Content uploaded by BinBin J. Pearce
Author content
All content in this area was uploaded by BinBin J. Pearce on Jul 26, 2017
Content may be subject to copyright.
379
1 Introduction
National oil companies (NOCs) are pivotal to the global oil mar-
ket. They control nearly 90 percent of the world’s crude oil reserves
and account for two-thirds of the world’s crude oil production (EIA
2009b ). China’s NOCs are especially important due to the role the
country plays as a major consumer of oil, but also as an increasingly
visible investor in oil producing projects around the world. A previ-
ous paper by the Baker Institute at Rice University (S. Lewis 2007 )
analyzed the Chinese oil industry as a whole, and others (Jakobson
and Zha 2006 ; Ma and Andrews-Speed 2006 ; Taylor 2006 ; Downs
2007a , 2007b ; Xu 2007 ) have focused on the overseas activities of
the Chinese NOCs. The following study, however, will concentrate
on the largest and most important of China’s NOCs: China National
Petroleum Corporation (CNPC). CNPC has been the central element
of the Chinese government’s efforts to manage its growing depend-
ence on oil and natural gas, which are essential to China’s economy
yet China has neither fuel in abundance.
CNPC is often portrayed in extremes. While many Western obser-
vers see it as a puppet of the state that scheme s to hoard energy resource s
abroad, the Chinese government has been persistent in characterizing
the rm as a transparent and internationalized oil company. The real
picture of CNPC is, of course, much more complicated. Its relation-
ship with the state, its nancial performance, and the non- nancial
9 China National Petroleum
Corporation (CNPC): a balancing act
between enterprise and government
I would like to acknowledge the support of David Victor, Mark Thurber, and
Kathy Lung from the Program on Energy and Sustainable Development at
Stanford University for the completion of this work. In addition, Yuan Ming
Dong of the Development Research Center of the State Council provided much
of the background information on CNPC. Xiaojie Xu, formerly of CNPC ,
and Mikkal Herberg, of UCSD, provided valuable comments and insights for
improving this report.
, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511784057.012
Downloaded from https://www.cambridge.org/core. ETH-Bibliothek, on 26 Jul 2017 at 13:20:34, subject to the Cambridge Core terms of use
BinBin Jiang380
aspects of its operations are the result of more than ve decades of
political upheaval, social change, and economic reforms since the
ascent of one-party rule in China. As a result, CNPC is an entity of
contradictions and inconsistencies. It is a victim of China’s incom-
plete transition between a planned economy and a free market econ-
omy as well as a driver of economic change. It is a forward-thinking
and technologically advanced enterprise and one that lacks transpar-
ency and accountability in its actions.
In order to understand CNPC as a whole, the enterprise must be
understood in its historical context – both in its relationship to the
Chinese state and to the outside world. This is explored in section 2 of
the chapter. Section 3 examines the relationship between CNPC and the
state, with particular attention to three perspectives: 1) the current cor-
porate governance and internal organization of the rm; 2) the overseas
expansion of the rm; and 3) the subsidies structure that is supported by
CNPC and the government. Section 4 focuses on CNPC’s nancial per-
formance and compares it with PetroChina and publicly listed Western
companies. Section 5 looks at non- nancial indicators of rm perform-
ance, namely the technical capacity and expertise of CNPC.
This chapter makes three main arguments. First, the nancial
performance of CNPC represents two realities. One reality is the
reported performance of PetroChina, the public arm of CNPC listed
on the Hong Kong, Shanghai, and New York stock exchanges, as
a nancially successful enterprise whose of cial performance rivals
any international oil company (IOC). Another reality is CNPC’s
nancial performance excluding PetroChina’s earnings. CNPC with-
out PetroChina is dramatically less ef cient than IOCs. The rm’s
performance is also much more dif cult to assess because not all of
its activities and nancial information are disclosed. As an arm of
the Chinese state, CNPC is also preoccupied with state goals that
extend far beyond producing oil, such as generating employment and
securing particular oil supplies for China. Even though CNPC owns
86 percent of PetroChina’s shares and PetroChina is completely con-
trolled by CNPC, the goals of these interwoven, but separate, entities
are divergent. PetroChina portrays an NOC that is autonomous,
competitive, and modern; CNPC is a conventional NOC that holds
the interests of the country as the top priority and acts more like an
agency for the government than a business. This divergence re ects
the fundamental tension in the relationship between the company and
the Chinese government. While CNPC is trying to nd ways to earn
, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511784057.012
Downloaded from https://www.cambridge.org/core. ETH-Bibliothek, on 26 Jul 2017 at 13:20:34, subject to the Cambridge Core terms of use
CNPC (China) 381
its own autonomy, its relationship with the Chinese state is still a bal-
ancing act between assuring nancial ef cacy and the government’s
efforts to assure social stability and a cheap supply of energy needed
for economic growth .
1
Second, an increasing par t of C N PC’s operat ions is overseas. W h ile
most analysis has pointed to overt government support for such oper-
ations, in fact the Chinese government is increasingly wary about
such operations due to their political fallout (e.g., in Sudan). CNPC
is not driving overseas with the guidance of the Chinese state; rather,
business considerations draw the company overseas where it is able to
earn higher pro ts and enjoy greater autonomy than in its operations
on Chinese territory. Moreover, as a state enterprise that accepts a
lower rate of return on its risk-adjusted investments than comparable
private rms, CNPC has a major competitive advantage in attracting
the support of overseas governments. CNPC’s initial forays overseas
were inspired partly by the Chinese government’s goal of assuring
energy security, but that is no longer the main reason that the trend
has continued and grown.
Third, CNPC is among the few NOCs examined in this book that
have built a vast and highly competent research and development and
technological capacity. Over its history (and especially since 2003)
CNPC has invested in technologies that allow it to better tap the
Time
CNOOC
(1982)
China
Petrochemical
Corp.
(1983)
CNPC
(1992)
(under SPC)
China Star
Petroleum
Co. Ltd
(1997)
Sinopec Group
(1998)
Sinochem Int’l
(2000)
Sinopec Corp.
(1983)
(under SPC)
PetroChina
(1999)
Sinopec Star
Petroleum Co. Ltd
(merged with
Sinopec 2000)
State-owned
companies
Publicly-listed
Former Gov’t
agency
Existing Gov’t
agency
China Import
and Export Co.
(1951)
Ministry of
Petroleum
Industry
(1978)
Ministry of
Geology and
Mineral
Resources
Figure 9.1 Evolution of the Chinese oil industry.
Source: Updated, revised and redrawn from Wang ( 1999 ).
, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511784057.012
Downloaded from https://www.cambridge.org/core. ETH-Bibliothek, on 26 Jul 2017 at 13:20:34, subject to the Cambridge Core terms of use
BinBin Jiang382
increasingly dif cult geology it has at home as well as in technologies
that allow it to expand beyond its core competency (onshore explor-
ation and production) into other activities such as re ning and offshore
operations. However, CNPC is a highly compartmentalized company
and expertise developed in certain subsidiaries is often dif cult to share
with other parts of the company. This mode of organization probably
explains why CNPC is relatively good at managing its increasingly
complex and aging elds within an existing technological paradigm
but is still struggling to develop new business models and competence.
2 Overview and history of the Chinese oil industry
2.1 Overview of the Chinese oil industry
State-owned enterprises (SOEs) comprise a major portion of the eco-
nomic activity in China. Whereas market-oriented reforms have swept
across much of China, SOEs still dominate in critical “pillar” indus-
tries – including the petroleum industry, which supplies resources that
are seen as essential to the country’s economic growth and security
(Ma and Andrews-Speed 2006 ). CNPC is the largest part of the pet-
roleum pillar. As is typical in such industries, the central government
oversees most major decisions such as allocation of capital and stra-
tegic choices for oil and gas exploration. As part of a broader reorgan-
ization of SOEs aimed at boosting performance, the government is
partly loosening its grip on the oil industry. Among other initiatives in
the sector, CNPC sold shares in its major subsidiary, PetroChina, on
the New York Stock Exchange (Ma and Andrews-Speed 2006 ).
Today, there are ve major oil companies in China: China National
Petroleum Corporation (CNPC); China National Offshore Oil
Corporation (CNOOC); Sinopec; Sinopec Star Petroleum Co. Ltd.;
and Sinochem (see Figure 9.1 ). All China’s oil companies trace their
roots back to government agencies or government-run organizations:
the Ministry of Petroleum Industry (disbanded); Ministry of Geology
and Mineral Resources (still existing); China Import and Export
Company
2 (disbanded). The Ministry of Petroleum Industry was dis-
banded in 1978 to create three enterprises: CNPC and China Star
Petroleum Co. Ltd., which were responsible for all onshore explor-
ation and production, and CNOOC, which was the dominant player
in offshore exploration and production. Sinopec was responsible for
, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511784057.012
Downloaded from https://www.cambridge.org/core. ETH-Bibliothek, on 26 Jul 2017 at 13:20:34, subject to the Cambridge Core terms of use
CNPC (China) 383
downstream re ning and petrochemical sectors. All oil imports and
exports had to be processed through Sinochem.
CNPC, CNOOC, and Sinochem dominated the landscape until
1998 when modern reforms began. However, because the sector is a
“pillar” in the economy, reforms mainly have involved reorganizing
(and solidifying) state enterprise rather than risking loss of state con-
trol. After 1998, the government sought to boost productivity in the
sector by creating more vertically integrated NOCs; for example, it
forced Sinopec and CNPC to swap some upstream and downstream
assets so that both became vertically integrated. (The government
also forced CNPC and Sinopec to separate their core and non-core
assets so that both would be focused more singularly on the oil indus-
try.) These two large onshore NOCs became, in effect, a duopoly
with Sinopec focusing more on the southern and eastern regions of
the country and CNPC active in mainly the northern and western
regions .
PetroChina, a company listed on both the New York and Hong
Kong stock exchanges, is the holding company for CNPC’s most
Shanghai
CHINA
INDIA
NEPAL
MONGOLIA
Beijing
Hong Kong
East
China
Sea
Oil field
Gas field
TAIWAN
RUSSIA
KAZAKHSTAN
MYANMAR
N. KOR.
S. KOR.
JAPAN
VIETN.
LAOS
BANGLA.
BHUTAN
KYRGYZSTAN
PAK.
Daqing Fields
Tarim Basin
West-East pipeline (gas)
Figure 9.2 Map of major oil and gas elds in China.
Source for oil and gas eld data: Wood Mackenzie (2009b).
, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511784057.012
Downloaded from https://www.cambridge.org/core. ETH-Bibliothek, on 26 Jul 2017 at 13:20:34, subject to the Cambridge Core terms of use
BinBin Jiang384
attractive and nancially viable core assets.
