*The Economist* endorses ngdp targeting

…it makes sense to look beyond inflation—and to consider targeting nominal GDP (NGDP) instead…

A target for nominal GDP (or the sum of all money earned in an economy each year, before accounting for inflation) is less radical than it sounds. It was a plausible alternative when inflation targets became common in the 1990s. A target for NGDP growth (ie, growth in cash income) copes better with cheap imports, which boost growth, but depress prices, pulling today’s central banks in two directions at once. Nominal income is also more important to debtors’ economic health than either inflation or growth, because debts are fixed in cash terms. Critics fret that NGDP is hard to measure, subject to revision, and mind-bogglingly unfamiliar to the public. Yet if NGDP sounds off-putting, growth in income does not. And although inflation can be measured easily enough, central banks now rely nearly as much on estimates of labour-market “slack”, an impossibly hazy number. Most important, an NGDP target would free central banks from the confusion caused by the broken inflation gauge. To set policy today central banks must work out how they think inflation will respond to falling unemployment, and markets must guess at their thinking. An NGDP target would not require the distinction between forecasts for growth (and hence employment) and forecasts for inflation.

There is more here, congratulations to Scott Sumner and others…


I propose using the word "growth" when discussing NGDP with the public. It is no more dishonest that saying "inflation" when you mean CPI. And as the other thread shows, I am all for something that "copes better with cheap imports, which boost growth, but depress prices."

Careful. "Growth" is a rate of change. Let's not fall into the trap of defining the target as a rate of change. "Targeting" the rate of growth of NPDP and (wink, wink) allowing it to fall below target for arbitrarily long periods of time if keeping it on target wold require "unconventional" policy instruments as the Fed has done with "inflation" undermines the principle benefit of NPDG LEVEL targeting.

Hmm. I was thinking "growth" would just be NGDP change over a reasonable time-frame. I'd also think that conventional tools, interest rate and/or deficit spending, could do that. But we might face times, as recently when such conventional tools are not politically acceptable? NGDP (change) might be telling you need to borrow at low interest rates and spend, but everyone feels better with austerity even so?

It's not about matching a growth today, or in a given window, but matching a specific point, based on the moment yu started the level targeting.

So if we claim that we'll aim for 5% growth, and we start with an NGDP of 1 trillion, We should expect 1.050 at the end of the first year, and 1.1025 the second. If the first year the target was missed, we still aim for 1.1205: None of the silliness of inflation targeting, where undershooting for 10 years in a row is ok, and still leaves you with a 2% target the next year

The inflation gauge is not broken, just unused.

I think NGDP LEVEL targeting could be a clever way to get away from inflation RATE ceiling targeting. But if it, too, is subject to all sorts of ad hod constraints (inflation rates cannot be > 2%, Federal funds rate cannot be negative, QE can only be doled out in pre-announced amounts) and additional policy objective (preventing banks from "reaching for yield), it will fare no better than price level targeting did.

We can't get around the fact that there is a strong "permahawkery" current of opinon that technical fixes will not necessarily overcome.

If QE was done more democratically, via tax rebates to taxpayers, it would become extremely popular.

I think the whole ngdp targetting stuff is complete and utter rubbish.

If the economy wanted to grow at three percent and the central bank set the rate at 2% they would be running contractionary policies when they should not.

Inflation matters to investors and in particular to bond investors. An official policy by the Fed of turning a blind eye to inflation and instead targeting NGDP makes it more difficult for investors. This is compounded by the fact that the advocates of ngdp targeting seem to view it as a way to do an end run on the 2% inflation targets: who after all is going to oppose 5% growth in ngdp even though it means four percent inflation.

There is also the small problem that the Fed has been unable to reach its 2% target using the tools available to it. Changing to ngdp does not suddenly give the Fed new tools. It seems likely that the ngdp advocates would want the Fed to cross over into fiscal policy in a major way which appeals to their technocratic / bureaucratic ./ elitist hearts but is wrong politically and constitutionally.

Unable or unwilling? And no, Sumner opposes Corbyn's "people's QE".

Money is fungible.

As for inflation mattering more to investors: look at 2008.

