HAVE CHINA’S ENTERPRIZE REFORMS LED TO IMPROVED
EFFICIENCY AND PROFITABILITY FOR PRIVATIZED SOEs?
Gongmeng Chen, Michael Firth, and Oliver Rui*
July 2000
ABSTRACT
About twenty years ago, China set about reforming its moribund economy by introducing
certain elements of free market capitalist economics. One reform was the partial privatization of
many State Owned Enterprizes (SOEs) and listing the shares in them on the stock exchanges of
Shanghai and Shenzhen. The partial private ownership of these companies was supposed to act as a
spur to improve profitability, efficiency, investment, and growth. In contrast to the operating and
financial improvements recorded in privatizations in other countries, our study shows that
performance deteriorated after the privatization. Profitability and efficiency statistics indicate a
decline in the three years after privatization. Sales growth showed modest gains. Capital investment
improved and debt levels fell after privatization.
We argue that the privatizations have been
unsuccessful (in terms of profitability and efficiency) because the state continues to hold substantial
shareholdings in companies and because the state often controls the make-up of the board of directors.
State shareholders are not fully committed to profitability and efficiency because they have various
social and political objectives that hinder economic performance. Based on our results we argue that
the state and its entities should sell all the shares they own if the privatized companies are to function
effectively and efficiently.
I.
INTRODUCTION
The privatization and subsequent stock market listing of State Owned Enterprizes (SOEs) in
many countries has increased dramatically in recent years (Kikeri et al. (1992)). This increased
activity has been occasioned by changes in socio-political ideologies, the need for governments to
raise cash and reduce state subsidies, and the desire to increase the competitiveness, performance, and
technological growth of the enterprizes. Privatization programs range from those conducted in
capitalist societies with established stock markets (e.g. Britain, Australasia, and Western Europe), to
those conducted in economically developing countries (e.g. Bangladesh, India, South America,
Turkey), and to those conducted in (formerly) communist countries that have no recent history of
stock markets and no recent history of market economies (e.g. Eastern Europe and Russia). In some
privatizations the state disposes of all its ownership interest, while in others the state retains a
minority or even majority ownership stake. When the state retains some ownership, the process may
be termed partial privatization; in this paper we use the general term, privatization. Privatizations are
sometimes accompanied by a fresh issue of shares where cash is injected into the company; this gives
additional resources to the managers and allows expansion of activities.
An extensive literature has developed which analyses, both theoretically and empirically, the
occurrence and the economic performance of privatizations. Although many empirical studies have
found that privatized companies experience improved economic efficiency and improved profitability,
we caution against assuming these results apply to all countries. Reasons for initiating privatization
programs, privatization processes, characteristics of privatized companies, maturity of stock markets,
national economies, and political systems, differ across countries and so the findings from a study
based on one country cannot be automatically extrapolated to another country.
The purpose of this study is to examine the changes in economic performances of newly
privatized SOEs in the People’s Republic of China (PRC). The privatization and listing of SOEs is
an integral part of China’s state enterprize reforms. Although privatizations in the United States,
Canada, South America, Western Europe, Eastern Europe, and the ex Soviet Union have been studied,
there are few studies relating to China. The corporatization and privatization processes adopted in
China differ markedly from those in other countries and so the experiences and outcomes from
1
research based on those countries cannot be imputed to privatized SOEs in the PRC.
One
characteristic of Chinese privatized enterprizes is that new capital is raised when listing takes place.
Another characteristic is that the state usually retains voting control or has extensive influence over
the firm.
Our results show that China’s enterprize reforms have not led to improved economic
performance by privatized firms. Comparing the profitability and efficiency of SOEs before listing to
the profitability and efficiency after listing, we find a reduction in performance.
This finding
contrasts with non-China privatization studies which generally show improvements in performance
measures. Other comparisons show that China’s privatization program is associated with reductions
in debt levels, increases in capital expenditures, and just a small growth in sales. We attribute the
poor performance of privatized firms to the half-hearted nature of the enterprize reforms and the lack
of good corporate governance structures.
The paper proceeds by reviewing some of the existing literature on privatizations. There then
follows a description of the privatization processes used in China. The sample and research design
are then described. This is followed by a presentation and discussion of the results. Finally,
conclusions are drawn.
II.
PRIOR RESEARCH
Research has identified myriad problems with State Owned Enterprizes. For example, SOEs
tend to employ excess labor (Boycko et al. (1996)), tend to hire top management because of their
political connections rather than executive abilities (Krueger (1990)), and tend to have social and
political objectives which involve wealth redistribution rather than wealth creation. The privatization
of SOEs can potentially remedy these problems and create vibrant companies that compete
effectively in world markets. Privatization represents an organizational change that helps clarify
property rights, motivates owners toward a common goal of profitability and economic efficiency,
and establishes a system of financial incentives. Advocates of privatization argue that the discipline
of private ownership will lead to greater economic efficiencies (Megginson et al. (1994) and Shleifer
(1999)). Yarrow (1986), Caves (1990), and Shirley and Nellis (1991) contend that it is the increased
2
competition and increased managerial accountability that usually accompany privatizations, that lead
to improved financial performance. If there is no increase in competition and if the shareholder base
is very narrow (e.g. largely held by government), then privatizations will likely fail to achieve
operating and financial efficiencies. 1
The conversion of SOEs to limited liability companies brings with it a different set of
principal-agent problems and these can reduce the potential gains from privatization (Vickers and
Yarrow (1991)). Kay and Thompson (1986) and Wortzel and Wortzel (1989) also discuss problems
that privatizations have in achieving efficiency gains. Partial privatizations where the state maintains
control, are likely to be beset with the same problems that plague SOEs.
There is a substantial literature that reports the results of empirical studies on the economic
gains to privatizations. These studies include single country analyses (Menyah and Paudyal (1996),
Vickers and Yarrow (1988), Bishop and Kay (1989), Caves (1990), Barberis et al. (1996), Martin and
Parker (1995), Ramanadham (1993) and Thompson (1987)) and multi-country studies (Galal (1994),
Megginson et al. (1994), Boubakri and Cosset (1998), Dewenter and Malatesta (2000) and D’Souza
and Megginson (1999)). In general, this research uses accounting data to measure profitability,
productivity (e.g. profits or sales per employee), asset and sales growth, and leverage. To evaluate
the numbers from the newly privatized company, comparisons are made to other already-listed firms
that are of similar size and in the same industry (e.g. Boardman and Vining (1989), Atkinson and
Halvorsen (1986), Pryke (1982) and Caves and Christensen (1980)). Other evaluations involve time
series comparisons where the change in performance is measured from pre-to post-privatization (e.g.
Megginson et al. (1994), Boubakri and Cosset (1998) and D’Souza and Megginson (1999)). There is
some inter-country variability in the way privatizations are listed but there are also commonalities.
Common features are the underpricing of IPOs, the use of fixed price offers, preference given to
domestic investors, and provisions that seek to keep the privatized company controlled by domestic
investors. These features are said to further the political and economic policy objectives of the
governments concerned (Perotti (1995) and Jones et al. (1999)).
Galal et al. (1994) used in-depth case analyses of twelve privatizations in Britain, Chile,
Malaysia, and Mexico (three from each country).
They concluded that in eleven cases the
privatization was successful in achieving net welfare gains; on average, the gains amounted to 26
3
percent of pre-privatization sales. In no cases did employment levels fall and in three cases there
were significant increases. La Porta and Lopez-de-Silanes (1999) examined the performance of 223
Mexican non-financial SOE privatizations made in the period 1983 to 1991. Changes in performance
from the four years prior to privatization to the year 1993 (post-privatization) were compared with
changes for matched-company or industry benchmarks. La Porta and Lopez-de-Silanes reported large
gains in profitability and output and a 50 percent reduction in employment levels. They attributed the
improvement in profits to productivity and efficiency gains rather than to price increases accruing
from the exploitation of monopoly power. One characterization of the privatization of SOEs in
Mexico is that incumbent senior management is typically replaced and a new team of top executives
is brought in.
Megginson et al. (1994) investigated the pre- and post-privatization financial performance of
61 companies spread over 18 countries and 32 industries.
They found that firms increased
profitability, improved productivity and operating efficiency, increased dividends, increased sales,
increased capital investment, and lowered debt ratios. The study also showed that the number of
employees did not decrease and that there were significant changes to the board of directors.
Boubakri and Cosset (1998) using similar methodology to Megginson et al. (1994), examined the
change in financial and operating performance of 79 companies from 21 developing countries that
were involved in partial or full privatizations during the period 1980 to 1992. Their findings mirror
those of Megginson et al. (1994), showing significant increases in profitability, operating efficiency,
capital investment, sales, employee numbers, and dividends. There was also some evidence of a
reduction in debt ratios.
D’Souza and Megginson (1999) examined the change in economic performance of 85
companies from 28 countries and 21 industries, that were privatized in the period 1990 to 1996. The
methodology was identical to the research designs employed by Megginson et al. (1994) and
Boubakri and Cosset (1998) and the results were very similar. Economic performance measured as
profitability, labor productivity, and sales growth all showed significant improvement after
privatization.
Significant reductions in debt levels and significant increases in dividends were
reported. Capital expenditures relative to sales and relative to assets showed no significant growth
after privatization. Employment levels fell slightly after privatization and this contrasts with earlier
4
research by Megginson et al. (1994), Galal et al. (1992), and Boubakri and Cosset (1998). The
results were disaggregated on the basis of the degree of competition, share ownership,
industrialization of nations, changes in the composition of boards of directors, and changes of chief
executive officers.
While all partitions were associated with improvements in performance,
companies in non-competitive industries (e.g. utilities) and companies where the state reduced its
ownership to below 50 percent had the most striking increases in profitability and efficiency.
Improvements in profitability could be the result of privatized companies in monopolistic
markets (e.g. utilities) increasing sales prices or otherwise exploiting market power. D’Souza and
Megginson (1999) conclude, however, that the profitability gains are not the result of exploiting
market power. There was no evidence of price increases and no evidence of increased cash subsidies
from the government; in fact, they report that in every sample case where governments were
subsidizing the SOE before privatization, these subsidies were explicitly terminated afterwards.
Boardman and Vining (1989) compared the operating and financial performance of State
Owned Companies to the operating and financial performance of listed firms. Their data came from
the Fortune International 500 companies in 1982 and covered large firms across many countries.
They found strong evidence that listed firms performed better than government owned companies.
Frydman et al. (1999) examined private ownership and corporate performance in the Czech Republic,
Hungary, and Poland and concentrated on small and medium size firms. Comparisons were made of
corporate performance (profitability, labor, costs) across privatized companies and state owned
enterprizes during the early 1990s (1991, 1992, 1993). The results showed that privatized companies
had better performance (improvements in profits, etc.) than State Owned Enterprises. This outcome
is attributed to private ownership. Additional results indicated that outside shareholders, rather than
management shareholders and employee shareholders, were associated with better performance;
‘internally’ owned privatized companies (ownership by incumbent managers and employees) had
similar performance to SOEs. Frydman et al. (1999) found that privatized companies had higher
employment levels vis-à-vis state owned enterprizes. One additional finding was that much of the
improved performance for privatized companies was attributable to increases in revenues rather than
reduction of costs. Claessens et al. (1997), examining data from 706 Czech companies involved in
privatizations in the years 1991 and 1992, and Pohl et al. (1997), examining 6300 private and state
5
owned firms in seven east European countries between 1992 and 1995, also documented improved
performance following privatization. Frydman et al. (1999) and Boycko et al. (1995) summarize
other privatization research carried out on the transitional economies of Eastern Europe and Russia.