3 Investors bought them
eagerly, which offered an opportunity for CNPC to raise funds and
to engage more ac tively in overseas markets. CNPC retained manage-
ment of all non-core assets, such as hospitals and schools. The rm
controls 86 percent of PetroChina’s stock, making the state-owned
parent therefore, in PetroChina’s own words, the “ultimate control-
ler” of the company (PetroChina 2008 ).
Today, CNPC, Sinopec, and CNOOC together are responsible for
90 percent of China’s oil and gas production. While CNOOC’s near
monopoly of offshore activities allows it to operate in its own world,
CNPC and Sinopec have complementary skills and engage in a symbi-
otic relationship that fosters both cooperation and competition. CNPC
has more experience and resources invested in exploration and produc-
tion (E&P) but is relatively weak in the re ning aspects of the industry;
indeed, more than 10 percent of CNPC’s sale of crude oil and re ned
oil products is to Sinopec for distribution and retail marketing.
China’s NOCs have welcomed the government’s goal of maintain-
ing a near monopoly over the oil business. There are strict limitations
on private (including foreign) investment, such as in the wholesaling
of crude oil and re ned oil products as well as in the construction
and management of gas stations. In a few areas where the Chinese
government is not con dent that its NOCs can mobilize the needed
investment and technology (such as in some kinds of oil and gas
exploration, in the development and application of technologies for
increasing recovery rates in mature elds, and in some aspects of oil
re ning and coking) private investment is allowed, but only when for-
eign companies work in cooperation with a domestic rm, namely,
one of China’s NOCs (Shi 2005 ). Many offshore projects and some
gas projects t into this category, but China’s NOCs dominate the
bulk of the Chinese oil industry.
2.1.1 Oil production and reserves
CNPC’s 1.7 billion tons of recoverable reserves are two-thirds of
the country’s total and more than three times those of Sinopec and
six times the reserves of CNOOC. Eighty percent of its reserves are
booked in mature oil elds (such as the Daqing oil elds in north-
eastern China that are the traditional heart of China’s oil industry,
see Figure 9.2 ), and CNPC has developed exceptional competence in
managing such elds at relatively low cost. Although independently
, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511784057.012
Downloaded from https://www.cambridge.org/core. ETH-Bibliothek, on 26 Jul 2017 at 13:20:34, subject to the Cambridge Core terms of use
CNPC (China) 385
veri ed data does not exist, it is widely thought that total production
costs in CNPC’s mature oil elds are about $20/barrel – which com-
pare favorably with world-class performance. However, CNPC has
not seen the sharp rise in costs during the 2000s that Western com-
panies have experienced because it controls most of its value chain
and thus has less exposure to the vagaries of the international mar-
ket in oil eld services (Yuan 2007 ). Elsewhere in this book Peter
Nolan and Mark Thurber ( Chapter 4 ) distinguish oil elds and their
operators by risk . CNPC’s Daquing elds t rmly in the category of
low-risk mature oil production that most competent NOCs would be
able to operate, although some of the aging elds present novel com-
plexities . New developments in exploration are now leading CNPC
into territories such as remote deserts and arid plateaus as well as
mountainous regions, along with a new focus on smaller and more
complex elds; together, these developments are raising the rm’s
total production costs for new elds to perhaps around $40–50/
barrel (Yuan 2007 ).
Since about 1990, the Chinese government has encouraged greater
use of natural gas – in an effort to diversify the energy system and
reduce local pollution. Traditionally, the Chinese gas industry has con-
centrated on the gas elds in the Erdos Basin and also gas associated
with oil in the country’s maturing elds, notably around Daquing. In
the past few years, CNPC has put a higher priority on nding gas,
notably in the country’s far west (e.g., the Tarim Basin), with favor-
able results. At the end of 2005, the company held recoverable gas
reserves of 1.95 trillion cm (73.3 percent of the nation’s total) and its
leadership position is likely to remain into the future. CNPC projects
that it will nd enough new gas that its reserves will be 5.6 trillion cm
by 2020 (Yuan 2007 ) – or about twenty- ve years of production at
the expected rate of consumption that year (120 bcm per year).
2.1.2 Energy demand and consumption
Since the late 1990s, the demand for petroleum products has rapidly
increased, especially for light and middle distillates used to fuel the
country’s increasing mobility by automobile and truck (see Figure 9.3 ).
At the same time, due to a scarcity of resources and technical limita-
tions, oil production has stagnated. From 1998 to 2005, production
grew on average only 1.5 percent while demand soared at 7.5 percent
annually (IEA 2008 ). In 1993 China became a net oil importer. This
, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511784057.012
Downloaded from https://www.cambridge.org/core. ETH-Bibliothek, on 26 Jul 2017 at 13:20:34, subject to the Cambridge Core terms of use
BinBin Jiang386
pattern is expected to continue into the future, although a proper
assessment is dif cult because Chinese of cial projections have a his-
tory of underestimating demand. Other projections envisioned that
about half of China’s oil would be imported in 2010, 65 percent by
2020, and three-quarters in 2030 (IEA 2008 ).
2.2 History of the Chinese oil industry
2.2.1 The beginning
China’s oil industry has always been intertwined with the functions
of the Chinese government. From the start of Communist rule in 1949
energy was seen as a strategic concern (Zhang 2004 ). Mao Zedong
ordered the 57th Division of the 19th Army of the PLA to be reorgan-
ized into an “oil corps” (1st Division of Oil) (Zhang 2004 ). The unit
was charged with the goal of quickly developing areas for exploration
and production of oil to allow self-suf ciency; by the early 1960s that
goal became urgent as the diplomatic relations disintegrated between
China and the Soviet Union, which supplied more than half China’s
critical re ned oil products (Downs 2000 ).
Communist central planning allowed few exceptions to the rule that
the center controlled all resources. The State Planning Commission was
ultimately responsible for determining the national plan for the demand
0
1,000
2,000
3,000
4,000
5,000
6,000
7,000
8,000
9,000
1965 1970 1975 1980 1985 1990 1995 2000 2005
Thousand barrels daily
Others
Fuel oil
Middle distillates
Light distillates
Oil production
Figure 9.3 Oil production and oil consumption by product group,
1965–2007.
Source: BP ( 2008 ).
, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511784057.012
Downloaded from https://www.cambridge.org/core. ETH-Bibliothek, on 26 Jul 2017 at 13:20:34, subject to the Cambridge Core terms of use
CNPC (China) 387
and supply of oil, allocation for processing, and oil imports and exports.
To help deploy the plan that the State Planning Commission penned,
in 1955 the government created the Ministry of Petroleum Industry
(MPI) (Wang 1999 ). As the headquarters of China’s nascent oil indus-
try, MPI was responsible for making investment decisions (e.g., the
location and timing of exploration and production projects), coordin-
ating the transportation and distribution of oil products, and allocating
oil resources to local-level entities known as Petroleum Administrative
Bureaus (PABs). Each of these local agencies managed a particular oil
eld and was responsible for meeting the production targets that the
State Planning Commission set out in the national plan.
The urgent, militarized effort to nd local production yielded early
successes. China soon started to produce enough oil to become self-
suf cient; by 1970 it was a net exporter. During the 1970s, China
sold oil (below market price) to Japan in order to decrease Japanese
demand for Siberian hydrocarbons and divert cash away from the
Soviet Union, which China feared might attack the northeastern
part of the country. Starting in the 1980s, the oil elds tapped for
the country’s surge in production began to mature, and growth in
production slowed. At the same time, massive and broader economic
reforms led to tremendous economic growth; GNP rose 8.6 percent
annually from 1979 to 1991 (Steinfeld 1998 ). Much of the growth in
productivity and employment occurred outside the state sector – nota-
bly within township and village enterprises (TVEs) that predominated
in southern rural areas; industrial state-owned enterprises, including
those that dominated the oil sector, were largely untouched. By 1988,
the entire petroleum industry was losing money; its problems included
lack of investment in exploration and production of oil as well as
mounting debt (Zhang 2004 ).
In an attempt to improve performance of state enterprises, in 1988
the central government abolished the MPI and assigned its assets and
responsibilities to CNPC.
4 (It implemented similar reforms in other
sectors, such as coal and nuclear power.) This partial corporatization
put CNPC in charge of not only operations (onshore and shallow
water oil and gas exploration and production; overseas cooperation)
but also some regulation that MPI had previously overseen, such as
setting standards for the petroleum industry.
Through these reforms, CNPC shared with the government the
task of planning. In consultation with the State Council, which
, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511784057.012
Downloaded from https://www.cambridge.org/core. ETH-Bibliothek, on 26 Jul 2017 at 13:20:34, subject to the Cambridge Core terms of use
BinBin Jiang388
devised all the country’s central economic plans, CNPC was in charge
of long-term planning, including control over investment decisions
within China and overseas; the planning bureau of CNPC was in
ch a r ge of busine ss s t rategy plan n i ng and p rodu c t ion pl a n n ing (inc lud-
ing regulation of inputs and outputs from the oil industry) in conjunc-
tion with the State Planning Commission and State Economic and
Trade Commission (SETC). The State Planning Commission’s plan
also required that CNPC coordinate with Sinopec, the downstream
state-owned oil company, to draft the detailed supply and transpor-
tation elements that were included in the master plan. Funding for
CN PC came from the a rray of sou rces t hat were used to nance opera-
tions of most large state-owned enterprises: government budget, state
bank loans, domestic bond issues, and overseas sources channeled
through the Ministry of Finance and the Bank of China. CNPC, at
this point, was not a real business that could reinvest its pro ts to
grow. It required the assistance of the state in order to be operational
and it operated with a “soft budget” because many of its loans, for
example, were issued with an ambiguous expectation of repayment.
Nearly every aspect of the company was interwoven with the state
and party apparatus. For example, the State Council and the Chinese
Communist Party (CCP) Department of Organization appointed the
president and vice presidents of CNPC. A “party committee” within
CNPC appointed the rest of the top management team. The structure
of CNPC remained almost completely unchanged between 1988 and
1998, although various minor reform policies were implemented to
improve performance of the oil industry as a whole .
2.2.2 Reforming CNPC
As the Chinese economy soared in the 1990s, there were many reforms
aimed at raising the economic ef ciency of the oil industry. These
reforms demonstrated the government’s recognition of the inherent
aws within the system of state-dependent companies; its reform strat-
egy, in general, sought to give these enterprises greater autonomy as
well as exposure to market-like competition. But this strategy was lled
with internal contradictions. Notably, the trend toward autonomy had
created efdoms that made it hard to preserve central planning (still the
bedrock of China’s economy) and also sowed chaos in the oil industry.
In 1998, the government abandoned many of these reforms with a new
strategy that recentralized some of CNPC’s authority.
, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511784057.012
Downloaded from https://www.cambridge.org/core. ETH-Bibliothek, on 26 Jul 2017 at 13:20:34, subject to the Cambridge Core terms of use
CNPC (China) 389
One area of particular chaos was the operation of the PABs.
