Training the Party

Charlotte Lee considers organizational changes taking place within the contemporary Chinese Communist Party (CCP), examining the party’s renewed emphasis on an understudied but core set of organizations: party-managed training academies or “party schools.” This national network of organizations enables party authorities to exert political control over the knowledge, skills, and careers of officials. Drawing on in-depth field research and novel datasets, Lee finds that the party school system has not been immune to broader market-based reforms but instead has incorporated many of the same strategies as actors in China’s hybrid, state-led private sector. In the search for revenue and status, schools have updated training content and become more entrepreneurial as they compete and collaborate with domestic and international actors. This book draws attention to surprising dynamism located within the party in political organizations thought immune to change, and to the transformative effect of the market on China’s political system. charlotte lee is Associate Director of the China Program at Stanford University’s Shorenstein Asia-Pacific Research Center. Prior to joining the Stanford China Program, she taught courses in Chinese politics, comparative politics, and international relations at Hamilton College and the United States Air Force Academy. Her research focuses on the institutions of authoritarian regimes, public bureaucracies, organizations, and contemporary Chinese politics. She holds a doctorate in political science from Stanford University and a Bachelor of Arts in political economy and Asian studies from the University of California, Berkeley.

Training the Party Party Adaptation and Elite Training in Reform-era China

Charlotte P. Lee Stanford University University Printing House, Cambridge CB2 8BS, United Kingdom

Cambridge University Press is part of the University of Cambridge. It furthers the University’s mission by disseminating knowledge in the pursuit of education, learning and research at the highest international levels of excellence. www.cambridge.org Information on this title: www.cambridge.org/9781107090637 © Charlotte P. Lee 2015 This publication is in copyright. Subject to statutory exception and to the provisions of relevant collective licensing agreements, no reproduction of any part may take place without the written permission of Cambridge University Press. First published 2015 A catalogue record for this publication is available from the British Library Library of Congress Cataloguing in Publication data Lee, Charlotte P., 1969– Training the party : party adaptation and elite training in reform-era China / Charlotte P. Lee. pages cm Includes bibliographical references and index. ISBN 978-1-107-09063-7 1. Zhongguo gong chan dang. 2. China – Politics and government – 1949 I. Title. JQ1519.A5C349 2015 324.2510075–dc23 2015004121 ISBN 978-1-107-09063-7 Hardback Cambridge University Press has no responsibility for the persistence or accuracy of URLs for external or third-party internet websites referred to in this publication, and does not guarantee that any content on such websites is, or will remain, accurate or appropriate. Contents

List of ﬁgures vii List of tables ix Acknowledgments xi

1 Introduction 1 2 The organizational landscape: party schools’ development and organization 28 3 Managing the managers: party schools as a pipeline to higher ofﬁce 55 4 Fusing party and market: introducing market-based incentives to the party school system 89 5 The entrepreneurial party school: party school responses to reforms 123 6 Adaptation measured: content analysis of party school training 157 7 Conclusion: risks and limits to party school reforms 178

Appendices A Number of party schools, by locale and national share of leading cadres 202 B Note on sources and research methods 203 C Central Party School organization 208 D City Z training allocations, 2008 209 E Descriptive statistics and robustness tests of PSM presented in Chapter 3 212 F Central Party School Mid-Career Cadre Training Classes descriptive data 221

v vi Contents

G International partnerships, central and provincial-level party schools 224 H Categories for coding training syllabi 227 References 229 Index 247 Figures

2.1 Local party organs involved in cadre education (district-level example) 36 2.2 Government organs involved in civil servant and cadre education (district-level example) 37 3.1 Frequency of administrative rank, 2003 China GSS 67 3.2 Prematch propensity score distribution 74 3.3 Postmatch propensity score distribution 74 4.1 Central-level organizations of cadre training 101 4.2 Share of total training, by school type 117 4.3 Central Party School trainee volume, 1977–2000 119 4.4 Timeline of domestic training providers 121 6.1 Percent of CPS training class time dedicated to partybuilding content, 1983–2007 160 6.2 Percent of CPS training time dedicated to orthodox theory, 1983–2007 163 6.3 Percent of CPS training time dedicated to the theories of reform-era leaders, 1983–2007 164 6.4 Percent of CPS training class time dedicated to management content, 1983–2007 165 6.5 Percent of CPS training class time dedicated to brieﬁngs, 1983–2007 166 6.6 Percent of CPS training class time dedicated to policy and law content, 1983–2007 167 6.7 Mid-career cadre training class training content, 2007–08 174 B.1 Cadre population and GDP per capita, coastal region 205 B.2 Cadre population and GDP per capita, central region 205 B.3 Leading cadre population and GDP per capita, coastal region 206 B.4 Leading cadre population and GDP per capita, central region 206

vii viii List of ﬁgures

B.5 Leading cadre population and GDP per capita, all regions 207 B.6 Cadre population and GDP per capita, all regions 207 E.1 Absolute distance between the propensity scores of each treatment observation and nearest neighbor match 216 Tables

1.1 The organization of public ofﬁcials in China, 1998 9 2.1 Highest educational attainment of cadres, by percent of 2003 survey respondents 32 2.2 Provincial-level party school leadership, 2009 40 2.3 City Z planned training classes, 2008, by type and enrollment levels 44 2.4 Percent increases, CPS mid-career cadre trainees, leading cadres, and CCP members, select years 51 2.5 CPS Mid-Career Cadre Training Class enrollment, 1980–2000 51 3.1 Descriptive summary of cadre and general population 66 3.2 First-stage probit regression results for propensity score estimation 69 3.3 Marginal effect for signiﬁcant control variables 71 3.4 Total cadres in treated and control groups, by propensity score block 73 3.5 Estimation of average treatment effect on the treated (ATT) enrollment in a non-degree party school training class 75 3.6 Estimation of average treatment effect on the treated (ATT) party school degree 76 3.7 Estimation of average treatment effect on the treated (ATT) university degree 76 3.8 Estimation of average treatment effect on the treated (ATT) party school training and university degree 77 3.9 Promotion patterns of Central Party School trainees, 1995 and 2000 classes 80 3.10 Career tracks of CPS trainees in stalled careers, 1995 and 2000 classes 81 3.11 Administrative rank, CPS Mid-Career Cadre Training Class alumni, 2010 84

ix x List of tables

3.12 Occupation category, CPS Mid-Career Cadre Training Class alumni, 2010 85 3.13 Comparing average ages, by rank 86 4.1 Summary of local party school expense categories and income sources 98 4.2 Provincial-level party schools and administration institutes 108 4.3 A selection of training bids, by locale and awardee 115 5.1 2001 Income and Expense Report, county-level party school in Province B 126 5.2 Local party school entrepreneurial activities 132 5.3 Provincial party school partnerships with Chinese universities 139 5.4 Shanghai party school and administration institute international partnerships 141 6.1 Central Party School training syllabi analyzed, by year 159 6.2 Shanghai party school pedagogies, by percentage of total class time 172 6.3 Description of mid-career cadre training classes, 2007–08 173 B.1 Administrative level of interviewees 204 B.2 Interviewee type 204 B.3 Interviewee gender 204 E.1 Pairwise correlations between independent and control variables 213 E.2 Descriptive summary of variables 214 E.3 T-tests for equality of means across treatment and control groups, before and after matching 214 E.4–6 Estimation of average treatment effect on the treated (ATT) using different matching algorithms 217 E.7 Probit regression results, effect of party school training on promotion 218 E.8 Marginal effect for signiﬁcant control variables 219 F.1 Age at the time of training 221 F.2 Gender 221 F.3 Home province at the time of training 221 F.4 Educational attainment 222 F.5 Percent with party school degrees 223 F.6 University major 223 Acknowledgments

When I began this project ten years ago, I had little idea that the road would be so delightfully peopled with teachers and friends, both old and new, who would bring much insight to the undertaking. Along the way, I have become indebted to many individuals on both sides of the Pacific, 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 during these years. This book first took shape as my dissertation at Stanford University, anditacquireditspresentformduringmytimeintheGovernment Department at Hamilton College, in the Department of Political Science at the US Air Force Academy, and at the Shorenstein Asia- Pacific Research Center at Stanford. Heartfelt thanks go first to Jean Oi. Her formidable skills in the field and in the classroom will continue to shape my approach to the study of China. I also could not have completed this project without the encouragement of two other readers at Stanford, 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. I am grateful for Mike’s big questions and his 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 Alice Miller, who has taught me volumes about the intricacies of elite politics in China. Joe Fewsmith sharpened the project by facilitating discussions with thoughtful audiences. I thank David Shambaugh for his generosity with documentary sources. The process of writing this book was also due in no small part to the support of my colleagues at Stanford, Hamilton, USAFA, and beyond. Thanks go to Yuen Yuen Ang, Xiaobin He, Xiaojun Li, and Kay Shimizu for reviewing drafts and advising on data analysis. Special thanks also to scholars from “across the Bay,” Leslie Wang and Rachel Stern, for including me in their China circles and providing invaluable guidance.

xi xii Acknowledgments

I am fortunate to be surrounded by so many talented China researchers, and I draw much inspiration from them. My deep thanks and appreciation go to each of my many friends and acquaintances in China. They remain anonymous here, but without their time and insights this book would not have been possible. I am grateful to have met so many Chinese teachers, officials, 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. I thank in particular Professors Chen, Fan, Hu, Mu, Xiang, Yu, and Yuan for their energetic guidance throughout my fieldwork. In a field where the data collection challenges are daunting, I am grateful for the research assistance of Kiki, Chelsea, Haihui, Shaorong, Xiaoya, Luke, and Zhaofen. My research would not have been possible without financial and administrative support from many organizations. The National Science Foundation, US Department of Defense Minerva Initiative, and Hamilton College all supported my research at various stages. At Stanford University, I remain grateful to the faculty and staffs of the Freeman Spogli Institute for International Studies, Shorenstein Asia- Pacific Research Center, and Center for East Asian Studies. At Cambridge University Press, I am thankful for the care with which my editor, Lucy Rhymer, her assistant, Fleur Jones, and a terrific copyediting team shepherded my project through the review and publication process. I also thank two anonymous reviewers for their careful reading of the manuscript, which improved the book in many ways. Finally, I am deeply grateful to many personal friends and family members for their boundless generosity and support. The entire “village” has kept me well nourished all these years. I remain especially grateful to my parents, Jean and Ken, and my brothers, Eric and Kennon, who have been a part of my explorations of China from the very beginning. I thank my husband, Nik, and daughter, Lesara, without whose humor, patience, and love this project would be much diminished. 1 Introduction

From its revolutionary roots to more recent reforms, China’s modern political system has prompted lively debates about regime durability. In the wake of the Soviet Union’s collapse, attention turned to the possibility of the demise of the Chinese Communist Party (CCP), driven by factors such as uncontrollable centrifugal pressures, demographic change, and institutional decay.1 The party’s dominance in the nearly three decades since the fall of other communist party-led regimes around the world, while defying some predictions, indicates that party institutions created for revolutionary purposes can negotiate key transitions. These transitions include responding to the ruling party’s current agenda of administrative reform and modernization. Understanding these shifts and the party’s durability over time necessitates an examination of the institutional underpinnings of this rising global power. Institutions have often been the object of inquiry in the study of authoritarian systems. Designing, constructing, and maintaining institutions of governance are vital to the state-building process, if not synonymous with it. Political institutions that constrain elected ofﬁcials in democracies are often established in autocratic contexts to serve the dictator’s (or leaders’) bid to stay in power. Such institutions facilitate the ordering of state and society and extend the coercive capacity of the ruler, and they do so across time and space. That institutions in authoritarian regimes often possess a complexity on par with their democratic

1 In policy journals, Minxin Pei (2002) has noted that the CCP’s growing weakness lies 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). Rowen (2011) predicts signiﬁcant political and/or economic change by 2020. 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. Chang (2001) focuses on incompatibilities between the Maoist state and the global environment. 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 2007). A discussion of those experts who are pessimistic versus optimistic about China’s political future can be found in Shambaugh (2008a: Chapter 3).

