this relationship appears quite robust. Why the debate? In a word, inflation, which determines the difference between NGDP and RGDP. Everyone recognizes the relationship between nominal and real growth, but there are questions about whether a central bank can push NGDP higher and what that means for RGDP. Actually, there’s little doubt that the Fed’s ability to engineer a higher NGDP. Thinking otherwise is to question a central bank’s powers to raise inflation. Presumably, that point isn’t in doubt. What is unclear, at least in the minds of some (most?) economists is the connection between a central bank’s overt policy to push inflation higher for the express purpose of raising RDGP. The fear is that with a policy to raise NGDP, the Fed will unleash a permanently higher level of inflation.
There are also questions about whether NGDP that’s “artificially” raised can meaningfully lift RGDP in the short term. David Andolfatto, a vice president at the St. Louis Federal Reserve, recently asked on his blog Macromania: “What is the theoretical underpinning for NGDP targeting? And what is the empirical evidence that leads them to believe that an NGDP target right now is a cure for whatever ails us right now?” Economist David Beckworth responds that “David Andolfatto Can Feel More Confident About NGDP Targeting” by reviewing the case in favor of the policy as outlined by Beckworth and others of his persuasion.
Meanwhile, James Hamilton considers the possibilities with targeting a higher NGDP, but is still worried about the “logistics” and so he is “convinced that it is a mistake to ask too much from monetary policy.” Scott Sumner counters that Hamilton’s concerns are misguided.
And so it goes. But it’s all academic unless Bernanke and company undergo an attitude adjustment. On the surface, it seems like the Fed chief would be a natural supporter of NGDP targeting. As Paul Krugman recently reminded, Bernanke wrote papers in favor of the idea. Of course, it’s one thing to explore monetary policy as a professor and quite another to weigh the pros and cons of a given policy as the head of the central bank for the planet’s largest economy. Noah Smith explains:
I think Bernanke is dealing with a severe case of model uncertainty. Think about it. A professor’s job is to say “Here is a way the world might work.” A policymaker has to say “OK, I am going to act as if the world works this way.” The latter requires a LOT more faith in the model’s correctness than the former. It seems highly likely to me that Fed Chairman Bernanke does not believe in Professor Bernanke’s theories enough to make big bets on them.
Empirical facts, it seems, only go so far in monetary affairs. “The more I read about monetary policy,” Smith writes, “the more convinced I become that humankind does not really understand it very well.” Smith continues:
The fact is, we just don’t know what monetary policy is the best. Maybe QE is a good idea (I think it is!). Maybe a rule like NGDP level forecast targeting is a good idea (I am skeptical but it doesn’t sound too bad). Or maybe the amount of QE needed to produce a noticeable movement in employment is so huge that it really would cause serious inflation. Maybe monetary policy operates with “long and variable lags,” as Milton Friedman suggested, meaning that it’s very difficult for the Fed to know the consequences of its actions. I am not economically illiterate. I can easily find, read, understand, and explain a paper supporting any of these contentions. But at the end of the day I’m willing to bet you that I won’t really know how right the paper is. At best, my opinions will probably only have shifted slightly. I am guessing this because I’ve never read a monetary policy paper that convinced me that “OK, this has got to be how the world works.”
So I think that Ben Bernanke has been paralyzed into inaction by the realization that, his academic papers aside, he doesn’t really know if QE would be good or bad.
That may or may not be true, but until (if) Bernanke changes his mind about policy, it’s (still) all academic.
This post originally appeared at The Capital Spectator and is posted with permission.
3 Responses to “Does History Support NGDP Targeting Now?”
As can clearly be seen in the first chart, the relationship between NGDP and RGDP is least correlated in the post-WWII period during the period of high inflation and monetarism (1970's through early 1980's).
For those convinced that NGDP targeting will be successful, the task is to explain why changing policy to promote higher inflation today will not cause a breakdown in the correlation of the past 30 years, similar to the 1970’s. Otherwise it seems perfectly reasonable to expect that NGDP targets will be met with increasing inflation, not real growth. http://bubblesandbusts.blogspot.com/2012/05/ngdp-…
Fortunately, that all got sorted out decades ago: if the economy is operating at full capacity (the natural rate of unemployment) then an increase in NGDP ultimately results in only an increase in inflation. So, when NGDP was raised well above the economy's capacity for production in the 1970s and early 1980s, the result was the Great Inflation.
Targeting NGDP can bring the economy up to capacity, but not beyond it in the long run.
I think it is perfectly reasonable to ask questions of NGDP level targeting, but I have a bit of a disagreement as to whether we asked, or were allowed to ask all the appropriate questions of hard PCE inflation targets before Bernanke actually did it. There is plenty of evidence that it was something Bernanke strongly wanted to do, and it is worth asking whether it is delivering the goods as promised. My feeling is that it isn't and likely complicated a bad situation in 2008-9, putting the policymakers out on a limb in the ZLB. If we're going to challenge something, it really needs to Bernanke's version of the world of monetary policy rather than just letting it go if it has some relatinship to the miserable state of affairs in which we currently find ourselves – because like this article says, it's one thing to be a professor and say this is how the world might work, and quite a nother thing if it actually doesn't work that way when one gets to implement it.