"Unable or unwilling?"

I think that the evidence from the US and Japan points pretty clearly at "unable".

Pppffft. Increasing the money supply by 10 or thirty percent is not "unable". It's unwilling. As Scott Sumner points out, Japan increased the monetary base faster in the early 1970s.

you can increase the money supply all you want without necessarily increasing nominal or real gdp if velocity is falling.

Sure. Money is not income. But when produced in great enough quantities, money always has a positive effect on income. If the Fed started printing $10 trillion Federal Reserve Notes by the day, you think that would have no effect on NGDP?

I was disappointed that Corbyn's "People's QE" was not going to be QE sent directly to citizens, but instead be money once again controlled by the state and invested in infrastructure.

"If the Fed started printing $10 trillion Federal Reserve Notes by the day, you think that would have no effect on NGDP?"

How are they going to get that money into the economy? If the Fed massively increased the money supply and triggered large inflation then NGDP would probably go up but real GDP (which is what I really care about) might well go down.

When Real GDP goes down, but Nominal GDP is strong, the Fed is truly impotent and supply-side reforms are needed. But that's not the situation we're in today! The Fed is not impotent, and NGDP is weak.

Stable NGDP and falling RGDP is a more sensible measure of monetary impotence than base increases or "low" rates relative to RGDP/unemployment. It doesn't do much good to assess the monetary stance from a specific action at a discrete time.

Actually it pretty clearly points to "unwilling". It was easy enough for Abe to increase inflation when he really wanted to. If you think the U.S. or Japan are using all available tools to increase inflation to their supposed targets there is something wrong with you. They tapered QE didn't they?

If NGDP was falling but the price level was rising in such a way that NGDP was below target, target shouldn't the Fed turn a blind eye to inflation at that conjuncture? Which historical conjuncture does jorgensen think NGDP targeting would have led to sub optimal policy?

>If the economy wanted to grow at three percent and the central bank set the rate at 2%

Why in the name of the lord would the CB target 2% NGDP growth? Nobody is advocating that.

The MM's have gotten more right during/post-crash than any other "school", including what happened in the UK, in the EZ, in the US, in JP and other places. Who else has that kind of track record?

The whole approach of NGDPLT is to leave some headroom (the inflation part) above expected real growth so that ebbs and flows can happen without deflation.

If the Fed starts raising while inflation is still so far sub-target, we'll likely experience a recession that will be chalked up to "bad luck". We can avoid that.

We like Sumner because of posts like this:


Inflation is a component of NGDP. Think it as a growth-adjusted inflation target.

Remember, inflation will actually be lower than today during high growth.

Absolutely this is something Scott can and should be proud of.

Sure - you spout nonsense for a long time and some people start treating you like a serious person, I suppose he can be "proud" of that.

So the logic of it is instantly appealing to anyone who actually understands central banking and who doesn't have ulterior motives. That its taken this long to get mainstream public press tells you what?

It tells you that the opposition comes from those who did not want more stimulative monetary policy in 2008-2015. Time will tell if "permahawks" retain their allegiance to tight money if a Republican is ever in the White House during a recession. Opposition to deficits evaporated pretty quickly after 2000.

"It tells you that the opposition comes from those who did not want more stimulative monetary policy in 2008-2015."

Exactly how could we have had more monetary stimulation than we did?

NGDP targeting assumes that the Fed could have done more - what more?

We play a game where we think monetary stimulus is the only (acceptable) kind.

$40 trillion per hour QE, -3.5% nominal Federal Funds Rate.

Could have not done interest on reserves in the midst of a liquidity crisis, for one.

The Fed held the Fed Funds Rate at 2% through most of 2008 because of inflation fears, even after Lehman failed for lack of liquidity. At that point, the bottom finally fell out, and even after that they started paying interest on reserves of 1% until mid December. To paint 2008 as a period where monetary policy was pushed as far as it could go is revisionist history.

"To paint 2008 as a period where monetary policy was pushed as far as it could go is revisionist history."