1 2 Introduction counterparts is not surprising, but the purposes of these nondemocratic institutions are at all times conditioned by the political context of which they are a part, that is, to sustain authoritarian rule. Given this core assumption, the task lies in discerning the functions served by a particular authoritarian institution and its impact on the individuals who guide and are guided by it. An additional undertaking lies in evaluating an institution’s capacity for coping with the uncertainty, 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 the displacement of systemic missions with more local, organization-specific objectives, and also 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.2 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 information (Cox 2009), and limiting economic predation by the autocrat (Gandhi 2008a; Wright 2008). 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 the dictator’s willingness to abide by nonarbitrary rules of the game.3 Parties are instrumental in solving this credible commitment problem. They 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, 2007). One mechanism for this is the allocation of party membership and positions; the

2 Surveys of the literature on the underlying logic for party formation include Magaloni and Kricheli (2010) and, on elections, Gandhi and Lust-Okar (2009). 3 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. Introduction 3 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 private entrepreneurs does not imply the end of high demand for party office. Critically, parties lengthen the regime’s time horizon for survival. Because of this expectation of regime durability, parties structure intra- elite conflict by offering elites the promise of “medium and long-term gains despite immediate setbacks” (Brownlee 2007: 12). A stable outcome may result as parties generate expectations that they will remain in power, which in turn promotes elites’ willingness to invest in development (Olson 1993).4 Single-party rule solves several additional problems of governance. Ruling 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 1999b).5 Party members cannot “retreat to the barracks” as military leaders might. Even more, by dispersing authority over a broader political base, parties provide a counterbalance to the threat of military coup (Geddes 2008). Parties also engineer outcomes with a “tragic brilliance”: the general population may accept corruption and suboptimal policies because of the party’s ability to maintain widespread patronage networks (Diaz-Cayeros et al. 2003). 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 creating ruling parties are engaging in several trade-offs – dispersing authority, investing resources in party

4 While party creation may mitigate this problem of incredible commitments, there are limits to this institutional strategy. Reforms in the USSR failed because the authorities of the Communist Party of the Soviet Union (CPSU) 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). 5 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 of 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 a revolution. However, after the CCP became China’s 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 Introduction organization rather than repression, and so on – this institutional choice ultimately enhances the durability of the regime. Forming a ruling party appears to be a successful 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.6 Among the institutions that a dictator may choose to establish or maintain, 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 book focuses on problems of party organization, particularly the organizations located within a ruling party and the individuals who guide those organizations.7 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, andhowtheselaythefoundationsfor stable single-party rule. Rather than treating parties as monolithic institutions, this study maps a more interior terrain. Its point of departure and focus is on variation in intraparty organization. This requires a look at 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

6 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 across 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, regional, duration, and other institutional variables to find that single-party regimes are significantly more likely to endure than other authoritarian regime types (pp. 30–2). The Hadenius and Teorell dataset, which spans 1960 to 2003, lends additional support to the Geddes finding. 7 More generally, I use Hannan and Freeman’s(1984)definition of organizations as “special corporate actors. Like other corporate actors, they are structures for accomplish- ing collective action as well as repositories of corporate resources. Unlike other collective actors, organizations receive public legitimation and social support as agents for accom- plishing specific and limited goals” (p. 152). Introduction 5 facilitate the complete control of state and society.8 Lenin’soriginal conception for the party was of an organization led by “professional revolutionaries” who were promoted from within the “rank and file” membership.9 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.”10 In contradiction to egalitarian ideological commitments, the party would be “an organization which of necessity is centralized” and governed by hierarchy.11 Thebureaucraticcentralismthat Lenin’s party eventually embraced was done unapologetically (Wolfe 1984:24–6, 192–95).12 The Communist Party of the Soviet Union (CPSU) 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 70 years (Klugman 1989). In theory, it was also to possess the organizational flexibility to respond to unforeseen circumstances and contingencies. With the global diffusion of Leninist principles, these parties have become highly structured and complex organizations, including extensive functional differentiation of constituent parts.13 A range of subparty organizations play a supporting role in the maintenance of the party’s political authority: propaganda bureaus, party personnel departments, courts, unions and other mass organizations, party schools, and the like.

8 See Lenin’s 1902 essay, “What Is to Be Done? Burning Questions of Our Movement” (Lenin [1964]: 347–529). 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: 148–52). 9 “What Is to Be Done?” (Tucker 1975:75–7). 10 Ibid., p. 90. 11 Ibid., p. 86. 12 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). 13 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 back- wardness 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). 6 Introduction

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. Leninist systems are characterized by a critical relationship that is often overlooked in general studies of parties in autocracies: party management of the state bureaucracy. Party organization, specifically party integration with and dominance over the bureaucracy, constitutes a source of political power (Barnett and Vogel 1967;Selznick1960). As the most prominent example of an extant ruling party formally organized along Leninist lines, the CCP maintains and reinforces its organization through party penetration of the state.14 While there have been attempts to draw an analytical and empirical line between the party and state in China (Zheng 1997), in practice the two political bodies remain intertwined.15 Existing work on the Chinese case characterizes the relationship as suffused with bargaining and negotiation (Lampton 1987, 1992; Naughton 1992); a reflection of elite conflicts (Dittmer 1978); and, above all, distinguished by party domination and coordination (Harding 1981;Li1994;Schurmann1970). In this sense, the state 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, or bureaucrat of the party and/or state apparatus, 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. In this sense, the party presents an organizational means to solve a political selection problem. This function is both separate from and part of the elite bargaining function noted previously. The party is an organized

14 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. 15 E.g., ofﬁcials occupy party and government ofﬁces simultaneously, the government funds party bureaus, and party and government training centers are often integrated on the same campus. Introduction 7 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. More simply put, “Leading cadres are at the head of the reform train [in China]. We must develop these leaders, otherwise reforms will be fruitless.”16 In light of the critical role played by those institutions that control who joins the party elite, this book will focus on party strategies of both bureaucratic management and political control.

Controlling China’s political elite 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. 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, super- iority, and domination” (Farmer 1992: 2). This is consistent with Putnam’s(1976) emphasis on those who “influenc[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. Drawing from Mills’ (1959) precedent, in which the “power elite” are those in positions of authority, this study employs a positional 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 vice-county magistrate or equivalent are considered members of the political elite within China.17 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 may overlook to some degree the increasing diversity in Chinese society, where entrepreneurial talent, global connections, and political authority may be interconnected but separate bases of political influence.

16 Interview 120, Central Party School professor, February 2008. 17 Leading cadres in China are those ranked at the vice-county (fuxianji or fuchuji) level or above. See COD (1999), p. 589, for a discussion of these deﬁnitional issues. 8 Introduction

In a Leninist system, cadres are responsible for party and government work at various administrative levels 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 Peoples’ Republic of China’s (PRC).18 “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), who had some specialist ability, and who carried out ‘mental’ rather than ‘manual’ labor” (Burns and Bowornwathana 2001: 23).19 More bluntly, a cadre is anyone who “eats the state’sgrain” (chi guojia liang shi).20 At present, the Chinese bureaucracy, in all its organizational variety, comprises over 40 million individuals.21 Table 1.1 offers a sketch of the size of the entire bureaucratic system and levels within the system. 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.22 A slightly larger population of “leading cadres” (lingdao ganbu) 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 leading cadre class produces the policies that the rest of the bureaucracy must implement (Burns 1989a, 2006). In 1998, 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. Controlling promotion to and within this latter group, the senior cadre ranks, is a crucial arena for the party’smaintenanceof“organizational health” (Nee and Lian 1994). This is especially critical in a system as decentralized as China’s (Landry 2003). Inability of higher-level authorities to manage party and government reformers is tantamount to a loss of party

18 Interview 112, Central Party School professor 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 state ministry system. 19 Burns draws from Strauss’ distinction between “lettered official” (wenguan), public servant (gungwuyuan), and cadre (ganbu) and Lee (1991) for this definition of cadre. 20 Interview 112, Central Party School professor, February 2008. 21 In 2005, the size of the cadre population was 47.78 million, which represents about 3.1 percent of the total population (Ang 2012). 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. 22 Changes in nomenklatura are reviewed in Burns (1994, 2003). Introduction 9

Table 1.1 The organization of public ofﬁcials in China, 1998

“Leading” (lingdao) Administrative level Administrative rank Population size cadres? Examples

Township and Section (ke) level ~46 million No Section head in a below and below county-level ministry, township party secretary County Deputy department 500,576 Yes County party (fuchu) and secretary, department (chu) mid-level level supreme court judge City Deputy bureau 45,688 Yes City mayor, (fusiju, fuditing) provincial and bureau (siju, party school diting) level principal Provincial Deputy ministry 3,665 Yes Ministry head, (fubu) and politburo ministry (bu) member level Central Premier (zongli) Yes

Source: COD (1999) and Heilmann et al. (June 2000). 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, 2008b;Wang2002;Xiao2002). Elite personnel decisions are a paramount responsibility of the party (Naughton and Yang 2004). Complicating this, 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 10 Introduction particular circumstances and challenges at the onset of reforms in the late 1970s, as the Chinese state was “growing out of the plan” (Naughton 2007:92–3). Comparatively low bureaucratic turnover during the post- Mao economic transition, which commenced in 1978, 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, but political constraints made a purging of party managers unfeasible. 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. 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). Hence, the rallying cry was to develop a “revolutionary, younger, better educated, and more technically specialized” (geminghua, nianqinghua, zhishihua, zhuanyehua) cadre corps.23 This bureaucratic transformation was to take place in the context of unprecedented economic development. With the initiation of reforms in 1978, China’s economy underwent dramatic change in terms of growth and industrial development. Ofﬁcial annual growth rates averaged 9.98 percent for the period 1978 to 2011.24 Industrialization also took off in urban special economic zones and through “local state corporatist” strategies in the countryside (Liao et al. 1999;Oi1992, 1998a). The economic miracle presented by contemporary China has a seemingly incongruous basis in a single-party authoritarian regime, which begs

23 See Deng Xiaoping’s “Opening Speech at the Twelfth National Congress of the Communist Party of China,” 1 September 1982, available online at http://archive.org/ stream/SelectedWorksOfDengXiaopingVol.3/Deng03_djvu.txt, accessed December 20, 2012. See also Manion (1985b) for a discussion of the personnel policies resulting from these “four transformations” (si hua). This idea of well-trained and professionalized cadres leading the modernization drive was repeated in a speech before the Politburo, 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 March 6, 2010. 24 World Development Indicators, World Bank, available online at http://data.worldbank. org/country/china, accessed December 11, 2012. Introduction 11 further examination of how the ruling CCP has maintained organizational discipline during this period of rapid and apparently successful economic liberalization.25 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 the 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. With their totalizing emphasis on party control over all political, economic, and social activities, Leninist party-led systems traditionally rely on monopsonistic control by the party over labor markets.26 Over the course of market reforms, skilled labor that was formerly dependent on state entities for upward mobility found new 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 the data from a survey of post- communist countries taken in the early 1990s: bureaucratic turnover ranged from a low of approximately 25 percent in Russia to 51 percent in the Czech Republic during 1988–93.27 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. Occupational change among cadres from 1988 to 1993 was much lower when compared to their transitioning European and

25 China’s combination of state-led development and unprecedented growth has challenged assumptions about the correlates of economic prosperity. 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 1999b; Przeworski et al. 2000). 26 I.e., the party was the sole buyer in a labor market comprising many sellers of political, managerial, and administrative talent. 27 See the 1993 Social Stratification in Eastern Europe after 1989: General Population Survey. These figures were calculated by counting the survey respondents who reported cadre occupations in both 1988 and 1993. I defined ‘party and state managers’ as individuals reporting occupations that fell in the ‘legislator and manager’ category of the survey (ISCO codes 1000 to 1319). The countries surveyed included Bulgaria, Czech Republic, Hungary, Poland, Russia, and Slovakia. Unfortunately, this data does not capture whether communist party-era cadres opted for non-party and non-state sector professions over the course of transition or were ousted by incoming political elites and forced to turn to private sector alternatives. 12 Introduction

