Manufacturing in 2050 (Ib)

choi

Raymond Zhong, A Blocked Path to Development; 'It's more of a "Made in the world" kind of thing rather than "Made in India." '  WSJ, Nov 25, 2015 (front page).
http://www.wsj.com/articles/for- ... ns-rocky-1448374298


Quote:
(a) "The US and Europe—and East Asia more recently—first got rich because of their factories. Over time, as incomes rose and their economies became more sophisticated, they shifted into modern services like health care and finance.

"But today, parts of South Asia, Africa and Latin America are failing to create thriving manufacturing sectors even though their wages remain low. Manufacturing employment and output are peaking and declining at vastly lower levels of income and development than they did in the West.

"Economists’ worry is that the factory-led model of advancement—which, for more than a century, has offered the quickest route out of poverty—is simply no longer available to today’s poorest nations.

(b) "In the early 1960s * * * manufacturing output in India was around 12% of the economy. The share peaked at 19% in the 1990s, and has since fallen to around 17%.  

"South Korea’s manufacturing sector, by contrast, grew to 36% of the economy by 2010 from 3% in the early 1960s. The path was similar in China, where manufacturing also accounts for around a third of the economy.

"Africa looks more like India. In South Africa, manufacturing was 15% of output in 1962 and peaked at 25%—in 1981. By 2011, the share was closer to 18%. Factory activity in fast-modernizing Ethiopia hasn’t managed to grow beyond 6% of the economy. In Tanzania, it peaked at 13% in 1976 and dropped since to around 10%.

(c) The latecomers to industrialization suffer from poor infrastructure and governance. "But experts including Mr Rodrik say there are also deeper forces at work.

"Factory automation and robotics are reducing the need for unskilled workers from the countryside to staff assembly lines.

"Industrial latecomers now have to compete against China, whose massive, integrated manufacturing machine has made it the world’s factory floor and created a huge barrier to entry.*  Tariffs are falling and trade is becoming freer, making it tougher for developing countries to shelter their producers from foreign competition.

* This is how this article describes what India faces in order to expand its manufacturing: "One headwind is China itself. Bargain-price Chinese goods, produced at titanic scale, mean that even with India’s factory labor costs at around $5 an hour versus almost three times that in China, manufacturers have to work harder to compete than they would have a decade ago."

(d) "When manufacturing peaked as a source of jobs in the US in 1953, it employed 26% of American workers, and overall per capita income was around $17,700 in today’s dollars. By 2010, manufacturing accounted for around 9% of US jobs.

"Factory employment topped out later in the UK, France, Italy and Japan, but at around the same income level or higher.

"In places that started the process later, manufacturing yielded less. By the time manufacturing in South Korea accounted for its highest proportion of jobs, in 1988, incomes there were around $12,700, after adjusting for differences in purchasing power. Brazil peaked in 1986, at $8,700.

"In India, factory employment started to decline as a share of employment when income was around $3,300. And for Nigeria, Kenya and Ghana, the figure was closer to $2,000.

"In many of these countries, manufacturing’s share of economic output continued rising even after the employment share peaked. As plants become more productive, they make more with fewer workers.

(e) "India’s annual trade deficit with China has ballooned to $48 billion, or 2.5% of India’s gross domestic product. * * * A decade ago, the trade gap was less than $1.5 billion.

choi

Note:
(a) Take notice quotation (b) is all about "output" or "economy," which are another way to say "GDP."

(b) "Harvard economist Dani Rodrik, who began compiling data on manufacturing world-wide a few years ago, says he is seeing growing evidence of what he calls “premature deindustrialization”—the idling or shrinking of manufacturing sectors as a share of the economy in poor countries like India that never industrialized very much in the first place."

This WSJ article is the retelling, in ways an ordinary person may understand (by converting “1990 international dollars” to 2015 US dollar, for example), of a paper :
Dani Rodrik, Premature Deindustrialization. Institute of Advanced Study, Princeton University, January 2015 (draft; Economics Working Paper No 107).
https://www.sss.ias.edu/files/papers/econpaper107.pdf
(i) Read Abstract (page not numbered), which stated, "The hump‐shaped relationship between industrialization (measured by employment or output shares) and incomes has shifted
downwards and moved closer to the origin."

There is a peak, so it must be hump-shaped. If you must see it, go to
page 31
for
Figure 1 Simulated manufacturing shares as a function of income (in GDP per capita in 1990 international dollars).
(ii) For an acronym "manemp," go to page 6: "For completeness I will use all three measures in this paper, denoting them as manemp (manufacturing employment share) * * *"  (Ignore two other acronyms -- nommva and realmva -- because the Wall Street Journal article does not mention the last two.)
(iii) The essence of the paper is at page 15: section “VI. Premature deindustrialization”:

"Let us denote these peak levels by manemp* and realmva*. [That is what the asterisk means: 'peak.' Here, peak manemp]  There is evidence that suggests these peak levels are reached at lower levels of income as well. Denote that level of income by y*. [Again, y is income (per capita GDP measured in PPP) and y* is peak income.] Our baseline regressions capture the downward shift in the manufacturing hump over time, but not the possibility that the curve may be moving closer to the origin as well. Figure 5 [at page 37]"
(iv)
(A) The only graphic in print (but not online) is titled “Failure to Launch.”  Underneath the graphic is: "Source: Groningen Growth and Development Centre, University of Groningen."
(B) Rodrik’s paper is more detailed, where page 29 has “Appendix A: Country and variable coverage in the GGDC 10-sector Database."

Here it is:
GGDC 10-Sector Database.
www.rug.nl/research/ggdc/data/10-sector-database
(Click the first blue item under the section heading “Data” will open an Excel file. The bottom horizontal line displays “notes” and we should click “dataset” to its right)

In the new “dataset” page, one item (column D) at the top horizontal bar is “Variable.”  Each country (which is grouped by continents) has three variables (in the order of “VA,” “VA_Q05” and “EMP’” each variable spans decades).
(C) My own estimation (because I did not calculate for every year), judging from (B), is Taiwan’s manufacturing employment -- Country: TWN; Variable: EMP; “Manufacturing” (column H) divided by Summation (column P) -- peaked at 1987 with manemp* at 34.85% (compared with S Korea’s in 1989 at 29.16%).


(c) Dani Rodrik
https://en.wikipedia.org/wiki/Dani_Rodrik
(1957- ; Turkish; Descended from a family of Sephardic Jews)
(d) “Mr [prime minister Narendra] Modi governed the state [Gujarat] for more than a decade before he became prime minister. * * * Gujarat * ** is the only Indian state where manufacturing’s share of the economy has even approached East Asian levels, according to an analysis by Mr. Subramanian and economist Amrit Amirapu. Factories contributed 22.7% to Gujarat’s output in 2011, compared with a third in some East Asian countries.  Yet here [Gujarat], too, job growth isn’t keeping up with the explosion of young people training for factory work. Manufacturing’s share of employment peaked in 1984 at 5.4%.”

From 2001 t0 2014, Modi was chief minister of the state of Gujarat.
https://en.wikipedia.org/wiki/Gujarat