Fair enough. I stand corrected. The Fed could have done more, sooner, in 2008. I had a quick look at https://research.stlouisfed.org/fred2/series/FEDFUNDS and the Federal Funds rate was falling like a stone throughout 2008. I know that I personally put about half a million dollars into the market in October 2008 thinking the bleeding had stopped (I was wrong). By January 2009 the effective rate was down to 0.15.

The big thing would have been for the Fed to stop implying that they are satisfied with current nominal progress. Bernanke and Yellen have done some inscrutable things, such as linking contractionary expectations to calendar dates, while simultaneously releasing forecasts showing target underperformance until 2017. They released the ambiguous Evans Rule, and then made it less ambiguous by pointing the Phillips Curve and consistently overestimating NAIRU.

As a result, the market had to discover that 2% PCE inflation was not a target with symmetric monetary responses, but a ceiling. The search for the true acceptable channel invited unnecessary volatility, while also adding vulnerability because it hints that the Fed is not interested in a nominal level defense. In the next big recession, a Bernanke/Yellen type might allow the level to collapse again, and then reset the inflation growth rate target at another low channel. Maybe then, at least, the future chairperson will be explicit with the true target.

I'm confused by the discussion of stimulative monetary policy suddenly jumping to fiscal deficits.

You do all realize that the only way to target a higher NGDP growth rate while at the zero bound is with fiscal stimulus?

Sumner's pounding on the table to target a higher NGDP growth rate while simultaneously attacking the value of fiscal stimulus has never made one iota of sense.

Who wants to take the opposite side of this bet: Any developed country that adopts NGDP targeting will adopt helicopter money.

"You do all realize ..."

No they don't.

"Who wants to take the opposite side of this bet: ... "

Not me. :-)


-As the good Scott Sumner says, which way is more effective at tripling NGDP in a year: fiscal or monetary stimulus?

I think the answer should be obvious.

Fiscal stimulus! That's a good one. Had no effect during the 1930s, had no effect in 2013, had no effect in the 1990s, had no effect in 2008. Yet, somehow, "the only way to target a higher NGDP growth rate while at the zero bound"? You have evidence for that, right?

QE is helicopter money.

Also failed spectacularly in Japan. Who would have believed in 1991 that the Japanese growth would be zero for the next 25 years?

"....the bet.." ??

This is economics science, not gambling
Keynesian central banking has never failed to deliver

Since apparently not everyone understands what "helicopter money" means, it is monetary financing of a fiscal expansion.

QE is monetary financing independent of fiscal policy, ie substituting monetary financing for debt financing without changing public spending levels.

I'm not advocating fiscal expansion, I'm telling you that public authorities can't target a higher rate of NGDP from the zero lower bound without fiscal expansion.

@jorgensen - And it looks like none of them will take the bet either.

"public authorities can’t target a higher rate of NGDP from the zero lower bound without fiscal expansion."

Of course they can. The Fed can bid on every asset in existence. Fiscal spending could fall to zero and the Fed could still target NGDP.

Central banks can always inflate.

What you're describing is a central bank that determines fiscal policy by lending directly to the real economy.

The central bank can't "always" do that because of legal restrictions, and where there are none, the likelihood that legislators would enact them if the central bank crossed traditional lines into fiscal policy.

Besides, it would be a totally stupid policy. You don't like fiscal stimulus, but you want the FOMC to design industrial policy by buying corporate bond issues at politically determined prices? This is how you counter that NGDP targeting from the ZLB can work?

The Fed, in the crisis circumstances of late 2008 and 2009, established the Commercial Paper Funding Facility and purchased 700 Billion in commercial paper as a way of injecting liquidity into that market. I think that was an appropriate and prudent thing to do (me and Bagehot :-) ) .

More recently, the Fed bought mortgage backed securities as a direct intervention in the housing market. I think that was either over the line or dangerously close to the line of crossing into fiscal policy. Even though I think Congress is bat shit crazy I am uncomfortable with the bureaucrats trying to grab the reins.

Somewhere in between those two (for me) is the possibility of the Fed making deals with state and local authorities that the Fed will buy 20 and thirty year bonds issued to pay for infrastructure repair and replacement if the private sector will not take them.