Russian counterparts. In one representative national survey, only 7.3 percent of Chinese bureaucrats left the party or government ranks during that five-year period.28 From the onset of reforms in 1978, bureaucratic turnover is not much higher. Only 15.8 percent of survey respondents reported leaving their party and/or state posts by 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 1993 was due almost entirely to retirement; only one cadre reported joining the private sector during the period between 1978 and 1993.29 These patterns may be explained by the particular 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 2009;Solnick1998: Chapter 7). There also exists the possibility that individuals retained their official office while “moonlighting” in private ventures. 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 China’s administrative class to cope with the implementation and management of economic and social reforms.30 How has the party adjusted its organizations of bureaucratic control and management to account for building a new political elite? In a Leninist system in particular, party organizations designed for a command economy and ideologically disciplined cadre corps would seem outdated and out of place in a decentralized market economy, one with increased autonomy for decision-makers. The organizational puzzle posed by the case of the CCP is the persistence of seemingly anachro- nistic party organizations in the post-Mao period. Organizations forged during and for a revolutionary context have limited purchase in the management of a state no longer bent on revolution but rather focused on routine. Scholars have unpacked the many reforms contributing to

28 These percentages are calculated from the 2003 China General Social Survey. I defined “party and state managers” as individuals responding to a survey question on administrative rank in 1988 and reporting non-party or non-state occupations in 1993. 29 Over the 1978 to 1993 period, 28 individuals reported retiring out of a survey population of 183 individuals. 30 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). Introduction 13 the “remaking of China’sleviathan” and focused in particular on reforms of the administrative state.31 The remaking of political bodies within the party has occurred more slowly than this administrative transformation, and deeper political reforms have lagged behind state reforms in China’s push to modernize its economy. Changes have proceeded slowly and cautiously, but it has been impossible for the party to ignore the pressures to adjust. 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 increased wage fund, and so on become irrelevant to success in a marketized context” (Hanson 1995: 312). The pivotal issue becomes how to retrain these bureaucrats, and the strategy adopted has implications for the political and administrative future of China’s party-state. While the state education system could take on some of this burden of re-educating managers, regular universities and schools might not promote entirely “correct” ideas.32 In a field interview with a provincial-level party school teacher, he declared, “Ideological training must be preserved. You can’t have liberal-minded (ziyou zhuyi) university teachers teaching cadres; this task can’t be divided.” To continue exerting party control over individual bureaucrats, it would seem logical for party officials to draw on existing organizational resources. CCP organizations forged during the early, underground days of party activism have persisted into the present and offer one solution to the question of how

31 Scholars have identified six waves of administrative reforms and restructuring that have swept through the post-Mao Chinese state (Wang 2010). For an overview of the reforms carried out between 1979 and 1982, see Burns (1983). In 1982, efforts to reduce the number of state agencies resulted in a decrease of State Council–managed agencies from 100 to 61 (Yang 2001;Yongnian2004). After the 13th Party Congress of 1987, there was again an effort to streamline agencies under the State Council, and personnel cuts numbered in the thousands at the central level (Yang 2001:24–6). In 1993, fiscal recentralization asserted the center’s control over provinces, and there were accompanying reforms to institutions of taxation. This year also marked the creation of the State Economic and Trade Commission, a powerful economic bureaucracy headed by Zhu Rongji. Conservatives such as Premier Li Peng, however, blocked efforts to carry out more dramatic administrative reforms (Yongnian 2004:90–3). Major changes took place in the late 1990s, and government restructuring in 1998 reduced central government minis- triesfrom40to29(Yang2004). By the 2000s, reformers had shifted focus to public service provision and transparency. In addition to these concerns, reforms carried out in 2003 also deepened structural reforms initiated in 1998. Administrative reforms carried out in 2008 saw the creation of five “super-ministries” in an effort to rationalize government functions. 32 Interview 211, provincial party school professor, May 2008. 14 Introduction the bureaucracy shall be reformed into a politically appropriate but pro- fessionally competent “organizational weapon.” A key issue is how to reform these revolution-era organizations to match current needs. 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 are an understudied but critical component of the organizational life of communist party systems. These schools exist to “inculcate the desired attitude to the Party” on the part of new party recruits (Meyer 1961: 112). Crucially, they are responsible for the ideological training of revolutionary cadres.33 Schools embody the party emphasis on discipline, correct thought and action, and organizational unity. They are sites for reinforcing individual commitment to the party.34 In principle and to some degree in practice, party schools provide the organizational space for forging ideal cadres. Of importance is how schools carry out these functions and how they remain relevant in changing contexts. In the case of China, party schools provide 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 2,800 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.35 The centrality of party schools within this organizational landscape is due to their exclusivity, since they have historically been sites for educating party members and

33 In his comparative research, Meyer finds similarities in school structure, training requirements, and training content across party schools of Europe. See Meyer (1961), pp. 159– 69, 206–9. 34 As such, they “are in no way divorced from the whole process of the molding of the Communist. Rather, they occupy in that process key points, nodes of intense development. Far from being regarded as periods of retirement from ‘the class struggle,’ from the constant pressing day-to-day existence of the cadre Communist, they are conceived and organized to carry that urgency of continuous commitment to an even higher level” (Meyer 1961: 162). 35 This includes cadre schools located in party organs such as state-owned enterprises, 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). By 1982, this was reduced to 8,000 (Paltiel 1990: 588). Introduction 15 officials.36 These schools are a prime example of Leninist party organizations that would appear incompatible with a market economy and the changed domestic and global circumstances facing reforming China. Yet, reforms in cadre training have been one way to meet the demands imposed by economic change and modernization. This study of party schools argues 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 a deliberate embrace of market mechanisms by central party authorities and the introduction of organizational competition, or redundancy, to the system.37 Decades of reform have yielded more dynamic, relevant party schools. The process of marketization carries risks for party authorities, however. While party schools continue to serve critical roles in cadre education and promotion, incentives now exist at the organization level for party schools to embrace market opportunities, sometimes at the cost of party discipline. Through case studies and extensive fieldwork, this book makes several contributions to the state of knowledge on political institutions in China. First, this investigation probes how party authorities have induced organizational change within the party school system and how this provides traction on the larger, multifaceted story of CCP survival in the reform period. This study builds upon the observation of many China scholars (e.g., Miller [2008], Shambaugh [2008a]) that the party has embarked on a broad-based institution-building project. Arguably, this has been in

36 Party schools such as those located within universities are also tasked with training party activists. 37 Recent scholarship has placed party schools and cadre training in the context of general processes of party adaptation. Party schools reflect the party’s efforts to study and learn from cases of failed communist party-led reforms elsewhere (Shambaugh 2008a: Chapter 7, 2008c). With these historical lessons in mind, recent passage of training- related legislation reasserts organizational discipline over cadre ranks. 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 June 6, 2006 and the “2004 CCP Party School Work Trial Regulations,” available at http://wlzx.hdpu.edu.cn/zzb/Article_Show. asp?ArticleID=126, accessed June 6, 2006. Scholarship has also moved beyond central training organizations in 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 (2009) has probed how party schools reflect the party’s efforts to redefine socialism in a changed context. 16 Introduction 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 2009: 18). My contribution is one of specifying mechanisms and processes: I unravel why we continue to observe significant investment in party schools 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 mesolevel, or institution-driven, view of how the party pursues its fundamental desire to not only survive through present reforms but remain their central architect. Second, it demonstrates with survey 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. In so doing, it 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 in the selection of cadres, specifically by measuring the effect of party school training on cadre promotion. The third contribution of this study is to propose and assess specific indicators of organizational adaptiveness. The content of party school syllabi has shifted over time, and one question is whether to view these changes as adding to or detracting from processes of adaptive change. Through field visits and interviews conducted in the local party schools of provinces in the coastal and central regions of China, I present a ground- level understanding of change within the party. This allows for both a vertical and horizontal examination of how these 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.

Sources of adaptive capacity Scholars of Chinese politics have examined various pathways to institutional change in reforming China, some of which are bottom-up in orientation and others which are elite-led. Informal institutions devised by local actors in response to state strictures can become the drivers of formal institutional change. This was the case with the blooming of grassroots capitalist activity in the reform period, compelling party officials to recog- nize and then co-opt capitalist practices within the party (Tsai 2006). This pathway of change is particularly striking due to the ability of local non- state actors to drive change upward and throughout China’s political and economic system. However, such change is precluded in arenas of political Sources of adaptive capacity 17 life which, short of reforms initiated within the party itself, are closed to non-party actors. Initial shifts in ideology, for example, are party-led. A second body of work has focused on the role of local experiments in stimulating systemic change (Heilmann 2008a, 2008b;Heilmannand Perry 2011). Local experiments, which have roots in the party’sparticular historical experience, are one means for authorities within the Chinese political system to assess new policy directions and reproduce those which have potential for nationwide implementation. This type of change also relies on grassroots action which higher authorities may choose to replicate as part of a larger national agenda. While policy experiments have affected a wide range of policy arenas and are intraparty in orientation, they are limited in their ability to move central political organs of the party. Third are studies of top-down, elite-led institutional change. These have traced the search by party leaders to avoid the mistakes of communist parties elsewhere (Shambaugh 2008a) and conceive of institutional change as an indicator of systemic rationalization and greater inclusiveness by the ruling party (O’Brien 1990). Such studies offer detailed historically grounded analyses of change but do not embed the Chinese case in a broader comparative framework. The present study seeks to identify causal mechanisms with more general applicability. In the post-Mao period, large-scale 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 dysfunc- tion that Merton observed across bureaucracies is creating incentives for “adaptive efficiency,” that is, “an institutional structure that in the face of ubiquitous uncertainties ... will flexibly try various alternatives to deal with novel problems that continue to emerge over time” (North 2005:154).38 Such adaptiveness is distinct from and a subset of observed organizational change. While organizational change implies that some dimension

38 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 anticipa- tion 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” (Tucker 1975: 110). 18 Introduction of a unit is different in period t from period t+1 or t−1, adaptation speaks to the ability of an organization or a system to anticipate or respond to environmental change such that organizational changes achieved over time enhance that unit’s likelihood of surviving in a new time period. Adaptation, when it is either anticipatory or reactive, is not without risks. An organization, in the process of attempting to adapt to changes in its environment, may choose unwisely and inadvertently set in motion the conditions for its decline (Hall 1976; Zammuto and Cameron 1985). There is also a degree of uncertainty to environmental change such that an adaptive change made in response to one shift in the environment may lead to organizational maladjustment in the face of a different environmental context. Of interest is how the CCP has generated adaptive solutions to novel challenges within existing party organization. Leninist parties such as the CCP have incorporated many features of a Weberian bureaucracy. Such a bureaucracy, indispensable to the modern state, draws on rules, offices, and expertise to govern bureaucratic behavior and administration. Organizational rationality derives from functional specialization across bureaucratic offices. Leninist parties comprise several of these features, notably the organization of party units according to the various needs bound up in the 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, creating bureaucracies within the bureaucracy. A central organizing principle of such systems 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). Lenin’s institutional innovation on the Weberian bureaucracy, the creation of a hierarchical party system populated by “vanguard” revolutionary party members, and one that would guide society out of a capitalist- led state system, would seem an unlikely candidate to weather through significantchangessuchassystem-wideeconomictransition.Asthe collapse of Leninist party systems in Eastern Europe and the Soviet Union testify, the internal structure of these party-states proved incapable of withstanding the various stresses of party-led reforms and external change. Sources of adaptive capacity 19

Organizational redundancy: replacing monopoly with partial competition 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 interorganiza- tional 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).39 Redundancy, taken to the realm of governance and public administration, refers to the introduction of additional agencies to fulfill an organizational goal previously mono- polized by a single agency.40 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 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, that is, competition as a means to

