Dealing with Non-Performing Loans: Political Constraints and Financial Policies in China

Author(s): Victor Shih

Source: The China Quarterly, No. 180 (Dec., 2004), pp. 922-944
Published by: Cambridge University Press on behalf of the School of Oriental and African
Studies
Stable URL: http://www.jstor.org/stable/20192411
Accessed: 14-12-2017 12:01 UTC

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide
range of content in a trusted digital archive. We use information technology and tools to increase productivity and
facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected].

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
http://about.jstor.org/terms

Cambridge University Press, School of Oriental and African Studies are collaborating
with JSTOR to digitize, preserve and extend access to The China Quarterly

This content downloaded from 111.68.96.110 on Thu, 14 Dec 2017 12:01:26 UTC
All use subject to http://about.jstor.org/terms
Dealing with Non-Performing Loans: Political
Constraints and Financial Policies in China*
Victor Shih

Abstract The stagnation in banking sector reform is puzzling given the Chinese
leadership's seeming resolve to reform other sectors of the economy. This article
develops a political explanation of why reform oriented central bureaucrats have
centralized financial power without liberalizing the banking sector. The starting point
of this explanation is that top bureaucrats value political survival as much as other
members of the Politburo. Thus, they make policies based on both political and
economic considerations. This framework is tested on three cases related to China's
non-performing loan (NPL) problem: the politicization of the NPL problem, policies
designed to slow the creation of NPLs and policies aimed at decreasing the pool of
NPLs. The findings strongly suggest that political considerations play a large role in
shaping financial policies in China.

Reform is crossing the river by feeling for stones (moshi guohe). Yet, the
stones that Chinese leaders feel for are not sound economic policies but
political survival. The current that threatens to topple leaders is not policy
failure but political attack. To stand firm on the stones, leaders do not
hesitate to move sideways or even step back away from the current. The
objective is to remain standing, not necessarily to cross the river.
From a pure policy standpoint, the path of Chinese financial reform
appears haphazard and does not unfold in a linear fashion towards
marketization. Seen through the lenses of power politics, however, it was
shaped by a series of short-term political calculations aimed at strength
ening the power of individual leaders. This article demonstrates this logic
through the case of Premier Zhu Rongji's policies towards the estimated
3.3 trillion RMB (US$400 billion) in non-performing loans (NPLs).1
The methodology employed is one of policy analysis similar to the one
developed by Skocpol.2 For all policies, I first present policy alternatives
obtained through over 70 interviews and numerous internal documents.
Examining a set of serious policy alternatives addresses the problem of
selecting on the dependent variable prevalent in policy analysis.3 Again
relying on interviews and internal documents, I pinpoint the political and

* I would like to thank Roderick MacFarquhar, Elizabeth Perry, Regina Abrami, Mary
Gallagher, William Hurst, Kun-chin Lin and Matt Rudolph for reading and commenting on
preliminary drafts of this paper. I am responsible for all the errors.
1. Terry Chan et al, "US$400 billion needed to recapitalize China's banks," Standard &
Poor's CreditWire, No. 4 (June 2001). This NPL figure represents the approximate mean of
estimated NPL ratios by various groups, which range between 30 and 70% of all loans
outstanding.
2. Theda Skocpol, Protecting Soldiers and Mothers: The Political Origins of Social Policy
in the United States (Cambridge, MA: Harvard University Press, 1992).
3. By serious policy alternatives, I mean those proposed by a State Council organ, which
should have some notion of which have a reasonable chance of succeeding.

? The China Quarterly, 2004

This content downloaded from 111.68.96.110 on Thu, 14 Dec 2017 12:01:26 UTC
All use subject to http://about.jstor.org/terms
 Dealing with Non-Performing Loans

distributive motives behind the chosen policies. These are compared with
the policy predictions of the alternative explanations to gauge the relative
explanatory power of each framework. Finally, I examine the net effects
of the policies and compare them with the predictions of various alterna
tive explanatory frameworks to determine which one can most accurately
predict policy outcomes. The policies I discuss fall into three general
areas: the decision to politicize the NPL problem, policies aimed at
stemming the creation of new NPLs and measures to digest the existing
pool of NPLs.

Alternative Explanations
Previous explanations of economic reform in China centre on the
diffusion of market-oriented ideas, the erosion of communist ideology
and the rise of determined technocrats. According to the ideational
framework, because Chinese leaders had an intrinsic desire to reform,
they were willing to experiment with a wide range of reforms.4 Set-backs
stemmed from incomplete information on how to implement a particular
reform, not from entrenched interests that blocked it. With the diffusion
of ideas on the market economy, "mistakes" should decrease over time.
In a slightly different version of the ideational model, set-backs in reform
were primarily caused by the resistance of ideological conservatives in
the Party.5 According to this framework, reform should proceed smoothly
after the ideologues' passing.
Shirk introduces a politically sophisticated version of China's reform
path. Still assuming that Chinese leaders, especially Deng Xiaoping,
preferred reform, she argues that reformers divided and conquered the
anti-reform coalition by offering a subset of status quo beneficiaries
selective benefits via "particularistic contracts."6 Thus, inconsistencies in
the reform path were caused by both incomplete information and the
costly process of building pro-reform coalitions. Although Shirk envis
ages a non-linear movement towards reform, the zigzagging path would
eventually turn China into a fully market economy as those in favour of
the market economy overwhelm the anti-reformers.7 Pei adds to Shirk's
model by pointing out that reformers often took advantage of economic
crises to carry out important reforms.8
Although the above frameworks come from different theoretical tradi
tions, they all predict a general movement towards the market and place
enormous faith on the spread of market oriented ideas or the rise of

4. Barry Naughton, Growing out of the Plan: Chinese Economic Reform 1978-1993 (New
York: Cambridge University Press, 1996).
5. See, for example, Richard Baum, Burying Mao: Chinese Politics in the Age of Deng
Xiaoping (Princeton: Princeton University Press, 1994), Joseph Fewsmith, Dilemmas of
Reform in China: Political Conflict and Economic Debate (Armonk, NY: M.E. Sharpe, 1994).
6. Susan Shirk, The Political Logic of Economic Reform in China (Berkeley: The
University of California Press, 1993).
7. Ibid. p. 344.
8. Minxin Pei, "The political economy of banking reforms in China, 1993-1997," Journal
of Contemporary China, Vol. 7, No. 18 (1998).

This content downloaded from 111.68.96.110 on Thu, 14 Dec 2017 12:01:26 UTC
All use subject to http://about.jstor.org/terms
The China Quarterly

reform oriented elite to implement reform. These explanations have been

extremely powerful in explaining the giant sea-change that has taken
place in many sectors of the Chinese economy. Yet they have a harder
time explaining the chronic lack or slowness of reform in certain sectors.
According to these explanations, favourable conditions for fundamental
financial reforms were all present at the beginning of the Zhu administra
tion. The Asian financial crisis had placed unprecedented power in the
hands of the new Premier. Zhu and his technocratic followers were
heralded as tough reformers by everyone from fund managers to the Wall
Street Journal.9 On the ideological front, the secretary general Jiang
Zemin legitimized private industries at the 15th Party Congress.10 Fur
thermore, China's financial elite had by then accepted the superiority of
the market economy.11 Observers of China therefore had great expecta
tions of what the Zhu administration could accomplish.
On the surface, the broad range of changes seems to confirm these
optimistic predictions. The Big Four state banks centralized their manage
ment, drastically decreasing administrative intervention from the local
government.12 They also adopted independent credit approval, as well as
the international five-category loan evaluation. As a result of these
efforts, the Big Four state banks certainly look very different today from
how they were at the beginning of the Zhu administration.
Despite these changes, when one examines the distribution of capital in
China, two underlying indicators remain virtually unchanged. First, state
banks, including both state commercial banks and the policy banks, still
dominate the financial sector. As seen in Figure 1, state bank share of
lending and deposits remains at essentially the same level. The seeming
decline in state bank share in 2001 was the product of a new statistical
grouping that moved the Bank of Communication and the China Inter
national Trade and Investment Corporation (OTIC) Bank from among
the state banks to the joint-stock banks, without any actual changes in
their ownership structure. Secondly, among state banks, the share of loans
to the non-state sector remains at a dismally low level. As seen in Figure
2, the Zhu administration saw only tepid growth in state bank lending to
the non-state sector, which hovered around the 10 per cent mark. What
went to the non-state sector, even including leakage from the state sector