Ehrlich et al. (1994) argue that productivity growth and efficiency gains to privatizations are
significantly enhanced if the state sells all of its stake in a SOE. Their empirical analysis was based
on panel data of 23 international airlines with varying levels of state ownership over the period 1973
to 1983.
Dewenter and Malatesta (2000) compared the performance of government owned and
privately owned firms. Their data consisted of 154 government firms and 1346 listed firms from 45
countries; data were extracted from Fortune Magazine for 1975, 1985, and 1995. They found that
government owned firms were less profitable, more highly levered, and more labor intensive, than
listed firms. A time-series analysis showed that privatized companies improved profitability prior to
the privatization but the evidence was mixed on whether profitability improved after listing.
Preparing for public listing appeared to give companies a spur to improving profitability but
subsequent to listing many performance measures declined.
Vining and Boardman (1992) and Megginson and Netter (1998) review empirical evidence
from studies across many countries and conclude that private companies perform better than State
Owned Enterprises 2 . Overall, the empirical studies suggest that privatization improves the operating
and economic performance of companies and that the changes represent socially beneficial
improvements in productivity, growth, and entrepreneurial activity. D’Souza and Megginson (1999)
conclude that privatization ‘works in a wide variety of countries, industries, and competitive
environments’ (p. 1433) and therefore in almost every institutional setting. It should be noted,
however, that none of the cited studies have used data from China 3 and we argue below that
privatizations in the PRC are very different from those in Eastern Europe, Mexico, and elsewhere.
III.
PRIVATIZATIONS IN CHINA
6
China’s political leadership in the post-Mao era realized that the centrally planned economic
system under communism had severe shortcomings and that the social welfare of its people was
likely to deteriorate. Under the paramount leader, Deng Xiaoping, the government initiated major
economic reforms which incorporated some elements of the capitalist or market economy systems.
These reforms were intent on increasing the self reliance of individuals, increasing innovation,
improving economic efficiency, and decreasing the financial demands placed on the state in
supporting SOEs and in supporting the welfare of people. Central pillars of the reforms were the
reduction or elimination of government control of markets (e.g. removal of production quotas and
product specifications, removal of price controls, enhancing the mobility of labor, and allowing the
freedom for individuals to engage in business activities) and the corporatizing of selected SOEs
where profitability becomes a major objective.
Initially, the reforms focussed on introducing competition in the market place, removing
production quotas and price controls, and giving more decision making powers to managers. There
was no change, however, in the ownership of SOEs (Gao (1996) and Rawski (1994)) and no change
in the heavy social burdens placed on enterprizes (Hu (1997)). 4 These reforms were instrumental in
enabling the Chinese economy to grow rapidly. At the same time, SOEs became increasingly
unprofitable (Lin et al. (1998) and Cao et al. (1999)) and were plagued by economic inefficiency and
under-investment. 5 Creating markets for products and giving discretionary decision making powers
to managers, were not, in themselves, sufficient to optimise resource allocation. This led the state to
the view that private ownership of companies may be needed in order to increase the economic
efficiency of SOEs. A policy was therefore established to convert selected SOEs into companies
with share capital and to sell the shares to institutional and individual investors who acted
independently of the state. The transfer, or more accurately the partial transfer, of ownership to
financial institutions and individuals reduces the power or influence of the state to interfere with
management decisions and will make company executives more accountable for their actions. Unlike
some other ex-communist countries, where the governments privatized 100 percent of its SOEs, the
Chinese government still wants to maintain ultimate control of the economy and so the new system,
termed the ‘socialist-market economy’, is a mixture of free enterprize and state planning and control.
Privatization of most SOEs in China are actually partial privatizations where the state retains
ownership of the majority of the shares.
7
To help operationalize the economic restructuring, the state and provincial governments select
various SOEs to be corporatized. This involves reorganizing SOEs and forming them into limited
liability companies legally distinct from the state and with ownership represented by share capital.
Initially, the share capital is owned by the state and various entities of the state. The next step is the
privatizing of selected corporatized SOEs. Here, shares in the companies are sold to individual
investors either by a new issue of shares (thereby injecting cash into the firms) and-or by existing
shareholders (the state and its entities) who reduce their investment stakes. 6 Although shares are sold
to individual investors, the state and its entities usually retain ownership of 50 percent or more of the
issued shares. The first shares issued to individuals occurred in 1984. These shares paid a stated
dividend and there was no ready secondary market for buying and selling the shares. This lack of
marketability for the shares dampened investor interest in privatizations.
In order to increase
marketability, and thus to make investment more appealing, two stock exchanges were created in the
early 1990s.
The Shanghai Securities Exchange (SHSE) opened in December 1990 followed in July 1991
by the Shenzhen Stock Exchange (SZSE). These exchanges trade A-shares and B-shares as well as
Government debt securities. A-shares are restricted to PRC citizens and are denominated in renminbi
(RMB). Beginning in 1992 some companies were allowed to issue shares to non PRC citizens; these
shares, restricted to foreigners, are designed as B-shares and they are denominated in U.S. dollars (on
the SHSE) and Hong Kong dollars (on the SZSE). A- and B-shares issued by the same company
carry equal rights and are comparable in all respects except for who can own them (A-shares for
domestic investors and B-shares for foreigners) and the currency in which they are traded and pay
dividends.
The segmentation of the A-share and B-share markets is effective as witnessed by the vastly
different market prices at which they trade (Poon et al. (1998)). 7 In 1993 certain SOEs were allowed
to issue shares to foreigners and obtain listings in Hong Kong (termed H-shares) and in New York
(N-shares); other foreign listings are planned for London, Sydney, Tokyo, Singapore, and elsewhere.
The issue of B-, H-, and N-shares helps raise foreign currency, raises the visibility of China in the
global financial industry, allows foreign practices and expertise to be observed and learnt, and
8
develops an alternative source of finance for China’s future needs. While the amount of capital
raised by issuing foreign shares is quite small so far, the future capital needs to modernize China’s
industry and its infrastructure are so immense that substantial foreign investment will certainly be
needed. The B-, H-, and N-shares issued to date may be viewed as a necessary learning experience in
tapping global financial markets and for these markets to learn about China.
Although there have been more than 700 privatizations by the end of 1998, the vast majority
of SOEs remain 100 percent owned by the state and its various agencies and entities. 8 The current
economic plan calls for an increase in the rate of privatizations. While many SOEs wish to privatize,
the Government has restricted the numbers.
The state wishes to maintain an orderly flow of
privatizations so as to not overwhelm the fledgling stock markets. Equally, the state is concerned that
the SOEs have the appropriate economic health, have an adequate management structure, and have
good business prospects, before they are allowed to issue shares to the public. An overall quota of
privatizations is set by the State Planning Commission, the People’s Bank of China, and the China
Securities Regulatory Commission (CSRC); this quota is allocated to individual ministries and
geographical provinces based on the state government’s national and regional development plans.
Selection of SOEs to be listed are based on economic and financial needs, commercial viability,
political objectives, society concerns, and guanxi (personal relationships). The SHSE and SZSE
analyze listing applicants to check their financial well-being and to ensure they have the necessary
track record (which normally includes having made profits in the three years prior to listing).
Privatized companies have three major categories of shares. First, there are shares owned by
government ministries and departments, and central and regional state governments. These shares are
non-tradable and they often account for more than 50 percent of the issued shares. Second, there are
legal entity shares that are restricted to other SOEs, institutions, and the foreign partners of
corporatized foreign joint ventures. Legal entity shares cannot be traded on the SHSE and the SZSE
although some limited buying and selling can be made through the Securities Trading Automated
Quotation System (STAQ) which is a country-wide computer-based exchange that started in July
1992. 9 Third, there are individual shares and these can be traded on organized stock exchanges (the
SHSE and SZSE for A- and B-shares, the Stock Exchange of Hong Kong for H-shares and the New
York Stock Exchange for N-shares). Companies aim for a wide spread ownership of A- and B-shares.
9
This aids the development of the stock exchanges and it also enables more individuals to participate
in capital ownership. In general at least 25 percent of the issued shares must be sold to individual
investors if the company has individual shareholders. Some A-shares are sold to company employees
although they generally represent but a small fraction of the total A-shares issued. Employee shares
are often issued with the condition that they cannot be traded for a certain period of time (often for
one year after issue); thereafter they are tradable just like any other A-shares. On average, for listed
companies, about one third of the issued shares are owned by the state, one third by legal entities, and
one third by individuals (A- and B-shareholders) and employees. Legal entities may be government
(or quasi-government, or SOE) owned and in these cases the entities may have objectives that
coincide with those of the state.
Normally only part of an SOE is corporatized and then privatized. Generally it is the
profitable operating assets and trade liabilities that are carved out into the privatized company.
Unprofitable operations along with what may be called societal or welfare operations such as
education, housing, and medical, are retained by the parent SOE or are hived off to the local
municipality or other state institution. 10 The decisions relating to what assets and operations are
transferred to the company are made by the state and state entities and are based on commercial and
political factors. Profitable and commercially viable operations are transferred to the company as the
state wants listed firms to be successful as further IPOs of SOEs will be made in the future.
Privatizations that perform poorly and suffer financial distress will make investors wary of
subscribing to future public offerings of SOEs. Additionally, poor corporate profitability and the
attendant poor stock price performance may deter SOEs from attempting to list and will be seen as a
sign of failure of the economic reforms and the move to a market economy.
One characteristic of privatizations in China is that the state, including its various ministries,
departments, and regional governments, tends to own or control more than 50 percent of the issued
shares. This contrasts with privatizations in many other countries where governments often end up
with no ownership of a company. The Chinese government is reluctant to relinguish its overall
control of both the economy and the companies that participate in it. Ostensibly the companies have
autonomy and operate in de-regulated markets, but in practice the government through its majority
share ownership, can vet decisions of management and can influence management to take courses of
10
action to achieve political and social objectives that may be detrimental to corporate profitability. It
is difficult to gauge the frequency of and the amount of pressure placed on a company’s top
management by the government.
While the state shareholders may have objectives that, in some circumstances, detract from
the profitability and efficiency goals of privatized SOEs, the role of legal entity shareholders is also
problematic. On the one hand some legal entity shareholders are financial institutions or profitoriented companies and they desire the privatized firm to maximise efficiency. On the other hand
some legal entity shareholders are effectively controlled by the state, regional government, or nonprivatized SOEs 11 and these may have objectives that depart from wealth maximisation. These legal
entity shareholders may be easily persuaded by the state to adopt non-wealth maximisation objectives
for their investments in privatized companies; this persuasion may include subsidies or other
economic rewards from the central or regional government. One characteristic of the legal entity
shareholders in a specific company is that each of them tends to own a sizable number of shares.
This gives them the economic motivation to analyse the company’s performance and to pressure
management to make changes if the operating and financial results are poor. If the combined
ownership by the legal entity shareholders is large, then they may have the voting power to remove
company management if performance is poor. Of course, if legal entity shareholders are under the
control of the state, regional government, or SOEs, they may use their voting power to install a
management team that suits the state’s objectives. Qi et al. (1999) report that Shanghai listed
companies have higher returns on equity if the legal entity share ownership is high. Similar evidence
is reported by Xu and Wang (1998). These authors did not examine changes in profitability and
changes in profitability from before privatization to after privatization. Legal entity shareholders
may therefore be investing in already-profitable companies rather than motivating company
management to become profitable.
