State-owned enterprises in China: How big are they?


This page in:

(Blog Admin note: please be aware that Gao Xu is no longer working for the World Bank and cannot follow up on new questions or comments on his posts.)

In China's industrial sector, the shrinking share of state-owned enterprises (SOEs) is evident. This is due in part to the stronger growth of non-government-owned enterprises.

When communicating with my friends outside China, their misconception of Chinese State-Owned Enterprises (SOE) always surprises me.  It seems that although Chinese SOEs play a major role in the Chinese economy, they continue to remain mysterious to most outsiders. For some people – even some long-time observers of China – Chinese SOEs are best described as dying dinosaurs that continuously absorb resources from the economy but produce little economic value. However, this impression is far from the case in nowadays.

That’s why I feel obliged to write something about the reality of this myth. In this blog post and the following ones, I’ll try to profile Chinese SOEs with data mainly from industrial sector, addressing questions regarding their size, profitability, leverage, and so on, hopefully presenting a clear picture to readers.

Today, I will tackle the question of how big are Chinese SOEs. This question can be rephrased in two different ways. First, how big are Chinese SOEs as a whole in the Chinese economy? Second, how big are individual SOEs compared with non-SOEs? Starting with the former question:

What is the share of SOEs as a whole in the Chinese economy? It depends on the criteria one chooses. This share is really negligible in terms of the enterprise number, but quite substantial in terms of assets.

The Second National Economic Census conducted in 2008 reveals that of all the 208 trillion RMB total assets of the secondary and tertiary sectors (industrial and service sectors), 63  trillion – or 30 percent of total – was held by SOEs. (SOEs here correspond to state sole funded corporations and enterprises with the state as the biggest share holder.) Meanwhile, in terms of enterprise number, there were 154,000 SOEs at the end of 2008, only accounting for 3.1 percent of the total enterprise number. Hence, the big picture is clear: SOEs control a substantial part of total enterprise assets in China despite the fact that their total number is marginal. As a corollary of this observation it follows that the average size of SOEs is much bigger than that of non-SOEs. This is indeed the case. In terms of average assets, SOEs are equal to 13.4 times of non-SOEs.

To analyze the developments, we now turn to the industrial enterprise survey dataset. As the National Economic Census is conducted once in 5 years, it is hard to study the dynamics with only the census data. Fortunately, there is another dataset that continuously dates back to 1999 readily available to us – the industrial enterprise survey data. This dataset is constructed and released by China’s National Statistical Bureau, covering all SOEs and other enterprises with annual sales larger than 5 million RMB in the industrial sectors. (Note: under the current foreign exchange rate, US$1 equals 6.8 RMB.) As non-state-owned small industrial firms (annual sales smaller than 5 million RMB) are excluded from the survey, this is a biased sample. However, enterprises covered by the survey in total account for almost 90 percent of total industrial value-added, suggesting this dataset is good enough to give a big picture of the overall developments.

The industrial survey data corroborates the Economic Census. In Figure 1, the dark blue line shows the share of SOEs number in total surveyed industrial enterprises, juxtaposed with the asset share shown by light blue line. At the end of 2008, SOEs only accounted for 4.8 percent of total enterprise numbers but controlled 43.8 percent of total assets owned by surveyed firms. Again, the data suggests that the average size of SOEs was much bigger than that of non-SOEs (15.5 times to be specific).

The shrinking share of SOEs in the industrial sector is evident in figure 1. This is on one hand due to the stronger growth of non-SOEs, and on the other hand, with less quantitative significance, reflects the fact that more non-SOEs that used to be excluded from the survey have crossed the 5 million RMB sales threshold and thereby entered into the sample. Once one takes into account the fact that the sample covers 90 percent of the industrial sector, or 39 percent of the national economy, the diminishing SOE share in the sample clearly points to a bigger role played by the non-SOEs in the Chinese economy.

It is worthwhile to note that the decline in the share of SOEs is much bigger than that in the share of SOE assets. Since 1999, the share of SOEs has declined from 37 percent to less than 5 percent in terms of numbers, and from 68 percent to 44 percent in terms of assets. This is to a large extent a result of the SOE reform – specifically the “grasping the big, letting go of the small” strategy – carried out in the past decade. Since the reform started in late 1990s, most small SOEs have been privatized or filed bankruptcy to lessen the burden of the government, while larger ones have been subsidized (directly or implicitly) and/or merged to hopefully create stronger firms controlling the commanding high of the national economy. This reform strategy is a big success in terms of creating large SOEs. As what is shown in figure 2 by the dark  blue line, the average asset size of industrial SOEs increased from 134 million RMB in 1999 to  923 million in 2008, expanding by 589 percent in 9 years. Meanwhile, the average assets of non-SOEs only moderately increased from 36 million to 60 million, up by a dwarfed 67 percent.

