For most of last year, China’s official PMI was above the Markit Chinese PMI report. Now the situation is reversed with Markit saying conditions are better than China’s official report. What’s going on?

China PMI 2016-08-03

Please consider China PMIs: a Tale of Two Surveys.

China’s purchasing manager indices, which take the pulse of factory activity by surveying companies, are a puzzle. There is the official government measure and a private one, produced by Caixin-Markit, a media and research group partnership — and each paints a very different picture.

In the first half of 2014, for example, the official chart stayed above 50 while the Caixin PMI showed manufacturing slumping. That was the year China’s growth hit a 24-year low of 7.4 per cent, missing the government’s target of 7.5 per cent, and property prices started to fall.

The March to June quarter this year brought a recurrence of that: the official measures showed manufacturing growing — but only just — while Caixin-Markit showed manufacturing falling. But In July’s PMI the two measures swapped sides: Caixin now showed the highest manufacturing growth since February 2015, while the government showed manufacturing starting to slump.

So what gives?

The divergence is explained in part by the difference in the companies being surveyed. Markit, which calculates the Caixin-Markit index, send out their questionnaires to more than 500 mainly small, private businesses. The government has a much bigger sample size of 3,000 companies, by far the bulk of which are state-owned enterprises.

Regional differences also matter. Liu Liu, an economist at investment bank China International Capital Corporation, suggests July’s uptick in the Caixin PMI is not down to differences in business size — the official PMI for SMEs also slumped last month — but to do with the fact that many of the companies surveyed by Caixin are based on the booming east coast.

In short, both measures are correct in that they reflect their constituents’ activity — but it is the private survey that gives the truer sentiment of factory floors that are not protected by the state.

The survey methodologies are so different neither represents a true picture.

Ignoring large companies totally while measuring a small sample of small businesses in a region that is booming when the rest of the country that isn’t, is like measuring drops of water in a glass that is full while ignoring three other glasses that are bigger, and empty.

Neither of these reports accurately represents what’s happening in China.

Mike “Mish” Shedlock