Individual A-shareholders want companies to follow wealth maximising objectives but they
do not, in general, have the expertise to critically analyze a company’s performance and they do not
have the time or other resources needed to discipline managers. In order to establish a wide base of
shareholders and in order to create liquid stock markets, the numbers of shares allocated to a single
individual investor is quite small. The diffuse nature of the ownership of A-shares means individual
11
investors don’t find it cost-effective to monitor management. A-shares held by individuals have
extremely high turnovers indicating that investors typically own a stock for little more than two
months before selling. The short term horizons for individual investors is another reason why they
have little motivation to monitor management and force changes if performance is poor.
Foreign B-shareholders tend to be financial institutions based in Europe, Hong Kong, Japan,
and North America. These investors have the resources to analyze company performance and they
have experience (from elsewhere) in trying to effect operational and management changes if
profitability and efficiency are poor.
Foreign shareholders tend to own quite small stakes in
companies (because the number of B-shares issued by a company tends to be limited) and so this
might deter some B-shareholders from actively monitoring and disciplining managers.
Another characteristic of listed privatized firms in the PRC is that top management often have
little or no share ownership. This characteristic contrasts sharply with IPOs of private companies and
of former SOEs in other countries. Typically there are no reward systems based on economic
performance and so the lack of incentives provides no motivation for managers. The state often
reassigns managers every few years and so this discourages executives from taking a long term
perspective. The executive directors of the company are usually the top management from when the
firm was a State Owned Enterprize. Thus the top management do not have much experience of
operating in market economies where profit is a major corporate objective. This incumbency of
management is due to vested interests and to the general lack of executives who have the necessary
experience and proven skills to run the privatized firms. Top executive managements are often
appointed by the state and business acumen plays no role in the selection of managing directors or
CEOs. Non-executive directors are usually former executives from the SOE or are appointed because
of their political connections and political standing.
The government is reluctant to hire top
management from outside China to run the newly privatized SOE companies and this extends to
ethnic Chinese from Hong Kong and ethnic Chinese from other countries (the salaries of top
management are very low and foreigners, whether ethnic Chinese or not, do not find the executive
positions financially attractive). Privatized companies also have supervisory boards in addition to the
board of directors. Members of the supervisory board often include a majority of representatives
from the company’s workforce. Other members are often political appointees. These supervisory
12
board members have varying objectives but profitability may be somewhat low on their list of
priorities. The supervisory board oversees the activities of the board of directors and seeks to ensure
that the charter laws and shareholder resolutions of the company are adhered to.
The special characteristics of Chinese privatizations described above (share ownership
structure, appointment of top executives, and management reward and incentive systems) imply there
may be different objectives and motivations in running companies when compared to privatizations
in other nations. While the overriding goals of privatization are to modernize industry, improve
growth, satisfy customers, and improve economic efficiency, the share ownership and management
shareholdings characteristics may mitigate the economic gains from converting SOEs to public
companies.
This paper examines whether the strong international consensus that improved
profitability and improved efficiency accompanies privatizations, extends to the People’s Republic of
China. The results will shed some light on whether China’s unique style of enterprize reform has
been successful in transforming SOEs into efficient and profitable firms.
IV.
RESEARCH DESIGN
In order to investigate the impact of privatizations on the economic performance of SOEs we
compare the financial performance of the companies prior to listing to the performance after listing.
The research design we adopt is similar to those employed by Megginson et al. (1994), Boubakri and
Cosset (1998) and D’Souza and Megginson (1999); the similarity in research designs enables us to
compare the results from China to the results from international data. Alternative approaches to
examine the issue are not used because of sample and data limitations. Comparing privatized
companies’ performances with control groups of other listed companies is not used as all listed firms
have been recently privatized. Comparisons with non-privatized SOEs are not used as there are no
data available.
The economic variables we examine are measures of profitability, growth, capital investment
expenditures, and debt ratios. Data on employment levels before listing and changes in these levels
are not publicly available. Lack of employment data means that many productivity measures used in
other studies (e.g. sales per employee) cannot be calculated. Changes in dividends, which are
13
investigated in other studies, are not examined as very few China SOEs pay dividends prior to the
IPO. The variables for the before privatization period are measured as the averages of the variable
for the three years immediately prior to listing (t-1, t-2, t-3 where t is the year of listing; data for year
t-1 come from the financial accounts for the year ended immediately prior to listing). A variable for
the post listing period is the average of that variable for the three years subsequent to the listing year
(years t+1, t+2, t+3). Year t+1 is the first complete year after listing. Averaging over three years is
similar to the approaches adopted in other studies (e.g. Boubakri and Cosset (1998) and Megginson et
al. (1994)). IPO prospectuses give summarized accounting data for three years prior to listing 12 ; thus
our maximum averaging period is restricted to three years. Averaging over three years in the post
listing period allows us to include IPOs made up to 1995.
The variables we use are measures of profitability (profit divided by sales, profit divided by
assets, and profit divided by shareholders’ equity), capital expenditures (capital expenditures to sales,
capital expenditures to assets), sales, asset efficiency as measured by total asset turnover (sales
divided by total assets), debt ratios (total debt to assets, long term debt to shareholders’ equity). If
privatization spurs economic efficiency and business growth and innovation, then we expect
profitability, capital expenditures, growth, and asset efficiency to increase. A directional hypothesis
for debt ratios is less clear although other studies have documented a reduction in borrowings. An
explanation for the reduction in debt ratios in non-China studies is that governments use debt rather
than equity to finance SOEs. SOEs are able to borrow at a lower cost (because of implicit or explicit
government guarantees) and can more easily borrow from the government coffers. This leads to
higher levels of debt financing. Once a SOE is privatized, these government loans or guarantees are
reduced or withdrawn. Debt levels are also high for Chinese SOEs but, in contrast to other countries,
the government continues to provide loans and explicit or implicit guarantees for loans after
privatization. This is due to the privatized SOEs remaining majority owned by the state and due to
other sources of finance (e.g. cash flow from operations) being insufficient for corporate needs.
Privatizations are accompanied by the issue of new shares thereby increasing shareholders’ equity
and in some cases the cash proceeds are used, in part, to repay debt. Based on prior findings and the
fact that IPOs increase equity capital, we hypothesize a reduction in debt ratios.
One characteristic of the listing of privatized SOEs is that fresh capital is usually raised by
selling new shares to the investing public (A-shares and B-shares). This contrasts with privatizations
14
in many other countries. The new equity capital is used for buying long term assets, reducing debt
levels, and increasing working capital. The sales or profits generated from these new investments
may take some time to appear and in these cases changes in profitability and efficiency ratios will be
‘biased’ downwards 13 vis-à-vis the results from countries where privatized firms do not raise new
capital. To adjust for this ‘bias’ and to make our results more comparable with non-China studies, we
run our tests excluding the new issue proceeds from total assets and shareholders’ equity. 14 This
adjustment procedure is used in the tests reported in Tables II to VI. We also run the tests where new
issue proceeds are included in total assets and shareholders’ equity.
As sensitivity tests we also a) compare a performance measure from the three years prior to
listing to the measure at year t+3 and, b) compare year t-1 to year t+3. Measuring performance at
year t+3 allows time for the new issue proceeds to be invested and may capture the sales and profits
generated by these investments.
The significances of differences in performances across pre- and post-privatization periods are
tested via parametric t-statistics and non-parametric Wilcoxon and sign statistics. The results are
partitioned on the basis of the companies having or not having foreign investors (B-, H-, or N-shares).
The presence of foreign investors is hypothesized to lead to enhanced performance. There are several
reasons for this hypothesis. Foreign investors will be mainly interested in stock market returns (for
the level of risk borne). 15 Unlike domestic investors, foreigners have many types of securities from
many countries in which to invest. Foreign investors are unlikely to have a sentimental or ‘loyalty’
attachment to China and instead they invest to maximise returns for given levels of risk. Because of
foreign investors’ strong focus on profitability and efficiency, this desire may filter down to company
executives who are even more motivated to perform well. Most foreign investors are institutions and
they have the resources to analyze, lobby, and influence the companies in which they own shares.
Many SOEs wish to issue B- (H- or N-) shares because this brings prestige to the managers and may
even bring ‘western-style’ management rewards (foreign travel, stock options, etc.).
The state
decides which companies can issue foreign shares and the government is likely to choose companies
which are expected to perform well and are effectively managed. The PRC wants to show to the
outside world that its economic reforms are successful and so it wants the B-share companies to
perform well.
15
Data on share ownership is also collected. In particular, the percentage share ownership held
by the state, the percentage ownership held by legal entities, and the percentage ownership held by
individuals, are collected for each IPO. The legal entity ownership includes shares held by other
SOEs and therefore indirectly influenced by the state. Individual shares are represented by A-shares
and B-shares. We use the share ownership data to partition the sample in the univariate analyses and
the ownership data is also used in cross-sectional regression models.
A.
Regression Models
In addition to comparing the mean performance measures before and after privatization, we
also develop pooled intertemporal and cross-sectional regression models to explain the changes in
performance from pre-privatization to post-privatization. Independent variables are the change in
GNP per capita (from the three years pre-privatization to the three years subsequent to privatization),
foreign ownership, ownership structure (percentage ownership by the state, legal entities, and
individual shareholders), and control variables representing company size, stock exchange, industry
sector, and year of listing.
Changes in the economy are expected to have an effect on the performance of privatized
SOEs. A positive relationship is therefore hypothesized between change in GNP and profitability,
sales growth, and asset efficiency. Foreign ownership and share ownership by individuals are
expected to be positively related to profitability and asset efficiency. State ownership may have a
negative relationship to profitability and efficiency. Size, listing exchange, industry sector, and
listing years are included as control variables. The regression model is:
DV = β0 + β1ΔGNP + β2FOR + β3STATE + β4PUBLIC + β5SIZE + β6EX + β7EST +
β8CONG + β9IND + β10COM + β11Y91 + β12Y92 + β13Y93 + β14Y94
(1)
where:
DV = dependent variable. The dependent variables are percentage changes in profit/sales (ΔROS),
profit/assets (ΔROA), profit/shareholders’ equity (ΔROE), real sales (sales deflated for
16
changes in inflation) (GRO), asset turnover (ΔASTURN), capital expenditures/sales (ΔEXP/S),
capital expenditures/total assets (ΔEXP/TA), debt to total assets (ΔD/TA), and long term debt
to equity (ΔD/E);
ΔGNP = percentage change in gross national product per capita from the three years prior to the
privatization to the three years after privatization. The GNP numbers are deflated for
changes in inflation;
FOR = a dummy variable taking the value one (1) if there is foreign ownership in the company,
otherwise FOR is coded zero (0);
STATE = percentage of shares held by the state after the privatization;
PUBLIC = percentage of shares held by individual investors (A-shares and B-shares);
SIZE = log of market capitalization of the company after listing;
EX = a dummy variable taking the value one (1) if the company is listed on the Shanghai Securities
Exchange, otherwise EX is coded zero (0);
EST = a dummy variable taking the value one (1) if the company is in the property sector, otherwise
EST is coded zero (0);
CONG = a dummy variable taking the value one (1) if the company is in a number of industries (and
is included in the Conglomerate or Composite index), otherwise COMP is coded zero (0);
IND = a dummy variable taking the value one (1) if the company is an industrial or manufacturing
firm, otherwise IND is coded zero (0);
COM = a dummy variable taking the value one (1) if the company is in the retailing and commercial
sector, otherwise COM is coded zero (0); and
17
Y91, Y92, Y93, Y94 are dummy variables representing the year of listing (1991, 1992, 1993, or
1994). FOR, STATE, PUBLIC, and SIZE are measured immediately after the listing of the shares.