Autonomy for the PABs rested on a 1981 government reform called
the “big contract” model. While MPI was still nominally in charge of
oil elds, it contracted with the central government to hand over 94.5
percent of actual oil output for domestic use but was allowed to keep
the rest for export. MPI was also permitted to retain the difference
in price between the higher international prices and lower domestic
prices – revenues it was supposed to invest in exploration and pro-
duction. MPI contracted with the PABs to actually produce the oil,
and these local agencies also received higher revenues from exceed-
ing quotas, which provided them with a strong incentive to push up
production. Each PAB faced a strong, individual incentive to boost
output rather than collectively follow the dictates of the central gov-
ernment. This policy, which was seen as a success because it seem-
ingly forestalled decline in production, was renewed in 1991. Under
the big contract system, performance indicators were put in place to
evaluate the performance of each subordinate enterprise.
5 However,
as with most systems that operate on soft budgets, the consequences
of failure to meet performance targets were never clear.
PABs were not only allowed to keep a share of pro ts but there were
also reforms that shifted the incentives for local managers of PABs.
Known as the “oil company” model of reform, these reforms corpo-
ratized the PABs, encouraged them to invest more of their resources
in the core elements of the petroleum industry and to contract out the
diversi ed fu nc ti on s ne ede d to supp or t th at m is sion , such as so ci al ser -
vices, utilities, and the manufacture of drilling machinery. While new
oil elds that came online after 1989 certa inly followed t his organiza-
tion (for example, the Tarim Basin, which opened that year, was the
rst region to implement the “oil company” model) it is unclear how
much success traditional PABs had with restructuring. In the 1990s,
as part of this dalliance with decentralization, CNPC also adopted a
policy that gave all its subordinate enterprises “legal person” status.
This meant that each enterprise would be responsible for its own prof-
its, losses, and growth – imposing, in theory, hard budget constraints
that are often essential to successful market-oriented reforms . For the
most successful enterprises, such as China’s largest PAB and oil eld
Daqing, the new status was encouragement to become more than just
a regional branch company . It was seen as an opportunity to develop
its own business strategy and even to work toward possible listing as
, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511784057.012
Downloaded from https://www.cambridge.org/core. ETH-Bibliothek, on 26 Jul 2017 at 13:20:34, subject to the Cambridge Core terms of use
BinBin Jiang390
a separate company in the future. In the partially reformed planning
system – where PABs were encouraged to follow their own instincts
yet central planning meant that all parts of the far- ung oil industry
were tightly interdependent – this policy of decentralization would
later prove disastrous as fragmentation of the oil industry started to
cause both nancial and management problems for CNPC .
In 1995, the government tried to improve performance of the oil sec-
tor by adopting a three-tier oil pricing system that would start to align
domestic prices with prevailing international prices. Lower (“grade I”)
prices were set for t ransfers from CNPC (which main ly produced crude
oil) to Sinopec (which focused on re ning); a second price (“grade II”)
prevailed within the state production plan ( zhi ling xing ji hua ), and
anythi ng t hat the PABs had lef t over could be sold at i nternational pr ice
levels. While the goal of this policy was admirable, its perverse impacts
unfolded as the government attempted to regulate similar products
at different prices. Sinopec experienced supply shortages. “Oil bro-
kers,” who theoretically should not have existed in the planning sys-
tem, emerged to control key decisions around allocation of effort and
rents. They intercepted low-priced oil (intended for use according to
government plan) and sold it at a higher rate for the private domestic
market. Sinopec was forced to import larger quantities of expensive
international oil even as the government subsidized the price difference
between “grade I” oil and international market prices.
During the late 1990s, the central government began to abandon
the big contract model by adopting a royalty-like approach to revenues
from oil production. The new policy would require oil enterprises to
“hand over” ( shang jiao ) a negotiated amount that includes the por-
tion of the pro ts and fees for using the reserves ( chuliang shiyong
fei ). As is typical with such reforms, the oil producers (PABs) found
themselves in dispute with their regulators (CNPC) over the proper
allocation of revenues from their operations – in particular, the size
of remittances to the Chinese government. CNPC had little control
or knowledge over capital expenditures and thus, as regulator, was
poorly equipped to set a royalty fee that encouraged rational produc-
tion. For example, PABs controlled many more services than just the
core functions of oil exploration and production. Despite attempts at
reform, these bureaus provided social services such as hospitals, elec-
tricity generation, and schools as well as ventures into many diver-
si ed businesses that were not on the books at CNPC even though
, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511784057.012
Downloaded from https://www.cambridge.org/core. ETH-Bibliothek, on 26 Jul 2017 at 13:20:34, subject to the Cambridge Core terms of use
CNPC (China) 391
CNPC nominally owned and controlled the PABs. Since these ancil-
lary enterprises were under no obligation to achieve a certain return
on their investments, they encouraged the PABs to increase spending
even when that destroyed value. In practice, however, the reform of
the big contract system toward the “hand over” policy did not really
affect behavior because the budget constraints remained soft; a por-
tion of the pro ts from solvent enterprises was remitted to the state
while state enterprises that suffered losses were given subsidies using
these same funds. Ironically, because the industry had become more
decentralized, CNPC was unable to monitor the escalating losses of
each PAB, which kept nding new ways to lose money while generat-
ing local political bene ts.
For a while, the autonomy of the PABs helped to boost the coun-
try’s overall oil production, but by the 1990s that surge in output had
run out of steam (see Figure 9.4 ).
In 1998, the government ordered sweeping reforms that funda-
mentally restructured the entire Chinese oil industry. The impetus
for change was the government’s desire to create a Chinese petrol-
eum industry that would be competitive with IOCs in the global
market. In addition, the need for change was also evident in the num-
ber of regional oil enterprises that were losing money. Indeed, the
–
25
50
75
100
125
150
175
200
225
1965 1970 1975 1980 1985 1990 1995 2000 2005
Million tonnes oil
1978
Ministry of
Petroleum
disbanded
Figure 9.4 Chinese oil production and consumption, 1965–2007.
Source: BP ( 2008 ).
, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511784057.012
Downloaded from https://www.cambridge.org/core. ETH-Bibliothek, on 26 Jul 2017 at 13:20:34, subject to the Cambridge Core terms of use
BinBin Jiang392
cost of oil production was rising above the prevailing domestic level
due to the failure to invest in research and long-term planning in the
industry. The central government, which owned CNPC, faced steep
declines in pro t and yet was unable to spur improvements in each of
the regional oil enterprises because CNPC had little control over the
actions of each region.
The reforms concentrated on three main actions. First, the “legal
person” status of the regional enterprises was revoked in an attempt
to recentralize decision making in the oil industry. This element of the
reform aimed at producing a more coherent strategy for the industry
overall and creating stronger incentives for better economic perform-
ance, although success on both fronts has so far been elusive.
6 Second,
CNPC and Sinopec swapped assets and became regional, vertically
integrated compa nies, mak ing each one a stronger competitor by hav-
ing access to complementary assets for which they were dependent on
one another before .
7 Third, core and non-core assets in each CNPC
production unit were separated from each other. T he core assets were
reconstituted as “branch companies” and were eventually oated as
PetroChina. The non-core assets were spun off where possible; where
not possible, they were kept together under CNPC’s ownership.
8 The
CNPC that exists today is the direct descendent of these reforms.
3 The state–CNPC relationship
As a state-owned enterprise, the key to understanding CNPC is in its
relationship to the Chinese government. That relationship is not easy
to summarize and thus here I examine it through two lenses. First,
I examine who really controls CNPC – in particular, its procedures
for corporate governance and internal decision making. Second, I
examine how CNPC and its listed subsidiary PetroChina have been
affected by the various attempts to liberalize the Chinese economy
and introduce market-like forces within state enterprises. While some
level of competition has been introduced to the industry by reforms
introduced in the past couple of decades, wariness about upsetting the
political order and completely liberalizing the energy sector has led
to policies that explicitly prevent these rms from behaving as truly
competitive and ef cient enterprises. One by-product of liberalizing
reforms has been a stronger incentive for the Chinese oil industry, in
particular CNPC, to invest overseas.
, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511784057.012
Downloaded from https://www.cambridge.org/core. ETH-Bibliothek, on 26 Jul 2017 at 13:20:34, subject to the Cambridge Core terms of use
CNPC (China) 393
3.1 Who controls CNPC?
CNPC’s relationship with the Chinese state is mediated by the
Communist Party of China (CPC). From the founding of communist
China in 1949 until today the CPC has traditionally exercised control
over the country’s economy. The ultimate priority of such one-party
political rule is to retain control of its institutions. In turn, the high-
est priority for managers within these institutions is to move up the
Communist Party’s political hierarchy. Thus the heads of state-owned
enterprises and other public institutions are, rst and foremost, stew-
ards of political ideology and the party line; economic ef ciency is sec-
ondary.
9 Creating incentives for ef ciency has not been easy because it
has required disentangling the functions of state enterprises from the
political and ideological missions of the state and party. Among those
efforts has been corporatization, which has aimed to create the right
incentives for managers of state enterprises even as ultimate control
over the enterprises has not been divested from the state and the CPC.
Below I look at the basic concepts that guide Chinese understanding
of corporate governance and then at the managers and decision mak-
ing within CNPC.
3.1.1 Corporate governance and control over senior managers
The concept of corporate governance came into the consciousness of
the Communist Party with the advent of economic liberalization poli-
cies in 1978. The idea refers to the institutional forms and mecha-
nisms that govern how managers control the enterprise and allocate
risks (McNally 2002 ). In theory, the embrace of corporate govern-
ance re ected a new goal of government in the post-1978 reforms – to
maximize ef ciency and minimize cost for state-owned enterprises –
which was pursued by transforming state enterprises into rms with
individual legal status and accountability. In practice, the theory
was never so simple and implementation proved to be quite complex.
Incentives given to managers were of ten not strong enough to encour-
age adequate changes in behavior. Ownership rights were poorly
de ned, which resulted in a “principal/agent problem” because the
people who stood to gain from sound business decisions were not
actually the decision makers themselves.
The most important force for change in the management of state
enterprises was the problem of losses. Traditionally, state banks
, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511784057.012
Downloaded from https://www.cambridge.org/core. ETH-Bibliothek, on 26 Jul 2017 at 13:20:34, subject to the Cambridge Core terms of use
BinBin Jiang394
covered losses by state enterprises with soft loans. As the nancial
situation of the SOEs (and their sponsors in the state banking system)
worsened, the government began to see corporatization as a way to
x both problems. Not only would reforms clarify ownership and
accountability but as the reform movement progressed, it was also
thought that public listing would yield an important source of nan-
cing for this restructuring and was especially attractive to govern-
ment masters because it shifted pressure away from state banks to
equity markets that were stuffed full of investors that wanted entry
into China. The public listing of PetroChina, for example, arose from
this logic.