With all these evidences in hand, we can now answer the question raised before: How big are Chinese SOEs? Even though their share in the economy continued to decline in the past decade, SOEs still make up a substantial part of the national economy – roughly controlling 30 percent of the total secondary and tertiary assets, or over 50 percent of total industrial assets. The average size of SOEs is much bigger than their non-SOE peers, with average assets of the former equaling over 13 times of the latter.

Now we know that Chinese SOEs have a relatively bigger average asset size, while they continue to account for a substantial part of the national economy. Naturally, one would ask what performances of these big SOEs are. This will be the topic of my next blog post – Chinese SOEs: how profitable are they?

Photo courtesy l_prusecki under a Creative Commons license.
(Blog Admin note: please be aware that Gao Xu is no longer working for the World Bank and cannot follow up on new questions or comments on his posts.)



Gao Xu

Macroeconomy - China

Damjan D
February 01, 2010

Greetings Mr. Xu

I have a feeling that my request will be covered in part two but I will submit it anyway. Can you talk about the impact that the large holdings of SOEs can have on inflating true value in markets?

Ex: One of the explanations for there being a bubble in the real estate market is that large SOEs are bidding the price of land too high for anything built there to be profitable in the long run.

Gao Xu
February 03, 2010

Thanks for the good question.

Regarding the recent price surge in China’s property market, I think the role played by the SOEs is overly exaggerated by the media. It is true that some SOEs, armed with big money borrowed from banks, are purchasing land at astonishing high prices. For example, Franshion Properties, a subsidiary of the Sinochem Group (a Chinese SOE ranked 170th in the Fortune Global 500 list) and a mere nobody in the real estate sector not long ago, won in an eye-catching land auction in mid 2009, beating other property giants. But the share of SOEs in the real estate sector is very small, less than 10 percent measured by any criteria. As this sector is clearly dominated by non-SOEs, it is fair to say that the impact of SOEs on the land price is marginal.

However, on the general asset price appreciation in the economy, it is hard to quantify the impact of SOEs, though they are clearly direct beneficiaries of the current credit boom. I’ll touch this topic in my future post about the leverage of SOEs.

February 18, 2010

Hello Mr. Xu, thanks for the insights! However, there are rumours that the Chinese statistics offices are not overly reliable. If the former Soviet Union is any guide, all their reporting would be driven by party political objectives - any views on that? Thanks!

March 10, 2010

Thank you for your perspective regarding SOEs. I is a bit opaque to the casual western observer. My question is this: How difficult and frequent is the privatization process for small to medium sized SOEs in China? Are the difficulties, such as they are, largely due to governmental obstacles, or are they more organic, market-driven impediments? Is the State heavily involved in the process?

Thank you very much in advance.

March 13, 2010

Dear Mr. Xu,

whilst you point out to the differences of SOEs, I somehow never really encounter any reference made to their geographical location. In my opinion this might be one of the determining factors why performance differences exists between the coast and the west as well as different levels of reliance on SOEs. It would be great, if you could elaborate this topic a bit.

Thanking you in advance.
Best regards

Fatih Oktay
March 20, 2010

Thank you Mr. Xu, for the great post...

Could you explain why "SOEs here correspond to state-owned enterprises and state sole funded limited liability corporations"? and not include some of say Share holding enterprises?

If you were calculating the share of the state, how would your calculation method differ?

Thanks and best regards

Gao Xu
March 23, 2010

Dear CrisisMaven, this is really an important question. In my mind, the Chinese statistics are not manipulated. Please see my blog post “Are Chinese statistics manipulated?” which will be put on the web in a few days for a lengthy response to your question.

Gao Xu
March 23, 2010

Unlike in Russia and Central and Eastern European countries, where massive privatization was pushed through by central governments in early 1990s, privatization in China has taken an unique pass since 1990s. Local governments, motivated by getting rid of fiscal burdens – SOEs under their control – sat in the driver’s seat in the privatization process.