B.
Sample
The study examines 275 privatizations made during the period 1991 to 1995. This represents
all privatizations made during the period where the IPO and the listing of the shares occurred within a
year of one another. 16 In a few cases some historical (pre-privatization) and/or some future (postprivatization) data are not available and so these privatizations are excluded. We collect data on
profits, sales, total assets, shareholders’ equity, capital expenditures, total debt, long term debt, and
share ownership. IPO prospectuses are used to collect data for the three years prior to listing and
annual reports and corporate announcements published in newspapers are used to collect data for the
three years subsequent to listing. Industry sector information come from the SHSE and the SZSE.
Sales numbers (used in calculating change in sales) are deflated by the Consumer Price Index
published by the State Statistical Bureau.
About 58 percent of the issues were made on the Shenzhen Stock Exchange with the
remainder made on the Shanghai Securities Exchange. Most of the listings took place in 1992 and
1993. Table I details the year and exchange of listing of the 275 privatized firms.
The Shanghai
Securities Exchange opened in December 1990 and the first year for IPOs is 1991. Because we use
three year’s data for averaging, our sample includes companies listed in 1995 and which had three
years of post privatization accounts data available by 1998.
TABLE I HERE
V.
RESULTS
18
Table II shows the mean and median performance measures before (column 2) and after
(column 3) privatization, the hypothesized change in performance (column 4), the mean and median
difference between before and after performance measures (column 5), and the percentage of
increases in performance measures (or percentage decrease in the case of gearing) (column 6). tstatistics, Wilcoxon tests, and sign tests are used to examine the statistical significance of changes in
performance (columns 5 and 6).
TABLE II HERE
A striking feature of the Table II results is that profitability and asset efficiency deteriorate
after privatization. While the mean and median returns on sales change very little (the mean changes
from 12.30 percent to 12.37 percent), the mean and median returns on assets and returns on
shareholders’ equity fall substantially. For example, the mean return on assets drops from 8.48
percent to 6.43 percent, while the mean return on shareholders’ equity falls from 22.13 percent to just
13.15 percent. Column 5 shows the mean and median changes for returns on assets and returns on
shareholders’ equity are negative and statistically significant at the 0.01 level. The deterioration in
profitability contrasts sharply with the results from other privatization studies that use data from other
countries (Megginson et al. (1994), Boubakri and Cosset (1998) and D’Souza and Megginson (1999)).
In absolute terms, the return on assets and return on equity are very low. During 1991 to 1998 the
average bank interest rate is about 9 percent and this exceeds the return on assets after privatization.
Asset turnover, which reflects the amount of sales generated by assets, falls dramatically after the
privatization. The mean falls from 1.04 to 0.71 and the median drops from 0.77 to 0.54. The
reductions in the mean and median are significant at the 0.01 level.
The capital expenditure investment ratios show statistically significant increases. One source
of financing for the increase comes from the proceeds of the issue of new shares which accompany
the listing. Another source of finance comes from cash generated from operations. The evidence
from the PRC is similar to that from other countries; privatized companies significantly increase
capital expenditures compared to when they were wholly government owned. Table II also shows
that sales revenues increase after privatization although the statistical significance of the change is
weak; only the mean change is significant (the median change is not significantly different from zero
19
and the proportion of firms increasing sales is not significantly different from 50 percent). The sales
numbers have been adjusted for inflation. The sales gains after privatization are less than those
reported in studies based on data from other ‘developing’ countries and from ‘developed’ countries
(Megginson et al. (1994)). Given that the privatization also raises capital to expand operations it is
very disappointing that real sales have not improved after listing.
Results from Table II demonstrate a significant fall in debt ratios after privatization. The
mean (median) debt to total assets ratio falls from 0.56 to 0.46 (0.59 to 0.45) and the mean (median)
long term debt to shareholders’ equity falls from 0.32 to 0.15 (0.15 to 0.07). The significant
reductions in debt ratios are consistent with the results from other studies. One characteristic of our
data is that the long term debt to shareholders’ equity is low when compared with other countries
(Megginson et al. (1994) and Boubakri and Cosset (1998)). Most debt of Chinese SOEs is in the
form of short term liabilities and non-interest bearing short term debt. A reduction in debt ratios is
expected as the new issue proceeds increase equity and in some cases part of the proceeds is used to
repay debt. As our post-privatization debt ratios are measured over three years, the results are
consistent with there being no substantial new debt being raised other than perhaps that used to repay
maturing debt. 17 The debt ratios are also calculated by excluding new issue proceeds from assets and
shareholders’ equity. In this case the debt ratios still fall, but by much less. This suggests the actual
debt levels fall somewhat after privatization.
In order to examine if the results are influenced by the ownership structure of the privatized
company, the data are partitioned on the basis of whether the company has also issued foreign shares
(Table III), by whether there is high or low public or individual ownership (Table IV), and by
whether there is high or low legal entity share ownership (Table V). t-statistics are computed to see if
there are differences between the partitions. In general we do not comment on these differences as
they are not the main focus of the study. More importantly for our study, however, are the tests for
changes in performance from before to after privatization.
Approximately one third of the sample companies have issued foreign shares as well as the
domestic A-shares. The foreign issues are made at the same time as, or very near to, the issue of the
A-shares. The overall results in Table III indicate that the directional changes in variables are the
20
same across the domestic invested companies and the foreign invested firms. Except for returns on
sales and sales growth, the changes are statistically significant. Similar to the results in Table II,
returns on sales are little changed while the other profitability ratios have declined significantly for
both the domestic and foreign invested companies. Table III suggests that China chooses more
profitable SOEs for foreign share issues; return on equity is significantly higher for foreign invested
firms than for domestic invested firms in the three years prior to listing. The asset turnover efficiency
ratios decline significantly (at the 0.01 level) for both domestic and foreign invested privatizations.
The evidence from Table III does not support the hypothesis that foreign invested companies are
more likely to improve profitability and efficiency. Although foreign investors are hypothesized to
press management to pursue profitability and efficiency, this pressure is of little use when faced with
the conflicting demands placed on companies by the state, which is invariably the majority
shareholder.
Capital expenditures increase substantially for both domestic and foreign partitions. Sales
growth shows little change although there is a statistically significant (at the 0.05 level) increase in
the mean of the sales growth of domestic companies. The domestic and foreign partitions both show
significant declines in debt ratios.
TABLE III HERE
Companies are divided into three segments based on the percentage of shares held by
individual investors (A-shares and B-, H-, and N-shares). Table IV shows the changes in variables
for the companies with the highest percentage (in top one third) of individual shareholders and the
changes in variables for companies with the lowest percentage (in bottom one third) of individual
shareholders. The middle one third companies are omitted from the analysis. The results show that
the pattern of changes is the same across high individual ownership and low individual ownership
groups. Although there are arguments for expecting individual share ownership to exert more
pressure for improvements in profitability and efficiency, the results show that profitability and
efficiency fell significantly after privatization. Some of the top one third individual ownership
companies are still majority owned by the state and so the influence of individual shareholders may
be very limited.
21
TABLE IV HERE
Legal entity shareholders often own relatively large blocks of shares in a privatized company
and this gives them an incentive to closely monitor management. These shareholders also often have
the manpower and expertise to carry out such activities. The above scenario suggests high legal
entity share ownership will be associated with high profitability companies.
This association
assumes that legal entity shareholders are interested in maximising returns from their investments and
they therefore motivate company management to maximise efficiency and profitability. However,
legal entity shareholders include those that are owned or influenced by the state and those that are
SOEs; these types of legal entity shareholders may want management to pursue goals that conflict
with maximising profitability. It is difficult to disentangle the two types of legal entity shareholders
as disclosure of names and the ultimate owners of these investments is poor.
We partition the data by comparing performance measures across companies where the legal
entity shareholders have relatively high aggregate percentage holdings (in the top one third of our
sample) and where they have relatively low aggregate percentage holdings (in the bottom one third of
our sample). The results are shown in Table V. The changes in performances mirror those in Tables
II, III, and IV. Profitability decreases after privatization for both high and low legal entity ownership.
Asset turnover also declines after privatization implying a reduction in efficiency.
Capital
expenditures increase significantly which is consistent with expectations. Sales growth is limited.
While the mean sales growth increase is statistically significant for the high ownership sample, the
median value shows a decrease. Small sales growth is shown for the low legal entity ownership
sample but the increases are not significant. Finally debt ratios fall for both high and low ownership
groups.
TABLE V HERE
A.
Regression Results
22
Regression equation 1 results are shown in Table VI. The dependent variables are the
changes in variables from before to after privatization. The independent variables include controls
for state of the economy and the market capitalization at the date of listing. Industry and Year
dummy variables are also added as control variables.
The major variables of interest are the
existence of foreign shareholders and the relative ownership stakes of the state and of individuals.
The model also includes a control for the stock exchange on which the shares are listed.
TABLE VI HERE
The model fits are low although similar R-squares are reported in other studies. Change in
asset turnover has the highest R-square at 0.228 while the change in return on assets has the lowest at
0.061. ΔGNP and SIZE are not statistically significant (except for SIZE being significant in the
model of ΔROS). The state of the economy when the IPO was made and the size of the company are
not significant indicators of changes in profitability, asset efficiency, capital expenditures, sales
growth, and debt ratios. Companies that list in Shanghai tend to have higher returns on equity and
lower increases in capital expenditures divided by sales. There is also weak evidence that Shanghai
listed firms have greater decreases in long term debt to shareholders’ equity ratios.
The existence of foreign shareholders is associated with a greater reduction in profitability
when measured as profit divided by sales. The coefficients on FOR are not statistically significant
when modelling changes in return on assets and changes in return on shareholders’ equity. In general,
foreign ownership does not have a major influence on changes in profitability and changes in asset
turnover.
If the foreign shareholders have attempted to influence managers to follow profit
maximising strategies, then their efforts have come to nothing during the periods investigated in this
study. Companies with foreign shareholders have lower capital expenditure to assets increases
although the coefficient (-0.027) is only weakly significant. Larger decreases in debt ratios are
associated with companies that have some foreign ownership.
The level of state ownership has little association with the dependent variables although there
is a weakly negative relationship with sales growth. Individual share ownership has no statistically
significant associations with any of the performance measures save for a negative association in
23
explaining changes in the debt to assets ratio; this implies that higher individual share ownership, the
greater the reduction in debt. Alternative specifications for ownership variables, state (STATE), legal
(LEGAL), and individual (PUBLIC), are used in auxiliary regressions; LEGAL is the percentage of
shares held by legal entities. Using LEGAL in place of PUBLIC in regression equation 1 yields
similar conclusions as regards the importance of ownership variables in explaining changes in
performance.
The variables (STATE - PUBLIC), (STATE - LEGAL), (STATE + LEGAL -
PUBLIC), and (STATE - LEGAL - PUBLIC) are included in regression equation 1, one at a time.
These variables give the relative strength of control by the state over public or legal ownership.
LEGAL is also added to STATE in the third variation as we argue that legal entity shareholders are
often controlled by the state and may therefore be regarded as state shareholders. In general these
additional models yield few significant results for the ownership variables and so they are not
tabulated. For the relatively few significant results, capital expenditure increases are higher if public
ownership is relatively high (vis-à-vis STATE and STATE + LEGAL) and decreases in debt level are
greater if public ownership is high.
The industry and year dummy variables indicate some minor associations with performance
measures. EST, CONG, IND, and COM have negative coefficient signs for returns on sales, have
lower reductions in debt to shareholders’ equity, and have lower increases in capital expenditures.
