Scott Sumner  

Who will save China's rust belt?

PRINT
Reflections on my Latest RT Ap... Speech Tomorrow at San Jose St...

Now that Trump has decided to save America's rust belt, I thought it would be interesting to look at the issues facing China's rust belt. Here is the Financial Times:

North-eastern China is facing a demographic crisis as educated millennials abandon the industrial heartland, the country's worst-performing region.

Planning officials revealed this month that the economy of Liaoning, one of the three northeastern provinces, had shrunk 2.2 per cent in the first nine months of the year -- the largest regional contraction in China in seven years. , , ,

For younger workers, the slowdown is made worse by the region's extreme reliance on the state. Most new jobs in China are created in private companies but north-eastern China is home to the state-backed heavy industrial companies and state-owned farms that form the Communist party's traditional support base. In some cities, new jobs in government or state-owned enterprises only open when an older worker leaves, leading to a practice whereby parents or other family members will retire to create a slot for a younger relative.

In the 1990s, China's three north-eastern provinces saw net immigration of 360,000 people, but from 2000 to 2010, 2m left. . . .

China's 2010 census showed that the fertility rate of the north-east had dropped to only 0.75, too low to replace an ageing labour pool. More recent mid-cycle census data from 2015 has not yet been released, but is likely to show a further decline.


If that 0.75 fertility rate is accurate, it would be by far the world's lowest rate, for any large region.

Before fixing the problem, let's think about possible causes:

1. Trade: Many people claim America's rust belt has been devastated by imports. But China is the world's largest exporter of goods. It's also the world's largest exporter of steel. So trade does not seem to be the culprit.

2. Environmental regulations: But China has relative weak environmental controls, so that doesn't seem to be the problem either.

3. Neoliberalism: This region is the most state dominated in China, so neoliberalism doesn't seem to be the problem.

4. Declining output: China's steel output has soared in recent decades, rising to roughly 50% of global output:

Screen Shot 2016-12-04 at 9.35.35 AM.png
My conclusion? It seems like the "problem" is automation. China's manufacturing productivity is soaring, and that means that even producing 50% of the world's steel is no longer enough to keep a rust belt prosperous. In China, it is the free market, high tech cities like Shenzhen and Hangzhou that are booming.

How does this apply to the US? First, our steel output is down about 20% since 1970, even as consumption edged up slightly:

Screen Shot 2016-12-04 at 9.53.22 AM.png
So is that why steel employment has fallen? No, it's mostly productivity. Steel employment is not down 20%; it's down closer to 80%:

Screen Shot 2016-12-04 at 9.55.29 AM.png
Are there any lessons here for America? I think there are. People trying to save the rust belt are essentially Luddites. The only way to save those old industrial jobs is though restrictions on productivity growth. Sorry, but there is no other way.

You might think this is all common knowledge among economists. It's not. Every day I read economists talk about how trade has devastated America's rust belt. Nope, it's the productivity, stupid.




COMMENTS (14 to date)
EB writes:

There is no lesson for America or for any free society. I don't know how free Chinese people are to move within their own country today, but 25 years ago their freedom to move around was quite limited. In addition, the reform of state enterprises has been advancing but we still know little about how they have been managing their old employees and their children (yes, the enterprises were supposed to take care of them). If you know about any serious research on those two topics, please let me know.

DB writes:

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

DB writes:

@DB

Have factories consolidated?

Given that US industrial output has increased 10-20% since 1996 (although flat-lined since 2000) they must have.

[This comment is not written by DB to DB. The comment author seems to have mistyped his nickname.--Econlib Ed.]

Kerry Kaneko writes:

Excellent piece.

Technology - robots, automation, etc. - is replacing workers. Steel manufacturing is much more efficient, requiring less workers to do the same work: http://finance.yahoo.com/news/trumps-campaign-ignored-robots-and-automation-202156153.html; http://www.csmonitor.com/Business/2016/1102/American-jobs-are-going-to-robots-not-China; https://techcrunch.com/2016/10/09/industrial-robots-will-replace-manufacturing-jobs-and-thats-a-good-thing/; and so on and so on and son.

BC writes:

When I read the sentence about millennials abandoning northeastern China, I actually had the opposite reaction to @EB. Much has been made of the mass migration in China from rural areas to the cities. At the same time, geographic mobility seems to have declined in the US, explaining part of our stagnation. How is it that state-controlled China is benefiting from mobility in a way that relatively free America is not? It's also strange that mobility in the US is declining even as communications and (air) transportation has become more widely accessible, making it easier to be separated from relatives. Maybe, Americans don't *need* to move as much because more people can be physically remote from their workplaces?

