Is NGDPLT a perfect guard dog? A challenge.

Can anyone think of a single historical example, in any country, at any time, where NGDPLT would have failed as a guard dog?

Is targeting the level-path of Nominal GDP a perfect monetary policy? Almost certainly not. I would be really surprised if something as crude as NGDP turned out to be the very best of all possible targets in all possible circumstances. There are just so many different possible target variables and so many different possible circumstances. And I can imagine a world in which NGDPLT wouldn't be the best possible monetary policy, and build a model to "prove" it. But I want to ask this as an empirical question about the real world.

Even as an empirical question, perfection is a lot to ask. There may be no candidate who is ideal in all real world circumstances. All we can reasonably hope for is that we choose the candidate who is less imperfect than all the other candidates, and who makes fewer mistakes and smaller mistakes.

But I'm going to ask for empirical perfection anyway, just to see what happens.

In my previous post, I made the analogy between monetary policy target variables and guard dogs:

"What we are looking for in monetary policy is some target variable, and some numerical target for that target variable, so that deviations of the target variable from target act as a good guard dog that barks loudly and clearly when, and only when, there are undesirable real effects from the interaction between monetary policy, nominal rigidities, and whatever other shocks happen to be hitting the economy. (And that target variable has to be a nominal variable, of course, because of long run neutrality.)"

There are two ways a guard dog can fail: it can bark when there is no burglar; it can fail to bark when there is a burglar. Type 1 errors; type 2 errors; (and type 3 errors, when you forget which is which.)

For example: the 1987 stock market crash, which was not followed by a recession, is an example of a type 1 error made by stock prices as a possible target variable for monetary policy. The guard dog barked loudly, but there was no burglar. Monetary policy was not too tight, despite what that guard dog said.

A second example: the 2008-9 Canadian recession, where the inflation rate failed to fall below target, is an example of a type 2 error made by the inflation target. There was a burglar, but the guard dog failed to bark. Monetary policy was too tight, despite what that guard dog said.

Can anyone think of a single historical example, in any country, at any time, where NGDPLT would have failed as a guard dog?

Let's see how much dirt we can find on our candidate.

Comments

[Benoit Essiambre]

I'm not an economist so I don't think I can help very much but I am quite curious to know what people can find. This would help answer more comprehensively the question I asked yesterday.

[Nick Rowe]

Benoit: yep. On my previous post your objection was that I had a sample of one! Which was a fair point. Now in truth my sample was a bit more than one (even if that one data point was a very clear data point), because I knew that NGDP had looked fairly good as a potential guard dog in other countries and other times as well. But now I want to draw the sampling net as wide as possible.

[Mike]

"A second example: the 2008-9 Canadian recession, where the inflation rate failed to fall below target, is an example of a type 2 error made by the inflation target. There was a burglar, but the guard dog failed to bark. Monetary policy was too tight, despite what that guard dog said."

Didnt inflation decline significantly during 2009? This chart from statistics Canada indicates Canada experienced deflation.

http://www.statcan.gc.ca/daily-quotidien/130719/dq130719a-eng.htm

[Vladimir]

Following on Mike's point, I think Nick means core inflation rate targeting. (Do you Nick?) Otherwise even something like a 1970s oil shock could in the short run drive NGDP over a target rate yet be contractionary.

[Nick Rowe]

Mike: Yep. But look at inflation in 2008, where it was well above target, and how it went back to target in 2010, and above target again in 2011. And yet the Canadian economy is barely back to normal now, in late 2013 (just as inflation falls below target again). Total CPI inflation moves around a lot on a short term basis, and so the Bank of Canada looks at core inflation as an "operational guide". Look at the graph for core CPI in this old post and you see it keeps growing steadily with no signal of the recession.

Let's ask the question another way: suppose the Bank of Canada had a crystal ball, which let it see that CPI (total and core) data in advance. When would it have tightened, and when would it have loosened? Then, given the lags in monetary policy, would that have done any good? Now ask the same question about the NGDP data.

[Mike]

If you look at m3. The stabilization of broad money coincided with the decline in inflation. So in that instance there appears to be no lag in MP.

http://www.tradingeconomics.com/canada/money-supply-m3

[RPLong]

How about the Industrial Revolution. NGDPLT would have kept growth far lower than it needed to be. Horrible guard dog.