Sales growth is the lowest for IPOs made in 1995.
B.
Sensitivity Tests
The change in performance measure used in Tables II to VI compare the average performance
in the three years prior to listing to the average performance in the three years after listing. Two
alternative measures are also used. Firstly, the average performance in the three years before listing
are compared to the performance in year t+3. Secondly, the performance in year t-1 is compared to
the performance in year t+3. These alternative measures allow time for the privatized firms to
restructure. The results from these two sensitivity tests are qualitatively similar to those reported in
Tables II to VI. When the new issue proceeds are included in total assets and shareholders’ equity,
the changes in performance (return on assets, return on equity, total asset turnover) are even worse
24
than those reported in Tables II to VI. The sensitivity tests confirm that profitability and efficiency
deteriorate after the privatization and listing of SOEs. 18
C.
Discussion
In stark contrast to arguments that privatization increases economic efficiency and motivates
managers to maximise profitability, our results unequivocally show that return on investment ratios
fall and that asset turnovers decrease. Our evidence is opposite to that found in empirical studies
examining privatizations in both economically ‘developed’ nations and in economically ‘developing’
nations (Megginson et al. (1994), Boubakri and Cosset (1998), D’Souza and Megginson (1999) and
La Porta and Lopez-de-Silanes (1999)).
Various partitionings of the data and cross-sectional
regression models fail to identify reasons for the deterioration in profitability and efficiency. The
existence of foreign ownership of shares, high levels of legal entity share ownership, and high levels
of individual share ownership, do not mitigate the poor performance of privatized companies.
Although these ownership characteristics are hypothesized to be positively related to performance
because of the monitoring activities of the investors, the empirical support is very weak. Corporate
governance structures based on pressure applied on management by investors, do not appear to have
much influence in improving the profitability and efficiency of privatized SOEs. We also examine
the economic performance of 14 privatizations where individual shareholders collectively own more
than 50 percent of the shares. Although the detailed results are not tabulated in the paper, economic
performance deteriorates after privatization. The state has a minority stake in these companies but it
still has significant influence through the appointment of managers and directors.
A reduction in profitability after listing has also been reported in some studies of IPOs (Firth
(1997)). Although Chinese privatizations also raise new capital upon listing, we argue that they are
very different from IPOs in the U.S. and other ‘developed’ countries. IPOs in other countries are
owned by wealth-maximising entrepreneurs and they operate in competitive markets with limited
government influence. These IPOs typically continue the corporate objectives already in place before
the listing.
In sharp contrast, the privatizations of Chinese SOEs are supposed to involve
fundamental changes in business philosophy. We believe our study is more in the ‘privatization of
25
SOEs’ strand of research than the ‘IPOs of private firms’ strand. Nevertheless there are some
similarities in the characteristics of Chinese privatizations and IPOs in ‘developed’ countries.
We argue that China’s styles of privatizations have inhibited the incentives and motivations
associated with free market policies. As one example, the state and its entities generally have voting
control of the company by retaining more than 50 percent of the issued shares. This investment is
usually long term and, indeed, because the shares are not tradable on the SHSE and SZSE, it is
difficult to sell them. Although the government appears wedded to the idea of competition and free
markets, the reality is that social and political factors sway the guidance and influence the state exerts
on companies. The social disruption caused by the privatizations (for example reducing staff levels)
and other parts of the overall economic reforms, may lead the government to influence companies’
staffing levels and staff benefits.
Although privatizations represent a move towards individual
ownership of companies, the fact is that in the majority of cases the state retains voting control. The
objectives of the government will sometimes depart from those of profit maximisation and hence
economic efficiency will be compromised. Legal entity shareholders often have similar goals as the
state.
Typically the senior and junior management of privatized SOEs are the same as before
privatization. Studies in other countries, however, have concluded that changing top management
(retiring or firing incumbents and recruiting new executives) is a crucial factor in the economic
success of privatized SOEs (Barberis et al. (1996), Frydman et al. (1999) and La Porta and Lopez-deSilanes (1999)). This characteristic is absent in the privatization of SOEs in China. Managers are
often political appointees who have little or no experience of running businesses. These managers
are schooled in the old ways of protected state ownership, highly regulated markets, and receiving
and following directions from government or regional ministries. They are ill-equipped to handle the
sudden change to a market economy, competition, and profit objectives. These executives tend to be
ideologically bound to the government 19 and to the systems in place before the move to a market
economy. Senior management in the privatized SOEs tend to have little or no share ownership in
their companies and so they may be less motivated towards achieving shareholder wealth
maximisation.
Although reward practices are not widely disclosed we believe bonus-incentive
schemes based on corporate profitability and-or share price appreciation are rare or are of a small
26
magnitude 20 ; this contrasts with business practices in free market economies where such incentive
systems motivate management toward shareholder objectives. The lack of pay incentive schemes and
performance indicators provides no incentives for managers to maximise profitability.
Pay levels of senior managers in privatized SOEs are comparatively low. According to a
1999 article in the Manager’s Magazine, the average annual total compensation in 1998 (salary,
bonus, housing, medical, social security, and company shares) for a top executive was RMB60,000
(approximately US$7,000). This amount had doubled from the early 1990s. We find that the average
annual compensation of 495 CEOs of listed Chinese companies was RMB50,000 in 1998. There is a
policy that top managers earn no more than two to five times the income of the average employee in
their organization; in some companies senior executives earn less than the average employee. This
pay structure for privatized SOEs provides little incentive for managers and so some executives turn
to illegal means to increase their personal income and wealth. As an example, the Eastern Daily
newspaper (15 March 1999) reported the case of Zhu Shijian the former president of Red Tower
Cigarette Corporation. He turned a small cigarette workshop into the largest cigarette manufacturer
in Asia and the second highest tax payer in China. In 1998 the before tax profit of Red Tower was
RMB20 billion (US$2.3 billion) but Mr. Zhu had an annual salary of just RMB36,000 (about
US$4,300). In order to supplement his income it is alleged that he accepted bribes totalling US$1
million and investigators claimed he had RMB6 million in assets from unknown sources. Zhu
Shijian was sentenced to life imprisonment in January 1999. This is just one case uncovered by the
authorities; there are scores of other reported cases. It is widely believed that there are many other
cases of corruption by managers that go undetected in China and these all detract from the economic
performance of privatized SOEs (Qian (1996)). In addition to direct monetary losses from corrupt
practices, shareholders also suffer because of senior managers’ pre-occupations with designing
schemes to maximise their own utility.
The incumbent senior managements of many privatized companies lack the necessary skills
and experiences to successfully run their firms in the new economic environment. 21 It is difficult to
rectify this problem in the short term as there is a severe lack of motivated managers skilled in
reorganizing SOEs into profitable and efficiently run companies. So even if China has the political
will to retire the heads of SOEs, it will be impossible to replace all of them with highly skilled
27
alternatives. Because of the practice of the state appointing top managers in companies, no free
standing market for managerial labor has developed.
Even if companies are given complete
autonomy to hire senior executives, it will take time before there is a pool of high caliber managers
and it will take time to establish a liquid labor market for managers. Other countries may not face
such difficulties because they have a smaller number of privatizations and because they may have a
larger pool of skilled managers that can be recruited (for example in Britain and Western Europe
there is a large free enterprise sector from which managers can be recruited. Additionally, many
privatizations in (say) Europe have recruited managers from other countries; China may be more
reluctant to recruit globally and they may find only moderate success if they engage in an
international search for executives).
We believe the twin themes of weak corporate governance (accompanied by continuing state
interference) and poorly motivated management (allied with weak incentive systems) has resulted in
the deterioration of profitability and economic efficiency in privatized companies. Our conclusions
are echoed by comments from senior advisers and academics in the PRC. Jinglian Wu, a senior
economic adviser to the Chinese government, comments that corporatized companies ‘have failed to
put in place adequate corporate governance’ (Wu (1999, p.3)). He also states that government
agencies previously supervising the SOE companies ‘continue to intervene with day to day
management even after corporatization and managers’ incentives are inadequate’ (p.3). Lin et al.
(1998) in reviewing the state owned enterprise reforms, are critical of the weak corporate governance
structures and the lack of incentives, which they view as the hallmarks of recent privatizations.
Remedying weak corporate governance and improving managerial incentive systems are the keys to
improving the economic performance of privatized companies.
VI.
CONCLUSIONS
Privatizating SOEs is a major plank of China’s economic reforms. The conversion of SOEs
into profit maximising companies with significant non-government ownership is seen as a way to revitalize industry, enhance technological advancement and growth, and reduce or eliminate subsidies
28
from the state.
Successes in these pursuits will manifest themselves in the form of increased
profitability, improved efficiency, increased capital expenditures, and growth in output.
The operating and financial results from China’s privatizations have been very poor and are
far from the hopes and aspirations of the economic planners. One consequence of the declining
profitability and efficiency is that investors and especially foreign investors may be less inclined to
subscribe to new IPOs in the future. The withdrawal of these investors will severely curtail the
privatization and listing plans of SOEs and thus put added strains on government coffers as it seeks to
modernize China’s industry. Poor economic performance also implies that listed companies will find
it more difficult to raise fresh equity capital when the need arises.
While product markets have become more open and competitive, the reform of SOEs has
been half-hearted. Privatization of SOEs is supposed to be a spur to increased profitability, efficiency,
growth, and investment. We argue that China’s privatizations have not been able to rid enterprizes of
government interference and control. This is because the state in its various guises controls many
privatized companies via shareholding influence and by board and management representation.
Other shareholders such as legal entities and individuals have not so far been forces for disciplining
managers. Legal entities are often under the control or influence of SOEs or the state and so their
interests are similar to those of the state. Individual shareholders have neither the expertise nor the
motivation to actively monitor management. In addition to weak shareholder monitoring, company
managers tend to have little experience of operating in competitive markets and in some cases they
may feel antipathy, if not hostility, to the concepts of shareholder wealth maximisation. These
managers and directors were senior managers of the SOE before privatization.
We contend that in order to realize the full benefits from privatization, the state and its entities
need to sell all of their shares to individuals and non-government affiliated institutional investors.
Government ownership of shares sometimes impedes and at other times inhibits management from
pursuing policies designed to maximise profitability and efficiency. China’s privatization policy
cannot be described as a success because the privatizations typically leave majority ownership in the
hands of the state and the majority of the board of directors are the same as before.
While
privatizations have been said ‘to work’ in almost all other countries (D’Souza and Megginson (1999))
29
they have abjectly failed to work in China. China’s enterprize reforms have not so far led to
improvements in efficiency and profitability.
The poor performance of privatized companies has not gone unnoticed by the Chinese
government and by the national media.
This has led to an intense policy debate within the
government as to the way to proceed. Some leaders have called for a rolling back of the enterprize
reforms and for a strengthening of central planning and government control of organizations and
industries. They contend that privatizations are a failure and are simply not suitable for China. Other
leaders, however, argue that the poor performance of companies is because the privatizations are
half-hearted and that the state needs to give complete autonomy to the firms in order for the reforms
to work. The outcome of this policy debate will help determine the prosperity of China as it enters
the new millenium. 22 Hopefully, our results will help inform the current on-going debate and will
serve as counterweight to those who wish to revert to a centrally planned economy.
30
REFERENCES
Atkinson, S.E. and R. Halvorsen, 1986, The relative efficiency of public and private firms in a
regulated environment: The case of U.S. electric utilities, Journal of Public Economics 29,
281-294.
Barberis, N., M. Boycko, A. Shleifer, and N. Tsukanova, 1996, How does privatization work?
