PARTY ADAPTATION, ELITE TRAINING, AND POLITICAL SELECTION
IN REFORM-ERA CHINA
A DISSERTATION
SUBMITTED TO THE DEPARTMENT OF POLITICAL SCIENCE
AND THE COMMITTEE ON GRADUATE STUDIES
OF STANFORD UNIVERSITY
IN PARTIAL FULFILLMENT OF THE REQUIREMENTS
FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY
Charlotte P. Lee
June 2010
http://creativecommons.org/licenses/by-nc/3.0/us/
This dissertation is online at: http://purl.stanford.edu/jn431dw0841
© 2010 by Charlotte Ping Lee. All Rights Reserved.
Re-distributed by Stanford University under license with the author.
This work is licensed under a Creative Commons Attribution-Noncommercial 3.0 United States License.
ii
I certify that I have read this dissertation and that, in my opinion, it is fully adequatein scope and quality as a dissertation for the degree of Doctor of Philosophy.
Jean Oi, Primary Adviser
I certify that I have read this dissertation and that, in my opinion, it is fully adequatein scope and quality as a dissertation for the degree of Doctor of Philosophy.
Beatriz Magaloni-Kerpel
I certify that I have read this dissertation and that, in my opinion, it is fully adequatein scope and quality as a dissertation for the degree of Doctor of Philosophy.
Michael McFaul
Approved for the Stanford University Committee on Graduate Studies.
Patricia J. Gumport, Vice Provost Graduate Education
This signature page was generated electronically upon submission of this dissertation in electronic format. An original signed hard copy of the signature page is on file inUniversity Archives.
iii
iv
Abstract
How and why have seemingly outdated political organizations of the Chinese
Communist Party (CCP) persisted through China’s past three decades of economic
liberalization and social transition? In answering this question, I focus on the CCP’s
nationwide system of political training academies, or party schools, to piece together the
story of these schools’ role within the party apparatus and subsequent adaptation to
changed incentives and circumstances. By investigating the logic of party organization
and examining sub-party actors, this dissertation seeks to address broader questions
regarding how ruling parties in authoritarian regimes are able to generate incentives for
both institutional continuity and adaptation.
During market reforms, CCP leadership has invested in reviving and updating
party organizations of bureaucratic management. In the case of the party school system,
party authorities have sought to adapt these core political organizations to a market
context not by insulating them from market forces, but by linking organizational survival
to market savvy. Central party authorities have induced adaptation by opening party
schools to local and national market forces. This study thus uncovers a market-based
path by which organizations within a Leninist party persist through – and even thrive on
– economic transformation. I draw on organizational theories of competition and
redundancy to explain the logic of these developments – and resultant “party
entrepreneurialism”.
Two sets of findings emerge from my research. First, the party school system in
China continues to be an important site of political control over individual bureaucrats.
My analysis of survey and career history data reveals that party school enrollment
v
increases the likelihood of attaining a higher administrative rank and more rapid
promotion up the career ladder. In the principal-agent relationships which suffuse
China’s hierarchically organized political system, the party must solve the problem of
how to select bureaucratic agents at all levels of the system. This selection problem
comes prior to the monitoring of agents and is particularly salient in the highly
competitive bureaucracy of the Chinese party-state. By employing a matching method on
large-N survey data and analyzing an original dataset of the career histories of central-
level officials, I find that party school training constitutes a pipeline to high office.
Second, party schools have fulfilled this selection function while responding to
the demands of multiple markets. These include the local and national markets for goods
and services opened under the reforms of the past thirty years and a training market
created by central party authorities. Party authorities have harnessed market forces in
order to generate incentives for traditionally closed party schools to turn outward for
income, innovative training content, and new partners. Content analysis of an original
dataset of Central Party School training syllabi from 1985 to 2007 reveals the extent of
change in party school training content. Findings from field visits to the party schools of
two provinces, one Special Economic Zone, and the Central Party School in Beijing
uncovers the means by which these organizational actors have leveraged the limited local
autonomy granted to them and become highly entrepreneurial. One result has been an
expansion in the diversity of educational and profit-making activities on party school
campuses, developments that both complement and conflict with the core function of
these schools as elite training grounds.
vi
These findings illustrate the balance that China’s ruling party has sought to strike
between encouraging organizational reform while maintaining preexisting institutional
arrangements for managing political elites. The “party entrepreneurialism” that has
developed in the party school system has implications for central-local relations in China
as well as strategies of organizational adaptation within the CCP. Market opportunities
embed party schools more deeply in local economies, and this has the potential to
strengthen the ties between schools and local constituents – at the risk of compromising
school responsiveness to central dictates. At the same time, the marketization of cadre
training has generated strong incentives for party schools to search actively for innovative
solutions to challenges from party and non-party competitors. Over the past three
decades of reform, party schools have proven to be nimble political and economic actors:
school leaders have developed an entrepreneurial spirit, in the process retaining the
relevance of their organizations within the party and profiting from the market-based turn
that internal party reforms have taken.
vii
Acknowledgements
When I began my graduate studies six years ago, I had little idea that the road to
producing a dissertation would be so delightfully peopled with teachers and friends, both
old and new, who would inject much wisdom and encouragement into what is otherwise
a long, wearying process. Along the way, I have become indebted to so many individuals
– at Stanford, in China, and from my alma mater, UC Berkeley – and each has
contributed to my research and thinking in ways that are difficult to measure. I am
grateful for the generosity of ideas and spirit that many mentors, friends, and family have
bestowed on me these years.
Heartfelt thanks goes to my advisor, Jean Oi, whose energy, unflagging curiosity,
and deep knowledge of Chinese politics have been an inspiration to me. I still remember
feeling exhausted just watching Jean conduct one interview after another in the townships
and villages where rural China has taken off, far from the glamour and resources of the
big cities. Her formidable skills in the field and in the classroom will continue to shape
my approach to research and teaching.
I also could not have completed this project without the encouragement of my two
other dissertation readers, Beatriz Magaloni and Mike McFaul. Beatriz challenged me
always with her questioning and thinking about how the China case fits into our
understanding of authoritarian systems, and she did so while achieving an admirable
work-life balance. I am grateful for Mike’s big questions and his amazing ability to
support my work while handling calls from the “situation room.”
Special thanks go to Andy Walder for pushing my research in new directions and
chairing my dissertation defense with humor and care. Alice Miller has also taught me a
viii
great deal about how to put together a terrific class while teaching me volumes about the
intricacies of elite politics in China. I also thank her for bringing her expertise to the
table during my dissertation defense. Although he was not on my dissertation committee,
I am also indebted to Larry Diamond for his mentorship over the years and admire his
ability to bridge the policy and scholarly worlds.
The process of writing this dissertation was also due in no small part to the
support of my colleagues at Stanford and beyond. I thank the members of my
dissertation writing group, Bethany Lacina, Jed Stiglitz, Luke Condra, and Roy Elis for
their gentle comments in the face of glaring flaws. Thanks also go to my
interdisciplinary writing group, including Daniel Kreiss, Johanne Eliacin, Rekha Balu,
and Zhanara Nauruzbayeva, for their fresh perspectives. I thank Leslie Wang for her
critical spirit and introduction to the Berkeley graduate student community.
In the social sciences at Stanford, I am fortunate to be surrounded by a collegial
circle of China researchers, and the brilliance of my colleagues continually inspires me.
Thanks go to Yuen Yuen Ang, David Barmore, Chris Chan, Joo-Youn Jung, Jen Haskell,
Xiaobin He, Xiaojun Li, Chao-Chi Lin, Paul Ling, Sung-Min Rho, Kay Shimizu, Mee
Smuthkalin, Jeremy Wallace, and Rachel Wu for their support and help throughout.
My deep thanks and appreciation go to each of my many friends and
acquaintances in China. They must remain anonymous here, but without their time and
insights this dissertation would not have been possible. I consider myself fortunate to
have met so many Chinese officials, teachers, and students who were willing to share
their experiences and, in so doing, illuminate some of the challenges and solutions
devised within their complex political system.
ix
In a field where the data collection challenges are daunting, I am grateful for the
research assistance of Kiki, Chelsea, Haihui, Shaorong, Xiaoya, and Zhaofen.
My research would not have been possible with without financial and
administrative support. The National Science Foundation and, at Stanford University, the
Freeman Spogli Institute for International Studies and Center for East Asian Studies
provided grants to support my research and dissertation write-up. I also thank Chandelle
Arambula, Eliana Vasquez, and Tram Dinh for being oases of sanity.
Many personal friends and mentors provided words of wisdom, lent sympathetic
ears, and opened their homes to me over the years, especially during my various trips
spent crisscrossing China on a student budget. I thank Nancy Rogers for her far-sighted
advice. I also thank my many dear friends from the 1997-98 UC EAP program in China.
Over the years, we have shared an abiding interest in China’s rise.
Finally, I am deeply grateful to my immediate and extended family members for
their boundless generosity and support. When I was in the Peace Corps in Romania and
typing graduate school applications on a borrowed computer, my parents, Ken and Jean
Lee, put the documents in the mail and got the whole process started. My brothers, Eric
and Kennon, have been able problem-solvers along the way and constant sources of
laughter. The entire “village” has kept me well-nourished and grounded all these years.
Finally, I am indebted to my partner, Nik Crain, who endured through countless midnight
practice presentations and managed every crisis with aplomb. Without his patience,
fortitude, and love, this project would be much diminished.
Charlotte Lee Cupertino, CA
x
Table of Contents
Acknowledgements vii
List of Tables xi
List of Figures xiii
Chapter 1: Introduction 1
Chapter 2: The Organizational Landscape 38
Chapter 3: Managing the Managers 64
Chapter 4: Fusing Party and Market 109
Chapter 5: Organizational Strategies of Adaptation 146
Chapter 6: Conclusion 186
Appendices
A. Number of party schools, by locale and national share of leading cadres 223
B. Note on sources and research methods 224
C. City Z training allocations, 2008 228
D. PSM descriptive statistics and robustness tests 230
E. Central Party School Young Cadre Training Classes descriptive data 235
F. International partnerships, central and provincial-level party schools 238
G. Categories for coding training syllabi 240
References 242
xi
List of Tables 1.1 Turnover in party and state managers, 1988 to early 1990s 12
1.2 Turnover in Chinese party and state managers 13
2.1 Provincial-level party school leadership 49
2.2 City Z planned training classes, 2008, by type and enrollment levels 54
2.3 Percent increase, over previous year, in CPS Young Cadre trainees, 61 total leading cadres, and total CCP members, select years
2.4 CPS Young Cadre Training Class enrollment, 1980-2000 61
2.5 Highest educational attainment of cadres, by historical period 62
2.6 Educational attainment of leading cadres after 1982-84 reforms 62
3.1 The organization of public officials in China, 1998 70
3.2 Descriptive summary of cadre and general population 79
3.3 Probit regression results for propensity score estimation 82
3.4 Marginal effect for significant control variables 84
3.5 Total cadres in treated and control groups, by propensity score block 85
3.6 Estimation of average treatment effect on the treated (ATT) 86 Enrollment in a non-degree party school training class
3.7 Estimation of average treatment effect on the treated (ATT) 87 Party school degree 3.8 Estimation of average treatment effect on the treated (ATT) 88 University degree 3.9 Estimation of average treatment effect on the treated (ATT) 88 Party school training and university degree 3.10 Occupational categories of cadres, by rank 89 3.11 Cadres who do not experience promotion, 2000 and 1995 classes 96
xii
3.12 Career tracks of CPS trainees in stalled careers (1995 and 2000 classes) 97 3.13 Total trainees, by CPS Young Cadre Training Class 101 3.14 Current administrative rank, CPS Young Cadre Training Class alumni 103 3.15 Current administrative level, CPS Young Cadre Training Class alumni 103 3.16 Current occupation category, CPS Young Cadre Training Class alumni 104 3.17 Comparing average ages, by rank 105 4.1 Summary of local party school expense categories and income sources 120 4.2 Provincial-level party schools and administrative institutes 129
4.3 A selection of training bids, by locale and awardee 136 5.1 2001 Income and Expense Report, Province B’s CYL School 150 5.2 Local party school entrepreneurial activities 156 5.3 Provincial party school partnerships with Chinese universities 162 5.4 Shanghai party school and administration institute international 164 partnerships 5.5 Central Party School training syllabi analyzed, by year 167 5.6 Comparison of CPS and provincial-level party school training content 179
xiii
List of Figures 2.1 Local party organs involved in cadre education 47
2.2 Government organs involved in civil servant and cadre education 47
3.1 Frequency of administrative rank, 2003 China GSS 80
3.2 Distribution of propensity scores for treated and control groups 84
3.3 Pre-match propensity score distribution 85
3.4 Post-match propensity score distribution 85
4.1 Central-level organizations of cadre training 122
4.2 Share of total training, by school type 137
4.3 Central Party School trainee volume, 1977-2000 139
4.4 Domestic training market entrants, by year and type of organization 141
5.1 CPS training content by category, percent of total training time (1985) 168
5.2 CPS training content by category, percent of total training time (2007) 168
5.3 Percent of CPS training time dedicated to orthodox theory 170
5.4 Percent of CPS training time dedicated to the theories of reform-era 171 leaders
5.5 Percent of CPS training time dedicated to party 172
5.6 Percent of CPS training time dedicated to party building 174
5.7 Percent of CPS training time dedicated to management 176
5.8 Percent of CPS training time dedicated to policy briefings 177
5.9 Percent of training class time dedicated to case studies 178
1
Chapter 1 Introduction
Political institutions that constrain elected officials in democracies are often
established in autocratic contexts to serve the dictator’s bid to stay in power. Such
institutions facilitate the ordering of state and society, extend the coercive capacity of the
ruler, and they do so across time and space. Designing, constructing, and maintaining
institutions of governance are vital to the state-building process, if not synonymous with
it. That institutions in authoritarian regimes often possess a complexity on par with their
democratic counterparts is not surprising, but the purpose(s) of these non-democratic
institutions are at all times conditioned by the political context of which they are a part,
i.e., to sustain the dictator’s continued rule. Given this core assumption, the task lays in
discerning the functions served by a particular authoritarian institution and, beyond this,
evaluating this institution’s capacity for coping with the uncertainty, the unforeseeable
circumstances, and contingencies that all rulers in power must eventually confront.
These matters of institutional design and adaptiveness are complicated by the “sunk
costs” that accompany the creation of any institution, by tradition and the inertia that may
resist pressures for change.
Elections, legislatures, and parties are among the most prominent and well-studied
examples of political institutions adopted by authoritarian leaders.1 The channels through
which they contribute to regime survival are several: by co-opting potential opposition
(Gandhi 2008b; Gandhi and Przeworski 2007; Lust-Okar 2006); managing elites in
opposition groups (Blaydes 2008; Lust-Oker 2005; Tezcur 2008); providing political 1 Surveys of the literature on the underlying logic for party formation include Magaloni and Kricheli (2010) and, on elections, Gandhi and Lust-Okar (2009).
2
information (Cox 2009); and limiting economic predation by the autocrat (Gandhi 2008a;
Wright 2004). More generally, institutions allow the dictator to make a credible
commitment to sharing power with supporters (Magaloni 2008). This solves a core
dilemma facing all dictators. In order to remain in power, the dictator must rely on some
base(s) of support, but these groups are unwilling to back a dictator who may, once in
power, renege on promises. To generate confidence in his decision to extend benefits to
those who provide loyal service, a dictator may create “power-sharing institutions” that
over time generate some confidence in a dictator’s willingness to abide by non-arbitrary
rules to the game.2
Parties are instrumental for solving this credible commitment problem. Single
ruling parties in particular allow the dictator to make credible commitments to loyalists
by promising access to locally-generated revenues or future privileges in exchange for
service in the present (Lazarev 2005; Lazarev 2007). One mechanism for this is the
allocation of party membership and positions; the privileges of party office present an
intertemporal solution to the dictator’s commitment problem. This present-service-for-
future-privileges arrangement has been tested empirically in the Soviet regime where the
party controlled all political, economic, and social mobility, but this monopoly is not a
necessary condition for the arrangement to remain credible. As the Chinese case attests,
the emergence of an economic elite that is separate from the party does not imply the end
of high demand for party office. Importantly, these institutions lengthen the regime’s
time horizon. Because of this expectation of regime durability, parties also structure
2 A dictator nonetheless possesses, in theory, the authority to abolish an institution at will, though the threat of revolt by dissenting elites and/or the general population presents one potential deterrent. It is also the case that institutions may, by design or over time, obtain their own authority, resources and legitimacy, all of which serve as bulwarks against arbitrary dissolution.
3
intra-elite conflict by offering elites the promise of “medium and long-term gains despite
immediate setbacks” (Brownlee 2007: 12). A party’s success at maintaining this
intertemporal bargain thus coincides with the party’s time horizon (or expectations of the
durability of its rule) and concomitant willingness to invest in development (Olson
1993).3
Within the menu of institutional choices facing a dictator, party creation helps the
dictator to solve several additional problems. Parties, unlike collective leadership under
military rule, generate strong incentives for party members to support the authoritarian
status quo because these party members and cadres depend on the party for rents (Geddes
2006).4 Even more, by dispersing authority over a broader political base, parties provide
a counterbalance to the threat of military coup (Geddes 2008). Parties are also guilty of
the “tragic brilliance” in which corruption and suboptimal policies are accepted by the
general population because of the party’s ability to maintain widespread patronage
networks (Diaz-Cayeros, Magaloni, and Weingast 2003). Studies have found that in
China, the narrower extension of party patronage to economic elites forges the credibility
that encourages private investment (Gehlbach and Keefer 2008).
While those who invest resources in party creation are engaging in several
tradeoffs – e.g., dispersing authority, investing limited resources in party organization
3 While party creation may mitigate this problem of incredible commitments, there are limits to this institutional strategy. Reforms in the USSR failed because CPSU authorities were unable to commit credibly to a long-term growth strategy. Instead, the party maintained discretion over whether or not to adhere to growth-promoting policies (Litwack 1991). 4 I use the term ‘cadre’ to refer to individuals who hold positions of authority – though not necessarily ranked positions – within the bureaucracies of a Leninist party-state. Lee (1991) spells out the evolution in this term, describing cadres as “people whose high level of political consciousness qualified them to assume responsibility for specific political tasks. In this original sense, cadres are the leaders, in contrast to the masses, who are the followers in a revolution. However, after the CCP became the ruling party, the meaning of cadre expanded to include all those who were paid from the state budget but not engaged in productive manual labor. Thus, the current Chinese concept of cadre includes two analytically distinct categories: the political elite and the functionaries staffing the huge party-state apparatus,” (p. 4).
4
rather than repression – this institutional choice ultimately enhances the durability of the
regime. Party creation appears to be a successful institutional strategy: in the post-World
War II period, authoritarian regimes led by a single party have enjoyed long durations of
rule in comparison to authoritarian counterparts without a ruling party.5 Among the
institutions that a dictator may choose to establish or maintain, then, ruling parties are
perhaps the most critical for understanding questions of authoritarian resilience.
While acknowledging that parties serve these important functions of elite
management and mass mobilization, this project shifts the analytical focus to problems of
party organization. In reviewing the state of knowledge on one-party regimes, Magaloni
and Kricheli (2010) state that ruling parties serve two critical functions, “a bargaining
function, whereby dictators use the party to bargain with elites and minimize potential
threats to their stability, and a mobilizing function, whereby dictators use the party
machine to mobilize mass support,” (pp. 124-5). In much of the existing literature, there
is less emphasis on drilling down into the party itself and asking questions of party
structure, the consequences of organizational design, and how these intermediate choices
lay the foundations for stable single-party rule. Rather than treating parties as monolithic
institutions, this project attempts to map a more interior terrain. Its point of departure and
focus is on variation in intra-party organization. In particular, this project will argue that
organizations located within the party contribute to solving an important selection
5 Among the authoritarian types identified by Geddes, single-party regimes have persisted for, on average 34 years, which compares favorably with the averages of 10 years for military and 18 years for personalist regimes. Regimes exhibiting characteristics of these ideal types, or hybrids, last the longest in her accounting. These averages span the period 1946 to 2000. See Geddes 2003, p. 78. Smith (2005) argues that this effect is due to the outliers of Mexico under PRI rule and the USSR, but Magaloni’s separate accounting, with its finer-grained breakdown of authoritarian regime types, supports the original Geddes (1999) finding. Brownlee (2007) also controls for economic, region, duration, and other institutional variables to find that single party regimes are significantly more likely to endure than other authoritarian regime types (pp. 30-32).
5
problem. To flesh out the contours of this problem first requires a detour through the
organization – and organizational problems – of perhaps the most highly structured of
single-party regimes, those led by Leninist parties.
Inside Lenin’s “organizational weapon”
Because of their emphasis on organization and hierarchy, Leninist party systems
present an ideal case for probing the purposes, risks, and advantages of particular
decisions in party-building. The revolutionary, and eventually totalitarian, aspirations
that motivated the creation of these parties translated in practice to party organization that
would facilitate the complete control of state and society.6 Lenin’s original conception
for the party was of an organization led by “professional revolutionaries” who were
promoted from within the “rank and file” membership.7 He wrote that “the only serious
organizational principle for the active workers of our movement should be the strictest
secrecy, the strictest selection of members, and the training of professional
revolutionaries.”8 In contradiction to egalitarian ideological commitments, the party
would be “an organization which of necessity is centralized.”9 The bureaucratic
centralism that Lenin’s party eventually embraced was done unapologetically (Wolfe
6 See Lenin’s 1902 essay “What is to be Done? Burning Questions of Our Movement” (Lenin 1964: 347-529). Nearly two decades later, in his 1918 essay “The Chief Task of our Day,” he calls for the Bolsheviks to learn from the German model, which he saw as driven by “principle[s] of discipline, organization, harmonious cooperation on the basis of modern machine industry, and strict accounting and control.” Party control of the media and cultural expression is discussed in Lenin’s “The Party Organization and Party Literature,” (Tucker 1975: 148-152). 7 “What is to be Done?” (Tucker 1975: 75-77). 8 Ibid., p. 90. 9 Ibid., p. 86.
6
1984: 24-26, 192-95).10 In conceptualizing the party he was to lead, Lenin was also
aware of the importance of flexibility in political organization. As he wrote during the
first years of the twentieth century,
It would be a grievous error indeed to build the Party organization in anticipation only of outbreaks and street fighting, or only upon the ‘forward march of the drab everyday struggle.’ We must always conduct our everyday work and always be prepared for every situation.11
The Communist Party of the Soviet Union became the organizational embodiment
of the pragmatic recommendations bound up in Lenin’s political vision. The
party was to coordinate political functions, distribute economic power, and play
the crucial centralizing role in the command economy and politically closed
system that endured for over seventy years (Klugman 1989). In theory, it was also
to possess the organizational flexibility to respond to unforeseen circumstances
and contingencies.
Following the diffusion of Leninist principles across Russia’s borders and around
the globe, these parties have become highly structured and complex organizations, with
extensive functional differentiation of constituent parts.12 A range of sub-party
organizations play a supporting role in the maintenance of the party’s political authority:
propaganda bureaus, party personnel departments, courts, unions and other mass
10 In his early theorizing about the organization of the party, Lenin held democratic practice to be a secondary concern, since “'broad democracy' in Party organisation, amidst the gloom of the autocracy and the domination of gendarmerie, is nothing more than a useless and harmful toy,” (Lenin 1964: 479). 11 Tucker, p. 110. 12 In his collected letters (Tucker 1975), Lenin expresses some antipathy toward the “bureaucratic bog” of Russia (“Letter of January 1922 to A.D. Tsiurupa”, pp. 717-18). His complaint was one of the impotence of the citizen in the face of bureaucratic authority: “The complete lack of rights of the people in relation to government officials and the complete absence of control over the privileged bureaucracy correspond to the backwardness of Russia and to its absolutism,” (“The Tasks of the Russian Social-Democrats, p. 10). While pointing out the obstacle of the bureaucracy, he is also pragmatic: “Bureaucratism cannot be ‘sent packing’ from a peasant country, cannot be ‘swept from the face of the earth.’ One can only reduce it by slow, stubborn effort,” (“Letter of May 1921 to M.F. Sokolov”, p. 714.
7
organizations, party schools, and the like. The central committee of a ruling communist
party becomes the principal to these various organizational agents, and this relationship is
mirrored at lower administrative levels in the system, forming overlapping chains of
principal-agent relationships. This parallels the principal-agent relationship between
higher-level cadres and their subordinates, for example the principal role played by a city
party committee over agents in a county or township located within the city’s jurisdiction.
The pervasiveness of these hierarchical relationships within the political system, at both
the individual and organizational levels, provides the structural basis for governance and
the distribution of political power.
Hierarchical Leninist systems are characterized by a critical relationship that is
often overlooked in general studies of parties in autocracies: party management of the
bureaucracy. Party organization, specifically party integration with and dominance over
the bureaucracy, constitutes a source of political power (Barnett and Vogel 1967;
Selznick 1960). As the most prominent example of an extant ruling party formally
organized along Leninist lines, the Chinese Communist Party maintains and reinforces its
organization through party penetration of the bureaucracy.13
While there have been attempts to draw an analytical and empirical line between
the party and bureaucracy in China (Zheng 1997), in practice the two political bodies
remain intertwined.14 Existing work on the Chinese case characterizes the relationship as
suffused with bargaining and negotiation (Lampton 1992; Lampton 1987; Naughton
13 Drawing on the cases of England and the city-state of Venice, Gonzalez de Lara et al. (2008) make the interesting argument that the possessors of administrative power, not the threat of citizen revolt, may constrain rulers. For autocrats, then, control over the bureaucracy and those segments of society with administrative capacity is a critical cooptation strategy. 14 E.g., officials who occupy party and government offices simultaneously, government funding of party bureaus, party and government training centers integrated on the same campus.
8
1992); a reflection of elite conflicts (Dittmer 1978); and, above all, distinguished by party
domination and coordination (Harding 1981; Li 1994; Schurmann 1970). In this sense,
the bureaucracy in China is not a “neutral layer” between the ruling party and the
governed but rather an instrument in the service of political power holders (Massey 1993).
At the individual-level foundations of this arrangement, who becomes a cadre is
of fundamental and paramount importance. Since “the formation of cadres is a basic task
of communist organization,” (Selznick 1960: 19), it becomes vital for party authorities to
manage who may enter and move up the ranks within this vast bureaucratic apparatus. In
this sense, the party serves a function not accounted for in the literature: it presents an
organizational means to solve a political selection problem. This function is both
separate from and part of the elite bargaining function studied by other scholars. The
party is an organized means for selecting those who are most likely to advance party
goals. In the case of China, it is in the interest of CCP authorities to devise effective
instruments for controlling bureaucrats and party organs at various levels of
administration because disciplined party agents are more likely to implement party
policies, hence extending the scope of party rule over a hierarchically organized political
terrain. In light of the critical role played by those party institutions that control who
joins the party elite, this project will focus on party strategies of both bureaucratic
management and political control.
Through interlocking but functionally specific bureaucratic organizations, a
Leninist ruling party attempts to control several overlapping groups of key political actors:
party members, rank-and-file party and government cadres, and senior (leading) party
and government cadres. Controlling promotion to and within this latter group, the senior
9
cadre ranks, is a crucial arena for the party’s maintenance of “organizational health” (Nee
and Lian 1994). This is especially critical in a system as decentralized as China’s
(Landry 2003). Inability by higher-level authorities to manage party and government
reformers is tantamount to a loss of party authority. The collapse of the Soviet Union
reinforced for Chinese party authorities the danger in, among other things, a decline in
party discipline (Shambaugh 2008a; Shambaugh 2008b; Wang 2002; Xiao 2002).
Elite personnel decisions – and the selection processes this implies – are a
paramount responsibility of the party (Naughton and Yang 2004). Furthermore, authority
relations between party managers and their subordinates are dynamic. While these
relationships are moderated by the institutions that authorities use to monitor and control
subordinates and the flow of information between levels, they are subject to the dictates
of new generations of leaders and system wide shocks – such as the transition to a market
economy, technological change, new global balances of power, and shifting international
alliances.
Placing China in context: High growth, low bureaucratic exit
While the tasks of political elite selection and party organization must be
confronted in all single-party authoritarian regimes, the CCP faced particular
circumstances and challenges at the onset of reforms, as the Chinese state was “growing
out of the plan” (Naughton 2007: 92-93). Comparatively low bureaucratic turnover
during the post-Mao economic transition generated pressures for internal updating of
cadre administrative skills. Party leaders, beginning with Deng Xiaoping, realized the
need to engineer a bureaucratic transformation to meet the demands of a market transition,
10
but political constraints were such that another purging of party managers was not a
feasible policy option. The normalization of party politics in the aftermath of the
Cultural Revolution (1966-76) and the legitimacy wielded by the old revolutionary cadre
generation limited the range of alternatives. At the same time, the demands of an
assertive economic modernization program were straining the human resources of a
political system designed to manage a command economy.
With the implementation of liberalizing economic and social reforms under Deng,
the party faced a problem: Chinese leaders realized that the party comprised a high
number of public managers with outdated and irrelevant skills. The rallying cry was to
develop a “revolutionary, younger, more educated, and more technically specialized”
(geminghua, nianqinghua, zhishihua, zhuanyehua) cadre corps.15 There existed a cadre
class that suffered from “one high and two lows” – bureaucrats were, on average, too old
(i.e., their age was too high) while their education and professional skills were
insufficient (Lee 1983: 676).
This bureaucratic transformation was to take place in the context of
unprecedented economic development. Since the initiation of reforms in 1978, China’s
economy has undergone dramatic change in terms of growth and industrial development.
Official annual growth rates stand at 9.6 percent on average for the period 1978 to
15 See Manion (1985) for a discussion of the personnel policies resulting from these “four transformations” (si hua), Deng also wrote in an essay dated November 2, 1979, “But the problem we face is the shortage of younger, professionally competent cadres. Without them it will be impossible to carry out the program of modernization.” See Deng Xiaoping, “Senior cadres should take the lead in maintaining and enriching the party’s fine traditions,” available online at http://english.peopledaily.com.cn/dengxp/vol2/text/b1360.html, accessed 24 February 2010. This idea of well-trained and professionalized cadres leading the modernization drive was repeated in a speech before the Poliburo, where Deng’s opening remarks linked China’s economic development and political advancement with the “urgent need to discover, train, employ and promote a large number of younger cadres for socialist modernization, cadres who adhere to the Four Cardinal Principles and have professional knowledge.” See “On the reform of the system of party and state leadership,” available online at http://english.peopledaily.com.cn/dengxp/vol2/text/b1460.html, accessed 6 March 2010.
11
2007.16 Industrialization has also taken off in urban Special Economic Zones and
through collective and “local state corporatist” strategies in the countryside (Liao, Pan,
and Zeng 1999; Oi 1992; Oi 1998a). This authoritarian state-led development has
challenged assumptions about the correlates of economic growth. Modernization
theorists posit a positive, causal relationship between economic development and
political liberalization, (Lipset 1981; Rostow 1960: Chapter 2), and this has found some
support in more updated analyses of the correlates of democracy (Geddes 1999;
Przeworski et al. 2000). The economic miracle presented by contemporary China has a
seemingly incongruous basis in a single-party authoritarian regime, which begs further
examination of how the ruling CCP has remained relevant and maintained organizational
discipline during this period of rapid and apparently successful economic liberalization.
One way to place the Chinese case in its comparative context is to contrast the
problem of low administrative turnover in China with the transformations taking place in
other communist party states such as the Soviet Union and Eastern Bloc countries of
Central and Eastern Europe. This is an imperfect comparison due to the simultaneous
political and economic transitions that took place in Western communist party states, but
in all cases bureaucratic transformation was a requirement for successful economic
reforms. In each country, engineering a revolution in bureaucratic talent was also
complicated by the lure of new market opportunities. As part of their emphasis on party
control over all political, economic, and social activity, Leninist party-led systems
traditionally rely on monopsonistic control by the party over labor markets.17 Over the
16 “Industrialization on track: report,” China Daily, 29 January 2007, http://www.chinadaily.com.cn/china/2007-01/29/content_794936.htm, accessed 20 December 2008. 17 I.e., the party was the sole buyer in a labor market comprising many sellers of political, managerial, and administrative talent.
12
course of market reforms, skilled labor that was formerly dependent on state entities for
upward mobility found new exit options in newly created non-state sectors. When
compounded with political reforms and, ultimately, revolutions such as those in Eastern
Europe and the Soviet Union, there was considerable turnover in the bureaucratic ranks.
This is observed in data from a survey of post-communist countries taken in 1992 and
1993.
Table 1.1 below presents rates of bureaucratic turnover, ranging from a low of
approximately 25 percent in Russia to 51 percent in the Czech Republic by 1993.
Table 1.1: Turnover in party and state managers, 1988 to early 1990s
Country
Percentage of individuals in cadre positions in 1988
who reported the same occupation in 1992
Percentage of individuals in cadre positions in 1988
who reported the same occupation in 1993
N Bulgaria 66.9 64.6 131 Czech Republic 58.7 49.0 208 Hungary 66.7 57.4 129 Poland 76.0 68.7 179 Russia 78.7 74.9 211 Slovakia 67.3 59.5 153
Source: Social Stratification in Eastern Europe after 1989: General Population Survey, 1993. I defined ‘party and state managers’ as individuals reporting occupations which fell in the ‘legislator and manager’ category of the survey (ISCO codes 1000 to 1319). Unfortunately, this data does not capture whether communist party-era cadres opted for
non-party and -state sector professions over the course of transition or were ousted by
incoming political elites and forced to turn to private sector alternatives.
China, in comparison, realized much lower rates of bureaucratic turnover during
its reform period despite the option for cadres with managerial experience or connections
to “jump into the sea” (xia hai) of capitalism. Table 1.2 below summarizes occupational
change among cadres from 1988 to the early 1990s as well as change from the onset of
reforms in 1978 to the early 1990s.
13
Table 1.2: Turnover in Chinese party and state managers
In China,
…continued to report a party or government rank in 1992
…continued to report a party or government rank in 1993
N Percentage of individuals in cadre positions in 1978 who… 85.8 84.2 183
Percentage of individuals in cadre positions in 1988 who… 93.7 92.7 381
Source: China General Social Survey 2003. I defined ‘party and state managers’ as individuals responding to a survey question on administrative rank in 1988 and 1978 and reporting non-party or -state occupations in 1992 and 1993.
Even as the CCP withdrew from its monopoly on economic opportunity, exit by
bureaucrats to the private sector was rare. Turnover from 1978 to 1992 and from 1988 to
1992 was due entirely to retirement.18 Only one cadre reported joining the private sector
during the period between 1978 and 1993; the rest was due to retirement. This pattern
was repeated in the cadre population from 1988 to 1993, with only two individuals
reporting private sector occupations by 1993 and the remainder leaving the cadre ranks
due to retirement.19 These patterns may be explained by the particular fiscal incentives in
place for cadres to stay in the system and realize benefits from profit-sharing contracts
with party authorities and/or party-sanctioned extrabudgetary revenues (Ang 2009b;
Solnick 1998: Chapter 7). Such high retention rates may bode well as an indicator of
party legitimacy, but this low turnover pattern has left party leaders with the problem of
how to retrain the administrative class to cope with the implementation and management
of economic and social reforms.20
18 This is 26 and 24 individual retirees, respectively. 19 This was 28 retirees in 1978 and 26 retirees in 1988, respectively. 20 One possibility was recruiting those with relevant management skills from the newly-created private sector. Scholars have examined how the party has attempted to absorb capitalists in the reform era (Dickson 2003; Tsai 2005; Zheng 2006). In 2000, Jiang Zemin’s “Three Represents” declaration that the party should represent the most productive forces in society (i.e., capitalists) reflects an important moment in party adaptation. Still, China does not yet have a “revolving door” between political office and the private sector, and entrepreneurs rarely become cadres (Interview 82, with a Central Party School professor, December 2007, and Interview 212, city organization department official, April 2010).
14
This project will address whether and how the party adjusts its organizations of
bureaucratic control and management to account for new elite selection criteria and the
increased decision-making autonomy that accompanies market reforms. In a Leninist
system in particular, party organizations designed for a command-based economy and
ideologically disciplined cadre corps would seem outdated and out of place in a
decentralized market setting. This prompts several questions aimed at probing strategic
choices made by an authoritarian ruling party: How does the party maintain the relevance
of its political organizations during a period of economic transition? In the case of China,
has the CCP replaced, reformed, or otherwise rejuvenated inner party organizations
during the past three decades of economic growth? Throughout this transition, how does
the party maintain control over the bureaucracy? In particular, how has the CCP sought
to reshape and control its powerful cadre ranks in such a way that the party’s
administrative capacity is not diminished?
The organizational puzzle posed by the case of the CCP is the persistence of
seemingly anachronistic party organizations in the post-Mao period. The party needed to
adjust its Leninist organizations to a market context in which institutions that centralize
decision-making, such as the plan and the collective, have been replaced in large part by
market mechanisms. The old standbys, Marxist-Leninist tenets and Mao’s writings,
would not be enough to guide cadres’ administrative decisions in a market economy
characterized by expanding trade, new forms of industrial production, and increasing
global exchange. “For managers, entire careers spent learning how to maneuver through
the planning bureaucracy to obtain scarce materials, to lower plan targets, to lobby for an
15
increased wage fund, and so on become irrelevant to success in a marketized context,”
(Hanson 1995: 312).
While the education system could take on some of this burden of re-educating the
administrative class, regular universities and schools might not promote entirely ‘correct’
ideas.21 In seeking to exert party control over individual bureaucrats, it would seem
logical for party officials to draw on existing organizational resources. An examination
of the CCP’s cadre training system reveals how the party has sought to retrain, manage,
and select administrators during a period characterized by dramatic economic growth,
low exit from the cadre ranks, and the need for skilled public managers.
Party schools, party reform
Party schools, the organizations responsible for the ideological training of
revolutionary CCP cadres, serve as a well-situated case for examining how the party has
generated incentives for party organizations to respond and adapt to the new demands of
the post-Mao reform period. Numbering nearly 3,000 campuses nationwide, party
schools constitute an extensive network of training academies for China’s political class
(Appendix A). Party schools are the anchor within a larger category of organizations
charged with cadre training.22 The centrality of party schools within this organizational
landscape is due to their exclusivity, since they are officially reserved for party members
and, within this population, those holding party office.23 These schools are a prime
21 See, for example, comments by a provincial party school teacher: “Ideological training must be preserved (baozhang). You can’t have liberal-minded (ziyou zhuyi) university teachers teaching cadres, this task can’t be divided [between institutions],” Interview 211, May 2008. 22 This includes cadre schools located in party organs such as SOEs, socialism schools, and communist youth league schools. At one time, the total number of cadre training organizations numbered over 11,000 (Central Party School yearbook 1985; Shambaugh 2008). 23 Party schools such as those located within universities are also tasked with training party activists.
16
example of Leninist party organizations that would appear incompatible with a market
economy and its accompanying social context. Yet, economic reforms and
modernization in China have brought into focus the need for new administrative capacity
within the bureaucracy of the party-state, and reforms in cadre training would be one way
to meet these demands, however unclear the path to realizing these new party objectives.
I argue that by altering incentives while leaving Leninist party organization intact,
the CCP has managed in the post-Mao period to induce organizational adaptation that has
bridged, however incompletely, the disjuncture between new realities and prior
institutional arrangements. As the following analysis of party schools will demonstrate,
this adaptation is a result of deliberate marketization by central party authorities and the
introduction of organizational competition, or redundancy, to the system. Before diving
into the logic behind these changes, I first explore in more depth why party schools
constitute key sites for understanding the mix of organizational continuity and
adaptiveness observed in the CCP.
Placing party schools in context: Changing forms of political control
It is not the case that all institutions of political control within the party have
remained as robust as party schools during the reform period. In this sense, the story of
party school survival provides important clues regarding internal processes of party
reform. The resilience of party schools, and the criticality of cadre training for China’s
transition, stands in contrast to the waning of other key institutions of political control,
17
i.e., political campaigns (yundong).24 The relative decline of campaigns in the economic
transition period highlights the centrally-directed nature of organizational change and the
capacity of central party leaders to shape the geography of political institutions in
accordance with new preferences and realities. Campaigns, which were most effective
during the “mobilizational phase” of the party, have waned in centrality as the party
completes its transition to an “institutionalization phase” (Li and Bachman 1989: 91; Li
and White 1988).
In the past, the staging of periodic political campaigns, as a form of “internal
remedialism” (Harding 1981), had been a powerful means for the CCP to communicate
“policy winds” and shape the actions of party members, bureaucrats, and citizens.
Campaigns entailed an intensification of activity in order to meet goals that were often
economically and politically transformative in intent. Such mass mobilization served
several functions: political socialization, policy implementation, and as a way to realign
party ranks (Zheng 1997: 153-158). Individual performance during campaigns – as
indications of political intelligence and ability to meet campaign goals – factored into the
career prospects of political leaders (Bo 2002). During the Mao era, political
mobilization requiring broad public participation and demonstrations of party loyalty by
bureaucrats created a tension between mass unrest and party control, but this has
historically been resolved in favor of party authority (Townsend 1969). Selznick
likewise views Soviet campaigns through a cynical lens, arguing that mass mobilization
was a tool of elite control, deployed to manipulate an “unstructured” and “alienated”
population that has been “absorbed into the organization” (Selznick 1960: 288-9). In the
24 Cell (1977) defines a campaign as “organized mobilization of collective action aimed at transforming thought patterns, class/power relationships and/or economic institutions and productivity (p. 7). For a list of campaigns under Mao, see Cell, Appendices 1-3.
18
end, despite the intense levels of organizational activity required by political campaigns,
scholars have found that this political tactic did little to realize the goal of genuine
attitude transformation (Whyte 1974).
Chinese citizens exhausted by the seemingly endless campaigns under Mao found
relief when Deng rose to power, though there remained the reflexive deployment of a
nationwide anti-bureaucratic campaign in 1977 to unify the bureaucracy in the wake of
the Cultural Revolution (Morgan 1981). In 1979, Deng called an end to reliance on
political campaigns and exhorted comrades to turn instead to the tasks of economic
development:
In addition to economic work, the Party committees perform many other kinds of work, but many issues involve economic affairs. .... Instead of conducting campaigns, such endeavors should be accomplished through routine and chiefly economic work.25
On the heels of Deng’s call for the cessation of campaigns, party authorities issued
official condemnation of the intense political mobilization of the Cultural Revolution and
sought to restore order to bureaucratic ranks (CCP Central Committee 1981).
Campaigns were not eliminated entirely from the political landscape, as illustrated
by party mobilization to enforce family planning policies and maintain the vanguard
nature of the party, but they have waned in frequency and intensity (Manion 1985;
Shambaugh 2008a; White 1990).26 Institution-building within the party has replaced ad
hoc political mobilization, representing a normalization of political life. Campaigns are
25 “Some Comments on Economic Work,” Speech by Deng Xiaoping, 4 October 1979, published online by the People’s Daily, Selected Works of Deng Xiaoping, Volume II, 1975-1982. http://english.peopledaily.com.cn/dengxp/vol2/text/b1330.html, accessed 21 May 2007. As early as 1956, Liu Shaoqi stated, in his political report to the 8th CCP Congress, “Mass struggle must give way to the rule of socialist law and order.” This statement was given in the context of political struggle between Liu and Mao, which Mao ultimately won. 26 O’Brien and Li reference rural public opinion survey results to argue that “campaign nostalgia” has emerged, most often with respect to cadre malfeasance, though the breadth of this sentiment across the general population is unknown (O'Brien and Li 1999).
19
no longer intended to engineer a wholescale transformation of society so much as target
particular political issues. Anti-corruption campaigns are one example. These have
persisted in the post-Mao period, but they differ from the mass campaigns of the past by
relying instead “on short bursts of hyper enforcement by state and party agencies,
wherein the ‘masses’ are asked to report corrupt cadres but are not allowed to take on an
active or leading role,” (Wedeman 2005: 93, fn. 1). Political campaigns have left their
organizational legacy, since the organization of small groups continues today in more
institutionalized settings.27
More orderly forms of party management, such as cadre training, have risen in
prominence. Like campaigns, training is a form of ex ante political control: it offers
general prescriptions for action and increases local awareness of general legal (or
normative) boundaries set by higher level authorities.28 These qualities are particularly
suited for transitional environments in which local circumstances are shifting rapidly and
local agents must act autonomously based on knowledge of central goals.29 Cadre
training, like campaigns, provides individuals with incentives to promote central political
goals. “In the cadre party, with its heavy emphasis on indoctrination and institutional
character-formation, this means that party members may be relied on to carry out party
policy even under conditions which do not permit direct control over the member by
regular party organs,” (Selznick 1960: 65). Unlike cadre education at training schools
27 See Whyte (1974) for a discussion of the (ideal and actual) purpose of small groups under Mao. The study groups he describes engage in more coercive activities such as intense group- and self-criticism, which have waned in the present: the organizational form remains, though the content has changed. 28 On the distinction between ex ante versus ex post controls, see Huber and Shipan (2002). 29 In his seminal study of behavior among U.S. forest rangers, Kaufman identifies training as the vehicle for shaping rangers’ decision making under highly autonomous circumstances (Kaufman 2006). Few public servants are as self-directed or isolated as rangers, but in times of heightened uncertainty higher-level authorities must rely more heavily on the knowledge of local agents and grant them more decision making authority.
20
and universities, however, campaigns are not amenable to organizational competition and,
by implication, lack the capacity for local innovation that these organizations may
develop.
An alternative explanation: Party schools as a reward for party service
In the reform period, central CCP leadership has tackled two critical tasks:
attempting to maintain internal control over ‘leading’ public managers and party organs
while seeking to engineer a bureaucratic transformation. Cadre training, and the party
school system in particular, brings together these tasks. Before attempting to map the
contours and mechanisms of this control and adaptation, I must first address a key
critique. There exists the prospect that party schools are outdated institutions with an
irrelevant mandate, anachronisms in China’s market-oriented development agenda. This
suggests the null hypothesis that party schools are inconsequential for understanding
internal dynamics of party change, that training is epiphenomenal to the true bases of
party authority. At the individual level, this implies that training has a negligible effect
on the careers of upward bound officials (who are already on track for promotion).
A potential problem with a proposed study of cadre training – and the party
school system at its core – lays in the risk that such institutions exist solely as a reward,
or “club good” for the party faithful. Historical accounts link the initial creation of
China’s party school system to its Soviet counterpart, which suggests international
institutional mimicry as the justification for their existence (Chen 2007: Chapter 4).30
30 Party schools are not unique to China and have existed in the Soviet Union and its political satellites in Central and Eastern Europe and Asia. In the USSR, party school education was mostly remedial and not a significant channel to the most prestigious and lucrative offices at the central level (Farmer 1992). Party school counterparts in Russia and Eastern Europe have, remarkably, outlasted the party system: the central
21
From this perspective, the party school system derives legitimacy from an (extinct)
external source and may thus be both hollow in purpose and motivated by the need to
continue perpetuating a myth of its own usefulness in order to survive (Meyer-Sahling
2006).
From this perspective, the primary purpose of party schools is to provide material
rewards to party managers who have demonstrated loyalty to the party. Ritual and
reward suffuse the training experience. For example, county-level trainees at the Central
Party School must swear allegiance each day of training to the Chinese flag.31 Other
trainees who attend programs at the new cadre executive academy at Jinggangshan are
issued revolutionary-style fatigues and backpacks and led on treks up Jinggangshan in an
attempt to capture the spirit of revolutionary survivors before them.32 These efforts
reflect an attempt to reignite a revolutionary fervor among cadres too young to have lived
through the civil war leading up to the founding of the PRC in 1949, but this has the
effect of reducing the dense relationship between myth and experience to “revolutionary
tourism” (Pieke 2009a: 184). On the reward side of the equation, the material comforts
of party school training assignments are difficult to ignore. During in-residence training
classes at party schools from the central level down, cadres enjoy state-of-the-art leisure
facilities which range from indoor basketball courts to hotel-style banquet halls.
If cadre training at party schools is primarily a reward for loyal service to the
party, two important inconsistencies arise. First, cadres’ performance in training
Russian party school in Moscow has since become a school of public administration and joined the market for professional higher education. Romania’s central party school, the Stefan Gheorghiu Academy for Training and Development of Leading Cadres, also reinvented itself as a secular public administration academy in 1989. 31 “Five thousand county officials big training-in-rotation,” Southern Weekend, 4 January 2007. 32 Interview 114, Central Party School professor, party building department, February 2008.
22
programs are recorded in their personnel files, and these training records are reviewed
during regular evaluations.33 The Trial Regulations on Cadre Education and Training
Work (2007) state that “each cadre education and training department must in
accordance with cadre management powers create and perfect a cadre education and
training file and submit cadres’ training situation and evaluation results,” (Art. 43).
Thesis papers and field reports written by cadres during training must also be submitted
to these personnel files, though interviewees have noted varying degrees of seriousness in
the grading of these papers. More importantly, the personnel management agencies of
the party, organization departments, will send monitors to observe and record cadres’
performance during training.34 Overall student performance in training, including
appearance (biaoxian) throughout discussion sessions, field investigations, and on-site
visits, all are submitted to each cadre’s personnel file. In short, while there are elements
of ritual to party training, there are also organizational links between training programs
and a cadre’s career prospects within the party and government.
Second, the view that party schools should be dismissed as peripheral
organizations of marginal purpose is inconsistent with internal and external efforts to
update the relevance of party school training for reform-era public managers. Party
schools have sought to adjust their pedagogy, training materials, teaching staff, and
training content to meet perceptions of shifting needs in a rapidly changing context. This
suggests an organizational objective that extends beyond ritual or reward. Schools may 33 A certain number of points are allocated to training in recent performance evaluation rubrics. Interview with a county-level city organization department cadre, Interview 206, May 2008. According to this official, individual cadres have a “study and training” (xuexi peixun) component to their evaluation package. Individuals require at least 60 points to pass, and training performance is worth up to 20 points. 34 Only one interviewee, at a department (chu)-level training at the CPS, said that discussion sessions were tape recorded (Interview 40, October 2007). Interviews with party school officials and organization department officials at the city level and below indicated that organization department monitors are sent to take handwritten notes.
23
beat the drum of Marxist-Leninist-Maoist doctrine and provide some sort of retreat
experience for hard-working cadres, but they may also fulfill purposes above and beyond
these. The entrepreneurial streak that party schools exhibit today in both training content
and income generation would be even more puzzling if these schools were only places of
respite. Party schools on one level are the embodiment of the perks of party membership,
but evidence suggests that they serve additional, more important functions for the ruling
party, e.g., managing the career chances of upwardly mobile cadres and conveying party
goals to these increasingly autonomous, market-oriented agents of the party.
In short, party schools are more than a “perk” of party office. By drilling down
into these core party organizations, I seek to explore how the party school system has
adapted to the changed conditions of the reform period while contributing to the
continuity and stability that are significant features of the CCP’s bureaucratic structure.
One key driver of party school adaptation is the introduction of new incentives for
internal organizational innovation. These incentives have spurred the reform of existing
party organizations. Rather than destroy and rebuild its bureaucracy as other communist
party states did in their dual transitions to democratic rule and market capitalism, the
Chinese party-state has transformed its bureaucracy from within, through a focus on
reforming specific party organizations. It has done so through the introduction of
competition, the logic of which is spelled out in organization theories on redundancy.