9. Kathy Chen, "China's economy chief plans to speed financial reform - Zhu Rongji
emerges as leading candidate to take prime minister post," The Wall Street Journal, 23
February 1998, Nisid Hajari, "Zhu's leap forward when China's maverick economic czar
takes over the government, his open, gunslinging style is certain to change the way the country
operates," Time International, 16 March 1998.
10. Jiang Zemin, "Gaoju Deng Xiaoping lilun weida qizhi, ba jianshe you Zhongguo tese
shehui zhuyi shiye quanmian tuixiang ershiyi shiji" ("Raise high the great banner of Deng
Xiaoping theory and comprehensively push forth the task of building socialism with Chinese
characteristics towards the 21st century"), Renmin ribao (People's Daily), 12 September
1997.
11. With the exception of one veteran banker, everyone I interviewed, including central
government officials, local officials and bankers, agreed that the market economy was superior
to any other economic system.
12. The Big Four state banks are the Industrial and Commerce Bank of China, the Bank
of China, the Construction Bank of China and the Agricultural Bank of China.

This content downloaded from 111.68.96.110 on Thu, 14 Dec 2017 12:01:26 UTC
All use subject to http://about.jstor.org/terms
 Dealing with Non-Performing Loans 925

o share of SB loans a share of SB deposit

.4

.2
1997 1998 1999 2000 2001 2002

Figure 1 : Share of State Bank Lending and Deposit to Total Lending an

Deposit, 1997-2001 (quarterly)
Note:
1 = 100%.
Sources:
Division of Statistics of the People's Bank of China, Zhongguo jinrong tongji: 1997-1999
(China Financial Statistics: 1997-1999) (hereafter ZJT) (Beijing: China Financial Publisher,
2000), National Bureau of Statistics, China Monthly Economic Indicators 2001: 3 (March)
(Beijing: China Statistical Press, 2001), People's Bank of China, Baogao yu tongji shuju
(Reports and Database) (People's Bank of China, 2003 [cited 24 June 2003]); available from
http://www.pbc.gov.cn/baogaoyutongjishuju/.

to the non-state sector, in all likelihood remained well below 50 per

cent.13
It is also unclear whether the Zhu administration made significant
progress on the non-performing loans issue, the problem that motivated
the new financial policies in the first place. It remains difficult for
outsiders to discern whether the total volume of NPLs has decreased
significantly and whether the creation of new NPLs has slowed down
significantly. After the transfer of NPLs to asset management companies
(discussed below), the government repeatedly revised the remaining NPL
ratio. In 1999, the People's Bank of China governor, Dai Xianglong,
announced that NPL ratio was at 20 per cent, which meant that the

13. According to official PBOC statistics, some 20% of all loans outstanding as of
September 2003 went to the non-state sector. However, many loans to state entities end up
in private hands. According to a 1999 internal report published by the PBOC, the financial
system as a whole allocated around 42% of resources into the non-state sector. However, the
report states that allocation to the non-state sector by the Big Four state banks remains
extremely low. See Project Team at the Research Bureau of the PBOC, "Guanyu jinrong jigou
dui feiguoyou jingji daikuan qingkuang de diaocha baogao" ("Research report on the situation
of financial institutions lending to non-state economy"), Jinrong tongji yufenxi {Financial
Statistics and Analysis), No. 9 (1999).

This content downloaded from 111.68.96.110 on Thu, 14 Dec 2017 12:01:26 UTC
All use subject to http://about.jstor.org/terms
926 The China Quarterly

1997 1998 1999 2000 2001 2002

series

Figure 2: Share of State Bank Lending to the Non-State Sector,

1996-2001 (quarterly)
Note:
1 - 100%.
Source:
Division of Statistics of the People's Bank of China, ZJT, National Bureau of Statistics,
China Monthly Economic Indicators 2001: 3 (March).

transfer of NPLs to asset management companies basically would have

solved the problem.14 In July 2001, the Bank of China first broke rank
with the official line and admitted that its NPL ratio was 28.78 per cent,
even after the transfer of NPLs.15 Dai further revealed in November 2001
that NPLs still totalled 26.6 per cent of loans outstanding at the Big Four
state banks.16 At the beginning of 2002, the People's Bank of China
(PBOC) announced for the first time that NPLs were declining. However,
during a March 2002 meeting with international investors, Dai hinted that
the NPL ratio might be as high as 30 per cent if calculated by Western
accounting standards.17 The constant revisions by the PBOC partially

14. Lorien Holland, "Cleaning house: China takes aim at state banks' bad debts," Far
Eastern Economic Review, 6 May 1999.
15. Peggy Sito, "Bank of China strategy sees bad-loan ratio drop by 1.77 percentage
points," SCMP, 9 July 2001.
16. Lester J. Gesteland, "Bad debt remains a problem for banking sector," ChinaOnline
News, 1 November 2001.
17. In Western financial accounting, loans are classified into five categories depending on
the financial health and future business prospects of the borrower. Loans in China are divided
into four categories: normal (zhengchang), overdue (yuqi), congested (daizhi) and bad
(daizhang), based on the time period for which loans and interest payments are overdue.

This content downloaded from 111.68.96.110 on Thu, 14 Dec 2017 12:01:26 UTC
All use subject to http://about.jstor.org/terms
 Dealing with Non-Performing Loans

reflect the government's own uncertainty about the actual NPL level,
especially during the transition towards the new five-category system to
classify loans.
Bankers and government officials also admitted that there is no telling
whether the NPL ratio will once again take off in the next few years.18 As
discussed below, the policy environment created after 1998 encouraged
banks to provide long-term loans to fixed-asset projects, which can
remain solvent in the short run but are potentially very risky in the long
run. For example, a 1997 State Development and Planning Commission
(SDPC) internal survey found that only 5 per cent of state sponsored
construction projects could completely repay interest and principal. A full
20 per cent of these "key point" projects could repay neither.19 More
recently, a government researcher revealed that the drive to maintain 7
per cent growth had forced the government and state banks to support
many second, third and even fourth rate projects with highly questionable
qualifications.20 Should a substantial portion of these long-term loans
become non-performing in the future, the government will have to engage
in even more deficit spending.
Two compelling alternative explanations remain to account for the
stagnation in financial reform. The first is an extension of Shirk's
anti-reform coalition framework while the second is an explanation of
bureaucratic capture. According to rent-seeking theory, stagnation in
reform is caused by the opposition of rent-seekers, or small groups that
benefit from the status quo.21 Thus, the lack of substantial financial
reform during the Zhu administration might have stemmed from oppo
sition by powerful anti-reform lobbies. In contrast to this portrayal, the
State Council successfully implemented reform that substantially weak
ened a politically powerful group: provincial government officials. Finan
cial centralization in 1998 deprived local officials of access to cheap
credit from banks, drastically increasing their dependence on the central
bureaucracy for funds. Under Shirk's framework that reformers played
one group of rent-seekers against another, one would expect reformers in
the State Council to eradicate the rent-seeking opportunities of another
group after the defeat of the provincial officials.22
In reality, after financial control was taken away from the provinces,
the Zhu administration left the power in the central bureaucracy, which
poured funds into large SOEs and state projects. Suggestions of liberaliz
ing interest rates and legalizing private banks, which would have given

18. Interviews in Beijing: 10 October 2000, 14 May 2001, 23 June 2001.

19. "Diaocha biaoming jinyou 5% zhongdian jianshe xiangmu ke huanben huanxi"
("Study shows that only 5% of key-point construction projects can still repay interest and
principal)," Qiye juece cankao {Enterprise Decision-Making Reference), No. 124 (1997).
20. Tongsan Wang, "Predicting China's economic future and the national economic policy
moving forward: 2002," paper presented at the Chinese Academy of Social Science visit to
the Center for Business and Government at the John F. Kennedy School of Government,
Cambridge, MA, 7 December 2001.
21. Mancur Olson, Jr, The Logic of Collective Action (Cambridge, MA: Harvard
University Press, 1965).
22. Shirk, The Political Logic of Economic Reform.

