This is another one of a long line of papers purporting to show that state-owned enterprises are more inefficient than other forms and therefore the Chinese financial system is vastly misallocating capital.
Like other such papers, it has a number of methodological flaws:
1) It doesn’t correct for the differences between state-owned enterprises and non-state owned firms. The big one is that money losing SOE’s are responsible for a great deal of social welfare spending. If you remove those social obligations from SOE’s, it might make the firm more efficient, but someone else needs to pick up the tab.
2) The other problem is that China has a huge number of tiny SOE’s which don’t make that much money in contrast to a small number of huge SOE’s which do. The result of this is that if you do a factors analysis, you pick up the huge number of tiny SOE’s which are losing money, but I suspect that if you weight things by industry size, you’ll find that the bulk of bank loans are no longer going to these small SOE’s.
3) Finally, something that is consistent in these studies is that “mixed” or “collective” enterprises do as well as “private” ones. This makes no sense if you believe that the cause of inefficiency is state ownership, but it makes perfect sense if you assume that there is some “cherry picking” going on. Basically, any SOE which isn’t hopeless is going to find investors, and not be an SOE.
One thing that I’d like to do at some point is to dig into the NBS data, and write a paper on this. The trouble is that I’m already booked to write three papers, and part of the reason I’ve limited myself to three papers this year is so that I don’t try to write ten.
This has some relationship to my affinity to Austrian economics. Austrians look at numbers with a lot of skepticism and focus on the idea that we don’t really know that much about economics. This sort of skepticism about quantitative data, I find refreshing among economics. In the literature trying to analyze quantitative performance of Chinese firms, I’ve rarely found that the author goes in and questions the numbers and the meaning of the numbers.