Since I posted "Chinese Government Releases Treasure Trove of Statistics on 农民工 (Nongmingong)," many people have approached me with the same question - "should we believe what the Chinese government is telling us about unemployment amongst the domestic migrant labor force?"
The answer is yes and no.
Yes - From the research I have conducted thus far as well as the literature that is coming out on the floating population, these numbers seem quite plausible. While the statistics released by the government don't indicate that the situation is mind-blowingly bad, they do paint a pretty solemn picture. At this point, we have no reason to believe that the Bureau of Statistics is misleading us in any major way.
No - As mentioned in previous postings, generating an accurate number for the size of the floating population, let alone its sub-components, is next to impossible. Anyone who works with, knows, and/or researches China's migrant workers is aware of the fact that many of them, especially those who are younger, don't officially exist. Their parents, for a variety of different reasons (more often than not associated with the One Child Policy), never registered their birth with the government. Though this is most common among young children, I have run into some teenagers and young-adults who still haven't applied for a hukou (residency permit). With all these "invisible" men and women floating around (I have read that there could be as many as 80 million), it is impossible to get a perfectly accurate look at the scale of current migrant labor unemployment.
Ultimately, I think we should accept the numbers released by the Bureau of Statistics at face value, but do so recognizing that they are far from perfect. Anyone who has studied economics knows that though GDP and the unemployment rate feature many flaws, they are still useful as measures that can be used to assess a situation and react to it with sound public policy. I recommend viewing the recently released statistics on migrant unemployment in China in the same light.