This content downloaded from 111.68.96.110 on Thu, 14 Dec 2017 12:01:26 UTC
All use subject to http://about.jstor.org/terms
The China Quarterly

the private sector more control over financial resources, were repeatedly
vetoed by Zhu. The stagnation of reform was also not a product of Zhu's
weakening power. Although Fewsmith points out that Zhu had lost much
influence after his failure to reach an accord with the United States on the
World Trade Organization (WTO) issue, he regained substantial power
with the prosecution of the Yuanhua case.23 Most interviewees agreed
that Zhu had regained his status as "boss" (laoban) in the financial sector
by late 2000.
Bureaucratic capture, which argues that bureaucratic interests dictate
the implementation of financial policies, provides another compelling
alternative explanation.24 The centralization efforts in 1998 indeed
brought an increase in power to the State Economic and Trade Com
mission (SETC), the State Development and Planning Commission
(SDPC), the PBOC and the Ministry of Finance (MOF). Yet Zhu's
policies also created enormous problems for some of these agencies in the
medium and long term. For example, officials in the MOF well knew that
transferring non-performing loans from banks to asset management com
panies would create enormous fiscal pressure on the MOF in the future.
Despite the potential problem, top officials eagerly supported this policy
because it solved an enormous problem without jeopardizing the short
term budgetary balance. Overall, when bureaucratic preferences were at
odds with the promotion imperatives dominating top officials in the
various ministries, the desire for more power often carried the day.

Politicians in the State Council

"No one cares what happens in the next administration" (meiyou ren
guan xiayijie de shi), one interviewee casually remarked, as if this was
the most obvious fact.25 One often associates this mentality with political
incentives in a democracy because officials in authoritarian political
systems are insulated from popular pressure and thus have longer time
horizons.26 Nevertheless, many researchers and bankers I interviewed
expressed their cynicism and frustration with the short-sighted financial
policies adopted by the central government.27 From a theoretical perspec
tive, the financial policy pattern during the Zhu administration would be
much more understandable if it is assumed that the Premier's primary
objective was to remain in power.
Holding on to power was by no means an easy task. As Fewsmith
notes, Zhu was nearly ousted after the failure to reach an accord on the

23. Joseph Fewsmith, China since Tiananmen: The Politics of Transition (Cambridge:
Cambridge University Press, 2001), p. 225.
24. David M. Bachman, Bureaucracy, Economy, and Leadership in China: The
Institutional Origins of the Great Leap Forward (Cambridge: Cambridge University Press,
1991).
25. Interview in Beijing: 24 November 2000.
26. Yasheng Huang, Inflation and Investment Controls in China: The Political Economy
of Central-Local Relations During the Reform Era (New York: Cambridge University Press,
1996), p. 326.
27. Interviews in Beijing: 24 November 2000, 3 January 2001, 14 May 2001.

This content downloaded from 111.68.96.110 on Thu, 14 Dec 2017 12:01:26 UTC
All use subject to http://about.jstor.org/terms
 Dealing with Non-Performing Loans

WTO with the United States and after the Belgrade bombing in 1999.28
In the spring of 2001, the policy research centre of the Central Com
mittee, an organization reputedly under the control of Jiang's prot?g?
Zeng Qinghong, sent out a request to all government research organiza
tions to evaluate economic performance during the Zhu administration.
According to interviewees, this kind of request, particularly from the
Central Committee, was highly unusual and constituted nothing short of
a challenge to the Premier himself.29 More recently, the removal of
Construction Bank president Wang Xuebing, a long-time Zhu prot?g?,
might have been an attempt to undermine his authority.30
Even if Zhu had intended to be a one-term Premier all along, he had
good reasons to maintain his political power until the end of his tenure.
According to the rich literature on factional politics, a top Chinese leader
needed to maintain his influence up to the end of his life to advance any
personal or policy objectives.31 As Huang puts it succinctly, "decisions
are made according to the vision of those who have prevailed in the
power struggles rather than through the due process ...,"32 Thus, if Zhu
wanted to maintain any measure of political influence, he would have to
continue his political life beyond his formal retirement. While his post-re
tirement activities seem more muted than that of the other leaders in his
cohort, it would be mistaken to think that he has no role in the current
policy process, particularly in matters of promotion. Thus, his action in
office did not just reflect his desire to establish a glowing historical
legacy, but also his wish for some degree of post-retirement political
authority.
What could he do in office to maximize his political influence both
during and after his tenure? Two useful strategies are consolidating policy
power and making oneself the indispensable problem-solver. These
complement each other and serve as both an insurance against unexpected
political attacks and an investment towards maintaining post-retirement
influence. Consolidating financial power to the central level would first
allow the Premier to use the distribution of financial resources as a
political bargaining chip at the Politburo level, building a support co
alition through disbursing funds to the pet projects of other leaders. With

28. Fewsmith, China since Tiananmen, p. 211.

29. Interviews in Beijing: 19 May 2001, 26 June 2001.
30. It is probable that Wang's removal is related to the power struggle between Premier
Zhu Rongji and his predecessor Li Peng. Wang had been head of the Bank of China branch
in New York. Although the Chinese government had discovered irregularities as early as 2000,
Wang was merely transferred to the Construction Bank and remained in power until the US
accusation. His removal was widely seen as an embarrassment to Zhu. See Hu Shuli and Gu
Wei, "Zhongguo yinhang Niuyue fenhang fasheng le shenme? 'Bu anquan' he 'bu kekao' de
xingwei yanxu le jin 10 nian" ("What has happened to the BOC New York branch? Unsafe
and unreliable practices that lasted nearly 10 years") Caijing, 5 February 2002.
31. Lowell Dittmer, "Chinese informal politics," China Journal, No. 34 (1995), Andrew
Nathan, "A factionalism model for CCP politics," The China Quarterly, No. 53 (1973),
Andrew Nathan and Kellee Tsai, "Factionalism: a new institutionist restatement," China
Journal, No. 34 (1995).
32. Jing Huang, Factionalism in Chinese Communist Politics (New York: Cambridge
University Press, 2000), p. 5.

This content downloaded from 111.68.96.110 on Thu, 14 Dec 2017 12:01:26 UTC
All use subject to http://about.jstor.org/terms
The China Quarterly

total control over financial resources, Zhu could bargain either for
political support in the event of a setback, such as the failed WTO
negotiations, or for the promotion of a close prot?g?, which would
increase his post-retirement influence. One high level interviewee point
edly stated, "Zhu values the power to distribute financial resources
through loans and stock-listings, and he has used the power as chips in
the political game."33 Secondly, firm control over the financial sector
allowed the Premier and his followers to solve pressing problems for the
regime and thereby accumulate "administrative accomplishments"
(zhengji).
As the post-reform era experience has shown, the perceived ability to
solve problems is a sure tactic to gain promotion and retain power. Zhu
himself rose through the bureaucratic ranks by repressing inflation in
1993 and by solving the triangular debt problem in the early 1990s.
During his administration, Zhu and his prot?g?s gave the appearance of
tackling a series of problems, ranging from chronic state-owned enter
prise (SOE) indebtedness to the decline of rural income. Delegating
financial authority to his prot?g?s in various agencies allowed them to
solve problems, which increased their chances of promotion. Two of
Zhu's core followers, Zhou Xiaochuan and Wang Qishan, especially
benefited from accumulating administrative merits during the Zhu Rongji
administration. Zhou Xiaochuan's tough exercise of regulatory power at
the Central Securities Regulatory Commission (CSRC) earned him a
lateral promotion to head the PBOC, while Wang Qishan's role in curing
"financial chaos" in Guangdong yielded a high profile promotion to the
mayoral post in Beijing. This ultimately bolstered the Premier's influence
both during and after his tenure.
Because central bureaucrats have a strong incentive to deal with
politically pressing issues, they tend to ignore longer-term structural
problems. Moreover, because their administrative accomplishments are
evaluated by short-term results, they tend to concoct short-term solutions.
The Premier's Politburo colleagues do not care what happens five years
down the road either. In fact, interviewees pointed out that everyone from
the Premier down had a strong disincentive to carry out fundamental
reforms that could potentially jeopardize short-term stability or prevent
solutions to more pressing issues.34 As the examples of Hu Yaobang and
Zhao Zhiyang demonstrate, signs of instability and chaos provided one's
political enemies the perfect excuse to launch a political attack. For
ministers, reforms that went out of control could jeopardize their chance
of promotion.
Given the incentives facing central bureaucrats, what policies can the
State Council be expected to adopt with regard to NPLs? First, it could
be expected to politicize policy problems in order to enhance central

33. Interview in the US: 2 May 2002. The exact location is not disclosed to protect the
interviewee's identity.
34. Interviews in Beijing: 10 October 2000,14 May 2001,23 June 2001; in the US: 2 May
2002.

