Home Political science
Measurement Issues and General Patterns
Table 1.2 provides a glimpse of the changing landscape of nonfarm economic organizations in the post-Mao era. It shows that the number of public enterprises increased substantially before the mid-1990s but experienced contraction in the next two decades, when private and quasi-private entities demonstrated significant growth, especially during the last decade shown. To go beyond this broad view for a closer assessment of the magnitude of private economic activities, it would be useful to examine data on three major indicators: assets, output value, and workforce. Unfortunately, the Chinese government has never published full and detailed information that can be used to
Table 1.2 Number (millions) of industrial and commercial organizations
Notes: (a) SOEs include traditional (pre-1994) SOEs, state-controlled companies and solely state-owned limited liability companies; (b) figures in parentheses are shares of joint ventures in foreign-invested enterprises.
Sources: SICA 1994, 2004, 2014; SFYCICA.
cross-tabulate these indicators with different ownership types and compare the results over time.29
In the case of assets, no systematic data exist on the farming sector (Pei Changhong 2014), where farmland has remained publicly owned and been contracted out to rural households since the dismantling of the people’s commune system in the late 1970s and early 1980s, while other farm assets have been mainly privately owned. Limited information is available on the assets of nonfarm economic organizations. Before the accounting system reform in 1993 (Wu and Patel 2015), no balance sheet data were reported by domestic industrial and commercial enterprises, as the old accounting system was designed to record and track fund use and flow for a centrally planned system based on public ownership. Nor were land and intangibles (copyrights, trademarks, brand names, etc.) valued and included as asset items. Except for the industrial sector and nonfinancial SOEs, most of the asset data collected since 1993 are not released. It was not until 2004, when the first economic census was conducted (and then followed by the second and third ones in 2008 and 2013 respectively), that asset data were collected on all economic organizations in the secondary and tertiary sectors, thus providing a baseline for assessing the relative significance of public enterprises. However, the government has withheld much of the ownership- related information that can be used for cross-tabulation, with the data on the financial sector being most closely guarded. Despite these limitations, 29. The quality of published data is another issue, which cannot be addressed in this book. See Holz 2014 and Rawski 2001 for assessments of the quality of China’s GDP data.
from the data available it is still possible to discern some general patterns and trends.
Table 1.3 shows that the shares of public enterprises in industrial assets have declined. Although the scope of the time series varies for different periods, it is still clear that the decline was more drastic around the turn of the century and that collective enterprises have lost much of their significance.
Table 1.3 Shares (%) of public and nonpublic enterprises in industrial assets, 1993-2014
Table 1.3 Continued
Note: Figures in parentheses are shares in equity capital.
Sources: CSY(various years), CECY2004, 2008, 2013, FYC201$; SSIC 1995, vol. 1; CIESY2006.
Table 1.4 Shares (%) ofpublic enterprises in secondary and tertiary sector assets
Note: The shares of SOEs in all secondary and tertiary sector assets as reported in the statistical summaries of the three economic censuses do not include those of state-controlled joint-stock companies and therefore cannot used as indicators for all state sector entities. The alternative measure reported in this table is derived by dividing the total assets of nonfinancial SOEs with the total assets of nonfinancial economic organizations reported in the three economic censuses reported in the Finance Yearbook of China.
Sources: http://www.stats.gov.cn/tjsj/tjgb/jjpcgb/; CECY2004, 2008, 2013; FYC2010, 201$.
The shares of “quasi-private” industrial entities stabilized after 2004, whereas the expansion of private enterprises continued until 2010. The more encompassing information reported in Table 1.4 confirms the overall shrinkage of SOEs after the turn of the century. It also suggests that the decline of SOEs may have eased in the postprivatization years, though. But there is no sign of a trend reversal—even during 2008-2013, which was the focal period of theguojin mintui debate. The widening gap between industrial SOEs’ shares in assets and equity capital (reported in parentheses in table 1.3) after 2007 further indicates that the assets of SOEs consisted of increasingly more liabilities than those of private and quasi-private organizational types. An implication of this is that the easing in SOEs’ declining asset shares during the past decade may have been at least in part aided by heavier use of financial leverage under continued preferential treatment by the state-dominated banking system.