39 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 lies in the nature of the incentives driving competitive systems. Competitive, as opposed to non-competitive, redundancies offer 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 threat to survival. 40 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. Downs (1967) 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. Interagency competition may develop accidentally or by fiat. These are not mutually exclusive pathways, and both may play a role in the development of competition in a bureaucracy. 20 Introduction induce organizational change and adaptation to new circumstances. In their study of interservice 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 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 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. This is a logical response to the uncertainty that waxes and wanes in different political conditions. 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 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 debated how to cross the river of economic change. The party leadership has proceeded 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. 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 incentivize 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 at the system level, but the participation of private market actors is not a necessary precondition for these processes to unfold. Sources of adaptive capacity 21

Some additional design considerations accompany the decision to build a redundant bureaucratic system. Competition will lead to the highest levels of creativity when three conditions prevail.41 Downs (1967) asserts that competing agencies must be close enough in purpose that their funding derives from the same sources. This transforms 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. Significant overlaps in human capital may decrease overall creativity. 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 entrantsofferalternativesthatareofsimilarorhigherqualitythanthe original monopoly agency. If competition presents an inferior standard, this has the perverse outcome of lowering standards throughout the system. Several problems can attend the introduction of competition to a bureaucracy. There 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. These uncertainties can include whether innovations preserve the existing system or plant the seeds of instability. In a hierarchical political system such as China’s, innovations often carry the risk of strengthening the hand of locales against that of the central state. Another concern is the opportunity cost of devoting resources to a redundant function when those resources might be committed elsewhere. The problem here is the difficulty in assigning costs to a given outcome as well as observing the counterfactual case. There are also deeper considerations such as coping with the possibility of market failure and the suitability of redundant systems for nonexclusionary goods.42 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;

41 Bendor (1985) explores the criticality of independence across agencies, under the assumption that nonindependence might risk 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. 42 Landau 1991 raises but skirts these issues in his advocacy for introducing redundancy as a virtue and not a sin of public administration design. 22 Introduction

Felsenthal 1980). Overall, these critiques present some limits on the universal application of redundancy to government functions. Finally, bureaucratic competition 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 and 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. Whether and how monopolistic agencies resist the introduction of competition is an additional focus of the empirical chapters of this book. In sum, 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.43 Aversion to innovation by monopoly agencies within the CCP lies at the heart of the party’s concern with the old state of affairs in cadre training. As this study demonstrates, central party authorities deliberately turned to market- based competition to induce change in these party organizations.

Bringing in market processes Interagency competition is one among several changes that have affected cadre training in China. Over the past three decades, reforms within party bureaus have taken a market turn. This marketization encompasses a bundle of processes that have resulted in the creation of a new organizational environment for cadre training. Ideally, markets comprise three interrelated processes: free exchange between buyers and sellers for a good or service, prices dictated by supply and demand, and free entry and exit of market actors. All of these characterize, to some degree, cadre training in China today. Marketization has remained incomplete due to

43 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 exceptional- ism in this regard. Sources of adaptive capacity 23 significant interference by the party and continued dominance of party actors. The response of party schools to these changes is also shaped by additional market opportunities which have emerged in the reform period. The market in cadre training exists alongside more general markets for the goods and services that entrepreneurial school leaders may choose to offer: leases for plots of school land, facilities rentals, tourist services, and so forth. Subsequent chapters will detail developments in both of these markets. There is now a broader range of sellers of cadre training services, including Chinese universities, training schools managed by various bureaus of the party, and schools located abroad. These sellers alter- nately compete for or are allocated training contracts. Buyers of cadre training content have diversified as well. These actors now include bureaus of the party-state, private-sector entrepreneurs, and everyday citizens. All have become consumers of the myriad services offered on party school campuses. Prices for training courses continue to be subject to negotiation between party actors, but party schools must also compete with bids from outside sellers. No longer are exchanges dictated by one- and five-year training plans. Still, party authorities retain the authority and ability to interfere with these exchanges, as training plans and ad hoc dictates from central authorities still determine, to a degree, the activity of party schools. Importantly, there exist distortions in the free entry and exit of training providers. While non- party providers may enter and exit at will, party schools may not shutter their doors. Party schools remain a privileged category of training providers. Many of these schools are still guaranteed some minimal floor of training revenue, though schools often supplement these with additional entrepreneurial activities. Among market processes, subjecting party schools to competitive pressures has generated the strongest incentives for organizational change. This is because “markets promote high-powered incentives and restrain bureaucratic distortions more effectively than internal organization” (Williamson 1989: 150). Since some degree of risk accompanies market- based competition, the stakes are higher than for a monopolistic bureaucracy. 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 is ascertaining how many agencies, or how many providers of a service, is optimal. To resolve this, it is possible 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, actors will continue to enter the 24 Introduction market for a particular good or service until there is no longer a marginal gain for additional entrants. 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 heightened market-based competition for resources, encompasses several interrelated activities that include updated service provision, programmatic innovation, and the search for lucrative new ventures.44 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. All indicate the party’s capacity for significant organizational rethinking behind the veneer of a relatively unchanged political structure.

Précis of study This study argues that the CCP has selectively enhanced the adaptability of subparty organizations by employing market mechanisms to incentivize organizational change. Party schools, as relatively understudied sites of political control and bureaucratic management, offer a window into the restructuring of incentives and how the CCP has exhibited surprising adaptability in the face of signiﬁcant economic and social change. Competition between providers of cadre education has spilled beyond the boundaries of the party-state, but a key motivation has been to improve the adaptive capacity of party organizations. This argument

44 This study departs from previous studies of competition within the Chinese bureaucracy in several ways. Existing work has 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 People’s Republic of China’s (PRC) intellectual property rights protection regime and finds that redundant systems are more effective at monitoring compliance with state regulations. Crucially, competition in cadre training brings market principles to party bureaucracies with principally political functions. 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. Second, 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. 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. Solinger (1992) and Duckett (2001), among others, have noted the entrepreneurial nature of both party officials and new private actors in the early reform period. This study approaches party-based entrepreneurialism from a different tack. In the party school case, official incentives for risk-taking activity now structure the behavior of party officials. Précis of study 25 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? Have there been unintended consequences, either welcome or not? Are party schools still relevant? Findings at two levels of analysis offer answers to these questions. Individual-level career patterns and the “treatment” of party school training on career paths show that party schools remain an organized means for the party to manage critical processes of political elite selection. In carrying out this selection function, the party school system has been subject to competitive pressures. School-level analysis will map out organizational responses to centrally driven reforms and new policies. Chapter 2 begins with an overview of the party school system, its history, and organizational context. Existing research on party schools is classified into roughly two groups: studies that focus on the functions of the Central Party School (CPS) and those that look at the school system beyond Beijing. Scholars have focused on party schools as indicators and drivers of ideological change within the party. This study, however, takes a different tack and emphasizes processes of organizational change as they unfolded throughout the system, in the CPS and beyond. This chapter also presents an intraparty comparison to demonstrate that not all party institutions have fared as well as party schools in the reform period; party schools have become more robust while other Mao-era institutions of political control, such as the campaign, have waned in importance. In light of the reform-era investment in cadre training, Chapter 3 explores the theoretical and empirical relationship between cadre training and elite selection. In the principal–agent relationships which suffuse the Chinese political system, the party’s selection problem comes prior to other problems, more commonly studied, in a principal–agent relationship (i.e., moral hazard, which is solved by monitoring, rewards, and sanctions). This chapter tests whether selection for training at a party school constitutes a channel for promotion to higher cadre office. By employing a matching method on survey data, to control for selection bias, this chapter presents 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 considers mechanisms for selection, including screening and signaling. Chapter 4 shifts the level of analysis to discuss the marketization of cadre training, uncovering how market mechanisms were introduced to the party school system. Beginning in the mid-1980s, different sets of 26 Introduction 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 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 that a combination of collaboration and rivalry characterizes the relationship between the organizations that now compete for cadre training contracts.45 This chapter also discusses an important precondition to this marketization strategy, that is, limited fiscal and administrative decentralization. Chapters 5 and 6 peer inside party schools to unpack the various school responses to competition and the development of an “entrepreneurial sensibility” within these organizations (Eisinger 1988). 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 schools’ training outputs. Changes observed in the party school system have parallels in the commercialization of China’s media, though differences exist due to variation in the core missions of these organizations. Chapter 6 presents indicators of adaptive change and applies them to content analysis of training syllabi from party schools at the central and local levels. Taken 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 have adapted to two imperatives: maximizing income streams in a new market economy and updating the content of cadre training. The concluding chapter considers the implications of these changes. One result of party schools’ search for new income-generating projects has been greater embeddedness in local economies. This trend 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

45 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. Précis of study 27 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 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.46 The particular approach chosen by the CCP, that is, 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 by party institutions, at the same time retaining the party’s hold on the loyalties and careers of ambitious cadres.

46 A critique of the rigidities of the socialist system can be found in Kornai (1992). 2 The organizational landscape Party schools’ development and organization

Throughout its reform period, the CCP has held firm to the principle of “party management of cadres” (dang guan ganbu) and from this 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 on how the party has exercised authority over the career paths and political knowledge of its leading managers and administrators. This chapter places party schools in their political and institutional context through a discussion of the history, organization, and management of party schools. Examining the development of the party school system demonstrates that while these schools are grounded in the early ideals of the party, they now reflect and embody the pragmatic objectives of more recent organizational reforms. Across these changes, party schools have carried out a trinity of responsibilities: maintaining the party line, conveying major policy winds, and training officials in those skills that advance party priorities. Existing scholarship has focused most on the ideological function served by party schools.1 Early analyses centered on the Beijing-based Central Party School’s role in developing and conveying the party line. The CPS possesses considerable ideological authority and has been the site of breakthroughs in party theory. Under Hu Yaobang’s leadership of the Central Committee (1982–87), the CPS became a locus of formulation of “practice as a sole criterion of truth” and debate of the post-Mao “two whatevers” ideological stances (Schoenhals 1991). The CPS also carries symbolic weight as the site of major declarations of party doctrine. Party leaders use the school as a platform for delivering key speeches on the party line. Jiang Zemin elaborated on his “Three Represents” there in May 2000, followed by Hu Jintao’s call for a “harmonious society” in

1 An early study of cadre training identiﬁed “ideological remolding” as the foremost purpose of party schools (Tang 1961:7–9). A review of the literature on party schools, both pre-1949 and after, can be found in Pieke (2009), p. 35, fn. 19.

28 Creating the party school system 29

2005 (Dickson 2008: 76; Zheng 2010). There has been a tradition of the party general secretary giving a major speech at the CPS in advance of a party congress as a means to foreshadow major themes and initiatives taken up at the congress.2 In a June 2007 speech at the CPS, for example, Hu Jintao expounded on the major themes of the 17th Party Congress, several months before the October convening of congressional delegates (Fewsmith 2007: 7; Shambaugh 2008a: 111–15).3 Other studies have considered party schools’ responsibility to train ofﬁcials 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 faculty enjoy in promoting their research and policy recommendations on political reform (Fewsmith 2008). Prominent faculty 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.”4 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).5 There exists less knowledge on the larger system of cadre training institutions anchored by the CPS. In limiting their scope to the Central Party School, many studies do not consider when, why, and how cadre training more broadly may be a useful instrument of political control for succes- sive generations of party leaders.