This content downloaded from 111.68.96.110 on Thu, 14 Dec 2017 12:01:26 UTC
All use subject to http://about.jstor.org/terms
 Dealing with Non-Performing Loans

power. After the politicization of a financial problem, one would expect

the top bureaucrats to adopt policies that maximize central power,
maximize the apparent improvement of the situation without jeopardizing
their ability to resolve other pressing issues, and minimize the short-term
cost of solving the problem with little regard to the long-term conse
quences. In hindsight, these are precisely the policies that the Zhu
administration adopted toward the NPL problem.

Crisis and Command: Politicizing NPL and the Rise of Zhu

As the 15th Party Congress confirmed the transition of power from

Deng Xiaoping to Jiang Zemin in autumn of 1997, the most dynamic
economies neighbouring China were toppling one after another. At first,
it did not seem to affect China. After all, China did not have currency
convertibility and could not fall victim to massive capital flight. Less than
a week after the 15th Party Congress at the First Plenum, however, Party
secretary general Jiang Zemin declared "the prevention and resolution of
financial risk is an important and urgent task in our national economic
work."35 Jiang was not just paying lip service. Bank officials uniformly
attributed major changes in bank policies in 1998 to the shocking
demonstration effect of economic collapse followed by political chaos.36
While the Politburo had reasonable cause for alarm, the new Premier also
highlighted the issue to obtain banking centralization, finally.
Even before the Asian financial crisis, a small group of government
experts had realized that years of treating banks as another government
treasury had produced mounting NPLs and liquidity problems in the
banking system. In 1995, for example, an expert at the Central Party
School estimated that the NPL level had reached between 25 and 50 per
cent of all loans outstanding, or between 21 and 43 per cent of GDP at
that time.37 Yet, the issue had only been a topic of discussion among
experts rather than a top priority for the regime.38 In 1996, the PBOC
finally issued the "rules on lending" (daikuan tongze), which systemati
cally addressed the questions of loan quality, loan classification and NPLs

35. Zhu Rongji, "Shenhua jinrong gaige; fangfan jinrong fengxian; kaichuang jinrong
gongzuo xin jumian" ("Deepening financial reform; prevent financial risks; create a new phase
for financial work"), in Document Research Center of the Chinese Communist Party Central
Committee (ed.), Xin shiqi jingji tizhi gaige zhongyao wenxian xuanbian {A Selection of
Important Documents for Economic Structural Reform in the New Period, hereafter Xin shiqi)
(Beijing: Central Document Publisher, 1998). Zhu quotes Jiang in his own speech.
36. Interviews in Beijing: 9 October 2000, 10 October 2000, 13 December 2000, 27
October 2000, 15 April 2001.
37. Zhou Tianyong, "Yinhang daizhang he huaizhang shi daozhi shehui dongdang de
youhuan" ("Bank overdue loans and bad loans are worries that cause social turmoil"), Jinrong
cankao {Financial Reference, hereafter JQ, No. 1 (1995).
38. Bi Zhiyong, "Guanyu zai qiye zhaiwu chongzu zhong ruhe weihu yinhang zhaiquan
de tantao" ("A discussion on how to preserve bank rights in the process of enterprise
restructuring"), Jinrong yanjiu baogao {Financial Research Report, hereafter JYB), No. 43
(1997).

This content downloaded from 111.68.96.110 on Thu, 14 Dec 2017 12:01:26 UTC
All use subject to http://about.jstor.org/terms
The China Quarterly

for the first time.39 Nevertheless, the political leadership did not raise
these issues at all during the Central Economic Work Conference at the
end of 1996.40 Although recognized as one of the many problems
confronting the regime, the NPL issue was clearly not at the top of the
elite agenda.
To deal with the worsening NPL problem, a series of measures were
proposed by State Council technocrats. One group of researchers sug
gested that the central government set up a commission for co-ordinating
investment and lending to ensure the efficiency of fixed asset invest
ment.41 Another group suggested replacing the credit plan with a national
system to monitor the asset-to-liability ratio of all financial institutions
around the country.42 Finally, a PBOC researcher urged the government
simply to forgive the interest payment of a substantial portion of the
NPLs without provisioning in order to "enliven" (gaohuo) loans.43 No
proposal seriously considered the total centralization of the banking
system.44
Before these policies could be tested, the Asian financial crisis elevated
the NPL issue from an accounting annoyance to the top of the regime's
agenda. Although the crisis had been alarming, Zhu Rongji, who was due
to begin his tenure as Premier, and State Council technocrats also
magnified it to increase the centre's financial discretion. Just as the crisis
occurred, a research team at the headquarters of the Industrial and
Commercial Bank of China had been preparing a research report on NPLs
and local redundant investment. The report first circulated internally
among government officials in mid-1997 before the full development of
the crisis.45 However, its political significance quickly increased as Asian
economies collapsed one after another. Its troubling findings of rampant
redundant investment, non-performing loans and banks with chronic
liquidity problems gave central bureaucrats the perfect device to connect

39. People's Bank of China, "Daikuan tongze" ("Rules on lending"), in The Office of Bank
of China et al. (eds.), 1996 Nian jinrong guizhang zhidu xuanbian {Financial Regulations and
Systems for 1996) (Beijing: Finance Publisher, 1997).
40. Luo Guanxing, Zhao Ran and He Jiazheng, "Zhongyang jingji gongzuo huiyi zai
Beijing zhaokai (1996)" ('The Central Economic Work Conference convenes in Beijing
(1996)"), Renmin ribao, 24 November 1996.
41. ICBC Research Group on Problems of Enterprise Bankruptcy, "Guanyu jiejue chongfu
jianshe youguan wenti de diaocha baogao" ("A research report concerning solving the
problem of redundant construction"), Qiushi {Seeking Truth), No. 3 (1998).
42. Xie Hangsheng, "Guanyu quxiao xindai guimo guanli de xianshi tiaojian yu shishi
buju" ("The conditions and steps in implementing the termination of management over
lending quantity)," JYB, No. 10 (1997).
43. See Zhang Qian, "Woguo buliang daikuan de zhuangui tezheng ji 'mianxi huoben'
panhuo de sheji" ("The transitional characteristics of our country's non-performing loans and
designs to enliven them through 'recovering principal through cancelling interest"'), JYB, No.
30 (1997).
44. Analysts from outside China had made the proposal to centralize the banking system.
For example, see Jun Ma, "Monetary management and intergovernmental relations in China,"
World Development, Vol. 24, No. 1 (1996).
45. Zhan Xiangyang, Lun Zhongguo buliang zhaiquan zhaiwu de huajie {On the
Dissolution of Bad Debt and Bad Debt Obligations in China) (Beijing: China Financial
Publisher, 2000), p. 262.