[Frank Restly]

Nick,

I guess I don't understand the question. A guard dog should bark before the burglar is standing in your bed room with a gun to your head. The guard dog should be a leading indicator not a coincident indicator - thank you Sparky for barking but I can feel the cold steel against my temple. How do deviations from a GDP target function as a leading indicator?

"And that target variable has to be a nominal variable, of course, because of long run neutrality."

A monetary policy target variable has to be money neutral because of long run neutrality. It does not have to be a nominal variable. Even nominal variables like GDP are measured in the units of account (dollars, pesos, etc.). A central bank could instead target a unitless ratio - $ GDP per $ Debt or a relative price $ Barrel of oil per $ Ounce of gold.

[Lord]

A gold standard would have to have been an even worse guard dog, and a coincident indicator is still better than a lagging or absent one.

But I can see better ones, such as NGDP per capita PLT or NGDP per workforce PLT, along the lines of SRW latest Interfluidity post.

[Tom Brown]

"But I want to ask this as an empirical question about the real world."

Nick I can't believe you're sullying your pristine blog by requesting pedestrian empirical information about the real world! The real world?? This is where I come to read about economies consisting only of gold, silver and haircuts!! Not the real world. You disappoint me Nick! ;^)

But seriously, this sounds like a GREAT exercise for Mark Sadowski... that guy seems to be able to come up with examples or counter examples to anything at the drop of a hat!

[Tom Brown]

BTW, you ever finish thinking about Sumner's MOA vs MOE examples? I'd love to see a response to DOB too:
http://worthwhile.typepad.com/worthwhile_canadian_initi/2013/08/banks-and-the-medium-of-exchange-are-both-special-or-neither-special.html?cid=6a00d83451688169e2019aff322b8a970d#comment-6a00d83451688169e2019aff322b8a970d

[Mark A. Sadowski]

RPLong
"How about the Industrial Revolution. NGDPLT would have kept growth far lower than it needed to be. Horrible guard dog."

Really?

The dates for the industrial revolution vary but it is generally considered to start in the late 18th century. The Second Industrial Revolution is generally considered to run from 1870-1914.

NGDP is estimated to have been $189 million in 1790, $7,812 million in 1870 and $36,831 million in 1914:

RGDP is estimated to have been $4,351 million in 1790, $121,309 million in 1870 and $598,287 million in 1914 in 2009 dollars.

The GDP deflator is estimated to have been 4.34, 6.44 and 6.16 in 1790, 1870 and 1914 respectively:

http://www.measuringworth.com/usgdp/

NGDP growth averaged 4.76% in the first Industrial Revolution and 3.59% in the second. RGDP growth averaged 4.25% in the first industrial revolution and 3.69% in the second. Inflation averaged 0.49% in the first Industrial Revolution and (-0.10%) in the second.

So if NGDP growth had been set at 5.0% a year during the Industrial Revolutions it probably would have led to slightly higher but still low inflation in the first Industrial Revolution, and low inflation instead of grinding deflation in the second (which is a period which includes something which is popularly referred to as the Long Depression (1873-96)).

[Tom Brown]

Mark, can you find the examples Nick is looking for?

[Mark A. Sadowski]

Tom Brown,
I’m insufficiently motivated to take up Nick’s challenge because I simply don’t believe there is a good counterexample.

For example, if memory serves (I’m not sure about the exact cutoff), in every recession in the US since the end of WW II year on year NGDP growth dipped below 0.8% with the exception of the 1974-75 recession, and everytime it dipped below 0.8% there was a recession. So not only is NGDP a guard dog that has a near perfect record of recession barking, it doesn’t bark when there isn’t one.

As for the 1974-75 recession, that was a giant negative AS shock related to the energy crisis that was further compounded by tight monetary policy. The recession certainly would have been less severe with NGDPLT even if it were it not completely avoided.

[Tom Brown]

Nick, I think you're in trouble on this challenge if Mark is giving up already!!