Creating the party school system A tour through the history of cadre training in the CCP and PRC demonstrates that party-led cadre education has experienced an uneven trajectory over time. Early, limited efforts to institutionalize cadre training were followed by the decade-long disruption of the Cultural Revolution, and

2 Jiang Zemin gave speeches at the CPS prior to the 15th and 16th party congresses, held in 1997 and 2002, respectively. Hu Jintao and Xi Jinping continued this practice. 3 Themes included a reaffirmation of market reforms and building of “inner party democracy” as well as reiteration of Hu’s “scientific development concept.” 4 “Beijing Brain Trusters Propose a New Path for the Political Reform in China,” Yazhou Zhoukan, May 27, 2007. See also, “CPC National Congress to Launch New Resolution on Intra-Party Democracy,” Hong Kong Hsin Pao, March 23, 2007. 5 In the present period, scholars from the CPS’s Institute for International Strategic Studies have engaged with research centers such as the Woodrow Wilson Center in Washington, DC, and the University of Kentucky, Louisville, on topics ranging from US foreign policy to US political institutions. CPS officials and faculty have also conducted study visits with academic institutes of the US military such as the Asia-Pacific Institute for Security Studies. 30 The organizational landscape the reform period has seen 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 (1912–49). 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 Marxist-Leninist historical narrative and accompanying weapons of the party – such as how to formulate class-based rhetoric, disseminate propaganda, and establish base camps – took place in party schools. While the concept of training schools for cadres and other revolutionary actors, such as workers and peasants, has existed since the early years of the CCP, these organizations have undergone contraction and expansion from the first half of the twentieth century to the present. Precursors to party schools existed within China, during the tumultuous period of warlord and Guomindang rule, and abroad. Early sites of cadre training included study societies of the May Fourth Movement and the Self-Education College (zixiu daxue) founded in Changsha by Mao and other leaders in 1921 (Price 1976:11–52).6 At the same time, Soviet schools such as Moscow’s Communist University of the Toilers of the East and Communist University of the National Minorities of the West received Chinese communist party organizers. The CPSU also provided inspiration for early party school curricula, particularly the theoretical component of training classes (Tang 1961). With the Soviet example to draw from and the imperative to build organization skills among CCP leaders, the movement to establish a party school system within China gained momentum. As the CCP developed its membership base in the 1920s, leaders envisioned a division of labor across party members: cadres devoted to internal party organizational matters and subordinate members responsible for mass activity in the cities. Given this strategy, the first party schools were established in Hunan province by party members and Communist Youth League members in 1923 (Wang 1992:33–4).7 By 1926, CCP leadership created a domestic party school system, modeled after the Soviet Union’s, for training party cadres responsible for either internal party matters or mass work. These early schools offered short courses, lasting from two

6 Price notes that students studied Marxism-Leninism through “sit and talk” (zuotan hui) sessions, a format still used in party schools today, and by 1922 the college was engaged in training classes for party leaders. 7 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. Histories of the party school system are available in English (Pieke 2009: Chapter 2) and Chinese (Chen 2007) scholarly work. Creating the party school system 31 weeks to three months, on mass work and party organization, with a heavy emphasis on mastering theoretical texts.8 With the collapse of the party’s urban strategy and subsequent retreat to bases in the countryside, party leadership retained the institution of party schools for cadre training. During this period, military exercises and field work, both components of party school training programs today, replaced the earlier overemphasis on study of theory (Ibid.: 78–81).9 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). From the formation of the party in the 1920s, its decimation at the hands of the Guomindang, retrenchment during the civil war, and eventual triumph in 1949, party schools remained a feature of party organizational life. Price points to the logic underlying this persistent dedication of party resources to cadre training during these first decades of party development. She notes, Of course, the Chinese Communists have had other options for enforcing compliance. Their leadership policy – in theory – could have placed more emphasis on material rewards or external coercive pressure such as the Soviet system of terror. However, the Chinese Communists did not take power through a rapid seizure of well-developed institutions. Nor did they have access to large material or man- power reserves. To meet the demands of over twenty years of revolution, war, and nation-building, they chose to maximize their organizational resources, strengthening the links within their own chain of command and their leverage over local society. For this type of setting education was probably the least expensive means of upgrading Party leaders’ commitment and skills. It was also the most suitable way, in the absence of strong channels of communication and control, to keep in line elites scattered over an extensive area. The circumstances surrounding the party’s development, and the choices facing its early leaders, had institutional implications that carried forward into the contemporary period. 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 imperative for this was clear: a high share of cadres with low levels of formal education (Table 2.1).

8 Early CCP party school curricula can be found, in outline form, in Wilbur and How, eds. (1972), pp. 97–8, 130–4. 9 Early party schools were forced to relocate to the Soviet Union after the CCP’s retreat in 1927. In the countryside, training content shifted to rural issues. Party school course topics included “theory of the peasant movement,”“methods of organizing peasant associations,”“peasant self-defense corps,” and others. See Price (1976), p. 86, fn. 9. 32 The organizational landscape

Table 2.1 Highest educational attainment of cadres, by percent of 2003 survey respondents

Cultural Revolution, Education level Early PRC, 1950–65 1966–76 Reform era, 1977–

No formal education 5.0 0 0 Elementary 20.0 34.0 3.0 Middle school-level 20.0 17.0 12.5 vocational 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 Total 20 47 279

Note: 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.

Development of human capital, and pipelines for channeling this capital to the machinery of the party-state, became part of this early state- building agenda. As Manion notes, there was a burst of activity in this arena: the creation of systems for hiring and promoting cadres, training programs, and party and government personnel bureaus (Manion 1985: 206). These early party organizations lent structure and discipline to the young state. Drawing from émigré interviews, Whyte details cadre education during the 1960s as characterized by a “strict political atmosphere” in certain work units, with an emphasis on doctrinal study, personal political evaluations, and peer criticism (Whyte 1974: Chapter 5). During these early years of the republic, 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. In a field interview, a retired cadre from the Ministry of Foreign Affairs reminisced about these cadre training experiences: During this early period, there weren’t any rules. Leaders would work hard to arrange speeches. These would be reports on trips abroad, or what a cadre learned about external trade. These lectures might also be on laws, regulations, how to handle a particular situation ... These were just lectures, with no questions or discussion afterwards. Whether we had these activities depended on how active Creating the party school system 33 and energetic a particular work unit leader was ... [As for cost,] everything was free until the 1990s, when honoraria started being offered. Maybe 500 yuan for a talk in the 1990s, and now it must be 1000 yuan. Leaders of young cadres would also assign them to a mentor who would advise them on relevant skills for the job. My mentor helped [me] with translating skills.10 This system depended largely on the initiative of individual managers, and no formal rules governed, for example, the number of lectures to organize or mentors to assign. Two provincial party school teachers commented, Party schools like ours were not very broad in scope before the reform period. We would have short-term classes, maybe 15 to 20 days in duration. The point was to study party documents, major party meetings, and party congresses, to unite thought (tongyi sixiang).11 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.12 Then, during the antibureaucratic campaign of 1977–80, 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 late 1970s, training and continuing education programs comprised a new front in the leadership’s multipronged approach to reform. During this period, 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 party line and thus is of major strategic importance” (Central Organization Department 1983: Appendix 3, 67).13 In these declarations, there remains a clear ideological purpose to training, though subsequent reforms have used cadre training to advance the administrative modernization 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

10 Interview 26, December 2006. 11 Interviews 93 and 94, December 2007. 12 Interview 114, Central Party School professor, party history department, February 2008. The “five old topics” (wu ge lao men) were Marxist philosophy, Marxist economics, scientific socialism, party history, and party building. 13 This follows the enshrining, in the 1982 party constitution, of the clause that “Cadres must receive training, assessment and examination” (quoted in Paltiel 1990, p. 587). 34 The organizational landscape generation of young party cadres, the party school is shouldered with great responsibility.”14 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. Notwithstanding efforts to institute retirement norms, human capital development had taken place within existing party ranks. 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. One party school teacher recalled, Under Hu Yaobang’s leadership, in the late 1970s there was more focus and direction. You had degree programs offered in scientific socialism, Marxist philosophy, party history, and Marxist-Leninist political economy. Cadres would enroll in these because they had promotion in mind. Party school teachers would enroll to build their education credentials. At that time, it was hard to get into universities, the university system was at a low capacity. There was a need to develop talent, and this was one way.15 Officially, May 1982 saw the birth of the national party school system, beginning with the movement by Central Party School president Wang Zhen (1982–87) to standardize (zhengguihua) cadre training throughout the country (Wibowo and Fook 2006: 145).16 This standardization has been protracted and met with limited success. As one CPS professor of party history remarked, “Standardization started in 1983, and we have been trying to normalize the system for 25 years!”17 A district-level party school vice-principal bemoaned what he saw as a continued absence of standards across the party school system: There is a lack of systematic inspection standards, of means to evaluate training. Whatever the school head wants, that’s what goes. In party schools, there are no regulations or guidelines for teaching materials, class content, teacher qualifications, in how teaching plans are established. The management of schools is not uniform, down to the diplomas they issue.18 These complaints aside, there is some uniformity in the system today. At a minimum, local party schools all organize Mid-Career Cadre Training Classes and short training classes following major events such as party

14 “Promote the Party School Education to a New Level,” People’s Daily, June 7, 2000. 15 Interviews 93 and 94, December 2007. 16 Standardization included attempts to unify party school training content and the types of classes organized by local schools (Interview with CPS party history professor, Interview 112, February 2008). 17 Interview 120, February 2008. 18 Interview 195, May 2008. Organization and oversight of the party school system 35 congresses. There remains variation in the ﬁnal lineup of training classes mandated by local party committees and conceived by the schools themselves. This outcome has a logic of its own, as standards constitute a relatively weak form of organizational coercion (Brunsson 1999). As imperfect as they may be in practice, these decades-long efforts to modernize and standardize the party school system must also be viewed in schools’ larger organizational context.

Organization and oversight of the party school system Monitoring constitutes a critical means for ensuring accountability between party organizations and higher party authorities. The importance of sound monitoring processes increases when an organization possesses certain characteristics, for example, if it is responsible for generating information used by actors throughout a political system or when the organization is charged with political work (Hannan and Freeman 1984: 153). Accordingly, local party schools are enmeshed in a variety of formal and informal relationships with party and government bodies located at the same administrative level. 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 offer a description of the local party and government organizations involved in cadre training. Multiple oversight mechanisms serve to ensure that the boundaries of permissible activity are not crossed by these party organizations, and monitoring is critical given certain organizational traits of party schools. 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. Within these organizational arrangements, local party schools are subject to less formal monitoring by higher-level party schools, party personnel (zuzhi, or organization) departments, and finance bureaus. District party committee

District District District discipline District party organization propaganda and inspection school department department department

Theoretical education Party’s clean Often combined with: Cadre bureau section government education office District administration institute • Coordinates with the party school on leading • Maintains delegation of lecturers on party theory • Responsible for anti- District socialism school cadre training ( ) corruption education • Responsible for leading cadre study group content within work units E-education office

• Responsible for Party member education distance learning department programs

• Responsible for CCP Cadre online study member education at the credit office district and lower levels

• Responsible for online training system

Figure 2.1 Local party organs involved in cadre education (district-level example) District government

District District District government administration personnel ministry training District finance institute department centers

Civil servant section • Organizes training of Ex: District Tourism • Allocates funds to party non-leading cadres and Training Center and government organs civil servants for cadre/civil servant • Generally located on the • Collaborates with training party school campus and administration institute on shares staff civil servant training

Figure 2.2 Government organs involved in civil servant and cadre education (district-level example) 38 The organizational landscape

These vary in terms of their ability to influence party school activity. First, party schools are embedded in a web of relationships with other party schools. These relationships are characterized as advisory in nature (zhidao guanxi).19 Such consultative relationships stand in contrast to the more coercive, binding control that exists in other parts of the party bureaucracy. Customs (haiguan) bureaus, for example, must follow the dictates that flow down through the system from the center to the localities. Relations across party schools are the weakest among the organizational ties that schools will have with other party organs. One provincial-level party school professor remarked, Between schools, there’s only a professional, advisory relationship, and this advisory relationship is empty (xu). We don’tgetfinancial assistance or guidance from higher party schools, the relations are soft (ruanxing). In more recent years, grassroots party schools have wanted more leadership from higher party schools. So we have had training classes for grassroots party school teachers and leaders. We do a lot of trainings [for students] from outside the province. These add to our school’s profit because sending schools have to pay for their students to come here.20 As part of this advisory relationship, there is a sharing of training plans and materials. The Central Party School’s textbooks are circulated widely throughout the system, though the CPS cannot mandate that other schools use these materials. Relations with other party organs have the potential to be more coercive. The relationship between party schools and local organization and finance bureaus hinges on the annual formulation of training plans. While the center will issue five- and ten-year training plans to be implemented nationwide, specific classes are determined at the local level. The process is initiated within the party committee. A locale’s party committee, in consultation with the local party school, organization department, and other relevant bodies such as the education and personnel departments, will draft annual training plans and targets. Actual implementation of training classes assigned to a party school resides with school administrators. School leaders have incentives to meet plan targets not only because their performance evaluations are linked to plan implementation but because funding from the local finance department accompanies each training target in a plan. One county-level party school vice-principal outlined these relationships through the example of a recent training course for rural development:

19 This is stated in Article 11 of the CCP Party School Work Regulations, available at www. sina.com.cn, accessed November 13, 2008. The Chinese equivalent for top-down com- pulsory control is lingdao guanxi. 20 Interview 211, May 2008. Organization and oversight of the party school system 39

The county party committee had in its annual training plan to organize a training for township and village party secretaries on “building a new socialist countryside.” This was a big policy last year and is still a big policy this year. The county organization department cadre training section was put in charge of contacting the county party school and county rural development office. For the training, two hundred trainees were shown seven test sites, and we had expert teachers from Tsinghua University, the county party secretary, rural NGOs ...The training plan was sent down by the party committee after the Spring Festival [in February] and the organization department did all the coordinating. The [training class] took place in one month, from July to August. The county government finance department paid for it all, I think the budget was about 500 yuan per student, or 100,000 yuan in all. This was a big project. It was a very successful training; the trainees were very moved by what they saw in the test villages.21 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 2.2 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 in 2009, Wang Mingfang, is also a provincial vice-party secretary with responsibility for partybuilding and ideological work.22 Five vice- principals are responsible for day-to-day management of the school.23 In other provinces, such as Guangdong, the current 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. Under Hu, the Guangdong party school is managed by a team of seven: one standing vice-principal and six vice-principals.24 Again, this integration between party committees and party school leadership is within-locale and under- lines local party schools’ autonomy from higher-level party bureaus. These local lines of authority are balanced by structural checks and monitoring mechanisms between party schools and various party

21 Interview 49, November 2007. 22 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), www. china.com.cn/xxsb/txt/2007-04/09/content_8089276.htm, accessed April 17, 2014. Beginning in 2011, Wang Mingfang became the president of the provincial CPPCC while maintaining his leadership roles in the provincial party committee and party school. 23 See www.ahdx.gov.cn/sm2111111124.asp for a listing of Anhui Provincial Party School leaders, accessed June 18, 2009. 24 The Guangdong Provincial Party School leaders and party committee members are listed at www.gddx.gov.cn/xyxk/xrld.htm, accessed June 18, 2009. 40 The organizational landscape

Table 2.2 Provincial-level party school leadership, 2009

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

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 principals and their positions on the Central Committee and other party organs. Source: Author’s dataset.

authorities. Such intraparty monitoring ensures greater consistency throughout the party school system than might be assumed when looking only at the formal lines of authority. Monitoring between party authorities and party schools ensures some degree of quality control. Within these relationships are three major types of supervision: sending down work teams and observers, convening party school heads for training classes and conferences at higher levels, and imposing reporting requirements. First, 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 Organization and oversight of the party school system 41 lower-level schools.25 The Central Party School has been known to increase the number of work teams it sends out after important events, such as a party congress, to ensure that local schools are aware of special congress-related training sessions that they must organize and carry out.26 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 assign minders. For some classes, which are divided into small groups, each group will have a minder from the organization department and party school.27 When higher- level party committee members 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 visitor and host to share information about what is happening at each level.28 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.29 These meetings comprise celebratory and networking activities as well as the substantive exchange and ﬁling of ofﬁcial reports.30 At the subnational level, provincial-level party schools also convene training classes for heads of city and county party schools.31

25 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, and culminated in a banquet lunch and afternoon departure. 26 Interview 108, retired Central Party School professor, February 2008. 27 Interview 208, provincial party school professor, May 2008. 28 Ibid. 29 Records exist for the first “National Party School Work Meeting” convened in 1979, followed by a second in 1983 (Jiang et al. 1983). Other meetings were reported in each of their respective CPS yearbooks (1985, 1990, 1992, 1995, 1996, 2001). 30 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. Field notes May 2008. 31 See, for example, the two-week “Province-wide City, County (District) Party School Principal Training Class” organized by the Yunnan Province Party School in 2007; syllabus available online at www.ynce.gov.cn/ynce/site/school/article001.jsp?ArticleID=17824, accessed July 2011. 42 The organizational landscape

Third, party school heads also send reports of school affairs up the system, to the local finance bureau and the local organization department.32 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.33 Taken together, these monitoring strategies illustrate how party schools are connected to the larger party school system and key bureaus at the local level, 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 2009: 122). Throughout this hierarchical system, there is monitoring and control of party school activities, within which schools retain some autonomy. One city party school department head emphasized the relative weakness of these controls: Work units have to send leaders to training when the organization department orders it. They have to fulfill their quotas (ming’e). If not, they will run into trouble during inspection (kaohe) time. Party schools have a series of inspection meetings: an annual meeting to report to the local finance bureau, a meeting with the organization department to fill out forms and [report] statistics, a “teaching and study” meeting (jiaoxue hui) to exchange teaching methods, another meeting to talk about research. Overall, monitoring is pretty passive (beidong).34

Types of training and target students Understanding the varieties of training offered by party schools reveals the tension between efforts to impose system-wide discipline and allow for local variation. Despite attempts in the early reform period to standardize party school training outputs, schools continue to offer a broad range of educational programs and training sessions.35 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.36 Ofﬁcially, these residential and non-residential programs were intended to raise the

32 Interview 176, city party school teaching department head, April 2008. 33 At the CPS in 2010, the head vice-principal, Li Jingtian, reports to the principal, Xi Jinping, who sits on the Politburo Standing Committee and CCP Secretariat. 34 Interview 176, April 2008. 35 Party school organization reflects this diversity. See Appendix C for the organization of offices and departments within the Central Party School. 36 The syllabi from these early CPS undergraduate training classes are available in the 1985 CPS yearbook, pp. 224–32. Types of training and target students 43 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.37 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 schools provide course support and serve as marketing and distribution points.38 In 2007, distance degree programs were to cease admitting new students, but interviews yielded vague responses regarding the enforcement of this decree. Graduate degree programs, furthermore, continued to exist at the central and provincial levels. The second category, comprising non-degree training programs, is enor- mously 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– 5). 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.39 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.40 Core training courses include those which are for orientation or on-the-job training purposes, but they can also include more select classes for mid-career

37 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 (often province-level) campuses, 20–25 percent to schools below branch campuses (often city-level), and the remainder (50–55 percent) to the county- or district-level “tutorial stations” (fudao zhan). Zhu Hongjun, “Central Party School Calls for the End of Correspondence Degrees,” Southern Weekend, November 29, 2007. 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 (pp. 75–80). 38 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 both 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. While instructors were the same across programs, presumably there was some difference in course content. Interviews 10 and 11, November 2006. 39 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. 40 These include “outside-the-plan” (jihua wai) training classes. 44 The organizational landscape

Table 2.3 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 1,490 2,295 3,785 students (39%) (61%) Enrollment, in student-months 518 417 935 (55%) (45%)

Note: Row percentages in parentheses. Source: Internal school document. 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 2.3 offers an example of the allocation of training courses, by type, scheduled by a city-level party committee for 2008.41 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, since core classes last longer on average. 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.42 The first population includes graduate students and cadres taking degree courses on weekends or through correspondence courses, mostly to pass a standardized exam, fulfill an employment requirement, or acquire credentials needed for promotion. 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.43 Cadres enroll in party

41 See Appendix D for a full listing of this city’s annual training plan. 42 Interview 17, Central Party School professor, November 2006. 43 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 web page, available at www.yimendj.gov.cn/Article/ ShowArticle.asp?ArticleID=216, accessed June 30, 2009. Types of training and target students 45 school degree programs with these job requirements in mind. Although they are expected to pay from personal funds, in reality their workplace often produces a scholarship or other financial aid. Graduate students, concentrated at the central and provincial party schools, focus on pre- dictable disciplines such as party history and socialist theory, but there are also more contemporary specializations such as law and finance. A party school degree is a common feature of the official biographies of many high-level party officials, given the attractiveness of the part-time degree options offered by these schools.44 Officials of any rank may enroll in party school degree programs (though the more prestigious the party school, the more expensive the tuition), and these programs are also open to members of the general public. Enrollment in training programs is also diverse. Core training programs include routine periodic (lunxun) training classes for a variety of cadre populations such as those working in particular bureaucracies or areas of expertise and those with minority or reserve cadre status. Promising cadres with the potential to rise through the ranks are also invited to a key-point training for “mid-career cadres” (zhongqing nian ganbu) that is common across virtually all party schools. Invitation to training programs may be a matter of negotiation with direct superiors or at the behest of supervising organization departments.45 Advanced (jinxiu) trainings exist for leading cadres, such as the CPS’s 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. Compared to auxiliary classes, core classes tend to have more defined target students. The general rule of thumb is that schools will train cadres up to two ranks below the school’s administrative ranking. For example, provincial-level party schools have traditionally trained city- and county- level cadres. Recently, higher-level schools have been expanding their bases of potential cadre students. Since the early 2000s, for example, the Central Party School has organized nationwide training classes for

44 Li (2008a), in his analysis of 538 “fifth-generation CCP leaders,” finds that 24 percent of leaders obtained postgraduate degrees from party schools (pp. 71–2). Of the 25 current leaders in the CCP Politburo (the party’s highest decision-making body) five hold degrees from the Central Party School located in Beijing. 45 Interview 65, Central Party School department chief training class (chuji peixun ban) student, November 2007; Interview 66, provincial party school vice-principal, November 2007. 46 The organizational landscape county magistrates, in addition to its lineup of classes for bureau-, provincial-, and central-level cadres. Auxiliary training programs have no generalizable target populations of cadres and can include cadres of all ranks and functional responsibilities.46 This increased variety and specialization of party school training classes reflects the emphasis on more targeted cadre training in the reform period. These changes are a function of the frag- mentation that accompanied the creation of a training market, discussed in the following chapters, as well as efforts by party schools to position themselves well within that market. These trends, and the incentives underlying them, push against early reform efforts to standardize party school training outputs.

Reform-era focus on cadre training Party authorities have continued to emphasize cadre training as a core project of the party-state. They have turned to party schools as an instrument for supporting China’s modernization and economic reform. Developments in three areas attest to the growing importance of cadre training for governance in China: declarations in speeches and policy documents, investment in training infrastructure, and growth in training programs for the most promising “mid-career cadres” (zhongqing nian ganbu). Notably, institutions of cadre training such as party schools, administrative institutes, cadre executive academies, and cadre schools fall squarely under the leadership of the Central Committee, the State Council, the Central Organization Department, and the Ministry of Personnel, respectively, not the Ministry of Education. Such organizational lines reinforce the importance of cadre training for the party to exert control over cadres and bureaucrats, beyond the building of bureaucrats’ educational credentials. Ofﬁcial declarations of support for cadre training have spanned generations of CCP leadership. Interwoven throughout efforts to codify and create more systematic cadre training programs has been ofﬁcial acknowledgment that the ability of the party-state to implement 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

46 A party school will generally not enroll cadres of an administrative rank higher than the school’s administrative rank. For example, a city-level party school may convene auxiliary classes for city-, county-, and even township-level cadres, but provincial- and higher-level cadres would not be a target population. Reform-era focus on cadre training 47 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.47 In a subsequent generation of CCP leadership, Party Secretary Hu Jintao stated the following in his 2007 speech before the 17th Party Congress: 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.48 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).49 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.”50 Within this new framing, “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 official pronouncements. The 2003 Regulations on the Selection of Party and Government Officials specified that candidates for a leadership post

47 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 March 21, 2007. 48 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,” October 15, 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.” 49 See 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 and “CPC Issues Document on Ruling Capacity,” Xinhua, September 27, 2004, www. chinadaily.com.cn/english/doc/2004-09/27/content_378161.htm, accessed March 6, 2009. Hu Jintao also uses this term in his 2007 Party Congress report. 50 “Zeng Qinghong Speaks at Central Party School on Building up Governing Capability,” Xinhua, September 24, 2004. 48 The organizational landscape at the county level or above must have (1) a bachelor’sdegreeorhigher, (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).51 For the period 2003–08, the Central Organization Department, the powerful bureau charged with party personnel matters, declared that over 100,000 government and party leaders from the county level up must undergo three months of training over the ﬁve- year period.52 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.”53 Trial regulations issued in 2006 emphasized the linkage between cadre training, building the party’s governance capacity, and creating a “learning party.”54 Under Xi Jinping’s administration (beginning in 2012), cadre training has continued to be a means for building party loyalty and forming “strategic” public managers.55 The 2013–17 National Cadre Education and Training Plan is similar to the 2006–10 Plan with respect to guiding principles, training content, and target trainees.56 But there are some differences between the two documents. The newer plan places less emphasis on the organizational innovation and overseas exchange

51 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. As noted in the 2002 Regulations on the Selection and Appointment of Party and Government Leading Cadres, training could take place at “party schools, administration schools, or other training institutions recognized by the organization (and personnel) departments” (Section 2, Article 5), available online at www.china.com.cn/policy/zhuanti/17da/2007-08/21/content_8720821.htm, accessed December 2, 2012. The 2014 Regulations on the Selection of Party and Government Officials also notes that officials must complete required training, but they do not specify duration. Interviewees indicated that some officials receive their prepromotion training after assignment to a new post, but this “make-up training” (bu xun) does not appear to be a common practice. Interview 65, former Central Party School trainee, November 2007 and Interview 191, county-level party school vice-principal, May 2008. 52 This figure includes “about 500 provincial-ministerial-level leaders, 8,800 prefecture- division-level leaders, and 100,000 county-department-level leaders [who] must participate in these training programs” (2006–10 Nationwide Cadre Education and Training Plan, Section 3). 53 Chinese Yearbook of Ideological and Political Work (2001), pp. 50–4. 54 2006 Cadre Education and Training Work Regulations (Trial), available online at http:// politics.people.com.cn/GB/8198/60906/, accessed December 18, 2007. 55 Xiao Shan, “The Content of Cadre Education and Training,” People’s Daily Online, June 25, 2014, available at http://dangjian.people.com.cn/n/2014/0625/c117092-25195692. html, accessed July 2, 2014. 56 The full text of the 2013–17 plan is available at http://politics.people.com.cn/n/2013/ 0929/c1001-23069508.html, People’s Daily Online, September 29, 2013. Reform-era focus on cadre training 49 highlighted in the 2006–10 Plan.57 Under Xi, there has also been an effort to increase party discipline during cadre training and education. Official regulations issued in February 2013 present a list of do’s and don’ts during training, with a focus on barring misuse of office. Cadres are not to accept gifts, use public funds for banqueting during classes, or ask others to complete their assignments.58 These regulations dovetail with Xi’s push to rein in corruption throughout party ranks during his early years in office. Because of the reach and scope of cadre training, it is an essential arena for transmitting such messages. Additionally, the proliferation of training courses and operators during the reform period has motivated central officials to reassert control over cadre education as a time for policy transmission and party- and skills-building, not the building of power bases among officials.59 Speeches and policy documents may offer a sense of trends and intentions; stronger evidence lies 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 US dollars.60 While systematic financial data are difficult to obtain, these figures give a sense of the infusion of funds to party school development. Enrollment levels are another indicator that cadre training is an increasingly important aspect of cadre management. From 1997 to 2004, for example, a total of 20.2 million trainees passed through training

57 For example, the 2006–10 plan discusses an “innovative training model” (Section 6) and includes a full paragraph on overseas training (Section 4), all of which are minimized in the newer plan. 58 These are among other stipulations regarding prohibited behavior during training. See Xinhua, March 18, 2013, “Central Organization Department Issues ‘Regulations on Advancing and Strengthening Student Management in Cadre Education and Training.’” 59 Regulation 8 cautions against building “coteries” (xiao quanzi) among training class alum. Regulations on Advancing and Strengthening Student Management in Cadre Education and Training, February 19, 2013. 60 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. 50 The organizational landscape programs at the provincial, city, and county administrative levels.61 Since the onset of reforms in 1978, the CPS has trained in excess of 50,000 medium- and high-ranking cadres.62 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 “Mid-Career Cadre Training Classes” and other selective, but regularly scheduled, training programs.63 Mid-Career Cadre Training Classes (zhongqing nian ganbu peixun 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 are available, the pace at which Mid- Career Cadre Training Classes grew, at the central level, generally outstripped the growth in the total number of leading cadres in the country (Table 2.4). This is a somewhat noisy trend, since there was a drop in the 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 these Mid-Career Cadre Training Classes. The CPS’s mid-career cadre training enrollment has increased from 1981 to 2000, and the share of total CPS trainees of the training class has also increased (Table 2.5). This suggests that party authorities have focused more narrowly on building the political knowledge, networks, and career prospects of a select group within the leadership class. It also appears that the CCP has placed increasing emphasis on training those leading cadres with the potential to provide long tenures of service to the party. Funding data are not available, but these are resource-intensive classes, since trainees are housed on campus for the duration of the training program (10 and a half months or 23 months, depending on

61 “Accomplishments in Party Organization Work Since the 15th Party Congress,” February 12, 2004, www.xfdjw.gov.cn/show.asp?id=79, accessed November 3, 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, aver- aging a rather high 177 trainees per training session. 62 “China’s Central Party School Trains 50,000 Ofﬁcials in 30 Years,” Xinhua, October 2, 2007. 63 Interview 216, Central Party School professor, May 2008. He noted that core general training programs in the CPS’s Training Department (peixun bu), which organizes these “mid-career cadre training classes,” were receiving more emphasis than the more targeted training classes organized by the Advanced Training Department (jinxiu bu). Placing party schools in context 51

Table 2.4 Percent increases, CPS mid-career cadre trainees, leading cadres, and CCP members, select years

Percent increase over Percent increase over previous year, CPS previous year, total Percent increase over mid-career cadre number of leading previous year, total Year trainees cadres in the country 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

Sources: CPS Yearbooks, 1985–2001, Central Organization Department 1999, Xinhua.

Table 2.5 CPS Mid-Career Cadre Training Class enrollment, 1980–2000

Mid-Career Cadre Total Central Party Mid-Career Cadre Training Class, as a School enrollment, Training Class, in percent of the CPS Year in trainee-months trainee-months 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. whether the training class is for one or two years) and sent on multiple study trips throughout the country and abroad.

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. The resilience of party schools stands in contrast to the waning of other key institutions of political control, particularly political campaigns 52 The organizational landscape

(yundong).64 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 has entered an “institutionalization phase” (Li and Bachman 1989: 91; Li and White 1988). While the ﬁrst phase entailed building elite and mass support for a reordering of Chinese society under the party’s vision, institutionalization called for more routinized, rule-bound modes of governance. 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 intensiﬁca- tion 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 rea- lignment of party ranks (Zheng 1997:153–8). 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 (1949–76), political mobilization requiring broad public participation and demonstrations of party loyalty by bureaucrats created a tension between mass unrest and party control, but this had 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 had been “absorbed into the organization” (Selznick 1960: 288–9). In the 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 deployment of a nationwide antibureaucratic campaign in 1977 to unify the bureaucracy in the wake of the Cultural Revolution. In 1979, Deng called an end to reliance on political campaigns and exhorted comrades to turn instead to the tasks of economic development:

64 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. Placing party schools in context 53

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.65 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, maintain the vanguard nature of the party, and stamp out corruption, but they waned in frequency and intensity through the Jiang and Hu administrations (Manion 1985; Shambaugh 2008a; White 1990).66 Institution-building within the party has replaced ad hoc political mobilization, representing a normalization of political life. Campaigns are no longer intended to engineer a wholescale transformation of society so much as target particular political issues. Anticorruption campaigns are one example. These have persisted in the post-Mao period, especially under Xi Jinping’s early administration, 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.67 Like campaigns, training is a form of ex ante political control: it offers broad prescriptions for action and increases local awareness of general legal (or normative) boundaries set by higher-level authorities.68 These

65 “Some Comments on Economic Work,” Speech by Deng Xiaoping, October 4, 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 May 21, 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. 66 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). 67 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 with greatly diminished force. 68 On the distinction between ex ante versus ex post controls, see Huber and Shipan (2006). 54 The organizational landscape qualities are particularly suited for transitional environments in which local circumstances are shifting rapidly and local agents must act auton- omously based on knowledge of central goals.69 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 and universities, campaigns are not amenable to incentives to innovate, such as organizational competition. In this sense, they lack the capacity for local innovation that schools may develop.

Conclusion Party schools have long been a fixture of party life. Over the party’s tumultuous history, these institutions have carried out several key roles in partybuilding work. At present, they contribute to articulating and maintaining the party line, conveying policies throughout the party- state, and enhancing the quality of the CCP’s human capital. Official declarations and documents, funding increases, and targeted investment in mid-career cadres reveal the central role of cadre training in party personnel management. New investments in cadre training suggest the efficacy of party schools for carrying out their trinity of partybuilding responsibilities. These schools are in a prime position to continue expanding, as there is a great deal of flexibility and organizational autonomy in the kinds of training outputs that party schools may produce. As the next chapter will detail, these schools are significant organizational pipelines for moving bureaucratic talent to leading positions within the party and government.

69 In his seminal study of behavior among US 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. This increases the risk to authorities of grassroots resistance (Oliver 1991), and training can serve to mitigate such risks. 3 Managing the managers Party schools as a pipeline to higher ofﬁce1

In 1999, Huang Zhijia enrolled in a correspondence degree program jointly organized by the Hubei provincial party school and Central Party School. After two years of study, he received the equivalent of an undergraduate college degree, with a major in law. By 2004, Huang had launched a successful career in the court system, attaining the offices of judge (shenpan yuan) and president of the local court (jiceng fating tingz- hang). Then his career unraveled. In 2006, he discovered that his party school degree was not sufficient qualification for his position. When he attempted to take a routine judicial exam, authorities barred him from taking the exam on the grounds that he did not meet the prerequisite of holding a university degree in law. The Hubei Ministry of Education corroborated this decision, confirming that the “party school is outside of the national education system.”2 Angered, Huang sued the Hubei Province Party School and Central Party School for damages. He argued that they had made fraudulent claims about the validity of their correspondence degrees. The Jianghan District of Wuhan agreed to hear the case in 2007, but the case did not go to trial.3 His lawsuit garnered national media attention, and he soon became famous for using the legal system to tackle issues with higher education providers. As a result of increasing public visibility, Huang fell out with party authorities. In

1 Portions of this chapter reprinted with kind permission from Springer Science+Business Media: Studies in Comparative International Development, Party Selection of Officials in China, vol. 48, num. 4, 2013, pp. 356–79, Charlotte Lee. 2 Southern Weekend, “Hubei Judge Sues Party School for Illegally Issuing Degrees,” November 29, 2007. 3 Caijing, “Turmoil Following a Judge Suing the Central Party School,” January 10, 2008. This is not the first time party school degree holders have had their degrees rejected by state authorities or state schools. In 2006, a student could not take exams because her Zhejiang party school degree was deemed insufficient by state authorities. Similarly, a party school degree holder was not allowed to take an entry exam for a master’s program in Jiangsu because his party school undergraduate diploma was not considered a valid qualification. See “Hubei Party School Diploma Holder Sues the Party after Beijing Refused for Judicial Exams,” Legal Weekly, April 21, 2008.