Ascertaining the share of public enterprises in economic output is equally difficult. In 1993 the Chinese government replaced the Soviet-style “material product system” (MPS) with the Western-style “system of national accounts” (SNA) for accounting and output reporting. Since then it has compiled and reported gross domestic product (GDP) as the main indicator of economic output and reconstructed the data series for the years before 1993. But, again, the government has withheld information on many economic sectors, organizational types, and time periods. As a result, it is not possible to derive directly from published data an economy-wide view of the changing contributions from public and nonpublic entities. Nevertheless, one can still piece together a partial picture from two limited time series—one on the industrial sector and the other on nonfarm economic activities in rural areas.
Table 1.5 shows that the GDP contributions by publicly owned industrial enterprises and by rural collective enterprises both declined from their peak levels in the mid-1990s. Collective industrial enterprises and collective enterprises in the rural nonfarm sector experienced steady declines in both absolute terms and relative to the all-inclusive groups. In contrast, the decline of industrial SOEs reached a bottom in 1997, when massive privatization began. Interestingly, their GDP share edged up in the ensuing decade, though this was not enough to reverse the decline of the public sector’s share in both industrial GDP and total GDP (see the next paragraph). It is unclear whether the trend continued after 2007 and whether it took place among nonindustrial SOEs as well. It also remains to be carefully investigated whether the limited bounceback during and after massive privatization was due to improvement in governance, monopoly positions held by remaining SOEs in upstream industries that rode on the economic expansion in the postprivatization era (Wang 2015), or both.
For GDP shares of public and nonpublic entities in the economy as a whole, no published information can be used for making a direct calculation. Estimates are legion, though. Table 1.6 provides a brief summary. It
Table 1.5 Shares (%) of contributions by public enterprises to GDP
Notes: (a) Contributions from industrial enterprises in the public sector are defined as shares of industrial value-added in GDP; (b) the scopes of coverage for publicly owned industrial enterprises during different periods are the same as those indicated in table 1.3; (c) breakdown figures on industrial value-added are unavailable for the years before 1992 and after 2007.
Sources: CSY(various years); CIESY2006, 2007, 2008; CTEY(various years); CTEAPPY(var- ious years).
seems clear that the overall share of publicly owned entities has declined. The two most recent estimates—by Pei Changhong (2014) and jointly by the World Bank and the Development Research Center of the State Council (2014, 104)—put the share of the nonpublic sector and that of the
Table 1.6 Estimates of contribution (%) by the nonpublic sector to GDP
Note: Li Chengrui was director (1981-1984) of the State Statistical Bureau; Fan Gang is director of the National Economic Research Institute (http://www.china.com.cn/chinese/ OP-c/277946.htm); Huang Mengfu was chairman of the All China Federation of Industry and Commerce (http://theory.people.com.cn/GB/49154/49155/3981648.html); Pei Changhong is director of the Institute of Economics, Chinese Academy of Social Sciences.
nonstate sector at 71% (in 2012) and 70% (in 2010-2014) respectively. If one uses the latter as the benchmark and deducts 2% from the non-SOE portion as the share of collective enterprises, then a more conservative estimate for the share of the nonpublic sector in total GDP would be 68% for 2014.
There are more detailed and systematic (albeit not without gaps) data series about the workforce and its distribution among organizations of different ownership forms. Given such data availability and considering the importance of employment in socioeconomic life and to political decision-makers,
I will focus more on this indicator for assessing the relative magnitude of the nonpublic sector in the remainder of the book. The summary statistics in tables 1.7 and 1.8 reveal several important facts. First, public enterprises made substantial contributions to nonfarm job creation from 1980 through 1995, yet the result was not strong enough to maintain their initially dominant position in nonfarm employment.