[Tom Brown]

Common Mark... wouldn't it be fun to put on the black hat for a day? Give it your best shot at playing the bad guy? ;)

[RPLong]

Mark, thanks for the numbers. What happens when we keep people spending at 5% annual increases between the two industrial revolutions? Would the foregone savings have impacted the second, or would the second have been equally as large despite a smaller capital stock?

I'm not a macroeconomist so forgive me if those are dumb questions.

[Mark A. Sadowski]

It's not question of good or bad. In my opinion Nick's challenge is a Snipe Hunt. (I'm no chump.)

[Gustav]

What would have been the impact of adding a mere 1.4% a year in inflation to the economy during a 44 year period of economic tumult and structural adjustment?

One answer is, the economy would have been more stable as a result.

A competing answer is, a NGDPLT target might have produced all manner of unintended consequences.

For instance, goods prices want to decline as production scale and transportation efficiency increases. What is the impact on relative prices of an NGDPLT target? On income distribution and political outcomes? What happens to the term premium, maturity transformation and financial intermediation if inflation becomes volatile? How do international payments work if inflation and therefore the dollar FX rate are more volatile? What about international capital flows? If they are disrupted, how does that affect the financing of long term projects such as the railroads?

The burglar might be the least of the homeowner's worries.

[Mark A. Sadowski]

RPLong,
"What happens when we keep people spending at 5% annual increases between the two industrial revolutions? Would the foregone savings have impacted the second, or would the second have been equally as large despite a smaller capital stock?"

Investment in capital stock is a form of spending and is a component of GDP. In fact investment is adversely impacted by output gaps. In fact I strongly suspect investment as a proportion of GDP would have been higher (and consumption lower) in the second industrial revolution if NGDP growth had been steady at 5% because capital utilization would have been more consistently higher. Thus the rate of growth in potential GDP likely would have been higher.

[apt]

I am not sure, not even close to be sure, about this, but I think that it would increase the already high inefficiency of the Eurozone, and it would make room for more and more inequality between the countries' competitiveness. I know this is not an historical example, and well, it's just a try.

[vincecate]

Whatever rules are supposed to govern the central bank are changed or ignored to get hyperinflation. When central banks are setup people realize that if the central bank funds the government deficit this would be bad, so they always make rules to prevent that. But the government appoints the people that run the central bank and the government makes the laws that control the central bank. So if the government is spending twice what they get in taxes, like say Japan, and the government needs the central bank to buy their bonds, it always happens. Note how the politicians got their man running the central bank and how the "bank note rule" that limited money creation is now ignored. This is why NGDPLT would fail too. The currency would still die a hyperinflation death at some point.

http://howfiatdies.blogspot.com/2012/10/faq-for-hyperinflation-skeptics.html

[Mark A. Sadowski]

apt,
"I am not sure, not even close to be sure, about this, but I think that it would increase the already high inefficiency of the Eurozone, and it would make room for more and more inequality between the countries' competitiveness. I know this is not an historical example, and well, it's just a try."

First of all, just to make sure we're on the same page, NGDP in the eurozone is way below trend:

http://research.stlouisfed.org/fred2/series/EUNGDP

So NGDPLT would advocate that the ECB conduct expansionary monetary policy in order to get NGDP back to trend. How would this impact the differences in trade competitiveness between the periphery and the core of the eurozone?

For that I turn the microphone over to Paul Krugman, who answers that question far better than anyone I've ever seen (I just wish he talked about NGDP and the AD-AS Model more often):

http://krugman.blogs.nytimes.com/2012/07/29/internal-devaluation-inflation-and-the-euro-wonkish/

July 29, 2013

Internal Devaluation, Inflation, and the Euro (Wonkish)
By Paul Krugman

"I’ve been writing for a long time about how the euro area needs more inflation. But I suspect that many readers don’t quite see how this ties into the macro story. So here’s something that may or may not clear things up — a stylized little model linking euro inflation and the adjustment problem to overall monetary policy. It’s very stylized, making some obviously untrue but I think still useful assumptions, and I have been finding that it clarifies my own thinking, at any rate.

So, imagine a currency area with just two countries, Spain and Germany, which I’m going to represent in an aggregate supply-aggregate demand framework.

