Desktop version

Home arrow Political science

  • Increase font
  • Decrease font

<<   CONTENTS   >>

Broad Trends of Change

Ascertaining the magnitude and change of public ownership in the postMao Chinese economy is a challenging task. There are three major sources of measurement errors. First, the official classification of public versus private economic activities has been evolving and contains ambiguities concerning mixed-ownership organizations. Second, there are reporting errors regarding the true nature of ownership for a sizable number of economic organizations. Third, there is a lack of systematic information on the major indicators that are useful for a comprehensive and accurate assessment of the private sector’s significance in the economy. Despite these problems, it is still possible to use existing data to piece together a clear enough picture of the general trend in China’s ownership transformation.

Evolution of Ownership-Based Classifications

Before the start of economic reforms there were only two ownership-based categories of economic organizations outside farming (which was run by people’s communes): state-owned enterprises and collective enterprises. In the 1980s the categorization was expanded to includegetihu, private enterprises, joint (rural) household enterprises, Chinese-foreign joint ventures, wholly foreign-owned enterprises, joint ownership enterprises between public enterprises (state-state, state-collective, and collective-collective), and employee shareholding cooperatives. In government statistics, however, these heterogeneous organizational forms were oftentimes lumped together under an “other” category, especially before 1998.

Into the 1990s the categorization of economic organizations became even more complicated. The main sources of complication lie in the formation of new, mixed-ownership organizations under the shareholding system codified by the Company Law in 1994 and in the creation (under the concurrently adopted Ordinance on the Administration of Company Registration) of a new type of owners called “legal persons,” or institutional owners, which may be entirely publicly owned, entirely privately owned, or co-owned by public and private owners with various distributions of shareholding and control rights. Starting from 1998, however, the National Bureau of Statistics (NBS) has used the label “state-controlled companies” to refer to limited liability


companies and joint-stock companies where the state holds the controlling shares (above the 50% threshold or by agreement). In official (especially NBS) statistics, these companies are defined as SOEs and often grouped together with traditional state-owned sole proprietorships not (yet) (re)organized according to the Company Law. Solely state-owned limited liability companies organized according to the Company Law are sometimes listed as a subcategory under “limited liability companies” and therefore are identifiable. Limited liability companies and joint-stock companies founded or controlled by natural persons only or by legal persons with natural persons as the only founders or controlling stakeholders are categorized, along with private sole proprietorships (other than getihu) and partnerships, as “private enterprises” (siying qiye).[1] These measures help reduce some of the ambiguities concerning the classification of shareholding entities. But there remain difficulties in ascertaining the controlling stakeholders in Chinese-foreign joint ventures and those in shareholding companies that are not explicitly categorized as state-owned or state-controlled companies or as “private enterprises.”

In the survey of aggregate data below I examine ownership change by defining the public sector as consisting of entities that are explicitly classified by the government as state-owned enterprises, state-controlled companies, solely state-owned limited liability companies, urban and rural collective enterprises, and joint ownership entities between public enterprises.[2] I use the term “quasiprivate enterprises” to denote shareholding companies not classified by the government as SOEs, collective enterprises, or private enterprises. These entities, which I will call “label-less” shareholding companies, have various combinations of individual and institutional owners and controlling stakeholders that do not meet the above-mentioned “natural person test” for inclusion as private enterprises. I also include in this category all Chinese-foreign joint ventures, as it is difficult to ascertain from available data the extent of ownership and control rights held by the public sector partners in these entities. I treat all other economic organizations as being private or predominantly private, includinggufen hezuo qiye, or employee shareholding cooperatives, which are categorized by the government as a form of “collective” enterprise. A special kind of shareholding entity, these cooperatives pool truly private funds from members and thus fundamentally differ from traditional collective enterprises, which have nominally “collective” but individually indivisible (among members of the collective) assets and operate under an external “public” authority as the holder and controller of such assets. Table 1.1 summarizes the organizations in each of the three categories defined above.

A problem with this method of classification is that the size of the public sector may be underestimated by the inclusion as “quasi-private enterprises” of some “label-less” shareholding companies where the controlling stakes are held by institutional owners that are actually state-owned or controlled.[3] This distortion is likely to be more pronounced for the years since 2003, when the tide of massive privatization subsided and an increasing number of the remaining SOEs began to reorganize and expand their businesses by adding subsidiary companies, forming alliances, investing in existing companies, and even acquiring private companies (Fan and Hope 2013). It is important to recognize the resultant ambiguities in ownership and control. Yet it is also important not to treat these state-invested companies as similar to SOEs,[4] which are directly controlled by state authorities and faced with greater constraints on the pursuit of private interests and agendas. In fact the statistical distortion associated with “label-less” shareholding companies with state investment may be limited by a number of factors, thus not undermining the general validity of the “quasi-private” categorization.

