China’s numbers are wrong!

Well, that is no surprise, since they have ALWAYS been wrong to one extent or another!

The most cited example in aluminium is the production number. NBS does not count “illegal” smelters such as the Shandong Weiqiao plants. These smelters never went through the proper approval process, so they are not included in the production data. The NBS production number therefore undershoots the real number. The other example is the NBS monthly production number for the first few months of each year. This has been well documented by Paul Adkins of AZ China. Paul would tell you that the Q1 numbers do not make any sense on a tonnes-per-day basis.

Other numbers do not take into account discontinued series or they simply do not add up!

It is sure embarrassing for the NBS. But, everyone knows the lack of data accuracy inside China, even the policy makers themselves. New Premier Li Ke Qiang even went so far as to call China’s official numbers as being “for reference only.”

If one turns up at a macro economic conference in China, he or she would find that the standard ‘two men walk into a bar’ joke has been replaced with NBS data manipulation jokes. Policy makers tend to give their own private estimates on the data during meetings and conferences. After talking to employees who have worked at the NBS or other Chinese industry associations, one would find out that ‘number making’ and ‘poor methodologies’ are ongoing practices.

Does it matter? Yes and No. For accuracy’s sake, the numbers do not paint an accurate picture of reality; they are therefore misleading on many occasions. When numbers get so grossly out of whack, they can result in misallocation of capital. Arguably, the inability for foreign investors to read China clearly has been a result in part of these inaccurate data. However, the numbers provide more of a guidance than an accurate pin on any one asset, socio-economic phenomena or productive activity. The system can work with numbers that are inaccurate provided they are not completely off-track.

The market tends to adjust the released numbers to deal with poor quality data; the methods of adjustments vary from one user to the next. However, we would argue that poor data are part of the risk to the Chinese economy, because without good data, there is likely to be poor decision making by policy makers and investors. The lack of data on shadow banking is a good example of how poor data undermines risk management and market well-being.

Other areas of questionable data include:

• Electric power - Overall power demand is up 4.3% YTD, yet, thermal power utilisation went down 7.2%. Thermal power utilisation is highly measurable, since it requires coal. According to the NBS, the difference between demand and supply was made up by hydropower, which made a miraculous 16% improvement in the same period. Just as well it is harder to measure waterflow.
• Steel - Steel production has been reported at record levels, though demand has barely changed. According to Anne Stevenson-Yang of J Capital Research, this is because plants which do NOT report to the Industry Association magically increase their production levels at the same time that those who do report to CISA are closing blast furnaces.
• PMI - At least 3 times in the last 12 months, the China Logistics Association has changed its sampling and methodology for measuring PMI, but has never disclosed how they changed, nor bothered to produced any comparison data.
• Retail Sales - This is a major sector for China’s economic hopes, and has been reporting growth of 12% YoY. Yet the number of sub-sectors being dropped from the measure continues to grow. Since December 2012, food, grain, oil, meat, tobacco, liquor, books and other items have all been dropped from the measure. Coal metals and fertiliser have not been in the Retail Sales figures since 2010.

How is it that the NBS can be so wrong? Perhaps the best way to understand how official data works in China is what Anne Stevenson-Yang says. “The modern NBS is like a priesthood, providing moral guidance to the work of data collection. Is news at the grassroots is not harmonious, best that it should simply go unreported.”