March 07,2010

Lies, Damned Lies, and Chinese Statistics

By Huang Yiping, Beijing

The sententious statement, "There are three kinds of lies: lies, damned lies, and statistics," which Mark Twain attributed to former British Prime Minister Benjamin Disraeli, would seem to be suitable to describe the gap between China's current crop of statistical data and its real economy. The difference can't be put down to simple statistical error or local phenomenon. It is a systemic problem.


In the current economic statistics system, problems arise in three areas: (1) Critical data are not accurate, e.g. how much does GDP grow? (2) Important data are not collected, e.g. the National Bureau of Statistics (NBS) does not report month on month data, which accurately reflect the recent economic trends in the market and are very important for investment and government decision-making; (3) Data that are reported are often useless, e.g. the reported urban unemployment rate does not reflect the real situation in the labor market.

The process of estimating China's GDP bears on social stability and even the political future of local leaders, and is inevitably subject to political influence.


Both domestic and foreign investment experts point out that the credibility of China's GDP data is not high. The NBS reported GDP grew 7.8% in 1998, surprisingly close to the government target of 8%. Economists and investment professionals making their own judgments believed that China's economic growth in 1998 was below 2%. What's worse, provincial GDP growth rates are almost every year reported as higher than the national average. It is absurd and inconceivable that a government known for its sweeping measures and administrative efficiency can't provide accurate GDP data. Moreover, statistical departments also publish misleading data, such as urban fixed-asset investment and total retail sales, but don't collect data on investment and consumption.


The weight of current housing costs in GDP, over 30% in the US, is reported as slightly more than 10%, far lower than most people's actual burden. The distortion of the housing price index is the most serious. An obvious fact is that fewer and fewer citizens now feel they can afford to buy new housing, but NBS data show that the growth rate of housing prices over the past decade has been below that of per capita income. In 2009, housing prices in Beijing and Shanghai rose ferociously, but this has not been reflected in the government's price index.


The reported urban unemployment rate, slightly over 4%, has not changed over the years, but it is painfully apparent that this does not accurately reflect labor market volatility. The data is estimated based entirely on registered unemployed workers, which is meaningless with the privatization of many state-owned enterprises and 150 million migrant workers in cities. It is in fact a waste of taxpayers' money and certainly misleading for economic policy-making.


Another significant deficiency of China's economic data is the lack of month-on-month data. The international economic circle attaches great importance to month-on-month data because they more accurately reflect the latest economic trends.

How to develop sensitive macro-control if economic sectors don't know what is happening?

(The author is a professor at Peking University's Center for China Economic Research)

Click to Get New TextCan't read this text? Please click the image!
Please verify the text in the image.