China Research

The quality of Chinese GDP

Soon it’s time again for China to publish its outcome for Chinese GDP growth in 2014 – only a few weeks after the end of the past calendar year. A couple of years ago, this important statistical event happened even before year end. Of course, one may wonder about the speed until publication of this very complex and comprehensive statistical indicator with all its aggregates. In most developed countries we are talking about 6-12 weeks’ delay – and the gigantic country of China does it in (probably) less than four weeks!

This conundrum may have various reasons – or combinations of it:

-Statistical methods for really technically and representatively covering the whole country of China with its entire production/demand are still underdeveloped (which would be logical considering the current – on average – lagging stage of the institutional development in China);

– collecting statistics from all distant parts of the country would be too time-demanding despite all possible efforts; marginal benefits of more detailed data would probably be too low;

– rapid statistical input is needed for preparing the traditional National Congress in the beginning of March;

– there exists some benign or even malign statistical quality neglect/misleading (look, for example, at the usually very small quarterly variations/changes – a quite stable growth pattern that is not common in our part of the world!).

Historical experience or rumours can be added in this context with certain – but academically insufficient – evidence/experience: in times of an overheated Chinese economy GDP growth may be even higher than officially shown; in the opposite case, (foreign) China experts often believe that the officially published, more dampened GDP-growth numbers are still overestimating reality. Is the latter happening these days?

Unsophisticated reactions by financial markets, their analysts and journalists      

Financial analysts are – wrongly – too frequently focusing on very minor changes compared to the previous numbers and/or deviations from expectations. Statistical moves of GDP growth from 7.4 to 7.3 or 7.2 percent may mean nothing – apart from what has been said above(i.e. that the downturn actually may have been larger than national accounts were telling). By the way, a similar conclusion can be drawn when the frequently applied Chinese Purchasing Manager Index (PMI) changes from 50.4 to 50.6 or something like that.

Also many journalists tend to make too rapid conclusions by following (too) rapid interpretations of financial market analysis. Most financial analysts – particularly in our part of the world – are not really familiar with the Chinese conundrum of statistics and the insufficient transparency culture. My feeling is (almost) that the safer analysts feel about many statistical numbers from China, the less they really know about the second largest economy in the world.

Summary:
The probably most relevant conclusions from my reflections above are

– that Chinese GDP numbers should be treated with necessary caution, and

– that quantitative long-term forecasts on China should be treated carefully since they are based on uncertain statistics and methods.

I still have my doubts when foreign economists praise the progress of Chinese economics statistics (even if there may be some slight qualitative improvement going on, as our annual China Survey Panel from April 2014 showed). Statistical progress would be good for China as well.

This article is about the technical part of GDP calculation and its shortcomings. More exciting in the future, however, will be to what extent Chinese decision-makers will be able to improve the qualitative composition of GDP, i.e. to achieve economic growth that favors the Chinese people’s well-being and future. I will come to this important issue pretty soon.

 

Hubert Fromlet
Senior Professor of International Economics, Linnaeus University
Editorial board

 

Back to Start Page