2 Inducing organizational adaptation: Competition and redundancy
The ideal Weberian bureaucracy, indispensable to the modern state, is at its core a
legal-rational structure. Rules and expertise govern bureaucratic behavior and
24
administration, while organizational rationality derives from functional specialization
across bureaucratic offices. Leninist parties incorporated many of these features, notably
the organization of party units according to the various functional needs bound up in the
revolutionary transformation of society and, ultimately, in the more mundane tasks of
governance. Each core task of the party-state would have its own bureaucratic proxy, a
bureaucracy within the bureaucracy. A central organizing principle is bureaucratic
monopoly according to functional domain. While this lends coherence to the
organization of the party-state and facilitates the assignment of both responsibility and
blame, it is problematic from the standpoint of adaptability to change. A monopoly lacks
strong incentives to innovate since there is inelastic demand for its output. In this sense,
monopolies have a predisposition for “the quiet life,” and innovate rarely because they do
not employ the same “diversity of processes” found in a competitive system (Niskanen
1971: 161).
In the post-Mao period, broad-based systemic changes have tested the party’s
flexibility and adaptiveness. The general problem to overcome is one of “trained
incapacity”, where party functionaries reach a “state of affairs in which one's abilities
function as inadequacies or blind spots. Actions based upon training and skills which
have been successfully applied in the past may result in inappropriate responses under
changed conditions. An inadequate flexibility in the application of skills, will, in a
changing milieu, result in more or less serious maladjustments,” (Merton 1968: 252).
The antidote to the individual- and organizational-level dysfunction that Merton observed
across bureaucracies is creating incentives for “adaptive efficiency”, i.e., “an institutional
structure that in the face of ubiquitous uncertainties … will flexibly try various
25
alternatives to deal with novel problems that continue to emerge over time,” (North 2005:
154).
The CCP response to these institutional weaknesses has been to modify the
Weberian and Leninist emphases on functional monopoly: with the onset of reforms,
central party authorities have promoted inter-organizational competition to cope with new
economic and social uncertainties. This reflects the logic that the reliability of a system
of imperfect, and hence fallible, parts may be increased through the introduction of
competition or redundancy (Landau 1969).35 Redundancy, taken to the realm of
governance and public administration, refers to the introduction of additional agencies to
fulfill an organizational goal previously monopolized by a single agency.36 It applies in
all cases where agencies “make some contribution to the achievement of the system’s
goal, but this contribution is blurred because some other element(s) make(s) a similar
contribution,” (Felsenthal 1980: 248). In this sense, redundancy is the introduction of
slack, or additional resources, to a system (Landau 1991). This slack then generates the
35 Redundancy is often used interchangeably with competition, but they are not precisely the same thing. Bendor (1985) notes, “All competitive structures are redundant, but the converse is not true; there are non-competitive types of duplication,” (p. 54). Competition is thus a subset of the possible universe of redundant systems. The difference lays in the nature of the incentives driving competitive systems. Competitive, as opposed to non-competitive redundancies, offers stronger incentives to individual agencies. Competitive systems imply a rivalry between actors, since they must compete for finite resources, and this serves as a stronger incentive to search for alternatives and innovate in the face of problems. The drive to innovate is reduced or even nonexistent in systems where agencies function in parallel without any mutual threat to survival. 36 The most detailed analyses of bureaucratic competition have focused on public administration and institutional design in democracies, but the principle is not dependent on the regime type of a polity. In Landau’s classic 1969 essay, he raises the many examples of redundancy in the US political system: “separation of powers, federalism, checks and balances, concurrent powers, double legislatures, overlapping terms of office, the Bill of Rights, the veto, the override, judicial review, and a host of similar arrangements,” (p. 351). See also Mittal (2008) for a historically grounded discussion of the founding fathers’ intention to embed redundancies in the US political system in order to increase the adaptive efficiency of the structure overall. However, the decision to introduce redundancy or competition in a bureaucracy is also not limited to democracies. Inter-agency competition may develop accidentally or by fiat. Downs (2008) points out that redundancies, at the agency level, arise most often when there are unclear boundaries between agencies; this provides the conditions for the pursuit of expansionist agendas. Redundancy may also be introduced to a system through executive order. These are not mutually exclusive pathways, and both may play a role in the development of competition in a bureaucracy.
26
reserve capacity that enables a system to become more tolerant of failure. Redundancy
thus produces two important results: increased system reliability and incentives for
organizational adaptation.
While reliability, or the mitigation of system failure, is the more widely
researched benefit of redundancy (Streeter 1992), this study focuses on a second, but
equally important outcome: competition as a means to induce organizational change and
adaptation to unforeseeable circumstances. In their study of inter-service rivalry in the
US military, Enthoven and Rowen (1959) argue that “human limitations being what they
are, there is good reason to believe that a decentralized competitive system, in which
people have incentives to propose alternatives, will usually meet this test [of developing
comprehensive defense capabilities] more effectively than a highly centralized system,”
(p. 5). Competition increases the diversity of perspectives brought to bear on a particular
issue, which increases the chances of discovering new alternatives. Because competition
entails some ambiguity in the jurisdictional boundaries between bureaus, some blurring
of organizational purpose, this “loosens structure” and “facilitates an expansion of the
range of possible organizational responses to problems” (Lerner 1978: 20). One of the
most powerful effects of the introduction of competition to a system, then, is to stimulate
change in preexisting actors.
The rationale behind the introduction of competition is to raise a system’s overall
capacity to generate multiple alternatives for solving a problem. In this sense, it presents
a logical response to the uncertainty that waxes and wanes in different policy contexts.
Furthermore, competition induces rival agencies to search more aggressively for
alternatives. By increasing the number of agencies focused on a task, a greater number of
27
possible alternatives are considered and pursued in the interest of fulfilling a system wide
goal. High uncertainty obtains in the case of post-Mao China. During this period, party
authorities have run into deep uncertainty over how to cross the river of economic change.
The party leadership has set policy by “feeling for stones” each step of the way, and this
oft-invoked metaphor captures the party’s heightened uncertainty over policy and
governance matters in recent decades. Introducing competition to areas deemed critically
important to party rule thus increases confidence in the ability of the system to weather
through unpredictable environments.
A case study of cadre training in China contributes to understanding the processes
of creating a competitive system where there was previously monopoly. Introducing
competition, and not only uncompetitive redundancy, to party organizations in China has
resulted in party entrepreneurialism. Party entrepreneurialism, as a response to market-
based competition for resources, encompasses a bundle of interrelated activities that
include service provision, programmatic innovation, and the search for lucrative new
ventures. Some of these activities reflect significant changes in the substance of cadre
training in China, and other activities are more limited (and local) in scope.
This study departs from previous studies of competition within the Chinese
bureaucracy in several ways. First, previous studies have not examined the introduction
of competition to party organizations with primarily political, rather than economic,
purpose(s). Mertha (2006), for example, maps the emergence of a “policy enforcement
market” across the PRC’s intellectual property rights protection regime and finds that
redundant systems are more effective at monitoring compliance with state regulations.
Second, existing studies do not explore how the introduction of competition, in the form
28
of a partially open market, leads to significant organizational evolution within party
organizations. By peering inside party schools and mapping the range of organizational
activities therein, this study offers concrete indicators of change. Third, the competition
presented here is broader in scope and spills beyond the boundaries of the bureaucracy:
the market for cadre training encompasses party, state, private, domestic, and
international actors.
Crucially, competition in cadre training brings market principles into the heart of
party activities. In Mertha’s examination of copyright and trademark enforcement, the
entry of competition is highly curtailed by central authorities. Cadre training, in contrast,
saw the creation of a training market in which virtually any educational institution could
compete. This mix of public and private actors, all vying for the privilege to train
China’s political elite, should in theory yield new approaches to training itself. Finally,
more than policy implementation is at stake in this process. Marketization of cadre
training addresses issues of organizational survival as well as competing visions for the
skills and loyalties that the party’s elite should possess.
It is important to note here that the introduction of competition to a particular
bureaucratic function does not imply privatization. Competition may take place solely
between government and/or party bureaus, and this should still yield the outcome of
greater system adaptiveness and innovation. Introducing private actors is one option
among many for diversifying the range of players in a competition, which in turn should
increase the pressures for organizational creativity (Miranda and Lerner 1995).
Subjecting a bureaucratic agency to competition implies greater diversity in
organizational activity, the search for an edge over rivals, and ultimately some innovation
29
at the system level, but the participation of private market actors is not a necessary
precondition for these processes to unfold.
There is further logic to the introduction of market-based competition. First,
“markets promote high-powered incentives and restrain bureaucratic distortions more
effectively than internal organization,” (Williamson 1989: 150). Because some degree of
risk accompanies market-based competition, the stakes are higher than for a monopolistic
bureaucracy. From the perspective of participating agencies and institutions, the risk of
failure implies that they must compete for scarce resources. Second, a market in which
competitors enter and exit freely may mitigate the problem of determining when there are
a sufficient number of actors in a system. One predicament facing planners of a
redundant system is ascertaining how many agencies, or how much redundancy, is
optimal. One solution is to apply a satisficing principle, or ceasing expansion when some
minimum threshold of competition is reached (Simon 1979). A market configuration
presents a more self-regulating solution. Where there is market-based competition,
agencies will continue to enter the market for a particular good or service until there is no
longer a marginal gain for additional entrants. This depends, however, on the freedom of
market entry. As will be discussed in the case study of the party school system, there is
some interference by central party authorities and hence only partial marketization.
Some additional design considerations accompany the introduction of redundancy
to a bureaucratic system. Competition will lead to the highest levels of creativity when
three conditions prevail.37 Downs (1967) asserts that competing agencies must be close
enough in purpose that their funding derives from the same sources. This transforms
37 Bendor (1985) explores the criticality of independence across agencies, under the assumption that nonindependence might risk to the spillover of failure across agencies, but he finds that nonindependence is difficult to achieve in practice and that the usefulness of redundancy still obtains in cases of overlap.
30
competition into a zero-sum scenario, which raises the stakes for success and agency
survival. Second, these agencies must be distant enough in purpose that there is no
significant exchange of personnel between them. Third, rival agencies must possess
discretion over which programs to pursue (Bendor 1985). Krause and Douglas (2003)
have also argued convincingly that competition is effective only when new entrants offer
alternatives that are of similar or higher quality than the original monopoly agency. If
competition presents an inferior standard, this has the perverse outcome of lowering
standards throughout the system.
There exist several problems with the introduction of competition to a
bureaucracy. First is the possibility of unpredicted interactions between agencies in a
competitive system and the unknown outcomes these may produce. Ironically, while
redundancy may be introduced to mitigate uncertainty, it can introduce uncertainties of
its own. Another concern is the opportunity cost of devoting resources to a redundant
function when those resources might be committed elsewhere. The problem with
assessing this critique, however, is the difficulty in assigning costs to a given outcome as
well as observing the counterfactual case. There are also the deeper considerations such
as coping with the possibility of market failure and the suitability of redundant systems
for nonexclusionary goods.38 There is no easy means to dismiss these issues. Safeguards
against market failure will inevitably constrain the extent of competition that is possible
or safe to introduce into a system. Nonexclusionary goods, on the other hand, may be
amenable to competition. Classic examples, such as defense and security, do contain
high degrees of overlap in agency jurisdiction (Bendor 1985: 3-22; Felsenthal 1980).
38 Landau 1991 raises but skirts these issues in his advocacy for introducing redundancy as a virtue and not a sin of public administration design.
31
Overall, these critiques present a sobering limit on those who would propose the
universal application of redundancy to government functions.
While the discussion thus far has emphasized the advantages of competition to
bureaucratic management, it cannot be imposed without expectation of resistance or
complication. Introducing redundancy to a system entails a series of strategic decisions.
Principals must first decide whether or not to create a redundant system at all, whether to
assign agents to similar tasks or otherwise determine the range of choice, and then agents
must choose how much effort to expend based on their particular policy preferences
(Ting 2003). This sets up the expectation that the “old guard” will resist the introduction
of competition. The tendency toward bureaucratic monopoly in Leninist systems would
also seem to advantage those agencies in a monopoly position. Whether and how
monopolistic agencies resist the introduction of competition is thus another focus of this
project’s empirical case study.
In sum, introducing competition may appear to fly in the face of the bureaucratic
tendency toward monopoly, particularly in a highly centralized, authoritarian regime.
Yet, as Bendor (1985) points out, this preference is not based on empirical tests of the
various advantages of monopoly over competition in matters of public administration:
“the empirical warrant for monopoly in government … is virtually nonexistent,” (p. 252).
Crucially, Niskanen (1971) finds that a monopolistic public bureau is not more efficient
than systems with overlapping or competitive bureaus.39 Aversion to innovation by
monopoly agencies within the CCP lies at the heart of the party’s concern with the old
39 He defines efficiency in terms of the production costs for a good or service, though he notes that the problem of oversupply still exists in competitive bureaucratic systems. I also note here that he uses a Weberian definition of bureaus as organizations that do not allocate any difference between revenues and costs as personal income, which is violated in the case of Chinese bureaus. See Ang 2009 for a discussion of Chinese exceptionalism in this regard.
32
state of affairs in cadre training. As this study demonstrates, central party authorities
deliberately turned to market-based competition to induce change in party organizations,
indicating the party’s capacity for significant organizational rethinking behind the veneer
of a relatively unchanged political structure.
3 Précis of study
Across authoritarian regimes, parties fulfill – to varying degrees – rulers’ search
for institutional solutions to the basic and centrally motivating question of how to survive
in office (Geddes 1994). Parties make possible the credible commitment that is
instrumental in a ruler’s attempt to co-opt elites and the general population. Parties are
also a means to solve the elite selection problem that exists in hierarchical political
systems. The contribution of this project is to identify how the party, as a complex
organization comprising many sub-party bureaucracies, addresses the latter problem.
That the party does so in a changing environment has brought about, in the Chinese case,
a shift in the incentives that structure the decisions of key party organizations of elite
selection. China thus presents an important case for understanding these questions of
party organization and change within its broader context of domestic and global
transformation. Previous research on the Chinese case has drawn attention to
adjustments in the criteria guiding the selection of political elites, but there has been less
attention on the organizational changes taking place within party agencies. Party schools,
as relatively understudied sites of political control and bureaucratic management, offer a
window into this restructuring of incentives and how the CCP has exhibited surprising
adaptability in the face of significant economic and social change.
33
I will argue in this study that the CCP has selectively enhanced the adaptability of
sub-party organizations by applying a principle more commonly studied in democracies,
inter-organizational competition. In the Chinese case, this competition has been broadly
market-based and spilled beyond the boundaries of the party-state, but the central
motivation has been to improve the party’s adaptive capacity. This argument raises
several questions: What were the processes by which party authorities introduced
competition to a bureaucratic realm previously dominated by one set of party
organizations? Who was allowed to compete and why? What have been the
organizational responses to this competition? What are some ways to measure
organizational change? Has the goal of stimulating organizational innovation been met?
Have there been unintended consequences, either welcome or not? Are party schools still
relevant? In answering these questions, I will present findings at two levels of analysis.
To show that party schools are an organized means for the party to manage processes of
political elite selection, I will consider individual-level career patterns and the
“treatment” of party school training on career paths. At the same time, in carrying out
this selection function the party school system has been subject to competitive pressures.
To understand these changes, I embark on a school-level analysis to map out
organizational responses to centrally-driven policies.
Chapter two of this dissertation will begin with an overview of the party school
system, its history, and organizational context. Existing research on party schools is
divided into roughly two groups: studies that focused on the functions of the Central
Party School (CPS) and, more recently, those that have looked at the school system
beyond Beijing. This scholarship has viewed party schools as indicators and drivers of
34
ideological change within the party. My study takes a different tact and emphasizes
processes of organizational change as they unfolded throughout the system, in the CPS
and beyond. In so doing, I introduce a key selection function that these schools are able
to carry out, beyond the dissemination of the party’s ideology. I build upon the
interpretation that party schools are part of the party’s continual state-building project –
but identify more precisely the actors, motivations, and processes driving organizational
changes in the post-Mao period.
Chapter three will explore the theoretical and empirical relationship between
cadre training and elite selection. In the principal agent relationships which suffuse a
hierarchical political system such as China’s, the party’s selection problem comes prior to
the other problems, more commonly studied, in a principal-agent relationship (i.e., moral
hazard, which is solved by monitoring, rewards and sanctions). This chapter will test
whether selection for training at a party school constitutes a channel for promotion to
higher cadre office. By employing a matching method on large-N survey data, this
chapter will present findings from analysis of a national sample of individuals on an
administrative and/or political career track as well as results from an original dataset of
the career histories of Central Party School trainees. It will also consider mechanisms for
selection, including screening, signaling, and the role of networks.
Chapter four shifts the level of analysis to discuss the marketization of cadre
training, unraveling the processes by which competition was introduced to the party
school system. Beginning in the mid-1980s, different sets of preexisting and new
organizations were allowed to enter a cadre training market. At the same time, party
schools were also allowed to engage in market activity that extended beyond their core
35
training work. These two sets of market opportunities emerged via top-down, center-led
processes which local actors then seized for local gain and to effect system wide change.
Some intentionality can be deduced from central policy documents, while field interviews
reveal a more mixed terrain: a combination of collaboration and rivalry characterizes the
relationship between the organizations that now compete for cadre training contracts.40
This chapter will also discuss an important precondition to this marketization strategy,
limited fiscal and administrative decentralization.
Chapter five peers inside party schools to unpack the various school responses to
competition and the development of party entrepreneurialism. Party school leaders have
pursued a variety of income-raising schemes, some of which exist purely for pecuniary
gain, while others attract income as well as improve the quality of the school’s core
training outputs. This chapter will also consider change in the content of party school
training through content analysis of Central Party School training syllabi from 1985 to
2007. Together, these varieties of party school activity demonstrate the range of
organizational responses to competition. Site visits to training organizations at the
central, provincial, city, and county levels form the basis for case studies of party school
adaptation across regions with varying levels of economic development (Appendix B).
In all locales, party school adaptation is a function of organizational responses to two
markets: the market opportunities created by Deng’s liberalizing economic reforms and
the pressures presented by a second market in which a variety of party-approved
organizations compete for trainees. Schools must thus adapt to two imperatives:
40 Field work focused on localities located within a coastal Province A and inland Province B. Appendix B provides an overview of my field research strategy, a summary of interviewee data, and comparative analysis of field sites.
36
maximizing income streams in a new market economy and updating the content of cadre
training.
The concluding chapter will consider the implications of these changes. One
result has been greater search by party schools for new income-generating projects,
which in turn embeds party schools more deeply in local economies. Political
embeddedness results because training signals to ambitious cadres the skills that they
must demonstrate in order to move upward within the party ranks. Economic
embeddedness occurs through the pursuit of revenue streams from local partnerships.
This, in turn, speaks to larger questions of the tension between party efforts to remain
relevant and at the forefront of China’s economic development while avoiding the danger
of granting too much autonomy to local actors.
Looking beyond the China case, the theory and findings presented here offer an
explanation for how a hierarchical ruling party may develop the capacity to adapt to
systemic shocks and uncertainty. In the case of China, change initiated in one realm has
created pressures for adaptation in others: the decision to introduce market reforms to
China’s state-managed economic sector has motivated shifts in the organizational
geography and survival strategies of political institutions. This dynamism challenges
accounts of the brittleness and inertia of communist-party led systems.41 The particular
approach chosen by the CCP, introducing market incentives to organizations of political
control, suggests the diffusion of market principles beyond the economic realm to the
political. In creating a training market to introduce competition to the party school
system, the party leadership has sought to put in place incentives for continual adaptation
41 A critique of the rigidities of the socialist system can be found in Kornai 1992.
37
by party institutions, at the same time retaining the party’s hold on the loyalties and
careers of ambitious cadres.
38
Chapter 2 The organizational landscape
As discussed in the previous chapter, institutions of elite management located
within the ruling party are crucial for maintaining bureaucratic order in authoritarian
regimes. Controlling selection to party and government office remains a crucial part of
the foundation upon which party rule is built. As stated in CCP study materials, “the
party manages cadres” (dang guan ganbu), and from this principle flows the party’s
monopoly over the pathways to political authority (CPS 2004). Among the CCP’s cadre
management strategies, cadre training – its organization, the content of training programs,
and how it has changed over time – presents untapped insights regarding how the party
has exercised authority over the career paths and political knowledge of its leading
managers and administrators.
Cadre training has been a topic of study by scholars of contemporary Chinese
politics, and this chapter will first situate the present study in existing scholarship on
party schools. The second part of this chapter will “set the stage” and place party schools
in their political and institutional context through a discussion of the history, organization,
and management of party schools. An historical examination of the development of the
party school system demonstrates that while party schools are grounded in the early
ideological ideals of the party, they now reflect and embody the pragmatic objectives of
more recent organizational reforms. In this chapter and the next, I will argue that cadre
training is viewed by party authorities as a vehicle for updating the knowledge of party
and government bureaucrats and, furthermore, that party schools are training academies
with a measurable impact on the career development of bureaucrats throughout the party-
39
state. Framing the function of party schools’ in this way differs from existing scholarship,
which has focused instead on how party schools, as incubators of party ideology,
reinforce CCP authority.
2 Contribution to existing scholarship
Early scholarly analyses of party schools centered on the Beijing-based Central
Party School’s role in developing and conveying the party line.42 As such, there was less
emphasis on its flagship role within a broad-based web of cadre training institutions.
Unquestionably, the CPS possesses considerable ideological authority and has been the
site of breakthroughs in party theory. Under Hu Yaobang’s vice-presidency, the CPS
became a locus for formulation of “practice as a sole criterion of truth” and debate of the
post-Mao “two whatevers” ideological stances (Schoenhals 1991). During the reforms
and intellectual debates of the 1980s, the CPS was the institutional base for independent-
minded intellectuals within the party ranks (Ding 1994). The CPS also carries symbolic
weight as the site of major declarations of party doctrine: Jiang Zemin elaborated on his
“Three Represents” there in 2001, followed by Hu Jintao’s call for a “harmonious
society” in 2005 (Zheng 2010). Prominent scholars in both the CPS and the National
School of Administration, also located in Beijing, have presented new ideas for political
reform within the party and ways to move from “harmonious society” to “harmonious
socialism.”43 In June 2007, at the CPS, Hu Jintao expounded on the major themes of the
42 A review of the literature on party schools, both pre-1949 and after, can be found in Pieke (2009a), p. 35 fn. 19. 43 “Beijing Brain Trusters Propose a New Path for the Political Reform in China,” Yazhou Zhoukan, 27 May 2007. See also, “CPC National Congress to Launch New Resolution on Intra-Party Democracy,” Hong Kong Hsin Pao, 23 March 2007.
40
17th Party congress, several months before the October convening of the congressional
delegates (Fewsmith 2007: 7; Shambaugh 2008a: 111-5).
Other studies have considered the development of the CPS’s responsibility to
train central-level leaders on core party policies, but these works have remained Beijing-
centric (Fewsmith 2003; Wibowo and Fook 2006). Scholars have also noted the
importance of the CPS as a think tank and the positional advantage that party school
scholars enjoy in promoting their research and policy recommendations on political
reform (Fewsmith 2008). In limiting their scope to the Central Party School, however,
these studies do not consider when, why, and how cadre training more broadly may be a
useful instrument of political control and change for successive generations of central
party leaders.
Another body of research has placed party schools and cadre training in the
context of general processes of party adaptation. Party schools thus reflect the party’s
efforts to study and learn from cases of failed communist party-led reforms elsewhere
(Shambaugh 2008a: Chapter 7). With these historical lessons in mind, recent passage of
training-related legislation reasserts organizational discipline over cadre ranks.44 These
studies have also moved beyond Beijing and considered diversity in the organizational
actors that contribute to cadre training. Greater access to field sites has enabled more
detailed observations of how immersive training experiences contribute to the building of
a distinct cadre identity (Tran 2003). Through a vertical case study of the Yunnan party
school system, Pieke (2009a) has probed how party schools reflect the party’s efforts to
44 Examples of these official declarations include the “Resolution on the CCP’s strengthening and improving party school work in the twenty-first century”, available at http://wlzx.hdpu.edu.cn/zzb/Article_Show.asp?ArticleID=125, accessed 6 June 2006 and the “2004 CCP Party School Work Trial Regulations”, available at http://wlzx.hdpu.edu.cn/zzb/Article_Show.asp?ArticleID=126, accessed 6 June 2006.
41
redefine socialism in a changed context and contribute to the party’s larger “neo-
socialist” state-building project.45
This dissertation makes a separate but related argument to these more recent, local
and system wide studies. First, this study supports and expands upon existing work by
demonstrating with new data the mechanisms by which party schools contribute to the
party’s management of human resources, in particular the exercise of party control over
cadre careers. This study draws out assumptions that are implicit in existing studies to
test whether party schools play a significant role in the construction of China’s political
elite. Beyond the ideological import of these schools, this project maps out their function
within the organizational machinery of a rational Leninist party-state. By measuring the
effect of party school training on cadre promotion, I show how these schools generate
positive incentives for cadres to adhere to the dictates of the ruling party, beyond any
individual-level commitments to ideology.
Second, I depart from extant literature by investigating how party authorities have
induced organizational change within the party school system itself and how this provides
traction on the larger, multi-faceted story of CCP survival in the reform period. This
study builds upon Shambaugh and Pieke’s observation that the party has embarked on a
broad-based institution-building project. Arguably, this project has been in progress
since the founding of the republic, but in its current form it is characterized by
“interlocking patterns of neo-socialist marketization, bureaucratization and party
building,” (Pieke 2009a: 18). My contribution is one of specificity: I unravel why party
schools, and not other institutions of party management such as political campaigns, were
45 By neo-socialism, Pieke is referring to “the creation of a strong, effective and modern bureaucracy … while at the same time maintaining the party’s Leninist control,” (Pieke 2009a: 18).
42
developed in the reform period. In this narrative, I consider the logic underlying the
decision to introduce market-based competition rather than apply bureaucratic reforms of
a less radical nature. All of this is to lend a meso-level, or institution-driven view of how
the party acts on its fundamental desire not only to survive through present reforms but
remain their central architect.
My third contribution to the literature is empirical. Through field visits and
interviews conducted in the local party schools of provinces in the coastal and central
regions of China, I present a more ground-level understanding of organizational change
within the party. This allows for both a vertical and horizontal examination of how these
key party organs work, across regions and administrative levels. Such an empirical
approach provides a more complete picture of the incentives, responses, and risks
underlying political change in China.
3 An historical overview
A tour through the history of cadre training in the PRC, where the party school
system has remained a central and enduring feature, demonstrates that party-led cadre
education has experienced an uneven trajectory over time: from early, limited efforts to
institutionalize cadre training, followed by the decade-long disorganization of the
Cultural Revolution, to more recent, top-down efforts to unify the party school system.
Cadre training has long been a component of CCP policy, beginning with revolutionary
education during the Republican period (1912-1949). For parties initially guided by a
Marxist-Leninist ideology, the dissemination of theory to party members and leaders was
a prerequisite for organizing revolution. Training in the Leninist historical narrative and
43
accompanying weapons of the party – such as how to formulate class-based rhetoric,
disseminate propaganda, establish base camps – took place in party schools. While the
concept of training schools for cadres and other revolutionary actors (e.g., workers,
peasants) has existed since the early years of the CCP, these organizations have
undergone expansion and contraction from the first half of the twentieth century to the
present.
The first party schools were established in Hunan province by party and
Communist Youth League members in 1923 (Wang 1992: 33-4).46 What was to become
the Central Party School was established in 1933 in Ruidian, Jiangxi province, and
persisted through the Long March to become, by 1955, the Advanced Central Party
School (Zheng 2010). In the 1950s, during the party’s laying of organizational
foundations in the new republic, most provinces saw the construction of a provincial-
level party school. During this first decade of Communist party state-building, leaders
attempted to create a system for training, recruiting, and educating cadres. “The period
of the early to mid 1950s saw the major development of systematic cadre management
practices. The party specified more clearly the criteria for cadre appointments and
promotions, there was an increased effort to expand training to help cadres acquire
necessary technical and political credentials, and a formal network of parallel Party and
state institutions for cadre management was set up,” (Manion 1985: 206, citing Harding’s
Organizing China). Drawing from émigré interviews, Whyte details cadre education
during the 1960s as characterized by a “strict political atmosphere” in certain work units,
46 During this period, training consisted of small groups studying subjects such as ‘capitalism and China’, ‘workers’ movement’, ‘rural movement’, ‘social revolution and people’s revolution’, and ‘world revolutionary history’. They also engaged in propaganda writing. Detailed histories of the party school system are available in English ((Pieke 2009a: Chapter 2) and Chinese (Chen 2007) scholarly work.
44
emphasis on doctrinal study, personal political evaluations, and peer criticism (Whyte
1974: Chapter 5). During these early decades of the republic, then, there existed the
notion to codify cadre education via a network of party schools, but training plans and
general operating principles were as yet unsystematic.
By the late 1970s, after reopening from the closures of the Cultural Revolution
(1966-76), cadre training was still relatively ad hoc: cadres attended occasional field
investigation reports by elder cadres and were assigned mentors by their direct
superiors.47 This system depended largely on the initiative of individual cadres with
management authority, and no formal rules governed, for example, the number of
lectures to organize or mentors to assign. Local schools focused on staging short-term
classes for the study of party documents, major party meetings, and party congresses.48
Before Deng’s reforms, training consisted of two parts: the “five old topics”, all based on
fundamental Marxist theory, and the study of party documents.49 Then, during the anti-
bureaucratic campaign of 1977-1980, party leaders shifted from political means for
rectifying the bureaucracy (e.g., struggle sessions, media-intensive propaganda
campaigns) to emphasis on rational managerial means such as retiring old cadres and
enhancing technical training (Manion 1993; Morgan 1981).
With the triumph of Deng’s reformist camp and the onset of economic
transformation in the 1980s, central leaders emphasized the significance of cadre training:
“The needs of modernization require large-scale, well-planned training of cadres to
improve the quality of the cadre ranks. Cadre training … guarantees the continuity of the
47 Interview with a Ministry of Foreign Affairs cadre, Interview 26, December 2006. 48 Interviews with provincial party school professors, Interviews 93 and 94, December 2007. 49 Interview with a CPS professor, party history department, Interview 114, February 2008. The “five old topics” (wu ge lao men) were Marxist philosophy, Marxist economics, scientific socialism, party history, and party building.
45
party line and thus is of major strategic importance,” (Central Organization Department
1983: Appendix 3, 67). In these declarations, there remains a clear ideological purpose to
training, though subsequent reforms have placed cadre training within the larger
administrative modernization project of the party. Hu Jintao, the leader of China’s fourth
generation of party officials and also the president of the Central Party School from 1993
to 1998, declared, “As required by the Three Representatives theory, establishing a team
of high-quality party cadres is the key to the development of our party and our country in
the new century. … To train a new generation of young party cadres, the party school is
shouldered with great responsibility.”50 Compounding this training burden was the need
to rotate bureaucrats through remedial programs and prepare them for the changes
brought on by the economic liberalization and social reforms of the post-Mao period. As
detailed in the previous chapter, low turnover in China’s party and government
bureaucracies lent additional urgency to retraining efforts.
During this period, there was also an effort to take the loosely organized party
school system and create a more coherent approach to cadre training. Officially, May
1982 saw the birth of the national party school system, beginning with the movement by
Central Party School President Wang Zhen (1982-1987) to standardize (zhengguihua)
cadre training throughout the country (Wibowo and Fook 2006).51 This standardization
has met with limited success. Local party schools all organize young cadre training
classes and short training classes following major events such as party congresses, but
there remains variation in the final lineup of training classes mandated by local party
50 “Promote the party school education to a new level,” People’s Daily, 7 June 2000. 51 Standardization included attempts to unify party school training content and the types of classes organized by local schools (Interviews with CPS party history professors, Interviews 112 and 120, February 2008).
46
committees and conceived by the schools themselves. Before exploring each of these
types of training classes, it is necessary to set the stage with a brief foray into party
schools’ larger organizational context in the contemporary period.
4 Organization and oversight of the party school system
Local party schools are enmeshed in a variety of formal and informal
relationships with party and government bodies located at the same administrative level.
At the same time, oversight of party schools cuts horizontally through locales as well as
vertically through various party bureaucracies. To situate party schools in their
organizational contexts, Figures 2.1 and 2.2 below offer a description of the local party
and government organizations involved in cadre training.
On official organization charts, a local party school is under the managing authority
(zhuguan bumen) of its local party committee. This within-locale authority relationship
contributes to the organizational autonomy enjoyed by party schools. That a local party
school’s affairs are managed by a party organ located at the same administrative level
highlights the local nature of party school supervision. If a party school wishes to
experiment with new programs, for example, the local party committee stands to benefit
from any positive outcomes generated by these ventures, even if they are not necessarily
consistent with central directives. Less successful ventures, furthermore, can be
contained within the local party bureaucracy. This arrangement has the effect of aligning
the interests of local party committees with their respective party schools and generating
strong incentives for local, rather than system-wide, development.
47
Figure 2.1: Local party organs involved in cadre education
District government区政府
District personnel
department区人事局
District government ministry training
centers部门培训中心
Civil servant section公务员科
• Collaborates with administration institute on civil servant training
Ex: District Tourism Training Center旅游培训中心
District finance区财政局
• Allocates funds to party and government organs for cadre/civil servant training
District administration
institute区行政学院
• Organizes training of non-leading cadres and civil servants•Generally located on the party school campus and shares staff
Figure 2.2: Government organs involved in civil servant & cadre education
District party committee区委
District organization department区委组织部
District propaganda department区委宣传部
Cadre bureau干部局
• Coordinates with the party school on leading cadre training
E-education office电化教育办公室
• Responsible for distance learning programs
Cadre online study credit office
网上干部学分考核办公室
• Responsible for online training system
Theoretical education section
理论教育科
• Maintains delegation of lecturers on party theory (讲师团)•Responsible for leading cadre study group content within work units (领导干部中心组的学习)
Party member education department党员教育处
• Responsible for CCP member education at the district and lower levels
District discipline and inspection
department区纪委
Party’s clean government education
office党风廉政教育室
• Responsible for anti-corruption education
District party school区委党校
Often combined with:District administration
institute区行政学院
District socialism school区社会主义学院
48
At the same time, local party schools are subject to less formal but nonetheless
powerful monitoring by higher level party schools, party personnel (organization)
departments, and finance bureaus. First, party schools are embedded in a web of
relationships with other party schools. These relationships are characterized as advisory
in nature (zhidao guanxi).52 Such consultative relationships stand in contrast to the more
coercive, binding control that exists in other aspects of the bureaucracy. Customs
(haiguan) bureaus, for example, must follow the dictates that flow down through the
system from the center to the localities. Instead, the local party committee, in
consultation with the local party school, organization department, and other relevant
bodies such as the education and personnel departments, drafts annual plans and training
targets. Actual implementation of training plans resides with party school administrators.
Integration of personnel within a given locale also colors the relationship between
local party committees and party schools. As in many other realms of Chinese political
organization, party school principals hold concurrent posts. Table 2.1 below lists the
names and concurrent offices held by provincial party school leaders.
One of the vice-party secretaries for a locale is often, but not always, the titular
head of the local party school, while day-to-day administration is carried out by one or
more vice-principals. In Anhui, for example, the principal of the provincial party school,
Wang Mingfang, is also a provincial vice-party secretary; party school management is
carried out by five vice-principals.53 In other provinces, such as Guangdong, the current
52 This is stated in Article 11 of the most recent CCP Party School Work Regulations, available at http://www.sina.com.cn, accessed 13 November 2008. The Chinese equivalent for top-down, compulsory control that interviewees repeated again and again was lingdao guanxi. 53 See http://www.ahdx.gov.cn/sm2111111124.asp for a listing of Anhui Provincial Party school leaders, accessed 18 Jun 2009. Wang Mingfang’s concurrent status as vice-party secretary within the provincial party committee can be found in the Central Party School’s Study Times (xuexi shibao), http://www.china.com.cn/xxsb/txt/2007-04/09/content_8089276.htm, accessed 18 June 2009.
49
Table 2.1: Provincial-level party school leadership
Provincial-level Unit Party School Principal Concurrent Post(s) Anhui Wang Mingfang Provincial vice-party secretary Beijing Wang Anshun City vice-party secretary
Political-legal committee party secretary Chongqing Zhang Xuan City vice-party secretary Gansu Liu Weiping Provincial vice-party secretary Guangdong Hu Zejun Provincial organization department head
Provincial administration institute head Hainan Liu Qi Provincial organization department head Heilongjiang Zhou Tongzhan Provincial vice-party secretary
Provincial CPPCC party committee vice-secretary Hubei Yang Song Provincial vice-party secretary
Wuhan city party secretary Hunan Zheng Peimin Provincial vice-party secretary Jiangsu Zhang Lianzhen Provincial vice-party secretary
Provincial CPPCC president Jilin Wang Rulin Provincial vice-party secretary Liaoning Zhang Chengyin Provincial vice-party secretary Qinghai Luo Huining Provincial vice-party secretary Shandong Jiang Yikang Provincial party secretary Shanghai Shen Hongguang City organization department head Shanxi Xue Yanzhong Provincial vice-party secretary Sichuan Ke Zunping Provincial organization department head Tianjin Zhang Gaoli City party secretary
City administration institute head Yunnan Yang Chonghui Provincial vice-party secretary Si Xinliang Provincial organization department head
Sources: Author’s dataset. Notes: Leadership information not available for Fujian, Guangxi, Guizhou, Hebei, Henan, Jiangxi, Inner Mongolia, Ningxia, Shaanxi, Xinjiang, and Tibet. See (Guo and Shan 2009) for a list of provincial-level party school presidents and their positions on party committees, the Central Committee, and other party organs. principal of the party school and administration institute, Hu Zejun, is the head of the
provincial organization department, an office that is typically a vice-party secretary
position on the provincial party committee. Below Hu, the Guangdong party school is
managed by a team of seven: one standing vice-principal and six vice-principals.54 Again,
this integration between party committees and party school leadership is within-locale
and underlines local party schools’ independence from higher level party bureaus.
54 The Guangdong Provincial Party School leaders and party committee members are listed at http://www.gddx.gov.cn/xyxk/xrld.htm, accessed 18 June 2009.
50
These local lines of authority are balanced, however, by structural checks and
monitoring mechanisms between party schools and various party authorities. Such intra-
party monitoring ensures greater consistency throughout the party school system than
might be assumed when looking only at the formal lines of authority that exist within
locales. Monitoring between party authorities and party schools ensures some degree of
quality control. There are three major types of supervision in place: sending down work
teams and observers, bringing party school heads to training classes and conferences at
higher levels, and imposing reporting requirements.
Work groups, as the most obvious form of top-down inspection, are sent down
from organization departments and higher level party schools, generally one
administrative level up, to observe affairs in junior counterparts.55 The Central Party
School has been known to increase the number of work teams it sends out after important
events, such as the most recent Seventeenth Party Congress, to ensure that local schools
are aware of the special congress-related training sessions that they must organize and
carry out.56 This “police patrol” monitoring occurs frequently in the system, though the
regularity of such inspections is unclear.
For important training classes, party school planners can also expect local
organization departments to designate a minder for each small group formed within a
training class. For some classes, small groups will have minders from both the
organization department and party school.57 When higher-level party committee members
55 Pieke 2009 (Chapter 3) offers a detailed overview of the organizational measures in place to organize and monitor cadre training in a locale. During one field visit to a county-level party school in November 2006, I observed part of a work team visit from the nearby city-level party school. The visit lasted half a day, with the team meeting throughout the morning with school officials, then culminated in a banquet lunch and afternoon departure. 56 Interview 108, February 2008, retired CPS professor. 57 Interview 208, May 2008, provincial party school professor.
51
are invited to schools to give speeches at the beginning or conclusion of training classes,
this is another form of examination (kaohe) and allows both parties to share information
about what is happening at each level.58
Second, party school heads will be invited to partake in “research and discussion”
classes at higher level party schools. According to CPS yearbooks, such classes were
held in 1991 and 1995 for the principals of city (prefecture, diting)-level party schools.
These small classes, for around 50 school principals and ranging from one to three weeks
in duration, were organized not by the CPS’s core training department (peixun bu) but
rather by the theory (lilun bu) and advanced training (jinxiu bu) departments . The
Central Party School also hosts nationwide meetings in which provincial and city party
school heads submit reports to peers and central leaders.59 These meetings comprise
celebratory elements as well as the substantive exchange and filing of official reports.60
Third, party school heads also send reports of school affairs up the system, to the
local finance bureau and the local organization department.61 Party school vice-
principals (chang wu fu xiaozhang, i.e., the vice-principal in charge of day-to-day affairs)
also file reports with the school principals who are generally vice-party secretaries on
local party committees.62 Taken together, these monitoring strategies illustrate how party
schools are connected to the larger party school system and key bureaus at the local level,
58 Ibid. 59 Records exist for the first “National Party School Work Meeting” convened in 1979, followed by a second in 1983 (Jiang, Zhang, and Sheng 1983). Other meetings were reported in each of their their respective CPS yearbooks (1985, 1990, 1992, 1995, 1996, and 2001). 60 I was invited to observe the closing banquet at one such reporting conference. This meeting was convened by a provincial-level party school and hosted by a county-level party school in Province A. In addition to a lunch banquet, the meeting allowed provincial party school leaders to collect reports from city, district, and county schools that were participants in the provincial school’s distance degree programs. See field notes May 2008. 61 Interview 176, April 2008, city party school teaching department head. 62 At the CPS, the current head vice-principal, Li Jingtian, reports to the principal, Xi Jinping, who sits on the Politburo Standing Committee and CCP Secretariat.
52
despite the advisory nature of the relationships. Party schools thus “operate more as a
cluster of hierarchically embedded networks rather than an impersonal bureaucratic
hierarchy,” (Pieke 2009a: 122). Throughout this hierarchical system, party schools all
organize certain general categories of training classes.
5 Types of training
Despite attempts in the early reform period to standardize party school training
outputs, party schools continue to offer a broad range of educational programs and
training sessions. One basic breakdown is to distinguish between degree and non-degree
programs. Degree programs are a more recent offering, beginning in 1983 with two-year
undergraduate (benke) training classes at the CPS and the creation in 1985 of distance
education programs (hanshou jiaoyu) that, under CPS leadership, radiated downward
through the system.63 Officially, these residential and non-residential programs were
intended to raise the educational level of cadres in the post-Mao period and answer
Deng’s call for a more educated cadre corps. With time, distance education degree
programs have involved party schools from the county to the central levels in profit-
sharing networks.64 As such, they have become an additional revenue stream for schools
and supplement transfers from local government finance departments. Generally, higher
level schools issue diplomas and set curricula and testing standards, while lower level
63 The syllabi from these early CPS undergraduate training classes are available in the 1985 CPS yearbook, pp. 224-232. 64 Pieke 2006 offers a detailed description of the profit-sharing arrangement between the CPS and Yunnan party schools at the provincial, city, and county levels. A county party school principal in Shandong informed me that 60 percent of tuition from distance coursework programs remains at the county school, while 40 percent is sent up to provincial and central levels. Interview 232, July 2008. This is somewhat consistent with a newspaper article report that the CPS correspondence program allocated tuition accordingly: 10 percent to the CPS, 15 percent to branch campuses, 20-25 percent to schools below branch campuses, and the remainder to the “tutorial stations” (fudao zhan). Zhu Hongjun, “Central Party School calls for the end of correspondence degrees,” Southern Weekend, 29 November 2007.
53
schools provide course support and serve as marketing and distribution points.65 In 2007,
distance degree programs ceased to admit new students, but interviews yielded vague
responses regarding the enforcement of this decree. Graduate degree programs,
furthermore, continue to exist at the central and provincial levels.
The second category, comprising non-degree training programs, is enormously
diverse. Training defined by Chinese government documents includes four types of non-
degree programs, ranging from new hire training classes to less defined “special topic”
training programs (Zhang 2005: 233-235). Local party schools tend to use a different
categorization system. Core (zhuti) training courses are mandated by the local party
committee and supported by funding from government coffers.66 Auxiliary (fei zhuti)
courses vary more across locales and depend on what the party school leadership wishes
to offer, in consultation with the local party committee.67 Core training courses include
those which are for orientation or on-the-job training purposes, but they can also include
more select classes for young cadres with promotion potential. Auxiliary classes often
address special topics or are geared for specific party bureaucracies. All of these non-
degree party school programs vary in duration, level of study, and content. Table 2.2
below offers an example of the allocation of training courses, by type, scheduled by a
city-level party committee for 2008.68
65 One consequence of this is some interesting overlap in the educational experiences of local officials. In one inland county that I visited, a township party secretary was enrolled in a provincial party school degree program while a party discipline committee cadre in a different township had opted for the CPS distance degree program, though it was more expensive. While most of the coursework consisted of independent study, they went to their local county party school for periodic lectures and program support. The county party school was the local study center for these provincial and central party school distance degree programs. Interviews 10 and 11, November 2006. 66 These are also referred to as “inside-the-plan” (jihua nei) training classes, in accordance with annual training plans approved by each local party committee. 67 These are also referred to as “outside-the-plan” (jihua wai) training classes. 68 See Appendix C for a full listing of this city’s annual training plan.
54
Table 2.2: City Z planned training classes, 2008, by type and enrollment levels Core training
classes Auxiliary training
classes Total
Number of training classes 20 20 40 Enrollment, in number of students
1,490 (39%)
2,295 (61%) 3,785
Enrollment, in student-months 518 (55%)
417 (45%) 935
Note: Row percentages in parentheses. Source: Internal school document.
Out of 40 scheduled classes, an equal share of courses was dedicated to core versus
auxiliary programs. While the total number of cadres enrolling in special topic training
classes is greater than for core classes, the total student-months dedicated to core training
is higher, implying longer core classes.
In terms of which type of training class a cadre might attend, a basic division of
students at party schools are those 1) who are paying for degrees, 2a) sent to core training
classes and 2b) sent to auxiliary training classes.69 The first population includes graduate
students and cadres taking degree courses on weekends or through correspondence
courses, mostly to pass a standardized exam and/or fulfill an employment requirement.
Some are allowed to take an extended leave of absence to complete a degree, as is the
case for party school teachers interested in earning a graduate degree. This has bearing
on promotion prospects insofar as promotion to certain ranks requires particular levels of
education.70 Cadres enroll in party school degree programs with these job requirements
in mind, and though they are expected to pay from personal funds, in reality their
69 Interview, Central Party School teacher, Interview 17, November 2006. 70 By 1983, the Central Organization Department had called for all leading cadres under 40 years of age to improve their educational credentials by attaining at least a middle school education, technical high school degree, or technical college degree within three years. All future leading cadres were to have at least a high school degree (Manion 1985, p. 223, fn. 66). By 2006, cadres promoted to the leading ranks (county- or department-level and above) required at least a tertiary-level education (daxue zhuanke yi shang wenhua chengdu). See “Questions and answers on cadre promotion policies and regulations,” Yimen Organization Department webpage, available at http://www.yimendj.gov.cn/Article/ShowArticle.asp?ArticleID=216, accessed 30 June 2009.
55
workplace often produces a scholarship or other financial aid. Graduate students,
concentrated at the central and provincial party schools, focus on predictable disciplines
such as party history and socialist theory, but there are also more contemporary
specializations such as law and finance.
Core training programs include routine, periodic (lunxun) training classes for a
variety of cadre populations such as those working in mass organizations, minority cadres,
and reserve cadres. Promising cadres with the potential to rise through the ranks are also
invited to a key point training for “young cadres” that is common across all party schools.
Interviewees have indicated that invitation to training programs may be a matter of
negotiation with direct superiors or at the behest of supervising organization
departments.71 Advanced (jinxiu) trainings exist for leading cadres, such as the CPS’
three-month training for cadres at the provincial (ministry) level and five-month training
program for bureau-level cadres. The CPS has an advanced training department
dedicated to crafting these classes at the central level, while local party schools often
lump together advanced and other core training classes as part of the work of their
education departments.
Auxiliary training programs have no generalizable target populations of cadres
and can include cadres of all ranks and functional responsibilities. This increased variety
and specialization of party school training classes reflects a broader, more general
emphasis on cadre training in the reform period. The next section will discuss the various
ways party authorities have supported cadre training and turned to party schools as an
instrument for supporting China’s modernization and economic reform.
71 Interviews, CPS department chief training class (chu ji pei xun ban) student, Interview 65, November 2007; provincial party school vice-principal, Interview 66, November 2007.
56
6 Reform-era focus on cadre training
Developments in three areas attest to the growing importance of cadre training in
contemporary China: declarations in speeches and policy documents, investment in
training infrastructure, and growth in training programs for the most promising “young
cadres” (zhongqing nian ganbu). Notably, institutions of cadre training such as party
schools, cadre schools, administrative institutes, and cadre executive academies fall
squarely under the leadership of the Central Committee, the State Council, the Central
Organization Department, and the Ministry of Personnel, not the Ministry of Education.
This implies, in turn, the positioning of cadre training as a mechanism for party control
over cadres and bureaucrats, not simply a means to increase professional competence.
Official declarations of support
Interwoven throughout efforts to systematize and codify cadre training over the
past 20 years, there has been official acknowledgement that the ability of the Chinese
government to prepare its agents for implementation of reforms hinges on the quality of
training programs. Deng Xiaoping declared in 1980,
The current problem, in a nutshell, is not that we have too many cadres but that their training does not match their work, and that too few of them have specialized training in their particular field of endeavor. The solution lies in education. One way is to open schools and training courses for cadres, another is self-education. It is essential for everyone to devote serious effort to study.72
Turning to the most recent generation of CCP leadership, Party Secretary Hu
Jintao declared in his speech before the 17th Party Congress in 2007,
72 Selected Works of Deng Xiaoping, “The Present Situation and the Tasks Before Us,” speech given in Beijing on January 16, 1980, available online at http://english.peopledaily.com.cn/dengxp/vol2/text/b1390.html, accessed 21 March 2007.
57
We must strengthen Party building in all respects. Ideologically, we will focus on fortifying the convictions of Party members. Organizationally, we will put emphasis on bringing up Party members and cadres of quality … We will continue to train cadres on a large scale, making full use of Party schools, schools of administration and cadre academies to substantially improve the quality of cadres … We will develop modern distance learning programs for Party members and cadres in rural areas throughout the country.73
Such pronouncements are part of a larger CCP initiative to transform itself from a
revolutionary party to a ruling party with an emphasis on its ability to “govern” (zhizheng)
and become a “learning party” (xuexi xing zhengdang).74 Zeng Qinghong, president of
the CPS from 2002 to 2007, declared that “Party schools at all levels [are] the main
channel for training the core of high-standard governing personnel, the main front for
promoting the building of a learning-type political party.”75 Within this new tact,
“advancing party members’ training on the advanced nature of the party is the most
important measure in strengthening the party’s governance capacity,” (Dai 2006: 317).
Training has been singled out by the party leadership as a vehicle for party reinvigoration
and to raise the oft-invoked but vaguely defined notions of cadre “quality” (suzhi) and
“ability” (nengli).
A slew of policy documents have sought to give administrative flesh to such
ambitious official pronouncements. First, the 2003 Regulations on the Selection of Party
73 Hu Jintao, Report to the 17th Party Congress, Section XII, “Comprehensively carrying forward the great new undertaking to build the party in a spirit of reform and innovation,” 15 October 2007. This section is also where Hu uses the term, “learning party” (xuexi xing zheng dang) to refer to the CCP’s new direction. As early as 2003, he called for raising the quality of party and government cadres in a speech entitled, “Implement the Strategy of Human Resources and a Strong Country, Firmly Uphold the Principle of the Party Managing Human Resources.” 74 The key document is the CCP Central Committee’s “Decision on the Enhancement of the Party’s Governing Capacity”, adopted at the 4th plenum of the 16th Party Congress in 2004. See “CPC issues document on ruling capacity,” Xinhua, 27 September 2004, http://www.chinadaily.com.cn/english/doc/2004-09/27/content_378161.htm, accessed 6 March 2009. Hu Jintao also uses this term in his 2007 Party Congress report. 75 “Zeng Qinghong speaks at Central Party School on Building up Governing Capability,” Xinhua, 24 September 2004.