On demand, I’ll make two assumptions I don’t believe. The first is that the ECB can determine nominal GDP for the euro area. Under liquidity-trap conditions, this is a very problematic assumption, and I don’t mean to drop my skepticism for other purposes. For right now, however, it’s useful, I think, to use nominal GDP as a proxy for the whole range of possible expansionary policies the ECB might follow.

By assuming that the ECB chooses nominal GDP, we get an aggregate demand curve for the euro area as a whole: Py = Y, where P is the price level, y is real GDP, and Y is the target nominal GDP. Now for my second assumption: Cobb-Douglas preferences, so that a fixed share of total spending falls on Spanish and German output respectively. This doesn’t have to be true, and surely makes no difference to the final conclusion — but it means, given the overall target, individual-country AD curves reflecting that fixed division of the total.

Meanwhile, on the supply side, I’ll assume that prices can rise but not fall (because of downward nominal wage rigidity), so that the AS curve in each country is a backwards L.

Given all that, the current situation looks like this:

[Graph]

Spain is deeply depressed, while Germany is close to full employment.

How can Spain restore normal employment? The current European strategy is “internal devaluation”, which means expecting Spain to cut wages and thereby restore competitiveness; this can be represented as a downward shift in Spain’s AS curve (with the new curve in red):

[Graph]

The trouble with this strategy is twofold: it’s really, really hard to get wage cuts, and deflation in Spain makes the problem of debt overhang worse.

What’s the alternative? Aggressively expansionary monetary policy, which shifts the AD curves of both countries to the right. This raises output and employment in Spain, but leads to inflation in Germany:

[Graph]

I’d argue that this is a much better outcome. But it does mean a rise in the overall euro price level, because of the asymmetry between the effects of higher demand and lower demand in the face of wage stickiness. In practice, because these developments would play out over time, following the path of higher demand would mean that the ECB would have to accept temporarily higher inflation.

And of course the difference between these two strategies demonstrates what a really bad idea it is for the ECB to have a mandate that only takes account of price stability, with no consideration for the real economy.

Just in case you’re wondering, this little model does not in any way contradict what I’ve been saying in other euro analyses — it’s just a different angle of approach. But it may be helpful. And trivial as the model is, I think it suggests just how important it is that the ECB find some way to escape its stable-price shackles."

[Frank Restly]

Mark,

"How can Spain restore normal employment? The current European strategy is internal devaluation, which means expecting Spain to cut wages and thereby restore competitiveness; this can be represented as a downward shift in Spain’s AS curve (with the new curve in red). The trouble with this strategy is twofold: it’s really, really hard to get wage cuts, and deflation in Spain makes the problem of debt overhang worse."

How can Spain restore normal employment? Simple, expansionary fiscal policy funded with equity in lieu of debt. It lowers the after tax cost of servicing private debt, it generates demand for goods produced domestically via government expenditures, and it short circuits any attempt of a "bond market revolt".

Spain still controls it's own fiscal policies last time I checked.

[Mike]

I seems headline inflation is a good measure to target. It don't think it fluctuates excessively. In the depths of the recent crisis which was rather extreme the most it moved in one year was from 1.5% to -1%. Money is neutral so a 0% target would be the best IMO.

[Mark A. Sadowski]

Frank Restly,
How would a Spanish fiscal stimulus correct Spain's trade competitiveness imbalance relative to fellow eurozone member Germany (for example)? (Recall that was essentially the gist of apt's original question.)

It wouldn't, would it? In fact, if you think about it using Krugman's AD-AS Model explanation, it's clear that it would almost certainly make it worse.

[Seamus Hogan]

Nick: I don't know the data, but the place I would look for a NGDPLT guard dog that barked when it shouldn't have would be the developing tiger countries in the period when their rapid growth stared (Japan first, S. Korea et al later, China more recently). In Western countries, how about the monetary response to the OPEC shock and productivity slowdown of 73/74?

[Nick Rowe]

Seamus: I don't know the data either, but you might have a point there. Suppose they had been targeting (say) 5% NGDPLT. Given the very rapid and sustained RGDP growth rates from industrialisation and adopting modern technology, RGDP growth rates of (say) 10% would have implied 5% deflation rates. There is that "good deflation vs bad deflation" debate, but 5% deflation does sound rather a lot, and might have made the ZLB a binding constraint. (I take it as given we cannot say that the Tigers' rapid growth was totally from an unsustainable money-fuelled boom, even though, obviously, it couldn't last forever given "catch-up".)