This content downloaded from 111.68.96.110 on Thu, 14 Dec 2017 12:01:26 UTC
All use subject to http://about.jstor.org/terms
 Dealing with Non-Performing Loans

what was occurring in the rest of Asia with what could happen in China.46
Moreover, this report blames China's NPL problem primarily on the local
government and on bad industrial policies made by Zhu's predecessor, Li
Peng, although Zhu himself had been actively involved in financial
policies during the Li Peng administration.
To reinforce the crisis mentality, a flood of reports from various State
Council research organizations followed to tie the crisis with China's
NPL problem.47 Under this barrage of information, the Politburo decided
during the First Plenum to give Zhu enormous discretion to change the
financial system. In October 1997, the Central Committee and the State
Council jointly issued the "Notice concerning deepening financial reform,
rectifying financial order, and preventing financial risk" (Guanyu shenhua
jinrong gaige, zhengdun jinrong zhixu, fangfan jinrong fengxian de
tongzhi). In one fell swoop, local branches of the state banks and of the
PBOC were removed from the local Party committees' jurisdiction and
placed under the newly formed Central Finance Work Committee. Hun
dreds of locally controlled trust and investment companies and under
ground banks were closed.48
To ensure compliance with these stunning announcements, Zhu, who
was due to become Premier in the spring of 1998, chaired an emergency
Central Finance Conference in November 1997 with the main theme of
"preventing financial risk" {fangfan jinrong fengxian).49 In his main
remarks, Zhu first reminded his audience of central and local government
officials that they were in the midst of a putative national crisis, which
required unity and obedience to the centre.50 This was to generate a crisis
mentality, whereas previous financial crises in Latin America had merely
been discussed among academics and bureaucrats.51
Provincial officials were clearly outraged by this sudden surge of
central power. Zhu had to offer words of comfort to the disgruntled local
leaders: "(the centralization of monetary authority) does not mean that we
do not trust local officials. It is just that the centre and localities should
each divide up our labour accordingly."52 Besides these words, however,
Zhu also offered two concessions to the provinces to stem their anger.
First, he allowed the provinces to charter more local banks to help local
development.53 This was merely a token gesture because Zhu had just
closed down hundreds of local financial institutions and deprived local
government control of the state banks. The second concession was a

46. ICBC Research Group, "A research report."

47. See, for example, Bureau of Economic Prediction of the State Information Center,
"Yao yifa zhili jinrong" ("We should rectify finance according to law"), Caijing kuaixun
{Headlines of Finance and Economics), No. 18 (1998), Guangdong Branch of the PBOC,
"Dongnanya jinrong weiji dui woguo jinrongye fazhan de qishi" ("The lesson of the
South-east Asian financial crisis on our country's financial development"), JYB, No. 7 (1998).
48. Zhu Rongji, "Deepening financial reform."
49. The routine Central Financial Conference had already taken place in January 1997.
50. Zhu Rongji, "Deepening financial reform."
51. Interviews in Beijing: 8 October 2000, 10 October 2000, 13 December 2000.
52. Zhu Rongji, "Deepening financial reform."
53. Ibid. Interviews in Beijing: 20 March 2001, 14 May 2001.

This content downloaded from 111.68.96.110 on Thu, 14 Dec 2017 12:01:26 UTC
All use subject to http://about.jstor.org/terms
The China Quarterly

promise to use the new central power to fund the development of western
provinces. At the December 1997 Central Economic Conference, Zhu
promised that "... the economic development of the central and western
areas will not be affected."54 By doing this Zhu also made use of his
new-found power to solve the perceived problem of regional imbalance
for the regime.
The centralization of the financial system was further confirmed when
the Industrial and Commercial Bank of China report was published in
Qiushi (Seeking Truth), the official publication of the Central Committee,
days after Zhu had become Premier.55 This left no doubt about the new
banking centralization policies. The article's condemnation of industrial
policies under the Li Peng administration also demarcated the new
administration from the previous one, although both pursued similar
policies in bailing out large SOEs.
The Asian financial crisis should have alarmed the Chinese leadership
because it showed that state-led growth might come at the price of
financial instability. Instead of concluding that fundamental reform was
needed, the State Council framed it into a crisis that demanded unparallel
central power. Previous proposals on how to deal with the NPL issue
were either abandoned or incorporated into the new drive for central
power. As shown below, the enormous financial power concentrated at
the centre ultimately did not bring a fundamental solution to the NPL
problem.

Stemming the Creation of NPLs

Zhu's main strategies to stem the creation of NPLs included the
centralization of the banking system, the imposition of strict NPL re
duction quotas on the state banks and improved risk evaluation in the
banking system. Instead of accompanying these measures with policies
that gave banks more independence and made lending more market
oriented, Zhu used centralized banking power to fulfil other political
goals, namely keeping large SOEs afloat and inland development. Thus,
while the creation of NPLs has slowed in the short run, capital allocation
in China is still largely driven by state policies.
After the First Plenum, centralization measures were quickly devised
and implemented. The central government set up a separate vertical Party
committee system within the financial sector which appointed and moni
tored bank managers. At the top of this new hierarchy was the Central
Finance Work Committee headed by then vice-premier Wen Jiabao. This
institutional change was finally able to remove local state banks from the
grasp of the local government, although informal influence and the giving

54. Zhu Rongji, "Renzhen shiqie shiwu da jingshen; jiji luoshi guanyu jingji tizhi gaige
de ge xiang buchu" ("Conscienciously realize the spirit of the 15th Party Congress; actively
implement the various arrangements concerning economic structure reform"), in Xin shiqi.
55. ICBC Research Group, "A research report."

This content downloaded from 111.68.96.110 on Thu, 14 Dec 2017 12:01:26 UTC
All use subject to http://about.jstor.org/terms
 Dealing with Non-Performing Loans

of "face" continued between local officials and bank managers.56 It also

served as the backbone for a series of other centralization efforts. For one,
PBOC provincial offices were abandoned in favour of regional offices
that oversaw banking activities in several provinces. This measure was
designed to elevate regional PBOC offices above the reach of the
provincial government.57
For the Big Four state banks, the establishment of a Party committee
within each bank meant that the headquarters in Beijing could finally
order local branches to do its bidding without local interference. In
concert with Zhu's drive to lower the NPL ratio in China, the headquar
ters of the Big Four banks also lowered the lending authority of local
branches from billions of RMB to hundreds of millions of RMB.58
Finally, headquarters of state banks offered local branches interest
slightly above deposit rate to encourage them to pass deposits up to the
headquarters level.59 Given strict NPL quotas, local managers were eager
to let headquarters worry about making loans.
Zhu also set harsh targets to lower the ratio of NPLs and ordered state
banks to implement responsibility systems so that managers who autho
rized loans could be punished with wage reduction and employment
termination.60 PBOC governor Dai Xianglong further specified this policy
by ordering the Big Four banks to lower NPL ratio by 2 to 4 per cent
annually depending on the quality of their portfolios.61 The net effect of
these policies was a new autonomy for the state banks from local
influence and the concentration of enormous financial resources at the
central level. Efforts were also made to bring state banks more in line
with international practice in risk assessment and loan classification.
These changes, if combined with market-promoting policies, could have
engendered fundamental reforms in the financial system.
Despite new policies that moved the banking sector towards commer
cial independence, interviewees unanimously pointed out that the under
lying logic of the state banking system was still one of following
commands from the central government. Bankers took credit risk seri
ously simply because the centre had issued a quota, not because they truly
cared about profitability. If Zhu's command was withdrawn or if the

56. Interviews in Beijing: 8 October 2000, 10 October 2000, 27 November 2000; in

Shenyang: 25 December 2000; in Dalian: 21 May 2001, 22 May 2001, 23 May 2001; in
Fuzhou: 8 June 2001.
57. Local officials and bank managers in Beijing reported that PBOC redistricting served
only to confuse jurisdiction in bank monitoring. Others thought that creating PBOC regions
was not nearly as important as establishing Party committees within banks. Interviews in
Beijing: 10 October 2000,3 January 2001; in Dalian: 22 May 2001,23 May 2001; in Fuzhou:
11 June 2001.
58. Interviews in Beijing: 10 October 2000; in Dalian: 22 May 2001, 23 May 2001.
59. Ren Derong, "Dangqian huobi zhengce chuandao shouzhu de biaoxian, cheng'en, yu
duice" ("The manifestation, causes, and solution to the blockage to current monetary
policies"), JC, No. 6 (2000).
60. Zhu Rongji, "Deepening financial reform."
61. "Guoyou yinhang shangshi san nian zhinei meixi" ("The chance for state banks to list
shares publicly within three years is minimal"), Jingji da cankao {The Great Reference of
Economics, hereafter JDQ, No. 4 (2001).