58
and Government Officials specified that candidates for a leadership post at the county
level or above must have 1) a bachelor’s degree or higher, 2) at least three months of
training in a party school or other executive training program, and 3) at least five years of
work experience (Li 2007a: 24).76 Second, for the period 2003-08, the Central
Organization Department, the powerful bureau charged with party personnel matters,
declared that almost 110,000 government and party leaders from the county level up must
undergo three months of training over the five-year period (Li 2006).77 Third, in 2004
the COD issued a document to renew focus on cadre training (Central Organization
Department 2004). This was because “strengthening and improving party school work
is … an urgent need of the party” as it faces “global change on many fronts.”78
Performance during training classes was also explicitly linked to cadre evaluation.
A 2006 set of trial regulations for cadre education and training called for a given
student’s “attitude and appearance (taidu he biaoxian), grasp of political theory and
policy, job-related knowledge, and cultural knowledge” to be recorded and referenced
during annual performance evaluations (Central Organization Department 2006: 122-135;
Ministry of Personnel 2006: Section 7). These regulations went so far as to assert that “in
the future, a cadre’s education and training will become one of the key means for
76 The three month training requirement was stipulated as early as 1995, in the Central Committee’s Interim Regulations on Selection and Appointment of Party and Government Leading Cadres. Training could take place at “party schools, colleges of administration, or other training institutes,” (Bo 2004: 82). Interviewees indicated that some officials receive their pre-promotion training after assignment to a new post, but this “make-up training” (bu xun) does not appear to be a common practice. Interview 65 with a former CPS trainee, November 2007 and Interview 191 with a county-level party school vice-principal, May 2008. 77 This figure includes “about 500 provincial-ministerial-level leaders, 8,800 prefecture-division-level leaders, and 100,000 county- and bureau-level leaders [who] must participate in these training programs,” (2006-10 Nationwide Cadre Education and Training Plan, Section 3). 78 For a text of this document, see the Chinese Yearbook of Ideological and Political Work (2001), pp. 50-54.
59
assessing the cadre’s work and promotion prospects,” (Ministry of Personnel 2006:
Section 7, Article 41).
New investment
Speeches and policy documents may offer a limited sense of trends and intentions;
stronger evidence lays in the dedication of real resources to cadre training. Central
finance has earmarked more funds to the general category of cadre training: “In 2005, for
instance, the Ministry of Finance raised its own (non-capital construction) spending for
cadre training at the central-level by 10.2 percent,” (Pieke 2008: 8).
It is unclear how evenly this funding is spread across training programs, but a
building boom is also taking place across local party schools. In site visits to county- and
district-level schools in the wealthy coastal Province A, all the schools were in the
process of being rebuilt from the ground up on both larger chunks of state land and with
expanded facilities. During a visit to one county-level school in this province, the vice-
principal boasted to me that the school’s new campus cost 85 million yuan, or
approximately 10.6 million dollars.79 While systematic financial data are difficult to
obtain, these figures give a sense of the infusion of funds to party school development.
Targeting young cadres
Training program enrollment levels are another indication as to whether cadre
training is an increasingly important strategy of cadre management. During the period
1997 to 2004, for example, a total of 20.2 million trainees passed through training
79 Half of the investment came from selling the old party school property and the other half was obtained from a direct transfer from the county finance department. Interview 195, May 2008.
60
programs at the provincial, city, and county administrative levels.80 Since the onset of
reforms in 1978, the CPS has trained in excess of 50,000 medium- and high-ranking
cadres.81 There is evidence that within these general statistics, more targeted investment
is being made in specific training classes. One CPS interviewee indicated that there has
been increasing emphasis on “young cadre training classes” and other selective, but
regularly scheduled, training programs.82
Young cadre training classes (zhongqing ban) were noted by interviewees as the
most significant training classes for rising cadre talent. These long-term training
programs, organized at party schools of all administrative levels are invitation-only
classes for those cadres, often in their thirties or forties, with the brightest futures in the
party. For the years where data is available, the pace at which young cadre training
classes grew, at the central level, generally outstripped the growth in the total number of
leading cadres in the country (Table 2.3 below). This is a somewhat noisy trend, though,
since there was a drop in growth of trainees and a spike in the growth of leading cadres in
the mid-1990s. It is not the case that these classes were expanding due to general
increases in the recruitment of new party members, as CCP membership grew at a
relatively slower pace than Young Cadre Training classes.
80 “Accomplishments in party organization work since the 15th Party Congress,” 12 February 2004, http://www.xfdjw.gov.cn/show.asp?id=79, accessed 3 November 2007. This figure is for total training-person sessions, not total individuals trained; some cadres may have attended more than one training session during this period. A total of 114,000 training sessions were organized nationwide during this period at the three levels, averaging a rather high 177 trainees per training session. 81 “China’s Central Party School trains 50,000 officials in 30 years,” Xinhua, 2 October 2007. 82 Interview 216 with a CPS teacher, May 2008. He noted that core, general training programs in the CPS’ Training Department (peixun bu), which organizes these “young cadre training classes”, were receiving more emphasis than the more targeted training classes organized by the Advanced Training Department (jinxiu bu)
61
Table 2.3: Percent increase, over previous year, in CPS Young Cadre trainees, total leading cadres, and total CCP members, select years
Sources: CPS Yearbooks, 1985-2001, Central Organization Department 1999.
The CPS’ young cadre training enrollment has increased from 1981 to 2000, and
the training class’ share of total CPS trainees has also increased (Table 2.4). This
suggests that party authorities have focused more narrowly on building the political
knowledge, networks, and career prospects of a select group within the administrative
class. Funding data are not available, but these are resource intensive classes since
trainees are housed on campus for duration of the training program (10 and a half months
or 23 months, depending on whether the training class is for one or two years) and sent
on multiple study trips throughout the country and abroad.
Table 2.4: CPS Young Cadre Training Class enrollment, 1980-2000
Year
Total Central Party School enrollment, in
trainee-months
Young Cadre Training Class, in trainee-
months
Young Cadre Training Class, as a percent of
the CPS total 1980 13,920 605 4.3 1981 17,055 1,573 9.2 1982 11,207 2,101 18.7 1990 9,309 1,595 17.1 1991 12,205 1,958 16.0 1993 8,150 1,892 23.2 1994 8,493 2,123 25.0 1995 7,300 2,189 30.0 2000 8,098 2,552 31.5
Source: CPS Yearbooks, 1985-2001
Year
Percent increase over previous year, CPS
Young Cadre trainees
Percent increase over previous year, total
number of leading cadres in the country
Percent increase over the previous year, total
CCP members 1981 160 9.5 2.4 1982 33.6 8 1.9 1991 22.8 5.4 2.2 1994 12.2 3.2 3.8 1995 3.1 10 1.9
62
These findings suggest that the CCP has placed increasing emphasis on training
those leading cadres with the potential to provide long tenures of service to the party.
Tables 2.5 and 2.6 below capture these changes since the founding of the PRC.
Table 2.5: Highest educational attainment of cadres, by historical period
Education Level
Early PRC, 1950-1965
Cultural Revolution,
1966-76 Reform era,
1977- No formal education 5.0 0 0 Elementary 20.0 34.0 3.0 Middle school-level vocational 20.0 17.0 12.5 Middle school 30.0 14.9 15.5 High school 20.0 14.9 15.2 Technical school 0 4.3 1.5 Three-year college 5.0 10.7 23.0 Undergraduate 0 4.3 11.4 Graduate and higher 0 0 0.6 N 20 47 279
Notes: Figures are in percentages. Individuals are sorted into periods based on the first year for which they reported a cadre occupation. Source: 2003 China General Social Survey Table 2.6: Educational attainment of leading cadres after 1982-84 reforms
Administrative level
Percent college
educated before 1982-84 reforms
Percent college
educated after 1982-84 reforms
Ministers and vice ministers 38 59 Directors and deputy directors of the State Council 35 52 Directors and deputy directors of central party organs 43 53 Heads and deputy heads of bureaus of central party organs 50 56 Provincial party secretaries, governors and vice governors 20 44 Directors and deputy directors of provincial bureaus 14 51 Municipal leaders 14 44 County leaders 14 47 Source: Lee 1991. Official declarations and documents, funding increases, and targeted investment in early-
career cadres lend support to this argument. The creation of entirely new systems of
cadre training organizations, discussed in the following chapters, also supports – and
problematizes – this argument. When considered alongside educational data that Chinese
political elites are now much more highly educated than their counterparts of three
63
decades ago, these changes suggest that reform-era emphasis on cadre training is bound
up in the elite transformation that has developed over recent decades (Li and White 1990).
As noted in the previous chapter, this transformation has also occurred with relatively
low bureaucratic exit to the private sector, which stresses the importance of retraining
efforts.83 Forming cohorts of public managers capable of coping with new market-based
realities from the planned economy generation also sets China’s experience apart. Party
schools are implicated in this bureaucratic transformation as sites where new bureaucrats
are trained and selected for higher office. As the next chapter will detail, these schools
are significant organizational channels for moving bureaucratic talent to leading positions
within the party and government.
83 To ensure turnover of old revolutionary cadres, the enforcement and diffusion of retirement norms have been an important development; see Manion (1993).
64
Chapter 3 Managing the Managers
Studies of single-party regimes have begun to unpack the “black box” of the party
to understand the informational challenges that party leaders face and the strategies they
pursue to address these problems. Hierarchical, single-party systems must solve the
problems bound up in the principal-agent relationships that pervade the party and, for
party-state systems, characterize the relationship between the ruling party and its state
bureaucracy. Understanding the delegation of responsibilities in this relationship requires
an examination of the interests of each set of actors as well as the institutional context in
which policy decisions are made. For the party managers who are principals to the agents
charged with implementing policies and making decisions at the street level, there exist
two problems, one of “hidden action” by agents on the job and a second of “hidden
information” (Braun and Gilardi 2006; Epstein and O'Halloran 2006; Milgrom and
Roberts 1992; Miller 2005). Problems of “hidden action”, or moral hazard, are mitigated
through monitoring and the structuring of incentives.84 Both “police patrol” and “fire
alarm” mechanisms, which draw upon the monitoring abilities of higher authorities and
receiving publics, respectively, may curb agents’ temptation to shirk their administrative
duties (Lorentzen 2008; Lupia and McCubbins 1994; McCubbins and Schwartz 1984).
More generally, principals may devise both ex ante and ex post strategies for coping with
these agency problems (Huang 1995; Schubert 2002).
84 Huang (2002b) presents an overview of the direct and indirect controls in place to cope with problems of hidden action in the Chinese bureaucracy. A related discussion of the inability of authorities in the Soviet Union to solve the moral hazard problem within the bureaucracy and the contribution of “hidden action” by bureaucrats to regime breakdown, see Solnick 1998 and Solnick 1996.
65
The problem of hidden information arises prior to and compounds the difficulties
presented by moral hazard. Private information possessed by agents about their abilities
hinders the work of higher level party officials since the ability of superiors to set
objectives for agents depends on private information about agents’ abilities (Miller 2005:
138-158). Hidden information also affects personnel decisions regarding hiring, firing
and promotion and can result in adverse selection. In political labor markets affected by
adverse selection, the presence of hidden information by the bureaucratic agents
competing for official posts can result in two undesirable outcomes for the party:
increasing the cost to the party of supplying services or decreasing the revenues the party
is able to collect (Milgrom and Roberts 1992: 149-159). Findings in studies of executive
and legislative control over the U.S. bureaucracy suggest that appointment processes,
which include decisions about structure as well as personnel, have the largest influence
on policy outcomes: “Only if there is slippage at the appointment stage do either agency
discretion or after-the-fact political control take on any importance,” (Calvert,
McCubbins, and Weingast 1989: 605).
Furthermore, among the various external and internal influences on a bureaucrats’
compliance with organizational objectives, the most important are inherent to the agents
themselves. That is, selecting agents with professional, functional preferences that are
consistent with the objectives of the organization yields more effective bureaucratic
administration than the monitoring used to address problems of hidden action. In this
sense, “the process of selecting and indoctrinating bureaucrats is the process that
matters … [and] the problem of adverse selection trumps the problem of moral hazard,”
(Brehm and Gates 1997: 202).
66
There exist two alternatives for mitigating this selection problem: inducing agents
to reveal private information or devising strategies to screen for certain types of agents.
These differ in terms of the initiating party. Agents may signal their abilities in a bid to
differentiate themselves from less qualified contenders, for example through education
credentials (Spence 1973). Signaling may be a powerful mechanism for overcoming
selection issues, but they depend on the accuracy and credibility of a signal for conveying
information. An educational signal is accurate so long as low productivity (or less
desirable) workers are not able to or uninterested in expending the effort to attain the
higher levels of education that signal high productivity.
In contemporary China, a credentialing frenzy has taken hold of the bureaucratic
class. Party and government bureaucrats now signal their abilities through ever more and
higher educational credentials (Lee 1991; Li 2007b). The expansion of higher education
and the emergence of technocratic norms have contributed to this movement. However,
signaling by bureaucratic agents may break down for at least two reasons. First, there are
limited ways an individual can signal certain unobservable qualities that may be valuable
to the party, such as commitment to the political system. Political authorities may instead
opt to design ways to screen for these qualities. Second, credentials may not signal
ability accurately. Following the unrest of the Cultural Revolution, in which the
educational sector shut down and a “lost generation” of uneducated citizens entered the
market for political office, many high ability bureaucrats did not possess requisite
credentials. One consequence was a pool of political and managerial talent lacking
outward signs of competence. While many of these individuals have sought to obtain
67
higher education degrees in the decades since, the variation in degrees and degree-
granting institutions presents yet another complication.85
In addition to agent signaling, principals may devise and rely upon screening
mechanisms to differentiate between the abilities of competing agents. From this
perspective, selection for enrollment in an educational institution may also serve a
screening purpose (Arrow 1973). Schools serve as “double filters” – in the selection of
students and the evaluation of students over the course of their schooling. In an
authoritarian political system where the ruling party wishes to screen agents on a variety
of dimensions – objective managerial skills, policy expertise, and party loyalty – the use
of party-controlled screening mechanisms remains a politically logical move for party
authorities (Zhao and Zhou 2004). Political criteria, which may render the screening
strategy adopted less efficient than relying purely on educational credentials (Stiglitz
1975), nonetheless explains the persistence of party-managed organizations for training
and credentialing bureaucrats.
This selection problem becomes even more salient during a period of transition,
when central party decision makers begin to value one type of agent over another. In
communist party systems, the transition from planned to market economies demanded a
new type of public manager. While party membership may continue to serve a
preliminary screen, political loyalty is not enough. Public managers in a market economy
face new challenges: enforcing vastly changed and increasingly complex regulations,
forming relationships with contractors in the new private sector, and, in some cases,
making shrewd decisions about public investment and enterprise.
85 As the analysis below will show, degrees from party schools are often sought by cadres but are not correlated with an increased likelihood of promotion.
68
Scholars have documented the bureaucratic transformation that has taken place to
cope with the demands of economic transition (Lee 1991; Walder 1995). Bound up in
this Deng-era push for “a more educated and specialized” bureaucratic class have been
efforts by party authorities to retain control over who is selected to move from a party
activist to party manager role – who shall represent the party, carry out the tasks of
governance, and also benefit from the privileges of office. Controlling promotion to and
within the cadre ranks is crucial because cadres – and leading cadres in particular – make
the decisions that radiate throughout the millions of party members charged with toeing
the party line and maintaining party rule over the general population (Walder and Zhao
2006).
In this chapter, I will first detail the scope of the selection problem facing CCP
leaders. Then, drawing on data collected in the 2003 China General Social Survey, I will
test whether party school training has an effect on promotion to higher office. This
chapter also examines patterns in the career trajectories of cadres trained at the Central
Party School to determine which posts in the party and government such high-level party
school alum eventually occupy. As pipelines to higher office, party schools constitute an
organizational means within the party to control access to positions of authority.
2 Scope of the selection problem
Who is a member of the political elite in China? Scholars have identified this
population in general by employing a variety of criteria, beginning with the vague
parameter of those in possession of “decisive” political power (Smith 1979: Appendix A)
or those who enjoy “exclusivity, superiority, and domination” (Farmer 1992: 2). This is
69
consistent with Putnam’s (1976) emphasis on those who “influence[e] the policies and
activities of the state, or (in the language of systems theory) the … authoritative
allocation of values,” (p. 6). These definitions, which have the advantage of comparative
application, are difficult in practice to apply defensibly to particular cases. In earlier
work, Mills’ (1959) defines a “power elite” as those in positions of authority. I draw
from this precedent and employ a positional, or occupational, approach to defining the
political elite in China. Those members of the party and state bureaucracy who have
attained some “leading” rank at the level of county magistrate or equivalent, are
considered members of the political elite within China, though attaining such rank often
requires marching up the grassroots ranks of the party-bureaucracy or civil service. The
disadvantage of this approach is its emphasis on formal title, as opposed to informal
bases of power, which is risky in an increasingly diverse Chinese society where
entrepreneurial talent, global connections, and political authority may be interconnected
but separate bases of political influence.
In a Leninist system, cadres are responsible for party and government
administration at various levels of government and across functional areas of
specialization. This population of party and government managers is then divided into
increasingly smaller and exclusive ranks, at one time up to 25 ranks in the PRC.86 “To
distinguish itself from the Nationalist Government that it had fought in a civil war, the
CCP referred to its functionaries by the generic term ‘cadre’ (ganbu), regardless of
whether they worked for the party, the Government or the army. In this usage, cadre
referred to those who had a certain level of education (initially secondary school level),
86 Interview 112 with a Central Party School teacher and party historian, February 2008. Today, the ranking system has been streamlined to two ranks per administrative level, and this system is compatible with the hierarchy within the ministry system.
70
who had some specialist ability, and who carried out ‘mental’ rather than ‘manual’
labor,” (Burns and Bowornwathana 2001: 23).87 More bluntly, a cadre is anyone who
“eats the state’s grain” (chi guojia liang shi).88
At present, the Chinese bureaucracy, in all its organizational variety, comprises
over 46.5 million individuals.89 Table 3.1 below offers a sketch of the size of the entire
bureaucratic system and levels within the system:
Table 3.1: The organization of public officials in China, 1998 Administrative Level
Administrative Rank
N “leading” (lingdao) cadres?
Examples
Township and below
Section (ke) level and below
~46 million No Section head in a county-level ministry, township party secretary
County Department (chu) level
500,576 Yes County party secretary, mid-level supreme court judge
City Bureau (siju, diting) level
45,688 Yes City mayor, Provincial party school principal
Provincial Ministry (bu) level Yes Ministry head Central Premier (zongli)
3,665 Yes
Source: COD (1999) and Heilmann et al. (June 2000). At the very top of this hierarchically-organized system is a stratum of individuals whose
appointments are managed by the Central Committee of the CCP.90 A slightly larger
population of “leading cadres” (lingdao ganbu) comprises officials from the county and
department levels on up. This group, collectively, possesses local policymaking and
allocation authority for the party and state. Leading cadres maintain the party’s political
dominance and the state’s administrative authority. This nomenklatura and leading cadre
class produces the policies which the rest of the bureaucracy must implement (Burns 87 Burns draws from Strauss’ distinction between “lettered official” (wenguan), public servant (gungwuyuan) and cadre (ganbu) and Lee (1991) for this definition of cadre. 88 Interview 112, Central Party School party history department, February 2008. 89 For a breakdown of this figure by funding type, see Ang 2008. Shambaugh distinguishes party cadres, which number 6.9 million, from state cadres, which number 33.6 million. He cites a 2002 Central Organization Department source for these numbers. These differ from the Ang figures, which derive from a 2003 Ministry of Finance publication and include public service unit employees. 90 Changes in nomenklatura are reviewed in Burns (1994; 2003)
71
1989b; Burns 2006). In 1998, this category of political and administrative leaders, or
“leading cadres”, totaled 549,929 individuals (Central Organization Department 1999).
In other words, the more than 45 million public officials in China must be sifted through
to produce an elite decision-making corps of fewer than 1 million.
Walder (2004) sums up the selection process at work within this bureaucratic
population and the critical importance of party and bureaucratic discipline:
The political elite of 500,000 cannot rule the country unless it can retain the obedience of 40 million state cadres — especially the 38 per cent of whom are Party members. … If the elite maintains the discipline of state bureaucrats and the allegiance of Party members, it can withstand challenges from other groups in society, even in periods of economic hardship and social upheaval. If, however, challenges from other groups stimulate a defection of the Party membership and parts of the state bureaucracy, the elite is in real trouble (this is what happened in China, temporarily and briefly, in May 1989).
This discussion illustrates what is at stake should the party lose control over its elite cadre
ranks, though smart political selection is aimed at mitigating this problem.
CCP authorities thus face an enormous selection problem: they must devise
strategies for choosing from among the 46 million low-ranking cadres at the lowest,
deputy-section level to promote to the leading ranks at the county (department), city
(bureau), provincial (ministry), and central levels. Fewer than 600,000 individuals
populate these elite levels of the party and government hierarchies, which translates to a
selection rate of less than two percent. Low bureaucratic turnover, as discussed
previously, also contributes to the tight competition for elite posts. To determine who
will be the one in a hundred to earn a place among the leading cadre ranks, the CCP has
in place the regulatory and organizational framework to vet cadres through party schools.
72
Cadre training at party schools is not the only solution, however. Across
authoritarian systems, there exist a variety of pathways for resolving this selection
problem. Performance evaluation presents one alternative. Cadre evaluation has evolved
considerably from a system of annual reviews based on vague performance standards to
an increasingly regularized and rule-based institution, though it is still weakened by
patronage relations (Chow 1988; Chow 1993; Heimer 2006; Minzner 2009).
Modifications to evaluation criteria over time also indicate “adaptive learning on the part
of principals” (Whiting 2004: 101). Evaluation systems are a significant aspect of the
CCP’s overall ability to monitor cadre behavior, but they require costly, police-patrol
monitoring. The recent cadre evaluation procedures tested and promulgated in 2006 are
even more onerous on local party personnel bureaus, as they call for internal party
assessments along with feedback from the public (Guo and Gore 2007). Furthermore,
evaluation objectives are difficult to identify as the efforts of local officials do not relate
directly or immediately to local developmental outcomes (Landry 2003). Monitoring and
evaluation of cadre behavior has been further complicated by the entry of market
institutions, which create incentives for opportunistic activity while placing greater stress
on the party’s monitoring capacity (Nee and Lian 1994).
Another alternative for identifying political talent is through controlled elections.
Elections are a common feature of non-democratic regimes; during the post-World War II
period, 98 out of 101 non-democracies held at least one legislative or presidential
election.91 In PRI-ruled Mexico, for example, elections were employed nationwide for the
selection of local and national leaders and to provide party leaders with information on
91 Analysis of the Przeworski et al. dataset in which regime type is coded based on the presence of competitive elections (Przeworski et al. 2000).
73
public opinion and candidates’ mobilizational abilities (Magaloni 2006). In the Chinese
case, elections began as village-level experiments in the 1980s and have since become
compulsory exercises nationwide. As a selection mechanism, elections have been limited
to the very lowest levels of governance and are proceeding slowly, on a trial basis only,
to the township and county levels.92
Selection through party schools may also coexist with patronage networks and the
privileges accruing to hereditary ties. Clientelism in CCP court politics has long been a
topic of study, and there is also evidence of the benefits of a “princeling” background for
high political office (Li 2008a; Nathan 1973).93 This reflects the mix of impersonal and
personal in China’s bureaucratic culture, where there exists the “paradox of bureaucratic
impersonality and personalization of administrative ties” (Zhou 2010). While these
personal networks are difficult to trace in practice, their influence may be bound up in the
processes by which individuals are invited to and evaluated in a party school training
program. Party school training thus reflects the distillation of many disparate pieces of
information about a particular candidate and her likelihood to perform well in higher
office. From this informational perspective, selection through party school training
programs carries advantages over selection though alternatives such as performance
evaluation. In short, a nationwide network of schools provides the systematic, orderly
option that is consistent with the CCP’s drive to deepen the institutionalization of party
governance.
92 Studies of elections in China are extensive, and local party officials may use election results to gather information on local political talent. Reviews of this literature are numerous (Alpermann 2001; Epstein 1996; Gadsden and Thurston 2001; Kelliher 1997; Manion 1996; Oi and Rozelle 2000). 93 This is despite official COD regulations, for example those against employing family members in the same work unit. See Bo (2004).
74
All of these control mechanisms point to the dense, overlapping institutions that
contribute to the crucial task of personnel management. Whereas elections are risky and
performance evaluations are resource-intensive, vetting candidates through party schools
may capture whether a candidate possesses various ingredients for political success in
China. These include the desire and ability to signal political ambitions, access to
political networks, and managerial potential. As the organizations where these various
individual qualities may be observed and developed, party schools are key sites for
building China’s administrative and political elite.
3 Party school training: A pipeline to higher office
There is both official and unofficial support for the argument that party school
training performs a gatekeeping function. The Chinese popular press has reported, “on
[the Central Party School] campus even taxi drivers stop for pedestrians because they
don’t know whether that someone crossing the street in front of them may be the CCP’s
next party secretary.”94 Such recent, and often positive, coverage of the Central Party
School in the popular press, domestic and foreign, suggests more critical testing of
competing hypotheses regarding the substance of party schools’ role within the CCP.95
In the provisional 2006 Cadre Education and Training Work Regulations, a cadre’s
education and training record is considered an “important factor for promotion” (Article
41).96 Development of cadre training and education is also a recommendation proffered
94 Southern Weekend (Nanfang Zhoumo), “China’s most special school – the Central Party School,” 8 November 2007. 95 For coverage of the party school system in English language outlets, see for example Melinda Liu and Jonathan Ansfield, “Life of the Party,” Newsweek, 30 May 2005. 96 For a full text of the trial regulations, see http://wlzx.hdpu.edu.cn/zzb/Article_Show.asp?ArticleID=858 http://www.sz.gov.cn/rsj/zcfggfxwj/zcfgjgfwjqt/rsbzgbywpx/200809/t20080917_50492_4369.htm, accessed August 2008. Training requirements applied to SOE managers as early as 1991, when a decision
75
in the internally-circulated (neibu) 2004 publication, Research on Certain Questions in
Building a Governing Party (zhizheng dang jianshe ruogan wenti yanjiu). As one official
expressed, “Training is critical to promotion because we are a one-party system, and the
party needs training to control human resources.”97 Media accounts further claim that
“one-third of Young Cadre Training Class graduates [from the Central Party School] are
promoted to the provincial (ministry) rank.”98 Young Cadre Training Classes, which all
party schools convene each year in accordance with local training plans, remain the most
selective and prestigious programs for ambitious party officials.
Despite these assertions, the actual effect of party school training on a cadre’s
promotion prospects remains unexamined. Furthermore, whether party schools serve this
functional purpose within party organization remains untested in extant literature. The
existence of “party-sponsored upward mobility” does suggest that party school training,
as an educational option nested within a larger universe of party-related educational paths,
serves a sorting function in the context of party personnel policy (Li 2001; Li and Walder
2001). To date, no scholarly work on party schools has tested whether schools are an
organizational strategy for addressing the selection problem inherent to a large and
complex bureaucratic system. This chapter thus presents several empirical tests of the
argument that party school training constitutes a channel for individuals to move up the
cadre ranks. This in turn investigates whether and how Leninist mechanisms of cadre
control remain relevant in the transition to a market economic context.
was made at a National Enterprises Leader Training Conference: “Starting in 1993, all leaders of the nation’s large- and medium-size enterprises would not be qualified for their leading positions unless they received a certificate of professional training issued by the National Economic and Managerial Cadre Training Committee,” (Liu 2001: 107). 97 Interview with a retired central-level ministry cadre, Interview 34, October 2007. 98 Zhao Xue, “China’s most special school – the Central Party School,” Southern Weekend, 8 November 2007.
76
This line of reasoning stands in contrast to the null hypothesis, discussed
previously, regarding the purpose of party schools in the CCP’s organizational universe.
It may be the case that party schools are only a reward for cadres who may have earned
the opportunity to withdraw from work responsibilities for several days or weeks to relax
on a party school campus. In this sense, time spent at party schools constitutes a “club
good” to be shared among party managers. In interviews, some cadres also expressed a
belief that party school “vacations” are part of the perks of a cadre occupation. “Maybe
10 to 20 percent of all section-level cadres will be chosen to attend the city [party
school’s] Young Cadre training class, and being sent to this is really a reward of some
sort.”99 From this perspective, invitation to a party school training class does not
necessarily indicate potential to climb the administrative ranks. Instead, the invitation to
train at a party school is an end unto itself.
To adjudicate between these hypotheses, national survey data (from the 2003
China General Social Survey) will be used to test the following:
Hypothesis 1 (null) Party schools as rewards: If a cadre attends a party school training program, then s/he is not more likely to be promoted to a higher administrative rank.
Hypothesis 2 Party schools as pipeline to higher office: If a cadre attends a party
school training program, then s/he is more likely to be promoted to a higher administrative rank.
The latter hypothesis implies that party school training of any amount has a positive
effect on a cadre’s promotion prospects.
Results from tests of these hypotheses suggest that enrollment in different types of
party school training programs affects whether a cadre is promoted and how quickly.
99 Interview with a city-level cadre, January 2008, Interview 104. This sentiment was echoed in another interview with a towship-level cadre, May 2008, Interview 196.
77
The first test considers the subset of cadres within a national survey of Chinese citizens
(N=589). I employ this data to ascertain whether the “treatment” of training at a party
school has a positive effect on career advancement. Selection bias poses a
methodological challenge for identifying the effect of training, and I employ propensity
score matching in an attempt to correct for this. Results from estimating the average
treatment effect of training show that individuals who have been selected for a party
school training program are more likely to reach a higher administrative rank compared
to individuals with similar characteristics but no training experience.
This test captures the general, positive effect of all party school training programs
on career advancement, but field interviews suggest that the most important training
program for upward bound cadres is the Young Cadre Training Class organized by party
schools throughout the system. A second empirical test considers a subset of the leading
cadre population, those who have attended the Central Party School’s version of this class.
Analysis of the career histories of this elite population demonstrates that individuals who
participate in this highly selective training course reach high office at younger ages than
the national average, thus laying the groundwork for long tenures of service to the party.
4a Analyzing the effect of training in the 2003 China General Social Survey
Ideally, promotion serves two functions within organizations: to match
individuals of certain skills with appropriate office and to generate incentives for
improved performance (Baker, Jensen, and Murphy 1988). In trying to identify the effect
of training on promotion, selection bias arises as a methodological problem. It may be
the case that cadres with certain characteristics are more likely to be selected for party
78
schools and promoted, and these confounding factors confuse the true effect of party
school training on cadre promotion. To determine the effect of training, the
counterfactual question to answer is, what would have happened to the careers of those
who received the treatment of party school training if they had not received treatment?
Conversely, what would have happened if those who were not sent to training were given
a party school training opportunity? One way to estimate the effect of a treatment is
through experimental design: randomly assigning treatment to an individual and none to
another, identical individual. When such an experimental approach is not possible, as in
the present study, an alternative is to create a counterfactual group by matching survey
respondents along a set of observable characteristics.
I employ propensity score matching, a technique to reduce selection bias by
predicting via a probit regression and a set of observed predictors the probability that an
individual will be in a treatment versus control group. Pre-treatment characteristics are
summarized in a propensity score for matching purposes. After determining propensity
scores for each individual, randomly drawn nearest neighbor matching from the common
support region is employed: an individual in the treatment group is matched to an
individual in the control group based on the closeness of their propensity scores. In the
case of a tie between individuals in the control group, a match is randomly drawn
between tied controls for a given individual in the treatment group. Drawing from the
common support places an overlap condition on the individuals that may be matched and
maximizes exact matches while excluding more cases of inexact matching. The average
treatment effect on the treated (ATT) is then determined by differencing the mean values
on an outcome variable across matched individuals from the two groups.
79
Propensity score matching reduces, but does not eliminate selection bias entirely.
Bias exists so long as treatment is not assigned completely randomly to two individuals
with the same propensity score (Becker and Ichino 2002). Furthermore, it is limited in
that hidden biases may still remain because matching only controls for observed variables
included in the first-stage probit regression (Pearl 2000). Mismatch error is also difficult
to calculate using this technique (Lan and Rosenbloom 1992). Yet propensity score
methods, while imperfect, have been shown to produce results closer to experimental
treatment effects than nonexperimental estimates (Dehejia and Wahba 1999).
Data for this analysis were obtained from the China 2003 General Social Survey,
a representative national sample comprising 5,894 Chinese citizens. A total of 589
respondents reported careers subject to the party and government ranking system.
Table 3.2: Descriptive summary of cadre and general population Cadres
N=589 General Population N=5,894
Variable Mean SD Min Max Mean SD Min Max
Age 50.8 12.1 20 72 43.4 13.1 15 77
Total years of education 6.8 4.6 0 70 4.3 3.5 0 70
Years of CCP membership 22.4 13.1 0 56 19.6 13.2 0 56
Source: 2003 China GSS This survey was jointly conducted by the Hong Kong University of Science and
Technology Research Center and the Sociology Department of the People’s University of
China.100 Table 3.2 summarizes characteristics of the cadre sub-population within the
general population surveyed. Within this population, 156 (26.5 percent) were female and
408 (69.2 percent) were CCP members. This compares with a CCP membership of 18.6 100 A report of the CGSS sampling design and survey methodology is available at http://www.ust.hk/~websosc/survey/GSS2003e0.html. Tibet, Hong Kong, Macau, and Taiwan were excluded from the survey. The ratio of the urban sample size to the rural sample size is 5900:4100, giving an urban bias to the sample. First-stage egression estimates are presented with weights to compensate for this.
80
percent among the sampled population, of which 51.9 percent were female.101 Somewhat
contrary to Deng’s prescription for transforming the cadre population, cadres in this
sample were older than the general population. Consistent with Deng’s mandate,
however, this group was more educated and possessed longer tenures of CCP
membership compared to the general population. Mean comparison tests show that
differences in each of these areas are statistically significant at the 0.01 level.
The outcome of interest is whether, conditional on party school training, a cadre is
more likely to be promoted to the next higher administrative rank. A dummy outcome
variable was coded for the three administrative categories for which there are sufficient
survey respondents in the China GSS: no rank/section rank, section rank/department rank
and no rank/some rank. Due to small population size, a separate analysis was not
conducted on those who reported a bureau rank (ditingju ji) or higher in their careers
(N=15). The histogram below summarizes the distribution of administrative ranks over
the 589 surveyed cadres:
Figure 3.1: Frequency of administrative rank, 2003 China GSS
27
108
35
100
241
63
6
9
0
50
100
150
200
250
300
350
400
Below Section Section Department Above Department
Rank
Tota
l ind
ivid
uals
No party schoolAttended party school
101 The national numbers are lower: according to 2007 figures, total CCP membership stood at 5.6 percent of the total population. See “Chinese Communist Party membership exceeds 74 million in 2007”, Xinhua, 1 July 2008, available online at http://www.china-embassy.org/eng/gyzg/t470863.htm, accessed 28 December 2008. National population figures available from China Data Online.
81
A dummy variable which captures whether an individual attended a non-diploma
training program at a party school is hypothesized to have a significant, positive
correlation with the administrative rank outcome variables. These are the programs in
which cadres are chosen by local organization departments or work units, not the degree-
based programs with open enrollment.
The first-stage probit regression model will estimate individual-level propensity
scores of the treatment (non-diploma party school training) on several demographic
characteristics, including an individual’s age, gender, and total years of schooling.102
Individuals also reported their socio-economic status.103 Political control variables
included an individual’s party membership and that of his parents, whether or not an
individual served in the military, and whether an individual was “sent down” to the
countryside for hard labor during the Cultural Revolution. Finally, a dummy variable
coded for whether an individual self-reported frequent interaction with his direct
managers.104 Individuals’ home provinces were also coded as a set of dummies.
Descriptive statistics of these variables are reported in Appendix D. Table 3.3 below
gives the results from the first-stage probit regression analysis, where the dependent
variable is whether an individual was chosen for party school training.105
Several of these control variables were statistically significant. Unsurprisingly,
party membership was an important characteristic among those selected for party school
102 Alternate specifications for these control variables included total years in the workforce (instead of age) and highest educational level attained. 103 While self-reported and subjective, these dummy variables were used instead of income data. There were irregularities with self-reported income data, for example individuals reported monthly and annual incomes that were often inconsistent. 104 These data are unanchored. The nature of these interactions and the actual frequency as a measure of time per week, for example, were not given in the data. 105 This model was estimated using the pscore command in Stata.
82
training, though parents’ party membership was not. Those with military backgrounds
were less likely to undergo party school training, and this may be due to the separate
career track – along with related training at military academies – for individuals in the
military. Since the early 1980s, with policies to encourage the professionalization of the
PLA, there has been increased emphasis on the recruitment and training of military
officers at military academies (Zheng 1997: 228-231). Those with “sent down”
experiences were also less likely to attend a party school training class, but this was
marginally significant at the 10 percent level. It is unclear why these particular life
events might be negatively correlated. Some provincial-level dummy variables were
significant, both substantively and statistically, in the first-stage probit regression, but
there does not exist a theoretical explanation for why these particular locales might
Table 3.3: Probit regression results for propensity score estimation DV: non-degree party school training program (dummy) Coefficient
Robust SE
Constant -2.05 0.90** Age 0.001 0.01 Female -0.05 0.18 Education (total years) -0.01 0.02 SOE level (lower middle) -0.26 0.25 SOE level (upper middle) -0.12 0.24 SOE level (upper) -0.06 0.31 Political control variables (dummies) CCP membership 1.24 0.20*** Father’s CCP membership -0.02 0.18 Mother’s CCP membership 0.19 0.25 Military experience -0.44 0.18*** Sentdown experience -0.41 0.23* Frequent interaction with superiors 0.32 0.50 Provincial-level dummies Beijing 0.29 0.60 Tianjin 0.65 0.59 Hebei 0.70 0.66 Shanxi 0.04 0.86
83
Provincial-level dummies, continued CoefficientRobust SE
Neimenggu 0.95 0.71 Liaoning 0.04 0.65 Jilin 1.34 0.88 Heilongjiang 0.09 0.82 Shanghai 0.89 0.60 Jiangsu 0.71 0.64 Zhejiang 0.55 0.82 Anhui -0.33 0.63 Fujian 0.68 0.68 Jiangxi 0.63 0.74 Shandong 0.29 0.68 Henan -0.23 0.63 Hubei 1.41 0.61** Hunan 1.17 0.61** Guangdong 0.45 0.60 Guangxi 0.44 0.64 Chongqing 1.95 0.83** Sichuan 0.68 0.73 Guizhou -0.78 0.75 Yunnan 0.58 0.63 Shaanxi -0.46 0.79 Gansu 1.30 0.73* Iteration 0: log pseudolikelihood -351.59717 Iteration 1: log pseudolikelihood -279.76579 Iteration 2: log pseudolikelihood -275.81003 Iteration 3: log pseudolikelihood -275.73312 Iteration 4: log pseudolikelihood -275.73306 N 574 Wald chi2(45) 107.71 Prob > chi2 0 Pseudo R2 0.22
Notes: Significance codes: *** for p<0.01, ** for p<0.05, * for p<0.10. To ensure that the mean propensity score is not different for treated and control groups across blocks, the test of balancing property was satisfied for this model. This tests whether means of covariates are the same across blocks. Dropped categories: “Lower class” SOE level (dummy); Xinjiang regional dummy. Source: China GSS 2003 correlate positively with selection for a party school training class. The table below gives
a summary of the marginal effect of each statistically significant control variable. An
individual who is a party member, for example, is 44 percent more likely to attend a non-
84
degree party school training program than a non-party member, holding all other
characteristics constant.
Table 3.4: Marginal effect for significant control variables
Notes: Other control variables were held constant at their medians.
During the second stage of the propensity score matching procedure, individuals
are matched based on the proximity of the propensity scores assigned to them in the first
stage probit regression. The histogram and table below illustrate the distribution of
estimated propensity scores for individuals by treatment status. Treated individuals in
Figure 3.2 are tallied in the bins above the horizontal axis, while untreated individuals are
below the axis.
Figure 3.2: Distribution of propensity scores for treated and control groups
0 .2 .4 .6 .8 1Propensity Score
Untreated Treated
The histogram shows that there is overlap between the treatment and control groups for a
broad range of propensity scores. Treated individuals with a propensity score of 0.8 and
higher have no comparisons in the control group and are not included in estimations of
ATT limited to the common support. However, because of the broad overlap in
Variable Marginal
effect CCP membership 0.44 Army experience -0.16 Sentdown experience -0.15 Hubei 0.50 Hunan 0.42 Chongqing 0.69 Gansu 0.46
85
propensity scores across treated and untreated groups, it is possible to carry out
differences-of-means tests across the two groups and determine the ATT.
Table 3.5: Total cadres in treated and control groups, by propensity score block Propensity score block Control Treated Total
0 235 19 254 0.2 97 50 147 0.4 66 60 126 0.6 15 43 58 0.8 0 4 4
Total 413 176 589 The following kernel density graphs demonstrate the distribution of propensity
scores across the treatment and control groups both before and after matching. While the
kernel densities are not perfectly matched in the post-match graph, the distance between
the treatment and control groups has decreased after the matching process.
Figure 3.3: Pre-match propensity score distribution
01
23
Ker
nel d
ensi
ty o
f res
pond
ents
0 .2 .4 .6 .8Propensity score
Treated pre-match Control pre-match
Figure 3.4: Post-match propensity score distribution
0.5
11.
52
Kern
el d
ensi
ty o
f res
pond
ents
0 .2 .4 .6 .8Propensity score
Treated post-match Control post-match
To confirm that matches are balanced, the results of t-tests for the equality of
means in the treatment and control groups, both before and after matching, are presented
in Appendix D. There were no statistically significant differences in the means of control
variables across the treatment and control groups, post-match, with the exception of the
“lower middle SOE level” dummy. This variable was not significant during the first-
stage probit regression, however.
86
Using nearest neighbor matching over the common support, Table 3.6 below
presents the ATT on several binary outcomes of interest, whether a cadre was promoted
to a section, department, or any rank during his career.
Table 3.6: Estimation of average treatment effect on the treated (ATT) Treatment: Enrollment in a non-degree party school training class DV N treated N control ATT SE t-statistic Section rank (dummy) 170 86 0.221 0.078*** 2.861
Department rank (dummy) 170 66 0.137 0.060*** 2.266
Any rank (dummy) 170 96 0.229 0.062*** 3.708
Notes: Estimates obtained over the common support using nearest neighbor matching (random draw version). Bootstrapped standard errors are reported (200 replications) These results are statistically significant at the 99 percent level. For the two
administrative ranks with sufficient numbers of surveyed individuals, those who
participated in a party school training program had a higher likelihood of promotion than
individuals with similar observable characteristics who did not attend party school. To
interpret the ATT in substantive terms, an individual who attends party school is 22.1
percent more likely to attain a rank of section chief than a peer who did not attend a party
school training class, while this becomes 13.7 percent for attaining the rank of
department chief. Those who attended party school training were 22.9 percent more
likely to achieve some rank in comparison to untrained peers. Results from propensity
score matching confirm the core intuition articulated at the beginning of this section, that
party school training has a significant, positive effect on promotion up the ranks of party
and government office. It is also notable that the treatment of any non-degree party
school training includes all manner of programs, from core to auxiliary to vocational
classes, so the effect of training on promotion might be even higher for certain key point
87
classes. Data limitations are such that I am unable to determine the particular type of
training class beyond its degree designation.
In comparison, separate tests to determine the ATT of party school degree
programs found no statistically significant effect (Table 3.7).
Table 3.7: Estimation of average treatment effect on the treated (ATT) Treatment: Party school degree (dummy) DV N treated N control ATT SE t-statistic Section rank (dummy) 52 35 0.010 0.104 0.093
Department rank (dummy) 52 38 -0.064 0.100 -0.645
Any rank (dummy) 52 47 0.000 0.075 0.000
Notes: Shandong and Guizhou dummies were dropped from the first-stage probit regression (and Xinjiang included) in order to satisfy the balancing property. Estimates obtained over the common support using nearest neighbor matching (random draw version). Bootstrapped standard errors are reported (200 replications) This suggests that individuals who enroll in party schools for a degree are not more likely
to be promoted to higher administrative ranks, reinforcing the importance of selection
and invitation to a training program. Signaling the intention to serve in the party
bureaucracy by pursuing a party school degree does not appear to yield payoffs in terms
of climbing the career ladder.
On the other hand, individuals who attain a university degree do stand better
chances of promotion to the leading cadre ranks (Table 3.8). While a university
education does not have an effect on an individual’s promotion to the lowest section
ranks, this changes dramatically for the transition from section to department, or non-
leading to leading cadre positions. As the table below illustrates, individuals with a
university degree are 25.9 percent more likely to be promoted to the department rank than
peers without a degree. Overall, the findings are somewhat uneven: an individual with a
university degree at the lowest administrative ranks might not expect higher chances of
88
promotion, but there does appear to be a strong effect for the transition to a career as a
leading cadre.
Table 3.8: Estimation of average treatment effect on the treated (ATT) Treatment: University degree (dummy) DV N treated N control ATT SE t-statistic Section rank (dummy) 122 45 -0.138 0.090 -1.526
Department rank (dummy) 122 50 0.259 0.100*** 2.603
Any rank (dummy) 122 59 -0.057 0.061 -0.948
Notes: Estimates obtained over the common support using nearest neighbor matching (random draw version). Bootstrapped standard errors are reported (200 replications)
Finally, what is the combined effect of both party school training and higher
education? Previous studies have shown that the combination of party membership, as a
key political credential, and higher education open the doors to administrative positions
of authority (Walder 1995). Since party school trainees are a subset of the party
membership, this joint effect should be particularly salient for high administrative ranks.
The results from the data, however, are mixed (Table 3.9).
Table 3.9: Estimation of average treatment effect on the treated (ATT) Treatment: Party school training and university degree (dummy) DV N treated N control ATT SE t-statistic Section rank (dummy) 39 23 0.038 0.124 0.308
Department rank (dummy) 39 24 0.265 0.131*** 2.017
Any rank (dummy) 39 30 0.051 0.097 0.530
Notes: Estimates obtained over the common support using nearest neighbor matching (random draw version). Bootstrapped standard errors are reported (200 replications) The joint effect of party school training and higher education on the likelihood of
promotion to the leading cadre ranks is greater than for either of these credentials alone.
Again, there was no effect on the likelihood of promotion to the section rank or a higher
rank generally, but for the important transition from a section to department rank,
individuals with party school training backgrounds and a university degree were 26.5
89
percent more likely to be promoted than their untrained and non-degreed peers. These
findings are somewhat inconclusive, however, since only 39 individuals in the sample
hold both a university degree and have attended at least one party school training
program.
There are additional implications to the hypothesis that party schools prepare
cadres for increasing positions of authority in the party. Analysis of career histories shed
light on whether party school-trained bureaucrats are more likely to end up in key
political, as opposed to technical or administrative, occupations. Previous studies have
identified the emergence of “dual track” career patterns in the reform period (Li and
Walder 2001). These tracks capture the tension between authority and expertise in
bureaucratic relations. Individuals with technical expertise fill functional posts, whereas
individuals who pass through party-sponsored educational programs are channeled to
positions with decision-making authority (Table 3.10). The career trajectories of cadres
in the sample confirms this trend, where individuals with party school training are more
likely to land positions as party leaders rather than more technical positions.
Table 3.10: Occupational categories of cadres, by rank
Vice-section rank
Section rank
Vice-dept rank
Dept rank
Party school training? Y N Y N Y N Y N Occupation percent percent percent percent percent percent Percent percent CCP cadre 50.01 26.83*** 54.16 33.34*** 70 25.92*** 33.33 31.43 Law and order 5.56 3.66 4.17 5.66 10 3.7 0 2.86 Government 25 19.51 22.22 18.24 10 14.81 13.33 2.86* Economic planning 11.12 12.2 5.56 15.73** 0 11.11* 40 14.28** Education 0 6.1* 5.56 8.19 10 25.92* 13.34 17.15 Other 8.31 31.7*** 8.33 18.84** 0 18.54** 0 31.42***
Total 100 100 100 100 100 100 100 100 Notes: Code for significance of two-sample t-tests: *** for p<0.01, ** for p<0.05, * for p<0.10. Data: China GSS 2003
90
4b Testing mechanisms
While the analysis presented here suggests that party school training has some
effect on a cadre’s likelihood of promotion to a higher rank, the next step is to consider
the mechanisms driving this result. At least three mechanisms may be at play, all of
which assist party officials with making decisions of bureaucratic selection. First, it
might be the case that individuals attend these training classes in order to signal their
intentions to pursue careers in the bureaucracies of the party and state, and they might
lobby their superiors or organization department leaders in order to attain these
credentials. Additionally, it is possible that individuals are more likely to be promoted
because of their ability, demonstrated during training, to make strong contacts and forge
the political networks that might propel them to higher positions of authority. On the
other hand, it may also be the case that party schools serve a screening function for
higher party authorities. In this sense, these schools are more of a party-driven – as
opposed to cadre-driven – organizational strategy for solving the selection problem
facing the rulers of a country with a massive, expanding bureaucracy. These are not
mutually exclusive mechanisms and all three may converge on the experience of party
school training. The discussion below will consider each mechanism in turn, as well as
testing strategies to ascertain which mechanisms may be at play.
Signaling
In assisting personnel officials with hiring, promotion, and other selection
decisions, party schools may be used instrumentally by cadres to signal career intentions.
In a competitive market where bureaucratic positions are scarce, those of high ability
91
seek ways to indicate their potential and distinguish themselves from other competitors.
Party school training presents one means for ambitious party officials to signal potential.
In order to accomplish this, an individual might maneuver for an invitation to training
classes.106 If this were the case, then individuals attending party school training would all
be of a certain type, i.e., those with a high potential to serve the party well as bureaucratic
managers. For party schools to serve as a signaling device, it must be the case that only
high potential cadres attend, thereby signaling their type. The intuition behind the “party
school as signaling device” story is that the process of undergoing training does not
reveal any information about cadres’ potential or add to a cadre’s abilities. Rather, only
cadres of a certain type attend party schools.
This “high potential” might comprise many characteristics such as loyalty to the
party, managerial talent, and personal connections. Some of these characteristics are
observable and controlled for through the first-stage regression in the matching procedure
detailed above. For example, the regression analysis captured some qualities of an
individual’s political background: his party membership and parents’ CCP membership,
which may be one proxy for party loyalty or the “redness” of an individual’s background.
Education level also controls for managerial ability. In alternate specifications of the
probit regression, dummies for university majors in management and engineering were
statistically insignificant.
106 In the first-stage probit regression results, however, there was no significant correlation between an individual’s selection for party school training and whether or not she interacted frequently with her managers. It is difficult, however, to draw strong conclusions from this finding due to the lack of information the substance of these interactions and their actual frequency. Furthermore, the categories themselves are subject to interpretation by each respondent, as the questions do not include an anchoring measure.
92
The matching strategy employed in this chapter suggests, however, that party
school training serves other functions beyond signaling. This is because individuals who
wish to signal their type through enrollment in a party school training class are matched
to those with similar characteristics but who are not selected for training. As the results
from matching indicate, high-ability individuals sent to party school training are still
more likely to be promoted than high-ability matches not sent to party schools. This
supports the prospect that while signaling by cadres may be one mechanism at play, party
school training accomplishes additional functions.
Networks
Scholars have also examined the importance of schools for generating the
political networks, and political capital, that propel bureaucrats to higher office. Putnam
(1976) counts educational institutions among the channels for elite recruitment, alongside
parties, bureaucracies, and local governments, due to the professional network-building
that takes place in these organizations (pp. 49-52). France’s grandes ecoles are the
centerpiece of a long tradition in which public and private elite connections are forged
(Suleiman 1977). In his longitudinal study of political elites in Mexico, Smith (1979)
notes the pivotal role of a single national university in building the careers of public
officeholders from 1900 to 1971, since it is at this school where students “formed crucial
friendships and alliances” (p. 86).
China is no exception, and party-managed institutions of cadre training would be
a prime site for the creation of both administrative and personal networks. Party schools
are sites for cadres to reinvigorate “both the formal institutional arrangements and
93
informal relationships that make the party-state work,” (Pieke 2009a: 148-151). One
CPS professor, for example, noted glibly how training classes at party schools are a time
for cadres to “study, make connections, rest, and eat well (xuexi xuexi, lianxi lianxi, xiuxi
xiuxi, mishi mishi).”107 Another city party school vice-principal declared that party
schools “are the only place where networks can be built, which is important for
coordinating across different ministries. … These networks are stronger than college
networks because trainees are concentrated in a locale.”108
While party schools may serve this networking function, which is vital in China’s
bureaucratic and business worlds, evidence from field interviews remains mixed.
Interviewees tended instead to cite the social, rather than professional, value of networks
formed during party school training, though the boundaries between these are blurry.
One provincial-level official who attended several training classes at the Central Party
School, for example, could not recall any professional problems that he actually solved
by accessing networks from his party school training classes.109 Two vice-principals at a
city party school were also dismissive of the network-building potential of their party
school training classes. “Most training classes for cadres in leadership positions aren’t
really about building networks. Most students already know each other before they
arrive.”110 Another party school teacher corroborated this and pointed out how “cadres
107 Interview with a CPS professor, Interview 57, October 2007. 108 Interview 131, April 2008. Another interviewee, a teacher at a county-level party school, noted how training classes are a time for officials to “learn about the work in every department [in government],” (Interview 201, May 2008). Along this vein, during one party school training class that I observed in 2008 in Province B, each trainee was given a red-covered class directory (tongxun lu) at the class’s closing celebration. This particular class included mostly university party secretaries, some Communist Youth League party school principals, and the heads of military academies. All of these officials were working in the same inland province. (Participant observation at a party school training class closing banquet, April 2008.) 109 Interview 182, April 2008. 110 Interviews 202 and 203, May 2008.
94
are seeking broad networks, so they want to attend training classes as high up as possible.
They want more than local connections.”111 The success of connections thus depends on
the type of training class a cadre is sent to, the level of the party school, and her
individual initiative.
If access to elite networks and the opportunity to create all-important personal
networks are benefits of party school training, this effect is uneven across schools,
programs, and individual cadres. One party school trainee, who was sent to CPS for a
one-month, in-residence training class, noted in an interview, “Some students develop
very close relationships, but for me they were nothing special (yi ban de). For me,
networks were not very important in the class that I enrolled in, but this can be an
important aspect of a training class.”112 A county-level official who was sent to a city-
level Young Cadre Training Class reported the opposite experience. “Many of my
classmates [at the Young Cadre Training Class] were city officials, and after training in
2003, I can now just call them and they will take care of (anpai) things for me.”113 Party
school training classes have the potential to generate the personal networks which
enhance individual careers, but the effect appears to be uneven. Until more systematic
data are available to test the utility of these school networks, anecdotal comments from
field interviews paint at best a somewhat murky picture.
111 Former city-level party school teacher, Interview 215, May 2008. 112 Interview with a CPS trainee, now a county-level university official, Interview 65, November 2007. His training class at CPS was for cadres and administrators in universities. 113 Interview with a city-level party school trainee, now a vice-county-level official, Interview 88, December 2007. In an interview with the vice-principal of a provincial-level socialism school, the official also noted that longer classes (“classes longer than two weeks”) are more conducive to the formation of social groups and “social capital”. (Interview 95, December 2007)
95
Screening
Beyond signaling and network formation, selection for party school training may
correlate positively with upward mobility within the party-state apparatus because
training provides party officials with information regarding a particular bureaucrat’s
fitness for higher office. Furthermore, if party schools serve a screening function, it
follows that some individuals will not meet the criteria, either formal or informal,
established for bureaucrats to continue upward in their careers. To test this, one strategy
is to examine post-training career paths. If cadre training itself, and not the selection
process for training, serves a screening function, this implies that a certain number of
cadres may stall in the upward march of their careers in the periods following party
school training. Further analysis of bureaucrats’ careers captured in the China GSS and
the official biographies of Central Party School alumni reveals that not all bureaucrats
selected for training move to higher administrative ranks.
Within the population of bureaucrats in the China GSS, a total of 94 out of the
589 individuals in the sample, or 16 percent, were not promoted to a higher rank
following enrollment in a party school training class. Career histories collected on the
officials sent to key point Young Cadre Training Classes at the Central Party School
offers additional evidence (Table 3.11). These longer term classes, which last anywhere
from several weeks to one year, provide central authorities with an extended period of
time to observe the behavior – e.g., knowledge of party policies, leadership ability – of
select managers. While a certain degree of screening takes place during the selection
process that culminates in an invitation to the class, further screening takes place during
96
the one- and two-year training process at the Central Party School. The table below on
post-training promotion paths suggests that some screening does occur:
Table 3.11: Cadres who do not experience promotion, 2000 and 1995 classes Difference in rank,
(present rank - rank at time of training) N Percent
0 45 16.73 1 145 55.13 2 64 24.33 3 10 3.80
Total 264 100.00 Those individuals whose difference in rank, i.e., their rank in 2009 minus their rank at the
time of training, is zero have stalled in their careers. Based on available data,
approximately 17 percent of trainees are not promoted at all following their enrollment in
the elite CPS Young Cadre training class. This is similar to the “stall rate” in the general
cadre population sample. This figure is not surprising, given the intense competition to
reach higher ranks. While 1.26 percent of all cadres eventually become leading cadres
(department rank and above), only 1 percent eventually attain a bureau rank and this is
whittled down to 0.005 percent at the provincial/ministerial level. If, during ten or more
months of highly monitored training, a cadre lets slip any sign that he is unsuitable for the
responsibilities of national leadership, potential replacements are already being prepared
from among the hundreds toiling away in lower ranks.
Table 3.12 below lists the career tracks of CPS-trained bureaucrats who have
stalled in their careers despite the investment of one year of training in their skills sets.
Of these 45 individuals who were not promoted to a higher rank following their
enrollment in the CPS Young Cadre Training Class, the largest number were on a
propaganda career track. About one quarter of those who stalled were in economic
97
management positions, while the remainder were evenly distributed across science and
engineering, security, and law-related occupations.
Table 3.12: Career tracks of CPS trainees in stalled careers (1995 and 2000 classes) Career track N Percent Science/Engineering 5 11.36 Security 6 13.64 Law 6 13.64 Economic management 11 25.58 Propaganda 17 38.64 Total 45 100.0
In short, evidence exists to support the line of reasoning that party school training
serves some sort of screening function. This is in addition to or instead of a signaling
function; the tests presented in this chapter cannot adjudicate between whether the
signaling or screening function is more salient, only that party schools potentially serve
both functions. Further tests might also establish the criteria used to screen cadres within
these training classes. Creating tests to separate which of these mechanisms is truly at
play presents a challenge, given data constraints, but the present analysis suggests that
information about individual bureaucrats is revealed during party school training and
does have some effect on the subsequent achievement of higher office. In other words,
party school training is more than a device used instrumentally by ambitious cadres who
desire high office and more than a reward distributed to loyal servants of the party.
Training is also a device used by party officials to assess cadres’ potential.
4c One leading cadre’s story
Statistical tests illuminate patterns and tendencies, but individual stories lend
nuance to the life and career trajectories of cadres. The experiences of one such official,
randomly chosen from the 2003 China GSS data, add texture to the numbers presented in
98
the previous sections.114 This official became a leading, department-level (chuji) cadre in
1984, at a state-owned business in Guangdong. She was 45 at the time of promotion and
retired only 11 years later (in 1995). As this was a relatively late promotion, chuji was
the highest rank she acquired over her working lifetime. Such a career achievement was
all the more remarkable because this cadre reached a high rank despite some unlikely
personal background characteristics: she had only a middle school level of education,
neither of her parents was a party member, and she joined the party at the relatively late
age of 37. Her workplace relationships may have facilitated this advancement, however,
given that she reported “frequent” interaction with her superiors, subordinates, and
workplace peers.
She spent the entirety of her 37-year working life in a series of state-owned
enterprises (guoyou qiye) and state-owned businesses (guoyou shiye). Her occupations
ranged from teaching at an elementary school located within a state-owned business,
accounting at an SOE, and finally managing a state-owned business. The critical year in
her career history was 1978, when she switched from accounting to administration and
was promoted to a section-ranking position. This was followed in 1984 with promotion
to the department rank as a company leader (qiye fuze ren). The timing of this official’s
training history is illustrative: in 1978, the year of her first promotion, she attended a six-
month training program at a city-level party school. This was the only time she attended
a party school.
Taking such a condensed tour through one individual’s life history fleshes out the
preceding analysis in at least two respects. First, a party school training course was the
launching point for this official’s transition to an administrative-managerial career within 114 This was survey respondent 1224.
99
the state-owned sector. Data are limited, but it would be ideal to know what type of
training class this was and what kind of training content she might have been exposed to.
It is likely that her training was a combination of remedial education, orthodox theory,
and basic administrative coursework. Key point young cadre classes did not yet exist in
1978. Second, party school training preceded her promotion to the lowest, section-level
rank but did not precede her promotion to the higher, department level. This is consistent
with the statistical findings presented earlier, which show that the impact of party school
training is most pronounced, for rank-and-file cadres, at lower administrative ranks.
5 The select few: Central Party School Young Cadre Training classes
While analysis of the CGSS demonstrated that enrollment in party school training
programs increased the likelihood of promotion to higher administrative ranks, this
section will consider the effect of party school enrollment on the rate of promotion, i.e.,
whether party school training has an effect on the timing of promotions within an
individual’s career. Within China’s bureaucracy, an individual cadre’s career
development is a long and complex process, and those who are identified early in their
careers stand a stronger chance of achieving high office.115 By testing the following
hypothesis, this section will shift the focus to consider whether party school enrollment
increases the rate of promotion for selected public managers:
Hypothesis 3 Key point training classes place select cadres on a career fast track: After enrollment in a key point party school training program, cadres are promoted more quickly to higher ranks compared to peers not yet enrolled in such programs.
115 This is compounded by such factors as the enforcement of retirement laws under Deng.
100
This hypothesis was tested using an original dataset of the educational and career
histories of Central Party School alumni. GSS data are insufficient on at least two levels.
First, survey data do not indicate the type of training program that individuals were
enrolled in. As discussed earlier, there exists a wide variety of training program types,
from general new hire orientations to specific training classes for cadres in a particular
functional hierarchy within the bureaucracy. This section will compare students who
have enrolled in similar types of training programs rather than, for example, lumping
together those in new hire training programs with those in training programs for their
specific professional specialty. One starting point is studying Young Cadre Training
classes. These are convened in all party schools, at all administrative levels, and they are
an increasingly emphasized core (zhuti) training class for all party schools.116
Second, GSS data do not contain enough cases of individuals at each
administrative rank, especially the higher ranks, to test variation in the rate at which
individuals are promoted from one rank to the next. Using CPS name lists skirts this
problem to some degree because officials invited to CPS training classes have obtained,
at a minimum, department-level (chu) rank. Furthermore, their public biographies often
list the years they were promoted from vice-section chief on up. In the case of the CPS
Young Cadre Training Classes, students at the time of training were clustered in the vice-
bureau and bureau ranks, which are upper-mid-level managers. Depending on the quality
of each official’s published biography, this provides the opportunity to code careers from
the grassroots, vice-section level up to the ministry level for this population of cadres.
This empirical strategy is not without its own problems. It would be ideal to
obtain the career histories for individuals who enrolled in CPS Young Cadre Training 116 Interview with a CPS teacher, Interview 216, May 2008.
101
Classes and locate non-CPS-trained matches for each of these students. This may be
possible by identifying cadres with similar posts to the CPS trainees for the period just
prior to their enrollment. However, personnel data of this nature is not often in the public
record. Furthermore, even after determining matching criteria, locating such information
is difficult given the need to find records of public officials in 2000, the most recent year
for which a name list is available.
Class roll call lists were published irregularly in the CPS yearbooks (which were
also published irregularly), but names for students of the Young Cadre Training Class
(zhongqingnian ganbu peixun ban) were available for assorted classes from the inaugural
1980-81 class to the 2000-01 class listed in the most recent publicly available yearbook
(2001). This analysis will draw on a dataset of career histories for the training classes
held in 2000-01, 1995-96, and 1995-97. Table 3.13 below summarizes the total number
of trainees, according to published name lists.
Table 3.13: Total trainees, by CPS Young Cadre Training Class
Year Total
students1980-81 55 1981-82 143 1982-83 191 1990-91 145 1991-92 178 1993-94 172 1994-95 193 1995-96 199* 2000-01 243
Total 1519 *Note: The 1995-1996 class total includes 160 ten-month trainees and 39 two-year trainees. This latter group was in residence at the CPS from 1995 to 1997. Source: CPS yearbooks In compiling the educational and professional backgrounds of these individuals, I relied
most heavily on official biographies published online. Information was obtained from
102
government web pages whenever possible (e.g., ministry web pages, local government
homepages, Xinhua news service).
A descriptive summary of the trainees from two Young Cadre Training Classes is
available in Appendix E. The majority of trainees were men (81.9 percent) and in their
forties at the time of the training. The educational attainment of the trainees has
increased over time, with 24.6 percent of students in 1995 reporting graduate-level
education credentials, compared to 34.6 percent in 2000.117 A high number of trainees
also listed graduate degrees from party schools, most often the CPS but in rare cases
provincial party schools. In 1995, one third of trainees obtained their graduate degrees
from party schools, in majors ranging from Marxism to economic management.
Approximately one-fifth of the biographies for students of the 2000 class listed party
school degrees (19.8 percent), though the more vague “part-time graduate degree” was a
common phrase and degrees obtained from party schools could fall under this category.
A comparison of the university majors across the 1995 and 2000 training classes
finds relatively higher shares of economics, management, law, and humanities majors.
This is consistent with analyses of Chinese political elite that find a recent move toward
diverse educational backgrounds and away from more narrow technical training (Li
2007b; Li 2008b). Almost 10 percent of the 2000 class consisted of law majors,
compared to 7 percent listing engineering majors. This compares with 7 percent and 5
percent, respectively, for the 1995 class.
117 This includes those who were part-time graduate students (zaizhi yanjiu sheng), a common educational choice among Chinese officials.
103
The next series of tables summarize the current positions of 2000 and 1995 CPS
Young Cadre Training Class alumni within party and government hierarchies. Table 3.14
lists the current administrative rank of offices occupied by alumni.
Table 3.14: Current administrative rank, CPS Young Cadre Training Class alumni 2000 1995 Combined
N % N % N %Dismissed 2 0.8 3 1.5 5 1.1Department level 0 0 1 0.5 1 0.2Vice-bureau level 25 10.3 9 4.5 34 7.7Bureau level 109 44.9 37 18.6 146 33.0Vice-minister level 75 30.9 96 48.2 171 38.7Minister level 10 4.1 35 17.6 45 10.2Vice-premier level 0 0 2 1.0 2 0.5NA 22 9.1 16 8.0 38 8.6Total 243 100.0 199 100.0 442 100.0
In terms of administrative rank, the largest share of 2000 class alumni now hold bureau
level (sitingju ji) positions (44.9 percent). Alumni of the 1995 classes are clustered one
administrative rank higher, at the vice-minister level (fubu ji), and the combined number
of 1995 alumni at the vice- and full-minister level is 131, or 65.8 percent of the total class.
Turning to the administrative level of cadres’ current positions, most are clustered
at the provincial and central levels (Table 3.15). This is unsurprising, given the division
of labor across party schools. Party schools are responsible for training cadres at one
lower administrative level, hence city leaders are trained at provincial party schools while
city party schools absorb county-level leaders.
Table 3.15: Current administrative level, CPS Young Cadre Training Class alumni Combined 2000 1995
N % N % N %Central 130 29.4 75 30.9 55 27.6Provincial or provincial capital 242 54.8 121 49.8 121 60.8City 34 7.7 26 10.7 8 4.0County 1 0.2 0 0 1 0.5NA 35 7.9 21 8.6 14 7.0Total 442 100.0 243 100.0 199 100.0
104
Table 3.16 below summarizes the present occupation of CPS alumni. Though it is
common practice for officials to hold multiple titles – for example, a city mayor is often a
vice-party secretary for that city – the table below tabulates results based on the first
occupation listed for each individual in official biographies.
Table 3.16: Current occupation category, CPS Young Cadre Training Class alumni Combined 2000 1995
N % N % N %Party secretary 38 8.6 18 7.4 20 10.1Government head 42 9.5 20 8.2 22 11.1Party administrative assistant (mishu) 14 3.2 9 3.7 5 2.5Economic development and planning 69 15.6 40 16.5 29 14.6Security 9 2.0 4 1.6 5 2.5Foreign affairs 10 2.3 8 3.3 2 1.0Media 26 5.9 12 4.9 14 7.0Personnel 16 3.6 8 3.3 8 4.0Discipline and inspection 12 2.7 2 0.8 10 5.0Propaganda and party history 20 4.5 13 5.3 7 3.5Courts 19 4.3 13 5.3 6 3.0People's Congress 24 5.4 5 2.1 19 9.5CPPCC 18 4.1 7 2.9 11 5.5Other mass organization 9 2.0 6 2.5 3 1.5United Front 4 0.9 3 1.2 1 0.5Party school 3 0.7 3 1.2 0 0.0Education 26 5.9 18 7.4 8 4.0Social/cultural 24 5.4 15 6.2 9 4.5Scientific 21 4.8 16 6.6 5 2.5Dismissed/retired 5 1.1 2 0.8 3 1.5NA 33 7.5 21 8.6 12 6.0Total 442 100.0 243 100.0 199 100.0
The most common current occupational category is economic development and planning,
which includes leadership posts in powerful bodies such as central and local
Development and Reform Commissions (fagaiwei) and State Council State-Owned
Assets Supervision and Administration Commission (guoyou zichan jiandu guanli
weiyuanhui). Provincial and city party secretaries and government heads were also
common occupations, totaling 15.6 and 21.1 percent of 2000 and 1995 alumni,
105
respectively.118 Sensitive occupations, such as those in the security, personnel and court
systems were less common. This may be due to separate training systems for cadres in
these fields.119
To test the hypothesis that invitation to and enrollment in a party school training
program increases the rate of promotion to higher administrative ranks, cadre career
histories were coded in terms of the age at which an individual assumed office at a
particular rank. Table 3.17 below summarizes the average age, by rank, for cadres of the
CPS Young Cadre Training Class of 1995, as compared to national averages.120
Table 3.17: Comparing average ages, by rank 1995 CPS Class 1995 National Average121 Two-sample t-test N Average N Average Department level 91 34.8 407,207 48.0 *** Bureau level 124 43.5 35,620 53.2 *** Ministry level 26 54.7 2,459 58.9 ***
Mean comparison test significance codes: ***=p<0.01, **=p<0.05, *=p<0.1 Data sources: Author’s dataset and COD 1999
Findings provide support for the hypothesis. No cadres attend training at CPS if
they are ranked below the department level, hence the impossibility of evaluating 118 It is also important to note that these percentages represent absolute, rather than relative, totals by occupation category. It is not possible at present to control for the size of different bureaus. For example, the relatively low number of alumni assigned to posts in the United Front system may be due to the small size of this bureaucracy relative to others. Such data limitations prevent a more precise analysis of the share of posts in different occupations that are allocated to CPS graduates. 119 The Beijing Institute for International Studies, located across from the Central Party School, for example, is a training ground for foreign-oriented intelligence officers. Legal training for judges takes place at, among other institutions, the National Judicial College and, more recently, rapidly expanding law programs at public universities. Interview with a National Judicial College teacher, Interview 35, October 2007. 120 National data for cadres in 2000 were not available for comparison. 121 To calculate these averages, which were obtained from official statistics released by the COD, I had to make three assumptions. These data gave the age profile for leading cadre ranks (department, bureau, and ministry) in terms of five-year age ranges. Weighted averages were calculated assuming that the distribution of ages within each age category was uniform. Second, to take into account the youngest and oldest age categories, which were open-ended (“35 and under” and “60 and over”), I calculated a high and low national average. The low averages assumed that individuals could have attained office in a given rank as early as 25 or as late as 60; the high averages assumed 35 and 65 for these categories, respectively. The high age of 65 was calculated based on mandatory retirement ages. T-tests using these high and low averages were also significant at the same 0.01 level. Finally, because the COD data does not give age information for deputy-ranked cadres at each administrative level, I assumed that the share of deputy- and full-ranked cadres at the department, bureau, and ministry levels in the CPS trainee class mirrored the national share.
106
whether cadre age at the time of promotion is significantly lower for non-leading ranks.
CPS trainees were, on average 13 years younger than counterparts at the time they
obtained a department-ranked position. At the bureau level, this difference was 10 years,
while CPS-trained cadres were, on average, four years younger than counterparts when
they obtained ministry-level office. Two-sample t-tests found each of these differences
significant at the 0.01 level. This further reinforces the importance of party school
training for screening cadres with the potential for long tenures of service to the party.
Moreover, the time to promotion for leading ranks above the department level is
significantly shorter after training than before: a cadre can expect to work 4.78 years at
the vice-bureau level before promotion to the bureau level, if he has not yet attended the
CPS party school training, but this time is shortened to 3.17 years after training. For
bureau level cadres, those without CPS training experience must work, on average, 8.26
years before promotion, compared to 4.52 years after training. This represents a 45.3
percent reduction in time-to-promotion for bureau level cadres. These figures, while not
a full analysis of all available cadre trainees’ career histories, provide preliminary support
for the argument that party school training places cadres on a ‘fast track’ to promotion.
6 Conclusion
Hierarchical single-party systems such as China’s are capable of projecting the
party’s authority over large populations and territories. In perpetuating and reinforcing
this party organization, party authorities must confront the informational problems
inherent to the principal-agent relationships that pervade the party hierarchy and the
party’s relationship to the bureaucracy. One problem, that of hidden information
107
regarding a given agent’s type (and subsequent ability to serve the party well), presents a
selection problem for the party. To address this, the party has invested in a set of
political organizations for selecting cadres of the party and government and promoting
those with the most promise. Party-managed schools of cadre training thus present an
institutional means for party authorities to gather information on an individual
bureaucrat’s fitness for higher office. One objective of this organizational strategy is to
identify cadres with the potential for long tenures of service to the party, hence the
emphasis on screening cadres early in their careers. During the current economic
transition, furthermore, the party requires cadres with skills that may not be demonstrated
or developed in the political campaigns of the Mao period. With this historical context in
mind, there is a political logic to the persistence of seemingly outdated Leninist
organizations such as party schools.
This chapter has sought to identify and test the party’s organizational solution to
its selection problem. It is assumed, in studies of party schools, that these organizations
serve some function(s) within the party apparatus. This chapter has tested two competing
hypotheses regarding the utility of party schools for the party: whether these schools are a
reward for cadres, i.e., a “club good” of some sort, or constitute a pipeline to higher
office. The preceding analysis has drawn upon large-N survey data and an original
dataset of high officials’ career histories to support the claim that party schools serve the
latter function. Selection for party school training improves a cadre’s chances of
promotion to higher office. The mechanisms driving this outcome are less clear, though
evidence suggests that party schools may be both a means for ambitious, capable cadres
to signal their abilities as well as a means for higher party officials to screen for those
108
cadres who do not possess requisite skills or qualities. In this sense, at least two
processes are at play: first, high ability cadres may seek to distinguish themselves by
signaling their abilities with a party school training credential, but what happens during
training also has some bearing on a cadre’s career prospects.
Over the past three decades of economic and social liberalization, the party has
maintained formal strategies for managing the managers of the party and government.
These, in turn, lend the party organizational discipline. They are also indicative of the
strong political institutionalization that maintains order in periods of transformation
(Huntington 1968). At the same time, these findings are one piece of a multi-faceted
process. The promise of promotion is an incentive for cadres to seek invitations to party
school training classes, and the screening that takes place during training explains the
survival of these organizations well into the reform period, but the content of what is
taught in the schools themselves is equally, if not more, crucial. Understanding the
content of training, how this has changed over time, and how party schools themselves
have adapted to the changes of the reform period suggests what skills and attributes are
most desired in the cadres that the party now seeks to recruit. This will be the task of the
next two chapters, which unpack the various processes that have motivated dramatic
changes within the party schools themselves.
109
Chapter 4 Fusing party and market
Contemporary China remains one of the few surviving Leninist political systems
in the world today, and its global ascendance over the past three decades presents the
puzzle of how seemingly outdated political institutions have managed to coexist with and
guide the transition to a market economy. In accordance with Leninist party principles,
directives radiate from collective leadership at the CCP’s center.122 Yet, with the party
initiation of market reforms in the late 1970s, individuals at various levels of the Chinese
Communist Party’s (CCP) organizational framework have shifted some decision making
authority from the plan to market mechanisms. In order to maintain political control over
these economic changes, CCP leadership has stressed the importance of organizational
learning. One adaptive response of the CCP has been to connect market forces directly
with internal efforts to update party organizations.
In embarking on this initiative to transform the CCP, political leaders have begun
reforming sites of elite training and education situated within the party itself. Party
schools in particular are key point organizations of cadre training and would seem an
ideal site for centralized dissemination of updated information on the party line and
122 With respect to cadre training, policies and directives emanate from three central actors within the CCP: the Central Committee, the Central Organization Department, and the Central Party School. The Central Committee issues five-year training plans as well as major documents such as the call to revitalize the party school system in 1977 and the 1983 Decision on Standardizing Party School Education (Central Committee Document Number 14, available online at www.fjdx.gov.cn/document.asp?docID=6943, accessed 3 May 2010). The Central Committee comprises approximately 200 members who endorse major party decisions made by the Politburo (20-25 members) and, within the Politburo, its Standing Committee (currently 9 members). The Central Organization Department issues documents related to cadre management, and reforms in cadre training and education may have the COD stamp on them. One example, discussed in detail in this chapter, is the COD’s 1983 Notice. Finally, the CPS may also issue statements and documents that apply to the entire party school system, such as the 1982 speech by CPS principal Wang Zhen to regularize the entire party school system.
110
unifying beliefs throughout the system. This chapter will demonstrate that central
authorities, since the early 1980s, made a deliberate choice to decentralize decision
making over certain areas of party schools’ organizational management. The outcome of
these decentralizing policies was a more fluid state of affairs in which local school
decision-makers could act independently to avail themselves of opportunities granted by
central authorities. At the same time, there are limitations to this local autonomy. Local
schools are constrained by a variety of system-wide monitoring mechanisms and the
prerogative, retained by central authorities, to impose centralizing policies at will.
Because of the balance struck between these local and central actors, a system has
emerged in which local experimentation is offset by the imperative to comply, at least
minimally, with plans emanating from central authorities.
Documentary research and findings from site visits suggest that two markets have
reshaped the incentives facing party schools. The first market, which opened up with the
liberalizing reforms of Deng Xiaoping, created the conditions for party schools to engage
in entrepreneurial, income-generating activities. The second market, which has
developed more recently, has been the creation by central party authorities of a training
market whereby other organizations are allowed to compete with party schools for
training contracts. The result of these reforms is that local party schools possess limited
autonomy over training content but face harder budget constraints than in the past. In the
tension between maintaining central control and granting local actors the independence to
act on local knowledge and conditions, the pendulum has swung in favor of the latter.
This is moderated, however, by the disciplining effect of market competition: local actors
must now weigh whether to be responsive to local, national, and international factors.
111
One feature of the system, limited administrative and fiscal decentralization, has
been a necessary condition for the emergence of market-based incentives. Together,
decentralization and market forces are the channels through which the party has induced
party organizations of cadre training to adapt to the economic changes of the reform
period. These changes are significant in that they bring market forces into the party,
beyond those units – such as state-owned enterprises – which have a more direct
relationship with the new market economy. These changes also pit party organizations in
competition with each other for market share, which plants the seeds of intra-party
competition for scarce resources. This competition, however, is constrained by a
combination of top-down policy decisions and local market conditions.
This chapter will explore the central and local processes that have brought
together party schools and these two markets, and the following chapter will examine
party schools’ various responses and resultant organizational change. Party schools have,
in the process of opening up to market forces and rising to the challenge of competition,
become more deeply embedded in local markets and intertwined with a variety of public
and private actors. Competition has reduced party school “market share” and raised the
stakes for these organizations to search more intensively for creative solutions to
problems both fiscal and relating to their core training function.
2 Bringing in the market
In leading the complex transition from plan to market, central party officials had
begun, by the mid-1980s, to introduce market incentives to party and state organizations.
The resultant “bureaucratic entrepreneurialism” entailed “government officials engag[ing]
112
in for-profit economic activities, using means and resources available only to the state,
and in pursuit of objectives broader than purely business concerns,” (Gore 1998: 6).
These incentives included the hardening of previously soft budget constraints and the
relaxation of controls over self-raised revenues by various party and state units. The
party school system was no exception to these more general, macro processes. Driving
this story of marketization was a combination of deliberate choices made at many levels
throughout the system, e.g., new policies originating from central party authorities and
initiatives pursued by local party school leaders.
Fiscal and administrative decentralization constitutes a necessary condition for
these market processes to unfold. Decentralization, or the transfer of fiscal, policy,
and/or political authority from central to local governments, has been on the increase
across countries and regions since the mid-1970s (Dethier 2000; Rodden 2004).
Processes of decentralization in China began under Mao and have encompassed both
economic and administrative dimensions throughout the reform period.123 As the party
matured, decentralization posed less of a threat to organizational unity. As Selznick
observed in his study of Russian Bolshevik tactics, “One of the advantages of a firmly
established organizational character is the possibility of increasing decentralization
without sacrificing unity of policy or stability of command,” (Selznick 1960: 65).
Economic decentralization has political implications, and scholars have debated
whether the political authority of the party and central government have eroded with the
devolution of authority and economic resources to localities (Huang 1996; Oi 1992; Oi
123 There is a considerable literature on the relationship between decentralization and economic growth. The debate has ensued between those who argue for the growth-generating effects of “market-preserving federalism” (Montinola, Qian, and Weingast 1995; Qian 2003; Qian and Weingast 1997) and those who find no such relationship (Cai and Treisman 2006; Yang 2006).
113
1999; Yang 2006).124 Attempts to recentralize tax revenues through fiscal reforms in
1994 have led to strategies by locales to preserve locally-generated revenues (Jin and Zou
2003; Shi 2000; Zhang 1999; Zhang and Zou 1996). More recent 2001 reforms in
treasury management have streamlined the flow of revenues and expenditures through the
system (Ang 2009a). Despite efforts by the center to rein in localities and especially the
provinces, China remains a remarkably decentralized system: nearly 70 per cent of
government revenues is generated by localities, a level which places China among the
most decentralized of countries tracked by the IMF (Landry 2007: Chapter 1; World
Bank 2002).
Reforms leading to administrative decentralization and their implications have
been studied most extensively with respect to personnel appointments and monitoring
(Burns 1993; Huang 1995; Landry 2007; Manion 1985). Despite this large body of work,
scholars have paid less attention to the effect of these decentralizing trends on the
incentives facing local, sub-party organizations. Within the organizational geography of
cadre training schools in China, local party schools and administration institutes possess
limited autonomy in at least three areas: financial management, organizational structure,
and curriculum matters.
Limited local autonomy in financial matters
Within the organizational framework discussed in Chapter 2, funding for party
schools is localized. Rather than allocate funds at the central level and then transfer
124 Huang and Yang, for example, make the case that devolution of authority has been by design of the central government, which retains the ability to re-centralize authority. Oi, in her analysis of the rise of local state corporatism, finds a relative strengthening of local government in the early reform period, only to be followed by waves of recentralization efforts in the 1990s and again in the 2000s.
114
financial support downward through the system, local schools from the provincial to the
township levels receive funding via a horizontal transfer from their respective local
government finance department. This arrangement has been in place since at least
1983.125 In the pre-reform period, revenues and expenditures were centralized in a mode
common to command economies of the Soviet model. Sub-national administrative levels
sent up revenues and received budgetary allocations from central authorities (Ang 2009b:
Chapter 2). One implication of this more recent local fiscal independence is a harder
budget constraint, and schools are left to flourish or flounder as their local fiscal situation
allows.
Given disparities in levels of economic development across locales, there is
variation in how much local party schools receive from their respective public finance
bureaus. A teacher at the provincial capital party school in Province A reported a
stupendous 140 million (1.4 yi) yuan transfer from city finance in 2008, whereas the
school’s counterpart in Province B, the provincial capital of an inland province, received
13 million yuan.126 This difference is even more striking when considering the higher
number of leading cadres per capita in Province B compared to Province A.127 Such
variation is the result of relative financial booms and sluggishness across locales; no
significant central-level transfers reach party schools to equalize differences. These 125 See Appendix 3: Main Points of the National Cadre Training Plan in the Notice of the Organization Department of the Central Committee, 5 October 1983, in Burns (1989a, pp. 50-83). The appendix states that “the operating expenses of cadre colleges and schools should be incorporated in the budget of the departments in charge, which are to draw their funds from the financial departments at the same level. When budgeting the operating costs and investments in capital construction for cadre colleges and schools, party committees and governments at different levels should fulfill the Central Committee’s intent of gradually increasing the investment in cadre education so that funding in this area can be well taken care of and assured,” (pp. 73-74). 126 Interview 207, May 2008, and Interview 177, April 2008, with provincial party school teachers. Unfortunately, no nationwide numbers are publicly available, making systematic comparison beyond such interview data difficult. 127 For the period from 1981 to 1998, Province B’s total number of leading cadre’s as a percentage of the province’s population was 0.028, compared to 0.020 for Province A. Source: COD (1999).
115
horizontal disparities are also reflected vertically within the system. Schools at the
lowest, township level are left to languish from lack of local resources. With the
increasing reliance of schools on local sources of funding, the fate of party schools is
often a reflection of the robustness of local economies.
Central authorities have compounded the problem of disparities in local party
school development by placing pressures on locales to improve training and generate
their own funding for such initiatives. In what amounts to an unfunded mandate, schools
have been asked to advance Deng Xiaoping’s drive for a “revolutionary, younger, more
knowledgeable, and more specialized” cadre corps. Central documents call for capital
investments to be made to realize these goals and keep party schools on par with other
institutions of higher learning (Central Committee 2004). At the same time, these
upgrades must be done with the provision of funds from local finance bureaus and each
locale’s planning bureau should allocate capital funding in support of these training
endeavors (Ibid., Article 24). Such top-down pressures do little to relieve disparities
across locales.
Organizational experiments
Second, there is evidence of decentralization in administrative and local
organizational matters. Local party schools possess the authority to test novel
organizational arrangements. In Province A, for example, the provincial capital party
school is merging two county schools and six district schools under the roof of the city
party school.128 This organizational shuffling was conceived by a city vice-party secretary
and party school vice-principal in order to “improve teacher quality” and “consolidate 128 Provincial and city party school teachers, Interviews 207 and 208, May 2008.
116
resources”.129 Interviewees reported that this decision reflects concerns with the financial
viability of and staff qualifications at these lower level schools, so city leaders decided to
take matters into their own hands and combine school programming and management
under one party committee.
There is evidence of organizational consolidation elsewhere in the country. In
Shandong province, both city and county party schools are under the guidance of the
Shandong provincial party school (Guo and Shan 2009). Lower-level schools within this
province enjoy budgetary autonomy but receive administrative guidance from the
provincial party school. Whether or not the quality or solvency of schools will improve
with these experiments remains an open issue, but local schools do have the leeway to
pursue new organizational arrangements.
Local school control over training content
Third and finally, higher-level schools cannot impose their training content on
schools at lower administrative levels. Local party schools have some freedom to design
their own training curricula and teaching materials. This is illustrated by the Central
Party School’s publication of a set of books on core socialist theory, leadership, and party
history, ostensibly to guide curricula at local schools. When asked about these materials,
a city party school teacher responded, “We don’t use them here and the Central Party
School doesn’t have the authority (quanli) to impose their publications on us.”130
Similarly, a provincial party school head declared that “We use our own materials here,”
129 Ibid., Interview 207. 130 Interview 75, December 2007.
117
and complained that centrally-produced materials became outdated too quickly to keep
up with changing circumstances in the province.131
At the same time, there is consistency across schools in the general topics taught
to cadres. The vast majority of training classes will start with some nod to the guiding
party theory of the day, though schools will address these topics to varying degrees.
Party school officials have reported in interviews that they do turn to other party schools
in the system for guidance in curricula matters. One provincial official characterized the
relationship as one where schools will “imitate each other” (huxiang mofang).132 This is
in part because party schools transmit their training syllabi up and down the party school
system as part of their advisory, exchange-based relationship. Provincial schools will
look to the CPS and other provincial party schools for ideas on updating curricula and
pedagogy, though “the party school head has leeway to structure the school’s curriculum
and teaching methods, so there is some unpredictability.”133 Some teachers have
welcomed non-party school training providers as yet another source for guidance on
updating curricula.134
This discussion demonstrates the various areas where party school leaders may
exercise some discretion, though they remain constrained by a variety of monitoring
mechanisms. There are benefits to central authorities for the decentralization that local
autonomy implies. First, this reduces the monitoring costs associated with a heavily
centralized system, since each level is responsible for monitoring the activities of schools
located one or two administrative levels down. Second, sub-national units gain the
131 Interview 66, November 2007. 132 Interview 182, April 2008, Provincial party committee policy research office director. 133 Ibid. 134 Interview 143 and 144, April 2008, Provincial party school professors.
118
flexibility to respond to local circumstances rather than being bound to possibly irrelevant
central-level plans. This local innovation generates the model sites that central
authorities may decide to replicate nationally. In this sense, it is in the interest of central
authorities to encourage localized innovation and allow localities to absorb the risks from
such endeavors. After reviewing local efforts, the center can chose to disseminate
successful local experiments and either ignore or quash unsuccessful ones.135
Official incentives to “dive into the sea” of market activity
The dependence of party schools on funding from local finance bureaus has
implications for the financial and managerial decisions made by school leaders. Along
with many other segments of China’s population, this group of actors began to engage in
new income-generating activities as planned market controls loosened in the post-Mao
period. With “reform and opening” under Deng Xiaoping, China’s economy transformed
into one of “mixed ownership structure in which public ownership is dominant among
other types of ownership,” (Wang 2006). Because of the localized nature of party school
funding, it was in both the party’s and the schools’ interests for schools to develop
alternative revenue streams. Not only would this lighten the burden on finance bureaus,
but party schools themselves could leverage in-house resources and increase revenues
that would keep members of the danwei satisfied.
135 Official documents suggest that this central permissiveness and limited local autonomy is by design. As early as 1989, a published survey of party school education discussed the rationale for variation in content. “Because of differences between provinces, municipalities and autonomous regions, the party school education at different localities is different. Party school education in autonomous regions provides relevant training related to ethnic minority groups and strengthens the solidarity among different ethnic groups. In municipalities, party school training focuses more on urban development. In provinces, regional differences are reflected in party school education,” (Liu 1989: 114-118).
119
While national and local party school budget information is not publicly available,
some financial information comes to light during interviews at individual schools.
Because party schools are considered fully-funded public service units (quan’e bokuan
shiye danwei), they are in theory transferred enough funds from local finance bureaus to
cover operating expenses.136 In reality, transfers of public funds leave virtually no room
for organizational slack. This, in turn encourages revenue-seeking activity by school
leaders. For example, a local finance department may deliberately set the allocation level
to cover only a set percentage of the party school’s operating expenses and leave the
remainder for the school to raise through market activity. In Province B, for example,
one cadre stated that the transfer usually covers approximately 60 percent of operating
expenses.137
There also appears to be some regional variation in whether budgetary allocations
are trending up or down over time. In the relatively impoverished province of Yunnan,
budget allocations for local county schools have even been reduced in recent years,
leaving schools severely constrained (Pieke 2009a: 132-40). Provincial-level training
institutions in both Province A and B, did not report the sort of declines experienced
further west.138 In the provincial capital of Province B, party school officials reported
increases in transfers over time, though the school was still experiencing ever larger
budgetary shortfalls. This party school received a 13 million yuan transfer from the local
finance bureau in 2007, which was insufficient for an operating budget of approximately
25 million yuan. This shortfall of 12 million yuan, or 48 percent of the operating budget,
136 Ang 2008 offers a lucid description of the distinction between public service units (or extrabureaucracies) and party or government bureaus. 137 Interview with a city party school professor, Interview 177, April 2008. 138 Interviews, provincial socialism institute principal, Interview 95, December 2007; provincial capital party school teacher, Interview 177, April 2008.
120
would have to be covered through auxiliary (jihua wai, or “outside-of-the-plan”) training
classes, degree programs, and other revenue schemes.
Table 4.1 below offers a breakdown of general expense categories and expected
income sources for local party schools:
Table 4.1: Summary of local party school expense categories and income sources Expense Income sources General operating expenses Local finance bureau
School-generated revenue Capital construction Local finance bureau
Local planning/development and reform commission
Training expenses (e.g., teaching materials, site visits, studying in other institutions) If: Training is outside-the-plan If: Training is in local annual training plan
Trainees’ work units Local finance bureau and trainees’ work units
Salaries Local finance bureau Employee benefits (e.g. mobile phone service allowance, meal allowance)
School-generated revenue
Employee-related benefits constitute another party school expense category.
While local finance departments will transfer to party schools a certain amount for
employee salaries, there are other employee-related expenses that the local finance
department will not cover. These include employee benefits and salary supplements
(butie). Across party and government offices, there is wide variation in the level of
benefits enjoyed by employees. This is because such benefits and salary supplements are
linked to the fund-raising abilities of a particular office.139 In Province B’s provincial
capital, for example, each party school employee receives a different set of supplements
based on rank and whether that individual has civil servant status. On average, this
shortfall for the provincial party school is between 1000-1500 yuan per employee per
month.140 Supplements can include direct transfers to employees to cover expenses
139 A detailed discussion of the components of cadre wages and benefits can be found in Whiting (2006). 140 Interview 177, April 2008. This cadre went on to say that leaders can expect supplements upwards of 3,000 yuan a month.
121
associated with mobile phones, transportation, and meals with guests, among other
categories.141 In one wealthy county school, the school vice-principal receives an 1800
yuan monthly supplement for car maintenance alone: though he provides his own car, the
insurance, road tolls, and gas can come out of this supplement.142
In summary, central authorities’ conscious decision to decentralize the
management of party schools created a new context within which these organizations
functioned. Fiscal and organizational decentralization generated incentives for party
school leaders to pursue new ventures for pecuniary gain and organizational efficiency.
In budgetary matters, schools could utilize self-raised funds to supplement budgetary
transfers from local finance departments. This organization-level discretion extended to
other areas of operation. In the reform period, party schools leaders could also initiate
organizational changes and reforms in school programming and curricula. Along with
these new opportunities, however, came significant challenges. While any new ventures
or initiatives pursued by party schools carried attendant risks, these risks were
compounded by yet another development in the organizational environment within which
these schools operated. Beginning in the early Deng Xiaoping period, party schools were
exposed to yet another aspect of the market: competition.
3 Motivations to marketize training
Party schools once enjoyed a near-monopoly on the training of party and
government leaders.143 Over the course of diversifying activities in a new market context,
141 Interview 162, April 2008, two provincial party school teachers. 142 Interview 191, May 2008, county party school vice-principal and teacher. 143 Ministries and state-owned enterprises may also have party schools or cadre training academies, but these were traditionally for lower-ranked cadres. Furthermore, party schools in other ministry systems
122
the leaders of party schools began to face pressure on another front: competition from
other organizations and schools for the privilege to train government and party officials.
Figure 4.1below depicts the variety of central-level cadre training organizations that are
now in existence, many of which were established over the past three decades.
Figure 4.1: Central-level organizations of cadre training
Central Committee of the CCP
Central Organization Department
Central Party School
Cadre executive leadership academies:
Pudong, Jinggangshan,
Yan’an
United FrontChina Socialism
Institute
Four branch schools and cooperative
profit-sharing programs with
local party schools
Government ministry
training centers
State Council
National School of
Administration
UnitedFront
Department
Beginning in the 1980s and continuing through the present, new and existing institutions
were allowed, often by central decree, to plan and implement training programs for
cadres. These institutions included universities, administration schools, and new training
centers opened by other party units. In effect, these new organizational players were
given official sanction to claim training market share from party schools. The newest
additions include the National School of Administration and Cadre Executive Leadership
Academies.
(xitong) and SOEs are considered subsidiary schools (fenxiao) of the party school system (see, for example, the Beijing party school webpage for a listing of its subsidiary schools in ministries and SOEs and Beijing district party schools: http://www.bac.gov.cn/web/swdx/about/Link.aspx?NodeID=1&Link=区县站点导航&Tag=3#, accessed 3 July 2009). The training of leading (chu ji yi shang) cadres in particular was a traditional responsibility of the party school system. For an overview of the history of these schools by province, see General View of the Nation’s Party Schools (1996).
123
Motivations ranged from the desire to expand the system’s capacity to carry out
large-scale and regularized cadre training to a more specific dissatisfaction with the
quality of party schools. Capacity building was a relatively straightforward goal since
improvements in cadre training would fulfill Deng’s declaration to transform China’s
cadres. While there existed a dual-track system for sorting individuals into state
administrative versus party management careers (Walder, Li, and Treiman 2000), Deng’s
generation of leaders nonetheless perceived a shortage of managerial talent. On a
practical level, given insufficient capacity within the existing party school system, the
logical solution was expansion of cadre training and channeling promising bureaucrats to
new training academies (Liu 2001: 97).
Central-level policies calling for the expansion, and accompanying
institutionalization, of cadre training can be traced to the early years of Deng’s leadership.
Along with the 1983 Central Committee policy declaration to “standardize party school
education,” there was an early mandate to have all leading cadres and reserve cadres take
remedial theory and vocational training courses over the period 1983 to 1990 (Central
Organization Department 1983: 68-70). In the 2001-2005 National Cadre Education and
Training Plan, the Central Party School and National School of Administration were to
train 400 provincial-level cadres each year, totaling 2000 for the five year period. This
was combined with the stipulation that all cadres at that level should have 3 months of
training within a five year period. Since central cadres numbered over 70,000 by 2002,
branch campuses (fenxiao) would have to make up the difference (Yang 2002: 286-295).
This was logistically challenging and provided motivation to expand the system.
Accordingly, the training plan called for “speeding up the construction of cadre education
124
training bases and reforming each administrative level’s party school education as the
main channel for rotating party and government leading cadres through training.”144 As
recently as 2006, a minimum three-month training every five years was required for all
cadres. One provincial party school administrator complained how these requirements
strained school resources: “There are just too many grassroots cadres, it takes a lot of
time and resources to train all of them, and we [party schools] still cannot train them
all.”145 The numbers are revealing. Shambaugh observes that the CPS may train up to
5,000 cadres a year at present, so circulating 70,000 central-level cadres through this one
institution would take well beyond the 5 year cycle laid down in official training plans
(Shambaugh 2008a: 143).
While capacity concerns are politically palatable, there were also indications of
dissatisfaction with the quality of party school education at the central and local levels.
The end result has been a push to diversify the kinds of organizations engaged in cadre
training. During the early reform period, worries over party school deficiencies were a
manifestation of central-level displeasure with cadre management in general. At the 13th
Party Congress in 1987, Premier Zhao Ziyang reported,
The power of cadre management is over-concentrated and the people who handle personnel affairs lack professional knowledge; the methods are outdated and simplistic, which hinders the intellectual growth of talented people; the management system is flawed and there are no laws governing the way personnel are used.146
Concerns with the quality of party school training were addressed more directly in the
2001-5 National Cadre Education Training Plan, which sought organizational diversity as
the means to “raise training quality.” The plan stipulated that party schools “should
144 Section 4, 2001-2005 National Cadre Education and Training Plan 145 Interview 66, November 2007. 146 Quoted in Liu (2001), p. 67.
125
actively make use of administrative institutes, universities, research institutes and other
training resources” to carry out “multi-faceted, multi-level education and training.”
Likewise, party documents for internal circulation have recommended that “we should
expand the content of education for party cadres and improve education methods” due to
“the party’s changed mission and the request of cadres,” (CPS 2004: 142-145). In
official training plans, then, cadre training would and should include multiple actors.
As indicated in this 2004 CPS document, another source of the push for reforms
in the structure of cadre training came from cadres themselves. In evaluations of party
school-organized training programs, participating cadres complained about various
aspects of the experience, from the perceived old-fashioned thinking of the teachers to the
lack of relevance of training content for cadre work on the ground.147 Beginning in the
mid-1990s, the provincial-level party school in Province A began to have open
discussions at the end of training classes to collect student feedback. When brought
together for discussion, cadre-students voiced their grievances. They pointed to the
rigidity (si ban) of the training system and the conservatism of the teachers (laoshi fang
bu kai).148 One cadre complained to me, “Party school teachers are terribly boring and
didactic, they have outdated knowledge and are uninspiring. They only teach old theories
and are not creative. They must be the leftover teachers who couldn’t find work at real
schools!”149 This critique was not limited to students. A city organization department
official observed, “Some party schools, especially the grassroots ones, have low quality
teachers. These teachers aren’t interested in improving training, they are just interested
147 Interviews with a county-level civil servant (Interview 104, January 2008), county-level party school teacher (Interview 197, May 2008), provincial socialism school principal (Interview 95, December 2007). 148 Interview 208, Zhejiang party school teacher, May 2008. 149 Interview 104, city government cadre, January 2008.
126
in their own job security and benefits.”150 In Province B, a similar realization was
dawning: the party school system was becoming ossified in the reform era and
increasingly disconnected from problems that cadres were facing.151
But why a training market? Introducing market-based competition, rather than an
alternative arrangement such as a clear division of labor across training organizations,
suggests a search for the institutional incentives and framework that would promote
continuous organizational adaptation. There is the sense that central leaders considered
markets more capable of shifting resources quickly to match new developments. At the
founding of new cadre academies in 2005, for example, then-president Hu Jintao declared,
“The priorities of the training centers are to enhance the ruling awareness of the Party,
improve its art of leadership and governance, and strengthen its governing capacity. With
this foundation, the centers will continue to revise the training curricula, modernize the
training methodology, better allocate the training resources and improve the training
team, thereby increasing the overall training quality.”152 The invisible hand of the market
would provide strong incentives for organizational adjustment and, because of the
decentralized nature of market competition, respond to the unevenness of local conditions
in a way that top-down planning, with its informational disadvantages, could not.
4 Policies, processes, and the emergence of a training market
In the early 1980s, the Central Organization Department was pressing for the
construction of new cadre colleges and the integration and expansion of cadre training
150 Interview 212, May 2008. 151 Interviews 143 and 144, provincial party school teachers, April 2008. 152 Italics added for emphasis. CELAP webpage, “Congratulatory Letter by President Hu Jintao,” http://61.129.65.35/renda/node3284/node3286/userobject1ai39720.html, accessed 7 December 2008.
127
programs in universities and other educational institutions.153 Official declarations
spelled out these designs. “Starting in 1985, the enrollment of specialized cadre training
classes should be expanded step by step to reach 10 percent of the total annual enrollment
of institutions of higher learning and secondary vocational schools,” (Central
Organization Department 1983: 74). By 1985, keypoint colleges and universities were to
create advanced studies and training programs for cadres.154 There was also a call to
increase construction of cadre colleges, whose main task was vocational training.155 This
was part of the larger push to raise the educational credentials of cadres and follow
through on Deng’s mandate for a “better educated and more specialized” cadre class.
