It’s the 1970s, without the inflation
During the fourth quarter of 1980 and the first quarter of 1981, America’s NGDP soared at an annual rate of more than 19%. And yet unemployment hardly budged, slipping from 7.5% in September 1980 to 7.4% in March 1981. What went wrong?
I’d like to suggest that the past decade is basically the 1970s without the inflation, at least in one important respect. Before explaining why, let’s first consider the differences. The period from 1971 to 1981 saw roughly 11% NGDP growth, which included about 3% real growth and 8% inflation. That’s a lot more real growth than today. In addition, labor force participation was rising sharply, whereas in recent decades it has been falling. On the unemployment front, however, we are doing better today.
Now let’s imagine a policy counterfactual where, beginning in 1965, the Fed keeps inflation close to 2%. Here are my claims:
1. Under that monetary policy, we would have had roughly 5% NGDP growth from 1971 to 1981.
2. Real growth would have still been about 3%, with inflation about 2%.
3. The unemployment rate in late 1980 and early 1981 would have still been about 7.5%.
All of these predictions are based on the assumption that money is roughly neutral in the long run, that the Natural Rate Hypothesis is roughly (not exactly) true when NGDP growth avoids deep declines. And most importantly:
4. In that counterfactual, I believe that economists would have misdiagnosed the unemployment problem, assuming it reflected a sort of secular stagnation, a lack of aggregate demand.
We now know that what actually happened is that the natural rate of unemployment rose by a couple points between the late 1960s and the early 1980s. We don’t know exactly why (theories includes more women and minorities and young people in the labor force, oil shocks, deindustrialization, environmentalism, more generous benefits for not working, etc.), but clearly the natural rate was increasing. If 19% NGDP growth isn’t enough to make a dent, it’s fair to say that the high unemployment is not due to a “demand shortfall”.
Today, America is not faced with a higher natural rate of unemployment—it’s probably back around 4.5% to 5%. Instead, we are faced with a different structural problem—a decline in the labor force participation rate (LFPR). Among men, this rate has been trending lower for at least 5o years. We don’t know all the causes, but it’s clearly not just the Great Recession. And once again, people are assuming that a structural problem is due to a lack of demand.
I believe they will once again be disappointed. Our problems are much deeper than a demand shortfall. That’s not to say that monetary policy should not be more expansionary. Perhaps the economy still has a bit of slack, and the inflation rate has recently undershot its 2% target. I’m making a broader claim. The broad outlines of the 2016 economy are not going to change substantially with monetary stimulus. We might get unemployment down to 4.5%, and the LFPR might rise a bit more. And those would be very good things. But we are still going to be stuck with a trend rate of RGDP growth of barely over 1%, and a LFPR rate (for men) that is quite disappointing by historical standards. And that’s not even factoring in the lousy supply-side policies of the next president.
In the 1970s, we discovered the structural nature of the problem more quickly than today, because we had such high NGDP growth. Our current sub 2% inflation rate has indeed represented a modest demand shortfall, and that is a small part of the problem. But it doesn’t really come close to explaining the extent to which growth has slowed. Inflation also averaged about 1.5% during the first half of the 1960s, but growth was fine. “Lowflation” is a problem because it means the Fed is not hitting its target. But it explains at most only a small fraction of the very low RGDP growth rate since 2006, as well as the current very low LFPR.
PS. I believe that the Great Inflation is best described in terms of NGDP growth, not price inflation. That makes it easier to see that it was a 100% demand-side problem; RGDP growth was fine. Today I was pleased to see the Financial Times also using that metric:
According to Bank of America Merrill Lynch, the current global recovery is the least inflationary of all time, with nominal GDP (growth plus inflation) growing just 11 per cent in the last seven years.
That’s total! (Not the annual average.)
That was from an article discussing the new Chinese growth data. At the same time, there are a few “green shoots” for the global nominal economy. Chinese NGDP growth is up to 7.8%, from a low of (I think) about 5.8% at the end of last year. Chinese PPI inflation turned positive for the first time since early 2012, and CPI inflation is up to 1.9%. The overvaluation of the yuan seems to be ending. Eurozone inflation also has been rising, albeit from a very low level. I’m most pessimistic about Japan; they need to sharply depreciate the yen if they are serious about their 2% inflation target. Will they?
Chinese growth is expected to slow to 6% in 2017, as the housing boom cools. That’s still an excellent figure for a middle-income country with near-zero population growth. Consumption will lead the way next year, as housing construction slows. But China is by no means out of the woods. They still face a very tricky transition from a housing construction economy to a services economy. Lots could still go wrong–just not yet.
