from Follow the Money

Have China’s manufacturing powers been exaggerated?

April 15, 2006

Blog Post

More on:


In a word, no.

Many in the economics blogosphere have mentioned Guy de Jonquieres recent FT column, Which argues that the US - not China - is the world's manufacturing powerhouse.  And usually in a favorable light.   Let me chime in with a dissenting voice. 

De Jonquieres notes that China only manufactures ½ as much as the US.    But China's economy is only around one sixth the size of the US economy.   The world's largest economy should have the world's largest manufacturing sector if the term "advanced industrial economy" means anything at all.   

I am all for pushing against over-generalizations that get repeated so often that they become conventional wisdom.  The oft-stated argument that France isn't growing is one example.  In fact, France has grown faster than either Germany or Italy over the past few years, and France grew for the same reason the US grew: soaring real estate prices have pumped up domestic demand.  

But I would submit that the real story here is the growth in China's manucturing sector, not the fact that the US - the world's largest economy - continues to have the world's biggest manufacturing sector.  Or the fact that the US manufacturing did a bit better than Japanese manufacturing from 1995 on.  Japan, afterall, was in a prolonged slump; Japanese demand wasn't growing at all. 

I don't have the data in front of me, but China's manufacturing sector was probably only ¼ the size of US manufacturing only a few years ago.   Consider the trade data, which I do have in front of me.  In 2001, China ‘s average monthly exports of manufactured goods were around $22b.  Judging from the March data, China's 2006 manufacturing goods exports are going to average about $80b a month.   That strikes me as a big change.

For the record, the US  exports more manufactured goods than China.  It just won't be for long if China's current export growth continues.   By the end of this year, China will surpass the US.   

I also wish the argument that China's exports have little value-added was subject to a bit more critical scrutiny.    De Jonquieres echos the convertional wisdom:

"Indeed, to call China a manufacturing economy is something of a misnomer. In reality, it is the world's biggest final assembly shop, with minimal local value-added."

However, lots of the data that backs up this argument seems to come from 2002 or 2003.  This CEPII study is an example.   But a lot has changed since 2002 or even 2003.   In 2002, China's monthly exports averaged around $27b; in 2006, its monthly exports will average around $80b.  Even if Chinese "value-added" is constant, exporting more implies that the total value-added in China's export sector has grown substantially. 

Moreover, anecdotal evidence suggests a rise in Chinese value-added --particulary in 2005, when Chinese import grow slowed significantly.  The enormous amount of investment taking place in China means that the composition of China's output can change quickly.  More electronic components are made in China now, for example.   China is gearing up to produce high-quality steel for autos, not just the low-end stuff.   And so on.  

But the detailed data needed to do careful analysis of all this isn't out yet.  So everyone ends up working off what I suspect is "dated" data.  One big caveat: I didn't receive an advance copy of  the IIE's study that de Jonquieres cites, so I don't know what data it is drawing on. 

But logic tells us that the value-added in China's export sector is large enough to generate a trade surplus of around 5% of GDP (continuation of q1 export and import trends implies China's trade surplus will rise to $140b), pay for China's energy imports (another 5% of GDP, according to the IMF) and to pay for China's imports of the commodities and capital goods it needs for its domestic economic expansion.   

I suspect that the value-added in China's manufactured exports is probably now over 15% of China's GDP.    By comparison, total US exports of goods are around 7% of US GDP, and some of those exports have imported content.

I concede that I am working off inferences, not hard data.   But China's trade surplus - assuming it is real, not disguised capital inflows, as Stephen Green of Standard Chartered argues - didn't come out of nowhere.    China has been able to sell enough "assembly" to pay for a very rapidly rising bill for commodity imports, have enough cash left over to pay for a few Boeings, GE generators and the like and still have some left over to help China build up its reserves.   The US, by contrast, doesn't pay cash for much of anything anymore. 

Finally, much of the de Jonquieres column seems to be based on an Oxford Economics study.    That study was financed by the China Businesss Forum, itself a part of the US-China Business Council.    They hardly sound like a disinterested party.  I was a bit surprised - to be honest - that de Jonquieres didn't mention that the China Business Forum sponsored this study in his column.

I am not technically adept enough to comment on the methodology Oxford Economics used to assess the impact of China on US jobs, or the Mark Weisbrot critique Oxford Economics' study. Weisbrot argues that its focus on the overall benefits of the US-China trade ignores important issues about the distribution of those benefits.   I am inclined to agree: It doesn't take a lot of econometrics to figure out that the impact of China -- and others -  on the US manufacturing sector is one reason why home mortgage delinquencies are rising strongly in Indiana, Ohio and Michigan.

But I did think Oxford Economics was spinning that basic data fairly hard.   And when I feel like I am being spun on things I know well, I start to get suspicious about things I don't know as well

Let me mention a few reasons why I felt spun.

  • Oxford Economic's estimate for the 2005 Chinese current account surplus was on the low side-- $110b.    China's 2005 trade surplus alone was $100b - and in recent years the current account surplus has significant exceeded the trade surplus.   $130b strikes me as a more reasonable estimate. 
  • Oxford Economics opted to compare that surplus in dollar terms with that of Japan ($154b in 2005, according to the recent data) and Germany.  They didn't present the same comparison as a percent of GDP.   Such a comparison would show China in a much less favorable light.   Moreover, Germany is part of a broader currency union that ran a small current account deficit in 2005.  French and Spanish deficits offset Germany's surplus.    Looking at Germany alone is misleading.  
  • I understand presenting the data in dollar terms. I don't understand not presenting the same data as a percent of GDP.  Particularly when Oxford presented China's steel production in per capita terms ...
  • Oxford Economics noted that China's current surplus is small relative to the oil-exporters.  That is very fair.  But shouldn't the fact that China's surplus has risen in the face of an oil shock also be mentioned?  
  • Some data was presented in ways that rather obviously - if you know the data - skew the results.  For example, they presented a chart showing China's share of the US bilateral trade deficit.  It was 20% in 1997, and it is around 20% now.    What does that leave out?   The fact that overall deficit increased enormously over that time period, so the US bilateral deficit with China rose substantially as a share of US GDP.  Why not show the bilateral deficit with China - even the bilateral deficit with China adjusted for Hong Kong - as a percent of US GDP?
  • Oxford Economics presented a graph showing very rapid growth in US imports from China and relatively slow growth in US imports from the rest of Asia, but didn't include the (easily calculatable) growth in overall US imports from Asia.  I suspect there is a reason for that: the overall growth in US imports from Asia was rather impressive in 2004 and 2005, and showing that data would undermine the argument that Chinese is just doing the final assembly that used to be done elsewhere in Asia.   

China is, of course, doing much of the assembly that used to be done elsewhere in Asia.  But that isn't all that it is doing either.  There is a rather simple test.   Are US imports from Asia going up or going down as a percent of US GDP?  

The answer:  Overall US imports from the Pacific Rim went down sharply in 2001.  The tech bust  meant fewer electronics imports.  Add in a recession and 9.11.  But US imports from the Pacific Rim have been growing strongly since 2003.    Oxford Economics could have done the same graph I did last week.   Or they could have shown the evolution in the United States' overall trade balance with East Asia, as Saleh Nsouli of the IMF did.

They just didn't.    I doubt the resulting graphs would have pleased the China Business Forum.   Call me suspicious.

More on: