Once again, the official survey diverges from HSBC/Markit’s which I trust a lot more. The official PMI edged down to 50.4 in January from 50.6 in December. (Societe Generale reckons that if the official figure is seasonally-adjusted, it would come in at 50.8 rather than 50.4 — so an increase of o.2.) Here’s the HSBC/Markit report:
After adjusting for seasonal factors, the HSBC Purchasing Managers’ Index™ (PMI™) posted 52.3 in January, up from 51.5 in December, signalling a modest improvement of operating conditions in the Chinese manufacturing sector for the third successive month.
Output in the Chinese manufacturing sector also increased for the third month in a row during January. The rate of expansion accelerated from December to a marked pace. New orders rose for the fourth consecutive month, and at a solid pace that was the sharpest in two years. New export orders also increased, following a reduction in the previous month. Survey respondents suggested strengthened demand from clients in Europe and the US.
Backlogs of work increased during January, after a five-month period of reduction. Over 10% of panellists reported an increased level of work-in-hand, which was generally associated with new order growth. Employment levels rose, albeit marginally, for the second month in a row. (…)
However, inflation remained modest overall, with less than 12% of survey respondents reporting increased output charges.
Purchasing activity continued to increase in January. The growth rate of input buying accelerated from the previous month to a solid pace, with exactly 26% of survey respondents reporting an increase. Panellists attributed the rise in purchasing to new order growth. (…)
FT Alphaville deals with the diverging reports (What to make of the contradictory China manufacturing PMIs). Here’s the gist of it:
The official methodology changed a lot this month — primarily it has moved from a survey of 820 to now covering 3,000 companies. Furthermore, the China Federation of Logistics and Purchasing and the National Bureau of Statistics have merged some ‘unrepresentative’ sectors in the survey, reducing the total to 21 from 31. It’s not clear how this might affect the PMIs, which are a survey of month-on-month observations. (…)
Markit addressed the question of divergence in the two surveys last year, and insisted their survey methodology is superior, and showed some fairly clear evidence of the seasonality problem in the official surveys. As for which one is ‘best’, take your pick. Our impression from talking to pundits was that the official one was traditionally favoured because their sample was bigger, but sometime last year the HSBC/Markit PMIs become more popular, though as Liu’s comments above show, that’s obviously not a universal view, either.
As far as trends common to both surveys this month, there’s roughly three:
- Inventories are definitely down.
- Input prices for both surveys rose sharply, which Nomura says confirms their house view that inflation is going to be a key issue in the coming months.
- New export orders were lacklustre in both, rising only slightly in the HSBC survey and falling from 50 to 48.5 in the official PMI survey.