This content downloaded from 111.68.96.110 on Thu, 14 Dec 2017 12:01:26 UTC
All use subject to http://about.jstor.org/terms
The China Quarterly

political wind changed, bankers would immediately lapse back to carefree

lending.62 Meanwhile, the Zhu administration mobilized the centralized
financial apparatus to fulfil a host of policy objectives.
After the Big Four banks centralized, they became an integral part of
the three-year programme to get SOEs out of difficulties (san nian
tuokun). First, the policy of subsidizing technical innovation loans re
quired large and medium SOEs to submit proposals for improving
production technology to the SETC. After approval, the MOF would pay
for the interest of the loan until the completion of the project. In 1998
alone, 10 billion RMB in bond issuance was earmarked for this purpose,
which was expected to generate 200 billion RMB in loans.63 Although
banks were under no formal obligation to approve a technical innovation
loan, bank managers in reality competed with each other to lend to these
projects because SETC and MOF approval freed them from being held
responsible in the event of default.64 Because of these incentives, banks
approved over 90 per cent of all loans proposed by the SETC.65
Second, the provision of close-ended loans (fengbi daikuan)
specifically targeted loss-making large SOEs. Under this policy, banks
were instructed by the SETC to lend to loss-making SOEs that had a
production line with reasonable market and profit potentials. Presumably,
if funds were directed at improving the promising production line rather
than at wage arrears, an SOE would have a better chance of becoming
profitable. The State Council also occasionally demanded that state
commercial banks rescue certain industries. For example, in 1999, banks
were ordered to "lend" to qualified loss-making SOEs so that they could
repay fees owed to large state coal mines.66 Although this policy benefited
both large SOEs and state coal mines, banks were unlikely to see any
return on their "loans."
The combination of bank centralization, emphasis on risk prevention
and bias towards large SOEs also created an extremely favourable
environment for large SOEs and state corporations to obtain bank loans.
Under enormous pressure to maintain growth at 7-8 per cent, banks lent
to firms that had minimal risk in the short run, including large, strategic
SOEs and state corporations that had recently offered their shares pub
licly.67 Since most publicly traded state firms had also undergone debt-to
equity swaps, their asset-to-liability ratios were likely to be much lower.
Indeed, an internal decree of the Construction Bank strongly urged its

62. Interviews in Beijing: 10 October 2000, 10 November 2000, 27 November 2000, 14

May 2001, 19 May 2001; in Dalian: 22 May 2001, 23 May 2001.
63. Bureau of Economic Prediction of the State Information Center, "Duoxiang zhengce
chutai youxiao tuidong guoqi tuokun mubiao de shixian" ("Many policies implemented to
effectively realize the goal of state-owned enterprises getting out of difficulties") Zhongguo
hongguan jingji xinxi {Chinese Macroeconomic Signals, hereafter ZHJX), No. 38 (1999).
64. Interviews in Beijing: 15 April 2000, 27 November 2000; Shenyang: 25 December
2000; Dalian: 21 April 2001, 22 April 2001; Fuzhou: 7 June 2001
65. Interview in Beijing: 6 April 2001.
66. Bureau of Economic Prediction, "Many policies."
67. "Bu Ian dai ye bu xi dai" ("Do not lend freely but also do not lend too little"), ZHJX,
No. 9 (1998).

This content downloaded from 111.68.96.110 on Thu, 14 Dec 2017 12:01:26 UTC
All use subject to http://about.jstor.org/terms
 Dealing with Non-Performing Loans

branch banks to lend to listed firms for this precise reason.68 More
importantly perhaps, these companies enjoyed the protection and support
of the SETC.69 Thus, rather than evaluating firms based on their commer
cial or liquidity risk profiles, bankers lent to firms based on their
bureaucratic risk profiles.
The Zhu administration also effectively mobilized the banking sector
to fulfil another "campaign promise," the financing of the Go West
Campaign. Since the 15th Party Congress placed a high priority on
developing western China, bureaucrats at the State Council have viewed
it as a way to build up their own "administrative accomplishments"
(zhengji). PBOC governor Dai, for example, personally promised to
"actively oversee" (jiji ducu) state commercial banks to lend to state-di
rected infrastructure projects in the region.70 He also promised to finance
the mergers of loss-making SOEs in the west with profitable ones in the
east.71 By the end of 1999, over 1.5 trillion RMB in loans have gone to
western China. This was a 16 per cent increase from 1998, or twice the
national rate for loan expansion. Bankers gladly complied with the policy
pressure because they were much less responsible in the event of default.
In fact, state banks and even joint-stock banks routinely sent representa
tives to the SDPC to bid for state priority projects in western China.72
Finally, in order to ensure stability in poor regions, the State Council
still occasionally ordered commercial banks to make "peace and unity
loans" (anding tuanjie daikuan) to pay unemployed workers.73 While
mid-level bankers disliked these deliberate policies to help troublesome
SOEs, they recognized that senior managers needed to respond to the
policy demands of Premier Zhu.74 The combination of these central
lending initiatives gave rise to a rapid increase in long-term fixed asset
investment loans (Figure 3).
The centralization of the banking system gave Premier Zhu much more
power with which to fulfil political objectives. At the same time, it
slowed the creation of new NPLs in the short run because fixed asset
loans to large SOEs and state construction projects tended to have longer
maturity dates. Since most of these entities were also supported by direct

68. Construction Bank of China, "Guanyu jiaqiang dui gufenzhi gaizao qiye jinrong fuwu
de tongzhi" ("Notice on strengthening financial services to firms that have undergone share
system transformation"), in Construction Bank of China (ed.), Zhongguo jianshe yinhang
guizhang zhidu xuanbian: 1997 (A Selection of Regulations and Systems for the China
Construction Bank: 1997) (Beijing: Economic Science Publisher, 1998).
69. Chen Qiang, "Fei guoyou touzi de xianzhuang, zhang'ai ji duice jianyi" ("The current
situation and obstacles to non-state investment and suggestions of policy responses"), Jingji
yucefenxi (Economic Prediction and Analysis), No. 39 (2000).
70. "Jinrong jiang da li zhichi xibu kaifang lidu" ("The financial industry will
whole-heartedly support the force for the development of the west"), JDC, No. 7 (2000).
71. Bureau of Economic Prediction of the State Information Center, "Jinrong jiang cong
liu fangmian zhichi xibu" ("Finance will support development of the west in six ways"),
Caijing kuaixun (Headlines of Finance and Economics), No. 17 (2000).
72. Interviews in Beijing: 27 October 2000, 10 November 2000, 27 November 2000, 3
January 2001.
73. Interviews in Beijing: 8 October 2000, 27 November 2000, 3 January 2001; in
Shenyang: 26 December 2000.
74. Interview in Beijing: 8 October 2000, 10 October 2000.

This content downloaded from 111.68.96.110 on Thu, 14 Dec 2017 12:01:26 UTC
All use subject to http://about.jstor.org/terms
938 The China Quarterly

Figure 3: Share of Fixed-Asset Loans as% of Loans Made by State

Banks, 1996-2001 (quarterly)
Note:
1 = 100%.
Source:
Division of Statistics of the People's Bank of China, ZJT, National Bureau of Statistics,
China Monthly Economic Indicators 2001: 3 (March).

fiscal subsidies from the central government, they were likely to be able
to sustain interest payments for the first few years. In the long run,
however, it is unclear whether NPLs will continue to haunt the banking
sector. Firms which benefited from Zhu's tuokun programme are proba
bly taking central aid for granted, creating a moral hazard problem.
Already, one report found that large SOEs in Hubei were using around 10
per cent of received bank loans to speculate in China's highly volatile
stock market.75
Recently, an unconfirmed source reported that when the PBOC carried
out a limited investigation of loans over 100 million RMB made in 100
branches of the state commercial banks at the end of 2001, it found an
astounding 57.28 billion RMB in NPLs, most of which were generated
through recent loans to large SOEs and state construction projects.76 This
bodes extremely ill for the future of the Chinese banking sector. Never

75. Research Team of the PBOC Wuhan Branch, "Dui Hubei sheng xin fafang daikuan
qingkuang de diaocha" ("An investigation on new loans in Hubei province"), JC, No. 2
(2001).
76. Beijing Xin Bang Investment & Consulting Ltd, "Si da yinhang 570yi daikuan cheng
daizhang, zhongyang zhengjing" ("57 billion in loans from Big Four banks became NPLs;
the centre is shocked"), Xinbang caijing qingbao {Xinbang Financial Intelligence), 4 June
2002.