Thaomas writes:

Curious, everything I read about the decline of the rust belt points mainly to automation although there are estimates of job losses due to China opening in specific counties.

Harry Chernoff writes:

Scott:

Doesn't this mean that once employment in high productivity sectors like manufacturing declines towards the same low levels as manufacturing's high productivity predecessor - agriculture - we're going to be left with mostly workers in low productivity service sectors like government, education, retail, food service, and a lot of medical (e.g., home health care aides, long-term care nurses)?

Doesn't that also mean the combination of an overwhelmingly low productivity employment base (Baumol's disease writ large) and the current demographic headwinds is low RGDP growth for a very long time? And this is even before we get to Bob Gordon's view of long-term technological change, productivity, and sustainable RGDP growth rates.

Is this about how you see it?

E. Harding writes:

"The only way to save those old industrial jobs is though restrictions on productivity growth. Sorry, but there is no other way."

-Wrong. There is another way: promote manufacturing production to unprecedented heights. Had manufacturing output per capita in the U.S. kept pace with its 1946-1973 growth rate, while (indeed, especially if) manufacturing continued to decline as a share of GDP, nobody would be complaining about stagnant wages and unemployed manufacturing workers.

Scott Sumner writes:

Harry, Service jobs are not necessarily low productivity. But yes, I believe that we face a lower rate of trend growth going forward--slightly above 1% per year.

Growth has been 2% during the recovery, but I expect it to slow when the recovery is complete.

Harding, Who will buy all those manufactured goods that consumers don't want? If you want to save blue collar jobs, I'd reduce zoning so that more houses can be built.

Andrew_FL writes:
For younger workers, the slowdown is made worse by the region's extreme reliance on the state. Most new jobs in China are created in private companies but north-eastern China is home to the state-backed heavy industrial companies and state-owned farms that form the Communist party's traditional support base. In some cities, new jobs in government or state-owned enterprises only open when an older worker leaves, leading to a practice whereby parents or other family members will retire to create a slot for a younger relative.

This feels like it should've merited a "5" on your list of possible causes.

Fred Anderson writes:

I am not trained as an economist, so my comments may appear foolish to those better informed. However, I see my profit in getting educated by those more knowledgeable. That said;
Greater productivity, say in steel, means more steel for less work. If more steel doesn't lead to cheaper steel, then the steel makers (labor & capital) realize a sizable payoff for their greater productivity.
But in a competitive market, more steel likely will lead to cheaper steel. And at least some of the gains from increased productivity will be captured by the consumers of that steel. Since the consumers are the least wedded to the fortunes of the steel makers, they will have the upper hand in negotiations with them. And probably most of the gains from increased productivity will be captured by the consumers.
This would seem to be especially troubling if that steel goes largely to the export market. Here, Americans capture large savings on appliances made with that cheaper steel, while the Chinese steel makers get little of the payoff; perhaps slightly easier jobs, and a greater assurance of meeting their costs of capital.
I wonder if a Marxist mind set -- obsessed with the contest between labor & capital -- may not be especially prone to simply missing what's happening to (here, by) the consumers.

E. Harding writes:

"Who will buy all those manufactured goods that consumers don't want?"

-It's a supply-side issue. World oil consumption per capita has stagnated since the 1970s, and that in the U.S. has declined. Consumers will always want more manufacturing goods and oil. It's just that technological and other constraints came into effect during the 1970s.

"If you want to save blue collar jobs, I'd reduce zoning so that more houses can be built."

-People in the Zoned Zone vote Democrat, and the GOP platform supports local zoning as a measure against relocation of minorities into swing states, so I see no route to this happening through either party. Though it would be a good idea, and help reduce inequality. If I were to pick between zero difference in housing costs in all parts of America vs. manufacturing output per capita having continued its 1946-1973 trend in the U.S. (due to technological progress or otherwise) as a means of reducing blue-collar woes, I would definitely pick the latter.

Brian Donohue writes:

Excellent post. You are a credit to your profession.

Jose writes:

One could argue that there is no permanent reason why steel factory productivity should be higher in China than in the US, a priori.

Certainly one could argue that steel factories shifted to China for a reason (cheap labor?), but that is a condition that may not apply anymore.

It may be possible that factory productivity in the US could increase to the point where investors would like to put them back in Pennsylvania, although producing 1/5 less jobs. But if this is the case, incomes would go up, and spill over effects would happen.

Comments for this entry have been closed
Return to top