Broadly there were two tracks of market-oriented reform in China.
In the rst track, which began in the late 1970s and continued into
the 1990s, large SOEs were kept under state control although their
managers were given more autonomy and responsibilities. By contrast,
smaller SOEs were left to fend for themselves.
10 Seemingly, this rst
track of reform exposed segments of the Chinese industrial sector to
elements of market competition although, in practice, the real incen-
tives of a market were blunted. For example, many of the employees
that were red from the smaller SOEs were rehired by larger SOEs that
had been only partially reformed. In the oil industry, this approach to
reform meant that large, in uential companies such as CNPC came
under closer scrutiny of the central government, while PABs were left
to fend for themselves – at least until the chaos from decentralization
led to efforts by the government throughout the 1990s (notably in the
1998 reforms) to reassert authority through CNPC .
11 Nonetheless,
today CNPC is probably more ef cient as an enterprise due to this
legacy of decentralization. Even today, many of CNPC’s subsidiaries
remain fairly independent (especially regarding their investments in
R&D, which are detailed below); moreover, the process of decentral-
ization helped weed out some of the less ef cient subsidiaries (at least
those that could not secure state resources to sustain their operations)
while rewarding the more innovative subsidiaries. These reforms
included a new “Company Law,” which became effective in 1994 and
required that companies create a board of directors and hold meetings
of shareholders to elect directors and make major corporate decisions.
In addition, limited liability corporations were required to establish a
board of supervisors to provide independent oversight and evaluation
of the behavior of directors and managers.
, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511784057.012
Downloaded from https://www.cambridge.org/core. ETH-Bibliothek, on 26 Jul 2017 at 13:20:34, subject to the Cambridge Core terms of use
CNPC (China) 395
In parallel, but on a slower track and thus harder to assess from
today’s vantage point, the government adopted reforms that would
actually enforce accountability in corporate governance and allow
the government to become a more effective shareholder. During the
late 2000s, the government adopted a new three-tier state asset man-
agement system that would align the interests of the owners and
those who ultimately managed the companies; organizations on
each tier are held accountable to the organizations in the tier above.
Traditionally, any single state enterprise might be controlled by sev-
eral governmental agencies but none would actually own any of the
assets or be ultimately accountable. This created incentives for the
controlling ministries to act on the behalf of their constituencies
rather than the broader public interest or even to serve the govern-
ment as shareholder. The new system has attempted to align the
interests of management with owners by giving government a more
singular and decisive voice as shareholder. The rst (top) tier of the
system includes two new commissions that were created to supervise
the state-owned assets and regulate the nancial sector : the State-
Owned Assets Supervision and Administration Commission (SASAC)
and the China Banking Regulatory Commission (CBRC).
12 Under the
Company Law, SASAC’s main functions include promoting reform
of state enterprises as well as promoting better management of SOEs
and their integration into the Chinese economy.
13 By some interpret-
ations, SASAC has expansive and extensive control over the jewels of
the Chinese state-oriented economy. It implements its obligation at
large SOEs – such as CNPC as well as the other major oil compan-
ies such as Sinopec and CNOOC – through supervisory panels and
by exerting ultimate control over the choice of top executives . Below
SASAC is the second tier of the new corporate governance system,
which consists of the state holding corporations. The holding com-
pany’s role is akin to that of an institutional investor for the state
that aims to maximize the state’s return on its assets. CNPC is the
leading example of a holding company in the oil sector. The third tier
of the system consists of those SOEs that have been converted to lim-
ited liability corporations. Such corporations are owned by holding
companies – which appoint the boards of directors and supervisors –
although a portion of their shares can be listed.
The goal of the Company Law and the multi-tier system of state-
owned asset management was to “clarify property rights, establish
, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511784057.012
Downloaded from https://www.cambridge.org/core. ETH-Bibliothek, on 26 Jul 2017 at 13:20:34, subject to the Cambridge Core terms of use
BinBin Jiang396
clear powers and responsibilities, separate government and enter-
prise, and establish a scienti c method of management” (Jiang
1995 ). This complicated structure re ected the con icting goals of
boosting economic performance while not diminishing the govern-
ment’s hold on strategic industries. Indeed, although top managers
of each corporatized SOE are supposed to be elected via the board of
directors, they are instead usually appointed to the positions by the
Communist Party’s Organization Department (Tenev et al . 2002 ).
These managers, in turn, operate according to the incentives in their
career trajectories – they place the political interests of the party
above economic considerations for an enterprise. An additional drag
on performance is that local governments have shielded enterprises
from market forces by diverting money from public funds as subsi-
dies to these organizations. Since SOEs are usually an integral part
of the local economy, it is in the interest of the local government to
scuttle reforms that could cause unemployment or other forms of
social instability .
One interesting aspect of these reforms is that formal governance
and accountability of Chinese SOEs, at least on the surface, now look
a lot like Western companies. For example, CNPC now publishes a
corporate responsibility report every year, in addition to and separ-
ate from the annual report, which outlines safety and environmental
measures that the company is taking. In the last few years, CNPC has
adopted a management system focused on health, safety, and environ-
ment (HSE) that trains middle- and high-level management on these
goals. This shift re ects an effort to mimic best practice of the West;
however, the actual impact of HSE reporting is hard to ascertain.
These two tracks of reforms have not yet resulted in a fundamen-
tal change of the senior leadership in most SOEs, although that pat-
tern is evident even in publicly listed companies. As of the late 1990s,
independent directors on the board of directors with no ties to the
government or the company made up less than one-third of the board
of directors in Chinese SOEs (compared with around 75 percent for
Western private companies) (Mallin and Rong 1998 ). Today, after the
reforms have been under way for nearly a decade, not much is differ-
ent. Currently, of PetroChina’s twenty-four board members only six
are independent of the government and company .
14
Heavy insider in ltration of governing boards has produced
all the predictable problems. Notably, it is harder to expel poorly
, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511784057.012
Downloaded from https://www.cambridge.org/core. ETH-Bibliothek, on 26 Jul 2017 at 13:20:34, subject to the Cambridge Core terms of use
CNPC (China) 397
performing managers and to assure accurate reporting of nancial
and other performance indicators. Moreover, the performance cri-
teria by which managers are evaluated do not include traditional
metrics of stock price, shareholder returns, or economic value
added, but rather more subjective measures such as “improving
ideological and political work, enhancing Party conduct and anti-
corruption campaign” (CNPC 2007). The interpretation of this
message is that as long as managers ful ll their roles in keeping
the corporation in line with government and party expectations,
the managers would not only keep their jobs but also climb up the
Chinese SOE ladder (Liu 2006 ).
15 To be sure, the full range of incen-
tives that affect managers is complicated, but standard shareholder
performance plays only a minor role. Top managers receive bonuses
that account for a large share of their income, but some bonuses
are merely indexed to salaries (and thus not really bonuses). Some
are based on performance targets that make sense for the company
itself (e.g., net pro t, return on capital, and cost savings) and thus
partially align with shareholder interest. But shareholder perform-
ance (e.g., stock price) or customer satisfaction usually play little
role. More importantly, some performance indicators are not nan-
cial, but rather political goals such as “eliminating factors caus-
ing instability,” anti-corruption, “preventing occurrence of mass
commotion,” and “severe offences against party conducts” (Fan et
al . 2007 ). There is no real risk of managers being red unless a
major catastrophe occurs under their leadership. (Such catastro-
phes usually cause heads to roll for symbolic reasons as they are
rarely traced back to the singular decisions of particular managers.)
Because of their appointments by the highest levels of government,
poor business performance under top management’s direction does
not warrant a dismissal. Managers also cannot be voted out by the
board of directors without the nal approval of the State Council –
an important part of the government’s planning system that itself is
steeped in party control.
This system of control for top management positions helps to
explain why the population of managers is dominated by middle-aged
men with training in either engineering or economics and extensive
backgrounds in the oil and gas industry. Based on executive pro les
published by the companies, the same people are shuf ed between
CNPC and PetroChina (see Figure 9.5 ). Major changes in personnel
, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511784057.012
Downloaded from https://www.cambridge.org/core. ETH-Bibliothek, on 26 Jul 2017 at 13:20:34, subject to the Cambridge Core terms of use
BinBin Jiang398
are usually made on political grounds rather than the performance
record of the people involved – the most abrupt changes in leadership
are usually crisis-driven and geared toward protecting the reputation
of the central government rather than based on what was the best
choice for CNPC (Xin 2005 ). The signi cant reshuf ing of personnel
between 2002 and 2003, for example, corresponds with a major nat-
ural gas explosion that released a cloud of hydrogen-sul de gas that
killed 243 people over 28 towns in Kaixin County and the suburbs of
Chongqing Municipality at the end of 2003. Investigations following
the incident pointed to the absence of a critical blowout prevention
device in the well due to CNPC’s cost-cutting measures (Xu 2004 ).
The president of CNPC at the time, Ma Fucai, was forced to resign,
along with lower-level managers, and the six workers who followed
improper procedure at the worksite (Hubbell 2004 ; Jiang 2004 ).
According to CNPC’s 2003 annual report, Chen Geng replaced Ma
Fucai and more personnel were brought into the top management
team. A nother d isaster followed th is incident i n Nov ember 20 05 w hen
CNPC’s largest chemical plant, Jilin Petrochemical, contaminated the
Songhua River, the main drinking water supply of Harbin, the capital
of Heilongjiang Province, and also the Russian city of Khabarovsk.
Tap water was turned off for four days and affected 9 million people
in two countries (Li and Wang 2008 ). The president-in-residence once
again assumed responsibility and resigned. Heads are sacri ced to
garner credibility from the Chinese public, which has been inundated
with promises from senior government leaders that such disasters
would be averted .
In addition to the fact that reforms have not had much impact on
control over senior appointments, those reforms also have not had
the expected impact on decision making within CNPC. As already
discussed, the reforms starting in the late 1970s decentralized some
decision making within the oil industry – notably to the PABs. Even as
that control was reasserted by CNPC, the result has been an unusual
and fragmented decision-making process that re ects these twin, con-
icting pressures for and against central control.
On paper, today’s CNPC looks like a highly centralized institu-
tion modeled in the tradition of central planning. The CNPC central
planning bureau uses market analysis along with mid- to long-term
economic and technological projections to devise a ve-year plan that
optimizes use of the company’s resources. Next, each regional of ce
, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511784057.012
Downloaded from https://www.cambridge.org/core. ETH-Bibliothek, on 26 Jul 2017 at 13:20:34, subject to the Cambridge Core terms of use
CNPC (China) 399
Figure 9.5 Top CNPC leadership employment history.