This content downloaded from 111.68.96.110 on Thu, 14 Dec 2017 12:01:26 UTC
All use subject to http://about.jstor.org/terms
 Dealing with Non-Performing Loans

theless, at the moment, the official rate of NPL generation has slowed
down; large SOEs, flushed with cash, were generating operating profits;
billions were pouring into western provinces as construction loans.

Dealing with Non-Performing Loans

Besides devising policies to slow down the creation of NPLs, Premier

Zhu also devised ways to decrease the enormous pool of existing ones.
After taking office, Zhu immediately devised a two-pronged strategy to
deal with the problem. First, he recapitalized banks and wrote off a small
part of the NPLs. The main effort, however, was setting up asset
management companies (AMCs) to take over NPLs from state banks. It
is now clear that the emphasis on AMCs was intended to minimize
explicit fiscal deficit in the short run and to defer the NPL problem to the
future.
In 1998, the central government issued 270 billion RMB of special
financial bonds to banks. The government then took this money to
re-capitalize the same state banks which had bought the bonds. State
banks gained enormously from this process as they obtained both good
assets in the form of government backed bonds and higher capital ratio
from government injection. Instead of banks bearing the burden of NPLs
due to government policies, the NPL burden was transferred to the
government in the form of debt. Had this process repeated itself in
subsequent years and had banks been allowed to use government injec
tions to write-off NPLs, state banks would have been on their way to
being free of "historical burden."
Instead, however, the MOF continued to set the maximum write-off
amount to 1 per cent of all loans outstanding during a given year. This
regulation, in place since the early 1990s, was primarily designed to
control the amount of NPLs that was reflected on the budget deficit.
Banks using too much to write-off loans would be unable to hand "profit"
over to the MOF. Although the government briefly considered a gradual
increase of this ratio in 1998, it ultimately decided against it in order to
minimize explicit national debt.77 Making matters worse, state banks had
little authority over which loans to write-off. Under the tuokun pro
gramme, the SETC distributed much of the 40-50 billion or so RMB in
annual write-off money to politically connected large SOEs and state
corporations.78
Because existing policy on write-offs only resolved a tiny fraction of
the estimated 3.3 trillion RMB (37 per cent of 2000 GDP) in NPLs, the
Zhu administration devised an alternative policy to reduce NPLs.79 Again,

77. Interview in Beijing: 8 March 2001, 15 April 2001. "Jinnian jiang zhenzhong chuli
chongxiao daliang yinhang huaizhang wenti" ("[The government] will conscientiously deal
with and write-off large amount of bad debt in banks this year"), ZHJX, No. 19 (1998). This
rule finally changed in 2003 when the SETC disbanded.
78. Interview in Beijing: 14 May 2001; 23 June 2001. Bureau of Economic Prediction,
"Many policies implemented."
79. Chan et al, "US$400 billion needed."

This content downloaded from 111.68.96.110 on Thu, 14 Dec 2017 12:01:26 UTC
All use subject to http://about.jstor.org/terms
940 The China Quarterly

there were several proposals. The first would securitize NPLs into
government bonds and directly sell them to the public.80 This was
eliminated early, presumably because it required the government to
publicize SOE performance and to decrease the pool of money going into
the stock market. More importantly, it would drastically increase the
budgetary deficit. Another proposal called for the transformation of debt
to state investment (daigaitou). Government researchers recommended
various solutions within this broad category. For one, the central govern
ment could directly issue bonds to state banks and use the capital to
purchase NPLs from state banks. It could then convert NPLs into state
shares, thereby digesting the NPLs.81 This proposal, however, would also
immediately transform NPLs into explicit budgetary deficit. If the
government issued bonds to the state banks to buy NPLs, it in effect
transferred NPLs from the banks' balance sheet directly to the govern
ment's balance sheet. A second solution involved conducting debt-to
equity swap within banks and setting up asset management departments
within banks to oversee companies as shareholders. However, it was felt
that this solution would further tie banks to the fate of SOEs, encouraging
them to lend more to failing SOEs.82
The most promising option within the daigaitou category was to set up
some kind of financial company to take over AMCs which gave either
bonds or stocks to the banks in exchange for bad debts. This would clean
up the bank's balance sheet and separate bad debts from the banks.
Finally, the government considered using private financial companies to
purchase NPLs from banks at a highly discounted rate.83 The last option
would also require the government to inject huge sums of money into
banks to write-off the discounted amount after the sale of NPLs to private
entities. Again, the necessity of injecting large amounts of government
money and thereby increasing explicit deficit made this option extremely
unattractive to the Zhu administration.
The decision to set up state-owned AMCs came fairly quickly after
Zhu had become Premier. By the middle of 1998, the State Council had
decided to set up AMCs to absorb NPLs at face value and to recover as
many NPLs as possible. The four AMCs, Xinda, Changcheng, Huarong
and Dongfang, which took over NPLs from China Construction Bank, the
Agricultural Bank of China, Industrial and Commercial Bank of China,
and the Bank of China respectively, each received 10 billion RMB in
initial capital from the MOF.84 On that basis, the four AMCs issued 1.4
trillion RMB in financial bonds to the state banks and used the funds to

80. "Zhuanjia fenxi: 1997 nian jinrong fengxian you shei lai chengdan" ("Expert analysis:
who will bear the financial risk in 1997"), Qiyejuece cankao {Enterprise Decision-Making
Reference), No. 130 (1997).
81. Er Jianmin (ed.), Guojia jinrong anquan baogao {A Report on the Financial Safety of
the Country) (Beijing: Central Party School Publisher, 1999).
82. Er Jianmin, A Report, p. 221.
83. Ibid.
84. The People's Bank of China, Ministry of Finance and China Securities Regulations
Commission, "Guanyu zujian Zhongguo Xinda zichan guanli gongsi de yijian" ("Opinion
concerning the formation of Xinda Asset Management Company"), in Zhan Xiangyang (ed.),

This content downloaded from 111.68.96.110 on Thu, 14 Dec 2017 12:01:26 UTC
All use subject to http://about.jstor.org/terms
 Dealing with Non-Performing Loans

purchase 1.4 trillion in NPLs from the Big Four state banks at face
value.85 AMCs each had a charter of ten years and were suppose to
recover as many of the NPLs as possible through debt-to-equity swap,
bankruptcy and restructuring debt. At the end of the ten-year charter, the
MOF will issue bonds or inject government surplus to write-off the
remaining amount. In this manner, state banks replaced 1.4 trillion RMB
in NPLs with 1.4 trillion RMB in MOF-backed AMC bonds, thereby
getting rid of some two-fifths of the estimated 3.3 trillion in NPLs.
Meanwhile, the MOF did not have to list the 1.4 trillion in special bonds
on the official budget, since it merely guaranteed bonds issued by the
AMCs. The AMCs, on the other hand, were saddled with 1.4 trillion
RMB in NPLs. Although AMC officials initially resisted purchasing
NPLs at face value, the opposition soon dissolved as they realized that the
MOF was ultimately responsible for the pool of NPLs.86
After purchasing NPLs, the main strategy by which AMCs converted
bad asset into performing asset was debt-for-equity swap, or transforming
the debt owed by SOEs to share ownership for AMCs. Instead of
deciding which SOEs qualified for the swap themselves, AMCs received
SETC "recommendations." Because a debt-for-equity swap drastically
reduced a company's debt level, eager SOE managers and local officials
waited in a "long line of Red Hag sedans" in front of the SETC for the
privilege.87 In choosing beneficiaries, the SETC focused on large SOEs
and state corporations, especially those with personal connections and
factional ties.88 This policy configuration undoubtedly gave the central
bureaucracy and ultimately Zhu Rongji a wealth of political capital in the
Politburo.
In the short run, every bureaucratic interest, as well as the political
interest of Zhu Rongji, gained from the transfer of NPLs to AMCs. The
state banks and the PBOC were happy to replace NPLs with performing
assets backed by the MOF. AMC-issued bonds bore interest rates bench
marked to bonds issued by the policy banks, which increased bank
profitability significantly.89 Bankers could therefore claim increasing
profit as their administrative accomplishment. The MOF supported the
plan because it did not immediately have to make explicit the treasury's
obligation to the public. While the MOF would have to deal with the
remaining NPLs in ten years, top officials were mainly interested in

footnote continued

Lun Zhongguo buliang zhaiquan zhaiwu de huajie (On the Dissolution of Bad Debt and Bad
Debt Obligations in China) (Beijing: China Financial Publisher, 2000).
85. "At face value" means at the original amount of the loan plus accrued interest. In a
market economy, private asset management companies by definition never purchase a NPL
at face value because of the high risk profile of a non-performing loan.
86. Interview in Beijing: 14 May 2001
87. Xiangyang Zhan, On the Dissolution of Bad Debt, p. 199.
88. Interviews in Beijing: 4 December 2000, 14 May 2001. State Economic and Trade
Commission and People's Bank of China, "Guanyu shishi zhaiquan zhuang guquan ruogan
wenti de yijian" ("Opinion on certain problems on the implementation of debt-equity swap"),
in Zhan Xiangyang, On the Dissolution of Bad Debt.
89. Interview in Beijing: 19 October 2000.

