China, Statistics and the Environment

An interesting article in the NTY this week on a new report on water pollution in China. I find this all very interesting as the Party’s quest for growth at all costs seems to be imposing some staggering long term costs on the country. The article went from interesting to frustrating pretty quickly in a manner typical of anything involving Chinese statistics.

From the article:

“China’s government on Tuesday unveiled its most detailed survey ever of the pollution plaguing the country, revealing that water pollution in 2007 was more than twice as severe as was shown in official figures that had long omitted agricultural waste.”

DOUBLE?! So your statistics failed to include AT LEAST HALF of the appropriate sample?! I can understand the benefit, at least domestically of mitigating the horror of the environmental destruction that break neck industrialization is causing to the country, but this leads me to other more disturbing questions. How much omission or deception is there in statistics which matter on an international level such as the banking sector’s non performing loans? What about projects which are qualifying for Kyoto’s Clean Development Mechanism, or heaven forbid a true accounting of national emissions? The list could go on and on. Take your pick of any statistical reporting the government does. I wonder if I had simply failed to include half the sample in my stats classes if that would have been ok.

But it gets worse! Officials are attempting to explain how the problem isn’t really as bad as it seems:

“The census keepers had also used updated methodologies and reached many more parts of the countryside and industrial sites than had official statistics, which helped account for the much larger figure in the census, Mr. Zhang said. Were it not for the vastly expanded scope of the survey, the chemical oxygen demand level in 2007 would stand at only 5.3 percent higher than previously calculated, he said.”

So basically he’s saying that things aren’t so bad because if you just use a sample which is not at all a depiction of reality, then things don’t look as bad. Wow. Why even have statistical reporting anyway? Let’s just take samples which make everyone happy and then assume that there are no problems! What should be a call for massive government action is just being brushed off as a difference in reporting.

Things like this only add to my skepticism of anything the Chinese government publishes.