The universe of training institutions in the early reform period thus included party
schools and a variety of new actors in the push to modernize the cadre corps. However,
universities and cadre colleges were to boost technical skills, not stand in as authorities of
party doctrine.156 Party schools were still the pipeline for promotion to higher ranks
within the party.157 For example, only party schools had long-term training classes for
those bureaucrats tapped for top leadership positions. Party schools also dealt with this
153 This was not an entirely foreign proposition. Since the early 1950s, delegations of cadres were sent to the Soviet Union, Eastern Bloc, and small numbers of cadres were also selected for university education. See Liu (2001), p. 98. 154 These were to be funded by local finance departments. “Starting from 1984, financial departments at various levels should make unified appropriations to departments of education of funds for specialized cadre training courses and cadre training classes that are offered by institutions of higher learning and secondary vocational schools. These institutions and schools should no longer charge tuition for the sponsors of the trainees,” (Appendix 3: Main Points of the National Cadre Training Plan, Section 3: Expand the quota of cadre students enrolled by institutions of higher learning and secondary vocational schools, pp. 74-75). 155 Literacy and vocational courses were to constitute 70-80 percent of total credit hours, with Marxist theory filling the rest. Ibid., p. 73. 156 Pieke (2009) notes how the three most prestigious universities in China, all based in Beijing, have profited from the opportunity to train cadres in management and non-ideological coursework. In terms of training volume, in 2006 the School of Continuing Education at Tsinghua taught approximately 6000 cadres, projected to increase to 7500 in 2007 (p. 127). 157 All vice-ministers, for example, must attend the Central Party School prior to promotion to the minister rank. Interview 2, Beijing university vice-party secretary, September 2006.
128
challenge by drawing on party authority within universities and other educational
institutions. University party committees wielded considerable influence. At universities
as diverse as Xiamen University (in Guangdong province) and People’s University (in
Beijing), party secretaries could shape the direction of management institutes that were
formally responsible for secular public administration training (Williams 1993). These
efforts, however, could not prevent the entry of universities as competitors to party
schools in an emerging national training market.
Building government capacity: Administration institutes
By the late 1980s, a new set of institutions emerged to challenge the dominance of
party schools over cadre training. Beginning in 1987, the movement to build a system of
public administration academies gained momentum with a resolution passed at the 13th
National Party Congress.158 Shortly after, construction of the Beijing campus of the
National School of Administration began, and, due to delays wrought by the Tiananmen
protests, in 1994 the organization officially opened its doors (Huang 1993a; Li 1993).159
Organizationally, the NSA falls under the guidance of the State Council, State Education
Commission, and the Ministry of Personnel (Huang 1993b). Since then, a web of
administration schools has spread to the provinces and sub-provincial levels.
Officially, administrative institutes train government bureaucrats, while party
schools are responsible for the training of cadres in party units. In reality, however, the
boundaries between the two are blurred. Party schools have dealt effectively with the
158 “NSA: Cradle of China’s Public Servants,” People’s Daily (English service), 10 July 2000, http://english.people.com.cn/english/200007/10/eng20000710_45138.html, accessed 21 February 2006. 159 During the construction phase, as early as November 1988, the NSA was involved in trainings for secretaries and directors of ministry departments and provinces. Many of these took place on the CPS campus. See Huang 1993: 100.
129
entry of this competition through cooptation. From the provincial to county levels, party
schools have successfully merged with administration institutes. Table 4.2 below lists
whether provincial-level party schools and administration institutes are located on the
same campus.
Table 4.2: Provincial-level party schools and administration institutes
Place Merged? Year party school
established Year administration institute established
Anhui No 1951 1991 Beijing Yes 1950 1993 Chongqing Yes 1997 1997 Fujian No 1995 Gansu No 1938 1990 Guangdong Yes 1950 2001 Guangxi Yes 1950 1994 Guizhou Yes 1950 1997 Hainan Yes 1988 1994 Hebei No 1950 1987 Heilongjiang Yes 1948 Henan Yes 1949 1996 Hubei Yes 1954 1993 Hunan (in 2003) 1951 1953 Jiangsu Yes 1953 1992 Jiangxi (in 2001) 1950 1981 Jilin (in 2006) 1948 Liaoning Yes 1946 1990 Neimenggu Yes 1948 1995 Ningxia Yes 1958 1996 Qinghai Yes 1956 1993 Shaanxi No 1942 Shandong No 1938 1992 Shanghai (in 1989) 1949 1986 Shanxi Yes 1949 1995 Sichuan (in 2001) 1952 1997 Tianjin Yes 1949 2001 Xinjiang Yes 1950 2000 Xizang Yes 1961 1991 Yunnan Yes 1950 Zhejiang Yes 1949 1988
Source: Author’s dataset.
Most administration schools reside on the same campus as party schools, and it is
not uncommon for single school administrators to hold parallel, dual titles across both
organizations. The rationale given for this system of “two signs, one set of personnel”
130
(liang kuai paizi, yi tao renyuan) is to benefit from overlap in organizational goals and
resources, though the party school holds the higher status and authority.160 Combining
schools on one campus has not been exercised uniformly throughout the system; six out
of 31 provincial-level units have separate party schools and administration institutes
(Table 4.2).
Intra-party competition: organization departments join the training market
Despite their official mandates to engage in cadre training, universities and
administration institutes both lacked the stamp of party authority that party schools
enjoyed. A more recent entrant to the training market presents a challenge on this front.
The newest additions to the universe of cadre training schools are a variety of
organization department-sponsored training schools. At the national level is a trio of
cadre leadership academies. These academies are funded and managed by the Central
Organization Department, not the State Council or the Central Committee of the CCP,
and they possess near-equal status with the Central Party School in Beijing. They
represent lavish investments in elite cadre training, with the construction costs of the
Pudong campus alone estimated at US$100 million.161
Early efforts to expand the country’s organizational capacity for cadre training
differ from these most recent initiatives. The decision to pull universities and
administration institutes into the domain of cadre training was aimed more directly at
160 In a country where no official matters are deliberate, a basic gauge of the importance of administration institutes versus party schools is evident in resource allocation: when I interviewed administration institution and party school counterparts in their respective offices, the party school office was inevitably the larger and more well-appointed one. On the importance of order and rank in Chinese politics, see Macfarquhar 1971. 161 The figure given during an interview with a Pudong professor was 900 million yuan, or approximately US$112.5 million. Interview 187, April 2008.
131
developing the professional skills of cadres. Universities, as well as cadre colleges, could
enhance a cadre’s educational credentials, but it was understood that he needed to pass
through a party school training program on the way up the political career ladder. Later
efforts to create a new system of cadre training academies more explicitly challenged the
dominant role that party schools had enjoyed in cadre training and established an
alternative training model within the party. Central party authorities were, in effect,
using competition to discipline party schools, which represents a significant shift away
from the top-down control more conventionally employed in Leninist political systems.
The 16th party congress in 2002 saw the introduction of this intra-party
competition. At the congress, CCP leaders approved the establishment of three central-
level cadre executive leadership academies (CELA) in Pudong (Shanghai), Yan’an
(Shaanxi province), and Jinggangshan (Jiangxi province) to train the population of elite
cadres once channeled exclusively to the Central Party School.162 These academies
would remain distinct and separate from the party school system.163 Their creation, in the
words of one CELA administrator “represent[s] a new model of cadre training ... In
traditional [party school] training, there are too many traditional materials, too many
traditional teaching methods … Now the party schools have to produce results, otherwise
162 See the 18 Oct 2007 report filed by Ouyang Song, deputy head of the Central Organization Department, “Sound progress of the new and great project of party building since the 16th National Party Congress,” available at http://english.peopledaily.com.cn/90001/90776/90785/6285280.html, accessed 29 June 2009. The locations chosen are representative of the party’s bases of legitimacy: in a revolutionary past and economic future. Party school officials have also pointed out how this move by the COD was also related to a shift in the emphasis of training content. Market economics and law were emphasized as training topics from the time of the 16th party congress to the present (Interview 66, November 2007, provincial party school vice-principal; Interview 75, December 2007, city party school professor). 163 Similar to the funding situation for party schools, CELA, which are considered central-level training organizations, receive funding for inside-the-plan training classes from the Central Organization Department, via the central finance ministry, while individual work units or companies must bear the cost of training if they are commissioning auxiliary, or outside-the-plan training class. Interview 184, April 2008, CELA director.
132
they will be eliminated (taotai).”164 Making the competition explicit in policy documents,
recent central-level regulations have called for “a system of orderly competition”
(jingzheng youxu de jigou tixi) and again acknowledged that other institutions such as
universities and research institutes may “take on the task of cadre education and
training.”165
During the period between the 16th and 17th party congresses, local organization
departments also began establishing their own cadre training schools.166 In Province A,
the provincial capital’s organization department decided to boost grassroots training by
opening a district-level cadre training base (jiceng ganbu peixun jidi).167 In Province B,
the provincial organization department required party and government organs to send
cadres to the organization department-managed revolutionary education academy
(geming jiaoyu xueyuan).168 These new training bases divert cadre trainees, and the
training funds that accompany them, from party schools to other party organs.
Beyond presenting an updated model of cadre training, an additional motivation
for organization departments to create new training bases rests in concerns about a loss of
respect for party school training by cadres. Creating organization department-managed
training centers is, in the words of this organization department cadre,
to turn training into something serious again, to give training the stamp of the organization department’s authority. … Party schools are like nannies (baomu) while the organization department is the master (zhuren). Party schools don’t really have the authority to discipline cadres, so cadres don’t take party school training seriously. Cadres are like kids toward their nanny [in their relationship
164 CELA director, Interview 184, April 2008. 165 Chapter 5, Art. 27, Cadre education and training provisional work regulations, 2008. 166 Interview 211, May 2008, provincial party school teacher. 167 The organization department has been organizing training classes for grassroots cadres since 2002 but decided in 2006 to open a formal training center. Interview 212, May 2008, city organization department cadre. 168 The plans to build the academy were announced in 2004. Total construction costs are reported to be 65 million yuan.
133
with party schools]. This is unlike the cadres who come to the organization department’s training center, they are very well-behaved (guaiguai).169
Opening new, competing training centers managed directly by organization departments
is one means to remedy these perceived shortfalls.
New experiments, new competition
Organization departments have also experimented with other training schemes in
which party schools are brought in direct competition with organization department-
approved training providers. The emphasis in these new programs has been on flexibility
in training content and cadre choice. The Shenzhen city party school has been caught up
in one of the more radical training reforms in the country and serves as a model for other
schools. Since October 2003, city civil servants and party cadres have participated in a
“self selection” (zixuan) training program in which cadres may select training sessions
from a course catalog. By national law, bureaucrats must fulfill five days of training each
year, and in Shenzhen they may meet this requirement by selecting their own training
courses. Three training days consist of mandatory theory and party building classes and
the remaining two training days consist of elective courses. Basic theory classes, for
example, are still mandatory.170
These self-selected training classes are offered by local educational providers, all
approved by the city organization and personnel departments and a committee
comprising representatives from the city’s party committee, people’s congress, and
people’s political consultative conference (CPPCC). The six selected training providers
169 Interview 212, city-level organization department cadre, May 2008. 170 Shenzhen’s “self selection” training guidelines are available at http://www.szps.gov.cn/zxpx/detail.aspx?NewsID=556, accessed 9 August 2009.
134
include: the Shenzhen party school, the adult education school of Shenzhen University,
the training centers of the Shenzhen media group, Shenzhen’s international human
resources training center, the Shenzhen management learning institute, and a national
economic and information technology training center.171 Whereas the party school will
focus on classes such as “Research and questions in ‘Three Represents’ important
thought” and “Research on leadership capacity,” the city university will feature courses
from its MPA and MBA programs.172 The party school thus takes part in creating
elective classes (zixuan ban) that local bureaucrats may opt to participate in.
This training scheme has the advantage of flexibility in that class offerings may
shift in response to student demand and busy cadres may choose classes which fit their
schedules and interests. Furthermore, training organized in this fashion is less costly,
since students do not live on campus as they do in more traditional party school training
classes. While this training arrangement exists alongside other courses organized
exclusively by party schools, this “self-selection” training model has pitted party schools
in more direct competition with other training providers than existed in the past.
Delegations from cities in Guangdong province have visited Shenzhen to inspect this
training model, and it has received attention from the national-level management training
academy in Dalian, located in northeast Liaoning province.173 Whether this model will
spread to other locales, however, remains unclear.
171 “Shenzhen cadre ‘self selection’ training will start, six training institutions will take up positions,” Shenzhen Economic Daily, 8 Oct 2003, available online at http://www.southcn.com/news/dishi/shenzhen/shizheng/200310080733.htm, accessed 2009 August 8. 172 Ibid. 173 “Report on Guangdong study tours,” http://www.hnredstar.gov.cn/yueyang123/gbjy/gj_gjdt/t20060321_52030.htm, and “Several insights from Shanghai, Shenzhen, and Ningbo party school inspections” http://www.swdx.dl.gov.cn/yxlt/Document/5063/5063.html, accessed 8 August 2009.
135
Competition goes global
Beyond this domestic competition and experimentation, party schools must now
also contend with the entry of international training providers. As early as 2003, local
party committees, through their organization departments, were issuing public calls for
training proposals. The official motivation for initiating this bidding process was to
implement more effectively the goals stated in national cadre training regulations, and the
result has been explicit competition between training providers for local contracts.174 In
addition to bids from Chinese universities, party schools, and other training centers, these
calls attracted applications from universities located abroad. Table 4.3 below lists
various local party committees’ calls for training bids and the eventual recipients of
training contracts. A training class requested by the Wuhan organization department in
2008, for example, was awarded to UCLA for a training class on international trade.
The awarding of training contracts to such a variety of organizations reflects the
competitive pressures that party schools must respond to in order to retain a place in an
expanding training market. While there is some specialization, for example the
organization of training classes on “e-governance” by a national training center on
information technology, there does not appear to be a clear division of labor in the types
of organizations awarded various training contracts. Even general training programs that
party schools could once take for granted as part of their core responsibilities, such as
174 Official declarations of the motivations underlying the government procurement system for cadre training programs is available via www.chinabidding.org, the (only) website authorized by the State Development Planning Commission to post bidding notices. See http://www.chinabidding.com/zbjg-detailTwo-2534138.html and http://www.chinabidding.com/zbjg-detailTwo-2534139.html, accessed October 2009. The Guangdong provincial organization department’s webpage also refers to how a bidding process, which includes party schools, administration schools, cadre academies, universities, and research institutes will “improve training performance”. See http://www.gdzz.cn/javaoa/article/articleview_simple.jsp?act=view&articleID=999884555479ab276880f95f70f56215&catalogID=8cf83a421ad83bf7f8dfcf1c0374f696&path=%E5%B9%B2%E9%83%A8%E5%9F%B9%E8%AE%AD/%E5%9F%B9%E8%AE%AD%E5%8A%A8%E6%80%81, accessed October 2009.
136
training classes for young cadres, have been subject to open bidding. This was the case
in Heilongjiang province in 2009, when five separate organizations, none of which were
the provincial party school, received contracts for young cadre training classes.
Table 4.3: A selection of training bids, by locale and awardee Year Locale Training contract awarded to Training topic(s) 2003 Wuhan, Hubei Tsinghua University Public Administration 2004 Mianyang, Sichuan Southwest University of
Science and Technology Issues for female cadres
2006 Beijing >20 Chinese and overseas universities (such as Peking University, U Toronto, and Georgetown University)
Leadership and management Urban planning Law Economics
2006 Shenyang, Liaoning Korean university (not specified)
Rural development
2006 Sichuan National training center on information technology
E-governance
2008 Wuhan, Hubei UCLA International trade 2009 Beijing Beijing Party School Chinese culture 2009 Zunyi, Guangzhou College of Minorities Leadership for minority
cadres 2009 Hangzhou, Zhejiang Peking University Development, globalization,
economics 2009 Harbin, Heilongjiang People’s University School of
Education and Training Leadership
2009 Heilongjiang 5 institutions: Harbin Institute of Technology Heilongjiang University Peking University Marxism Dept Jilin University School of Administration Tianjin University
Young cadre training class
Source: Author’s dataset.
5 Party school market share
An examination of the allocation of training contracts in one locale, City Z in
coastal Province A, illustrates the impact of competition on party school market share
(Figure 4.2). While the city’s party school still retains a majority share of local training
contracts, it has had to cede ground to the various competitors that have emerged during
the reform period.
137
Figure 4.2: Share of total training, by school type
City party school61%
Central Party School
2%
Chinese university16%
University abroad5%
Other training center16%
Notes: Training share calculated as a percentage of total trainees planned for 2008. “Other training center” includes the following: Shenzhen Socialism Institute, the City Socialism Institute, Shanghai Socialism Institute, the City’s police academy, the cadre school at the provincial construction ministry, and the Central Discipline and Inspection Training Center.
Data on training class allocation were obtained from the city’s annual training plan. As is
the convention in locales throughout China, the local party committee, in collaboration
with personnel-related party and government officials, established a training plan for
2008. This plan detailed the total number of training classes that would be funded by the
city’s finance department for the year, including the substantive focus of each training
class. In 2008, the city’s party committee approved funding for 40 training classes,
totaling 3,785 local cadres. Classes ranged from large courses on the recently convened
Seventeenth Party Congress to more general “advanced” and young cadre training
courses. The party school was allocated 27 of these 40 training classes, though four of
these classes were to be organized jointly with other training providers.175 In terms of
trainees, these classes totaled an estimated 2,345 students, or 61 percent of the total
planned for that year. Other training sites included provincial universities, universities in 175 Partners included the Central Party School (1 class), a provincial university (2 classes), and the city personnel training and testing center (1 class).
138
Hong Kong and Singapore, and national universities such as Fudan (in Shanghai) and
Fujian Universities. Central training centers such as the Central Party School and Central
Discipline and Inspection Training Center were also granted training contracts.
There did not appear to be discernible patterns in which type of training
organization received a given training contract; the city party school was assigned both
general and more specialized classes (e.g., a public security section chief training class, a
grassroots letters and complaints training class). It might have seemed logical to assign
more technical classes, such as those regarding enterprise management, to universities
with cutting edge business schools, but the training plan does not seem to reflect such
logic. Fudan and Tsinghua Universities were assigned contracts for “enterprise manager
advanced research classes,” but the city party school was granted contracts for training
classes on agricultural technologies and industrial economic management. A full list of
classes and responsible training organizations is available in Appendix C. The pie chart
below summarizes the 2008 allocation of training classes in City Z.
Data on the party school’s share of training during the pre-reform period is not available.
However, universities in China and abroad claimed one-fifth of training contracts in 2008,
a share that would not have been possible prior to the creation of a training market in
China.
The effects of this competition are also observable at the apex of the party school
system, the Central Party School in Beijing. CPS publications offer a glimpse of
longitudinal trends in enrollment levels, information not as easily obtained at the
provincial and lower levels. Figure 4.3 below captures enrollment levels, by year, based
139
on name lists published in CPS yearbooks. The yearbooks, which span 1985 to 2001,
were not available each year but include class lists for assorted years from 1977 to 2000.
Figure 4.3: Central Party School trainee volume, 1977-2000
050
0010
000
1500
020
000
2500
0Tr
aine
e-m
onth
s
1975
1980
1985
1990
1995
2000
Year
Source: CPS Yearbooks, 1985-2001 The total number of cadres trained has been uneven over time, but trends are consistent
with the advent of competition during the reform period. Enrollment rose during the
period immediately after the tumult of the Cultural Revolution, which ended in 1976, as
party authorities sought to rebuild cadre ranks. This expansion was followed by a drop in
enrollment, beginning in the early 1980s, at the time that universities and other training
centers were encouraged to engage in the national project of educating and training
cadres. Since then, there has been a slow downward trend in CPS training capacity.
While annual figures are not available, it is possible to make some comparisons
regarding the training capacity at the new COD-managed academies and the CPS. The
Jinggangshan CELA reported training 9,993 cadres from the time of its opening in 2004
until September 2007, while the Yan’an CELA reported over training over 7,200 cadres
140
from its opening in 2005 until 2007.176 This translates to an annual average of nearly
2,500 cadre trainees at each school, which exceeds the CPS annual average of 1,509
trainees for the period 1990 to 2000.
Market gains and losses
Over the past two and a half decades, this training market has generated winners
and losers. The CPS and other elite party schools, such as those at the provincial and city
levels, have relinquished some market share to new entrants but remained afloat – and, in
some cases, thrived. This prosperity was evident in the county, city, and provincial-level
schools visited in Provinces A and B, which were all in the process of or had recently
completed significant improvements to campus facilities. These renovations were all
supported in part by revenues generated through local party school ventures.
There is some evidence, however, that a subset of party schools proven vulnerable
to competitive pressures. Given the localization of finance, schools at the lowest
administrative levels often languish from lack of resources.177 In field interviews, party
school leaders acknowledged the potential risk of a grassroots crisis in the party school
system.178 “Township party schools are basically local meeting halls with some signs [for
the schools] hung outside (gua paizi) so they can use the facilities for training village
leaders. They are empty institutions.”179 Higher level party schools are aware of a
potential crisis in grassroots training programs, but as a general rule, schools at different 176 On the Jinggangshan total, see “China’s Jinggangshan Cadre Academy trains nearly 10,000,” People’s Daily, 27 September 2007. On the Yan’an total, see “China’s Yan’an Cadre Academy reignites fire in a sacred revolutionary place,” China’s Personnel, 7 September 2007. 177 In his research of the Yunnan party school system from provincial to township level schools, Pieke finds relative paralysis of funding for the schools in counties and townships. See Pieke 2006, 2008 and 2009. 178 In 2008, for example, the Hubei provincial party school commissioned an investigation into the state of township party schools, but the report has not yet been made public. 179 Interview 180, former township vice-party secretary, April 2008.
141
levels remain administratively and financially separate. The atrophying of these party
schools demonstrates the disciplining nature of market-based competition and how the
struggle for training contracts has yielded uneven results. Yet, it is premature to draw
conclusions about the fate of higher level schools, as it is not entirely clear whether the
party school system will emerge victorious vis-à-vis new competitors.
In short, party schools throughout China have had to respond to new market-
based pressures: budgetary constraints and competition from multiple training providers.
This process of marketization began in the mid-1980s, when the COD welcomed
universities and other organizations to train cadres (Figure 3 below). Over time, there has
been an accumulation in the number and variety of organizations participating in training
activities once dominated by party schools. In the last decade, organization departments
throughout the country have opened their own training centers. One important result has
been the diversion of training funds to these various training providers. Some party
schools, however, have proven nimble and developed various coping strategies, the
subject of the next chapter.
Figure 4.4: Domestic training market entrants, by year and type of organization
Administration Institutes
Cadre Leadership Academies
At 13th Party Congress, nationwide network of
schools created to train government civil servants
At the 16th Party Congress, three keypoint academies
founded to address the revolutionary spirit and global outlook of cadres
ChineseUniversities
By central decree,allowed to dedicate up to
10 percent of annual enrollment to cadre
training
1985 1987 2002
142
6 Conclusion
This chapter has demonstrated that market forces directly shape the conditions
and decisions facing party schools. School administrators must now adjust to a certain
degree of financial insecurity and pressure to raise income beyond official budgetary
allocations. Compounding this problem, these once-cloistered party organizations must
also cope with new, market-based competition for training contracts. As a result of
several central policy decisions, party schools must contend with new training programs
offered by universities, administration institutes, and other training organizations in
China and abroad. One feature of political organization in China, limited decentralization,
has been compatible with and enabled diversification in the internal and external
environments of party schools. The provision of cadre training has fragmented, creating
a “many-headed” (duo tou) training landscape where there was once one core player.180
There is, furthermore, intentionality to this marketization: central party authorities
delineated, in a series of formal policy declarations and de facto practice, the new
political and economic contexts within which party schools would function in the post-
Mao period. The introduction of market forces such as competition and hard budget
constraints to cadre training “takes the market not only to the state apparatus, but even
further, right into the core of Leninist governance,” (Pieke 2009b: 23). These new
competitive pressures reflect a priority toward shifting resources and attention to the
professionalization of the cadre corps. This is an unfolding process and it remains
unclear whether China’s technocratic elite will ever move beyond a hybrid one in which
the professional and the personal coexist (Li and White 1990). Primers in contemporary
180 This description of cadre training was given during an interview with a county-level party school vice-principal, Interview 195, May 2008.
143
public administration now stand alongside lectures on the current party line. There has
also been a challenge to party schools’ ideological authority. While party schools have
sought to remain the party’s organizational authorities on the dissemination of party
doctrine to the cadre ranks, this distinction has eroded as a result of intra-party
competition. Through new training academies and revolutionary training schools,
organization departments have stepped into a space once commanded by party schools.
This change in status from being the only significant provider of bureaucratic
training and education to a market player among many competitors has echoes in the
Soviet experience. Party schools in the former Soviet Union have had to forge new
identities in order to adjust to the radically changed educational landscape in post-1991
Russia (Huskey 2004). The Soviet case differs from the Chinese in that the CCP has
moderated to some degree the entry of new players to the training market and, as the next
chapter will show, the party retains, to some degree, the ability to rein in party school
activities.
In some key ways, these developments mirror the “state entrepreneurialism”
(Duckett 2001) and “bureau-contracting” (Ang 2009b) that has emerged in government
agencies and other extrabureaucracies (shiye danwei) of the Chinese party-state. Similar
to the bureau-contracting observed in other agencies, party schools are now privileged to
generate revenues beyond official budgetary allocations and possess de facto property
rights over such extra-budgetary revenue. However, there are some key differences in
the marketization of the party school system. First, party schools must compete with
non-bureaucratic actors to secure training contracts. This is in contrast to the policy
awards that other extrabureaucracies receive, from higher authorities, to provide public
144
services in exchange for some share of the revenue. As the next chapter will demonstrate,
party school leaders exercise considerable discretion in devising income-generating
activities that suit local conditions. It is not necessarily the case that new ventures must
come about through top-down processes or with the explicit direction of higher levels. A
second difference concerns the relationship between party schools and their managing
bureaus (zhuguan bumen). Formally, party schools fall under the leadership of local
party committees, which grants a certain degree of autonomy to local schools. But as
noted earlier in Chapter 2, schools are immersed in a network of advisory and quasi-
hierarchical relationships with other party schools, personnel departments of the party
and government, and other party organs. These relationships are not as strictly
hierarchical as those imposed on other extrabureaucracies, for example the government
agencies responsible for public service provision. Third, these party ventures are not
“semi-legitimate” (Duckett 2001) and are instead part of official school reports to higher
authorities.
The legitimacy granted to schools’ income-raising activities parallels in some
respects the political and economic motivations driving “local state corporatism” in
China’s rural development (Oi 1992; Oi 1998b; Oi 1999). With the onset of economic
liberalization, party school leaders, similar to grassroots party secretaries, faced strong
incentives to engage in entrepreneurial, revenue-generating activities at the local level. In
at least one important respect, however, party school entrepreneurialism differs from
local state corporatism. Whereas local government debt and the pooling of assets are
common features of local economic developmental schemes, party schools in my
145
research do not take out loans to finance new ventures, nor is there a transfer of resources
either between schools or within-locale to support new entrepreneurial ventures.
All of these developments, central party authorities seemed to hope, would prod
party schools out of a Mao-era complacency and prompt significant organizational
change. In response to these various market-based incentives, school leaders must now
evaluate between and adopt appropriate strategies to cope with new pressures.
Competition also presents the potential for party organizations to shift away from the
“counter-bureaucratic” practices of the Mao period and toward more shrewd decisions
about organizational purpose and priorities (U 2007).
146
Chapter 5 Organizational Strategies of Adaptation
In 2006, when I first visited the county-level party school in a coastal province’s
County Y, the school campus exuded all the allure of socialist-style concrete drab: gray
dormitory buildings, cold concrete-floored hallways, flaking interior paint. By the spring
of 2008, when I visited again, the campus buildings appeared the same from the outside,
but there were renovations and changes taking place within. The meeting rooms had
glossy new furniture, there were new projectors in the classrooms, new air conditioning
units, and computers for all the school teachers.181 In pointing out these recently acquired
amenities, the school vice-principal noted the school’s initiative to “improve the teaching
environment” (gaishan jiaoxue huanjing), which was detailed in the school’s annual
report. The initiative was straightforward enough: it called for the funding of various
campus renovations through income from, among other things, property rentals. The
vice-principal proudly boasted, “We used to engage in these sorts of activities before, but
things have been going especially well in the past two years (zuijin liang nian zuo de
tebie hao).” The school was reaping the benefits of local entrepreneurial opportunities.
Even more, the vice-principal announced that “there is a large-scale project in the
works, a new building that will be built in the new development zone (kaifa qu) [of the
county seat]. The party school will be a central part (dangxiao weizhu), along with a
center for retired cadres and a training center for the people’s army (minbing xunlian
181 These upgrades represented investments totaling more than 300,000 yuan (or approximately US$37,500). This is a significant sum when compared to the approximately 200,000 yuan transfer from the local finance department that the school uses to cover annual operating expenses. Interview 232, county party school vice-principal, July 2008.
147
jidi).”182 He elaborated that the local government was planning the construction of a 22-
floor building to house multiple party and government organizations, though he was less
clear on the project timeline and budget. This ambitious plan was by no means unusual
across research sites. The intention to construct a new party school campus was well
beyond the planning phase in many of the schools I visited, from the county to the
provincial levels. Across Provinces A and B, schools were experiencing dramatic
upgrades in school facilities and building booms. What explains these developments, and
how are they connected to the market competition that has developed over the past three
decades?
As discussed in the previous chapter, not long after the onset of Deng Xiaoping’s
first liberalizing economic reforms, the party school system found itself shaped by a slew
of new policies and central directives. These changes in the policy “wind” modified the
internal stability and external environment within which schools operated. Party school
leaders found themselves confronting new challenges, including greater volatility in
funding matters and competition for training contracts. This chapter will detail how, in
response to these threats, schools instituted a variety of organizational changes.
In managing school finances, all variety of school employees – principals, faculty
and administrators – discovered the freedom to “dive into the sea” of entrepreneurial
activity. New ventures ranged from those that were more purely income-generating, such
as property rentals, to degree programs, which raised revenues as well as the profile of
party schools vis-à-vis competitors. All of these activities served, in multiple ways, to
increase the likelihood of party school survival in an increasingly crowded market.
Beyond entrepreneurial ventures, school leaders attempted to position party schools 182 Interview 232, July 2008.
148
within the training market as globally connected but ideologically authoritative training
centers. As part of this reorientation, they adjusted the content of what was actually
taught to cadres in an effort to remain at the forefront of the center’s drive to reshape
China’s administrative and political classes.
While the results of various strategies are mixed, that these changes were
implemented at all demonstrates an organizational adaptiveness and dynamism that is not
immediately apparent when one focuses on the seemingly static nature of the CCP’s
formal organization. Officially, party schools continue to reside where they have always
resided within the CCP’s bureaucratic structure, but to stop there would overlook
significant organizational responses to new market-based challenges. While it would
appear that party schools are at a disadvantage in comparison to the powerful
organization department-sponsored academies that have entered the training market, this
is not necessarily the case. “A bureaucracy need not be more powerful relative to its
opponent, but only that it have the capacity to carry out the functions over which it is
competing,” (Mertha 2006: 301). While other training systems had to begin from scratch,
party schools had the advantage of decades of experience to draw from.
The changes detailed in this chapter reflect how processes bound up in the
marketization of party schools have expanded the repertoire of schools’ activities,
educational and otherwise, and transformed them into diversified, well-connected
organizations requiring leaders with considerable entrepreneurial acumen. This in turn
suggests how political organizations within the party have moved away from the old
command-and-five-year-plan system of the past and toward a more hybrid model, one
that blends central mandates and party control with local, market-driven adaptation. By
149
bringing the market into its organizational heart, party authorities and local party
entrepreneurs have potentially hit upon a strategy that renews the relevance and vitality
of communist party organizations with a seemingly outdated mandate.
2 Staying afloat through alternative income streams
Due to the local nature of budgetary allocations for party schools, financial
shortfalls have become a reality for many schools. Pressures to “self-raise” funds have
prompted the search for independent, income-generating ventures. Party schools from
the county level up are obliged to develop some entrepreneurial spirit, though the success
of any given scheme is dependent on a range of factors, from local economic conditions
to the perceived prestige of a given party school. School administrators and teachers
have expressed frustration with the harder budget constraints of the reform period. As
one teacher expressed, “Party schools can’t rely on their self-raised income (chuangshou)
for survival, we must have a transfer from [government] finance. Earned income is hard
to get, charging fees for training classes is annoying (taoyan).”183 Despite the
organizational adjustments required to survive in the training market, schools have
responded – and some have thrived.
An illustration of the financial struggles facing one training organization and its
coping strategies can be found in a case study of Province B’s Communist Youth League
(CYL) training school. This school, along with its counterparts in other provinces, trains
grassroots party activists and unranked or section-ranked cadres at the county and city
district administrative levels, along with ranked cadres in provincial state-owned
enterprises. Transfers from the provincial finance department to the school were reduced 183 Province A capital party school teacher, Interview 207, May 2008.
150
significantly beginning in the 1990s. This was due to a policy shift which required
individual work units to cover a greater share of the training and living expenses for
cadres that they sent to the CYL school.184 The school has been suffering from dramatic
drops in trainee enrollment since. In 1990, at its peak, this school organized 17 classes
for over 2000 trainees. These figures had dropped to 5 classes for 320 students by
2007.185
With such a decline in core training class enrollment, the school leaders have
sought to plug up budget deficits through alternative means. The 2001 school budget is
provided below in Table 5.1.
Table 5.1: 2001 Income and Expense Report, Province B’s CYL School Income from government transfers (in yuan) Provincial finance bureau transfer 1 130 000 Provincial finance bureau transfer (additional supplement) 70 000 Irregular transfer from provincial finance 150 000 Total income 1 350 000 Expenses Salaries 960 000 Employee benefits 110 000 Insurance 150 000 Public (operating) expenses 1 340 000 Total expenses 2 560 000 Balance -1 210 000 Self-raised income sources Degree program tuition 800 000 Outside-the-plan training programs 50 000 Rent 380 000 Total self-raised income 1 230 000 Net balance 20 000
Source: Internal school accounting document.
184 Provincial CYL school vice-principal, Interview 180, April 2008. He went on to note that a one-week training at his school would cost about 350 yuan in training fees, not including room and board, and after this 1991 policy shift, work units were reluctant to send employees to training programs at this school. 185 Interview 167, CYL school vice-principal.
151
In 2001, the provincial finance bureau transferred 1.35 million yuan to the school, enough
to cover approximately 44 percent of total expenses. The finance bureau added two
additional supplements, totaling 220,000 yuan, to this initial transfer. Taken together, the
three finance bureau transfers cover 53 percent of the school’s annual expenses.186 Three
additional sources of income made up for the budget shortfall.
Income from degree programs
The first, alternative income source was tuition collected for degree programs
organized by the school. A look at the school’s history of degree programs reveals a
convoluted and confusing jumble of ventures. Intermittently since 1985, the school has
partnered with the provincial party school and city universities to offer coursework in
topics such as ideology and administration. These programs culminate in associates
degrees for students and cadres (da zhuan xueli). In 1990, the campus also became the
home of a provincial “youth political school,” an educational institution that enrolled
secondary and post-secondary students in basic coursework. However, from 1996
onward, the school decided to narrow its target audience to secondary students interested
in vocational programs. In 2006, the campus became the home of yet one more school: a
provincial “youth vocational college,” which could enroll ordinary students for three-year
vocational degree programs. At present, this school offers technical secondary school
diplomas in fields such as education management and computer science. Located on the
same campus, the vocational college’s degree programs co-exist alongside the CYL
186 This is compared to a finance transfer covering one-third of total income for the Beijing city party school, reported in Pieke (2009, p. 129).
152
school’s training programs and the income from the former provide a bulwark against
ever-tightening government transfers.187
More broadly, offering degrees is a common practice throughout the party school
system. These degree programs may assume a variety of forms, such as distance learning,
night school, or satellite-campus instruction for degrees granted by other institutions. The
Central Party School initiated this movement in 1985 with the creation of distance
learning degree programs that eventually involved provincial, city, and county-level
schools. These lower level schools joined the endeavor for a share of the profits.188
Officially, these early distance degree programs were intended to meet a particular set of
needs: to edify cadres from the Mao period who possessed low levels of formal education
as well as the “lost” generation of cadres deprived of a decade of formal schooling during
the Cultural Revolution. “At the start of the reforms, training was about general
education, since many cadres lacked fundamentals,” one CPS professor of party history
noted.189
Beyond meeting these official goals, there were additional benefits to party
schools’ various degree programs. They expanded the reach of party school education.
By 2002, CPS distance degree courses were being taught in approximately 2635 branches
187 According to the July 2007 edition of China Youth Daily, over 70 percent of independent CYL schools have this type of degree offering for ordinary citizens. The school’s official history indicates that vocational degree programs began under the ‘care’ of the provincial party committee and provincial government in 2003, and the provincial government formally approved the college’s credential in September 2006. 188 According to a retired head of the CPS distance degree program, the division of profits was as follows: 12 percent to the CPS, 18 percent to affiliated provincial-level party schools (fen yuan, or branch campuses), 25 percent to city party schools (xue qu, or study districts), and 45 percent to county-level party schools (fudao zhan, or tutorial stations). Profits were considerable. Whereas the overhead was approximately 1,000 yuan per student, tuition was in the neighborhood of 10,000 yuan. Interview 118, February 2008. 189 Interview 111, February 2008. The training curriculum for one of the Central Party School’s earliest combined training and degree class in 1984, for example, lists course topics such as basic sciences, math, logic, writing, and language arts (CPS 1985: 224-225).
153
(including fen yuan, xue qu, and fudao zhan) throughout the country.190 They also
signaled party schools’ willingness to compete with universities on their own terms, as
one other degree-granting option available to cadres – and members of the general public
– in need of secondary and tertiary education.
Partnership-based degree programs are common in the party school system, for
example between two or more party schools or between party schools and local
universities, to generate slush funds (xiao jin ku). The previous discussion of the CYL
school’s various degree offerings, in which programs and partnerships come and go,
reflects the deliberate creation of a space for school leaders to cope with insufficiencies in
government funding. This space exists in the broader national (and international) market
for educational services. The school’s experiments with different degree offerings were
not the actions of a rogue organization. On the contrary, this succession of different
schools and accompanying degree programs located on the party school’s campus is part
of the school’s official history. Each venture thus unfolded with government and party
approval, if not assistance. School leaders thus functioned in a context of official
permissiveness – or even encouragement – for seeking new income streams via the
private sector.
“Outside-the-Plan” Training Classes
A second source of income for this CYL school consists of “outside-the-plan”
training programs. These auxiliary training classes are often organized by the school in
cooperation with requesting work units. They may also originate from the school itself.
190 A listing of branch campuses and principals (fuze ren) is available in the CPS Correspondence Education Yearbook (Huang 2002a: 610-43). These branches exist in offices such as the Ministry of Railways, Finance Ministry, Chinese Academy of Sciences, and military bases.
154
CYL school leaders may decide to create a training class and advertise the course widely
to attract students whose sending work units then pay the training fees. By these two
routes for creating outside-the-plan classes, the school may procure some additional
income, though again the income is unstable and subject to negotiation. Even worse for
this particular CYL school, their outside-the-plan training classes are in competition with
other training organizations, such as provincial-level party schools and universities, and
they often lose out to these more prestigious institutions. By the late 1990s, the scope of
the CYL school’s training programs had narrowed because other, higher-ranked “party
schools were hungry for more trainees (dangxiao chi bu bao)” and had begun planning
training programs for cadres who once were the target audience for the CYL school.191
Rental income
The third source of income consists of rent collected from tenants of the school’s
on-campus dorms.192 This is a relatively straightforward and a common practice in many
party schools.193 In the case of the CYL school, rental income in 2001 was sufficient to
cover 15 percent of total school expenses. In Province B, the provincial party school
rents out rooms in the on-campus housing units to student exam takers every year from
November to March.194 In the case of a county-level school, the decision to rent rooms
191 Interview 180, April 2008. 192 Ibid. 193 At field visits to other schools, officials and professors noted (sometimes proudly) the rental of hotel and dorm rooms. This was the case in a provincial capital party school (Interview 140, April 2008, party school professor), a city party school (Interview 201, May 2008, party school professor), and a county party school (Interivew 232, July 2008, school vice-principal). 194 Interview 66, November 2007, provincial party school vice-principal.
155
for income on the school campus was not subject to pre-approval by the county finance
department, and the school could move forward quickly with such new ventures.195
The arrangement for renting out property can also vary across schools. In one
provincial capital, the management of the party school’s hotel and restaurant halls were
contracted out to a separate corporation. A party school teacher there pointed out, “the
school’s hotel is the city government’s asset (zichan), so the government can decide to
contract out the management and earn money from [these activities].” However, he went
on to note that even with the school’s reduced control over this revenue stream, “the
finance department is so pleased with us and the business we bring in,” and this gives the
school leaders more bargaining power when requesting budget outlays from the local
finance department.196
Beyond renting out residential units, party schools in both inland and coastal areas
are engaged in leasing out property for other uses and opening their facilities to a broader
range of clients. On a field visit to a city party school in Province A, a local private
insurance company had rented out the school dorms, cafeteria, and classrooms for a new
hire orientation program. One newly-renovated county party school had rented out its
hotel conference facilities for a local children’s chess tournament.197 Yet another school,
located in a tourist city in Province B, had created an office to provide logistical support
for tourists and out-of-town delegations of officials.198 Party school partnerships with
local entrepreneurs abounded. One school had a steamed bun stand located on campus,
while another had an English school for kindergarteners. The steamed bun stand,
195 Interview 232, July 2008, county party school vice-principal. 196 Interview 133, April 2008. 197 City party school visit, May 2008; county party school visit, May 2008. 198 City party school teacher, Interview 207, May 2008.
156
established around 2003, was proudly announced to me by the vice-principal of the
county party school as part of the school’s fundraising efforts to “improve the teaching
environment.”199 The English school for kindergarteners was established in 2003 on the
property of a county party school, a successful enterprise begun by an unemployed
factory worker turned entrepreneur.200 One provincial party school vice principal
expressed to me that they were “lucky to be located near a national highway” so they
could rent school property to various entrepreneurs on a shopping strip on campus.201
Based on these and other interviews, Table 5.2 below summarizes the range of
entrepreneurial activities observed on party school campuses.
Table 5.2: Local party school entrepreneurial activities Purpose Partnership level Activity Examples Income Update
content Local National
Degree programs and new schools
Distance learning programs, night school, vocational programs, graduate programs
X X X
“Outside-the-plan” training programs
Emergency management seminar X X
Renting property Dormitory rooms, strip mall units X X
Beyond satisfying the immediate goal of raising funds, these party school
activities are intertwined with and contribute to the development of entrepreneurship in
local economies. Schools leaders have been creative and astute at deriving pecuniary
benefits from the activities of local entrepreneurs. The various school ventures observed
at field sites, described in interviews, and documented in official reports all illustrate the
199 This initiative was noted in the vignette the beginning of the chapter. Interview 232, July 2008, county party school vice-principal. 200 Interview 192, May 2008, city party school teacher. In another interview with the school vice-principal, he reported an income of 900,000 yuan per year from rental property located on the school campus (Interview 191, May 2008). 201 Interview 66, November 2007.
157
broadened repertoire of activities that schools can engage in to diversify income streams.
Through these activities, party schools neatly fulfill the practical need to balance their
budgets. At the same time, they are fulfilling, in a more oblique manner, their mission as
cadre training organizations. The party-private sector partnerships that have resulted
from schools’ fiscal circumstances coincidentally dovetail well with the party-
entrepreneur alliance that Jiang promoted in his Three Represents doctrine. Looked at
from another perspective, the entrepreneurial activities of the party schools provide a
model of sorts for the cadre-trainees who pass through the schools in how to be
innovative and resourceful with organizational development.
Support for and opposition to party entrepreneurialism
Within the larger constellation of party and government organs connected to party
schools, the response to these income-generating ventures is mixed. Local finance
bureaus have little incentive to complain about business dealings between party schools
and the private sector. Income generated from these ventures decrease party school
dependence on government transfers as well as increase tax revenues. Matters are more
complicated with respect to central party authorities, however. Tensions over the
profitable nature of party schools’ distance learning and degree programs – and the
concern that these expanding programs distract party schools from their “core” work of
training officials – are evident in the abrupt call to end party schools’ distance degree
programs.202 This policy change, emanating from the Central Committee, has been de
202 This move is not necessarily because incomes from these degree programs are flagging. For example, a provincial capital party school professor bemoaned the school’s potential loss of over 10 million yuan in annual income from their correspondence degree programs (Interview 176, April 2008).
158
facto rather than publicly announced in official documents.203 The cue comes in part
from the omission of “correspondence degree programs” in official party school work
regulations.204 Accordingly, CPS officials recently declared that they would cease to
accept new correspondence degree applicants for undergraduate degree programs
beginning in 2008.205 While the Central Party School will take the lead in implementing
this policy, it will be carried out in a gradual national rollout. The actual effect of this
policy change on schools’ income is not clear, as party schools may still maintain
revenue flows through correspondence degree programs for graduate students, night-and-
weekend classes, and other non-degree offerings.
The motivations for this degree cancellation are several. Officially, CPS leaders
have asserted the fulfillment of correspondence degree programs’ mission to provide
older cadres with remedial education. Second, ending these degree programs is also
consistent with the narrow function of party schools not as general institutions of higher
learning but rather as focused sites for the transmission of skills and knowledge to
promising cadres. As one party school teacher pointed out, the degree programs are not
central to party schools’ organizational purpose, which according to him and official
203 While there are no publicly available official documents mandating the end of these degree programs, the registrar of the CPS correspondence degree office is quoted as saying that the “Central Committee ordered the CPS to cease organizing correspondence degree education.” See “Party school degrees difficult to cancel immediately,” Southern Daily, 9 October 2008. Zhao Changmao, the director of the CPS’s organization department, also stated that the CPS, “in accordance with Central directives” stopped accepting students for correspondence associate and undergraduate degree programs. See “Party schools gradually stopping acceptance of correspondence students,” Henan provincial education webpage, http://news.haedu.cn/GNYW/633609825171310000.html, accessed 19 July 2009. 204 This omission has been traced to the 2008 Party school trial work regulations. See CCP Party School Work Regulations (zhongguo gongchandang dangxiao gongzuo tiaoli), available at http://news.xinhuanet.com/newscenter/2008-10/29/content_10275191.htm, accessed 19 July 2009. 205 Zhu Hongjun, “Central Party School calls for stopping correspondence degrees,” Southern Weekend, 29 November 2007. Interviews 238 and 239, November 2006, county party school vice-principal and township organization department official.
159
declarations is to raise the quality (suzhi) and abilities (nengli) of cadres.206 On the other
hand, the cancellation may also be due to controversies stirred up by the lack of Ministry
of Education accreditation for party schools’ undergraduate degree programs and public
perceptions of party schools engaging in “degree selling”.207 The real reasons for the
cancellation lay somewhere between official declarations and the reports in the media.
Still, this cancellation policy has resulted in closing off one revenue channel for both the
central and local party schools and illustrates how the Central Committee of the CCP has
the wherewithal to interfere in party schools’ profit-generating ventures if they perceive a
threat to the reputation of the party school system and, by implication, the party itself.
3 Coping with competition in the training market
Beyond fundraising activities, party schools have responded in several ways to the
emergence of organizational competition for “training market share”. First, they have
stressed their unique position as institutions with deep party roots, in effect making a
reputational argument for government and party organs to continue enrolling their cadres
in party school programs. Based on their long history as core, “insider” party institutions,
party schools have also extended training programs to new, non-party audiences. Second,
they have sought to augment their program offerings through partnerships with secular
206 Interview 17, central party school teacher, November 2006. 207 See, for example, Liu Lan, “How could party schools sell diplomas?” Democracy and the Rule of Law (minzhu yu fazhi) 2004 (15). This article recounts how the Hainan provincial party school sold degrees from 2000 on and collected 37.5 million yuan from the enterprise. The Hainan party school webpage confirms that “there were some problems associated with the school’s management of its independent degree program,” (http://www.dx.hainan.gov.cn/html/intro.asp, accessed 2 June 2007). In one high-profile court case, a judge sued the Hubei provincial party school because he obtained a degree through their correspondence program in 2001, received a raised based on this advanced degree, and subsequently this degree was not recognized because of the school’s lack of accreditation. The judge is suing the school on the basis that they offered a fraudulent degree. See Southern Weekend, “Hubei judge sues party school for illegally issuing degrees,” 29 November 2007.
160
institutions of higher education. Third, in at least one case, a local party school has
sought to use back channels to thwart the emergence of new competitors. Finally, party
schools have updated their pedagogy and course content. Together, these changes reflect
the adaptability of the party school system – and the potential adaptability of party
organizations more generally – as a response to the introduction of competition, both
intra-party and with outside actors. This competition, in turn, reflects party leaders’
organizational strategies for responding to and initiating change within the party’s own
political structure.208
Maintaining school reputation
When asked about the pressures presented by new training schools and programs
springing up around the country, party school officials in interviews stressed how their
schools are preeminent sites of party study and doctrine. The Central Party School, on its
website, declares how it is the “most important venue for learning about the party’s
nature, and for studying and promoting Marxism, Leninism, Mao Zedong Thought, Deng
Xiaoping Theory and Three Representatives Important Thought.”209 One provincial
party school teacher averred that party schools provide an indispensable service to the
party: “There is some traditional content that you cannot let the market decide [whether
to teach or not]. There is too much diversity and freedom (ziyou) in universities, and the
208 There are similarities and differences between the intra-party competition discussed here and previous work on administrative bargaining and intra-bureaucratic bargaining. Earlier work has emphasized inter- and intra-bureaucracy bargaining in the context of policy and economic development. Consistent with Lieberthal and Oksenberg’s findings from studies of energy projects, the policy process in cadre training is “disjointed, protracted, and incremental” (Lieberthal and Oksenberg 1988: 24). While a similar competition for scarce resources is evident in the emerging cadre training market, Lampton (1987) emphasizes the consultative nature of bargaining in his case studies of water projects. The emphasis on fairness that Lampton finds is not, in my interviews, a component of the competition for training contracts. 209 Introduction to the CPS, htt://www.ccps.gov.cn/dxjj/index.jsp, accessed 29 June 2008.
161
central authorities will always emphasize cadre training in party schools … There will be
a division of labor based on market forces (shichang fenniu).”210 Certain routine training
programs, such as those following a party congress, do continue to be organized and
staged by the party school system. Following the 17th party congress in October 2007,
there was a rollout of training sessions on how to study the documents issued by the
congress, beginning with the CPS and cascading, in sequence, through party schools at
the various administrative levels, where teachers would hold study sessions for cadres
and other party school teachers below them.
Furthermore, party schools’ reputation as bastions of party authority continues to
hold new, prominent audiences in thrall. Recently, the CPS has become a highly desired
destination for entrepreneurs interested in understanding current party policies and future
directions in party doctrine.211 Local schools are also organizing training programs for
private enterprises interested in establishing their own party committees or strengthening
party ties.212 Bringing private entrepreneurs and managers into party schools also fulfills
the mandate in the Three Represents for the party to absorb the “most productive” – code
for entrepreneurial – segments of society. Forging partnerships
Reputation may not be enough to attract students, and party schools have also
strengthened their market position through partnerships. When the Central Organization
210 Interview 211, May 2008. 211 Private entrepreneurs are flocking to the CPS for 6-day training classes in which they learn about party history and policies from experts in the school and party think tanks. The cost for this program is 6,800 yuan. See He Huifeng, “Taking care of business at Central Party School,” South China Morning Post, 24 April 2006. The first group of entrepreneurs to train at the CPS, in 2001, was from Shanghai. See Li Weiping, “Shanghai entrepreneurs attend study program at Central Party School in Beijing,” Hong Kong Wen Wei Po, 10 Jan 2003. 212 Two party schools in Province A (one at the county and the other at the city level) had organized private enterprise training classes as early as 2005.