Table i.i Categorization of nonfarm economic organizations


Category in official statistics

(1) Public enterprises

State-owned enterprises (sole proprietorships not organized according to the Company Law)

State-controlled companies (limited liability companies and joint stock companies with controlling stakes held by the state)

Solely state-owned limited liability companies

Urban collective enterprises

Rural collective enterprises (township and village enterprises, or TVEs)

Joint ownership enterprises between SOE(s) and(/or) collective enterprise(s)

(2) Quasi-private entities

Limited liability companies not classified as SOEs or private enterprises

Joint stock companies not classified as SOEs or private enterprises Chinese-foreign joint ventures

(3) Private or predominantly private entities

Private sole proprietorships

Private limited liability companies

Private joint stock companies

Private partnerships

Joint (rural) household enterprises

Employee shareholding cooperatives

Wholly foreign-owned ventures

Foreign-invested joint stock companies

Joint ownership enterprises between private entities


First, there are clear regulations by the SASAC, SAIC, and NBS that classify state-invested companies into three types of shareholding: solely state- owned (guoyou duzi) (type 1), state-controlled (guoyou konggu) (type 2), and those with state-held shares (guoyou cangu) (type 3).26 Registering or

For a sample of these rules, see; http://; nels/ch0 0167/200 8 03/F9319286-1823-4AB1-BD24-77F879A8495F.htm; stats info/auto2072/201311/t20131104_4549 01.html.

reporting a type 1 or type 2 company as a type 3 company is a violation of the rules. This imposes a constraint on the extent of misidentification. Second, where the rules are violated or not followed closely for reasons other than administrative oversight, the companies concerned are likely to be less closely monitored by state assets supervision authorities and thus more prone to the manipulation by insiders and outsiders for private agendas (hence likely being “quasi-private”). Third, when a company with both private and state-owned shares is controlled by private owner(s), it tends to follow the logic of private ownership more than state agendas despite the lack of the registration label “private enterprise.” In fact it is not uncommon for SOEs to make investment in “label-less” shareholding companies that does not amount to a controlling stake.[5] While in general such investment may not cause the kind of statistical distortion mentioned above, potential problems may arise in a much less common scenario, where different SOEs constitute the largest shareholders of a “label-less” shareholding company that does not have a clearly established controlling stakeholder. But even in that case one needs to take precaution before coming to the conclusion of de facto state control, as it may be foiled by insiders (e.g., the management team) holding much smaller amounts of shares.[6]

  • [1] See and 0099 8.html.
  • [2] It should be noted that reporting errors about de facto private entities in the economy may causeexaggeration of the size of the public sector. From the late 1970s to the late 1990s, in order to containpolitical risk and leverage business opportunities, many private entities disguised themselves as publicenterprises (Shi 1993; Young 1995; Dickson 2008). There were (and still are) also companies registeredas “public enterprises” but in fact used by government officials and agencies as front organizations fortheir private profit-seeking activities (Duckett 1999; Lin and Zhang 1999; Lin 2001).
  • [3] A concern may also be raised about the predominance of public ownership in some Chinese-foreign joint ventures. Before 1993 the main organizational form of FDI was joint venture with publicenterprises, where control power often tilted toward Chinese partners (Pearson 1991; Huang 2002).When public enterprise partners indeed had greater say in decision-making, however, foreign partnerswere unlikely to resemble generally powerless small shareholders in large (especially listed) companies.Instead they could seek to protect and address their interests through negotiation and/or with thethreat of exit. So the term “quasi-private” still applies even when foreign partners were not predominantdecision-makers in joint ventures with public enterprises. Moreover, the situation has changed since themid-1990s. Firm-level data from the industrial sector, to be presented in the next section, reveal a subsequent decline of the importance of the joint venture form relative to wholly foreign-owned enterprises,along with a decline of the equity shares of public sector partners in joint ventures and a rise of therelative significance of (domestic) private-foreign ventures. As a result, underestimation of the significance of predominant public ownership in this group of quasi-private enterprises becomes a lesser issue.
  • [4] For an opinion that extends the notion of SOEs to “label-less” shareholding companies with stateownership, see
  • [5] China Life (an SOE), for example, was one of the founders and until mid-2015 one of the largestshareholders of the Minsheng Bank, a predominantly private joint stock company without an officialregistration status as private enterprise.
  • [6] A case in point is China Vanke, one of the largest real estate companies in the country. An SOEspin-off and a “label-less” joint stock company without a clearly established controlling stakeholder, ithas been controlled by its founder Wang Shi and his associates despite the presence of larger shareholders that are SOEs. For an interesting discussion of the recent disputes and struggles over control rightsat Vanke, see
<<   CONTENTS   >>

Related topics