Tags:
19. October 2016 at 06:32
Check that title. It the 1970s?
19. October 2016 at 07:09
> Instead, we are faced with a different structural problem—a decline in the labor force participation rate (LFPR). Among men, this rate has been trending lower for at least 5o years. We don’t know all the causes, but it’s clearly not just the Great Recession. And once again, people are assuming that a structural problem is due to a lack of demand.
I don’t necessarily agree with this depiction of events. Looking at the age-based breakdowns, the drop in LFPR is most pronounced for both men and women in the 15-24 age group, rather than in the prime-age 25-54 group. See https://fred.stlouisfed.org/graph/?g=7Ogu.
This could collectively reflect a greater emphasis on education, whereby men in particular could formerly find reliable work without formal postsecondary credentials.
However, I don’t for a second think that this is unrelated to the business cycle. The 15-24 LFPR rates for both men and women drop notably in the 1991, 2001, and 2008 recessions, not regaining their earlier levels.
There may be a hysteresis effect at play. When youth unemployment increases as part of a recession, businesses make a long-term shift in procedures away from less-skilled (uneducated) labour and towards higher-skilled (educated) labour, decreasing both youth hiring and provision of on-the-job training.
Tellingly, we can see the same business cycle effects in youth labour participation in the United Kingdom (https://fred.stlouisfed.org/graph/?g=7Oh8) from the 2008 recession, but that country evidently does not have the same sort of long-term decline in prime-age LFPR.
19. October 2016 at 07:45
One thing we seem to be learning is that OECD GDP is increasingly different from Gross Domestic Utility, what people actually care about. This was less of a problem in the past because technological changes moved slower, leisure had less utility, and labor was responsible for a larger proportion of output. So when you look at the official inflation numbers, you get crazy results like minimum-wage living standards being much higher in the 1960s than today. That’s one reason why some people like Lars and I believe OECD CBs are leaving a lot of real growth on the table right now by preferentially missing their inflation targets low.
Official Chinese NGDP numbers are whatever the Communist Party decided they would be several years ago, so they’re not very interesting except as a gauge of what the Communist Party leadership wants everyone to believe about China’s economy. Other measures suggest growth has slowed to around half the official number, but it’s very hard to tell, we can’t really even rule out growth higher than official numbers. Watch the reserves and the capital controls — the extent to which China is propping up the official numbers with debt eventually has to show up in a floating exchange rate, if so much yuan is being created.
19. October 2016 at 08:55
Both the 1970s and today have very similar productivity growth, especially in manufacturing.
Had there been no Great Inflation, I suspect unemployment would have been higher by a couple points. The natural rate of unemployment today is somewhere just under 4 percent, higher than 3.5 percent.
19. October 2016 at 12:45
For decades now, liberals have been telling us that we need to be more like Europe, with bigger governments and more regulation. What they left out, of course, is that per capita income was considerably higher in the U.S. than in Europe. But now that we’ve enacted much of the liberal agenda, economic actors here face mostly the same incentives Europeans have been responding to all these years. Why wouldn’t you expect per capita incomes to converge? And since European policies haven’t changed much, why wouldn’t you expect that the way that happens is U.S. incomes stagnate while the Europeans catch up?
19. October 2016 at 14:22
I wonder if the high minimum wage levels of the period had something to do with rising natural unemployment rates. Here’s one of a series of posts I did a while back comparing the number of workers at given wage levels today with the number of workers who had worked at those wage levels when minimum wages were higher.
http://idiosyncraticwhisk.blogspot.com/2014/05/revisiting-minimum-wage-workers-as.html
19. October 2016 at 15:33
How do you parse out inflation vs real growth in NGDP?
19. October 2016 at 21:24
Aging demographic is the reason people drop out of work. That and, as TallDave implies, they are voluntarily unemployed and prefer leisure time, like in Europe (France is a good example, but all of Europe is like that, in Greece the entire month of August is vacation time for employees). TallDave is right about China’s fake numbers: look at how ‘smooth’ the China GDP growth is post 2009. Very unnatural looking.
Sumner says: “Our problems are much deeper than a demand shortfall” – yet his NGDPLT addressed only demand shortfalls. It’s a tacit admission that NGDPLT fails. I’ll take that as a small first step victory.
20. October 2016 at 10:36
I wonder if the inflation of the 1970s didn’t serve some purpose in containing unemployment. According to data from Social Security, the ‘national average wage’ fell by 12% between 1972 and 1982 in real terms, way worse than anything seen in the recent recession. I don’t recall a similar level of bellyaching though. Maybe we were made of sterner stuff, but maybe money illusion played its part too.
20. October 2016 at 12:29
Randomize, Thanks I fixed it.