Name Former employment
Position within CNPC in 2007 Age
Educational
background
Jiang Jiemin Vice president (CNPC); Vice-
governor (Qinghai Province)
President (CNPC), President
(PetroChina); Vice Chairman of
the Board (PetroChina); Alternate
member (17th CPC, Central
Committee)
52 M.S., Economics
Zhou Jiping Vice president (CNPC); Director
(PetroChina)
Vice president (CNPC); Director
(PetroChina) 56 M.S., Marine
geologic structure
Duan Wende Vice president (CNPC) Vice president (CNPC); Director
(PetroChina) 55 Engineering
Wang Yilin
Vice president (CNPC); Senior
Executive (Xinjiang Petroleum
Administration Bureau; President
(Xinjiang Oil Branch, PetroChina)
Vice president (CNPC); Director
(PetroChina) N/A Ph.D., Engineering
Zeng Yukang Assistant president (CNPC) Vice president (CNPC); Director
(PetroChina) 57 B.A., Economics
Wang Fucheng Vice-president (Petrochina)
Chief of Discipline and Inspection
Group (CNPC); Vice president
(PetroChina)
58 B.A., Economics
Li Xinhua Deputy governor (Yunnan
Province)
Vice President (CNPC); Director
(PetroChina) N/A B.S., Chemical
Engineering
Liao Yongyuan Assistant president (CNPC) Executive director (PetroChina);
Safety director (CNPC) 45 M.S., Engineering
Wang Guoliang CFO (PetroChina) CFO (CNPC) N/A M.A., Accounting
Chen Ming General manager of the
supervisory dept (PetroChina)
Chief of Discipline and Inspection
Group (CNPC); Chairman of
Supervisors (PetroChina)
N/A B.A., Economist
and its subsidiaries submit investment plans to the central of ce. The
central CNPC planning bureau then pieces together this informa-
tion and creates an integrated plan for the whole company. In turn,
it submits this plan to the State Council as well as CNPC’s board of
directors for approval.
16 In reality, however, the planning process is
usually more complicated and less centralized than the of cial pro-
cess outlined here. While it is hard to understand the planning pro-
cess fully by looking from the outside, CNPC’s two most prominent
accidents, discussed above, have attracted enough scrutiny that they
also offer analysts a window on what really happens inside the com-
pany. Careful reporting on these accidents usually reveals that CNPC
suffers from a lack of standardization in work safety guidelines and
inadequate enforcement, especially within operational units not
, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511784057.012
Downloaded from https://www.cambridge.org/core. ETH-Bibliothek, on 26 Jul 2017 at 13:20:34, subject to the Cambridge Core terms of use
BinBin Jiang400
Table 9.1b. External decision-making governmental bodies related to
CNPC after March 2008
Ministry Main function
State Energy
Commission (SEC)
A high-level discussion and coordination
body whose speci c functions have not yet
been determined. Replaces the Of ce of
the National Energy Leading Group.
National Energy
Administration
(NEA)
Vice-ministerial component of NDRC,
successor to the Energy Bureau of the NDRC.
Even though this administration has more
capability than the Energy Bureau, it still
lacks authority for ultimate decision making.
Ministry of Land and
Resources
Approval of exploration and development
activities
Ministry of Commerce Approval of downstream distribution plans
State Environmental
Protection Agency
Approval of environmental impacts of projects
Source: Yu a n ( 2007 ).
Table 9.1a. External decision-making governmental bodies related to
CNPC before March 2008
Ministry Main function
Of ce of the National
Energy Leading Group
Involved in energy planning and
policymaking; creates a strategic and
integrated approach to energy
NDRC (Energy Bureau) Regulation of the petroleum industry,
organizational reform; petroleum
development planning; strategic petroleum
reserves; approval of new projects; set
price of petroleum
Ministry of Land and
Resources
Approval of exploration and development
activities
Ministry of Commerce Approval of downstream distribution plans
State Environmental
Protection Agency
Approval of environmental impacts of
projects
, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511784057.012
Downloaded from https://www.cambridge.org/core. ETH-Bibliothek, on 26 Jul 2017 at 13:20:34, subject to the Cambridge Core terms of use
CNPC (China) 401
immediately under the eyes of the CNPC in Beijing. Both of the inci-
dents occurred in areas that are geographically far away from the cen-
ter, and there was little oversight from the main corporation (Zhang
2004 ). Another window into company decision making is opened by
examining how new technology is devolved through the company.
Although CNPC has its own centrally operated R&D centers, many
of the important technical innovations in recent years have actually
come from individual subsidiaries focused on solving practical prob-
lems rather than an overall R&D plan that assigns speci c tasks to
individual branches of the government. I explore R&D in more detail
below.
The most visible evidence of CNPC’s efforts to recentralize internal
decision making is found in exploration and production. Prior to the
1998 reforms, individual oil eld operators (i.e., the PABs) would
decide when and where to drill new wells and how to manage the
eld; currently, those decisions are made by the central CNPC plan-
ning of ces (Yuan 2007 ). While eld-level decisions are made by
CNPC, CNPC does not, itself, fully control its own planning process.
Overall production levels for the sector and other long-term strategic
decisions are made in consultation with government agencies, such
as the ones listed in Table 9.1a and Table 9.1b . The Energy Bureau of
the National Development and Reform Committee (NDRC) has an
especially important role to play in the process due to its power to set
both upstream and downstream conditions by setting the price of pet-
roleum products in the Chinese market. In the next section I explore
these many relationships between the enterprise and the government
in more detail .
3.2 The state–CNPC relationship as a re ection of the
process of economic liberalization in China
A second way to explore the relationship between the state and the
NOC is by examining how the process of economic liberalization has
altered the relationships between the government and the oil enter-
prise over time. Here I explore that by focusing on three of the most
important aspects of that relationship: 1) the links between CNPC
and its major subsidiary PetroChina; 2) the decision to allow and
encourage CNPC to invest abroad; and 3) the setting and adjustment
of subsidies for oil products.
, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511784057.012
Downloaded from https://www.cambridge.org/core. ETH-Bibliothek, on 26 Jul 2017 at 13:20:34, subject to the Cambridge Core terms of use
BinBin Jiang402
3.2.1 The relationship between CNPC and PetroChina
CNPC is a holding company that is wholly owned by the government
and under the direct control of the State Council; in turn, CNPC
controls eighty-seven subsidiary enterprises, most of which special-
ize in engineering, technical services, and equipment manufacturing
(Zhang 2004 ). Of all CNPC’s subsidiaries, PetroChina is the only
publicly listed entity. Its shares are listed on the New York Stock
Exchange (NYSE), Hong Kong Stock Exchange (HKSE), and the
Shanghai Stock Exchange (SSE); thus it must conform to the man-
datory laws of the PRC Company Law and the securities laws and
regulations of Hong Kong and the United States (PetroChina 2008 ).
Indeed, by meeting these listing requirements, CNPC is able to claim
that PetroChina is as open and transparent as Western companies and
is therefore better able to gain shareholder trust.
While the governance structure of PetroChina may look similar to
other public companies and it must (on paper) conform with stringent
li sting rules, the actual prac tice is different. Notably, Pe t roCh i n a do e s
not vest control over crucial parts of the decision-making process
with independent directors. CNPC is able to elect the entire board of
directors without practical input of external shareholders; it is also
able to dictate the timing and amount of dividend payments (Ewing
2005 ). The lack of independent directors also means that the top
management of PetroChina is reviewed by its peers and associates,
people who operate within the same political and social networks –
a situation that is rife with con icts of interest. Moreover, because
directors are not obligated to act under a code of ethics required by
NYSE and other listing rules (Ewing 2005 ), they are probably able to
escape scrutiny for actions that are not allowed by directors in other
companies. Because its operations are still controlled exclusively by
CNPC, minority shareholder interests are poorly protected, although
such concerns have not urgently arisen because the interests of minor-
ity shareholders and CNPC happen to align: the former want growth
in share value and the latter needs to demonstrate credibility (for the
moment) in Western stock markets.
While CNPC and PetroChina are supposed to be formally separ-
ate, in fact there are blurry lines between the enterprises in nearly
every aspect of their operations. In 2007, CNPC was the largest
supplier and purchaser of products from PetroChina, such as con-
struction and technical services and supply of materials. CNPC
, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511784057.012
Downloaded from https://www.cambridge.org/core. ETH-Bibliothek, on 26 Jul 2017 at 13:20:34, subject to the Cambridge Core terms of use
CNPC (China) 403
and PetroChina also exchange assets, and some analysts believe
that such transactions are designed to keep PetroChina attractive
to private investors while shuf ing less desirable assets to CNPC.
In 2007, PetroChina acquired assets of the risk operation service
business from CNPC. In addition, CNPC is also a major creditor to
PetroChina through its subsidiary CP Finance. In 2006, guarantee
of borrowings by CP Finance totaled about 80 percent of the total
amount of borrowings for PetroChina. In 2007, CP Finance sup-
plied all of PetroChina’s state-backed debt. These loans, backed by
a de facto state guarantee, are made on the basis of low interest rates
(4.46–7.47 percent). In addition to these activities, PetroChina has
also acquired (in 2005) subsidiaries from CNPC, such as the re n-
ing and petrochemical business of CNPC, and 50 percent equity
interests in an overseas CNPC subsidiary, CNPC Exploration
and Development Company Limited (PetroChina 2008 ). In 2007,
PetroChina acquired the risk management services business of
CNPC. CNPC and PetroChina are not separate entities; rather, they
are more akin to divisions of the same company that are separated
by porous technical and legal barriers – with one division listed on
stock exchanges mainly for the purpose of raising money .
3.2.2 The decision to “go abroad”
Since CNPC’s rst overseas service contract – with Peru for the
Talahara oil eld in 1993 – the company’s overseas activities have
steadily increased with time. After more than a decade of devel-
opment, CNPC has established overseas oil and gas operations in
Africa, the Asia-Paci c, Central Asia-Russia, the Middle East, and
South America. In all, by 2007, CNPC had oil and gas interests in
twenty-six countries and provided petroleum engineering and tech-
nical services to forty-four countries. Overseas operations produced
60.19 million metric tons of crude oil and 5.36 billion cubic meters
of natural gas (Yuan 2007 ) in total, although the equity-adjusted vol-
ume of oil is much lower (less than 40 million metric tons).
Although what motivates CNPC to go abroad is under debate, an
important driver of China’s overseas oil policy in the past has been
the government’s concern with energy security. This concept is com-
mon ly u nd er st oo d a s “t he avai la bi li ty of en er gy at a ll ti me[s] in va ri ou s
forms, in suf cient quantities, and at affordable prices” (Xu 2004 ).