This content downloaded from 111.68.96.110 on Thu, 14 Dec 2017 12:01:26 UTC
All use subject to http://about.jstor.org/terms
942 The China Quarterly

providing a short-term solution to promote their career prospects.90 The

SETC, an agency under Zhu's strict control, gained control over the
debt-to-equity swap, which decided the fate of thousands of SOEs. In
fact, the central government found AMCs the ideal policy tool and
increased the amount of NPLs transferred to them from the originally
planned 400 billion to 1.3 trillion and finally to 1.4 trillion RMB.91 In this
process, Zhu "solved" a major problem and gained an enormous amount
of political capital without jeopardizing the short-term fiscal balance.
Despite the short-term success of the AMC programme to reduce
NPLs, rescue SOEs and reduce fiscal pressure, a main outcome was to
transfer fiscal pressure to the future. While official announcements opti
mistically claimed that AMCs would recover 30-50 per cent of the NPLs,
Western analysts and officials involved in setting up AMCs agreed that
the actual ratio is likely to be 10-20 per cent.92 Recovering transferred
asset in real estate proved to be extremely difficult because of collapsing
real-estate prices in some localities.93 Moreover, while debt-to-equity
swap gave AMCs sufficient preferred shares to pay interest on their
bonds, they ultimately needed to find buyers for their shares of SOEs to
repay the principal of the bonds.94
AMCs have thus far dealt with over 500 billion RMB in NPLs, but
cash recovery as of the end of 2003 was only 99 billion RMB, or 7 per
cent of the total.95 Attempts to sell NPLs to foreign institutional investors
yielded a return of 6 per cent, and the most recent offering of NPLs by
an AMC failed to attract sufficient bidders even after the assets were
greatly discounted.96 Despite these troubling indicators, State Council
officials were not perturbed because they knew that the main objective of
the NPL policies was to minimize short-term burden for the Zhu admin
istration.97 The next administration would have to take care of itself.

Conclusions
China's opening to the global economy and increasing participation of
Western-trained economists and managers in its financial system have

90. Interviews in Beijing: 23 June 2001, 20 May 2001.

91. Gordon Chang, The Coming Collapse of China (New York: Random House, 2001),
p. 133.
92. Interviews in Beijing: 13 December 2000, 23 June 2001. Bureau of Economic
Prediction of the State Information Centre, "Zichan guanli gongsi qianjin lu shang kunnan
duo" ("Asset management companies will face many difficulties in the road ahead"), Caijing
kuaixun, No. 42 (2000).
93. Ge Chong, "Hainan sheng fang dichan daikuan xianzhuang ji panhuo duice" ("The
current situation with real estate loans in Hainan and responses to enliven it"), JC, No. 2
(2001).
94. Zhang Wenkui, Ji Min and Xu Duo, "The effect of debt-to-equity swap on corporate
governance structure and an analysis on its prospect," China Development Review, 13 January
2002.
95. Andrew Collier, "Asset managers take 5 years to recover 36pc of bad loans," South
China Morning Post, 20 January 2004.
96. Karen Richardson, "Chinese banks' bad loans draw increased scrutiny," Wall Street
Journal, 15 January 2004.
97. Interviews in Beijing: 14 May 2001, 23 June 2001.

This content downloaded from 111.68.96.110 on Thu, 14 Dec 2017 12:01:26 UTC
All use subject to http://about.jstor.org/terms
 Dealing with Non-Performing Loans

laid more stones that lead to the distant shores of a marketized financial
system. Yet, the swift political current demands that bureaucrats in charge
of China's economy act as politicians and devise a rational strategy to
remain standing. In the highly uncertain political environment in China,
that strategy involves hoarding financial authority and devising policies
that give short-term fixes to pressing problems. Elite power politics
makes fundamental reforms that bear the risk of instability unpalatable to
everyone. As a result, the domination of the state in the financial sector
continues beneath a series of cosmetic changes.
The logic underlying this article is by no means novel in the field of
political science. A number of studies suggest that politicians' short-term
concerns constitute major obstacles to reform.98 Nevertheless, short-term
political concerns typically are not used to explain economic policies in
high-growth countries in East Asia.99 Within the China literature, this
article suggests a reform logic which departs from the conventional view
of the Chinese government, that it has a strong interest in carrying out
incremental reform that leads to complete marketization. While previous
frameworks do well in explaining instances of reform or decentralization,
the power framework does better in explaining centralization and reform
stagnation. Rather than negating the previous explanations of Chinese
reform, this work suggests that a more general framework is needed to
explain both the successes and failures of reform across various sectors.
This task requires a time-series, cross-sectoral analysis to pin down the
causal mechanism.
One preliminary observation is that the central bureaucracy is increas
ingly abandoning specific policy authorities in favour of more fungible
powers. Whereas complex rules made by the ministries used to govern
many aspects of firm behaviour, tightening central control over taxation
and monetary policies now arguably gives the central government just as
much leverage over major firms, while drastically reducing its monitoring
costs.100 Politically, also, rather than making complex preferential policies
to specific provinces or sectors, the top technocrats can now simply give
either fiscal subsidies or bank loans to the target recipients. Again, since
money is fungible, the top technocrat can earn political credit without
making substantial institutional investment. For example, the Open up the
West Campaign's massive injection of funds through the budget and the
banking system stands in stark contrast to preferential treatment of
Guangdong in the 1980s, which involved a mix of investment and
complex industrial and fiscal policies.
Moreover, the concentration of fiscal and financial power in a few

98. See, for example, Ann Krueger, The Political Economy of Policy Reform in Developing
Countries (Cambridge, MA: MIT Press, 1993), Adam Przeworski, Democracy and the
Market: Political and Economic Reforms in Eastern Europe and Latin America, Studies in
Rationality and Social Change (Cambridge: Cambridge University Press, 1991).
99. The exception is Kang' s work on Korea. See David Kang, "Bad loans to good friends,"
International Organization, Vol. 56, No. 1 (2002).
100. Zysman makes a similar observation about industry policy in European countries. See
John Zysman, Governments, Markets, and Growth: Financial Systems and the Politics of
Industrial Change (Ithaca, NY: Cornell University Press, 1983), p. 76.

This content downloaded from 111.68.96.110 on Thu, 14 Dec 2017 12:01:26 UTC
All use subject to http://about.jstor.org/terms
944 The China Quarterly

central agencies also provides the top technocrat with a virtual monopoly
over the allocation of state resources. In other words, the concentration of
rent allocation authorities in a few agencies under the Premier's direct
control makes it much harder for other Politburo members to play a direct
role in allocating rent. They become more dependent on the top tech
nocrat, which gives him much more political leverage over the rest of the
Politburo. Thus, reform in the financial sector might be a victim of the
central bureaucracy's simplification strategy. While a top technocrat is
willing to liberalize other sectors of the economy, the enormous pool of
savings in the banking sector renders it an indispensable policy and
political instrument. In China's increasingly monetized economy,
financial and fiscal institutions become even more crucial in the elite
political calculus.

This content downloaded from 111.68.96.110 on Thu, 14 Dec 2017 12:01:26 UTC
All use subject to http://about.jstor.org/terms