Back to 1990s, more than 2/3 of SOEs controlled by local governments (mainly small and medium SOEs) were loss makers, with a substantial part of them having negative equity. These SOEs had become burdens of local government financing. Hence, started in early 1990s, privatization as a solution to the SOE problem was pushed forward by local governments. As the privatization was banned by the central government during that time, most of the privatization took the form of insider privatization – selling equities to employees and managers – to circumvent political and ideological obstacles. It was not until mid 1990s that the privatization was finally endorsed by the central government. Since then, it is estimated by some researchers that 2/3 of China’s SOEs with state assets amounting to 11 trillion RMB (1.4 trillion USD) have been privatized.

If you want to know more about the privatization in China, I recommend you to a very good research paper called “What Makes Privatization Work? The Case of China.” You can download it at the following web address (…)

Gao Xu
March 23, 2010

Yes, the geographical location of SOEs and their regional differences look like very interesting topics. I’ll elaborate it in my future blog post called “State-owned enterprises in China: where are they?”

Gao Xu
March 23, 2010

Dear Fatih, I’m sorry for the confusion. It would be better if I put a precise definition of SOE in the post. SOEs in my blog post are defined as enterprises with the state as their biggest share holder. Those share holding enterprises controlled by the state are included in the classification.

March 25, 2010

Mr Gao:

Thanks for your articles on SOE size and profitability.

My quick question:

Per the National Economic Census, the number of SOEs decreased by 20% to 143,000 in 2009. (This is slightly different from your number of 154,000 but let's go with the Census number for now).

Per your chart the average assets per SOE increased from RMB 300 MN in 2004 to RMB 923 MN in 2009.

Putting these assertions together, the total SOE assets have grown from RMB 53 Trillion to RMB 132 Trillion in 5 years (from 2004 to 2009).... that's a growth of RMB 80 Trillion in assets (US $ 11.5 Trillion). Are you sure about these numbers?

Gao Xu
April 04, 2010

Hi Raj, thanks for your comment. I can see there are two misconceptions on the data definition here: (1) The National Economic Census data covers all the enterprises in the secondary and tertiary industries, which is more broader than the industrial enterprise survey data used in the charts of this blog post. (2) The SOEs defined in the National Economic Census is much narrower than the definition I used in the post. Those share holding companies with the state being the biggest share holder are not categorized as SOEs in the National Economic Census. But in my post, these firms are counted as SOEs.

Hence, the 143,000 in the National Economic Census is the number of firms solely funded by the state (excluding share holding enterprises controlled by the state) in the secondary and tertiary industries. The 154,000 I reported in the post is the number of firms controlled by the state (including share holding enterprises) in the secondary industry. So, you calculation is not correct.

June 02, 2010

would be appreciated for your to share your thoughts particularly on the CHinese construction SOEs and their performance if any.


January 13, 2011

Dear Mr Gao,

I am an economic researcher for a Beijing-based law firm, and I just wanted to say that I found the article above very well written and thoroughly researched - it was of great use to me.

I look forward to reading other articles of yours,


January 17, 2011

Dr. gao

Where do you see SOEs going in the future. Will they continue to increase in asset size proportionally faster than private firms or do you see that trend coming to a halt?

Similarly, how many SOEs will be left when the Chinese government sopts cutting their numbers? The two graphs you have produced seem to show no slowing down in the trends that have occurred in the last decade.

Your thoughts would be most appreciated,


February 03, 2011

Dear. Gao

Thanks for the instructive posting.
Could you let me know the reference of Fig.2?
Actually, I'm writing research paper and I'd like to identify the correct source for it.
It's not that easy for me to find out that graph in CEIC website.

Dong-Joon, Lim

Claudia Gabarain
February 04, 2011

Dear readers, it's encouraging to see Gao Xu's posts still generate comments and questions many months after being posted, but I'm afraid he's no longer working at the Bank and he cannot address new comments.
Cheers, and thanks for reading.

-- Claudia

March 29, 2011

It would be very much appreciated if you could pls let me know where I could find information and the names of Chinese SOEs including their listing either domestically or overseas.

Chum Sokha
February 05, 2016

Please explain me why SOE in China can't sign MOA within 10 working days from MOU?

Paul Godfrey Marley
October 18, 2021

How China managed to use state owned enterprises to run its economy

nick clement
October 18, 2021

Greetings, I am trying to understand the rationale for the smaller SOEs that may be connected to cities. Where do the profits go? Why aren't the profits distributed to the people of the city. Who gets the profits?