Thursday, March 25, 2010

Uncertain Coloration of Chinese Flags

Yesterday, we discussed a graph from Edward Chancellor's "Chinese Red Flags" report which showed an accelerating curve with more and more of the economy going in fixed investment, which seemed like clear evidence of a near-term unsustainability.

This morning, I spent a couple of hours trying to reproduce that graph, and the situation is, well, complicated.  The graph above shows my efforts.  The blue line is in the background to the graph, and is a direct cut and paste of Chancellor's "Illustration 2".  My first approach to compare this with official statistics was to look at the annual GDP breakdown, which includes figures for total capital formation, which ought to be similar to or a superset of fixed asset investment.  That, as a fraction of GDP, is the red line in the above.  As you can see, the pattern is pretty different, and much less alarming.

I emailed Andrew Hunt (credited as the source of the data), and he responded that the graph was based on monthly investment in fixed asset data from official Chinese stats as well as official GDP numbers (see also this comment from KLR). (Hunt also indicated that he didn't think either set of numbers was necessarily very reliable).

So I took the monthly data, picked the "Investment in Fixed Assets" and looked at the accumulated data through December, and divided that into the accumulated four quarter GDP for each year (eg here for 2009).  From the website, I was only able to get back to 2004, and that is the purple curve shown in the graph above.  It does not match the Chancellor graph, but does at least reproduce the sharp lurch upwards in 2007-2009.  However, it's also roughly consistent with the red curve in the region in which the two overlap, and thus is consistent with a story that the Chinese capital formation rate has been fairly stable over time, or only very slowly growing, and then lurched sharply upwards in a classic Keynesian approach to managing the great recession (as opposed to the speculative bubble narrative).

So I'm back to not knowing.  I cannot reproduce Chancellor's graph right now.


KLR said...

Listening at the moment to the FT 2006 Book of the Year: China Shakes the World: A Titan's Rise and Troubled Future -- and the Challenge for America. Great stuff, here's a pertinent excerpt:

In Germany,
for instance, state expenditures came to 47.8 percent of GDP in
2004. An accurate figure for China is more difficult to calculate because
of various categories of quasi-governmental financing, such
as funding by state banks for state infrastructure projects, which
is not normally included in the official figures. Nevertheless, narrowly
defined Chinese state expenditures, as a percentage of gross
domestic product, come to less than half of German levels.18

Perhaps that will be of guidance in nailing down this figure. Also: Chancellor wrote a terrific book on speculation, Devil Take the Hindmost.

Manolo said...

Some interesting stuff by Dian L. Chu about China, worth the read:

A tale of three Swan Songs: