Home / Wiki / C / Caixin China PMI

Caixin China PMI

Economics Markets and Financial TheoryUpdated 16 Nov 2026

The Caixin China PMI is a monthly indicator of business conditions in China, based on a survey of purchasing managers at several hundred companies. It is compiled by S&P Global, the data firm that produces purchasing managers' indexes for many countries, and for about a decade it carried the name of its sponsor, the Chinese financial news group Caixin Media. Separate indexes cover manufacturing and services, and a composite combines the two.

The survey is watched because it is independent of the Chinese government and appears within days of the end of each month. It is read alongside the official purchasing managers' index published by the National Bureau of Statistics. Because China is the world's largest manufacturer and the biggest buyer of many raw materials, the readings affect markets for commodity exports, Asian shares and currencies well beyond China itself.

01

How the index is built

Each month, managers at the surveyed firms are asked whether conditions have improved, worsened or stayed the same compared with the previous month, for items such as output, new orders, export orders, employment, prices and inventories. Each answer is turned into a diffusion index:

Index=Pup+0.5×Psame\text{Index} = P_{\text{up}} + 0.5 \times P_{\text{same}}

where PupP_{\text{up}} is the percentage of firms reporting an increase and PsameP_{\text{same}} the percentage reporting no change. If 30 percent report higher output, 50 percent no change and 20 percent lower output, the index is 30+0.5×50=5530 + 0.5 \times 50 = 55. A reading of 50 means no change on the month. Above 50 indicates expansion and below 50 indicates contraction, and the distance from 50 shows how widespread the change is. The figures are adjusted for seasonal effects, which matter greatly in China because of the shifting date of the Lunar New Year holiday.

The headline manufacturing index is a weighted average of five of these components:

PMI=0.30 N+0.25 O+0.20 E+0.15 S+0.10 I\text{PMI} = 0.30\,N + 0.25\,O + 0.20\,E + 0.15\,S + 0.10\,I
  • NN is new orders.
  • OO is output.
  • EE is employment.
  • SS is suppliers' delivery times, inverted so that slower deliveries, a sign of busy supply chains, raise the index.
  • II is stocks of purchased materials.

With component readings of 52, 51, 49, 50 and 48, the result is 15.6+12.75+9.8+7.5+4.8=50.4515.6 + 12.75 + 9.8 + 7.5 + 4.8 = 50.45, a marginal expansion. The services index is based on a single question about business activity.

02

Comparison with the official PMI

The official survey is larger, covering some thousands of enterprises, and gives more weight to big companies, many of them state-owned and concentrated in heavy industry. The private survey has a smaller sample with a greater share of small and medium-sized firms, many in the private sector and many dependent on exports from the coastal provinces. The two can therefore diverge. A gap between them is often read as a sign that conditions for smaller private firms differ from those of state-backed industry, for instance when credit is flowing mainly to large borrowers or when foreign demand is moving differently from domestic construction.

The official manufacturing figure is released on the last day of the month, and the private one usually on the first business day of the following month, with services a few days later. Analysts treat the pair as complementary and look at both before drawing conclusions.

03

History and sponsorship

The survey dates from the mid-2000s. It was sponsored for some years by the bank HSBC and published as the HSBC China PMI, at a time when the compiler was the British firm Markit. Caixin took over the sponsorship in 2015. Markit merged with IHS in 2016, and the combined company became part of S&P Global in 2022, which is why the index has appeared under several compound names. In 2025 the Caixin sponsorship ended and the series continued under a new sponsor's name with the same compiler and method, so recent releases may be labelled differently from the historical data.

04

How markets use it

The release is compared with the consensus estimate of economists. Because official Chinese statistics on gross domestic product appear only quarterly, monthly surveys are among the leading indicators used to judge the direction of the economy and the likelihood of policy support from the People's Bank of China. The sub-indexes on input and output prices give an early view of inflation pressure in factory goods, which matters for import prices in the rest of the world.

Reactions are strongest in assets tied to Chinese demand. Prices of iron ore and copper, the currencies of commodity exporters such as the Aussie, shares in emerging markets and the Chinese yuan all tend to respond to a large surprise. The export orders component is followed as a gauge of world trade, and it has drawn particular attention in periods of tariff disputes.

05

Limitations

The index measures the breadth of change, not its size: it shows how many firms report improvement, not by how much output rose. The sample is modest for an economy of China's scale, and monthly movements of a few tenths of a point are within the range of noise. Seasonal adjustment around the Lunar New Year is imperfect, so January and February readings are usually considered together. As with any survey of opinion, answers can be coloured by sentiment, and the index is best used with hard data on production, trade and credit across the business cycle.

Last updated 16 Nov 2026