Scott Sumner  

More evidence for a Great Stagnation

PRINT
Christina Hoff Sommers at Geor... Timothy Taylor on Net Debtor N...

The government recently estimated that real GDP rose at a 0.2% rate in the first quarter of 2015. Most forecasters overestimated the growth rate, while the Atlanta Fed was pretty close, with a 0.1% forecast. The media pointed to temporary factors like bad weather in Boston and a brief port walkout in LA. But those clearly were not the main issue, as they would not be expected to affect the level of GDP in the second quarter. And if RGDP bounced back to trend in the second quarter, then wouldn't we expect to see forecasts of roughly 5% growth for Q2?

The latest forecast from Atlanta's "GDPNow" program calls for 0.9% growth in the second quarter. Of course a lot could change, but so far there is no evidence that slow growth in 2015 merely reflects disturbances that occurred over a few week in February.

Here's something that is unlikely to occur, but worth contemplating nonetheless. GDP figures are almost always revised, and thus it would not be all that surprising if the final estimate of first quarter growth ends up slightly negative (of course it could also be even more positive.) Let's say there's a 1/3 chance it ends up negative. And suppose the second quarter also comes in slightly negative. What then?

In a macro sense it doesn't much matter whether the numbers are slightly negative or slightly positive. But in a political sense it's dynamite. Most people regard two consecutive quarters of falling GDP as a "recession." I've repeatedly argued that that view is nonsense, pointing to the phony "recession" in Japan during 2014---a year when it had robust job growth and a falling unemployment rate. (Well, robust job growth for a country where the working age population is plunging at 1.4%/year.) But that's just me; all the media insisted that Japan had a "recession."

Just to be clear:

1. I do not expect 2 consecutive quarters of falling RGDP. I'd guess the odds are 5% to 10% of that happening.

2. And even if it did, I would not regard that as a recession. I'd expect the unemployment rate to stay flat.

But as we saw with Japan, what I think doesn't matter. The media would scream "recession," and the Fed would be frightened from raising interest rates for many more years. There would be a sea change in how almost everyone thinks about macroeconomics in the US.

Even though I do not expect this to happen, I think it's quite plausible that we'll have two consecutive quarters of almost no growth. And that's almost identical to two slightly negative numbers. It tells me that the evidence for a Great Stagnation is getting stronger every day. In previous posts I've speculated that trend RGDP growth in the US may have fallen to around 1.2%, and about 3% for NGDP. Most economists think trend growth is considerably higher, and they may well be correct. But make no mistake, the early data for the US economy in 2015 is quite bad, and even though the last snow pile in front of my house didn't melt away until mid-April, the economic effects vanished several months ago. There are no more excuses.

PS. None of this should be viewed as absolving the Fed. NGDP grew only 0.1% in Q1, that's less (in per capita terms) than even George Selgin would recommend. But it's likely to pick up. My point is that growth is surprisingly slow for a period of falling unemployment. The juxtaposition of falling unemployment and slow RGDP growth cannot be explained in a demand-side model.

PPS. You might wonder if the 3.3% 2015 NGDP growth forecast at Hypermind contradicts my 3.0% trend forecast. Not so, as most people expect this to be an expansion year with some decline in the unemployment rate. Trend NGDP growth covers both expansion and recession years, and is therefore lower than the typical NGDP growth rate during an expansion year.


Comments and Sharing


CATEGORIES: Macroeconomics




COMMENTS (16 to date)
dlr writes:

Scott, given that you are so naturally suspicious of our general ability to separate the real from the nominal in NGDP growth, why aren't you more open to the idea that apparent declines in real GDP growth rates might reflect downward bias in our efforts to measure marginal growth amid technological change as opposed to the great stagnation you seem to favor?

Kevin Erdmann writes:

I'll second dlr. I think this can even explain the dot com "bubble". The market didn't so much overestimate the value of these new innovations. It's just that the value ended up mostly going to consumer surplus instead of producer surplus. This doesn't get picked up by our standard measures of economic activity, so we have the ironic development that the intelligentsia are worrying about the ever-increasing power of capitalists and economic stagnation in the midst of a revolution in consumer surplus.

It's a revolution so ubiquitous that the debate itself depends on it (such as this blog and comments themselves).

Scott Sumner writes:

dlr and Kevin, Good question, I'm actually more interested in NGDP growth, which is also undergoing a sharp decline in trend, and which has implications for interest rates, monetary policy, etc.