162
Department called for the implementation of cadre training programs by institutes of
higher learning, party schools began to initiate partnerships with universities in China and
abroad. These partnerships may take on a range of activities, from exchanging
delegations of scholars and administrators to party schools sending cadres abroad for
training. Table 5.3 below captures the variety of domestic partnerships established
between provincial-level party schools and universities.
Table 5.3: Provincial party school partnerships with Chinese universities
Province University partner Year partnership began
Anhui Tsinghua University 2009 Anqing Normal School 2009 Beijing Peking University
Beijing Foreign Language University School of Foreign Diplomacy Beijing Capital Normal School
NA
Chongqing Peking University Beijing Foreign Language University School of Foreign Diplomacy Beijing Capital Normal School
2009
Fujian Fujian Normal University 2008 Hubei Zhongnan Economics and Finance School
Hubei University Chinese University of Geosciences, Wuhan
1993
Hainan HKUST NA Hunan Chinese Agricultural University 2008 Hunan University 2005 Jiangxi Tsinghua University 2007 Liaoning Shenyang University 1999 Qinghai China Tibetology Research Center 2009 Shaanxi Hong Kong Open University 1996 Peking University NA Shanghai Finance University NA Shandong Shandong University NA Shandong University 1998 Shanghai Shanghai Jiaotong University 2003 Shanxi Shanxi University 2002 Yunnan Tsinghua University 2002 Zhejiang Chinese Academy of Social Sciences 2009
Source: author’s dataset; party school web pages, official announcements, and online news articles.
163
It is likely that only those partnerships that might elevate the status of party schools are
broadcast to the general public, i.e., partnerships with national universities, which implies
an underreporting of the true number of party school-university partnerships.
International partnerships are another way for party schools to improve their
profile as forward-looking, globally-connected organizations. Two decades ago, when
the training landscape began shifting, the CPS was the only party school with
international connections (Liu 1989: 110-114). The CPS took a leading role early on
with a research partnership, initiated in 1995, between its Center for International
Strategic Research and Harvard’s Fairbank Center for East Asian Research (Kirby, Ross,
and Li 2005: ix-x). Beginning in 2002, around 60 officials have been sent each year to
Harvard’s Kennedy School of Government for two-month training programs in public
administration.213 The CPS received a delegation in December 2007 from Louisville
University to give workshops on the US political system, though this was under the
purview of a partnership between the LU East Asian Studies Research Center and the
CPS Pacific Research Center.214 The purpose was to train party school scholars, which in
turn would enhance their ability to train cadres on the structure of foreign political
systems. The rationale for these partnerships is simple: “It’s mutually beneficial.
Foreigners are banking on the possibility of influencing the thinking of rising Chinese
leaders, and the Chinese government needs to develop its human capital.”215
Local schools have followed the example set by the CPS. The Wuhan city party
school, for example, lists on its webpage that it has exchange programs with the Moscow
National University, Far East Research Institute at the Russian Academy of Social
213 “China’s Central Party School trains 50,000 officials in 30 years,” Xinhua, 2 October 2007. 214 Interview 42, CPS professor. 215 Interview 34, October 2007, former central-level cadre.
164
Sciences, Hanoi Institute of Politics, and Lyon Institute of Public Administration.216
Some schools will also channel partnerships through their administrative institutes. Table
5.4 below lists the partnerships and exchange programs that the Shanghai city party
school and school of administration have engaged in since 2000. These programs vary in
their structure and have involved the exchange of school administrators, teachers, and
students. A full list of publicly-listed international partnerships by provincial-level party
schools is available in Appendix F.
Table 5.4: Shanghai party school and administration institute international partnerships School International Partner Year began Shanghai city party school Civil Service College in Singapore at least 2004 Ecole Nationale d' Administration
German University of Administrative Sciences Speyer (Germany) at least 2006
Kennedy School of Harvard University at least 2007 Maxwell School of Syracuse University at least 2000 Nagoya University (Japan) at least 2008 Sciences Po (France) University of Alberta (Canada) at least 2004 University of Georgia at least 2004 Shanghai administration institute Civil Service College in Singapore at least 2004 Ecole Nationale d' Administration at least 2004 Ecole Normale Superieure Humboldt University at least 2004 Kennedy School of Harvard University Maxwell School of Syracuse University Postdam University (Germany) at least 2004 Sciences Po (France) at least 2004 University of Alberta (Canada) at least 2004 University of Georgia
Source: Author’s dataset.
Organization departments, for their part, have not remained isolated.
International partnerships between the COD and institutions of higher learning abroad
include a master’s degree program for cadres at the Nanyang Technological University in
216 Wuhan party school introduction, available at http://www.whdx.gov.cn/xyjj.asp, accessed 1 June 2009.
165
Singapore and a global public policy collaboration between the COD, Peking University,
Columbia University, Sciences Po Paris, and the London School of Economics (Li 2006:
20). In general, these partnerships and the additional training that cadres receive through
them are consistent with the “opening up” promoted by Deng and subsequent leaders. As
one CELA administrator pointed out, “Our leaders need to be taught how to borrow from
the entire world’s best cases. As Deng said, it doesn’t matter if it’s a black cat or a white
cat, so long as the knowledge and models we adopt are the best.”217 In order to compete
in a global economy, cadres require an outward orientation and knowledge of other
systems; their party education has been tailored to address this particular need. From the
perspective of party schools, which are now facing pressure from increasingly
internationally-oriented universities and organization department schools, partnerships
are instrumental for lending additional cache and depth to the curriculum. In the end,
though, these partnerships increase the number and types of institutions involved in cadre
training, which service the goals of the party’s marketization strategy to create the
conditions for new content and expanded problem-solving capacity.
Outmaneuvering competition
In a less constructive vein, party schools have also dealt with competitors through
preemptive maneuvers. One provincial party school has sought to stave off the opening
of new training centers through connections and behind-the-scenes negotiations. One
leading university in Province A sought to build a cadre education academy in 2005, but
the provincial party school opposed the scheme. Party school leaders, along with
contacts at the Central Party School, challenged the project on the grounds that 217 Interview 184, April 2008, CELAP director.
166
“ideological training wouldn’t be preserved at the university.” 218 In the end, the key
party authority, the provincial organization department, would not approve the new
academy “because they couldn’t ensure the quality of the ideological training [at this
proposed center].”219
Such a gambit may limit the efforts of universities, which are not party organs,
but this strategy has limited leverage in the intra-party competition which has sprung up.
Preventing organization departments from opening training centers presents a thornier
problem. An organization department cadre noted that the provincial party school may
have been successful at blocking the university’s attempt to open a training center, but
“party schools don’t have any way to stop the organization department from building new
centers.”220
In summary, party schools have pursued several strategies to cope with the
competitive pressures arising out of the reform-era training market. These responses,
some rhetorical and others more substantive, have opened up party schools to new
partnerships and brought about adjustments in how party schools situate themselves
within the training market. Peering inside party schools, furthermore, reveals how these
schools have adapted to new constraints and opportunities in the actual content of what
they teach cadres in the new training market.
4 Change measured: Content analysis of party school training
One consequence of the introduction of market processes and pressures has been a
rethinking of what should be taught and how. By examining the substantive impact of
218 Provincial party school teacher, Interview 211, May 2008. 219 Ibid. 220 Interview 212, provincial capital organization department cadre, May 2008.
167
these market forces on party school curricula, it is possible to identify what kind of
change party authorities have ushered in with the opening of the market floodgates. Party
leaders have framed cadre training as integral to the process of party and government
bureaucrats’ successful management of China’s economic development and the CCP’s
continued monopoly on political authority. CCP leaders portray cadre training as the
vehicle for creating the kind of flexible, up-to-date bureaucratic corps that the party
requires. How do party schools reflect the new priorities of central party leaders? What
is taught in party schools? How has this changed over time? Examining the content of
party school training classes reveals how party schools have “kept pace with the times”221
and, in so doing, embody the organizational adaptability that has in part defined CCP
leadership for the past three decades.
This section considers changes in the training content of classes at the Central
Party School, the flagship within the party school system. CPS yearbooks contain
teaching plans for training programs organized over the period 1983 to 2000, and
additional training syllabi were collected during site visits. Table 5.5 below offers a
summary of the syllabi collected, by year.
Table 5.5: Central Party School training syllabi analyzed, by year Year 1983 1984 1985 1990 1993 1994 1995 2000 2007 Total
Training Syllabi 2 1 5 5 9 10 14 7 9 62
Because of the low number of syllabi available for 1983 and 1984, these two years were
excluded from analysis. The categories used for coding syllabi are given in Appendix G.
221 In a tour of the new Pudong CELA, Zeng Qinghong of the CCP Central Committee (and former Central Party School president from 2002 to 2007) declared, “One of the most important reasons behind the Party’s successful advancement of the transformations in economic growth, social administration and the Party’s governance is efficient cadre training, which makes leading cadres at various levels change their ways of thinking and working by keeping pace with the times,” (Xinhua, 15 May 2007).
168
One starting point is to compare the share of total training time, for a given
year, dedicated to various topics. Figures 5.1 and 5.2 below suggest that in some
areas of training content, change has been dramatic.
Figures 5.1 and 5.2: CPS training content by category, percent of total training time
2007
Management19%
Policy15%
Case studies32%
Vocational training
16%
Other4%
Orthodoxy9%
Reform-era theory
5%
1985
Orthodoxy31%
Reform-era theory
0%
Management6%
Policy26%
Case studies5%
Vocational training
27%
Other5%
Notes: Data obtained from CPS training syllabi. Study of party documents, including party-building documents, are included in the “policy” category. It appears that political theory – the foundational works of Marxism-Leninism, Mao
Zedong Thought, and the political declarations of paramount leaders Deng, Jiang, and
Hu – has decreased from 31 percent to 14 percent (adding together “orthodoxy” and
“reform-era theory”). Case studies and vocational training (which includes
coursework in accounting, law and economics) has increased from a combined total
of 32 percent to 48 percent. Management, a separate but related category which
includes training in crisis management, leadership, and speechmaking, among others,
has tripled from 6 percent to 19 percent of party school training time. However, these
figures present only a snapshot comparison of the two years that bookend the
available data and, as the preceding discussion will show, this comparison masks
169
some of the more interesting patterns (and noise) in examinations of longitudinal
trends over the two decades. Still, these changes in training content offer hints as to
what party leaders consider the most important knowledge and skills for cadres
serving at various levels in the system.
First, what has happened to the teaching of core ideology, proxied by the
study of party theory, in party schools? This is arguably the bread and butter of party
school training. In the 1983 Central Organization Department’s 8-year cadre training
plan, which called for the ‘regularization” (zhengguihua) of cadre training and
spurred a movement toward professionalization of the party school system, Marxist
theory was still accorded first mention in the COD’s vision for the content of cadre
training (COD 1983: 68). As a gesture to the ideological underpinnings of the
party’s rule, training programs often begin with a unit on “Fundamental Marxist
Theory” (CPS First Short-term Training Class in 1985).
By 1995, however, it is possible to detect a shift in the canon. The first
section in CPS training classes jumped straight to study of Deng Xiaoping’s writings
(1995 CPS Yearbook: 172-92), dropping the works of Mao, Marx and Lenin. Still,
CPS syllabi continue to couch the general goals of training classes in terms of the
long pedigree of thinkers shaping the Party’s ideology today. The 2007-08 Young
Cadre Training Class syllabus begins with a declaration of the objectives of the
course: “In accordance with the professional needs of provincial and ministry-level
leading cadre positions, we will systematically study fundamental questions of
Marxism-Leninism, Mao Zedong Thought, Deng Xiaoping Theory, ‘Three
Represents’ Important Thought, and contemporary world economy, world technology,
170
international law, global military affairs, global ideational trends, world religions,
etc.”222
Figure 5.3 illustrates the amount of training time, as a share of total training time,
dedicated to orthodox theory in the training syllabi collected. Orthodox theory in this
figure is defined as course content dedicated to Marxism-Leninism and “Mao Zedong
Thought”.
Figure 5.3: Percent of CPS training time dedicated to orthodox theory
Arguably, this is an area where party schools possess a competitive advantage over other
training organizations. At the same time, there may be pressure to push aside these topics
in favor of more professional training. A test as to whether orthodox theory is still the
core focus of party school training or more of a formality can be assessed by measuring
the actual amount of training time dedicated to the foundational thinkers. There is a
downward trend in the available data, from a peak of 30.7 percent of class time in 1985 to
9.3 percent in 2007.
222 Teaching plan, CPS 23rd One-year Young Cadre Training Class.
0.1
.2.3
Per
cent
of c
lass
tim
e
1980 1990 2000 2010Year
171
It may be that the decreasing emphasis on foundational thinkers has given way to
study of the ideas of more recent leaders. However, there also appears to be a general
decrease in the amount of training time dedicated to the theories and speeches of post-
Mao leaders from Deng Xiaoping to Jiang Zemin (Figure 5.4). Classes spent the most
time on Deng’s “reform and opening up” in the mid-1990s, but the overall amount of
time spent on studying the guiding words of top leaders has decreased from a peak of
20.6 percent in 1995 to 4.5 percent in 2007.
Figure 5.4: Percent of training time dedicated to the theories of reform-era leaders
In field interviews, party school teachers and officials have explained why there
might be a decrease in the attention given to orthodox party theory. One Central Party
School teacher mused,
[Content changed] because there was a realization that the old coursework was insufficient for solving the problems of the reform era. Marxism didn’t have the answers. Students needed to have a more global outlook (shijie yanguang). There was debate over these changes [in curricula], nothing public, but debate nonetheless. We are still debating these changes today. Some want to stay with the ‘five old subjects,’ some want issues in economic development and related theories to be central.223
223 Interview 114, CPS professor of party history, February 2008.
.05
.1.1
5.2
.25
Per
cent
of c
lass
tim
e
1980 1990 2000 2010Year
172
A second look at the data, however, tells a less conclusive story. There
has not been a steady decline in the study of core party ideology, more broadly
defined as Marxism-Leninism and the subsequent writings and declarations of
CCP leaders. A look at the total share of training time dedicated to party theory
shows a noisier trend (Figure 5.5), with declines in the mid-1990s and a
significant drop during the most recent year for which data is available.
Figure 5.5: Percent of CPS training time dedicated to party theory
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
1983
1985
1987
1989
1991
1993
1995
1997
1999
2001
2003
2005
2007
Year
Per
cent
of t
otal
trai
ning
tim
e
post-Mao party theoryMao party theoryMarxism-Leninism
The sharp increase in training time dedicated to foundational thinkers and party
orthodoxy in 2000 may be explained by two 20-week classes convened that year: the first
county party secretary training class (xianshi wei shuji jinxiu ban) and first western
region cadre training class (xibu diqu ganbu peixun ban). Both were non-routine training
classes and have not been organized since. Nearly half of the training time in these two
classes was dedicated to core party theory, from Marxism-Leninism to the writings and
speeches of Jiang Zemin. There appear to be varying motivations for the organization of
these two classes. First, a county-level training class was one manifestation of
centralization efforts begun during the Jiang Zemin administration. Given the convention
173
that party schools train cadres ranked one administrative level (or rank) lower, the CPS is
traditionally a training ground for central and provincial (ministerial) cadres. However,
five thousand county-level party secretaries and magistrates were sent to central training
schools in a five-year series of classes.224 Second, this special training class was also a
means to disseminate policy priorities bound up in building a “new socialist countryside”.
Similarly, the CPS training class targeting cadres from the western provinces came on the
heels of the 1999 rollout of the Western Development Program (xibu da kaifa jihua).
Trends are somewhat clearer but also puzzling with respect to training time
dedicated to party building, or course content focused on CCP doctrine and the party line.
Party schools in many ways embody and shape the party line of the day. Party members
study articles written by CPS leaders and party school-affiliated theorists to grasp the
evolution and future direction of party doctrine. By 2005, for example, leaders and
theorists in the CPS had developed the concept of a “harmonious society,” the idea
promoted by Hu Jintao to guide economic and social development under his
administration. Yu Yunyao, vice president of the CPS, has written at length about
bundling together this concept of social harmony with “scientific development,” two
terms often seen in official declarations.225 Shen Baoxiang, also a CPS professor, has
contributed to the ideological and developmental agenda and noted the many “historic
topics” that the party must confront: “implementing the concept of scientific development
across the board, building a harmonious society, building a new socialist countryside, and
224 A Southern Weekly article on this training class, for example, noted that this class enabled the “policy-making level [of government] to grasp grassroots officials” (juece ceng zhijie ‘zhua zhu’ jiceng guanyuan). See http://www.nanfangdaily.com.cn/zm/20070104/xw/szxw1/200701040006.asp, accessed 10 February 2008. 225 “PRC Central Party School Vice President on Building Socialist Harmonious Society,” Guangming Ribao, 18 May 2005.
174
building an innovation-oriented nation.”226 Declarations emanating from party schools
thus serve as a bellweather for shifts in party ideology, and it would make sense for
cadres to attend party schools to learn about these cutting-edge developments.
Yet, party building as a topic in CPS training classes has suffered from uneven
levels of attention over time (Figure 5.6). Party building is defined here as courses
dedicated to party history, the party’s constitution, the party line, and the study of central
party policies focused on party building. The trend is not altogether clear, but there is a
decline in the amount of time dedicated to these subjects beginning in the 1990s.
Figure 5.6: Percent of CPS training time dedicated to party building
.05
.1.1
5.2
.25
Tim
e de
dica
ted
to p
arty
-bui
ldin
g
1980 1990 2000 2010Year
Party-building Fitted Values
This downward trend may be due to crowding out by the more secular topics
(explained below) that have claimed increasing shares of training time, topics that are
compatible with the professionalizing ambitions of recent administrative reforms. This
trend may also be exacerbated, in more recent years, by the creation of organization
department-sponsored training programs in which cadres are exposed to immersive,
226 Shen Baoxiang, “Reform and Opening Up is Still the Idea that Decides China’s Destiny,” Study times, 20 March 2006.
175
“revolutionary spirit building” activities that in some ways substitute for studying the
party line. Such “experiential learning” may have claimed some of the ideological
ground once held by party schools, at least in terms of capacity to build “party spirit”.
One Central Party School teacher complained,
You have ridiculous training exercises [at these new CELA]. Cadres will be issued a revolutionary outfit and backpack and climb Jinggangshan. It’s so silly! In every 2- to 6-month long training class that the CPS organizes, trainees will have to spend 7 to 10 days at one of these academies and do this silly experiential learning (tiyan shi). Even township cadres have to go to these organization-department academies for revolutionary training.227
Such complaints present a stark contrast to the official story that party schools and
organization department training academies have a mutually supportive (hu wei buchong)
relationship.228 They also hint at an emerging division of labor, where party schools
remain sites for the serious study of party leaders’ theoretical development (hence
grooming trainees for leadership in party theory), while the organization department
handles matters of party esprit de corps.
Furthermore, in place of orthodox theory and party building, CPS training courses
now offer more diversity in content. Topics falling under the category of management
have claimed an increasing share of training time. Management-related content, in
contrast to the other categories discussed thus far, is more secular in nature and
administrative, rather than ideological, in focus. This category includes the following
topics: leadership theory or “the art of leadership,” crisis management, speechmaking and
227 Interview 114, CPS teacher of party history, February 2008. 228 A typical statement, repeated in interviews, can be found in official publications such as the Chinese Personnel Report: “Party schools, administration institutes and other cadre academies will each have their particular emphasis, be mutually supporting, and have a division of labor in cadre training.” He Xican, “Innovation in cadre education and training through the creation of ‘three academies’” available online at http://www.rensb.com/showarticle.php?articleID=10758, accessed 2010-03-09.
176
communication, media relations, principles of service-based government, public
administration, and strategic thinking. The figure below captures this increase, with
course time on general management skills consuming 16 percent of CPS training time in
2000. This year was significant as the inaugural year for a five-year rotational training
program to train all county magistrates at the central level, “the front line commanders in
the building of a new Chinese countryside.”229 Leadership and management training time
increased overall from 5.6 percent in 1985 to 18.9 percent by 2007.
Figure 5.7: Percent of CPS training time dedicated to management
0.0
5.1
.15
.2Po
rtion
of c
lass
tim
e de
dica
ted
to le
ader
ship
1980 1990 2000 2010Year
Leadership skills Fitted Values
Included in this shift to emphasizing more practical, professional skills has been
an effort to inform cadres on the state of various governance issues in China. Figure 5.8
gives the amount of training time dedicated to policy briefings. Experts are often invited
to the CPS to give briefings on some aspect of the “national situation” (guo qing). Such
training exercises reflect two opposing trends: a departure from orthodox theoretical
concerns and emphasis on practical knowledge as a key attribute of rising leaders.
229 Hu Jintao was quoted as saying this in a Southern Weekend article on this county-level magistrate training, “Five thousand county magistrate training by rotation,” 4 January 2007.
177
Figure 5.8: Percent of CPS training time dedicated to policy briefings
0.0
5.1
.15
.2Ti
me
dedi
cate
d to
brie
fings
on
the
natio
nal s
ituat
ion
1980 1990 2000 2010Year
natsitcps Fitted values
With the decrease in theoretical concerns, there have also been pedagogical shifts.
Content analysis of a combined dataset of national and local training syllabi (N=162)
revealed an increase in the amount of training time devoted to case study as a method for
informing cadres about the state of affairs in their own locales, elsewhere in China, and
globally. Figure 5.9 below indicates the amount of party school class time, on average,
dedicated to case studies, both domestic and international, across the CPS and local
schools.
A look at one case study used at the CPS in 2007 illustrates how leading cadres
are presented with a complex situation and then prompted to debate broader issues in
leadership. Case study 2007-015, “A party secretary’s controversial leadership style,”
tells the story of an anonymous city party secretary who uses heavy-handed tactics to
clean up a poor, crime-ridden county. This party secretary pushes through public works
projects by lowering wages and imposing corvee labor requirements on local villagers,
privatizes all state-owned enterprises in the county, and brings in the media to report on
corrupt subordinates. He claims that such rapid reform measures are necessary in order
178
for China to “walk in 50 years what the West walked in 300 years” (p. 7). There is
deliberate ambiguity in the “correctness” of the leadership practiced in this case study,
and cadres are asked via discussion questions posed before and after the case study to
reflect on their own management philosophies.
Figure 5.9: Percent of training class time dedicated to case studies 0
.05
.1.1
5Po
rtion
of c
lass
tim
e de
dica
ted
to c
ase
stud
ies
1980 1990 2000 2010Year
gcs Fitted values
Such case studies illustrate the participatory shift in training and the move away from
unidirectional, doctrinal learning. Developing such critical thinking skills is paramount
given China’s unpredictable, rapid local development and the aim of forging leaders
capable of making autonomous decisions which are nonetheless consistent with the
center’s developmental goals.
In sum, content analysis of CPS training syllabi allows for some measurement of
the change in what cadres are taught within the party school system. Training content is
a response to and a reflection of party leaders’ awareness that cadres often make
decisions under rapidly changing conditions and with a certain degree of autonomy.
Professional management skills are thus crucial and consume an increasing share of
training time. Less definitive evidence exists to support the conclusion that training time
179
dedicated to the theory of core party leaders – party orthodoxy – is on the decline. There
are also hints of a division of labor across training organizations, evident in the decreased
attention to party building at the CPS. It appears that the party school’s organization
department counterparts have moved toward more creative ways of instilling party spirit
in individual cadres, beyond the uninspired study and memorization of doctrine.
How much do local party school training classes differ from the Central Party
School? Data were insufficient to show trends over time for various topic areas, but a
comparison of CPS training content in 2007 with what was taught in the provincial-level
party schools of provinces A and B reveal similarities as well as striking differences
Table 5.6: Comparison of CPS and provincial-level party school training content
CPS
Provincial schools
combined
Of which: Province A
Province B
Year 2007 2008 2008 2008 Total syllabi 9 17 10 7
Percent of total training time dedicated to each category: Orthodoxy 9 2.7 2.2 3.5 Reform-era theory 4.5 2.8 0.4 6.1 Management 19.1 17.3 16.9 17.8 Party-building 11.5 13.3 15.8 9.6 Central policies 3.9 6.2 3.8 9.6 Local policies na 14.3 16.1 11.6 National briefings 19.4 8.7 9.2 7.8 Case studies 13.4 9.4 8.7 10.4 Economics 5.6 13.1 14.4 11.2 Law 5.6 5.7 4.6 7.2 Vocational skills 4.4 2.0 0.0 5.0 Other* 3.6 4.5 8.1 0.2 Total 100 100 100 100
*Other training content includes optional televised and weekly lectures (topics unspecified) and physical education. Source: Author’s dataset Analysis is based on the collected syllabi for core (“inside-the-plan”) training classes
organized by each school. Table 6 below summarizes the content across these provincial-
level counterparts to the CPS. Under the expectation that there will be some lag time
before local schools are able to incorporate training content from training syllabi
180
transmitted down through the system, the comparison below summarizes training content
from 2007 CPS and 2008 provincial party school syllabi.
What does this comparison suggest about the degree of central control over local
school training content? There are similarities in the allocation of training time to certain
topics, such as management, party-building, and legal studies. This suggests a system-
wide shift toward a hybrid identity for party schools as both traditional political
academies and more secular, forward-looking management schools. Local schools,
however, seem to be moving more quickly toward the latter identity. This is borne out by
local de-emphasis of party theory, a total of 5.5 percent of training time across the local
schools compared to 13.5 percent at the CPS (summing over the orthodoxy and reform-
era theory categories).230 There is also greater emphasis by local schools on official
policies, central and local. Policy analysis claimed 20.5 percent of local party school
training time, whereas this was only 3.9 percent at the CPS. There is also more emphasis
on economics at the local schools, in contrast to briefings on the national situation at the
central school, which reflects differences in the kinds of leadership decisions that
managers at these various levels may face in their daily work.
In the absence of longitudinal data, it is difficult to tell whether local schools have
been drifting further from their central model over the course of the reform period. The
evidence presented here indicates that local schools do enjoy some autonomy with
respect to training content, as there are significant departures from the central curriculum
across several topic areas.
230 The provincial schools make up for this difference through slightly more party-building content – e.g., the study of official party papers, visits to revolutionary historical sites. Party-building activities may have also increased this particular year as a build-up to the 17th Party Congress convened in October 2008.
181
5 Conclusion
In response to the challenges bound up in the marketization of cadre training,
party school leaders have implemented a variety of organizational reforms. These
strategies address the various dimensions to the problems presented by market
competition and less stable income streams: diversification of revenue-generating
schemes, the initiation of new programs and partnerships to cope with competition, and
adjustments in the content of training classes. Above all, party schools now have strong
incentives to search actively for solutions to ever-shifting circumstances. New
entrepreneurial ventures served important functions for the school leaders who were
responsible for the allocation of resources between official (training) and market
(entrepreneurial) activity: in maximizing revenues through new ventures, schools could
offer increased benefits to employees and generate the resources for reinvestment in the
school. These, in turn, would enhance a party school’s competitiveness. Through
content analysis of Central Party School training syllabi, the mandate to craft a
professional bureaucratic body is manifest in more management training and briefings on
local, national, and global developmental case studies. All of these changes have added
diversity to the training content received by local bureaucrats. This diversity, in turn,
represents a locally adaptive response to incentives generated by central party authorities.
Entrepreneurial activities by party schools have accompanied substantive shifts in
training content. A nationwide party school degree system, now officially defunct, was
the beginning of a stream of income-generating activities devised by party schools.
School leaders have leveraged their particular organizational resources in the market by
providing facilities and human resources to private actors in search of everything from
182
well-located rental property to event managers. Such entrepreneurship, which carries the
party stamp of approval, indicates that party organizations are making a fundamental shift:
rather than absorb entrepreneurial talent through the Three Represents, party leaders
themselves are embodying the new spirit of market-based competitiveness and the
innovation that comes with it.
At the same time, the most elite training classes emphasize general knowledge
rather than specialization. The breadth of bureaucratic positions and responsibilities is
so great, it is more logical for elite leaders to be groomed in the broad arenas deemed
most relevant to contemporary leadership. “The ability to synthesize, to adopt global
views and to make global decisions is considered the hallmark of the elite,” (Suleiman
1977: 142). As discussed in Chapter 3, party schools are part of the selection process for
political elite. The criteria for selection, and the institutions implicated in these selection
processes, has changed to reflect the new emphasis on a globally oriented, professional
party bureaucracy.
Another related outcome of the introduction of market forces to party
organizations has been an increasing integration of party schools in local economies.
Schools serve an economic function by contributing to party revenues and local
development. This is due to schools’ entrepreneurial drive for new income sources,
which in turn pushes them to turn outward and actively consort with a broad range of
local actors: private companies in search of rental facilities; private entrepreneurs seeking
out party school training to build professional networks; private corporations in need of
retail space; party and government cadres interested in taking part time degree programs;
party and government work units in search of made-to-order training classes; private
183
citizens seeking convenient and inexpensive degrees. These local transactions and
relationships all embed party schools more deeply in local economies. Party schools are
no longer closed, mysterious fortresses of party authority. Even more, this local
embeddedness suggests that party schools have carved out for themselves an expanded
role in local communities, beyond their traditional function as elite political training
academies. This has not been through a redefining of the role of party schools, but rather
by retaining the place of these schools within party organization and grasping new
opportunities.231 New functions have been layered atop the old. School leaders have
developed formal and informal ties to those key actors in the party and beyond the party
who can affect the conditions of party schools’ existence as increasingly local actors.
Local embeddedness also pays off within the national training market. Party
schools in model regions may serve as interlocutors for interested governments located
elsewhere in China but who are nonetheless interested in learning the secrets behind, for
example, the success of a locale in a particular sector of economic development, or
environmental standard enforcement, or for some other attention-grabbing reason. So
long as the local party school is able to maintain a convincing research profile or
knowledge of conditions in the locale, interested parties will inquire. Training classes
organized by party schools in the most economically developed coastal cities, for
example, provide cadres from inland regions with valuable reports and field experiences
on different economic development models. City-level party schools will often have
special research centers for investigating special topics. In one coastal, city-level party
231 This is similar to what Selznick pointed out in his study of the Tennessee Valley Authority, “In relations to the states, counties, and other local agencies, the TVA could not have operated successfully without framing its program within the existing pattern of government, including the powers and the traditional prerogatives of the local units,” (p. 55).
184
school, an in-house research center focuses on the development of the local private
sector.232 Local research conducted by school teachers and groups of cadre trainees are
then disseminated and stored in locally-distributed volumes of published field reports.233
This chapter also points to changing conceptions of the ideal leading cadre in
today’s CCP. In the past, selfless revolutionaries capable of carrying out Marxist
prescriptions for class-based conflict were the role models for aspiring cadres. Cadres
wore their “non-expert” status as a badge of honor.234 This was followed by a gradual
process of replacing revolutionaries with “semi-bureaucrats” and, more recently,
professional administrators (Lee 1984; Lee 1991; Li 1998; Vogel 1967). Training
content today paints a strikingly different ideal from the pre-reform period, or even the
early reform period. The model cadre of the present is a professional, entrepreneurial,
globally aware manager with a more artificial understanding of the party’s doctrinal
foundations. This new emphasis on professionalization, however partial, accords with
the view that a more effective bureaucracy should comprise (somewhat) independent-
minded mandarins rather than bureaucrats subject to the mandates of all-controlling
executives (Aberbach and Rockman 1988). Eddy U argues that Marxist-Leninist systems
of rule have generated “counter-bureaucracies”, or systems departing significantly from
Weberian ideals of rational bureaucratic rule (U 2007). What appears to be taking place
with the reform of cadre training and transformation of party schools is, in post-socialist
232 Interview 97, city party school teacher, Province A, December 2007. 233 When visiting one city-level party school, I was able to obtain an internal publication, “Materials from study and research activities” (xuexi diaoyan huodong cailiao) which included reports by cadre trainees. These contained data, sometimes comparative, that were obtained during field investigations on local political and economic developments (e.g., population issues, legal reform). 234 The idea was that so long as a cadre was “red,” a general background would nonetheless allow him to lead the experts. This was represented in the saying “experts must always be led by nonexperts” (waihang lingdao neihang). See Meiru Lu, p. 99, quoting page 1 of the Beijing Daily (Beijing ribao) dated 11 November 1979.
185
China, movement toward the professionalized bureaucracy that was so difficult to
achieve in the highly politicized climate of Maoist China. This professionalization,
however, should not be taken as an indication that politics are waning; rather, “the
promotion of technocratic rule serves enduring political agendas” (Pieke 2009a: 93) and
reflects a concerted attempt to deflect more destabilizing calls for democratization. The
project has become one of updating the “new socialist man” such that he can respond to
and generate prosperity in a changed context (Klugman 1989; Munro 1971).
Market processes in the party school system have thus served the interests of
central authorities by inducing party schools to contribute to the party’s project of
redefining the figure of the cadre in a market economy. The focus within the CCP on
developing “five capacities” (wuzhong nengli), which party leaders believe are crucial for
the party’s continued rule, require a new sort of cadre leader.235 This is because, in
grander terms, “the prosperity of our socialism and the future of China depend on
whether our Party can cultivate a generation of highly qualified leading talent to enhance
the Party’s governance capacity,” (CPS 2004: 133). In the new training market, party
schools are generating content that caters to those desired skills and abilities.
235 These capacities are identified in internal party documents as: possessing scientific judgment of domestic and international conditions; the ability to manage the market economy; the ability to cope with complex circumstances; the ability to rule based on the rule of law; and the ability to manage situations in a comprehensive fashion. See (CPS 2004)
186
Chapter 6 Conclusion
What is remarkable about the communist party-led states of the twentieth century
is the overall longevity of their rule. Among current regimes in the world today, one-
party autocracies have lasted for, on average, 25.5 years,236 while collapsed single-party
communist regimes endured for an average of 48.3 years.237 A certain institutional
stability underlay the persistence of these regimes, and an appreciation of this toughness
precedes consideration of the ultimate demise that claimed many of the CCP’s political
peers. China under CCP rule is fascinating not only because it endured through the
ideological collapse and tumult of the 1980s and 1990s which engulfed the communist
world, but because many of its Leninist party structures have withstood the tests of
economic and social transformation. This project took such institutional resilience as its
starting point.
In embarking on this project of mapping institutional survival in China, I began
with a puzzle. Why is it that the party school system, which has roots in early CCP
organization and reflects the Mao-era importation of Leninist party forms, continues to
survive more than three decades into the reform period? What role do these
organizations play in the maintenance of the CCP’s political authority, what challenges
have these organizations faced, and what coping strategies have they adopted? I sought
to expand beyond earlier findings that party schools are drivers of ideological change
within the party. In focusing on party schools to gain traction on broader questions of
236 This is for the period 1960-2003, according to the Hadenius and Teorell dataset, totdur1ny and regime1ny variables (Hadenius and Teorell 2007). 237 See Dimitrov 2010 (N=10).
187
party change, I have asserted that party schools are a valuable window into how CCP
organizations are capable of adaptive learning.
First, I sought to identify whether these party organizations contribute to solving a
particular, internal problem of single party rule. While previous studies have focused on
the ideological authority of party schools, this study has shown how these schools also
perform a critical gatekeeping function. As such, they constitute an important
mechanism of elite control. All authoritarian rulers must address an elite selection
problem, and this complicates the stark choice between allocating resources to “guns
versus votes.” Investing in sub-party organizations of elite selection is, in some ways, an
investment in both sticks and carrots, insofar as political elites are the party agents who
populate and manage the party’s repressive apparatus(es) and distribute the policy goods
that placate the general population. In short, the party must invest in managers.
In Chapter 3, I demonstrated that the party school system is a channel for
mitigating the selection problem facing party authorities. This suggests that cadre
training more generally fulfills a selection function within the party. Importantly, this
selection is anticipatory and represents an ex ante form of bureaucratic control. Within
the party’s repertoire of controls over the bureaucracy, selection mechanisms such as
training exist prior to those that monitor cadre behavior. Personnel selection is critical in
an authoritarian system where policymaking is a fragmented process, implementation
varies at the discretion of local bureaucrats, and monitoring remains weak. In a
bureaucratic system that is hierarchically organized but subject to decentralizing forces,
party schools are a manifestation of the unifying role of party organization in Chinese
governance. Party schools lend structure and regularity to the party’s vital need to
188
manage personnel. As such, they complement other institutions of political personnel
management within China. This control is an important link in the chain of authority
structures that support the CCP’s continued governance.
Throughout the reform period, party schools have also exhibited an adaptiveness
to changing conditions. In recent decades, school leaders have had to cope with
exogenous pressures imposed by central party authorities. They have responded to the
erosion of their monopoly position in the cadre training market and the introduction of
competition. Competition, by definition, introduces redundancy to the system.
Redundancy is primarily a safeguard against the failure of a bureaucratic system to fulfill
a task, given some probability that an individual agency might fail. Another benefit to a
competitive system, and the motivation cited by central party authorities in China, has
been the desire to update the incentives driving organizational survival and adaptation. A
key force for change has been the creation by central party authorities of a training
market whereby party and non-party organizations are allowed to compete with party
schools for training contracts. This has led to a system of overlapping agency
jurisdictions where one objective of a redundant system has been met, i.e., a more
innovative system overall. Party schools have responded to these new, centrally-
imposed competitive pressures and opportunities by exhibiting a certain degree of
organizational agility. Market competition has disciplined party schools with harder
budget constraints and induced them to search for more innovative training content and
new organizational directions. Many have become more outward-looking, even global,
organizations.
189
The opening of markets for goods and services, which were a key part of the
liberalizing reforms initiated by Deng Xiaoping from the 1980s onward, also created the
conditions for party schools to engage in entrepreneurial, income-generating activities.
This stands in contrast to concerns during the early reform period that bureaucrats would
exit the bureaucracy for private market opportunities and hollow out the bureaucracy of
managerial talent (Li 1998). Instead, party organizations themselves are incubators of
entrepreneurial activity. Party schools are now involved in a variety of activities to boost
revenues, from renting out facilities to partnering with local organizations. They now
have incentives to update both facilities and personnel to compete for clients and partners
from the public and private sectors. Certain features of the organization of the party
school system, e.g., the decentralization of funding and semi-autonomy over curriculum
matters, produce the conditions for entrepreneurial activity and competition. All of this
has generated greater variation in the breadth of school activities.
This study of the introduction of competition, or redundancy, to a particular
bureaucratic task has produced mixed findings. As expected, competition has led to a
general organizational search for activities that are most likely to increase the probability
of survival, at a minimum, and profitable enterprise beyond that. When uncovering the
actual range of activities pursued by organizations, however, the story takes on more
twists and turns. Party schools, in accordance with the benevolent intentions and
predictions that a proponent of bureaucratic redundancy would predict, invested in
updating training content. On the other hand, schools have also become entrepreneurial
actors in their own right and engaged in decidedly non-training activities. These
investments, while seemingly “off mission,” are school responses to the financial
190
pressures (and opportunities) in a world of decentralized funding. It remains difficult to
tell which set of activities is now more central to party school survival. Whereas the
updating of training content accords with political imperatives, entrepreneurial ventures
have the appeal of localized profit. Furthermore, in comparing the activities of party
schools across regions with varying levels of economic development, the most salient
variation appears to be across administrative levels rather than inter-regional. Schools at
the city levels and above, across inland and more prosperous coastal provinces, exhibited
a remarkable similarity in their embrace of market opportunities.
It is also noteworthy that processes of change affecting the party school system
have been additive in nature. Reforms in cadre training have been characterized by
layering on new organizations and actors rather than restructuring the management and
selection of bureaucratic personnel that takes place in party schools. What has occurred
thus far has been a rethinking of the incentives driving organizational decisions, not
necessarily a wholesale renovation of the bulky apparatus that is cadre training in China.
Competition was introduced under the rationale that it would improve the innovativeness
and adaptive capacity of existing key players, namely, the party schools. This in turn
suggests the degree of institutional inertia that exists in the system and the difficulty in
building new institutions as opposed to modifying existing ones (Genschel 1997; Hannan
and Freeman 1984). More dramatic institutional change is nonetheless possible, however,
given the many possible consequences of creating a competitive bureaucratic structure.
But the “politics of institutional change” are such that significant modifications are often
slow and hidden behind appeals to “restore tradition” or by gradual alterations to the
dominant ideology (Clemens and Cook 1999: 459). What has been observed here has
191
been an enlargement of the organizational space within which a set of tasks is carried out.
The function of party schools within the party apparatus remains unchanged, though the
content of the political messages that these schools convey has changed with the times.
This case study of the party school system also speaks to a larger debate regarding
the depth and scope of party adaptation to the consequences of economic reforms.
Whereas some have framed the CCP as an increasingly cloistered organization without
the capacity to manage new social and economic pressures, this project supports the
opposite conclusion.238 A party-led focus on building “administrative civilization” (Pieke
2009: 121) is in full swing. This project has specified how the construction of an
administrative civilization contains, beyond the ideological updating emphasized by
existing scholarship, organizational responses to new economic incentives and the
maintenance of the party’s institutions of elite selection.
The party has altered some core features of bureaucratic life under Mao, e.g.,
recruitment based on political credentials and mass campaigns, but this sea change has
not led to a discernable decrease in the internal stability of the party-state. With the
ascendance of the “second generation” of CCP leadership under Deng Xiaoping, the CCP
has displayed a certain degree of ideological flexibility in favor of pragmatic governance.
Writing early in the reform period, Harding (1981) debated whether overtures to
rationalize the bureaucracy were ephemeral or enduring. The verdict, based on the
238 See, for instance, Gordon Chang’s Coming Collapse of China (2001), which predicts that the CCP will not survive beyond this decade. Susan Shirk’s more recent and focused book on the insecurity of China’s leaders offers an analysis of both internal and external threats (Shirk 2006). In policy journals, Pei (2002) has noted that the CCP’s growing weakness lays in “the shrinkage of its organizational penetration, the erosion of its authority and appeal among the masses, and the breakdown of its internal discipline,” (p. 101). This dissertation provides evidence contrary to at least the latter two of these. Goldstone (1995) presents a neo-Malthusian argument, where population pressures, in combination with inadequate government capacity, will lead to significant political challenges to CCP rule. A discussion of the “pessimists” versus “optimists” on China’s political future can be found in Shambaugh (2009: Chapter 3).
192
evidence presented here, is in favor of the latter. The party school system, which has
long been an inner sanctum of ideological debate, illuminates the reach of efforts to
professionalize cadres and the pro-market form that party adaptation has taken. Whether
this marketization will extend to other areas of party management remains to be seen, and
the question also arises as to whether we might expect such a strategy carried out in other
authoritarian, single-party systems. The Chinese case, in which there has been party-led
marketization aimed at structures of the party itself, depends on necessary, but not
sufficient, pre-conditions such as a growth-oriented leadership that employed fiscal and
administrative incentives to effect a limited decentralization.
In brief, this study has explained how one enduring aspect of party management –
the functional role of a particular set of party organizations – persists alongside more
dynamic processes of internal adaptation. More simply put, the capacity for a
bureaucratic system to adapt to competitive pressures increases the likelihood that it will
continue to exist. In order to remain relevant in a new economic and social context, party
schools have adjusted to new market incentives by exhibiting such a capacity to adapt.
The processes underlying such organizational resilience are not immediately obvious
when focusing solely on formal bureaucratic structure or elite politics. This has been a
story of organizational change in response to shifting policy winds. Understanding
changing incentives and adaptation within this set of party organizations demonstrates
how seemingly anachronistic structures are able to persist and remain relevant in a
market context. While decentralization has been an important part of the story, central
193
authorities have driven these processes of change. As with previous realignments of
institutional and individual incentives, central party authorities led the charge.239
One aspect of party resilience that I have revealed in this study is the tension
between top-down, system-wide control and local school autonomy that exists within the
party school system. While party schools would seem an ideal site for centralized
dissemination of updated information on the party line and unifying beliefs throughout
the system, field observations at schools from the provincial to county levels were at odds
with this unifying function. Central authorities have allowed local party schools to create
their own training content and engage in independent, income-generating activities, and
these have added diversity to the training content received by local bureaucrats. These
organizational-level decisions indicate the tradeoff that central party leaders have made in
demanding, on the one hand, that sub-national party organizations impart content that is
consistent with central political goals but appropriate for local conditions while, on the
other hand, reducing direct funding for such change. Rather than centralizing the
financing of party schools and binding local party organizations more closely to central
mandates, the approach has been to retain a decentralized – and more flexible – structure
in the party school system. The party has opted to marshal market forces such as
competition and entrepreneurship to achieve the desired outcome of training content that
is locally relevant while still conveying central aspirations for China’s political and
economic development. The risk to this set-up is some loss of central control over the
end result in the localities. These choices, while risky, reflect a broader and significant
239 As noted in Chapter 4, these center-led and locally-grasped realignments have been studied in the context of the local state corporatism that characterized rural economic development, state entrepreneurialism within government ministries, and bureau-contracting in the public service units of the party and government.
194
party shift toward strategies that maintain the party’s wider relevance – and appeal – to
diverse local audiences. Party authorities are now interested in building a party that
represents and caters to the interests of a broad range of strategic segments of the
population, which now includes those who generate economic prosperity at the local
level.
Jiang Zemin, China’s third generation leader following on the heels of Deng
Xiaoping, emphasized that “whether a party and a state can develop excellent leaders to a
large extent determines the survival of the party and the state.”240 It would appear that
this is the core contribution of party schools and, more generally, institutions of cadre
training to the maintenance of CCP rule. In using party schools as channels for the
selection of bureaucrats – in effect, to control cadre careers – party authorities maintain
the capacity to shape the bureaucratic class. At the same time, the party has loosened
central control over these very same schools, releasing them to pursue market
opportunities and self-determine strategies for market competitiveness. In light of these
findings, I show how ruling parties are able to generate incentives for both institutional
continuity and adaptation. A system of competition-based redundancy safeguards against
system failure and generates the conditions for party entrepreneurialism. Significant
changes are taking place within the party as institutional incentives are adjusted and
reformed.
Zheng argues that “the CCP has become the major obstacle to state-building in
post-1949 China” (Zheng 1997: 255). This study suggests otherwise. Party authorities
have instead retained and reshaped organizational channels for selecting leaders of the
party and government. Existing party organizations have also been reformed to promote 240 People’s Daily, 9 June 2002.
195
new bureaucratic priorities, namely the importance of professionalized, globally-minded
public managers. As the preceding examination of Central Party School training content
has shown, cadres passing through party school training programs receive strong signals
that professional managerial skills is now a key criterion for promotion.
2 Preconditions
The emergence of party schools’ entrepreneurialism has depended on several
enabling conditions. The decentralization of fiscal and administrative responsibilities in
the party school system, detailed in Chapter 4, paved the way for schools to engage in
semi-independent activities for both pecuniary gain and to maintain party schools’
standing in a competitive training market. Beyond decentralization are other, more
broad-based changes that, in combination, were necessary but not sufficient conditions
for the changes observed. These critical phenomena include a growth-oriented leadership
and normative reorientation in favor of market reform.
As a backdrop to this story of party adaptation, it was imperative that the Chinese
leadership, particularly at the central level where key policy decisions originated, was
growth-oriented. Barring this, incentives to invest in the organizational innovation
observed in this study would be absent. If party authorities were instead focused more
exclusively on fidelity to the ideological goals of the Mao period rather than broad-based
economic growth, then there would not be the requirement that the party engineer a
bureaucratic transformation to match new economic goals. The emphasis on party and
government managers acquiring the professional skills to manage a particular set of
economic priorities arose along with a longer term, growth-enhancing outlook on the part
196
of party leadership. The bandit of the state had become stationary and taken an
encompassing interest in development (Olson 1993). Furthermore, this change in outlook
took the form of a commitment to shared growth, or growth that spilled beyond the
boundaries of the political elite. This more catholic dedication to general welfare was
mirrored in other single-party regimes of East Asia, many of which have also sought to
reform and professionalize their bureaucracies (Campos and Root 1996).
Crucially, China’s leaders were also inclined to pursue growth strategies through
pro-market reforms, which reflects a normative shift that has spilled beyond economic
organization and into the managerial logic of the party. The protracted process
culminating in the triumph of Deng’s reformist camp serves as some indication that there
was no inevitability to the market path. The rise of markets and unleashing of
consumption possibilities not possible under thirty years of Maoist rule paved the way for
party schools to “dive into the sea” of market opportunity. These activities granted
greater autonomy to local actors while potentially lightening some of the burden on local
government finance bureaus. The normative shift that this entailed – along with changing
expectations of bureaucratic competence – was critical for enlarging and shifting the
boundaries of the universe of feasible reform strategies.
While the priorities of the political leadership mattered, timing was also key. The
sequencing of reforms in the Chinese case and the decision to unbundle economic from
political reforms, in contrast to the Soviet and Eastern European cases, had implications
for the ability of the ruling party to control and respond to the unanticipated outcomes of
economic reforms before they led to wholescale destabilization of the entire political and
economic system. In this sense, I agree with Shirk’s (1993) analysis that a political logic
197
constrained but lent sufficient incentives for (possibly less efficient) economic reform
under CCP rule. Where my study departs from hers is in illuminating a different pathway
for policymaking. Policy outcomes may arise not only from elite, consensus-based
processes but rather through a more contentious, competition-driven process.
3 Risks
Marketization has led to reforms in the functioning – and priorities – of party
schools. What remains uncertain is whether the decision to introduce market
mechanisms might lead, inadvertently, to the unraveling of central party control over
cadre training. While the initiation of market processes hinged on central decisions, this
has the potential to unleash a train of events that might prove difficult to restrain. In the
reform period, there exist instances in which policy redirection has led to unanticipated
outcomes and subsequent central retrenchment. In the early reform period, for example,
the unleashing of local, collectively-owned enterprises led to explosive but unbridled
growth that central authorities sought to check through dramatic fiscal centralization
policies by the mid-1990s. This pattern has echoes in the political reforms shaping the
party school system. During the first decade of market reforms, schools expanded
rapidly their portfolios of non-training activities, particularly distance degree programs,
but central party authorities have recently attempted to rein in these practices while
leaving fundamentally intact initial pro-growth incentives. The decision to devolve
authority to local party schools thus speaks to this enduring tension between central
control and local interests.241 As with other policy areas, a balance has been struck. The
main risk to central authorities is whether the market, and its decentralizing tendencies, 241 See for instance the historical overview and thematic essays in Jia and Lin (1994).
198
will lead to an erosion of central party control over constituent parts. This might take
place on at least two fronts. First, there is the risk that party schools will displace central
goals in favor of their own, more local, interests. One implication is a reduction in local
party school responsiveness to central directives, though existing controls may ensures
some monitoring of local party school training content. Second, the partial nature of
marketization entails some risk that party school failure would hinder processes of party
personnel management, despite existing safeguards.
Choosing between profit and party service
Party schools are now striving to realize two distinct, though interrelated, goals.
First, as a link in the party’s transmission belt for sorting through cadres to promote to
higher office, schools must provide information on these individuals’ potential as well as
groom them for higher office. In carrying out this function, schools maintain a valued
place within the party’s universe of core institutions. Second, schools must supplement
this training work with additional ventures to remain financially viable. While these two
goals are complementary and, in many cases, can be satisfied simultaneously, it is unclear
what schools will choose when presented with starker choices. Distance degree programs,
inside- and outside-the-plan training classes attract income and, to varying degrees, fulfill
schools’ training mandate. Renting property and starting side businesses are more clearly
in the service of income generation. Given limited resources, it is unclear whether
schools will exhibit an increasing tendency over time to maximize the financial goal at
the cost of the functional training goal. While schools must strive to fulfill both, the issue
is one of relative resource allocation. The market has introduced choice to party school
199
decision makers, but each choice entails an opportunity cost. Bound up in the decision
making process is a calculation of the benefits in meeting central goals versus local needs,
and when these may be in conflict or agreement.