Helicopter money isn't really lending, which is a financial arrangement that increases velocity for a given base. If you are printing the money, then its distribution represents an increase in liabilities from existing money holders. The power of it comes from the credible debasement, not the method of distribution.

No, I'm merely describing a way to produce inflation.

A monetary sovereign can bid on anything it decides to. Legislators can tell the CB to do whatever legislators want.

Fed purchases are on the open market, they do not control prices.

You haven't set the terms of the bet dipshit

For the bet to be on, a country with GDP/capita over $30k has to explicitly adopt NGDP targeting.

If the country doesn't use fiscal expansion when it wants to accelerate NGDP growth, I lose.

And how will it be determined if fiscal expansion has been employed?

"You do all realize that the only way to target a higher NGDP growth rate while at the zero bound is with fiscal stimulus?"

Interest rates are not the only monetary policy tool, the mythical liquidity trap is just rank statism masquerading as Keynesian economics.


You're thinking needs some Yoga.

If QE does not result in greater economic growth or inflation, we still have the phenomenon of a central bank monetizing the national debt.

Who does not like monetizing the national debt without consequence?

Even if we assume that QE is temporary (as is likely), it still amounts to a tax cut in the long run.

The Economist is not actually written by economists, is it?

The Economist championing NGDPLT by Sumner is a futile and/or dangerous idea. First, NGDPLT is a harmless idea unless you overdo it, since money is neutral (Google Ben S. Bernanke 2003 FAVAR and note the conclusion therein*, from January 1959 through August 2001: the Fed has little influence over the real economy, at most between 3.2% to 13.2% of any increase).

Second, at some point NGDPLT is the same as the Taylor Rule or inflation targeting, which haven't worked now, so you must assume for NGDPLT to work that the economy 'springs to life' when certain rates of money printing occur, but remains below trend at other times. This is nonsense in most systems, it defies common sense.

Finally, NGDPLT is placebo and might work temporarily just for that reason: a study found that both raising the lights in a factory then dimming them back to the original level temporarily increased productivity. Nobody knows why, some studies fail to replicate the results, but the thinking is that people respond positively when somebody "just does something", like FDR supposedly did, even if it's just a fireside chat. So NGDPLT is an idea worth trying, until it is found out it does nothing special. The danger is that taken literally, if the Fed tries to print more and more money to reach some arbitrary target, and nothing happens, money no longer will be neutral but you might get hyperinflation. However, you'd need to have really excessive money printing for this to happen, accelerated money printing, which I doubt the Fed would do.

So Edmund Burke's criticism of the French Revolution is on point here with NGDPLT: while it might be good--for placebo reasons--to try it, in the same sense the French Revolution was a good idea in theory, but, in practice, it might lead to excess (excessive money printing, which might trigger hyperinflation, the exception to money neutrality), with the same results as the Reign of Terror.


* Measuring the Effects of Monetary Policy: A Factor-Augmented Vector Autoregressive (FAVAR) Approach * Ben S. Bernanke et al (2003) “Apart from the interest rates and the exchange rate, the contribution of the policy shock is between 3.2% and 13.2%. This suggests a relatively small but still non-trivial effect of the monetary policy shock. In particular, the policy shock explains 13.2%, 12.9% and 12.6% of capacity utilization, new orders and unemployment respectively, and 7.6% of industrial production” (page 24)

How hilarious is it that you keep quoting an article that says money is NOT neutral as evidence that it is superneutral

Ray has been shaping that comment for about three months now.

"at some point NGDPLT is the same as the Taylor Rule or inflation targeting"

Agreed. NGDP targeting is just new packaging for inflation targeting.

If you think this you have serious problems. If you have an NGDP target of 4% what is your inflation rate in year 1 with 3% RGDP growth and in year two with -1% RGDP growth?

Well Cliff, let me ask you this. How do you pick the NGDP target? What assumptions about growth and inflation go into arriving at the target and how do your expectations about real growth influence the NGDP target?

At best, NGDP is simply a way of trying to make a higher target inflation rate politically acceptable. But I, for one, do not believe in the magical properties that Krugman and De Long and others ascribe to inflation as a way to improve the economy.

Common sense is starting to prevail.