Second, the significance of public sector employment declined sharply in both absolute and relative terms during the ensuing decade, when massive privatization was in full force. Collective enterprises experienced much greater shrinkage than state-owned entities, as suggested by the rising percentages of the latter in the remaining public sector workforce. Although the pace of decline slowed down in the postprivatization years, the trend continued. By 2014 the share of public sector employment in the entire workforce had dropped below 10%.
Third, the overall significance of quasi-private enterprises in nonfarm employment rose steadily from the mid-1990s onward. But this was solely because of the growth of shareholding enterprises, as the share of Sino-foreign joint ventures actually shrank. In the meantime the share of the residual category—private or predominantly privately owned entities—also substantially grew, most probably passing the 70% level in 2014.
Fourth, there were variations between urban and rural areas and among different provinces. Table 1.6 shows that during the first decade of reforms, public sector employment dominated urban areas but experienced serious erosion in rural areas. Within the rural sector, there were also considerable variations among different provinces. In the next decade, the growth of employment outside the public sector continued in rural areas, which was paralleled by a similar but more speedy trend of change in urban areas. These changes not only extended into the decade following the start of massive privatization in 1997 but were accompanied by greater variability within urban and rural sectors, as indicated by the increase in the coefficient of variation.
To sum up, the foregoing survey illustrates a steady growth of private ownership in the post-Mao era. It also highlights a concurrent process of change, where public enterprises initially experienced expansion in absolute terms but subsequently declined, with collective enterprises under local authorities leading the way, both up and down, and at uneven paces across regions.
Table 1.7 Changing significance of public entities and quasi-private entities in nonfarm employment
Notes: (a) State-owned entities include both economic organizations and noneconomic organizations (including party and government agencies); (b) shareholding enterprises refer to limited liability companies and joint-stock companies that are not categorized as state-controlled companies or private enterprises.
Sources: CSY(various years); YICAC1992,1993,1994; CTEY(various years); CTEAPPY(various years); CECY2004, 2008, 2013; CSNCBU.
Table 1.8 Share (%) ofpublicly owned entities in secondary and tertiary
Notes: (a) 1985 is the earliest year for which provincial data on rural nonfarm enterprises are available: (b) CV stands for coefficient of variation, which is the ratio of the standard deviation to the mean.
Sources: CSY1986,1989,1999, 2009; SCTE; CTEAPPY2009.
Table 1.9 Sectors with dominant SOE shares (%) in economic census years
Notes: (a) Value-added data are unavailable, and sales data are used as proxies of (gross) output; (b) figures with “~” or “>” indicate approximate magnitude or range due to the lack of precise data; (c) figures in parentheses indicate drop of percentage below 50%.
Sources: CSY2005, 2009, 2014; CECY2004, 2008, 2013; FYC2006, 2015.
Relatively more SOEs survived the wave of massive privatization around the turn of the century, though their presence in the economy was also substantially reduced. When the SASAC was established in 2003 to consolidate control over the survivors, the state had retreated to a handful of strategically important sectors, as shown in table 1.9. Even there, the leading positions held by the remaining SOEs in some sectors after the tidal wave of privatization, such as coal, metallurgy, gas, water, and water transport, significantly eroded during the following decade. By 2013 the vast majority of industrial sectors,
Table 1.10 Number of industrial sectors with paid-in capital dominated by different owner groups, 2013
Note: The data cover all industrial enterprises with annual sales of no less than 20 million yuan. Source: CECY2013.
especially those identified by four-digit classification, were dominated by private and foreign-invested companies, as suggested by the statistics on paid-in capital in table 1.10.
The Growth and Organizational Patterns of FDI
Foreign investment is an important contributing force to privatization because the entry and expansion of foreign (private) capital not only dilute
Table 1.11 Selected statistics on the relative significance (%) of entities with FDI in the economy
Note: Figure in parenthesis is from the 2013 economic census.