Under the Mao-era system of central transfers, the primary problem was one of
party schools shirking their duties as training institutions. Introducing competition may
have solved that problem to some degree, as schools are clearly updating their training
content in response to market incentives, but another problem has arisen in its place.
Market opportunities present more immediate rewards to school leaders. Gaps in the
monitoring capability of the state increase the attractiveness of diverting energy to more
lucrative options. School staffs may choose to comply minimally with the less profitable,
but more collectively important, goal (selection and training) and divert resources toward
activities that serve the more locally beneficial goal (entrepreneurial ventures). Even for
tasks where schools are most inclined to fulfill the goals of central and higher level
authorities, such as the implementation of planned training courses, they now have
incentives to curtail the allocation of resources to these activities in favor of pursuing
their own entrepreneurial endeavors. The result is the displacement of one goal in favor
of another. While there are monitoring mechanisms in place to keep the range of party
school activity within certain bounds, the localization of financial support eliminated
vertical ties linking center to locality.
Heightened tension between local organizational interests and broader party goals
is one outcome of the loosening of central controls over the party school system. This
tension will continue to exist so long as party schools possess some discretion over local
organizational decisions. Whether this discretion can be taken away in a recentralization
200
effort, however, remains an open question. While it is difficult to envision party
authorities closing the doors to the training market that they have now opened to such a
broad array of actors, this does remain one possibility, however remote. As long as local
party schools are minimally compliant with central goals, however, the two goals of
center and locality will both continue to drive, to varying degrees, party schools’
organizational choices.
Local embeddedness eroding central mandates
While some compromise between central and local goals may take place as party
school leaders decide how to allocate resources across party-mandated training programs
versus other revenue-raising activities, this calculus is affected by the nature of schools’
revenue sources. As a result of the partial marketization of cadre training, party schools
are now in direct contact with a greater variety of constituents, which represent a broader
swathe of the population, and these constituents have, to varying degrees, an interest in
school survival. The party now has a stronger street-level presence. This presence,
furthermore, is more subtle than the imposition of democratically-elected grassroots
councils. Party schools are deeply embedded within the party apparatus, but they are
simultaneously building a locally relevant presence in response to competitive pressures.
This broadening of the party’s support base reflects a shift toward “inclusion”, whereby
“a ruling Leninist party’s perception that the major condition for its continued
development as an institutionalized charismatic organization is to integrate itself with,
rather than insulate itself from, its host society,” (Jowitt 1975: 72).
201
Post-Mao changes to the party school system have reverberated beyond the party,
to communities in which party schools are located. The entrepreneurial spirit taking hold
of party schools now generates incentives for localities to support party school
development. Schools are no longer closed bastions of party learning, where only party
members are permitted entry. Instead, school leaders are actively pursuing collaborations
with all manner of local partners. Beyond their conventional audience of cadre trainees,
party schools are now serving a more general audience. Everyday citizens have
incentives to see party schools strengthen and grow within their local economies. The
benefits to the party of a competitive training market parallel the benefits to local
consumers of increased competition in the provision of local services and products. One
advantage to the particular development of party school ventures is more general public
support for the expansion of these party organizations and, by extension, the continued
presence of the party as a local economic actor.
From this perspective, party schools are no longer mysterious party organizations
set apart from the fabric of local economies. Schools are now embedded in these
economies and providers of valuable services. They now have an interest in local
development. No longer can schools rely on a steady stream of income from cadre-
students assigned to periodic training courses. The demands of local groups and local
consumers now drive to some degree the activity of local party schools. In this emerging
dynamic, party schools are more accountable to and mutually dependent upon local non-
party actors. There exists, in this arrangement, the potential that schools will be captured
by local interests, particularly if schools are increasingly dependent on locally-derived
income to bridge the gap between government transfers and operating expenses.
202
This local embeddedness may tip the scales in favor of local interests, at the risk
of eclipsing the mandates of central authorities. With the changes prompted by
marketization – and the beguiling income-earning opportunities therein – party schools
now have added incentives to settle central-local tensions in favor of the local. School
leaders possess the means to be more responsive to local demands and circumstances,
given the autonomy generated by decentralization and marketization. Taking this logic to
its extreme, there is the risk that income generation through local entrepreneurial ventures
displaces the original purpose of party schools as core sites of cadre training. And if this
were the case, there exists the risk of central authorities losing control over the activities
and goals of party schools. Studies of organizational behavior and examinations of
central-local bureaucratic relations in the U.S., for example, have found that the more
dependent a local public agency on the local environment for resources, the greater the
relative importance of local influences on that agency’s decision-making (Pfeffer and
Salancik 1978; Scholz, Twombly, and Headrick 1991; Whitford 2002). One prescription
for coping with this potential agency capture by local interests is through stronger
institutions of accountability within the party or between the party and the public, but at
present reform efforts remain weak (Bardhan 2009).
The risk of failure
Exposing party schools to market forces has created winners and losers.
Marketization, by definition, implies a process whereby actors become entrepreneurial
and are subjected to risks (Hebert and Link 1988). While some schools, through a
combination of leadership, local economic conditions, and strategic partnerships, have
203
prospered in the training market, others have become hollow institutions. This is not
necessarily a surprising outcome, as agency failure is a recurrent theme in democratic
settings (Lewis 2002). In the Chinese case, significant reorganization of government
ministries has occurred periodically in the post-Mao period and entailed the eradication
of entire ministries.242 If party schools were to fail not by decree but through their own
inability to compete, the question arises as to whether party authorities are prepared for
other institutions to fulfill party schools’ role in elite selection. While this is the intention
behind the creation of a redundant system, an assessment of actual conditions and
potential outcomes reveals some departure from theory.
One key motivation for introducing redundancy to a bureaucracy is to safeguard
against a functional vacuum should one component of the system fail. Competition in
this context presents a form of reserve capacity. The training market abounds with
substitutes for party schools, but all of these are partial replacements. Other party
organizations, such as cadre leadership academies, may step into the space left by
uncompetitive party schools. This, however, raises an additional design flaw in the
emerging competition for cadre training contracts. The rise of organization department-
managed training academies, in combination with the powerful position already held by
organization departments in personnel matters, has the potential to concentrate further,
rather than diffuse, control over personnel management. This creates a further risk:
organization departments might become an even larger player in party management and
repeat, though on an even bigger scale, the descent into stagnation feared so much by
242 Prominent recent examples include the 1997 dismantling of 11 ministries and creation of high-level economic planning commissions and the 2003 merging of commissions and subsequent creation of a single economic planning and development entity. In personnel management, 2008 saw the creation of a “super-ministry” of human resources and social security, which replaced three state bureaus.
204
party authorities at the outset of the cadre training marketization project. In addition to
organization departments’ control over hiring, promotion, and dismissal of cadres, the
addition of control over training would consolidate personnel management and seem to
be at cross purposes with the original intent of the reforms.
Another possibility is the substitution of universities for party schools. In the
credentialing wave that has washed over China’s political elite, it would seem that
universities are well-positioned to become channels for elite selection. Universities are
sites for the recruitment of party members, and it may be a natural progression for
university credentials to replace party school training certificates. The appeal of
universities is further complemented by the Deng-era initiative to professionalize the
cadre class. Secular public administration and management programs are poised to
contribute to this trend. The risk this entails to the party, however, is a loss of control
over the political content that it still considers a critical part of the cadre training process.
That cadre training has this homogenizing effect, namely, imparting expertise to those
lacking in professional skills and enhancing the “socialist virtues” of technical experts
cannot be dismissed (Pieke 2009a: 142).
Among these two possible substitutes, then, there is the risk of greater
concentration on the one hand and too great a loss of political control on the other.
Obtaining data on the characteristics of trainees at organization department academies
and university management schools, if available, could speak to this issue. However,
data unavailability and the relatively recent entry of cadre leadership academies prevents
the assessment of longer term trends. Even without comparative data on these
alternatives, the party school system at present remains the most comprehensive, unified
205
system for cadre training. As such, it is positioned to continue serving as an institution of
party elite selection. Creating an alternate system with the breadth of party schools
would be a resource-intensive undertaking. This would be further complicated by the
difficulties in uprooting an entrenched bureaucracy within the party bureaucracy, one
vested with symbolic importance and a reminder of the ideological underpinnings of CCP
rule.
For at least two reasons, it appears that odds are in favor of the party school
system’s survival. First, marketization has been partial. Local party committees continue
to be responsible for the allocation, each year, of a portion of the total training classes
organized in a locale. I have not yet encountered the case of a party school receiving no
inside-the-plan training contracts. This is further buttressed by the center’s prerogative to
allocate training contracts to party schools in general. The implementation of a one-time,
five-year county magistrate training program at the central party school and cadre
executive academies is one illustration of this. This was also the case in the centrally-
mandated rollout of 17th party congress study sessions throughout the party school system.
Central control and interference provides a hedge against the possibility that party
schools will fade away in the presence of more well-endowed and globally-connected
competitors. That the center retains this coercive capacity further supports the argument
that redundancy in this Chinese context is primarily about adaptation and secondarily
about reliability.
Yet the past is not necessarily a predictor of the future in determining whether
party schools will continue to reinvent themselves or give way to the competitive
substitutes already chipping away at party school market share. During the first three
206
decades of reform, party school leaders have managed, on balance, to weather through
one organizational challenge after another. On one end of the spectrum are universities
with their global connections and professional management schools, and on the other end
are the organization department academies which may be equally, if not more, privy to
the inner sanctums of party power. Universities may rate the highest in terms of
providing the quality education that the party, and society more generally, desires on the
march toward modernity, but they are without the political substance to be found in those
training organizations embedded within the party itself. This presents a conundrum
which, for the time being, political authorities have been able to skirt.
I stress that the marketization of party schools is a dynamic process, one that in
theory must maintain some stability in process and function in order to remain effective.
At the same time, this stability might be derailed by the very competition which has now
become a driver of organizational adaptation. Competition, and the adaptation it
promotes, requires multiple players, and the system is, counterintuitively, stable so long
as there are multiple actors who eventually reach a division of labor, a division of the
market, or some other “steady state” in which interference from central authorities are
minimal. Should the system exhibit a tendency toward monopoly, then the benefits
derived from competition decrease accordingly. In comparative case studies of redundant
and nonredundant urban public transit systems in the US, for example, Bendor found
stable arrangements so long as political authorities did not demand an end to the
competition and organizational diversity was allowed to persist (Bendor 1985: 238-241).
207
4 Limits to party reforms
While opening up party schools to markets and competition may generate changes
in training content and embed party organizations in local economies, there are
significant limitations to the reach of these reforms. The marketization of party schools
does not create the mechanisms for coping with a particularly thorny aspect of
institutional design in China – the absence of strong bureaucratic accountability. By
accountability, I am referring to the mechanisms in place to ensure that the behavior of
organizations and individuals may be constrained and checked by the public and/or other
governmental bodies.243 While party schools are now punished if they fail to tailor
training content to market and party demands, this checks the performance of schools and
not the behavior of cadres themselves. In this sense, this study does not speak to the
ability of party authorities to reduce the bureaucratic corruption which is drawing both
domestic and international ire and which, as Lu Xiaobo argues, may be bound up in the
historical evolution and organization of the CCP itself (Lu 2000; Pei 2008; Wedeman
2004). While it remains an open question as to whether the CCP’s efforts to curb
corruption are to any measurable extent successful, it is noteworthy that China is now in
the “middle of the pack” on international corruption indices (Wedeman 2010).
Corruption may thus be a “growing pain” that the ruling party must confront in its rapid
developmental trajectory rather than a “crippling disease” (Ibid., p. 118).
Subjecting party schools to competition sends a signal that schools must
strengthen their contribution to the particular dimension of the state-building project
243 Rather than focus on the vertical and horizontal forms of accountability that are the conventional focus of research on democracies, Dimitrov (2010) theorizes that another dimension to accountability exists in the Chinese case: “proxy” accountability between citizens, local officials, and the central government, whereby the center punishes local officials who are unresponsive to citizen complaints.
208
within their domain. This is not the same as requiring schools to complement, in some
fashion, the still weak monitoring organizations of the party-state. In theory, party
schools could contribute to the monitoring of cadre behavior, constituting a form of
horizontal accountability within the party. This does happen to some degree in the
monitoring of cadre behavior (biaoxian) during training classes, but such monitoring has
little bearing on the range of activities that fall under official corruption. In terms of
vertical accountability, the marketization of the party school system detailed in this
project does not necessarily have any effect on the accountability of either these party
organizations or bureaucrats to a broader public. Marketization is instead about
improving the efficiency of cadre training itself, which has as one goal the creation of a
more competent bureaucratic class, broadly defined. Innovations in training do not
generate the incentives for political elites to refrain from seeking personal gain at the
expense of general welfare. A much denser network of administrative measures must be
in place for cadre training to contribute to the more normative goal of an accountable,
self-regulating bureaucracy. At present, this is lacking in the party and government
apparatus. This project is thus silent with respect to whether party entrepreneurialism can
address a more encompassing vision of sound governance.
A separate but related issue is the problem of individuals completely bypassing
legal, organized routes for obtaining office. Official selection processes are not the only
path to high office in China. One worrying phenomenon is the sale and purchase of
public office. There have been some public disclosures of such occurrences in the
Chinese media, but the true extent of this practice is unknown.244 While party schools do
244 Newspaper reports of this practice have detailed, for example, the sale in 2006 of eight leading cadre offices (three at the city-level (tingji) and five at the department level (chuji)) in Hunan and 110 county-
209
appear to play a role in cadres’ career advancement, it is beyond the scope of this study to
determine the relative reach of less formal routes for obtaining office. That there exist
these extra-organizational channels for obtaining office erodes the party’s
institutionalization efforts and presents a serious threat to organizational integrity.
Third and finally, this study is limited to unraveling the development and
consequences of competitive pressures in the Chinese bureaucracy, and it does not
address more evaluative questions regarding the efficiency of such redundancy.
Importing market principles to bureaucratic functions may introduce an internal
opportunity cost in that the efficiency gains from competition may still be less than the
loss associated with devoting resources to similar (or redundant) tasks. Instead, resources
could be applied more broadly to different bureaucratic functions. This critique, however,
presumes the ability to measure these costs and benefits when in fact they may be
difficult to compare either quantitatively or qualitatively. Perhaps this concern is also
less relevant in an authoritarian context where there exist no institutionalized channels for
outside parties to express dissatisfaction with organizational decisions or obtain
information on resource allocation. Even more, this overlooks the dual motivations for
introducing competition. In addition to efficiency considerations, there is the relative
importance of devising strong(er) incentives for the organizational innovation that
supports adaptive practices. In the end, efficiency considerations are difficult to evaluate
level offices in Anhui. See “Loudi city, Hunan, party standing committee’s Xie Wensheng is arrested for selling office,” 21st Century Business Herald, 22 Apr 2008, and “Anhui county party secretary receives largest bribe amount from selling office,” China Youth Daily, 10 Dec 2007. In response to this practice, the Central Commission for Discipline and Inspection has warned that selling and buying office will result in dismissal, Xinhua, 20 Oct 2008, http://news.sina.com.cn/c/2008-10-20/182816489732.shtml, accessed 22 Feb 2010.
210
and it is doubtful that a comparison of the party school system before and after market
reforms, notwithstanding data limitations, is unlikely to yield a clear verdict on this issue.
5 Whither the party?
Finally, we turn to questions of the party’s evolution and continuing relevance in
the post-Mao period. Uncovering recent developments in cadre training is significant for
understanding several broader transformations in the priorities and outlook of CCP
leaders. State-building goals have evolved from the Maoist focus on creating a utopian,
collectivist system in which the party enjoyed a monopoly over all material production
and social mobility. That totalitarian vision of the party-state has given way to one where
the uneven withdrawal of the state from social and economic realms is granted in
exchange for the party’s unrivaled political hegemony. In formulating and pursuing this
new developmental path, the CCP is now committed to an experimentation-based
approach in which local inequalities are tolerated as a byproduct of the tide of prosperity
that will eventually lift all boats.
In this policy context, local party organs now have strong incentives to be drivers
of change, derive benefits from market opportunities, and open up to new influences.
The boundaries around these local activities are drawn at the point where they do not
undermine in any overt way the legitimacy of the ruling party. These organization-level
changes show how the party is acting upon its resolution to become a “learning party”.245
245 Documents issued during the 16th Party Congress (in 2002) reflect the party leadership’s focus on becoming a “learning party” (xuexi xing zhengdang) and grasping new global and national trends. See “Resolution on the Strengthening of the Establishment of the Party’s Ruling Capacity,” available online at http://www.people.com.cn/GB/40531/40746/2994977.html, accessed 13 September 2009. This initiative has spilled into the work of the 17th Party Congress. See, for example, the official interpretation, dated 17 November 2009, of the decision passed down at the Fourth Plenary Session of the Seventeenth Party Congress on “How to Build a Learning Party,” available at http://www.gov.cn/jrzg/2009-
211
In an assessment of the party’s trajectory over time, party authorities and scholars have
presented “two transformations” as the catchphrase to describe the party’s transformation.
The first shift entailed the CCP transforming from a revolutionary to a ruling party. More
recently, it has morphed from a party directing the planned economy to a party leading
the processes of reform and opening up. This latter transformation has entailed a
capacity to manage the complexities of a market economy (Shi 2006: 92-93). In order to
carry out this second transformation, Chinese leaders have chosen to position the party as
open and willing to learn from different models of public management.
Understanding party adaptation, then, calls for understanding the institutional
means by which the party learns. Party authorities have invested strategically in building
adaptive capacity. The emphasis on cadre training reflects this new (self-given) identity
that the party seeks to embody and the concrete objectives that accompany it. Not only is
this “learning party” rhetoric evident in speeches by high leaders such as Hu Jintao, but
CCP investment in cadre training reflects a real commitment of resources. It is within the
bureaucratic class that the learning capacity of the party becomes manifest in policy
decisions and implementation on the ground. That party-led bureaucratic modernization
falls under the general category of activities that a learning party engages in will only
buttress party control over the reinvention of the Chinese state.
11/17/content_1466399.htm, accessed 12 February 2010. This document lays out general directions such as “first, turning study into a political responsibility” and “second, integrating the study of party theory, professional skills, and all types of new knowledge”. This is related to an earlier interpretation in which “reform and innovation” (gaige chuangxin) were the watchwords. See also the related interpretation, dated 1 November 2009, of the decision passed down at the Fourth Plenary Session of the Seventeenth Party Congress on “Party building must persist in reform and innovation,” available at http://www.gov.cn/jrzg/2009-11/01/content_1453837.htm, accessed 12 February 2010. One academic book, published in China, which elaborates upon the priorities of this “learning party” is Xie Chunhong’s Research on Contemporary CCP’s Building of a Learning Party-State (2009), which argues that “not to study is to retreat or die” (bu xue z etui, bu xue ze wang).
212
Second, the record of cadre training over the past three decades reveals how the
party is seeking new pathways for building the administrative capacity to manage a new
market-based economic and social reality. Above all, the CCP has had to adjust its
coping strategies to accommodate the uncertainties, pressures, and incentives of the
market. The party’s fate is now inextricably linked to the market. Marketizing cadre
training is a means to shore up the legitimacy of the CCP to diverse audiences, from the
cadre ranks to the communities where party schools are located. At the same time,
bringing market principles into party organization is part of the long, uneven reform
process of legitimizing market practices.
Accordingly, cadres must not only partake in all the rituals of party-building and
study the theoretical underpinnings of socialism with Chinese characteristics, but they
should be exposed to developmental models near and far. Case studies in governance
and lessons on the model economic projects throughout China all reflect the party’s
attempts to build a “managerial class” (Burnham 1962) more befitting China’s current
circumstances, where decollectivization, SOE restructuring, national markets and global
integration are the new imperatives of the day.246 Investing in administrative capacity is
one way to forestall the dangers of stretching scarce administrative talent too thinly over
many reform fronts (Rawski 1994: 274-5). The state administrative capacity that is under
construction, however, is not necessarily one that will seep evenly throughout the
political system. As this study has indicated, marketization, while generating strong
pressures and incentives, lacks the tidiness of a more centralized effort to engage in state
246 This transformation was observed by Brzezinski in the late 1980s, who wrote that the CCP is “becoming less characteristic of a revolutionary party claiming to be the representative of the dictatorship of the proletariat and more that of a modernizing party of the dictatorship of China’s emerging state-sponsored commercial class,” (Brzezinski 1989: 147).
213
capacity building. The very lowest levels of administration may not share in the fruits of
these efforts, indicating limits to the scope of the overall project.
As part of these shifts in identity and capacity, party authorities have striven to
avoid the mistakes of other single-party regimes and also learn from domestic policy
stumbles. While there have not been pronouncements as precise as the CCP’s decision
that Mao was incorrect 30 percent of the time (Spence 1999: 643), a pattern of openness
and retrenchment (fang/shou) has characterized the experimental, at times tentative
nature of the post-Mao reforms (Baum 1996: Chapter 1). Looking abroad, CCP scholars
and policymakers have noted the inability of the CPSU to combat deep economic
stagnation, contain centrifugal nationalisms, rein in rent-seeking bureaucrats, and resolve
myriad other domestic and international quandaries. Within the CCP, the most direct
avenue for prompting awareness and discussion of these issues in an organized yet
flexible way was through tried-and-true channels for allowing decision makers to discuss,
analyze, reflect, and ultimately apply key lessons. The party school system was and is
one such channel. The newest development in the party’s organizational evolution is that
there is now greater choice, beyond party schools, where this learning may be staged. In
short, this study accords with Shambaugh’s characterization of the current CCP as a party
that is adapting rather than withering away.
This study of CCP-led reforms also refines those arguments in the comparative
literature that seek to explain the resilience of the CCP in comparison to the unraveling of
communist party rule elsewhere. Solnick (1996; 1998) rightly points out that reforms in
China generated incentives for local actors to remain loyal to a CCP-ruled system, but his
focus on the monitoring capabilities of the party-state apparatus overlooks the crucial role
214
that the selection and training of bureaucrats played in the Chinese case. While
complementary to monitoring, selection is a prior concern. Furthermore, Solnick credits
the CCP’s avoidance of internal restructuring as a factor for regime stability, but I would
suggest that the introduction of market-based competition has been the driver of
significant changes which ultimately support the construction of an enduring party-state.
Counter intuitively, this resilience is a consequence of deliberate organizational
fragmentation rather than center-led consolidation. Whereas the foundations of stability
under the ancien regime of the USSR and satellite states of the Eastern Bloc were in the
party’s control over political, economic, and social resources, ironically the stability
achieved in one key aspect of party organization in China has been through a deliberate
relaxation of the party’s monopoly. In this particular realm, the party has moved beyond
a reflexive opacity and embraced, cautiously, a limited market-driven openness.
In other respects, pressures for institutional change in China are similar to elite
recruitment in the CPSU under Brezhnev’s leadership. In a study of elite selection and
socialization during this period, Klugman (1989) identifies pressures for new selection
criteria: “Somewhat more autonomy will be required to function efficiently in a more
loosely connected system, while increased discipline will require a more internalized
form of self-control” (p. 13). Comparable changes are taking place in reform-era China,
where higher level authorities use cadre training to instill and observe the “internalized
controls” which assist higher levels in coping with the increasing autonomy enjoyed by
sub-national bureaucrats.
These considerations of trends in CCP rule speak to broader issues in comparative
studies of authoritarian party resilience and breakdown. Rather than focus on the array of
215
internal party factions and cleavages that may affect party cohesion, this study has
dwelled on organization-level determinants.247 Organizational characteristics, in
particular the amount of competition and flexibility built into a bureaucracy, explain
overall bureaucratic performance to a greater degree than individual-level characteristics
such as values and norms (Punyaratabandhu-Bhakdi 1983). Organizational strengthening
in which flexibility is built in to facilitate the dissemination and support of broad
developmental goals moves the system away from breakdown. While a complete theory
of regime survival must focus on both elites and organizations as drivers of political
stability or instability, this study has focused on those observable structures of party
organization, as they relate to elite management, rather than attempt to penetrate the more
opaque realm of individual and collective preferences.
That the CCP remains the key political actor in the PRC today begs the bigger
question, how has the party remained in command? Pieke points to the pursuit of a “neo-
socialist” project within the party. This bundle of concepts and real policy priorities
features a strong bureaucratic apparatus, continued market creation, and, above all, the
maintenance of single party rule. This present study takes the latter factor in the neo-
socialist project, maintenance of single party rule, as the starting point for central
decision making in China. In this sense, a “political logic” (Shirk 1993) trumps the
economic or social. That this political logic has led to the marketization – and
rejuvenation – of Leninist party organizations rather than their quiet expiration attests to
the resilience of the party’s political project more generally. At the same time, I note that
247 The presence of or potential for elite disunity is presented as negatively correlated with regime strength, though studies in this tradition do not offer an analytical framework for knowing ex ante when a given split might prove serious for regime survival or be symptom of factional politics as usual. See, for example, O’Donnell and Schmitter (1986) on hardliner versus softliner splits and, on elite divisions in Chinese politics, Nathan (1973) on factions and Tang (1976) on informal groups.
216
transformations in party outlook are attempts – that outcomes are far from determined.
The adoption of a “learning party” identity is as yet in its early stages and incomplete, the
building of market-embracing party-state structures is progressing in the now-familiar
manner of crossing a river of uncertain waters.
In some respects, the substantive changes that have taken place in the party school
system reflect the broader shift from a “plan ideological” system toward a “plan rational”
mode of organization, though one that remains fundamentally political rather than
economic in orientation (Johnson 1982: 18-24). In China, the transformation to a “plan
rational” system remains incomplete. As this case study of the party school system
reveals, shifts in the incentive structures driving organizational behavior within the party
now reflect to a greater degree the intention to meet particular developmental goals rather
than the setting of plans and goals as ends in themselves. However, party authorities
retain the prerogative to tighten or relax their control over local organizations as they see
fit. Overarching political goals, as envisioned through an increasingly “plan rational”
lens, are several: maintaining the primacy of party control over bureaucratic selection,
creating an updated administrative and political elite. Because the emphasis is on the
overall effectiveness of party rule, rather than efficiency, developments may be rescinded
at any time.
6 Limitations and new directions
The initial spark for this project was a narrow question on the current state of
party organization. Over the course of exploring the fate of party schools, the
marketization of cadre training, and resultant party entrepreneurialism, many questions
217
remain unanswered and beg further investigation. It would also be remiss not to report
shortfalls in this present study, though, on an optimistic note, some of these limitations
may transform into new research projects.
Nonrandom site selection
As detailed in Appendix B, the site selection process for my field visits and
interviews was nonrandom. This reflects the realities and limits of in-depth field work as
a research strategy in China. Due to the sensitivity of this research topic, I relied on local
academic partners to introduce me to colleagues in party schools at various locales, and
from these networks I was able to gain entry to the world of cadre training. I hope future
studies might improve upon this rather idiosyncratic field research plan, but until there is
reliable and open access to party organizations in China, I will report my findings with
these caveats.
One saving grace may be to compare findings across studies. For example, the
more economically-advanced regions discussed in this study may complement findings
from research conducted in poorer areas (e.g., Yunnan province in Pieke 2009). I have
sought, whenever possible, to corroborate the findings from one site or assertions made in
interviews with comparable information obtained at other sites or in public documents.
This verification has, however, been most difficult with respect to financial information
revealed during interviews.
Leading factions and inter-bureaucratic politics
The marketization of cadre training and its effect on the party school system has
been to some degree a story of how sub-party organizations have confronted the
218
challenges and opportunities that arose out of their shifting political and economic
context. There remains a question as to the extent that inter-agency rivalries, specifically
between the party school system and the party’s organization department, drove the
decision to bring competition to the party school system. It is uncertain, for example,
how powerful a voice the CPS leadership has had within the most inner circles of party
authority, how this has changed over time, and whether this has varied with the particular
groups that have dominated successive administrations. The present study has drawn
heavily on official documents and sub-national interviews to piece together the story of
party school reform and adaptation. A fuller telling of this story would include the tracing
of the relationship between these two party agencies over time, the specific central
leaders involved, the transfer of personnel between these two bureaucracies, and an
assessment of the relative rise and fall of each bureaucratic system (xitong).
Local party school content analysis
Examination of changes in party school training content is limited to materials
collected from the Central Party School. I have attempted to draw on trends across some
local party school training syllabi and information provided by interviewees. However,
gathering documents older than two years, much less five or even twenty years, from
local schools has proven daunting. As such, this study has not systematically tracked,
over time, how local schools may have diverged from or mimicked central content.
Demand side of training
This study has focused on the suppliers of cadre training, and less consideration
has been given to the demand side of the training market. In interviews, there have been
219
hints that cadre training is becoming increasingly consumer-oriented. The most
progressive schools in model economic regions such as Shenzhen and Zhejiang are all
experimenting with training models where the preferences of cadre-students are taken
seriously. Interviewees at schools in both coastal and inland provinces A and B also
noted that shifts in the content of training classes were a response to cadre input.
However, cadre trainees are not the only consumers with the power to drive new
directions in training content. As local governments and party organs issue public calls
for training proposals and widen their searches to international providers, they will also
push training format and content in new directions. This study has not probed the process
by which local party organs and state ministries, as consumers of training content, set
their own particular training goals, choose from a menu of possible providers, and how
higher level authorities may influence that process. While it may be the case that the
center is the player with the most influence over the general tendencies in cadre training,
this may vary at different administrative levels in the system and across functional areas
of the bureaucracy.
Corporate culture and ideology
Further research is necessary to uncover the implications of party marketization
for the corporate culture that shapes elite decision making in China.248 Pursuing this
avenue of research would require delving into individual-level beliefs and perceptions
that are beyond the scope of this study. Insofar as party organizations are governed by
and contribute to the perpetuation of a particular corporate culture, it would appear that
marketization has the potential to undermine unification efforts. The new emphasis on 248 Kreps (1990) defines “corporate culture” as those principles that guide an organization’s actions, including the conveyance of those principles throughout a hierarchical system.
220
the professionalization of cadres and the waning of campaigns as tools of political control
would also seem to chip away at the strong corporate ties which bind together and unify
the ruling elite. Ironically, a clearly defined corporate culture is particularly salient
during China’s reform period because it establishes the standards for action in times of
uncertainty (Kreps 1990).
It is impossible to investigate the persistence and continued relevance of an elite
corporate culture in contemporary China without also probing the fate of one of the
reform era’s most significant casualties – orthodox Marxism-Leninism and Maoist
ideology. But what of ideology more generally as “a verbal image of the good society
and of the chief means of constructing such a society”?249 What place does ideology
have among political elites, in the party – and hence China – today? How do party
authorities determine who the believers are, and what beliefs are truly sacred? These are
only a few of the questions which might motivate an investigation into this labyrinth of
related issues. One starting point might lay in the assertion that ideology, as a “shared
mental model” improves group cohesion (Denzau and North 1994), but might this shared
model be distorted beyond recognition after constant reshaping and retooling in its march
through disparate locales? In the Mao period, the party school monopoly on cadre
training, in tandem with intense political campaigns, maintained the cohesiveness of
China’s political elite and socialized cadres to the changing practices and vocabulary of
party leadership. Additional fieldwork and even survey-based research can unravel how
these various forces – elite socialization, the maintenance of a party corporate culture,
and the dissemination of a guiding ideology – have fared in a marketized context.
249 See Downs (1957), p. 96.
221
Areas for comparative study
It is unclear how far the logic of market-based competition and party
entrepreneurialism has spread within the party bureaucracy, though its implications in
terms of organizational adaptation have bearing for virtually all aspects of party rule. At
its broadest, competitive redundancies could improve the state’s capacity for skillful and
effective policymaking. While most studies of redundancy have been limited to
democratic systems, there is no theoretical reason barring observation of the principle in
authoritarian contexts. Furthermore, if redundancy is a means to reinforce the overall
survivability of a system, then Chinese party authorities should apply it liberally and even
more readily in those areas deemed most critical to the party’s continued existence.
Some promising work has begun to emerge that supports this logic. Scholars have begun
to probe the processes and consequences of marketization in other party xitong, notably
the propaganda apparatus. Shirk (2007) considers the implications of the party’s
introduction of commercial incentives to media outlets, and others have argued that
content is still manipulated to maintain popular support for the CCP’s leadership
(Stockmann and Gallagher 2009). Other potential areas of research include effect of
competition on policymaking processes in environmental protection, the military, and
public goods provision more generally. Whether market forces have a subversive effect
or reinforce the regime’s legitimacy in what were once the most inner realms of the party
promises a more encompassing understanding of the reality that moderates and informs
theoretical possibilities.
222
In closing, I return to the underlying logic of organizational change observed
within the party itself. There exists a tension between the party’s loosening of
organizational controls in order to maintain a firm grip on the political elite that guide
Chinese state and society. Counter intuitively, central party authorities have reinforced
control over political elites by introducing competition and market forces to inner party
organizations. This study has shown how, within the party, there are now greater “realms
of freedom” (Oi 2004). It has also highlighted the party’s inner struggle to balance the
various centrifugal forces that have arisen out of the protracted transition to a market
economy. One critical development has been an increase in the autonomy of local
bureaucratic agents of the party-state. Local experimentation and innovation are one
positive outcome in the trade-off between central direction and local autonomy. While
the center has continued to dictate general policy goals, the means for achieving these
have been left to local agents that must now search more aggressively for viable solutions
or else lose ground to rivals. Limited decentralization and increases in the diversity of
local activities as well as the number of market players implies that the party is serious
about fostering the conditions for adaptive change. Party entrepreneurialism presents
further variation on the “local experiment to national policy” theme that has driven so
much change in modern China (Heilmann 2008: 25). It also presents a new twist on this
framework in that there is not necessarily the intention to unify local agents after a period
of experimentation. Diversity may be here to stay. The variation in organization and
training content observed across the party school system may be the intended outcome
rather than national unification.
223
Appendix A: Number of party schools, by locale and national share of leading cadres
Administrative Unit Number of Party Schools
Percent of total party schools in the country
Share of total leading cadres in the country, 1985
(percent)
Share of total leading cadres in the country, 1998
(percent)
Beijing 55 2.00 3.05 3.15 Tianjin 31 1.13 2.85 2.51 Hebei 156 5.67 4.88 4.47 Shanxi 117 4.25 3.95 2.98 Inner Mongolia 90 3.27 3.05 2.43 Liaoning 134 4.87 4.10 6.91 Jilin 77 2.80 3.19 3.35 Heilongjiang 147 5.34 4.95 4.43 Shanghai 77 2.80 2.90 2.97 Jiangsu 84 3.05 3.34 4.03 Zhejiang 94 3.41 2.11 3.38 Anhui 82 2.98 3.67 3.08 Fujian 54 1.96 2.17 2.69 Jiangxi 90 3.27 2.65 2.41 Shandong 126 4.58 3.94 6.49 Henan 137 4.98 4.55 4.04 Hubei 131 4.76 5.04 4.89 Hunan 124 4.50 4.81 4.30 Guangdong 127 4.61 3.85 5.54 Guangxi 95 3.45 3.91 3.10 Sichuan 233 8.46 5.64 6.05 Guizhou 74 2.69 2.48 2.15 Yunnan 121 4.40 3.00 2.99 Tibet 8 0.29 1.47 0.93 Shaanxi 95 3.45 4.11 3.52 Gansu 84 3.05 2.80 2.41 Qinghai 40 1.45 1.14 1.15 Ningxia 14 0.51 1.05 0.97 Xinjiang 56 2.03 5.35 2.69
Total 2753 100.00 100.00 100.00 Sources: COD 1999; 1985 DXNJ, numbers include provincial, municipal/prefectural, county, and central ministry party schools.
224
Appendix B: Note on sources and research methods
This dissertation draws on three major data sources: interviews, documentary
sources, and survey data. Over the course of 13 months of fieldwork in China from 2006
to 2008, I conducted 236 interviews with party and state officials on the topic of cadre
training. To understand organizational change at various administrative levels and across
locales with very different economies, interviews were focused in Beijing (central
government), a coastal Province A, an inland Province B, and a Special Economic Zone
(SEZ). Interview sites included party schools, administrative institutes, socialism
institutes, Communist Youth League schools, universities, and party and government
organs from the central to township administrative levels. Basic demographic
information about interviewees is summarized in the tables below. To protect the identity
of Chinese interviewees, I have cited interviews by interview number and year. Some
contacts agreed to be interviewed more than once to answer follow-up questions. When
appropriate, I have also noted the occupation or station of the interviewee. To compare
across sub-provincial localities, I strove to conduct interviews at the provincial capitals of
Provinces A and B and at least one city-level jurisdiction and at least one county-level
jurisdiction in each province.
During the early stages of this project, I conducted preliminary research at sites in
two coastal and two inland provinces. These initial visits were intended to assess the
quality of access to schools and officials. I ultimately focused on two provinces, one from
each region, and the decision was driven by practical data collection considerations.
There is, however, some basis for the selection of the two provinces which are the subject
of case studies in Chapters 4 and 5. Both provinces are average with respect to the sizes
225
of their general and leading cadre populations (Figures 1 to 6 below). However, they are
both economic high performers within their regions, which may bias the findings
presented here in favor of a more optimistic general assessment. Cities and counties
within each province were also selected nonrandomly.
In addition to field interviews, I collected published and unpublished documents
during site visits, library searches in China, and from government internet sites. When
visiting party schools, for example, I asked interviewees for copies of training syllabi and
training materials. Additional data was available online. Central Party School yearbooks,
Central Organization Department publications, newspaper articles in Chinese and English,
and online biographies were all valuable resources for constructing an understanding of
the institutional lay of the land, system wide changes, and local experiences. Mainland
libraries consulted include the National Library of China, the libraries of Tsinghua and
Peking universities, and the University Services Centre Library of the Chinese University
of Hong Kong. In constructing career histories of Central Party School alumni, I relied
most heavily on publicly-available official biographies.
Finally, statistical data were obtained from official government and party
yearbooks and the 2003 China General Social Survey.
226
Appendix B, continued: Descriptive data on interviewees Table 1: Administrative level of interviewees
Administrative Level N Percent Township 12 5.08County 38 16.10City 58 24.58Province 60 25.42Central 68 28.81Total 236 100
Table 2: Interviewee type
Occupation N Cadre 177 Party school teacher 90 Part-time party school teacher 20 Party school student, part- or full-time 32 Party school trainee 78
Note: these are non-exclusive categories. For example, some interviewees were both trainees and students at some point, or some party school teachers were also administrators (i.e., ‘1’s in the cadre and party school teacher categories). Table 3: Interviewee gender
N Percent Male 166 70.34Female 70 29.66Total 236 100
227
Appendix B, continued: Scatterplots of provincial research sites Figure 1. Cadre population and GDP per cap, coastal region
Province A
6000
7000
8000
9000
1000
011
000
GD
P p
er c
apita
(yua
n)
.0006 .0008 .001 .0012 .0014 .0016Number of cadres per cap
Figure 3. Leading cadre population and GPD per cap, coastal region
Province A
6000
7000
8000
9000
1000
011
000
GD
P p
er c
apita
(yua
n)
.00025 .0003 .00035 .0004 .00045 .0005Number of leading cadres per cap
Figure 5. Leading cadre population and GDP per cap, all regions
Province B
Province A
010
000
2000
030
000
GD
P p
er c
apita
(yua
n)
0 .0005 .001 .0015 .002Number of leading cadres per cap
Figure 2. Cadre population and GDP per cap, central region
Province B
4000
4500
5000
5500
6000
6500
GD
P p
er c
apita
(yua
n)
.0006 .0008 .001 .0012Number of cadres per cap
Figure 4. Leading cadre population and GPD per cap, central region
Province B
4000
4500
5000
5500
6000
6500
GD
P p
er c
apita
(yua
n)
.0002 .0003 .0004 .0005Number of leading cadres per cap
Figure 6. Cadre population and GDP per cap, all regions
Province B
Province A
010
000
2000
030
000
GD
P p
er c
apita
(yua
n)
0 .001 .002 .003 .004Number of cadres per cap
Notes: Data on cadre populations were obtained from COD 1999; all other data from provincial-level statistical yearbooks, China Data Online, 1998.
228
Appendix C: City Z training allocations, 2008
Class name Duration
(days) Enrollment Location(s) City party school, Young cadre training class 45 60 Central Party School
Spirit of the 17th Party Congress class 3 360 City party school City party school, City-level cadre advanced class 45 60 provincial university City party school, City-level cadre advanced class 45 60 provincial university
Promoting urban and rural development and building a new socialist countryside advanced research class
3 74 City party school
Building a civilized city and cultural production development research and discussion class
3 49 City party school
Environmentally conscious city and civilization building research and discussion class
3 52 City party school
Innovation and rapid transitioning economy research and discussion class
3 70 City party school
"City of innovation" research and discussion class
15 35 Hong Kong university
Social work management research and discussion class
15 35 Singapore university Tsinghua University,
Enterprise manager advanced research class 7 120
Fudan University Tsinghua University,
Enterprise manager advanced research class 7 120
Fudan University Agricultural technologies training class 7 50 City party school Democratic parties, Association of Industry and Commerce, and non-party representative training class
15 45 Shenzhen socialism institute
Industrial economic management leader training class
7 50 City party school
Personnel allocation training class 6 50 Fudan University Private enterprise party secretary demonstration class
3 40 City party school
Newly promoted section-level cadre training class
15 60 City party school
City party representative training class 2 100 City party school Party spirit education training class 3 City party school High-level crisis management research class 6 70 City party school Overseas Chinese Office cadre training class 7 50 Fujian university Family planning leading cadre training class 2 200 City party school Family planning leading cadre training class 2 130 City party school Grassroots letters and complaints leader training class
3 120 City party school
Non-CCP young cadre training class 30 55 City socialism institute, Shanghai socialism institute
229
Class name Duration
(days) Enrollment Location(s) Retired cadre training class 2 100 City senior citizen
university CCP activist training class 4 200 City party school Discipline and inspection cadre training class 7 50 City party school The city's "Low income rural household wellness project" cadre training
7 80 City party school
The city's "Low income rural household wellness project" cadre training
7 80 City party school
Public security section chief training class 7 480 City police academy
Retired cadre bureau chief training class 6 20 City party school Communist Youth League cadre training class 3 130 Another city Party affairs cadre training class 3 150 City party school City building leading cadre training class 15 30 Provincial
construction ministry, Cadre school
City military affairs civilian cadre training class 45 City party school, City personnel training and testing center
Party committee office affairs training class 3 120 City party school Theoretical reserve cadre training class 3 200 City party school Discipline and inspection party secretary training class
15 30 Central discipline and inspection training center
Source: Internal school document.