[Bob Murphy]

Nick,

This is a really neat post, and I plan on writing something next week on my blog about it. In the meantime, though, can you clarify *what it would look like* for NGDPLT to fail?

For example, my first thought was that we'd need to see a country suffer a typical business cycle, even thought NGDP kept growing at trend. But wouldn't Scott Sumner just say, "Well, that's clearly a supply-side recession. Nothing the central bank can do there." ?

Then, what if we found a country where the central bankers explicitly told the public they were targeting level NGDP, but then in practice they realized it was impossible, and a recession occurred because targeting NGDP level leads to recessions. Even here, Scott would point to the time when it broke down and say, "Ah, they stopped targeting NGDP level. I'm right again."

So I'm thinking it might be literally impossible to find an example of what you mean. In fact, I think Scott once wrote a post saying (somewhat tongue in cheek, but not really) that even if the Fed did what he said, and we got huge price inflation, that it would still be the right policy.

[Nick Rowe]

Bob: thanks! Good question, but I'm afraid I don't have a good answer. For example, it is always possible for someone to say that the Canadian 2009 recession wasn't really a recession, but was an optimal response to the rest of the world having a hissy fit, and that the best thing for Canada to have done was to have reduced output and employment, and increased unemployment, by exactly the amounts it did. That certainly doesn't sound plausible to me, but finding some cast-iron empirical criterion doesn't seem feasible. We just have to be like historians, and weigh the arguments each way.

[Frank Restly]

Mark,

"How would a Spanish fiscal stimulus correct Spain's trade competitiveness imbalance relative to fellow eurozone member Germany (for example)? (Recall that was essentially the gist of apt's original question.)"

Government intervention is always construed to generate aggregate demand - government spends money on goods. It can also be used to subsidize the production of goods - see farm subsidies in the United States. The financing decision becomes important here because a lot of Spanish Debt is held by German banks. If the Spanish government tries to subsidize production and fund it with debt, it could face retaliatory actions by the owners of it's debt - German banks. So the Spanish government should subsidize production with and fund those subsidies with equity instead.

[Bob Murphy]

Nick,

I think you're making my question deeper than it was. We can stipulate what a recession is. What I'm saying, though, is what do you mean when you are asking for an example of NGDPLT either falsely warning, or failing to warn, of a recession?

For example, do you want me to find a country that had a recession, even though the growth of NGDP was at trend throughout the whole episode?

Going the other way, do you want me to find a country that had NGDP grow above or below trend, and yet didn't have a recession?

See what I'm saying?

[Chun]

It is not really NGDPLT and it is not monetary policy implemented by central banks, but several developing countries' governments announce their annual NGDP growth goals. And they often fail to achieve them. Do they count?

[Mark A. Sadowski]

Frank Restly,
The problem with that solution is that it only works so long as you continue to subsidize production. In other words it would have to be continued indefinitely.

[Frank Restly]

Mark,

There is no reason to believe that it couldn't continue indefinitely. Farm subsidies in the U. S. have existed since the 1920's. Granted the political will must exist for the Spanish government to go that route, but high unemployment tends to bend political will.

I think you and I can agree that any solution implemented by the Spanish government must be internally financed. Otherwise, that solution could be undone by reciprocal measures from external financiers.

[JKH]

How would it have worked in the US at the peak of interest rates in 1981?

[Frank Restly]

JKH,

Who was your question for?

[JKH]

Sorry - for Nick or anybody - to the general point of the post, but I'm not even sure how to frame the question more specifically in the context of what was happening at the peak of interest rates in 1981

[Nick Rowe]

Bob: Basically, yes. Let me just change what you wrote very slightly:

'For example, do you want me to find a country that had a recession [or boom], even though the growth of NGDP was at trend throughout the whole episode?

Going the other way, do you want me to find a country that had NGDP grow above or below trend, and yet didn't have a [boom or] recession?'

where "boom" is defined as the opposite of recession: some sort of unsustainable over- or under-production due to bad monetary policy.

We will, of course, argue about whether a particular episode really was a boom or recession, or just a fluctuation that was caused by a meteorite strike and had nothing to do with monetary policy being too tight or loose.

[Nick Rowe]

JKH: IIRC, NGDP fell sharply below trend in the 81 recession. I think NGDP works well there. It barked, and there was a recession.

[apt]

But just imagine a situation where the Eurozone defines a NGDP target for the zone as a whole, with a given distribution among its members. But now also imagine this: countries like Greece, Portugal, Spain or Italy are lagging, so their real growth is far from desirable, and therefore they would need big amounts of inflation to fulfill their share. In that case, they would be left even further behind than they already are, or at least no improvements would be made.

On the other hand, imagine just a NGDP target for the zone as a whole, períod. Then if some countries lagged in terms of real growth, and could not get inflation because of their own output gap, then other "healthy" countries would have to "make up" for it. It could be great, and could do a lot in terms of competitivness. But what about in an extreme situation (probably unrealistically extreme, but possible)? Couldn't that lead to hyperinflation if the burden(a big one) was concentrated in only one country?

[W. Peden]

apt,

That's an imperfection of NGDP targeting, but it's also an imperfection of any macroeconomic regime, and can only be solved via supply-side policies e.g. free movement of labour, regional policies and inter-governmental subsidies.

[W. Peden]

(At the extreme, for example, even in a micro-state the price of one house in the most picturesque location in the nation can be shooting ahead when other property prices are falling dramatically. In that case, one cannot have a macroeconomic regime that prevents high inflation of the price of that particular house.

[apt]

I do agree that NGDP targetting would be the best choice, but the "challenge" was to find ways in which it was not perfect. It is not exclusive of NGDP targetting, but it is an imperfection of the regime. Also, I think that it brings and extra disavantadge: when you target inflation, it is just inflation, and it makes no sense for you to go for other countries' "share". But when you target nominal growth, it may make sense.
This does not mean I prefer the current regime, not even close to that, I am just raising some possibilities.

But if you ask me, I personally think that the ECB should also target M/W ratios, probably even in adition to NGDP. It is, as far as I am concerned, the best way to connect the real economy to monetary policy.

[Frank Restly]

Nick,

http://en.wikipedia.org/wiki/Recession#Definition

"The NBER defines an economic recession as: a significant decline in economic activity spread across the economy, lasting more than a few months, normally visible in real GDP, real income, employment, industrial production, and wholesale-retail sales."

And so is the question - Has there ever been a decline in real GDP (recession) without a decline in nominal GDP (barking dog). Yes there has been.

U. S. Nominal GDP (Beginning of 3rd quarter 1973): About $1.49 trillion
U. S. Nominal GDP (Beginning of 2nd quarter 1975): About $1.65 trillion

U. S. Real GDP (2009 adjusted dollars) (Beginning of 3rd quarter 1973): $5.44 trillion
U. S. Real GDP (2009 adjusted dollars) (Beginning of 2nd quarter 1975): $5.3 trillion

Dog (declining nominal GDP) did not bark warning of recession (declining real GDP).

[Declan]

WW2 in any major combatant (admittedly exaggerated by coming out of the Depression, but still).

Australia in the last decade. Any country with a resource boom (Arab states in 70s, Mongolia now) or a really big productivity shock (can't think of any offhand).

Basically anywhere hitting the NGDP target would require absolute nominal wage cuts, or even big cuts in nominal wage growth.

[W. Peden]

Frank Restly,

NGDP doesn't have to fall for NGDPLT to bark; there just has to be a sharp shift in the rate of NGDP growth. It is true, though, that there was a major supply-side elements to the 1974-1975 recession in the US.

[Frank Restly]

W. Peden,

"...there just has to be a sharp shift in the rate of NGDP growth"

Now you are into trend identification. But which trend - 5 year, 10 year, 30 year, 50 year? The rate of growth of nominal GDP fell during the 1973-75 recession, but that rate (7.5% NGDP growth rate in 1975) was about average from 1960-1970. And so trending from the 1960-1970, period, the nominal GDP growth rate did not fall off.