NGPDLT is about as close to a policy-less regime as we can come with fiat currency.

No, NGDP targeting is deeply, deeply stupid.

It tells you that if real growth accelerates, you should assume without any regard for circumstances that the growth acceleration is unsustainable and crush it by tightening.

It tells you that if real growth slows, you should assume without any regard for circumstances that credit is too tight.

And, at least as you're defining it, it tells that if NGDP growth is still below target after interest rates fall to zero, the central bank should make fiscal and industrial policy by lending directly to businesses (and consumers?) on political terms.

And at the same time you harp on how terrible Keynesian policy is? I'm no Keynesian, but they at least have some kind of logical consistency. You Sumnerites are koo koo for cocoa puffs.

Anyway, to get back to my original point.

NGDP targeting is never going to be used to fight an acceleration of real growth. If by some bizarre turn of events a country with NGDP targeting experienced real growth acceleration, it would abandon NGDP targeting immediately.

NGDP targeting is never going to be used during the middle of a normal cycle with interest rates above zero. Even if by some bizarre turn of events a country formally adopted NGDP targeting during a period of economic normalcy, the implementers would anyway actually target inflation and employment, not NGDP growth.

The only real attraction of NGDP targeting is its promise to generate higher spending at the zero lower bound when standard monetary policy and QE have failed. And the only way NGDP targeters can do increase spending at the ZLB is through fiscal policy, whether that be standard fiscal policy run by government or a de facto fiscal policy run by the central bank. Having the central bank generate spending by lending directly to businesses and consumers is just a dumber kind of fiscal stimulus.

All you guys are really doing is arguing for fiscal stimulus out of one side of your mouth and arguing against it out of the other. You don't understand what you yourself are saying.

Tom - a couple of points;

If real growth "accelerates" for some reason so that the real growth is above the NGDP target, then what you would see would be price deflation. The NGDP approach would be fine with this deflation occurring as long as NGDP was on track, i.e. there would not be any reason to raise or lower interest rates. This is better than what would happen with the current inflation targeting approach which would be trying to get inflation back back to target and would therefore be engaging in pro-cyclical stimulation. Another approach is to take the "wise man" running the Fed's "gut" view. Who knows how that would work? The gold standard would also be completely random as well, might be inflationary in such a circumstance, might be heavily deflation. Given these alternatives I know what I would prefer. What approach do you favour?

But realistically, if we set NGDP target at 5%, won't it be a nice surprise if real growth is consistently above 5%? If you really think this is going to be a frequent issue, and you hate price deflation for some reason, we can set the target at 7 or 8% if you insist.

On effectiveness after rates fell to zero, purchasing of government assets, such as debt seems pretty benign. In fact any fiscal conservative should welcome this approach. I suppose in the long run it could have an anti- "starve the beast" element, where a future government now has room to spend more but surely that is true of any debt reduction program, even one financed by taxes. Should we never try to reduce government debt because of this?

Chris, you've done a great job explaining why no central bank would target NGDP growth in normal times. They would seek the highest real growth rate possible, as you say. The bound would be some limit of acceptable inflation, never of real growth. Only of course there is no reason to imagine higher real growth causes disinflation or deflation, the opposite is true.

Again though you're all ignoring my central point. NGDP targeting is a leftist proposal. It aims to solve the lack of monetary policy power at the ZLB by allowing the central bank to step into fiscal policy in some way or another, usually simply by monetary financing of fiscal expansion aka helicopter money. The whole point of it is to overcome the long established separation of monetary and fiscal authority. NGDP targeting is a cousin of MMT and well to the left of Keynesians.

And here you are the Sumnerites, a bizarre little subsect of NGDP targeters who fervently believe in NGDP targeting and at the same time profess to be anti-left. You want to give one single public authority the power to create and arbitrarily distribute money because and at the same time you attack Keynesians for being too ready to concentrate piwer in the hands of public authorities.

I am going to pile on - ngdp advocates seem to assume that the Fed is single mindedly focused on the inflation number. I assume that in the real world the Fed is looking at estimates of real gdp, investment and unemployment in addition to the inflation number.

Comments for this post are closed