230
Appendix D: Descriptive statistics and robustness tests of PSM presented in Chapter 3 Table 1: Descriptive summary of variables Variable Obs Mean SD Min Max Party school 589 0.29 0.45 0 1partydum1 586 0.70 0.46 0 1totalyrsed 589 6.82 4.59 0 70soelvl2 589 0.34 0.48 0 1soelvl3 589 0.44 0.50 0 1soelvl4 589 0.09 0.28 0 1age 589 50.78 12.10 20 72female 589 0.26 0.44 0 1frqupdum 578 0.97 0.17 0 1army1 588 0.20 0.40 0 1sentdum 588 0.14 0.34 0 1fccp 589 0.31 0.46 0 1mccp 589 0.10 0.29 0 1
Table 2: Pairwise correlations between independent and control variables partydum totalyrsed soelvl2 soelvl3 soelvl4 age female frqupdum army1 sentdum fccp mccp partydum1 1.00 totalyrsed -0.04 1.00 soelvl2 -0.02 -0.08 1.00 soelvl3 0.03 0.05 -0.65 1.00 soelvl4 0.06 0.01 -0.22 -0.27 1.00 age 0.18 -0.32 0.07 -0.09 0.05 1.00 female -0.21 0.09 -0.11 0.09 -0.01 -0.11 1.00 frqupdum 0.08 -0.01 -0.05 0.10 0.05 0.00 0.01 1.00 army1 0.21 -0.17 0.03 -0.06 0.05 0.16 -0.23 0.01 1.00 sentdum -0.06 -0.01 0.07 -0.05 -0.02 0.11 0.09 0.01 -0.07 1.00 fccp -0.02 0.17 -0.06 0.10 -0.02 -0.39 0.11 0.01 -0.06 -0.05 1.00 mccp -0.04 0.11 -0.03 0.04 -0.02 -0.17 0.08 -0.01 -0.01 -0.01 0.35 1.00
231
Appendix D, continued. Table 3: T-tests for equality of means across treatment and control groups, before and after matching Mean Mean %reduct t-test Variable Sample Treated Control %bias |bias| t p>t Party school Unmatched 1 0 . . . Matched 1 0 . . . . CCP Unmatched 0.91538 0.54633 91.3 7.93 0 Matched 0.91538 0.89423 5.2 94.3 0.58 0.563 Education Unmatched 6.5077 6.6294 -3.1 -0.3 0.764 Matched 6.5077 6.3192 4.9 -54.9 0.37 0.714 SOElvl2 Unmatched 0.36154 0.36422 -0.6 -0.05 0.958 Matched 0.36154 0.475 -23.5 -4135.6 -1.86 0.064 SOElvl3 Unmatched 0.46154 0.42492 7.4 0.71 0.48 Matched 0.46154 0.36731 18.9 -157.3 1.54 0.124 SOElvl4 Unmatched 0.08462 0.07029 5.3 0.52 0.602 Matched 0.08462 0.07115 5 6 0.4 0.687 Age Unmatched 49.577 49.335 2 0.19 0.848 Matched 49.577 49.742 -1.4 31.5 -0.12 0.908 Female Unmatched 0.26154 0.30671 -10 -0.95 0.343 Matched 0.26154 0.27885 -3.8 61.7 -0.31 0.754 Frqupdum Unmatched 0.97692 0.96166 8.8 0.81 0.42 Matched 0.97692 0.98077 -2.2 74.8 -0.21 0.83 Army Unmatched 0.14615 0.18211 -9.7 -0.91 0.362 Matched 0.14615 0.13077 4.1 57.2 0.36 0.721 Sentdown Unmatched 0.1 0.15016 -15.2 -1.4 0.161 Matched 0.1 0.13654 -11 27.2 -0.91 0.364 Fccp Unmatched 0.37692 0.3099 14.1 1.37 0.173 Matched 0.37692 0.32308 11.3 19.7 0.91 0.365 Mccp Unmatched 0.09231 0.08946 1 0.1 0.924 Matched 0.09231 0.09231 0 100 0 1 Beijing Unmatched 0.05385 0.10543 -19.1 -1.73 0.085 Matched 0.05385 0.06346 -3.6 81.4 -0.33 0.743 Tianjin Unmatched 0.1 0.09904 0.3 0.03 0.976 Matched 0.1 0.05769 14.1 -4314.1 1.26 0.207
232
Mean Mean %reduct t-test Variable Sample Treated Control %bias |bias| t p>t Hebei Unmatched 0.01538 0.03195 -10.9 -0.98 0.329 Matched 0.01538 0.01538 0 100 0 1 Shanxi Unmatched 0.00769 0.01917 -10 -0.88 0.379 Matched 0.00769 0.00577 1.7 83.2 0.19 0.85 Neimenggu Unmatched 0.02308 0.01278 7.7 0.79 0.43 Matched 0.02308 0.02115 1.4 81.3 0.11 0.916 Liaoning Unmatched 0 0 . . . Matched 0 0 . . . . Jilin Unmatched 0.01538 0.00639 8.6 0.91 0.363 Matched 0.01538 0.03846 -22.2 -156.6 -1.15 0.252 Heilongjiang Unmatched 0 0 . . . Matched 0 0 . . . . Shanghai Unmatched 0.07692 0.0639 5.1 0.5 0.62 Matched 0.07692 0.04423 12.7 -151 1.1 0.271 Jiangsu Unmatched 0.04615 0.03834 3.9 0.38 0.705 Matched 0.04615 0.05385 -3.8 1.6 -0.28 0.777 Zhejiang Unmatched 0 0 . . . Matched 0 0 . . . . Anhui Unmatched 0.02308 0.06709 -21.3 -1.87 0.063 Matched 0.02308 0.01346 4.7 78.2 0.58 0.564 Fujian Unmatched 0.03077 0.03834 -4.1 -0.39 0.698 Matched 0.03077 0.03846 -4.2 -1.6 -0.34 0.736 Jiangxi Unmatched 0.00769 0.02236 -12.1 -1.05 0.292 Matched 0.00769 0.00962 -1.6 86.9 -0.17 0.868 Shandong Unmatched 0.03846 0.04792 -4.6 -0.44 0.663 Matched 0.03846 0.05962 -10.4 -123.6 -0.79 0.432 Henan Unmatched 0.01538 0.0607 -23.8 -2.05 0.041 Matched 0.01538 0.00769 4 83 0.58 0.563 Hubei Unmatched 0.16923 0.03514 45.2 5.02 0 Matched 0.16923 0.15192 5.8 87.1 0.38 0.705 Hunan Unmatched 0.12308 0.05431 24.3 2.52 0.012 Matched 0.12308 0.17308 -17.7 27.3 -1.13 0.258
233
Mean Mean %reduct t-test Variable Sample Treated Control %bias |bias| t p>t Guangdong Unmatched 0.06923 0.09904 -10.7 -1 0.32 Matched 0.06923 0.05962 3.5 67.7 0.31 0.753 Guangxi Unmatched 0.03846 0.03514 1.8 0.17 0.865 Matched 0.03846 0.04615 -4.1 -131.9 -0.31 0.759 Chongqing Unmatched 0.01538 0.00319 12.7 1.42 0.155 Matched 0.01538 0.01154 4 68.4 0.27 0.789 Sichuan Unmatched 0.02308 0.02236 0.5 0.05 0.963 Matched 0.02308 0.03269 -6.4 -1249.1 -0.47 0.639 Guizhou Unmatched 0.00769 0.04792 -24.6 -2.07 0.039 Matched 0.00769 0.00769 0 100 0 1 Yunnan Unmatched 0.04615 0.03514 5.6 0.55 0.584 Matched 0.04615 0.04038 2.9 47.6 0.23 0.82 Shaanxi Unmatched 0.00769 0.01917 -10 -0.88 0.379 Matched 0.00769 0.00577 1.7 83.2 0.19 0.85 Gansu Unmatched 0.03077 0.01278 12.3 1.29 0.196 Matched 0.03077 0.02308 5.3 57.2 0.38 0.703
234
Appendix D, continued. Histogram 1: Absolute distance between the propensity scores of each treatment observation and nearest neighbor match
050
100
150
Freq
uenc
y
0 .02 .04 .06Absolute difference in propensity score
Tables 4-6: Estimation of average treatment effect on the treated (ATT) using different matching algorithms DV: Section rank (dummy) Treatment: Enrollment in non-degree party school training class
Method N treated
N control ATT SE t-statistic
Nearest neighbor, random draw logit (attnd) 170 82 0.13 0.096 1.349 Nearest-neighbor, equal weight (attnw) 170 86 0.221 0.078 2.841 Stratification (atts) 166 378 0.116 0.061 1.91 Kernel (attk) 170 355 0.093 0.061 1.513 DV: Department rank (dummy) Treatment: Enrollment in non-degree party school training class
Method N treated
N control ATT SE t-statistic
Nearest neighbor, random draw logit (attnd) 170 70 0.105 0.071 1.489
Nearest-neighbor, equal weight (attnw) 170 66 0.137 0.058 2.343 Stratification (atts) 166 378 0.096 0.047 2.04 Kernel (attk) 170 355 0.072 0.05 1.442 DV: Any rank (dummy) Treatment: Enrollment in non-degree party school training class
Method N treated
N control ATT SE t-statistic
Nearest neighbor, random draw logit (attnd) 170 97 0.129 0.078 1.666 Nearest-neighbor, equal weight (attnw) 170 96 0.229 0.069 3.346 Radius (attr) 139 284 0.094 0.54 1.743 Stratification (atts) 166 378 0.132 0.051 2.588 Kernel (attk) 170 355 0.101 0.049 2.067 Note: Bootstrapped standard errors are reported (200 replications)
235
Appendix E: Central Party School Young Cadre Training Classes descriptive data Table 1: Age at the time of training
Total 2000 1995 N % N % N %
30-39 43 9.7 13 5.3 30 15.140-49 268 60.6 152 62.6 116 58.350-55 38 8.6 24 9.9 14 7.0NA 93 21.0 54 22.2 39 19.6Total 442 100.0 243 100.0 199 100.0
Table 2: Gender
Total 2000 1995 N % N % N %
Male 362 81.9 200 82.3 162 81.4Female 45 10.2 27 11.1 18 9.0NA 35 7.9 16 6.6 19 9.5Total 442 100.0 243 100.0 199 100.0
236
Appendix E, continued. Table 3: Home province at time of training Total 2000 1995
N % N % N %Beijing 12 2.7 4 1.6 8 4.0Tianjin 9 2.0 5 2.1 4 2.0Hebei 19 4.3 9 3.7 10 5.0Shanxi 12 2.7 6 2.5 6 3.0Neimenggu 9 2.0 5 2.1 4 2.0Liaoning 18 4.1 10 4.1 8 4.0Jilin 15 3.4 9 3.7 6 3.0Heilongjiang 9 2.0 3 1.2 6 3.0Shanghai 4 0.9 0 0.0 4 2.0Jiangsu 21 4.8 13 5.3 8 4.0Zhejiang 22 5.0 13 5.3 9 4.5Anhui 15 3.4 11 4.5 4 2.0Fujian 8 1.8 3 1.2 5 2.5Jiangxi 11 2.5 4 1.6 7 3.5Shandong 29 6.6 20 8.2 9 4.5Henan 16 3.6 10 4.1 6 3.0Hubei 13 2.9 4 1.6 9 4.5Hunan 13 2.9 8 3.3 5 2.5Guangdong 11 2.5 8 3.3 3 1.5Guangxi 8 1.8 3 1.2 5 2.5Hainan 7 1.6 3 1.2 4 2.0Chongqing 5 1.1 3 1.2 2 1.0Sichuan 12 2.7 6 2.5 6 3.0Guizhou 6 1.4 2 0.8 4 2.0Yunnan 8 1.8 4 1.6 4 2.0Shaanxi 12 2.7 6 2.5 6 3.0Gansu 8 1.8 4 1.6 4 2.0Qinghai 8 1.8 4 1.6 4 2.0Ningxia 4 0.9 0 0.0 4 2.0Xinjiang 8 1.8 2 0.8 6 3.0Tibet 7 1.6 3 1.2 4 2.0NA 83 18.8 58 23.9 25 12.6Total 442 100.0 243 100.0 199 100.0
237
Appendix E, continued Table 4: Educational attainment
Total 2000 1995 N % N % N %
Professional school 1 0.2 0 0.0 1 0.5Part-time specialized college 5 1.1 5 2.1 0 0.0Full-time specialized college 21 4.8 12 4.9 9 4.5University 71 16.1 36 14.8 35 17.6Graduate school 133 30.1 84 34.6 49 24.6Training apprentice 1 0.2 1 0.4 0 0.0Other 112 25.3 47 19.3 65 32.7NA 98 22.2 58 23.9 40 20.1Total 442 100.0 243 100.0 199 100.0
Note: The ‘other’ category includes party school graduate degrees. Table 5: Percent with party school degrees*
Total 2000 1995 N % N % N %
Undergraduate 10 2.3 4 1.6 6 3.0Graduate 114 25.8 48 19.8 66 33.2Total 124 52 72
*Notes: Many individuals only specify "part-time" graduate degree on their biographies, which may or may not be obtained from a party school. Cheng Li, in his analysis of the biographies of 103 Fifth Generation leaders, finds that 23 out of the 63 leaders with masters degrees obtained these through the CPS (Li 2008b). Table 6: University major
Total 2000 1995 N % N 5 N %
Science 12 2.7 5 2.1 7 3.5Engineering 27 6.1 17 7.0 10 5.0Computer application and software 3 0.7 2 0.8 1 0.5Medicine and pharmacology 2 0.5 2 0.8 0 0.0Agriculture, forestry, fisheries, animal husbandry 10 2.3 7 2.9 3 1.5Finance and economics 47 10.6 26 10.7 21 10.6Management and administration 34 7.7 18 7.4 16 8.0Law 38 8.6 24 9.9 14 7.0Social sciences 6 1.4 1 0.4 5 2.5Humanities 42 9.5 17 7.0 25 12.6Foreign languages 3 0.7 2 0.8 1 0.5Education and information systems 3 0.7 0.0 3 1.5NA 215 48.6 122 50.2 93 46.7Total 442 100 243 100 199 100
238
Appendix F: International partnerships, central and provincial party schools
School Partner Year partnership began
Central Party School Sozialdemokratische Partei Deutschlands (Germany) Instituto Español de Comercio Exterior Crawford School at the Australian National University National Defense University (US) Georgetown University The National Academy of Public Administration (Italy) National Academy of Politics and Public Administration (Laos) Sustainable Development Commission UK (SDC) National Political and Administrative Academy Ho Chi Minh (Vietnam) Brookings Institution Ministry of National Development (Singapore) Indian Institute of Public Administration
The Department of Indian Affairs and Northern Development (Canada)
Canada School of Public Service National Graduate Institute for Policy Studies (Japan) Malik Management Centre, Switzerland Harvard Fairbanks Center for East Asian Research 1995 Anhui n/a Chongqing n/a Fujian n/a Gansu n/a Guangdong Canada International Development Agency at least 2004 Guangxi Montreal University National Political and Administrative Academy Ho Chi Minh Guizhou n/a Hebei n/a Henan n/a Hubei n/a Hunan Ministry of Internal Affairs (Vietnam) at least 2005 Central Civil Service Academy (Korea) at least 2006 European Administrative School at least 2006 Civil Service College (Singapore) at least 2006
Jiangsu International Centre for the Study of East Asian Development
Kyoto University Jiangxi National Political and Administrative Academy Ho Chi Minh Ecole Nationale d'Administration University of Georgia at least 2004 Nanyang Technological University Jilin Jiangyuandao Human Resources Development Academy (Korea) National Academy of Public Administration (Belarus) Liaoning Loyola University Neimenggu n/a Ningxia n/a Qinghai n/a
239
School Partner Year partnership began
Shaanxi Party School Spanish Agency for International Development (AECID)
2008
Shandong n/a Shanxi n/a Sichuan n/a Tianjin n/a Xinjiang n/a Xizang n/a Yunnan n/a Zhejiang Civil Service College (Singapore) Beijing California State University University of Georgia Baldwin-Wallace College Far East Studies Institute of the Russian Academy of Sciences Northwest Academy of Public Administration (Russia) Sciences Po (France) Administrative College of NRW (Germany) Korean Research Institute for Local Administration (Korea) University of Canberra (Australia) Kanagawa University (Japan) National Academy of Public Administration (Vietnam) Shanghai University of Alberta (Canada) at least 2004
German University of Administrative Sciences Speyer (Germany) at least 2006
Civil Service College in Singapore at least 2004 Maxwell School of Syracuse University at least 2000 University of Georgia at least 2004 Kennedy School of Harvard University at least 2007 Nagoya University (Japan) at least 2008 Ecole Nationale d' Administration Sciences Po (France)
Russian North-Western Academy of Public Administration at least 2004
National Academy of Public Administration (Vietnam) at least 2004
National Political and Administrative Academy Ho Chi Minh (Vietnam) at least 2004
Korean Research Institute for Local Administration (Korea) Oxford University Sheffield Hallam University Netherlands Maritime University The National Academy of Public Administration (Italy) at least 2006 Milan Training Academy (Italy) at least 2006 The Institute of Public Enterprise (India)
240
Appendix G: Categories for coding training syllabi
Theory Marxist-Leninist theory Mao theory Deng theory – this includes ‘Socialism with Chinese Characteristics’ Jiang theory – this includes “Three Represents” Hu theory – this includes “Harmonious Society” Capitalist theory/non-Marxist western economic theory Liberal democratic theory General theories of socialism (no clear attribution) – e.g., socialist development, economy Minorities in a socialist system Central party policy & documents Central party policy – economic development (e.g., SOE, agriculture, 5-yr. plans) Central party policy – social reform Central party policy – party building (including united front, unions, etc.) 16th Party Congress 17th Party Congress Local party policy Provincial party policy – economic development Provincial party policy – social City party policy – economic development City party policy – social County party policy – economic and social (including building new countryside) Party history and party building Party history Party constitution Party line/party building Leadership studies Leadership theory/’art of leadership’ Management (including crisis management) Speechmaking and communication Media relations Principles of service government Public administration Strategic thinking Case study International – US International – Asia International – W Europe International – USSR and E Europe International – general cases/lessons from abroad International – international organizations Domestic – building new countryside Domestic – coastal economic development Domestic – inland economic development
241
Basic educational disciplines Basic social science Basic humanities Chinese history and culture Basic sciences Practical skills Investigation and basic writing (reports and composition) Budgeting and accounting Military training Foreign language Law General Administrative law Business/finance law Criminal law Labor law Property rights Other General interest lectures Ethics Elective course – via TV or internet Computer class
242
References Aberbach, Joel D., and Bert A. Rockman. 1988. Mandates or Mandarins? Control and
Discretion in the Modern Administrative State. Public Administration Review 48 (2):606-612.
Alpermann, Bjorn. 2001. The Post-Election Administration of Chinese Villages. The China Journal 46:45-67.
Ang, Yuen Yuen. 2009a. Centralizing treasury management in China: The rationale of the central reformers. Public Administration and Development 29 (4):263-273.
Ang, Yuen Yuen. 2009b. State, Market, and Bureau-Contracting in Reform China, Political Science, Stanford University, Stanford, CA.
Arrow, Kenneth J. 1973. Higher Education as a Filter. Journal of Public Economics 2:193-216.
Baker, George P., Michael C. Jensen, and Kevin J. Murphy. 1988. Compensation and Incentives: Practice vs. Theory. The Journal of Finance 43 (3):593-616.
Bardhan, Pranab. 2009. India and China: Governance Issues and Development. Journal of Asian Studies 68 (2):347-57.
Barnett, A. Doak, and Ezra F. Vogel. 1967. Cadres, Bureaucracy, and Political Power in Communist China. New York: Columbia University Press.
Baum, Richard. 1996. Burying Mao: Chinese Politics in the Age of Deng Xiaoping. Princeton: Princeton University Press.
Becker, Sascha O., and Andrea Ichino. 2002. Estimation of average treatment effects based on propensity scores. The Stata Journal:1-19.
Bendor, Jonathan B. 1985. Parallel Systems: Redundancy in Government. Berkeley: University of California Press.
Blaydes, Lisa. 2008. Authoritarian Elections and Elite Management: Theory and Evidence from Egypt. In Princeton University Conference on Dictatorships. Princeton University.
Bo, Zhiyue. 2002. Chinese Provincial Leaders: Economic Performance and Political Mobility Since 1949: M.E. Sharpe.
Bo, Zhiyue. 2004. The Institutionalization of Elite Management in China. In Holding China Together: Diversity and National Integration in the Post-Deng Era, edited by B. J. Naughton and D. L. Yang. Cambridge: Cambridge University Press.
Braun, Dietmar, and Fabrizio Gilardi, eds. 2006. Delegation in Contemporary Democracies. London and New York: Routledge.
Brehm, John, and Scott Gates. 1997. Working, Shirking, and Sabotage: Bureaucratic Response to a Democratic Public. Ann Arbor: University of Michigan Press.
Brownlee, Jason. 2007. Authoritarianism in an Age of Democratization. Cambridge: Cambridge University Press.
Burnham, James. 1962. The Managerial Revolution. Bloomington: Indiana University Press.
Burns, John P., ed. 1989a. The Chinese Communist Party's Nomenklatura System: A Documentary Study of Party Control of Leadership Selection, 1979-1984. Armonk and London: M.E. Sharpe, Inc.
243
Burns, John P. 1989b. Introduction: The Chinese Nomenklatura System. In The Chinese Communist Party's Nomenklatura System: A Documentary Study of Party Control of Leadership Selection, 1979-1984, edited by J. P. Burns. Armonk and London: M.E. Sharpe, Inc.
Burns, John P. 1993. China’s administrative reforms for a market economy. Public administration and development 13 (4):345-60.
Burns, John P. 1994. Strengthening Central CCP Control of Leadership Selection: The 1990 Nomenklatura. China Quarterly 138:458-491.
Burns, John P. 2003. "Downsizing" the Chinese State: Government Retrenchment in the 1990s. China Quarterly (175):775-802.
Burns, John P. 2006. The Chinese Communist Party's Nomenklatura System as a Leadership Selection Mechanism. In The Chinese Communist Party in Reform, edited by K. E. Brodsgaard and Z. Yongnian. London and New York: Routledge.
Burns, John P., and Bidhya Bowornwathana, eds. 2001. Civil Service Systems in Asia, Civil Service Systems in Comparative Perspective. Cheltenham, UK and Northampton, MA: MPG Books Ltd.
Cai, Hongbin, and Daniel Treisman. 2006. Did government decentralization cause China's economic miracle? World Politics 58 (4):505-535.
Calvert, Randall L., Mathew D. McCubbins, and Barry R. Weingast. 1989. A Theory of Political Control and Agency Discretion. American Journal of Political Science 33 (3):588-611.
Campos, Jose Edgardo, and Hilton L. Root. 1996. The Key to the Asian Miracle: Making Shared Growth Credible. Washington, DC: Brookings Institution.
CCP Central Committee. 1981. On Questions of Party History: Resolution on Certain Questions in the History of Our Party Since the Founding of the People's Republic of China: Beijing Review.
Cell, Charles P. 1977. Revolution at Work: Mobilization Campaigns in China. New York: Academic Press.
Central Committee. 2004. Decision on Facing the Twenty-First Century through Strengthening and Improving Party School Work. Beijing.
Central Organization Department. 1983. Notice of the Organization Department of the Central Committee. In The Chinese Communist Party's Nomenklatura System: A Documentary Study of Party Control of Leadership Selection, 1979-1984, edited by J. P. Burns. Armonk, NY and London: M.E. Sharpe, Inc.
Central Organization Department, ed. 1999. Party and government leading cadre statistical data compilation (dangzheng lingdao ganbu tongji ziliao huibian). Beijing: Party building reading materials publishing house (Dang jian du wu chu ban she).
Central Organization Department. 2004. CCP Central Committee Decision on Facing the Twenty-First Century through Strengthening and Improving Party School Work. Beijing.
Central Organization Department, Cadre Education Office. 2006. Reader on Trial Regulations on Cadre Education and Training (ganbu jiaoyu peixun gongzuo tiaoli (shixing), xuexi fudao): Dangjian duwu chubanshe.
Chang, Gordon G. 2001. The Coming Collapse of China. New York: Random House.
244
Chen, Guisheng. 2007. China's Cadre Education, 1927-1949 (zhongguo gan bu jiao yu (1927-1949)). Shanghai: Huadong Normal University Press.
Chinese yearbook of ideological and political work (zhongguo sixiang zhengzhi gongzuo nianjian). 2001. Central Party School Publishing House.
Chow, King W. 1988. The Management of Chinese Cadre Resources: The Politics of Performance Appraisal (1949-84). International Review of Administrative Sciences 54:359-377.
Chow, King W. 1993. The Politics of Performance Appraisal. In Public Administration in China, edited by M. K. Mills and S. S. Nagel. Westport and London: Greenwood Press.
Clemens, Elisabeth S., and James M. Cook. 1999. Politics and Institutionalism: Explaining Durability and Change. Annual Review of Sociology 25:441-66.
Cox, Gary W. 2009. Authoritarian elections and leadership succession, 1975-2004. CPS, ed. 2004. Research on certain questions concerning the building of a ruling party
(zhizheng dang jianshe ruogan wenti yanjiu). Beijing: Central Party School Publishing House.
Dai, Yanjun. 2006. The issue of strengthening the party's governance capacity (jiaqiang dangde zhizheng nengli jianshe wenti). In Compilation of CCP Central Party School Teaching Scripts 3: Studies of Marxism, Leninism, Mao Zedong Thought, Deng Xiaoping Theory, the Three Represents Thought and Other Issues (zhonggong zhongyang dangxiao jianggaoxuan 3: makesi liening zhuyi, maozedong sixiang, dengxiaoping lilun, sange daibiao zhongyao sixiang qianyan wenti yanjiu), edited by T. Shi. Beijing: Central Party School Publishing House.
Dehejia, Rajeev H., and Sadek Wahba. 1999. Causal Effects in Nonexperimental Studies: Reevaluating the Evaluation of Training Programs. Journal of the American Statistical Association 94 (448):1053-1062.
Denzau, Arthur T., and Douglass C. North. 1994. Shared Mental Models: Ideologies and Institutions. Kyklos 47 (1):3-31.
Dethier, Jean-Jacques, ed. 2000. Governance, Decentralization and Reform in China, India and Russia. Boston and London: Kluwer Academic Publishers.
Diaz-Cayeros, Alberto, Beatriz Magaloni, and Barry R. Weingast. 2003. Tragic Brilliance: Equilibrium Hegemony and Democratization in Mexico.
Dickson, Bruce J. 2003. Red Capitalists in China: The Party, Private Entrepreneurs, and Prospects for Political Change. Cambridge: Cambridge University Press.
Dimitrov, Martin. 2010. Popular Accountability and Regime Resilience in Contemporary China: Evidence from Citizen Complaints: Dartmouth College.
Ding, X.L. 1994. The Decline of Communism in China: Legitimacy Crisis, 1977-1989. Cambridge: Cambridge University Press.
Dittmer, Lowell. 1978. Bases of Power in Chinese Politics: The Fall of the Gang of Four. World Politics 31 (1):26-55.
Downs, Anthony. 1957. An Economic Theory of Democracy. New York: Harper. Downs, Eric S. 2008. Business Interest Groups in Chinese Politics: The Case of the Oil
Companies. In China's Changing Political Landscape: Prospects for Democracy, edited by C. Li. Washington, DC: Brookings Institution Press.
245
Duckett, Jane. 2001. Bureaucrats in Business, Chinese-Style: The Lessons of Market Reform and State Entrepreneurialism in the People's Republic of China. World Development 29 (1):23-37.
Enthoven, Alain, and Henry Rowen. 1959. Defense Planning and Organization. Santa Monica: Rand Corporation.
Epstein, Amy B. 1996. Village Elections in China: Experimenting with Democracy. In China's Economic Future: Challenges to U.S. Policy, edited by C. o. t. U. S. Joint Economic Committee. Washington, DC: M.E. Sharpe.
Epstein, David, and Sharyn O'Halloran. 2006. A Theory of Efficient Delegation. In Delegation in Contemporary Democracies, edited by D. Braun and F. Gilardi. London and New York: Routledge.
Farmer, Kenneth C. 1992. The Soviet Administrative Elite. New York and London: Praeger.
Felsenthal, Dan S. 1980. Applying the Redundancy Concept to Administrative Organizations. Public Administration Review 40 (3):247-252.
Fewsmith, Joseph. 2003. Where do Correct Ideas Come From? The Party School, Key Think Tanks, and the Intellectuals. In China's Leadership in the 21st Century: The Rise of the Fourth Generation, edited by D. M. Finkelstein and M. Kivlehan. Armonk and London: M.E. Sharpe.
Fewsmith, Joseph. 2007. Democracy is a Good Thing. China Leadership Monitor (22). Fewsmith, Joseph. 2008. A New Upsurge in Political Reform? - Maybe. China
Leadership Monitor (24). Gadsden, Amy Epstein, and Anne F. Thurston. 2001. Village Elections in China:
Progress, Problems, and Prospects: International Republican Institute. Gandhi, Jennifer. 2008a. Dictatorial Institutions and their Impact on Economic Growth.
European Journal of Sociology 49 (01):3-30. Gandhi, Jennifer. 2008b. Political Institutions Under Dictatorship. Cambridge:
Cambridge University Press. Gandhi, Jennifer, and Ellen Lust-Okar. 2009. Elections Under Authoritarianism. Annual
Review of Political Science 12 (1):403-422. Gandhi, Jennifer, and Adam Przeworski. 2007. Authoritarian Institutions and the Survival
of Autocrats. Comparative Political Studies 40 (11):1279-1301. Geddes, Barbara. 1994. Politician's Dilemma: Building State Capacity in Latin America.
Berkeley: University of California Press. Geddes, Barbara. 1999. What Do We Know About Democratization After Twenty Years?
Annual Review of Political Science 2:115-144. Geddes, Barbara. 2006. Paradigms and Sand Castles: Theory Building and Research
Design in Comparative Politics. Ann Arbor, MI: The University of Michigan Press.
Geddes, Barbara. 2008. Party Creation as an Autocratic Survival Strategy. In Dictatorships: Their Governance and Social Consequences. Princeton University.
Gehlbach, Scott, and Philip Keefer. 2008. Investment without Democracy: Ruling-Party Institutionalization and Credible Commitment in Autocracies.
General View of the Nation's Party Schools (quanguo dangxiao gailan). 1996. Beijing: Red Flag Publishing House.
246
Genschel, Philipp. 1997. The Dynamics of Inertia: Institutional Persistence and Change in Telecommunications and Health Care. Governance 10 (1):43-66.
Goldstone, Jack A. 1995. The Coming Chinese Collapse. Foreign Policy (99):35-53. Gonzalez de Lara, Yadira, Avner Greif, and Saumitra Jha. 2008. The Administrative
Foundations of Self-Enforcing Constitutions. American Economic Review 98 (2):105-09.
Gore, Lance L.P. 1998. Market Communism: The Institutional Foundation of China's Post-Mao Hyper-Growth. Hong Kong: Oxford University Press.
Guo, Liangping, and Lance L.P. Gore. 2007. China's New Cadre Evaluation System. East Asia Institute Background Brief 324.
Guo, Weigui, and Wei Shan. 2009. The Local Party Schools and China's Cadre Education and Training: Expectations and Realities. EAI Working Paper No. 151.
Hadenius, Axel, and Jan Teorell. 2007. Pathways from Authoritarianism. Journal of Democracy 18 (1):143-156.
Hannan, Michael T., and John Freeman. 1984. Structural Inertia and Organizational Change. American Sociological Review 49 (2):149-164.
Hanson, Stephen E. 1995. The Lenininst Legacy and Institutional Change. Comparative Political Studies 28 (2):306-314.
Harding, Harry. 1981. Organizing China: The Problem of Bureaucracy, 1949-1976. Stanford: Stanford University Press.
Hebert, Robert F., and Albert N. Link. 1988. The Entrepreneur: Mainstream Views and Radical Critiques. 2nd ed. New York: Praeger.
Heilmann, Sebastian. 2008. From Local Experiments to National Policy: The Origins of China's Distinctive Policy Process. China Journal 59:1-31.
Heilmann, Sebastian, and Sarah Kirchberger. June 2000. The Chinese Nomenklatura in Transition: A Study Based on Internal Cadre Statistics of the Central Organization Department of the Chinese Communist Party. China Analysis, Center for East Asian and Pacific Studies, Trier University (1).
Heimer, Maria. 2006. The Cadre Responsibility System and the Changing Needs of the Party. In The Chinese Communist Party in Reform, edited by K. E. Brodsgaard and Z. Yongnian. London and New York: Routledge.
Huang, Da-Qiang. 1993a. Promoting the Quality of the Chinese Government. In Public Administration in China, edited by M. K. Mills and S. S. Nagel. Westport and London: Greenwood Press.
Huang, Da Qiang. 1993b. Promoting the Quality of the Chinese Government. In Public Administration in China, edited by M. K. Mills and S. S. Nagel. Westport and London: Greenwood Press.
Huang, Shi'an, ed. 2002a. CCP Central Party School Correspondence Education Yearbook. Beijing: CCP Central Party School Publishing House.
Huang, Yasheng. 1995. Administrative Monitoring in China. China Quarterly (143):828-843.
Huang, Yasheng. 1996. Central-local relations in China during the reform era: The economic and institutional dimensions. World Development 24 (4):655-672.
Huang, Yasheng. 2002b. Managing Chinese Bureaucrats: An Institutional Economics Perspective. Political Studies 50 (1):61-79.
247
Huntington, Samuel P. 1968. Political Order in Changing Societies. New Haven, CT: Yale University Press.
Huskey, Eugene. 2004. From Higher Party Schools to Academies of State Service: The Marketization of Bureaucratic Training in Russia. Slavic Review 63 (2):325-348.
Jia, Hao, and Zhimin Lin, eds. 1994. Changing Central-Local Relations in China. Boulder: Westview Press.
Jiang, Huaxuan, Weiping Zhang, and Xiao Sheng, eds. 1983. Record of Important CCP Meetings, 1921-2006 (zhongguo gongchandang zhongyao huiyi jishi). Beijing: Central Documents Publishing House.
Jin, Jing, and Heng-fu Zou. 2003. Soft Budget Constraints and Local Government in China. In Fiscal decentralization and the challenge of hard budget constraints, edited by J. Rodden, G. S. Eskeland and J. I. Litvack. Cambridge: MIT Press.
Johnson, Chalers. 1982. MITI and the Japanese Miracle: The Growth of Industrial Policy, 1925-1975. Stanford: Stanford University Press.
Jowitt, Kenneth. 1975. Inclusion and Mobilization in European Leninist Regimes. World Politics 28 (1):69-96.
Kaufman, Herbert. 2006. The Forest Ranger: A Study in Administrative Behavior. Special Reprint ed. Washington, DC: Resources for the Future Press.
Kelliher, Daniel. 1997. The Chinese Debate Over Village Self-Government. The China Journal 37:63-86.
Kirby, William C., Robert S. Ross, and Gong Li, eds. 2005. Normalization of US-China Relations: An International History. Cambridge: Harvard University Press.
Klugman, Jeffry. 1989. The New Soviet Elite: How They Think and What They Want. New York and Westport: Praeger.
Krause, George A., and James W. Douglas. 2003. Are Two Heads Always Better than One? Redundancy, Competition, and Task Performance Quality in Public Bureaus. In Annual Conference of the Association of Budgeting and Financial Management. Washington, DC.
Kreps, David M. 1990. Corporate Culture and Economic Theory. In Perspectives on Positive Political Economy, edited by J. E. Alt and K. A. Shepsle. Cambridge: Cambridge University Press.
Lampton, David M. 1992. A Plum for a Peach: Bargaining, Interest, and Bureaucratic Politics in China. In Bureaucracy, Politics, and Decision Making in Post-Mao China, edited by K. G. Lieberthal and D. M. Lampton. Berkeley: University of California Press.
Lampton, Michael. 1987. Chinese Politics: The Bargaining Treadmill. Issues and Studies:11-41.
Lan, Zhiyong, and David H. Rosenbloom. 1992. Editorial: Public Administration in Transition? Public Administration Review 52 (6):535-537.
Landau, Martin. 1969. Redundancy, Rationality, and the Problem of Duplication and Overlap. Public Administration Review 29 (4):346-358.
Landau, Martin. 1991. On Multiorganizational Systems in Public Administration. Journal of Public Administration Research and Theory: J-PART 1 (1):5-18.
Landry, Pierre F. 2003. The Political Management of Mayors in Post-Deng China. Copenhagen Journal of Asian Studies 17:31-58.
248
Landry, Pierre F. 2007. Decentralized Authoritarianism in China: The Communist Party's Control of Local Elites in the Post-Mao Era. Cambridge: Cambridge University Press.
Lazarev, Valery. 2005. Economics of One-Party State: Promotion Incentives and Support for the Soviet Regime. Comparative Economic Studies 47 (2):346-363.
Lazarev, Valery. 2007. Political labor market, government policy, and stability of a non-democratic regime. Journal of Comparative Economics 35 (3):546-563.
Lee, Hong Yung. 1983. China's 12th Central Committee: Rehabilitated Cadres and Technocrats. Asian Survey 23 (6):673-691.
Lee, Hong Yung. 1984. Evaluation of China's Bureaucratic Reforms. Annals of the American Academy of Political and Social Science 476:34-47.
Lee, Hong Yung. 1991. From Revolutionary Cadres to Party Technocrats in Socialist China. Berkeley and Oxford: University of California Press.
Lenin, V.I. 1964. What is to be Done? Burning Questions of Our Movement. In Collected Works. Moscow: Progress Publishers.
Lerner, Allan W. 1978. On ambiguity and organizations. Administration Society 10 (1):3-32.
Lewis, David E. 2002. The Politics of Agency Termination: Confronting the Myth of Agency Immortality. The Journal of Politics 64 (01):89-107.
Li, Bobai. 2001. Manufacturing Meritocracy: Adult Education, Career Mobility, and Elite Transformation in Socialist China, Sociology, Stanford, Stanford.
Li, Bobai, and Andrew G. Walder. 2001. Career Advancement as Party Patronage: Sponsored Mobility into the Chinese Administrative Elite, 1949-1996. American Journal of Sociology 106 (5):1371-1408.
Li, Cheng. 2006. Reshuffling the Four Tiers of Local Leaders: Goals and Implications. China Leadership Monitor 18:1-22.
Li, Cheng. 2007a. The big shake-up. China Economic Quarterly 2 (1):24-28. Li, Cheng. 2007b. China's Leadership, Fifth Generation. Brookings Institution,
http://www.brookings.edu/articles/2007/12_china_li.aspx (accessed 8 Feb 2008). Li, Cheng. 2008a. China's Fifth Generation: Is Diversity a Source of Strength or
Weakness. Asia Policy 6:53-93. Li, Cheng. 2008b. Will China's "Lost Generation" Find a Path to Democracy? In China's
Changing Political Landscape: Prospects for Democracy, edited by C. Li. Washington, DC: Brookings Institution Press.
Li, Cheng, and David Bachman. 1989. Localism, Elitism, and Immobilism: Elite Formation and Social Change in Post-Mao China. World Politics 42 (1):64-94.
Li, Cheng, and Lynn White. 1988. The Thirteenth Central Committee of the Chinese Communist Party: From Mobilizers to Managers. Asian Survey 28 (4):371-399.
Li, Cheng, and Lynn White. 1990. Elite Transformation and Modern Change in Mainland China and Taiwan: Empirical Data and the Theory of Technocracy. The China Quarterly (121):1-35.
Li, David D. 1998. Changing Incentives of the Chinese Bureaucracy. The American Economic Review 88 (2):393-397.
Li, Kangtai. 1993. Administrative Law in China. In Public Administration in China, edited by M. K. Mills and S. S. Nagel. Westport and London: Greenwood Press.
249
Li, Wei. 1994. The Chinese Staff System: A Mechanism for Bureaucratic Control and Integration. Berkeley: Center for Chinese Studies.
Liao, Yuehui, Meichang Pan, and Yuqing Zeng, eds. 1999. The Development and History of China's Special Economic Zones. Shenzhen: Haitian Publishing House.
Lieberthal, Kenneth, and Michel Oksenberg. 1988. Policy Making in China: Leaders, Structures, and Processes. Princeton: Princeton University Press.
Lipset, Seymour Martin. 1981. Political Man. Baltimore: Johns Hopkins University Press. Litwack, John M. 1991. Discretionary Behaviour and Soviet Economic Reform. Soviet
Studies 43 (2):255-279. Liu, Jiaqi. 1989. Survey of Basic Principles of Party School Education (dangxiao jiaoyu
yuanli gailun). Beijing: Central Party School Publishing House. Liu, Meiru. 2001. Administrative Reform in China and its Impact on the Policy-Making
Process and Economic Development After Mao: Reinventing Chinese Government. Lewiston: Edwin Mellen Press.
Lorentzen, Peter L. 2008. Regularized Rioting: Strategic Toleration of Popular Protest in China: University of California, Berkeley.
Lu, Xiaobo. 2000. Cadres and Corruption: The Organizational Involution of the Chinese Communist Party. Stanford: Stanford University Press.
Lupia, Arthur, and Mathew D. McCubbins. 1994. Learning from Oversight: Fire Alarms and Police Patrols Reconstructed. Journal of Law, Economics, and Organization 10 (1):96-125.
Lust-Okar, Ellen. 2006. Elections under authoritarianism: Preliminary lessons from Jordan. Democratization 13 (3):456-471.
Lust-Oker, Ellen. 2005. Structuring Conflict in the Arab World: Incumbents, Opponents, and Institutions. Cambridge: Cambridge University Press.
Magaloni, Beatriz. 2006. Voting for Autocracy: Hegemonic Party Survival and its Demise in Mexico. Edited by M. Levi, Cambridge Studies in Comparative Politics. Cambridge/New York: Cambridge University Press.
Magaloni, Beatriz. 2008. Credible Power-Sharing and the Longevity of Authoritarian Rule. Comparative Political Studies 41 (4-5):715-741.
Magaloni, Beatriz, and Ruth Kricheli. 2010. Political Order and One-Party Rule. Annual Review of Political Science 13:123-43.
Manion, Melanie. 1985. The Cadre Management System, Post Mao: The Appointment, Promotion, Transfer and Removal of Party and State Leaders. China Quarterly 102:203-33.
Manion, Melanie. 1993. Retirement of Revolutionaries in China: Public Policies, Social Norms, Private Interests. Princeton, NJ: Princeton University Press.
Manion, Melanie. 1996. The Electoral Connection in the Chinese Countryside. American Political Science Review 90 (4):736-748.
Massey, Andrew. 1993. Managing the Public Sector: A Comparative Analysis of the United Kingdom and the United States: Edward Elgar Publishing Limited.
McCubbins, Mathew D., and Thomas Schwartz. 1984. Congressional Oversight overlooked: Police Patrols Versus Fire Alarms. American Journal of Political Science 28 (1):165-179.
250
Mertha, Andrew C. 2006. Policy Enforcement Markets: How Bureaucratic Redundancy Contributes to Effective Intellectual Property Implementation in China. Comparative Politics 38 (3):295-316.
Merton, Robert K. 1968. Social Theory and Social Structure. New York: Free Press. Meyer-Sahling, Jan-Hinrik. 2006. The Institutionalization of Political Discretion in Post-
Communist Civil Service Systems: The Case of Hungary. Public Administration 84 (3):693-716.
Milgrom, Paul, and John Roberts. 1992. Economics, Organization, and Management. Englewood Cliffs, NJ: Prentice Hall.
Miller, Gary J. 2005. The Political Evolution of Principal-Agent Models. Annual Review of Political Science 8:203-25.
Mills, C. Wright. 1959. The Power Elite. New York: Oxford University Press. Ministry of Personnel. 2006. Trial Regulations on Cadre Education and Training (ganbu
jiaoyu peixun gongzuo tiaoli (shixing)): Zhongguo fazhi chubanshe. Minzner, Carl F. 2009. Riots and Cover-Ups: Counterproductive Control of Local Agents
in China. University of Pennsylvania Journal of International Law 31:72. Miranda, Rowan, and Allan Lerner. 1995. Bureaucracy, Organizational Redundancy, and
the Privatization of Public Services. Public Administration Review 55 (2):193-200. Mittal, Sonia. 2008. A Necessary Precaution: The Separation of Powers and Political
Stability in America. In Midwest Political Science Association Conference. Chicago, IL.
Montinola, Gabriella, Yingyi Qian, and Barry R. Weingast. 1995. Federalism, Chinese Style: The Political Basis for Economic Success in China. World Politics 48 (1):50-81.
Morgan, Maria Chan. 1981. Controlling the Bureaucracy in Post-Mao China. Asian Survey 21 (12):1223-1236.
Munro, Donald J. 1971. The Malleability of Man in Chinese Marxism. The China Quarterly (48):609-640.
Nathan, Andrew. 1973. A Factional Model For CCP Politics. China Quarterly 53:98-114. Naughton, Barry. 1992. Hierarchy and the Bargaining Economy: Government and
Enterprise in the Reform Process. In Bureaucracy, Politics, and Decision Making in Post-Mao China, edited by K. G. Lieberthal and D. M. Lampton. Berkeley: University of California Press.
Naughton, Barry. 2007. The Chinese Economy: Transitions and Growth. Cambridge and London: MIT Press.
Naughton, Barry J., and Dali L. Yang, eds. 2004. Holding China Together: Diversity and National Integration in the Post-Deng Era. Cambridge: Cambridge University Press.
Nee, Victor, and Peng Lian. 1994. Sleeping with the Enemy: A Dynamic Model of Declining Political Commitment in State Socialism. Theory and Society 23 (2):253-296.
Niskanen, William A. 1971. Bureaucracy and Representative Government. Chicago and New York: Aldine Atherton, Inc.
North, Douglass C. 2005. Understanding the Process of Economic Change. Princeton and Oxford: Princeton University Press.
251
O'Brien, Kevin J., and Lianjiang Li. 1999. Campaign Nostalgia in the Chinese Countryside. Asian Survey 39 (3):375-393.
O'Donnell, Guillermo, and Philippe C. Schmitter. 1986. Transitions from Authoritarian Rule: Tentative Conclusions about Uncertain Democracies. Baltimore and London: Johns Hopkins University Press.
Oi, Jean C. 1992. Fiscal Reform and the Economic Foundations of Local State Corporatism in China. World Politics 45 (1):99-126.
Oi, Jean C. 1998a. The Collective Foundation for Rapid Rural Industrialization. In Cooperative and Collective in China's Rural Development, edited by E. B. Vermeer, F. N. Pieke and W. L. Chong. Armonk, NY: M.E. Sharpe.
Oi, Jean C. 1998b. The Evolution of Local State Corporatism. In Zouping in Transition: The Process of Reform in Rural North China, edited by A. G. Walder. Cambridge and London: Harvard University Press.
Oi, Jean C. 1999. Rural China Takes Off: Institutional Foundations of Economic Reform. Berkeley: University of California Press.
Oi, Jean C. 2004. Realms of Freedom in Post-Mao China. In Realms of Freedom in Modern China, edited by W. C. Kirby. Stanford, CA: Stanford University Press.
Oi, Jean C., and Scott Rozelle. 2000. Elections and Power: The Locus of Decision-Making in Chinese Villages. The China Quarterly:513-539.
Olson, Mancur. 1993. Dictatorship, Democracy, and Development. The American Political Science Review 87 (3):567-576.
Pearl, Judea. 2000. Causality: Models, Reasoning, and Inference. Cambridge and New York: Cambridge University Press.
Pei, Minxin. 2002. China's Governance Crisis. Foreign Affairs 81 (5):96-109. Pei, Minxin. 2008. Fighting Corruption: A Difficult Challenge for Chinese Leaders. In
China's Changing Political Landscape: Prospects for Democracy, edited by C. Li. Washington, DC: Brookings Institution Press.
Pfeffer, Jeffrey, and Gerald Salancik. 1978. The External Control of Organizations. New York: Harper & Row.
Pieke, Frank N. 2008. Marketization and centralization of cadre education and training. In Reinventing the Chinese party-state. Shenyang Normal University.
Pieke, Frank N. 2009a. The Good Communist: Elite Training and State Building in Today's China.
Pieke, Frank N. 2009b. Marketization, Centralization and Globalization of Cadre Training in Contemporary China. China Quarterly 200 (1):953-971.
Przeworski, Adam, Michael E. Alvarez, Jose Antonio Cheibub, and Fernando Limongi. 2000. Democracy and Development: Political Institutions and Well-Being in the World, 1950-1990. Cambridge: Cambridge University Press.
Punyaratabandhu-Bhakdi, Suchitra. 1983. Individual Values, Organizational Structure, and the Problem of Performance: Thailand as a Case Study. Public Administration Review 43 (6):510-519.
Putnam, Robert D. 1976. The Comparative Study of Political Elites. Edited by J. LaPalonbara, Prentice-Hall Contemporary Comparative Politics Series. Englewood Cliffs, N.J.: Prentice-Hall.
252
Qian, Yingyi. 2003. How Reform Worked in China. In In Search of Prosperity: Analytic Narratives on Economic Growth, edited by D. Rodrik. Princeton and Oxford: Princeton University Press.
Qian, Yingyi, and Barry R. Weingast. 1997. Federalism as a Commitment to Preserving Market Incentives. Journal of Economic Perspectives 11:83-92.
Rawski, Thomas G. 1994. Chinese Industrial Reform: Accomplishments, Prospects, and Implications. The American Economic Review 84 (2):271-275.
Rodden, Jonathan. 2004. Comparative Federalism and Decentralization: On Meaning and Measurement. Comparative Politics 36 (4):481-500.
Rostow, W.W. 1960. The Stages of Economic Growth: A Non-Communist Manifesto. Cambridge: Cambridge University Press.
Schoenhals, Michael. 1991. The 1978 Truth Criterion Controversy. The China Quarterly (126):243-268.
Scholz, John T., Jim Twombly, and Barbara Headrick. 1991. Street-Level Political Controls Over Federal Bureaucracy. The American Political Science Review 85 (3):829-850.
Schubert, Gunter. 2002. Village Elections in the PRC: A Trojan Horse of Democracy? Discourses on Political Reform and Democratization in East and Southeast Asian in the Light of New Processes fo Regional Community Building (19).
Schurmann, Franz. 1970. Ideology and Organization in Communist China. 2nd ed. Berkeley: University of California Press.
Selznick, Philip. 1960. The Organizational Weapon: A Study of Bolshevik Strategy and Tactics. Glencoe, IL: Free Press.
Shambaugh, David. 2008a. China's Communist Party: Atrophy and Adaptation. Washington, DC: Woodrow Wilson Center Press and University of California Press.
Shambaugh, David. 2008b. Learning from Abroad to Reinvent Itself: External Influences on Internal CCP Reforms. In China's Changing Political Landscape: Prospects for Democracy, edited by C. Li. Washington, DC: Brookings Institution Press.
Shi, Li. 2000. Efficiency and Redistribution in China's Revenue Sharing System. In Governance, Decentralization and Reform in China, India and Russia, edited by J.-J. Dethier. Boston and London: Kluwer Academic Publishers.
Shi, Taifeng, ed. 2006. Compilation of CCP Central Party School Teaching Scripts 6: Studies of Current Important Theory, Hot Issues, and Difficult Issues (zhonggong zhongyang dangxiao jianggaoxuan 6: dangqian zhongda lilun yu redian nandian wenti yanjiu). Beijing: Central Party School Publishing House.
Shirk, Susan L. 1993. The Political Logic of Economic Reform. Berkeley: University of California Press.
Shirk, Susan L. 2006. China: Fragile Superpower. New York and Oxford: Oxford University Press.
Shirk, Susan L. 2007. Changing Media, Changing Foreign Policy in China. Japanese Journal of Political Science 8 (01):43-70.
Simon, Herbert A. 1979. Rational Decision Making in Business Organizations. The American Economic Review 69 (4):493-513.
Smith, Benjamin B. 2005. Life of the Party: The Origins of Regime Breakdown and Persistence under Single-Party Rule. World Politics 57 (3):421-51.
253
Smith, Peter H. 1979. Labyrinths of Power: Political Recruitment in Twentieth-Century Mexico. Princeton: Princeton University Press.
Solnick, Steven L. 1996. The Breakdown of Hierarchies in the Soviet Union and China: A Neoinstitutional Perspective. World Politics 48 (2):209-238.
Solnick, Steven L. 1998. Stealing the State: Control and Collapse in Soviet Institutions. Cambridge: Harvard University Press.
Spence, Jonathan D. 1999. The Search for Modern China. 2nd ed. New York and London: W.W. Norton and Company.
Spence, Michael. 1973. Job Market Signaling. Quarterly Journal of Economics 87 (3):355-374.
Stiglitz, Joseph E. 1975. The Theory of "Screening," Education, and the Distribution of Income. American Economic Review 65 (3):283-300.
Stockmann, Daniela, and Mary Elizabeth Gallagher. 2009. Remote Control: How the Media Sustains Authoritarian Rule in China.
Streeter, Calvin L. 1992. Redundancy in Organizational Systems. The Social Service Review 66 (1):97-111.
Suleiman, Ezra N. 1977. The Myth of Technical Expertise: Selection, Organization, and Leadership. Comparative Politics 10 (1):137-158.
Tang, Tsou. 1976. Prolegomena to the Study of Informal Groups in CCP Politics. China Quarterly 65:98-114.
Tezcur, Bunes Murat. 2008. Intra-Elite Struggles in Iranian Elections. In Political Participation in the Middle East, edited by E. Lust-Okar and S. Zerhouni. Boulder and London: Lynne Rienner.
Ting, Michael M. 2003. A Strategic Theory of Bureaucratic Redundancy. American Journal of Political Science 47 (2):274-292.
Townsend, James R. 1969. Political Participation in Communist China. Berkeley: University of California Press.
Tran, Emilie. 2003. From Senior Official to Top Civil Servant: An Enquiry into the Shanghai Party School. Perspectives Chinoises (46):27.
Tsai, Kellee S. 2005. Capitalists Without a Class: Political Diversity Among Private Entrepreneurs in China. Comparative Political Studies 38 (9):1130-58.
Tucker, Robert C., ed. 1975. The Lenin Anthology. New York: W.W. Norton and Company.
U, Eddy. 2007. Disorganizing China: Counter-Bureaucracy and the Decline of Socialism. Stanford: Stanford University Press.
Vogel, Ezra F. 1967. From Revolutionary to Semi-Bureaucrat. China Quarterly:36-60. Walder, Andrew G. 1995. Career Mobility and the Communist Political Order. American
Sociological Review 60 (3):309-328. Walder, Andrew G., Bobai Li, and Donald J. Treiman. 2000. Politics and Life Chances in
a State Socialist Regime: Dual Career paths into the Urban Chinese Elite, 1949-1996. American Sociological Review 65 (2):191-209.
Walder, Andrew G., and Litao Zhao. 2006. Political Office and Household Wealth: Rural China in the Deng Era. China Quarterly (186):357-76.
Wang, Changjiang. 2002. Historical Lessons of Party Building from the Soviet Communist Party. In Compilation of CCP Central Party School Teaching Scripts: Important Issues of the World (zhonggong zhongyang dangxiao jianggaoxuan:
254
guanyu dangdai shijie zhongda wenti), edited by Y. Yu and C. Yang. Beijing: Central Party School Publishing House.
Wang, Huanchao. 2006. Social Development Theory and Several Major Issues in China's Social Development (shehui fazhan lilun yu woguo shehui fazhan zhong de jige zhongda wenti). In Compilation of CCP Central Party School Teaching Scripts 6: Studies of Current Important Theory, Hot Issues, and Difficult Issues (zhonggong zhongyang dangxiao jianggaoxuan 6: dangqian zhongda lilun yu redian nandian wenti yanjiu), edited by T. Shi. Beijing: Central Party School Publishing House.
Wang, Zhongqing, ed. 1992. General Description of Party School History (dangxiao jiaoyu lishi gaishu), 1921-1947. Beijing: Party School Publishing House.
Wedeman, Andrew. 2004. The Intensification of Corruption in China. China Quarterly 180:895-921.
Wedeman, Andrew. 2005. Anticorruption Campaigns and the Intensification of Corruption in China. Journal of Contemporary China 14 (42):93-116.
Wedeman, Andrew. 2010. Guilt and Punishment in China's War on Corruption. In Growing Pains: Tensions and Opportunity in China's Transformation, edited by J. C. Oi, S. Rozelle and X. Zhou. Stanford: Walter H. Shorenstein Asia-Pacific Research Center.
White, Tyrene. 1990. Postrevolutionary Mobilization in China: The One-Child Policy Reconsidered. World Politics 43 (1):53-76.
Whitford, Andrew B. 2002. Decentralization and Political Control of the Bureaucracy. Journal of Theoretical Politics 14 (2):167-193.
Whiting, Susan H. 2004. The Cadre Evaluation System at the Grass Roots: The Paradox of Party Rule. In Holding China Together: Diversity and National Integration in the Post-Deng Era, edited by B. J. Naughton and D. L. Yang. Cambridge: Cambridge University Press.
Whiting, Susan H. 2006. Growth, Governance, and Institutions: The Internal Institutions of the Party-State in China. World Bank Research Project on the Rise of China and India (Governance).
Whyte, Martin King. 1974. Small Groups and Political Rituals in China. Berkeley: University of California Press.
Wibowo, Ignatius, and Lye Liang Fook. 2006. China's Central Party School: A Unique Institution Adapting to Changes. In The Chinese Communist Party in Reform, edited by K. E. Brodsgaard and Z. Yongnian. London and New York: Routledge.
William A. Niskanen, Jr. 1971. Bureaucracy and Representative Government. Chicago and New York: Aldine Atherton, Inc.
Williams, Oliver. 1993. An Outsider's Perspective. In Public Administration in China, edited by M. K. Mills and S. S. Nagel. Westport and London: Greenwood Press.
Williamson, Oliver E. 1989. Transaction cost economics. In Handbook of Industrial Organization, edited by R. Schmalensee and R. D. Willig. Amsterdam: North-Holland.
Wolfe, Bertram D. 1984. Lenin and the Twentieth Century: A Bertram D. Wolfe Retrospective. Stanford: Hoover Institution Press.
World Bank. 2002. National Development and Sub-National Finance: A Review of Provincial Expenditures. Washington, DC.
255
Wright, Tim. 2004. The Political Economy of Coal Mine Disasters in China: "Your Rice Bowl or Your Life". China Quarterly:629-646.
Xiao, Guishu. 2002. Experiences and Lessons from the Socialist Construction in the Soviet Union. In Compilation of CCP Central Party School Teaching Scripts: Important Issues of the World (zhonggong zhongyang dangxiao jianggaoxuan: guanyu dangdai shijie zhongda wenti), edited by Y. Yu and C. Yang. Beijing: Central Party School Publishing House.
Xie, Chunhong. 2009. Research on Contemporary CCP’s Building of a Learning Party-State (dangdai zhongguo gongchandang jianshe xuexi xing zhengdang yanjiu): People's Publishing House.
Yang, Chungui. 2002. On Party School Education (dangxiao jiaoxue lun). Beijing: Central Party School Publishing House.
Yang, Dali. 2006. Economic Transformation and its Political Discontents in China: Authoritarianism, Unequal Growth, and the Dilemmas of Political Development. Annual Review of Political Science 9:143-64.
Zhang, Bailin, ed. 2005. PRC Civil Servant Law. Beijing: Zhongguo renshi chubanshe, Dangjian duwu chubanshe.
Zhang, Le-Yin. 1999. Chinese Central-Provincial Fiscal Relationships, Budgetary Decline and the Impact of the 1994 Fiscal Relations in China: An Evaluation. China Quarterly (157):115-141.
Zhang, Tao, and Heng-fu Zou. 1996. Fiscal Decentralization, Public Spending, and Economic Growth in China: World Bank.
Zhao, Wei, and Xueguang Zhou. 2004. Chinese Organizations in Transition: Changing Promotion Patterns in the Reform Era. Organization Science 15 (2):186-199.
Zheng, Shiping. 1997. Party vs. State in Post-1949 China: The Institutional Dilemma. Cambridge and New York: Cambridge University Press.
Zheng, Yongnian. 2006. The Party, Class, and Democracy in China. In The Chinese Communist Party in Reform, edited by K. E. Brodsgaard and Z. Yongnian. London and New York: Routledge.
Zheng, Yongnian. 2010. The Chinese Communist Party as Organizational Emperor: Culture, reproduction and transformation. London and New York: Routledge.
Zhou, Xueguang. 2010. The Institutional Logic of Collusion among Local Governments in China. Modern China 36 (1):47-78.