[scepticus]

NGDP may be a good guard dog during a general regime of population growth and upward trend in general energy usage and continuing efficiency gains. So, any time from the green revolution of the 18th century onwards.

Maybe in the regime in place in Europe from 1350 to 1650, one of flat population growth, plagues, and hard constraints on land usage and available energy would be less condusive to NGDPLT.

I think the net for counter examples needs to be cast a little wider, given that the benign conditions of recent history should not be taken as axiomatic.

[Andrew F]

Declan, Australia's real GDP growth over the past decade has consistently been below 5%, so a 5% NGDPLPT would still allow for low positive inflation, not deflation.

Maybe for China, a NGDPLPT rate of increase of 10% would be more appropriate, with a well-telegraphed ramp down to 5-6%. Nothing says that the rate has to remain set in stone, but the target path should be known well into the future (next 10 years at least) to set expectations.

[John Dougan]

How long do you think it would take for NGDPLT to become subject to Goodhart's Law (in any of it's related forms)? Is there something about NGDPLT that would prevent that fate? Could this be what happened to inflation targeting? (It used to work then gradually lost its information content as people adjusted to the perceived behaviors)

[Andrew F]

I don't see how Goodhart's Law even applies. What would that look like? Inflation expectations becoming a self-fulfilling prophesy so that it just happens to be the difference between real GDP and the NGDPLPT?

[scepticus]

"I don't see how Goodhart's Law even applies. "

Noah Smith at noahpinion disagrees:

"This seems to me to be an application of the second version of Goodhart's Law. If you set interest rates mechanically (i.e. according to some rule) based on current market expectations of inflation, you will change those expectations, and markets will move, requiring you to set interest rates differently according to your policy rule. It's possible that markets and policy might converge to some stable equilibrium, but also possible that market volatility might increase once it was known that policy was being set based on market prices themselves.

To link this to Goodhart's Law, if the Fed targeted the prices of inflation-linked assets, those prices would mostly contain information about expectations of Fed policy decisions (which the Fed already knows better than the asset market), rather than about non-Fed economic forces that might affect inflation or people's utility of stable prices (i.e. the things the Fed wants to use the asset prices to ascertain).

So Goodhart's Law, in its obviously true form, seems very important for policymaking. "

http://noahpinionblog.blogspot.co.uk/2013/04/two-versions-of-goodharts-law.html

[W. Peden]

Frank Restly,

I didn't say anything about trends.

[Declan]

Andrew F,

Where did I mention inflation? I said "absolute nominal wage cuts, or even big cuts in nominal wage growth."

In the mid 2000s Australia had pretty stable 4% wage growth and NGDP growth that peaked at 10%. To keep to even a 6% NGDP target and maintain the same real outcomes would therefore have required 0% wage growth. Implausible.

[Nick Rowe]

John Dougan: I think it is maybe possible that Goodhart's Law destroyed inflation targeting. Inflation targeting made inflation so sticky it could no longer signal monetary disequilibria. Could something similar happen to NGDP? I can't think why it should, but I can't rule it out either. But nihilism doesn't seem like a viable policy option.

[Andrew F]

Declan, I don't think that follows. I could be wrong. Why do you hold that wages would have to absorb all the difference in NGDP? Why not changes in investment?

[Declan]

Andrew,

I said "maintain the same real outcomes". To get NGDP growth down from 10% to 6% would require 4 % points slower growth in the NGDP deflator, therefore to keep real wages the same would require 4 % points slower wage growth.

Wages are not absorbing all the difference in NGDP, in the sense that nominal profits are also cut in the same proportion.

[Declan]

John and Nick,

Isn't downwards nominal wage rigidity enough to explain the current absence of deflation without bringing in Goodhart's Law? I struggle to see Goodhart's Law working at the level of individual price setters, but maybe that's just me.

[mickeyman]

Sometimes our guard dog would bark at plastic bags blowing down the street . . .

[apt]

I could think of Inflation targetting as a "smooth" modern-age illustration of the European Exchange Rate Mechanism (obviously in the way by which it can be broken). And I would say not that I can't see why it should, but that I can't see why it could not.