Chinese data generate more heat than light -- again

BEIJING (Reuters) - To adapt Winston Churchill’s description of Russia, Chinese statistics are a riddle wrapped in a mystery inside an enigma.

A worker walks past a large container used for steel production at the Shougang steel plant in Beijing January 22, 2009 which was closed down during last year's Olympic Games in an effort to reduce pollution. REUTERS/David Gray

And that is on the good days. Thursday was not one of them.

Keenly awaited growth figures for the fourth quarter of 2008 were as clear as mud to most analysts, who were left struggling to work out what is really happening in the world’s third-largest economy.

“The good, the bad, and the downright confusing” was how Stephen Green, head of China research for Standard Chartered Bank in Shanghai, titled his report on the data. The figures showed annual gross domestic product growth slowing to 6.8 percent from 9.0 percent in the July-September period.

Experience has taught economists to take Chinese statistics with a pinch, if not a packet, of salt.

The quality of the data may have improved in recent years, but it is conventional wisdom among China-watchers that the figures are prone to manipulation by a government intent on promoting stability and, hence, on minimizing data volatility.

In particular, they suspect the National Bureau of Statistics strives to avoid giving the impression of sudden lurches in activity: the resilience of Chinese growth after the 1997/98 Asian financial crisis is a prime case in point, skeptics say.

“This number has clearly been massaged (the NBS typically smoothes major turning points in the cycle) and the headline GDP figure reveals very little resemblance to the underlying nominal growth trends,” Glenn Maguire with Societe Generale in Hong Kong said in a report.

Imports fell faster than exports in the fourth quarter, while the drivers of domestic demand, including retail sales and fixed investment, are characterized more by resilience than cataclysmic weakness, Maguire said.

“Though things are clearly going to worsen from here, from a nominal bottom-up approach, growth does not appear to be as weak as these headline numbers suggest. So why report it so?”

His answer is that officials are juggling the data to set up a relatively firm statistical bounce in the second half of 2009.


Other economists, by contrast, were puzzled that the fourth quarter was so strong given weakness in factory output and power generation. The latter fell 7.9 percent in December from a year earlier, marking the third straight monthly drop.

Comments by one government economist fueled suspicions that all was not what it should be. Speaking before the GDP report, he said statisticians had concluded actual fourth-quarter growth was below 6 percent but would report it as being above 6 percent.

The economist, who declined to be identified, correctly predicted full-year GDP growth of 9.0 percent -- a figure, incidentally, that had been circulating in China’s financial markets and among officials for more than a week.

Dong Tao with Credit Suisse in Hong Kong said he did not rule out the possibility of some data massaging. But he pointed out that agriculture and services, which consume much less power than the industrial sector, have much steadier growth profiles.

Still, Ben Simpfendorfer with Royal Bank of Scotland in Hong Kong said GDP growth of 6.8 percent sat awkwardly with official figures for industrial production, which rose just 6.4 percent in the fourth quarter, a record low since the series began in 1994.

Moreover, RBS’s proprietary activity index, a proxy for GDP, is pointing to the lowest rate of growth in over a decade.

“The upshot is that there is a divergence between the quarterly and monthly data. GDP growth has slowed to its 2001 pace, but the monthly data indicate a more pronounced deceleration,” Simpfendorfer said.


The problem is that economists have to make do with piecemeal data. China’s quarterly GDP report is a supply-based figure. It includes only a real year-on-year growth rate and a nominal yuan level. Unlike in most developed economies, there are no real, expenditure-based figures.

“This means there are no quarter-on-quarter, seasonally adjusted, changes in key GDP inputs, in particular, exports, residential investment, and household consumption, with which to assess the result,” Simpfendorfer said.

The headache gets worse when the NBS revises annual GDP growth -- as it did last week for 2007, to 13.0 percent from 11.9 percent -- but does not update its quarterly growth profile.

“With the revisions in there, it’s hard to work out what 6.8 percent growth is,” said Green at Standard Chartered.

And what about inflation?

The NBS reported a revised 2007 GDP of 25.731 trillion yuan. Given Thursday’s estimate for 2008 of 30.067 trillion yuan, that points to nominal GDP growth of 16.9 percent last year and a GDP deflator of 7.9 percent.

Yet that would be a full 2 percentage points higher than the rise in the consumer price index and 1 percentage point more than the rise in the producer price index.

For all the doubts, economists agree that Thursday’s figures accurately portray the main underlying trend: China has slowed sharply, but it has not collapsed. And only time will tell how close the fourth quarter was to the bottom of the cycle.

“We don’t have a strong view either way on whether these numbers have been deliberately massaged,” Green said.

Editing by Kazunori Takada