Majromax, You said:
“When youth unemployment increases as part of a recession, businesses make a long-term shift in procedures away from less-skilled (uneducated) labour and towards higher-skilled (educated) labour, decreasing both youth hiring and provision of on-the-job training.”
I’d expect them to do the exact opposite.
Talldave, Are you sticking with the China/Mexico bet?
Harding, You said:
“Had there been no Great Inflation, I suspect unemployment would have been higher by a couple points.”
In 1980-81? I seriously doubt it? Why would unemployment have been 9.5% under a scenario of steady 5% growth in NGDP? How would wages have behaved?
Jeff, They still face much stronger disincentives to work in Europe.
Kevin, That might be part of it.
Pras, I used government data.
Ray, SMH
Brian, That could also occur under NGDP targeting, especially during periods when imported oil prices rose sharply. Imported oil is not a part of NGDP.
20. October 2016 at 13:34
“How would wages have behaved?”
-They would have grown at a slower rate than in real life, but not to the same extent NGDP growth would have been slower.
“Why would unemployment have been 9.5% under a scenario of steady 5% growth in NGDP?”
-My counterfactual wasn’t, to be clear, “steady 5% growth in NGDP”, but nominal GDP growth with all the NGDP growth ups and downs of the Great inflation, lowered by some constant number to average 5% over the decade. Under that scenario, yes, 9.5% in 1980-81 would have been plausible (we got over 8% in 1975). NGDP growth experienced a minor dip in 1980 which led the unemployment rate to shoot up.
21. October 2016 at 02:39
[…] The 1970’s Sans Inflation by Scott Sumner via The Money Illusion […]
21. October 2016 at 05:42
[…] The 1970’s Sans Inflation by Scott Sumner via The Money Illusion […]
21. October 2016 at 10:07
Scott — Well, I’ve been derelict in acquiring the wager’s physical avatar, but I certainly still hold with the principle. 🙂
Hopefully we live to celebrate my inevitable victory (haha).
I think you are more likely to win the bet than when we made it, but paradoxically I also think you are much more likely to win if China never adopts a floating currency and representative government, even though those are better policies, because in that case the Wikipedia entry for PPP GDP per capita will probably reflect the inscrutable official numbers.
https://en.wikipedia.org/wiki/List_of_countries_by_GDP_(PPP)_per_capita
Iraq’s position is interesting, it has suddenly become more productive than China despite the ongoing ISIS-led insurrections in Mosul and Anbar.
So far, China is still poorer the world average, so they have some ground to make up by 2036. We’ll see if their institutions can adapt well enough to get out of the middle income trap.
21. October 2016 at 12:47
TallDave, China will probably pass Brazil this year. Brazil is only 15% poorer than Mexico. You’ll lose way before 2036, maybe 2020.
21. October 2016 at 13:19
[…] The 1970’s Sans Inflation by Scott Sumner via The Money Illusion […]
21. October 2016 at 16:15
Scott — that would require China to grow at least 10% faster than Brazil this year. That seems unlikely, even using official numbers.
Do you see any sign China will hold truly open national, multiparty elections any time soon? How rich do you think China can get under its current political institutions?
22. October 2016 at 19:12
TallDave, Why is it so unlikely? Brazil will shrink by 3% or 4% this year.
23. October 2016 at 05:18
Hmm, well I don’t follow Brazil that closely, but the average of estimates seems to be around 1% growth. The average estimate for China seems to be around 5%.
At any rate, I certainly hope China avoids massive economic dislocation and/or violent political upheavals. I would be somewhat happier losing the bet in any case, since there are many more people in China than Mexico. I’m a bit surprised Mexico hasn’t done better, but petroleum has hurt them in the short run, while legalization in the US helps a bit in the medium term. At least it’s unlikely they’ll have to deal with Trump!
23. October 2016 at 05:34
In the latest Economist magazine, the consensus forecast for Brazil is minus 3.3% and for China it’s plus 6.6%.
24. October 2016 at 01:18
Remember, though, we’re talking in PPP terms.
It’s certainly a safe prediction that China will report something like 6.6% growth…
BEIJING — The investigators descend on government agencies and corporate boardrooms. They interrogate powerful officials and frequently rebuke them for lacking zeal. Most of all, they demand unflinching loyalty to President Xi Jinping and the Communist Party.
http://www.nytimes.com/2016/10/23/world/asia/china-discipline-commission-political-loyalty.html?_r=0
As alluded above, it will be very difficult to know what PPP GDP per capita in China really looks like until China has liberalized politically. Hopefully that will happen soon, but things seem to be going in the other direction at the moment.