Evidence from the Russian shops, Journal of Political Economy 104, 764-790.
Bishop, M. and J. Kay, 1989, Privatization in the United Kingdom: Lessons and experience, World
Development 17, 643-657.
Boardman, A.E. and A.R. Vining, 1989, Ownership and performance in competitive environments: A
comparison of the performance of private, mixed, and state-owned enterprises, Journal of
Law and Economics 32, 1-33.
Boubakri, N. and J-C. Cosset, 1998, The financial and operating performance of newly privatized
firms: Evidence from developing countries, Journal of Finance 53, 1081-1110.
Boycko, M., A. Shleifer, and R. Vishny, 1995, Privatizing Russia (MIT Press, Cambridge MA).
Boycko, M., A. Shleifer, and R.W. Vishny, 1996, A theory of privatization, The Economic Journal
106, 309-319.
Brandt, L. and X. Zhu, 2000, Redistribution in a decentralized economy: Growth and inflation in
China under reform, Journal of Political Economy 108, 422-439.
Cao, Y, Y. Qian, and B.R. Weingast, 1999, From federalism, Chinese style, to privatization, Chinese
style, Working paper, Stanford University.
Caves, R.E., 1990, Lessons from privatization in Britain: State enterprise behaviour, public choice,
and corporate governance, Journal of Economic Behavior and Organization 13, 145-169.
Caves, R.E. and L.R. Christensen, 1980, The relative efficiency of public and private firms in a
competitive environment: The case of Canadian railroads, Journal of Political Economy 88,
958-976.
Che, J. and Y. Qian, 1998, Insecure property rights and government ownership of firms, Quarterly
Journal of Economics 113, 467-496.
Claessens, S., S. Djankov, and G. Pohl, 1997, Ownership and corporate governance: Evidence from
the Czech Republic, Working paper, The World Bank, Washington DC.
Dewenter, K.L. and P.H. Malatesta, 2000, State-owned and privately-owned firms: An empirical
analysis of profitability, leverage, and labor intensity, Working paper, University of
Washington.
31
D’Souza, J. and W.L. Megginson, 1999, The financial and operating performance of privatized firms
during the 1990s, Journal of Finance 54, 1397-1438.
Du, H.W., N.W. Hu, and R.R. Yang, 1998, Research report on the Chinese economy: The state
enterprise reform in the transition to a market economy (Renmin University Press, Beijing).
Dyck, I.J.A., 1997, Privatization in Eastern Germany: Management selection and economic transition,
American Economic Review 87, 565-597.
Ehrlich, I., G. Gallais-Hamonno, Z. Liu, and R. Lutter, 1994, Productivity growth and firm
ownership: An analytical and empirical investigation, Journal of Political Economy 102,
1006-1038.
Firth, M., 1997, An analysis of the stock market performance of new issues in New Zealand, PacificBasin Finance Journal 5, 63-85.
Frydman, R., C. Gray, M. Hessel, and A. Rapaczynski, 1999, When does privatization work? The
impact of private ownership on corporate performance in the transition economies. Quarterly
Journal of Economics 114, 1153-1191.
Galal A., L. Jones, P. Tandon, and I. Volgelsang, 1994, Welfare Consequences of Selling Public
Enterprises (The World Bank, Washington D.C.).
Gao, S., 1996., China’s Economic Reform (Macmillan Press, London).
Groves, T., Y. Hong, J. McMillan, and B. Naughton, 1994, Autonomy and incentives in Chinese
state enterprises, Quarterly Journal of Economics 109, 181-209.
Groves, T., Y. Hong, J. McMillan, and B. Naughton, 1995, China’s evolving managerial labor market,
Journal of Political Economy 103, 873-892.
Hu, X., 1997., Reducing state-owned enterprises’ social burdens and establishing a social insurance
system, in H.G. Broadman, ed.: Policy Options for the Reform of Chinese State-Owned
Enterprises (The World Bank, Washington D.C.), 125-148.
Jefferson, G.H., 1998, China’s state enterprises: Public goods, externalities, and Coase, American
Economic Review 88, 428-432.
Jefferson, G.H. and T.G. Rawski, 1994, Enterprise reform in Chinese industry, Journal of Economic
Perspectives 8, 47-70.
Jin, H. and Y. Qian, 1998, Public versus private ownership of firms: Evidence from rural China,
Quarterly Journal of Economics 113, 773-808.
Jones, S.L., W.L. Megginson, R.C. Nash, and J.M. Netter, 1999, Share issue privatizations as
financial means to political and economic ends, Journal of Financial Economics 53, 217-253.
32
Kay, J. and D.J. Thompson, 1986, Privatization: A policy in search of a rationale, The Economic
Journal 96, 18-32.
Kikeri, S., J. Nellis, and M. Shirley, 1992, Privatization: The Lesson of Experience (The World Bank,
Washington D.C.).
Krueger, A.O., 1990, Government failures in development, Journal of Economic Perspectives 4, 923.
La Porta, R. and F. Lopez-de-Silanes, 1999, The benefits of privatization: Evidence from Mexico,
Quarterly Journal of Economics 114, 1193-1242.
Lau, L.J., Y. Qian, and G. Roland, 2000, Reform without losers: An interpretation of China’s dualtrack approach to transition, Journal of Political Economy 108, 120-143.
Li, W., 1997, The impact of economic reform on the performance of Chinese state enterprises, 19801989, Journal of Political Economy 105, 1080-1106.
Lin, J.Y., F. Cai, and Z. Li, 1998, Competition, policy burdens, and state-owned enterprise reform,
American Economic Review 88, 422-427.
Martin, S. and D. Parker, 1995, Privatization and economic performance throughout the UK business
cycle, Managerial and Decision Economics 16, 225-237.
Megginson, W.L., R.C. Nash, and M. Van Randenborgh, 1994, The financial and operating
performance of newly privatized firms: An international empirical analysis, Journal of
Finance 49, 403-452.
Megginson, W.L. and J.M. Netter, 1998, From state to market: A survey of empirical studies on
privatization, Working paper, NYSE and SBF Bourse de Paris.
Menyah, K. and K. Paudyal, 1996, Share issue privatization: The UK experience, in M. Levis, ed.:
Empirical Issues in Raising Equity Capital (Elsevier Science, Amsterdam).
Perotti, E., 1995, Credible privatization, American Economic Review 85, 847-859.
Pohl, G., R.E. Anderson, S. Claessens, and S. Djankov, 1997, Privatization and restructuring in
Central and Eastern Europe: Evidence and policy options, Working paper, The World Bank,
Washington DC.
Poon, W., M. Firth, and H-G. Fung, 1998, Asset pricing in segmented capital markets: Preliminary
evidence from China-domiciled companies, Pacific-Basin Finance Journal 6, 307-319.
Pryke, R., 1982, The comparative performance of public and private enterprises, Fiscal Studies 3, 6881.
33
Qi, D., W. Wu, and H. Zhang, 1999, Shareholding structure and corporate performance of partially
privatized firms: Evidence from listed Chinese companies, Working paper, Chinese
University of Hong Kong.
Qian, Y., 1996, Enterprise reform in China: Agency problems and political control, Economics of
Transition 4 (2), 427-447.
Ramanadham, V.V. (ed.), 1993, Privatization: A Global Perspective. (Routledge, London).
Rawski, T. G., 1994, Chinese industrial reform: Accomplishments, prospects, and implications,
American Economic Review 84, 271-275.
Shirley, M. and J. Nellis, 1991, Public Enterprise Reform: The Lessons of Experience (The World
Bank, Washington D.C.).
Shleifer, A, 1999, State versus private ownership, Working paper, Harvard University.
Tang, Y.L., L. Chow, and B. Cooper, 1996, Accounting and Finance in China: A Review of Current
Practice, Third edition (Pearson, Hong Kong).
Thompson, D.J., 1987, Privatization in the UK, European Economic Review 31, 368-374.
Vickers, J. and G. Yarrow, 1988, Regulation of privatized firms in Britain, European Economic
Review 32, 465-472.
Vickers, J. and G. Yarrow, 1991, Economic perspectives on privatization, Journal of Economic
Perspectives 5, 111-132.
Vining, A.R. and A.E. Boardman, 1992, Ownership versus competition: Efficiency in public
enterprise, Public Choice 73, 205-239.
Wortzel, H.V. and L.H. Wortzel, 1989, Privatization: Not the only answer, World Development 17,
633-641.
Wu, J., 1999., Enterprise reform and development in Mainland China, Paper presented at the Hong
Kong Securities Institute.
Xiang, B., 1998, Institutional factors influencing China’s accounting reforms and standards,
Accounting Horizons 22, 105-119.
Xu, X. and Y. Wang, 1998, Ownership structure, corporate governance, and corporate performance:
The case of Chinese stock companies, Working paper, The World Bank, Washington DC.
Yarrow, G., 1986, Privatization in theory and practice, Economic Policy 2, 324-364.
34
TABLE I
Data Sample: IPO Listings by Year and by Exchange
IPOs listed
Shanghai Stock Exchange
Shenzhen Stock Exchange
Total
1991
4
8
12
1992
65
18
83
1993
59
58
117
35
1994
23
19
42
1995
8
13
21
Total
159
116
275
TABLE II
Changes in Economic Performance from before Privatization to after Privatization
Performance
Measure
Mean
(Median)
Before
Privatization
Profitability
Return on sales
Mean (Median)
After
Privatization
Predicted
Change
Mean
(Median)
Change
Percentage of
Privatizations
that improved
performance
0.1230
(0.0879)
0.1237
(0.0909)
+
+
0.0007
(0.0030)
52.01
Return on assets
0.0848
(0.0763)
0.0643
(0.0593)
+
+
-0.0205**
(-0.0170)**
16.61**
Return on equity
0.2213
(0.1863)
0.1315
(0.1417)
+
+
-0.0898**
(-0.0446)**
32.29**
1.0368
(0.7724)
0.7090
(0.5403)
+
+
-0.3278**
(-0.2321)**
23.77**
Investment
Capital expenditures
to sales
0.1316
(0.0321)
0.4848
(0.1510)
+
+
0.3532**
(0.1188)**
83.19**
Capital expenditures
to total assets
0.0564
(0.0307)
0.0923
(0.0765)
+
+
0.0360**
(0.0458)**
0.4723
(0.3027)
0.5198
(0.3162)
+
+
0.0475*
(0.0135)
52.06
0.5551
(0.5904)
0.4554
(0.4513)
-
-0.0997**
(-0.1391)**
74.81**
0.3247
(0.1525)
0.1541
(0.0743)
-
-0.1706**
(-0.0781)**
63.50**
Activity
Total asset turnover
Output
Real sales (billion RMB)
Leverage
Debt to assets
Long term debt
to equity
72.89**
Statistical significance based on t-tests (mean differences), Wilcoxon tests (median differences) and
sign tests (percentage change). ** statistically significant at the 0.01 level; * statistically significant
at the 0.05 level
Before statistics are based on the average over three years prior to privatization and After statistics
are based on the average over three years after privatization. The year of the privatization is ignored.