Even if RGDP growth is mis-measured (and I have no opinion either way, as it can't be measured objectively) the government estimates do matter in all sorts of ways. They influence the solvency of Social Security. They influence the frequency with which the press will describe recessions. That's why I pointed to Japan. Just as they were the first with the zero bound, they were the first with the phony "recessions." But they will happen here too, it's just a matter of time.

Tyler said he was a revenue pessimist and a utility optimist. I'm also a revenue pessimist. Utility? I don't even have a clue as to how happy I am, how could I have any opinion on society?

Roger writes:

I third the above two comments. We are seeing a transition to an economy which is no longer measured well with traditional methods. Just about everything which is improving is getting damn near free.

I've repeatedly commented that things which would have cost me hundreds of thousands or millions of dollars a few decades ago are now free or near to it. A massive increase in utility for tens of millions of people shows up as a reduction in GDP.

18th and 19th century economists missed the great breakthrough which was occurring under their feet, and I suspect the same is true now for 21st century economists. Look at what people do today and what it doesn't cost.

Kevin Erdmann writes:

Of course you're right,Scott. We are probably all in agreement. The thing that keeps your solution from being accepted by the consensus is the deep-seated human desire to second-guess market-based information. People can't see targeting NGDP because sometimes they just know that prices are wrong, and somebody needs to be in charge of changing them.

Marc writes:

Prof. Sumner, aren't you an utilitarian? How does that exactly work without a way to measure utility even for yourself? Only slightly tongue in cheek here.

Scott Sumner writes:

Roger, You may be right, but I don't see any obvious way to determine whether you are right. Again, I'd focus more on things we do know:

1. NGDP growth

2. Government data on RGDP growth.

Leave actual RGDP growth to the philosophers.

Kevin, I agree.

Marc, Good question. All I can say is that we do the best we can in estimating the comparative utility occurring under alternative government policies, but without much confidence that we are getting it right.

Harun writes:

That port strike was not brief, and its disrupted supply chains seriously.

I had products delayed by 2-3 months due to that port strike, because even when it was over there was back-log.

Lorenzo from Oz writes:

1960, 1968, 1976, 2000, 2008. Years in which the incumbent Party lost the Presidency after two terms.

1988. Years in which the incumbent Party kept the Presidency after two terms.

Secular stagnation reduces the chances that 2016 will go against the established tendency.

Scott Sumner writes:

Harun, Thank's for that info. So when would you expect a return to the trend line? Q2? Q3?

Lorenzo, Yes, but is this statistically significant? Suppose I amended the dates:

1920, 1960, 1968, 1976, 2000, 2008

1904, 1928, 1940, 1988

Now it looks less significant. Not saying you are wrong, I just wonder if there is enough data there . . .

dire link writes:

Secular stagnation = "Baby Boomer Disease".

Nathan Smith writes:

Considering that the US dollar strengthened dramatically (by 15-20%) in the last quarter of 2014 and the first quarter of 2015, a sharp slowdown in US growth doesn't seem very troubling, nor like much evidence for a Great Stagnation. The fact that GDP held steady in the face of such a sharp loss of competitiveness seems like a sign of strength.

Andjuar Cedeno writes:

[Comment removed for supplying false email address. Email the webmaster@econlib.org to request restoring your comment privileges. We'd be happy to publish your comment. A valid email address is nevertheless required to post comments on EconLog and EconTalk.--Econlib Ed.]

Lorenzo from Oz writes:

My selection is since the Presidency became limited to two terms. Which makes a difference to the dynamics.

Jose Romeu Robazzi writes:

Prof. Sumner
Given what you posted here, would you change your recommendation of NGDPLT with 4.5% target ? Would you rather resort to some per capita indicator instead ?

Rick writes:

"My point is that growth is surprisingly slow for a period of falling unemployment."

Perhaps some peeling back of the unemployment layers is in order. For instance, the BLS' CPS survey directly asks people who are not in the labor force if they would want a job now. Of the 91,630,000 Americans not in the labor force, an all-time record, 6,538,000 answered yes to this question...a huge number that has increased 7.7% from a year ago. Typically this figure drops when the unemployment rate drops, but instead it's increasing.

Also, if labor productivity growth continues to decline, could the faux recession you describe become real? Unemployment might remain stable, but only because we're less productive.

Comments for this entry have been closed
Return to top