Sources: CSY1995,1999, 2005, 2009, 2015; CECY2004, 2008,2013; SFDIC2015.
the significance of public ownership but, as I will discuss later, embody and stimulate institutional change. The re-entry of foreign direct investment began in 1979, when the government allowed joint ventures to be formed between public enterprises and foreign companies. From 1979 to 2014 China received a total of US$1.59 trillion of foreign direct investment (CSY 2015, table 11.13). Table 1.11, based on limitedly available data, captures some facts about the important roles of foreign capital in the economy. It shows that FDI enlarged its presence, while public enterprises were in decline from the mid-1990s through the turn of the century. Its relative significance in the industrial sector and nonfarm employment may have peaked during the postprivatization years, when domestic private entities experienced faster growth. But its contribution to government revenue seems to have remained stable following massive privatization.
Spatially, FDI has concentrated in the coastal region, as shown in table 1.12. But it is important to note that over time FDI has spread both within the coastal region and among noncoastal provinces. In 1985 61% of the cities (inclusive of subordinate counties) in coastal provinces were recipients of FDI, whereas the percentage was only 26% for those in noncoastal
Table 1.12 Selected statistics on the geographic and sectoral distribution of FDI
Notes: (a) 1985 and 1995 were industrial census years, 2004, 2008, and 2013 were economic census years; 2014 was the year for which pertinent data are available from latest sources; (b) the percentages of industrial sectors with FDI for 1985 and 2014 are estimated using information on three-digit and four-digit industry classifications from the statistical summaries of the 1985 industrial census and the 2013 economic census respectively; the percentages for 1995, 2004, and 2008 are calculated directly from census data with four-digit industry classification.
Sources: CSY (various years); SSIC1985; SSIC199$; CECY2004, 2008, 2014; CCSY1985,1995, 2005, 201$; SICA; SFYCICA; data of the 1995 industrial census and the 2004 and 2008 economic censuses.
provinces. A decade later, these shares rose to 97% and 65% respectively, and the rise continued among noncoastal provinces during the next two decades. Interestingly, a substantial part of the spatial spread had already taken place before the massive privatization and China’s WTO accession after the turn of the century.
A similar pattern can be found in the sectoral characteristics of FDI, which only had a limited presence in the economy in the mid-1980s. By the mid-1990s, however, foreign capital had already propagated some 90% of the industrial sectors based on the four-digit classification. Table 1.12 further shows that from the mid-1980s through the turn of the century the distribution of FDI was increasingly skewed toward industrial activities, especially manufacturing. As will be shown later in the book, such a pattern was consistent with the fact that government economic policies (especially at local levels) focused on industrial development during that period of time. The trend gradually shifted during and after massive privatization, when the service sector rapidly overtook industries as the leading destination of FDI. China’s WTO accession in 2001 was a major catalyst, as it was followed by a lowering and removal of many barriers to foreign entry in service activities. Equally noteworthy, though, is the growing importance of urban development to the self-interest calculus of the gatekeepers and regulators of FDI—that is, local political actors—following the restructuring of fiscal relations in the mid- 1990s, as will be detailed in chapter 3.
The geographic and sectoral expansion of FDI during the first decade of reform was largely achieved through forming joint ventures with public enterprises. That organizational pattern was subsequently redefined by three developments in the next two decades, which substantially reduced the significance of public ownership in the expanding foreign sector. First, wholly foreign-owned enterprise became the predominant organizational form of FDI. Second, private-foreign partnerships overtook public-foreign partnerships as the predominant group of organizations among joint ventures. Third, within the remaining public-foreign joint ventures the share of public ownership has been in decline. As can be seen in table 1.13, the shares of joint ventures in the total number of FDI entities declined over time relative to wholly foreign-owned enterprises, which are more “private” than joint ventures with public enterprises. Table 1.13 further shows concurrent declines in terms of investment and workforce shares. It also shows that among the FDI entities
organized as joint ventures there was steady growth in both the importance of joint ventures with domestic private owners and the significance of foreign shares in joint ventures with public enterprises.