Historically – from the time the Chinese Communist Party took over
, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511784057.012
Downloaded from https://www.cambridge.org/core. ETH-Bibliothek, on 26 Jul 2017 at 13:20:34, subject to the Cambridge Core terms of use
BinBin Jiang404
in 1949 – the Chinese government has struggled to secure the sup-
ply of energy. Dependence on Soviet exports through the 1950s and
then the breakdown in Sino-Soviet relations were a stark reminder
of the need for reliable energy supplies; those concerns resurfaced as
production levels started to decline in China’s maturing oil elds due
to insuf cient levels of investment in exploration and production,
especially in the 1990s as economic reforms caused demand to soar.
The efforts have been on many fronts: alternative energy (e.g., renew-
able power) and energy ef ciency, as well as a global search for oil
resources that China could secure for itself.
Much of China’s effort to obtain oil overseas has married state-
owned CNPC with special state resources, notably development
assistance to the host country (Houser 2008 ). For example, China’s
Export-Import Bank extended loans to Angola and other African
nations during Premier Wen Jiabao’s African tour in 2006 (Faucon
2006 ).
In recent years the government has become more hesitant about
overseas investment, while CNPC has become more enthusiastic. In
fact, it is on the topic of overseas investment where we start to see
some of the tension between government and enterprise. Any major
overseas venture requires approval by the central government.
17 The
willingness of the government to offer such approvals is a subject of
great speculation by China scholars within the literature.
18 Some ana-
lysts, such as Erica Downs, claim that energy security is the main
driver for China’s “going out” strategy – by implication, the govern-
ment will be keen to offer approvals for overseas operations (Downs
2000 ). Others, such as Trevor Houser, argue that although Beijing
has encouraged overseas investment in the past, it is less enthusiastic
now due to the small effect such actions have on gaining energy secur-
ity for the country. Only about 10–20 percent of all the oil that is pro-
duced overseas actually makes it back to China (Dirks 2006 ).
19 The
crude that does make it back is also sour and heavy and thus harder
to re ne. The more pro table light, sweet crude tends to be sold in the
international market where pro ts are higher. Because re ning is a
losing business proposition in China due to the arti cially low prices
at which the government controls the downstream products, there is
little incentive for the enterprise to bring the crude back to the home
market in China.
20 How much of the different crude types CNPC
brings back for the domestic market is a point of tension between the
, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511784057.012
Downloaded from https://www.cambridge.org/core. ETH-Bibliothek, on 26 Jul 2017 at 13:20:34, subject to the Cambridge Core terms of use
CNPC (China) 405
Chinese government and CNPC. The exact proportion that is set-
tled on is a negotiation between the government and the company.
This situation, however, is in ux. The government raised the market
prices of fuel signi cantly in July 2008 (see below), which helped to
reduce the disincentive to bring valuable oil to China for sale. At the
time of writing, it is not clear what the future holds.
Our assessment is that CNPC’s continued press overseas is a
re ection of fundamental forces at work on the enterprise and not
simply the Chinese government’s effort to secure oil. In fact, gov-
ernment support for CNPC’s overseas push is likely to keep wan-
ing, not least because of the political troubles that arise in Sudan and
other places. Yet the business interest in moving overseas is strong for
CNPC. Overseas operations are more pro table (and more autono-
mous) than those within Chinese territory. And because as a state
enterprise CNPC is willing to accept a lower rate of return on its
investments, the company is able to outbid many private rms as it
competes around the world (Houser 2008 ). It regularly outbids India’s
NOC, for example, for CNPC’s pockets are much deeper, its political
assets such as development assistance are more diverse, and its tech-
nical skills more impressive (see Chapter 17 ). To be sure, development
assistance and other sweeteners from the Chinese state have helped
to open some markets, but CNPC’s business decisions are now the
dominant explanation for the push overseas. Energy security might
have started out as a main driver for China’s going out policy, but in
key places such as Kazakhstan, Venezuela, Iran, and elsewhere where
there is continuing strong Beijing involvement, it is certainly not the
main reason for why the trend has continued and grown .
3.2.3 Subsidies
Even though the industry is now organized in ways that formally
re ect the processes that prevail in much of the international oil
industry, in practice CNPC (and other Chinese NOCs) are still largely
bound by the practices of a planned economy. Inherent con icts arise
because of the dual roles that CNPC is expected to play – as both a
competitive (increasingly global) rm and an obedient state entity.
These tensions are evident in the role the rm plays as supplier of
tax revenue that the Chinese government uses, in part, to subsidize
re ners who are required to sell oil products at prices below their full
market cost. It plays a central role in the negotiation over the price
, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511784057.012
Downloaded from https://www.cambridge.org/core. ETH-Bibliothek, on 26 Jul 2017 at 13:20:34, subject to the Cambridge Core terms of use
BinBin Jiang406
of domestic oil products; those prices, in turn, determine the level of
subsidy the downstream sector will need each year to remain pro t-
able – a burden that CNPC principally bears through cross-subsidies
that are nanced through heavy taxation of its upstream arm. The
amount of tax that is paid is linked to the price of international oil
(see Table 9. 2 ), in part because the cost of subsidizing domestic oil
products rises with the price of oil. CNPC contributed RMB 198.5
billion in 2007, or 51 percent of total pro ts, and 0.01 percent of the
national GDP.
These downstream subsidies have been dif cult to reform because
they are popular with the growing urban middle class as well as rural
farmers who operate petroleum-fueled equipment. However, there
have been some moves toward fuel price and tax reforms. In June
2008, the government moved to increase retail gasoline and diesel
prices by 17–18 percent, the rst increase in eight months, following
months of spreading fuel rationing as re ners cut production to trim
deepening losses incurred by record crude costs. Tan and Wolak’s
careful analysis shows that Chinese prices are now reliably approach-
ing Western prices and the era of subsidies seems to be ending (Tan
and Wolak 2009 ).
4 Financial indicators of rm performance
There is no question that CNPC is a large company. It ranks fth on
Petroleum Intelligence Weekly ’s Top 50 Oil Companies list. During
the feverish interest in Chinese companies from 2006 to 2008 the
price of the small fraction (16 percent) of PetroChina’s shares that
Table 9.2. Petroleum industry extra pro t tax
collection rate
Crude price ($/barrel) Tax rate (percent)
40 ~ 45 20
45 ~ 50 25
50 ~ 55 30
55 ~ 60 35
60 and above 40
Source: CNPC (2007).
, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511784057.012
Downloaded from https://www.cambridge.org/core. ETH-Bibliothek, on 26 Jul 2017 at 13:20:34, subject to the Cambridge Core terms of use
CNPC (China) 407
trade publicly implied that PetroChina was the world’s largest oil
company (Forbes.com 2007).
Looking to the markets, then, CNPC seems to be a highly success-
ful company. But market valuation is hard to parse because so much
of CNPC’s management of PetroChina seems designed to manipulate
the latter’s value. Thus I look beyond the markets and attempt to
assess the company’s performance in four ways.
First, I compare CNPC and PetroChina to see if the latter’s list-
ing actually affects its performance. For many indicators, CNPC and
PetroChina appear, on paper, as quite similar companies. The big
difference is the number of employees: CNPC employs almost twice
as many people as PetroChina and therefore it has a lower revenue per
full-time laborer. Also, PetroChina has a much lower level of liquidity
compared with CNPC, which re ects that PetroChina borrows vast
sums of money from its parent company. In effect, CNPC is the nan-
cial capital of the enterprise and PetroChina is an operating company
that holds most of the pro table assets. Thus, PetroChina has a much
higher asset turnover ratio, indicating that it is almost twice as ef -
cient as CNPC in generating revenue (see Table 9.3 ). This almost cer-
tainly re ects the fact that CNPC still holds on to non-core assets that
are not nancially ef cient, while PetroChina is made up of only the
best assets from CNPC.
A second way to assess CNPC’s performance is to compare it with
a “best in class” benchmark, using ExxonMobil and examining
measures of liquidity, solvency, pro tability, and nancial ef ciency
(as de ned in Dobbins et al . 2000 ). In terms of return on equity,
ExxonMobil’s return is about double that of PetroChina and almost
three times the rate of CNPC (see Table 9.3 ). These comparisons are
dominated by oil operations, although all three of these enterprises
are increasingly investing in natural gas.
21
When we compare CNPC and PetroChina with ExxonMobil,
there are some striking differences. Most striking, rst, is that
ExxonMobil’s debt/asset ratio is double that of these Chinese com-
panies, even though ExxonMobil is widely seen in the industry as
one of the most debt-free companies (see Table 9.3 ). I interpret this
as CNPC/PetroChina keeping its formal debt at a lower rate because
it can rely on government as the ultimate backstop and there are
strong incentives to nance projects with revenues rather than debt.
22
CNPC’s close ties to government create little need to borrow money
, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511784057.012
Downloaded from https://www.cambridge.org/core. ETH-Bibliothek, on 26 Jul 2017 at 13:20:34, subject to the Cambridge Core terms of use
BinBin Jiang408
from outside sources and (unlike other impoverished NOCs that are
heavily relied upon by government for the purpose of raising funds,
e.g., KPC, PDVSA, and Pemex discussed in Chapters 8 , 10 , and 7 ,
respectively) the central government does not depend on CNPC (or
Table 9.3. Comparison of key ratios for CNPC, PetroChina, and
ExxonMobil, 2007
a
2007
CNPC
($m)
PetroChina
($m)
Exxon
Mobil ($m)
Liquidity
Current ratio 1.4 1.2 1.5
Worki ng capit al 22 , 995 4,726 27,651
Solvency
Debt/asset ratio 0.28 0.27 0.50
Equity/asset ratio 0.72 0.73 0.50
Debt/asset ratio 0.38 0.37 1.01
Pro tability
Rate of return on assets 0.12 0.18 0.23
Rate of return on equity 0.17 0.25 0.46
Operating pro t margin 0.19 0.15 0.15
Net income 16,218 19,432 40,610
Financial ef ency
Asset turnover ratio 0.63 1.22 1.61
Operating expense ratio 0.68 0.49 0.05
Depreciation expense ratio 0.003 0.052 0.031
Interest expense ratio 0.001 0.001 0.001
Net income from operations ratio 0.19 0.15 0.14
Number of employees 760,000 466,502 80,800
Revenue per full-time laborer 0.2 0.4 4.8
a Current ratio = current assets/current liabilities; indicator of short-term
debt servicing and/or cash ow capacity, extent to which current assets, when
liquidated, will cover obligations; Debt to asset ratio = Total liabilities/total
assets, proportion of total assets owned by creditors; Asset turnover ratio = Gross
revenue/total assets; how ef ciently assets generate revenue; Operating expense
ratio = (Total operating expenses – depreciation)/Gross revenue; proportion of
total revenues absorbed by operating expenses; Depreciation expense ratio =
Depreciation expense/gross revenue; Interest expense ratio = total interest/gross
revenue; Revenue per full-time laborer = Gross revenue/number of full-time
employees (Dobbins et al . 2000 ).