36
TABLE III
Changes in Economic Performance from before Privatization to after Privatization, Partitioned on the
basis of the presence of a Foreign Shareholding
Foreign/
Performance
Non-Foreign
Measure
Foreign
Mean (Median)
Before
Privatization
Mean (Median)
After
Privatization
Predicted
Change
0.1384
(0.0957)
0.1090
(0.0986)
+
+
-0.0294
(0.0030)
45.15
0.1179
(0.0821)
0.1287
(0.0859)
+
+
0.0108
(0.0038)
54.30
Return on sales
Non-Foreign
t-statistic on difference (F and NF)
Foreign
Return on assets
1.012
-0.475
Mean (Median)
Change
Percentage
of
Privatizations
that improve
performance
-1.204
0.0861
(0.0732)
0.0434
(0.0437)
+
+
-0.0428**
(-0.0295)**
10.44**
Non-Foreign
0.0869
(0.0771)
0.0449
(0.0495)
+
+
-0.0419**
(-0.0276)**
23.58**
t-statistic on difference (F and NF)
-0.524
-0.192
Foreign
0.2801
(0.2350)
0.0901
(0.0902)
+
+
-0.1900**
(-0.1448)**
11.86**
0.2002
(0.1805)
0.0671
(0.1221)
+
+
-0.1331**
(-0.0584)**
30.00**
2.801**
0.372
0.8250
(0.7310)
0.5216
(0.4775)
+
+
-0.3034**
(-0.2535)**
13.20**
Non-Foreign
1.0886
(0.7841)
0.6734
(0.5267)
+
+
-0.4152**
(-0.2574)**
21.83**
t-statistic on difference (F and NF)
-3.135**
-2.496*
Return on equity
Non-Foreign
t-statistic on difference (F and NF)
Foreign
Foreign
Total asset turnover
-0.103
-1.629
1.325
Capital expenditures 0.1517
to sales
(0.0326)
0.2274
(0.1622)
+
+
0.0756*
(0.1296)**
79.36**
0.1245
(0.0318)
0.5744
(0.1458)
+
+
0.4498**
(0.1139)**
84.54**
Non-Foreign
t-statistic on difference (F and NF)
2.623**
-0.790
37
-2.930**
TABLE III (Cont’d)
Foreign/
Performance
Non-Foreign
Measure
Foreign
Mean (Median)
Before
Privatization
Mean (Median)
After
Privatization
Predicted
Change
Capital expenditures 0.0589
to total assets
(0.0361)
0.0823
(0.0717)
+
+
0.0234**
(0.0356)**
71.23**
0.0554
(0.0302)
0.0960
(0.0822)
+
+
0.0406**
(0.0520)**
73.50**
1.991*
-2.010*
0.6811
(0.3494)
0.7160
(0.3111)
+
+
0.4019
(0.2933)
0.4537
(0.3212)
+
+
3.392**
0.456
0.6068
(0.6582)
0.4613
(0.4609)
-
-0.1455**
(-0.1973)**
89.04**
0.5363
(0.5717)
0.4532
(0.4479)
-
-0.0831**
-0.1238**
69.65**
1.488
0.568
0.4088
(0.2292)
0.1429
(0.0797)
-
-0.2658**
(-0.1495)**
75.34**
0.2941
(0.1220)
0.1582
(0.0743)
-
-0.1360**
(-0.0477)**
59.20**
Non-Foreign
t-statistic on difference (F and NF)
Foreign
Real Sales
(billion RMB)
Non-Foreign
t-statistic on difference (F and NF)
Foreign
Debt to assets
Non-Foreign
t-statistic on difference (F and NF)
Foreign
Long term debt
to equity
Non-Foreign
t-statistic on difference (F and NF)
3.356**
-1.452
Mean (Median)
Change
Percentage
of
Privatizations
that improve
performance
-0.269
0.0348
(-0.0384)
0.0518**
(0.0279)
47.54
53.91
-3.062**
-1.839
-3.185**
Statistical significance based on t-tests (mean differences), Wilcoxon tests (median differences) and sign
tests (percentage change). t-statistic on difference between Foreign invested SOE (F) and non-Foreign
invested SOE (NF). ** statistically significant at the 0.01 level; * statistically significant at the 0.05 level
Before statistics are based on the average over three years prior to privatization and After statistics are based
on the average over three years after privatization. The year of the privatization is ignored.
38
TABLE IV
Changes in Economic Performance from before Privatization to after Privatization; Partitioned on the
basis of High and Low Public Ownership of Shares
High/Low
Public
Ownership
Performance
Measure
Mean (Median)
Before
Privatization
Mean (Median)
After
Privatization
Predicted
Change
Mean (Median)
Change
Percentage
of
Privatizations
that improve
performance
High
Return on sales
0.1100
(0.0839)
0.0902
(0.0848)
+
+
-0.0199
(0.0010)
53.68
Low
0.1307
(0.0900)
0.1578
(0.1158)
+
+
0.0271
(0.0258)
47.43
t-statistic on difference (H and L)
-1.302
-1.345
High
0.0869
(0.0760)
0.0455
(0.0604)
+
+
-0.0413**
(-0.0156)**
25.30**
Low
0.0889
(0.0795)
0.0463
(0.0430)
+
+
-0.0426**
(-0.0365)**
11.32**
t-statistic on difference (H and L)
-0.140
High
0.1905
(0.1777)
0.0485
(0.1185)
+
+
-0.1420**
(-0.0592)**
29.26**
Low
0.2447
(0.2199)
0.0934
(0.1118)
+
+
-0.1514**
(-0.1081)**
16.27**
t-statistic on difference (H and L)
-2.672**
-0.824
High
1.0633
(0.7727)
0.6457
(0.4894)
+
+
-0.4175**
(-0.2833)**
17.94**
0.8659
(0.7384)
0.6492
(0.5102)
+
+
-0.2167**
(-0.2282)**
24.69**
1.055
0.326
0.0843
(0.0295)
0.2343
(0.1073)
+
+
0.1500**
(0.0779)**
84.88**
Low
0.1515
(0.0331)
0.3067
(0.2221)
+
+
0.1552**
(0.1890)**
82.27**
t-statistic on difference (H and L)
-0.048
-1.801
Return on assets
Return on equity
Total asset turnover
Low
t-statistic on difference (H and L)
High
Capital expenditures
to sales
0.806
0.042
39
0.261
0.341
-1.842
-0.188
TABLE IV (Cont’d)
High/Low
Public
Ownership
Performance
Measure
Mean (Median)
After
Privatization
Predicted
Change
Mean (Median)
Change
Percentage
of
Privatizations
that improve
performance
High
Capital expenditures
to total assets
0.0461
(0.0274)
0.0781
(0.0612)
+
+
0.0320**
(0.0338)**
71.73**
Low
0.0683
(0.0367)
0.1131
(0.0972)
+
+
0.0448**
(0.0605)**
71.57**
t-statistic on difference (H and L)
-0.074
-0.334
High
0.4399
(0.2960)
0.4749
(0.3550)
+
+
0.0351
(0.0590)
51.85
0.4255
(0.2485)
0.5007
(0.2974)
+
+
0.0751*
(0.0489)*
54.02
Real sales
(Billion RMB)
Low
t-statistic on difference (H and L)
High
Debt to assets
Low
t-statistic on difference (H and L)
High
Long term debt
to equity
Mean (Median)
Before
Privatization
0.040
-0.844
-1.366
-1.270
0.5539
(0.5863)
0.4389
(0.4284)
-
-0.1150**
(-0.1579)**
77.71**
0.5501
(0.6089)
0.4679
(0.4580)
-
-0.0821**
(-0.1509)**
68.42**
0.380
-0.738
-0.643
0.3069
(0.1627)
0.1486
(0.0718)
-
-0.1584**
(-0.0909)**
64.51**
Low
0.3386
(0.1594)
0.1265
(0.0831)
-
-0.2121**
(-0.0763)**
64.21**
t-statistic on difference (H and L)
-1.847
1.529
-1.400
Statistical significance based on t-tests (mean differences), Wilcoxon tests (median differences) and sign
tests (percentage change). t-statistic on difference between High (H) and Low (L) Public Ownership. **
statistically significant at the 0.01 level; * statistically significant at the 0.05 level
Before statistics are based on the average over three years prior to privatization and After statistics are based
on the average over three years after privatization. The year of the privatization is ignored.
40
TABLE V
Changes in Economic Performance from before the Privatization to after Privatization, Partitioned on
the basis of High and Low Legal Entity Ownership of Shares
High/Low
Performance
Legal Entity Measure
Ownership
High
Mean (Median)
Before
Privatization
Mean (Median)
After
Privatization
Predicted
Change
0.1562
(0.1206)
0.1360
(0.1300)
+
+
-0.0202
(0.0094)
50.85
0.1081
(0.0804)
0.1209
(0.0692)
+
+
0.0128
(-0.0112)
45.12
2.624**
2.466**
0.1054
(0.0852)
0.0565
(0.0532)
+
+
-0.0489**
(-0.0320)**
16.66**
0.0759
(0.0607)
0.0393
(0.0336)
+
+
-0.0367**
(-0.0271)**
20.43**
2.971**
1.889**
0.2545
(0.2310)
0.1178
(0.1292)
+
+
-0.1366**
(-0.1018)**
25.33**
0.2093
(0.1805)
0.0772
(0.1060)
+
+
-0.1320**
(-0.0745)**
20.48**
2.127*
0.217
0.9227
(0.7161)
0.5829
(0.4965)
+
+
-0.3397**
(-0.2196)**
20.27**
0.9785
(0.8089)
0.6688
(0.5587)
+
+
-0.3097**
(-0.2502)**
21.51**
Return on sales
Low
t-statistic on difference (H and L)
High
Return on assets
Low
t-statistic on difference (H and L)
High
Return on equity
Low
t-statistic on difference (H and L)
High
Total asset turnover
Low
t-statistic on difference (H and L)
0.143
Mean (Median)
Change
Percentage
of
Privatizations
that improve
performance
-0.611
-1.116
0.058
-1.239
-0.315
High
Capital expenditures 0.1800
to sales
(0.0301)
0.9540
(0.1539)
+
+
0.7740**
(0.1239)**
82.92**
Low
0.1087
(0.0321)
0.2189
(0.1394)
+
+
0.1102**
(0.1073)**
86.29**
0.136
1.552
t-statistic on difference (H and L)
41
0.440
TABLE V (Cont’d)
High/Low
Performance
Legal Entity Measure
Ownership
High
Capital expenditure
to total assets
Low
t-statistic on difference (H and L)
High
Mean (Median)
Before
Privatization
Real sales
(Billion RMB)
Mean (Median)
After
Privatization
Predicted
Change
Mean (Median)
Change
Percentage
of
Privatizations
that improve
performance
0.0610
(0.0306)
0.0942
(0.0796)
+
+
0.0332**
(0.0489)**
76.92**
0.0526
(0.0298)
0.0962
(0.0805)
+
+
0.0437**
(0.0507)**
73.62**
0.137
-0.177
-0.983
0.4192
(0.3113)
0.5250
(0.2993)
+
+
0.1057**
(-0.0119)
53.65
Low
0.4331
(0.2601)
0.4336
(0.3135)
+
+
0.0005
(0.0534)
54.43
t-statistic on difference (H and L)
-0.131
High
0.5081
(0.5332)
0.4391
(0.4559)
-
-0.0690**
(-0.0773)**
67.03**
Low
0.6113
(0.6319)
0.4789
(0.4579)
-
(-0.1324)**
(-0.1739)**
84.78**
t-statistic on difference (H and L)
-1.813
-1.748
High
0.2630
(0.1044)
0.0986
(0.0603)
-
-0.1644**
(-0.0441)**
59.34**
Low
0.4289
(0.2240)
0.1827
(0.0831)
-
-0.2461**
(-0.1410)**
76.08**
t-statistic on difference (H and L)
-0.660
-0.340
Debt to assets
Long term debt
to equity
1.849
2.502**
1.096
0.853
Statistical significance based on t-tests (mean differences), Wilcoxon tests (median differences) and sign
tests (percentage change). t-statistic on difference between High (H) and Low (L) Legal Entity Ownership.
