The world shows keen interest in China?s gross domestic product (GDP) when it is published at the end of each quarter or year. How are these data produced and just how accurate is the figure? Xu Xianchun, director of the Department of National Accounts under the National Bureau of Statistics of China responds.
Question: How was China?s GDP system set up?
Answer: In the period from the founding of the People?s Republic of China (1949) to the end of 1970s, a national economic index was used. This followed the same practice as the Soviet Union and the East European countries.
Following reform and opening-up, the service industries made rapid progress. To better serve the needs of managing the macro economy, the National Bureau of Statistics submitted a Report on Introducing Statistics for Tertiary Industry to the State Council in March 1985. This report, which was approved by the State Council, emphasised the merits of the GDP measure. From April of that year onwards, the general office of the State Council attached great importance to both national income and GDP statistics.
In the same year, the various responsible departments set up systems to collect data and generate GDP statistics at both national and provincial levels.
Question: How did the status of GDP then evolve in national economic accounting practice?
Answer: After China?s GDP made its debut, the country still depended on national income as its main index with GDP cast in a supporting role.
As reform and opening-up rolled out, the role of GDP became increasingly important in the work of those departments with macro economic analysis and management functions. Consequently, the National Bureau of Statistics elevated GDP from a minor index to a main index. By 1993 the national income index had been replaced by GDP as the core index of the national economy.
Question: How are the GDP figures generated?
Answer: China?s GDP is created in three steps. First comes tentative evaluation, then tentative verification followed by final verification.
Tentative evaluation is carried out early in the year following the year to be reported. Only limited data will have been gathered by that time. The available data is assessed and processed to generate the tentative figure.
Tentative verification is carried out at the second quarter of the year following the year to be reported. By this time most of the necessary information will be to hand. Compared with the earlier tentative evaluation stage, tentative verification is more reliable but further verification is still necessary.
Final verification is usually carried out at the fourth quarter of the year following the year to be reported. By then, all the data is to hand and the final and most accurate statistics can be reported.
Question: Will the published GDP remain unchanged after all three steps have been completed?
Answer: Annual GDP data could be adjusted after these three steps have been completed. This could be as the result of significantly different new data emerging or changes to the data categories.
Question: What kind of measures has the National Bureau of Statistics used to improve the accuracy of the GDP data?
Answer: We have worked to improve our means of obtaining the information.
The National Bureau of Statistics prefers to use sampling techniques rather than the full reportage which would tend to impose heavy administrative burdens with no guarantee of accuracy.
For example we have used sampling techniques to determine levels of production of grain and cotton. They are also employed for the small-scale industries and have an important role to play in ensuring the accuracy of the GDP data.
Question: Have the methods of collecting and verifying the accuracy of the statistical data changed much?
Answer: Traditional methods can be vulnerable to the effects of the human factor. In recent years more and more data have been forwarded directly to the National Bureau of Statistics for collation. This helps to reduce human intervention as far as possible.
In addition, the National Bureau of Statistics has been working to improve the scope of its information gathering activities. For example in 1999 it extended sampling to include small-scale industries.
(china.org.cn, translated by Li Liangdu, August 31, 2002）