Source: Compiled from C NPC, PetroChina, ExxonMobil 2007 Annual Reports.
, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511784057.012
Downloaded from https://www.cambridge.org/core. ETH-Bibliothek, on 26 Jul 2017 at 13:20:34, subject to the Cambridge Core terms of use
CNPC (China) 409
PetroChina) to serve as a fundraising vehicle because the value gen-
erated from their oil production is only a tiny part of the Chinese
economy. Despite higher debt, ExxonMobil is still a more pro table
business – almost twice as pro table as CNPC. The lower return on
equity re ects that CNPC/PetroChina is willing to take on projects
that have a lower rate of return (see Table 9. 3 ).
Among the striking facts in Table 9.3 is that CNPC’s expenses are
extremely high. Despite much lower wage levels in China, CNPC has
the highest operating expense ratio (PetroChina follows close behind)
of the three enterprises examined here because CNPC is saddled with
so many non-core operations. Moreover, ExxonMobil’s revenue per
full-time laborer is about twenty times higher than CNPC and ten
times higher than PetroChina. By a very wide margin, Exxon is a
much leaner and more ef cient operation than CNPC or PetroChina.
(Comparisons with other Western companies would yield similar
results.) In general, it seems that while both CNPC and PetroChina
are nancially viable, solvent companies, they are a lot less ef cient
than the best independent rms. That outcome re ects that the goal
of these NOCs is not principally generation of pro t: for CNPC a
principal goal is employment; for PetroChina a central goal is to look
attractive for equity investors. Perhaps the reality of this has set in for
some investors as the valuation of CNPC and PetroChina has fallen
in the past year.
A third way to assess CNPC’s performance is over time – to see if
reforms have plausibly changed investment and behavior that could
alter performance. For PetroChina and CNPC, absolute income and
assets have grown since the rst available data came out in 1998 and
2001, respectively. Much more interesting, however, are the changes
in nancial ef ciency over time. The nancial ef ciency of PetroChina
has improved slightly over the last decade, but the nancial ef ciency
of CNPC has not changed much. For example, the operating expense
ratio (operating expenses/net sales) for CNPC decreased slightly after
2001 and then actually increased by 2007. This ratio shows that the
amount of money spent in operating expenses in relationship with the
volume of sales has not improved. This is often seen as an indicator of
management ef ciency because management usually has greater con-
trol over operating expenses than over revenue. PetroChina also has
always maintained a much higher operating pro t margin than CNPC.
This points to the possibility that CNPC’s role as a major employer
and vessel of the state has not changed much, while PetroChina is
, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511784057.012
Downloaded from https://www.cambridge.org/core. ETH-Bibliothek, on 26 Jul 2017 at 13:20:34, subject to the Cambridge Core terms of use
BinBin Jiang410
increasingly responding to the incentive to become more ef cient in
the eyes of investors. Notably, revenue per laborer at PetroChina has
risen nearly tenfold over the last seven years; some of that improve-
ment probably re ects better performance but most is due to increas-
ingly effective efforts to shed excess workforce (comparable data is
not available for CNPC). Much of that shedding appears to have been
to CNPC and thus efforts within CNPC to become more ef cient
have been offset by the need to absorb unwanted elements cast off by
PetroChina.
A fourth way to assess performance is to look beyond nancial
indicators at how the company manages long-term investments.
Here I focus on the company’s technological capabilities including its
investment in human capital and R&D.
Sorting through the list of CNPC’s past technological achieve-
ments, it is apparent that this rm’s expertise lies in working with
mature elds as well as building understanding of complex geo-
logical formations found in China. Figure 9.6 shows a subset of prin-
cipal innovations compiled from CNPC’s annual reports over ve
years ( 2003 –2007). Within exploration and production, it is clear
that CNPC has invested much time in understanding lithologic and
carbonate reservoirs. According to the CNPC Sichuan Petroleum
Geophysical Prospecting Company, a subsidiary of CNPC, most
of the remaining reservoirs in China are likely to be of this com-
plex geology (CNPC 2005). It has also invested heavily in polymer
and chemical ooding as a means of enhanced oil recovery (EOR).
And it is investing in techniques to exploit elds with low perme-
ability reservoirs as well as heavy oils . Historically, most of CNPC’s
investment has been onshore, but it is now concentrating some atten-
tion on offshore options. Speci cally, the company plans to build
two semi-submersible drilling platforms by 2015 and another two
by 2020 in the South China Sea (Lei 2008 ). Among its many other
areas of innovation, CNPC is heavily involved in developing high-
strength steel materials for oil and gas pipelines – which re ects that
PetroChina operates and manages all major pipelines in China. In
general, CNPC ’s technology strategy since 2003 is aimed at develop-
ing: 1) technologies that allow CNPC to better adapt to the increas-
ingly dif cult geology it must tap, and 2) technologies that allow
CNPC to expand beyond its core competency (onshore E&P) into
other activities such as re ning and offshore operations.
, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511784057.012
Downloaded from https://www.cambridge.org/core. ETH-Bibliothek, on 26 Jul 2017 at 13:20:34, subject to the Cambridge Core terms of use
Figure 9.6 Technology development at CNPC, 2003–2007.
Source: CNPC Annual Reports 2003 –2007; Guo ( 2006 ).
Area of
Innovation
Exploration
Lithologic resevoir
exploration leading to new
reserves; Improvement of
seismic data in
mountainous and desert
areas
Lithologic resevoir exploration leading to
new reserves; Geo-East V1.0 integrated
seismic data processing and
interpretation software
Lithologic resevoir exploration
leading to new reserves
Carbonate reservoir
exploration
Lithologic resevoir
exploration leading
to new reserves
20072006200520042003
Development Polymer flooding (EOR)
Improved water flooding for high water
cut oilfields; Multi well combined thermal
recovery techniques for heavy oil wells
and steam flooding; Progression
development of techniques for extremely
low permeable resevoirs; Super high
pressure gas producing pipe strings
Polymer flooding and alkai-
suractant-polymer (ASP) flooding;
ASP-foam flooding; Development
of high-pressure condensate
fields; Temperature-controlled
viscosity acidizing (TCA)
Steam assisted
gravity drainage
(SAGD)
Integration of
development
techniques
Drilling
Increase of drilling rate by
50%; Underbalance drilling
for high-yield gas flows
Increased drilling rate; Developed under-
balanced equipment and mating
technology
Increase in drilling speed; 3000
HP rig for ultra-deep wells
(9000m+)
Massive application of
horizontal drilling; Gas
drilling; China
geosteering drilling
system (CGDS-1)
12000 m drilling rig
and matching top
drive system
Logging Array induction image
logging techniques
Development of EILog100 w/ capability of
open hold well logging and perforation,
coring operation; New remote metering
system for domestic unitized logging
equipment; developed logging data
processing and interpretation integrated
software LEAD 1.0
Industrial
application of
prestack seismic
reservoir
description
Surface
Engineering
Rapid synchronous root
pass welding (pipes);
Pollution treatment of
drilling fluid
Product development of X80 high
strength steel pipes
Combined oil-gas pipeline
system; X80 high-tensile pipeline
steel
High-resolution smart
magnetic flux leakage
(MFL) detector for
pipelines
X80 spiral welded
steel pipes
Refining and
chemical
engineering
Two riser catalytic cracking,
improve product recovery
rate by 1.6%; FCC gasoline
alken-reduction series
catalyst; Flue gas turbine
expander (33,000 kw)
Expansion of use of two riser catalytic
cracking; Ionic liquid catalyzer alkylation
technique; ABS production technology
LCC-1 high production propylene
catalyst and LCC-A aids, LCC-2
catalyst
LIP-100 and LIP-200
catalysts
https://doi.org/10.1017/CBO9780511784057.012
Downloaded from https://www.cambridge.org/core. ETH-Bibliothek, on 26 Jul 2017 at 13:20:34, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms.
BinBin Jiang412
According to PetroChina’s 2008 interim annual report, the com-
pany planned to spend RMB 1,700 million ($243 million) for research
and development in 2008 and spent RMB 1,613 million ($230 mil-
lion) in 2007. This investment, about 1.2 percent of net pro t from
2007, is comparable with other IOCs.
23 (BP, for ex a mple, is e stimated
to have spent 1.3 percent of net pro t on capital expenditures related
to research and development in 2007 (BP 2007 ).)
CNPC invests in new technologies both through its subsidiaries and
through centralized research programs. The highly compartmental-
ized nature of CNPC’s subsidiaries and tight capital controls issued at
the central level by CNPC have discouraged subsidiaries from making
large technology purchases from outside rms such as eld services
companies. Thus, CNPC’s subsidiaries self-organize to invest heav-
ily in technologies that allow them to be self-reliant. At times, how-
ever, CNPC designates a subsidiary to develop a certain technology
that would be made more widely available in the company. And for
some special projects CNPC creates “heavyweight teams” that operate
under special budgetary rules and appear to be an important engine
of innovation. These teams are still initiated by subsidiaries, although
CNPC headquarters may supervise them. The general manager of the
subsidiary is the head of such a team, made up of the best engineers
of the organization. The decisions made within this team can be fast-
tracked and do not need to follow normal procedures of approval. For
example, regular R&D projects need the approval of top management
from CNPC headquarters before important decisions can be made,
but this step can be bypassed in these cases.
Most engineers within CNPC are assigned to work in one oil eld
their entire careers. As a result, certain types of geologies are very well
understood by CN PC and knowledge of these elds is institutionalized
and well documented. It is easy, then, to build upon years of data and
experience to develop new ways of developing the same eld or other
elds of simila r geolo g y. This mode of organization probably explains
why CNPC is relatively good at managing its aging elds within an
existing technological paradigm but it appears to be relatively poor at
sharing resources between subsidiaries to foster further breakthrough
innovations. For example, in Figure 9.6 , we see that almost all of the
technologies that have been developed by CNPC from 2003 onward
are re nements of existing technologies with a focus on developing
existing elds (i.e., development and re ning technologies).
, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511784057.012
Downloaded from https://www.cambridge.org/core. ETH-Bibliothek, on 26 Jul 2017 at 13:20:34, subject to the Cambridge Core terms of use
CNPC (China) 413
5 Conclusion
The story of CNPC re ects China’s complex transition from a
planned to a more market-oriented economy. CNPC is the largest
state enterprise in a sector that China’s leaders have long consid-
ered as strategically vital to the national well-being. Thus CNPC has
been the locus of some reforms – notably the reshuf ing of the oil
industry in the late 1990s to create a semi-competitive duopoly of
CNPC and Sinopec – yet those reforms have never been so dramatic
that they might fundamentally alter the mode of operation of the
industry. Central leaders perceived reforms in the 1980s that had
been intended to boost performance as going awry when regional
production units (provincial administrative bureaus, or PABs)
started behaving autonomously. Decentralization had been the goal,
but when the central government saw rapid changes and chaos from
its policies it reasserted control. CNPC re ects an incomplete tran-
sition from a planned economy to a market economy. While it is
free to emulate Western companies in its internal operations such as
formal budget ac count i ng and investment in R&D, C N PC l i ke ot her
state-owned pillar industries is also obligated to work within the
framework of a planned economy without liberalized energy prices
and markets.