** statistically significant at the 0.01 level; * statistically significant at the 0.05 level
Before statistics are based on the average over three years prior to privatization and After statistics are
based on the average over three years after privatization. The year of the privatization is ignored.
42
42
43
44
TABLE VI
Regression Results
Performance Measure
Independent
Variable
Δ Return
on sales
Δ Return
on assets
Δ Return
on equity
Intercept
(ΔROS)
1.722
(ΔROA)
-0.118
(ΔROE)
-0.530
(-0.222)
0.176
(0.324)
0.004
(0.333)
0.016
(0.894)
-0.010
(-0.241)
-0.018
(-0.940)
-0.002
(0.214)
-0.015
(-0.710)
1.652
(0.807)
-0.012
(-0.737)
0.001
(0.011)
(-0.265)
0.137
(0.067)
-0.068
(-1.381)
0.002
(0.030)
-0.004
(-0.025)
-0.024
(-0.486)
0.084
(2.159)**
-0.205
(-2.368)**
-0.043
(-0.551)
-0.116
(-0.786)
-0.073
(-0.957)
ΔGNP
FOR
STATE
PUBLIC
SIZE
EX
EST
CONG
IND
COM
(-0.785)
1.947
(0.819)
-0.104
(-2.029)**
0.087
(1.158)
-0.106
(-0.632)
0.039
(4.784)***
0.007
(0.178)
-0.309
(-3.316)***
-0.162
(-1.872)*
-0.204
(-2.981)***
-0.134
(-1.655)*
Δ Capital
expenditures
to assets
(ΔEXP/TA)
-0.038
Δ Capital
expenditures
to sales
(ΔEXP/S)
-27.663
(-1.024)
30.865
(1.055)
-0.466
(-0.731)
-0.638
(-0.681)
-0.589
(-0.284)
0.117
(1.168)
-1.052
(-2.084)**
-0.348
(-0.295)
2.226
(2.089)**
0.289
(0.334)
0.405
(0.399)
(-0.057)
0.096
(0.133)
-0.027
(-1.774)*
0.034
(1.530)
-0.050
(-1.000)
0.002
(1.007)
-0.003
(-0.211)
-0.079
(-2.980)***
-0.058
(-2.315)**
-0.044
(-2.165)**
-0.029
(-1.191)
43
Δ Real
sales
Δ Debt to
assets
(GRO)
-0.294
(ΔD/TA)
-0.217
(-0.109)
0.166
(0.057)
0.013
(0.202)
-0.154
(-1.652)*
-0.201
(-0.957)
-0.010
(-0.870)
-0.025
(-0.486)
0.099
(0.882)
0.196
(1.807)*
0.095
(1.112)
0.126
(1.223)
(-0.181)
0.201
(0.155)
-0.101
(-3.714)***
-0.033
(-0.815)
-0.229
(-2.556)***
0.002
(0.574)
-0.001
(-0.024)
-0.006
(-0.125)
-0.002
(-0.033)
-0.011
(-0.306)
-0.016
(-0.373)
Δ Long-term
debt to
equity
(ΔD/E)
0.197
Δ Total
asset
turnover
(ΔASTURN)
-0.619
(0.052)
-0.475
(-0.116)
-0.164
(-1.912)*
0.010
(0.083)
-0.096
(-0.338)
-0.003
(-0.197)
-0.115
(-1.661)*
0.546
(3.649)***
0.377
(2.651)***
0.346
(3.029)***
0.410
(3.004)***
(-1.142)
6.330
(1.146)
0.124
(0.897)
0.082
(0.419)
-0.122
(-0.265)
-0.025
(-0.183)
0.058
(0.539)
0.145
(0.599)
-0.248
(-1.126)
-0.044
(-0.245)
-0.993
(-4.673)***
TABLE VI (Cont’d)
Independent
Variable
Y91
Y92
Y93
Y94
2
R
Δ Return
on sales
Δ Return
on assets
Δ Return
on equity
(ΔROS)
0.153
(ΔROA)
-0.012
(ΔROE)
0.176
Δ Capital
expenditures
to sales
(ΔEXP/S)
-0.436
(1.410)
(-0.145)
(0.911)
(-0.306)
0.127
-0.019
0.176
0.480
(1.648)
(-0.472)
(1.038)
(0.458)
0.015
0.009
0.132
-0.588
(0.197)
(-0.247)
(0.799)
(-0.567)
0.055
0.034
0.064
-0.419
(0.687)
(0.879)
(0.391)
(-0.381)
0.141
0.061
0.081
0.072
Δ Capital
expenditures to
assets
(ΔEXP/TA)
(0.104)
(3.047)***
0.048
(1.903)*
0.033
(1.315)
0.012
(0.447)
0.107
(ΔD/TA)
0.096
Δ Long-term
debt to
equity
(ΔD/E)
-0.067
Δ Total
asset
turnover
(ΔASTURN)
0.213
(1.595)
(-0.350)
(0.247)
0.084
-0.056
-0.090
(1.896)*
(-0.399)
(-0.200)
0.024
-0.155
0.325
(0.556)
(-1.126)
(0.730)
-0.006
-0.303
0.455
Δ Real
sales
Δ Debt to
assets
(ΔGRO)
0.313
(2.333)**
0.289
(3.019)***
0.184
(1.965)**
0.267
(2.648)***
0.089
(-0.135)
0.121
(-2.055)**
0.105
(1.022)
0.228
t-statistics in parentheses. *** statistically significant at the 0.01 level. ** statistically significant at the 0.05 level. * statistically
significant at the 0.10 level.
ΔGNP = percentage change in gross national product per capita from the three years prior to the privatization to the three years after
privatization. The GNP numbers are deflated for changes in inflation; FOR = a dummy variable taking the value one (1) if there is
foreign ownership in the company; otherwise FOR is coded zero (0); STATE = percentage of shares held by the state after the
privatization; PUBLIC = percentage of shares held by individual investors (A-shares and B-shares); SIZE = log of market
capitalization of the company after listing; EX = a dummy variable taking the value one (1) if the company is listed on the Shanghai
Securities Exchange; otherwise EX is coded zero (0); EST = a dummy variable taking the value one (1) if the company is in the
property sector; otherwise EST is coded zero (0); CONG = a dummy variable taking the value one (1) if the company is in a number
of industries (and is included in the Conglomerate or Composite index); otherwise COMP is coded zero (0); IND = a dummy variable
taking the value one (1) if the company is an industrial or manufacturing firm; otherwise IND is coded zero (0); COM = a dummy
variable taking the value one (1) if the company is in the retailing and commercial sector; otherwise COM is coded zero (0); and Y91,
Y92, Y93, Y94 are dummy variables representing the year of listing (1991, 1992, 1993, or 1994).
44
1
Note, however, that some empirical studies argue that privatized utilities, which operate in monopolistic
markets, significantly improve economic performance after the IPO.
2
In contrast, Martin and Parker (1995) and Caves and Christensen (1980) found no evidence to suggest private
companies out-performed government owned enterprizes in their studies using, respectively, British and
Canadian data.
3
There are, however, a number of studies that have examined China’s economic reforms at the macro level
and at the township-village enterprize level. See, for example, Che and Qian (1998), Jefferson (1998), Jefferson
and Rawski (1994), Jin and Qian (1998), Lau et al. (2000), and Brandt and Zhu (2000).
4
Hu (1997) reviews the social burdens of SOEs and discusses ways to ameliorate these costs.
5
In contrast, Li (1997) reports that the economic performance of SOEs improved significantly during the
1980s. This is based on a sample of 272 state enterprizes.
6
Privatizations in developed nations often involve the government selling its shares in the SOE and no new
cash is injected into the company. Unlike some privatizations in Russia and Eastern Europe, IPO shares in
China are issued for cash (that is there are no ‘free’ shares or share vouchers given out). The high rate of
personal savings by households in China gives people the wherewithal to invest in shares.
7
Foreign shares universally have lower market prices than the domestic shares. In many cases the A-shares
cost double or more the price of B-shares. The wide discrepancy in prices and the observation that foreign
shares are cheaper is opposite to evidence from other markets that have domestic and foreign shares.
8
There are about 300,000 SOEs in China (Cao et al. (1999)); about 15,000 of these are classified as large size
SOEs by the China Statistical Yearbook.
9
10
So far, very few companies have listed their legal entity shares on STAQ.
Privatized SOEs still have some social costs in their cost structures.
11
These and other corporate shareholders may use their influence to transfer wealth to themselves from the
privatized firm. This may be accomplished via setting non-market based transfer prices.
12
Chinese accounting standards are used. Although the amount of accounting disclosures have increased over
the period studied, the rules for computing earnings have been stable (Tang et al. (1996) and Xiang (1998)).
13
For example, the proceeds from a new issue of shares may be invested in new plant and machinery. Sales
and profits from the products produced by the new plant and machinery may take three or more years to occur.
While total assets and shareholders’ equity increase, there is no change in sales or profits in the first few years.
Performance and efficiency measures therefore fall.
14
This adjustment will create an upward bias in performance changes for those companies that invest new
issue proceeds very quickly and are able to reap sales and profits from the investments in the first few years.
15
However, some foreign investors may view their China investments as a learning experience and so they
may be ready to accept low returns on their initial investments.
16
IPOs made in the 1980s did not list until at least 1990 because the first stock exchange was not opened until
December 1990. If the IPO and its listing are more than one year apart, it becomes difficult to decide which is
year t, the year of privatization.
17
The state has a total quota for bond issuance and they give preference to non-listed companies. Up to 1998
only one listed company had been approved to issue bonds (Du et al. (1998)). Bank borrowing is available but
is subject to political considerations as well as commercial judgments. Most additional external financing
comes from equity rights issues.
45
18
Additional sensitivity tests examine whether the profitability results are caused by opportunistic earnings
management. Here, profits may be increased in year t-1 by the use of discretionary accounting accruals which,
by their very nature, reverse in the following year. This has the effect of ‘over-stating’ profits in year t-1 (or
year t) and ‘understating’ profits in year t (or year t+1). By ‘over-stating’ profits in year t-1 (or the forecast of
profits in year t) the issue price of the IPO may be increased. It is extremely difficult to estimate such earnings
management and this is especially so in China. Our sensitivity tests involve omitting years t-1 and t+1 from the
analyses as these are the years when the effects of IPO earnings management may be more prevalent.
Comparing profitability measures from year t-3 to years t+2 and t+3 and comparing years t-3 and t-2 to years
t+2 and t+3 yield results that are directionally the same as those reported in Tables II to VI (and the results are
statistically significant). These sensitivity tests corroborate our findings that profitability measures deteriorate
after privatization.
19
Many senior executives of SOEs are appointed, in part, because of connections to the government and to the
Communist Party.
20
For some counter evidence see Groves et al. (1994, 1995).
21
Dyck (1997) finds that replacing incumbent management is a key to successful privatization of SOEs in the
former East Germany.
22
The recently concluded annual plenary session of the Communist Party Central Committee (September 1999)
decided to slow down the privatization process. An official communique indicated that there will be fewer new
privatizations in the coming year, that Communist Party committees inside companies should play a more active
role in directing the affairs of SOEs, that subsidies will be paid to SOEs, and that more help will be given to
bankrupt state enterprizes. Interestingly, the communique also stated that there should be less government
interference in the running of existing privatized companies and that managerial pay and performance incentives
will be increased.
46