My assessment is that both CNPC and PetroChina are nancially
viable institutions yet are still highly inef cient relative to the best of
the IOCs. While CNPC has technical expertise in particular types of
geologies, it is still not able to prosper in many settings that are out-
side its traditional areas of competence due to compartmentalization
of the company and expertise. The company seems to be aware of this
problem and is investing – on par with major Western companies –
in technology development to become more versatile in the future.
However, as evident in many other NOCs, the company still has not
built the capacity to meld its internal technical capabilities with the
offerings from external rms.
Whether CNPC can become a truly competitive oil company – with
operations on the world market – has much to do with the way it is
treated by the Chinese state. If the Chinese government continues to
support the trend of allowing autonomy and increasing transparency
and a role for market forces in the sector, then CNPC is likely to take
advantage of new opportunities.
, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511784057.012
Downloaded from https://www.cambridge.org/core. ETH-Bibliothek, on 26 Jul 2017 at 13:20:34, subject to the Cambridge Core terms of use
BinBin Jiang414
Notes
1 Some CNPC executives will claim that the company is in no way obliged
to meet the demands of the government and that it is a fully privatized
entity. During the course of research for this chapter, however, I have
discovered much more evidence to the contrary.
2 The term “company” is misleading. In actuality, this was a government-
run organization.
3 Sinopec underwent similar changes. It formed Sinopec Group and within
that group the China Pet roleum & Chemical Corporation (which adopted
the name “Sinopec”) were formed. Sinopec was listed with Sinopec’s
core assets.
4 At this point, CNPC was a wholly state-owned oil company that had
status as a legal person and was under the direct control of the State
Council, which is the highest organ of state power in the PRC and is the
chief administrative authority. The premier chairs this organization and
is supported by vice-premiers, state councilors, the auditor-general, and
the secretary general. Heads of each ministry, commission, state admin-
istration, bureau, of ce, and state academic research institution are also
participants in this body.
5 These indicators included enterprise pro t and tax ratio of the industry,
oil and gas unit cost change ratio (between the beginning and the end of
the year), net pro t/asset ratio, state assets value increase ratio, reserve
and production ratio, labor productivity, capital expenditure for every
billion tonnes of recoverable reserves, capital investment of production
capacity for every million tonnes, debt/asset ratio, and social contribu-
tion ratio.
6 These reforms reversed the trend of increasing autonomy of regional
enterprises. Decentralization could have been part of the solution to
achieving economic success, but the government (and CNPC) failed
to simultaneously institute incentives that would guide the enterprises
t ow ar d a co m mo n g oa l of i mp ro vi n g e c on om ic p er f or m an ce . Wh en “l e ga l
person” status was revoked, subordinate rms were no longer allowed
to make investment decisions without the approval of CNPC; nor were
these local enterprises allowed to sell shares without CNPC approval;
none was allowed to operate overseas. CNPC also asserted control in
product pricing, marketing, and procurement. (These new responsibil-
ities created huge demands for information across the dispersed CNPC
enterprise; the company built an information technology system to help
it better monitor and manage these individual rms.)
7 The reshuf ing of the oil industry required CNPC to transfer twelve
upstream enterprises engaged in E&P to Sinopec, and Sinopec transferred
, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511784057.012
Downloaded from https://www.cambridge.org/core. ETH-Bibliothek, on 26 Jul 2017 at 13:20:34, subject to the Cambridge Core terms of use
CNPC (China) 415
nineteen re ning units to CNPC. The swap in assets improved Sinopec’s
supply security while allowing CNPC to have direct access to the mar-
ket and to build a brand around re ned oil products. Integration also
led CNPC and Sinopec to share the bene ts from government R&D
investments in upstream and downstream technologies. Separation and
reintegration around two enterprises gave each the freedom to deter-
mine its own supply and transportation system for optimizing perform-
ance within the enterprise rather than relying on government plans to
achieve that goal. This freedom fostered competition between CNPC
and Sinopec by putting the two on more equal footing.
8 The core and mostly pro table assets were grouped together and listed
in the New York and Hong Kong stock exchanges under a corporation
called PetroChina. The goal of the IPO was to raise funds for future
business development and also to improve management skills within
the corporation by forcing the company to adhere to strict international
guidelines and regulations, such as those that govern accounting and
reporting of nancial and performance data. The government realized
that this would be a necessary step to take in order to compete with
multinational companies. This was also an essential part of the strategy
to recentralize the petroleum industry because external accountability
would strengthen the hand of managers at PetroChina. Nonetheless, it
was a dif cult transition for some of the regional enterprises because
previously autonomous organizations now became production units
with little control over management, investment, or production targets.
The non-core assets had an even more dif cult time after the separ-
ation because many were already on the brink of bankruptcy before
the separation – sustained only by cross-subsidy from the pro table oil
production enterprises. Subsidies were provided for these companies
for three years, until 2002, after which they were exposed to market
competition.
9 One of the most striking ways in which this organizational pattern
manifests is through looking at how new products and technologies are
developed within the rm (Guo 2006 ), where local branches and sub-
sidiaries of the company often lead the way in many technical innova-
tions. This will be explored in detail in a later section.
10 At the same time, much of the country’s economic growth arose from
hybrid companies that were partially private and par tially state-owned –
often by local governments. Such companies, which my colleagues have
also referred to as “dual rms” because they combine both political
and commercial elements, have been a major feature of the Chinese
economic landscape (Victor and Heller 2007 ). Classic state enterprises
of all sizes lagged behind these dual rms.
, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511784057.012
Downloaded from https://www.cambridge.org/core. ETH-Bibliothek, on 26 Jul 2017 at 13:20:34, subject to the Cambridge Core terms of use
BinBin Jiang416
11 We focus in this article on the 1998 reforms, but the seeds for stronger
corporate control and management were sown in the early 1990s as
part of a broader push for economic reform. In 1993, two landmark
policy decisions set a foundation for the “modern enterprise system” in
China. First, the Central Committee passed its “Decision on Certain
Questions in Establishing a Socialist Market Economy Structure,”
which called for the formation of large enterprises across sectors and
regions. This legislation was a part of a recentralization movement that
aimed at shifting decision-making power from local enterprises back in
the hands of a central authority. (Decentralization had been the watch-
word in many industries, not just oil, and the effects were widely seen
by the central government as not encouraging – especially in heavy
industries with large network interdependencies between units, such as
electric power and integrated production of oil products.) Second was
the new “Company Law,” discussed in the main text.
12 CBRC governs all banking institutions in China and works closely with
SDRC (State Development and Planning Commission) to determine the
total number of shares that should be issued for SOEs.
13 www.sasac.gov.cn/n2963340/n2963393/2965120.html .
14 PetroChina Board of Directors 2008 , www.petrochina.com.cn/Ptr/
About _PetroChina/Executive_Pro les/.
15 The special role of the party helps explain why leadership in CNPC
differs from that of IOCs and even many other NOCs. Such positions
seem to be stepping stones to gaining a position in the highest author-
ity within the Communist Party – the Central Committee. The Central
Committee has 300 members and normally appoints the Politboro of
the Communist Part of China, the primary decision-making body in
China. This would be the equivalent of using board positions at Exxon
as a pathway to the US Senate or presidency. Looking at Figure 9.5 , we
can see that ve former and current members of the top leadership in
CNPC, or 25 percent of all the people who have served in such a pos-
ition, are af liated with the Central Committee. CNPC leadership is a
governmental position, rst and foremost, rather than one that princi-
pally seeks industry expertise and shareholder stewardship.
16 Approval of large projects (exploration and development of oil elds
that produce above 2 million tons annually, gas elds that exceed 3
billion cubic meters per year, and construction of re neries with an
annual capacity larger than 5 million tons) requires direct decisions by
the State Council and NDRC (Yuan 2007 ).
17 According to “Provisions on the Examination and Approval of
Investment to Run Enterprises Abroad,” the approval of new projects
overseas (including establishment of a new enterprise, equity purchases,
, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511784057.012
Downloaded from https://www.cambridge.org/core. ETH-Bibliothek, on 26 Jul 2017 at 13:20:34, subject to the Cambridge Core terms of use
CNPC (China) 417
mergers and acquisitions, etc.) is obtained from the Ministry of
Commerce and of ciated by a certi cate of approval for overseas invest-
ment. In 2004, NDRC announced the “Veri cation and Approval of
Overseas Investment Projects Tentative Administrative Procedures”
that stated that exploration investments over $30 million must be dir-
ectly approved by NDRC. For exploration investments above $200 mil-
lion and for purchases above $50 million, both the NDRC and the State
Council must submit their approval (Yuan 2007 ).
18 See Downs ( 2004 ); Lieberthal and Herberg ( 2006 ); Xu ( 2007 ) and
S. Lewis ( 2007 ).
19 The latest estimates have been as high as 20 percent (Mikkal Herberg,
private communication).
20 Yan, K. F. Interview with author, April 2007.
21 Over 90 percent of CNPC’s revenue comes from exploration, re ning,
and retail activities. Those same core assets are concentrated within
PetroChina. Historically nearly all of these revenues have come from
petroleum, but the natural gas segment is now growing the most rapidly
at 26 percent per year as the Chinese government is actively promoting
the use of natural gas by building infrastructure, such as the West-East
pipeline that aims to bring in supplies from western China, Russia, and
Kazakhstan to energy demand centers on the eastern coast. Oil produc-
tion from E&P petroleum is growing much more slowly (annual rate of
6.8 percent) as the domestic production output from very mature oil
elds stagnates.
22 Outside of state banks – for which formal accounting of true debt lev-
els is still poor – there are few mechanisms for debt nancing of large
oil projects inside China. In some countries the government has trans-
formed the NOC for purposes of raising debt on commercial markets –
notably Mexico’s Pemex (see Chapter 7 ) – but PetroChina’s creation
was not inspired with this purpose but rather for the purpose of encour-
aging equity investment in China’s oil sector.
23 While numbers are not available for all of CNPC, the level of funding
going to R&D is not likely to change since PetroChina accounts for
most (if not all) of the pro t.
, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511784057.012
Downloaded from https://www.cambridge.org/core. ETH-Bibliothek, on 26 Jul 2017 at 13:20:34, subject to the Cambridge Core terms of use