55 56 Managing the managers

2009, authorities shut down a blog that he had created to chronicle his lawsuits.4 Other cadres have had very different experiences with party schools. Take, for example, the experiences of Wang Fei.5 She 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 Wang reached a high rank despite a potentially inauspicious personal background: neither of her parents was a party member, and she joined the party at a relatively late age of 37. Her workplace relationships may have facilitated this advancement, however, given that she enjoyed “frequent” interaction with her superiors, subordinates, and workplace peers.6 Wang spent the entirety of her 37-year working life in a series of state-owned enterprises (guoyou qiye)andstate- owned businesses (guoyou shiye). Her occupations ranged from teaching at an elementary school located within a state-owned business, accounting at a state-owned enterprise (SOE), and finally managing a state-owned business. The critical year in her career history was 1978, when she switched from accounting to a general administrative career field and was promoted to section rank. This was followed in 1984 with promotion to the department rank as a company leader (qiye fuze ren). The timing of Wang’sparty school training 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, but it marked the beginning of a series of promotions that lasted until retirement. These two stories illustrate the variation in outcomes across the many types of cadre training courses and degree programs offered by party schools. A crucial difference between the judge’s and SOE cadre’s experiences is the nature of the coursework they completed. In the judge’s case, he obtained a degree by paying out of pocket for what he thought would be a career-enhancing credential. In contrast, the SOE employee was selected for a short-term, non-degree training class. In terms of timing,

4 Huang has since opened a lawsuit against the blog provider, Sina.com. Wang Heyan, “Hubei Judge Sues Sina for Arbitrarily Closing Blog,” Finance Network, June 8, 2009, available online at www.chinagfw.org/2009/06/blog-post_5641.html, accessed July 11, 2013, and “Hubei Activist Judge Huang Zhijia Sues Sina Gateway for Closing His Blog,” DW News, June 9, 2009, available online at http://politics.dwnews.com/news/ 2009-06-09/4958427.html, accessed July 11, 2013. 5 This is a fictitious name given to survey respondent 1224, who was randomly chosen from the 2003 China General Social Survey, from among those respondents who reported a cadre occupation in their career histories. 6 This was a self-reported response in the 2003 CGSS. Managing the managers 57 both went through party schools early in their careers, suggesting the usefulness of these schools for junior officials. Taking condensed tours through the two individuals’ experiences raises several questions regarding the relationship between party schools and promotion. First, it would appear that degree programs are not as career enhancing as training classes; the former are generally demanded by cadres, whereas the latter entail both cadre demand and selection by a supervisor and/or personnel authorities. Second, it seems a party school training course or degree can be the launching point for successful bureaucratic careers. A third pattern worth assessing is the relative effect of party school training on different points in a cadre’s career. Party school training preceded the SOE cadre’s promotion to the lowest section-level rank but did not precede her promotion to the higher department level. This implies that party school training is most pronounced, for rank-and-file cadres, at lower administrative ranks. Examining the experiences of others can illuminate whether additional aspects of this cadre’slifehistory,suchasher educational background and parents’ party membership status, are typical for cadres. While individual stories capture nuance and lived experiences, statistical tests can illuminate larger patterns and tendencies across the general population of cadres. This chapter will draw on cadre career histories culled from a national survey and an original dataset to answer the questions raised by these two very different experiences. More generally, such stories of personal investments in training and education speak to the ways that individuals may strive to signal their abilities to party authorities or the means by which party authorities identify and screen for talent. Hierarchical single-party systems such as China’s must solve the problems bound up in the principal–agent relationships that pervade the party. 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 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 1992). Problems of “hidden action,” or moral hazard, are mitigated through monitoring and the structuring of incentives.7 Both “police patrol” and “fire alarm”

7 Huang (2002b) presents an overview of the direct and indirect controls in place to cope with problems of hidden action in the Chinese bureaucracy. For 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 (1996, 1998). 58 Managing the managers mechanisms, which draw upon the monitoring abilities of higher authorities and receiving publics, respectively, may curb agents’ temptation to shirk their administrative duties (Lorentzen 2013; Lupia and McCubbins 1994; McCubbins and Schwartz 1984). Principals may devise both ex ante and ex post strategies for coping with these agency problems (Huang 1995). 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 1992: 138–58). 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 undesir- able 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: 149–59). Findings in studies of executive and legislative control over the US bureaucracy suggest that appointment processes, which include decisions about structure as well as personnel, have the largest influence on policy outcomes (Calvert et al. 1989: 605). It is critical to control selection to party and government office, which is foundational for party rule.8 Among the various external and internal influences on bureaucrats’ compliance with organizational objectives, the most important ones are inherent in the agents themselves. 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). Ideally, selection and promotion serve two functions within organizations: to match individuals

8 Eastman (1974) attributes the failure of KMT rule to weak political institutions and lax membership recruiting methods. He notes how, unlike the CCP, the KMT was, “prior to the assumption of governmental powers ... not a vital, tightly disciplined or uniformly idealistic revolutionary movement. Admission procedures had always been lax, ... and party leaders had been more concerned about the size than the quality of the membership” (p. 3). He points out that exams put in place to screen for administrative talent failed miserably: by 1935, out of the 1,585 individuals who passed civil service exams, only 8 were given posts of some responsibility. The vast majority of bureaucratic offices were instead granted through patron–client relationships (Eastman 1974:10–11). Managing the managers 59 of certain skills with appropriate offices and to generate incentives for improved performance. There exist two alternatives for mitigating this selection problem: inducing agents to reveal private information and 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 over- coming 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 higher-education degrees in the decades since, the variation in degrees and degree-granting institutions presents yet another complication.9 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

9 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. It is non-degree training courses that have a demonstrable effect on promotion outcomes. 60 Managing the managers 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). This selection problem becomes even more salient during a period of transition, when 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 iden- tifiers such as 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 many cases, making shrewd decisions about public investment and enterprise. 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 2004). Drawing on the data collected in the 2003 China General Social Survey, this chapter will test whether party school training has a measurable effect on promotion to higher office. This chapter also draws on an original dataset of elites to examine patterns in the career trajectories of cadres trained at the Central Party School. Findings include a summary of the party and government posts that such high-level party school alumni eventually occupy as well as rates of promotion. Findings support the intuition that party schools constitute an organizational means within the party to control access to positions of authority.

The party’s selection problem CCP authorities face an enormous selection problem: they must devise strategies for choosing from among the 46 million cadres at the lowest 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 2 percent. The party’s selection problem 61

Low bureaucratic turnover, as discussed previously, also contributes to the 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. Cadre training at party schools is not the only solution. Across authoritarian systems, there exist a variety of pathways for resolving this selection problem. Performance evaluation presents one alternative. Modifications to evaluation criteria over time indicate “adaptive learning on the part of principals” (Whiting 2004: 101). 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 subject to patronage relations (Chow 1988, 1993; Heimer 2006; Minzner 2009) and weak incentives to use evaluation rubrics for judging cadre work performance (Burns and Xiaoqi 2010). Evaluation systems are a significant aspect of the CCP’s overall ability to monitor cadre behavior, but they require costly police-patrol monitoring. Cadre evaluation procedures tested and promulgated in 2006 have placed an even greater burden 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). Cadre performance in a party school training program can serve as a shortcut for personnel officials facing the problem of too much information contained in personnel dossiers (dang’an). Since personnel dossiers in China contain all relevant documents on an individual’s personal history and professional performance, selection for and observation in a party school presents a useful summary of this store of information. 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 nondemocracies held at least one legislative or presidential election.10 Elections in authoritarian systems may serve multiple purposes such as generating rituals of citizen political participation, providing an institutional safety valve for

10 This figure is from the Przeworski et al. dataset, in which regime type is coded based on the presence of competitive elections (Przeworski et al. 2000). 62 Managing the managers channeling citizen discontent (while also providing political leaders with public opinion information), and distributing rents to elites (Blaydes 2011; Brownlee 2011; Hermet et al. 1978). They are also a means to sort through political entrepreneurs. In the case of China, elections began as village-level experiments in the 1980s and have since become compul- sory 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.11 Finally, there is the promotion of officials based on personal relationships, or guanxi, which present an extra-organizational route to high office. Systematic data on the impact of such networks on leadership outcomes in CCP politics has been elusive. There does exist some evidence of the benefits of a “princeling” or privileged political background for high office (Li 2008a; Nathan 1973), along with the importance of factional ties (Shih et al. 2012). The influence of networks may be bound up in the processes by which individuals are invited to a party school training program. While these schools may coexist with patronage relationships, they also impose impersonal party controls over the process of promotion. All of these control mechanisms point to the dense, overlapping institutions that contribute to party 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 and access to political networks and managerial potential. Party schools are sites where these various individual qualities may be observed and developed. These schools offer a 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 existing alternatives. A nationwide network of party-managed schools provides a systematic, orderly option that is consistent with the CCP’s drive to deepen the institutionalization of party governance.

Party school training: a pipeline to higher office There is both official and unofficial support for the idea that party schools are the place to look for the party’s present and future leaders. The

11 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). Party school training: a pipeline to higher ofﬁce 63

Chinese 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 become the CCP’s next party secretary.”12 These accounts claim that “one-third of Mid-Career Cadre Training Class graduates [from the Central Party School] are promoted to the provincial (ministry) rank.”13 Recent, and often positive, coverage of the Central Party School in the press, domestic and foreign, suggests the need to test hypotheses regarding the substance of the role of party schools within the CCP.14 Supporting the idea that party schools are gatekeeping organizations, performance during training classes has been explicitly linked to cadre evaluation. A 2006 set of trial regulations for cadre education and training called for a given student’s “attitude and performance (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: 122–35; 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 assessing the cadre’s work and promotion prospects” (Ministry of Personnel 2006: Section 7, Article 41).15 Development of cadre training and education is also a recommendation proffered in the internally circulated (neibu) 2004 publication, Research on Certain Questions in Building a Governing Party (zhizheng dang jianshe ruogan wenti yanjiu). Despite these assertions, the actual effect of party school training on a cadre’s promotion prospects remains unexamined. The existence of “party-sponsored upward mobility” suggests that party school training, as an educational option nested within a larger universe of party-related educational paths, serves a sorting function (Li 2001a;Li and Walder 2001). Related to this is identifying the criteria by which

12 Zhao Xue, “China’s Most Special School – The Central Party School,” Southern Weekend (Nanfang Zhoumo), November 8, 2007. 13 Ibid. 14 For coverage of the party school system in English language outlets, see, for example, Melinda Liu and Jonathan Ansfield, “Life of the Party,” Newsweek, May 30, 2005; “China Gives Peek Inside Elite School,” ABC Australia, July 5, 2010; Dan Levin, “China’s Top Party School,” NPR, March 7, 2012; and “Learning to Spin,” Economist, February 8, 2014. 15 For a full text of the trial regulations, see http://wlzx.hdpu.edu.cn/zzb/Article_Show.asp? ArticleID=858; www.sz.gov.cn/rsj/zcfggfxwj/zcfgjgfwjqt/rsbzgbywpx/200809/t200809 17_50492_4369.htm, accessed August 2008. Training requirements applied to SOE managers as early as 1991, when a decision 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). 64 Managing the managers cadres are selected for participation in these schools. One implication of this inquiry is to demonstrate whether and how Leninist mechanisms of cadre control remain relevant in the transition to a market economic context. Survey data will be used to test the initial hypothesis that party schools constitute a pipeline to higher office: if a cadre attends a party school training program, then he or she is more likely to be promoted to a higher administrative rank. While officials can be promoted without attending a party school, their likelihood of promotion is enhanced through the training programs of these schools. Results from tests of this hypothesis suggest that enrollment in party school training programs has a positive and significant effect on cadre promotion. Furthermore, the matching procedure employed in the empirical section below estimates the independent effect of party school “treatment,” separate from other factors that may drive promotion, and mitigates the problem of selection bias. This method tests the intuition that only cadres preselected for promotion attend party school. Related to this is determining whether a cadre’s performance during a party school training program provides information to higher party officials regarding that cadre’s suitability for promotion. Understood this way, party schools are more than organizations for carrying out the rituals of becoming a good CCP cadre or “new socialist man” (Munro 1971). These two functions, ritual and informational, are not mutually exclusive, and gathering information about a cadre’s fitness for office entails an assessment of his or her mastery of political scripts. This may be a proxy for political loyalty and submission to a broader corporate culture, all of which are salient for maintaining party discipline. The analysis presented here will also take up the issue of the ritual versus informational content of party school training by considering whether the process is driven by cadres seeking to signal political ambition (in which case the ritual argument is more salient) and/or by party officials seeking to identify political talent (and its many dimensions in the Chinese context). While these tests capture the general positive effect of all party school training programs on career advancement, field interviews suggest that the most important training program for upward-bound cadres is the Mid-Career Cadre Training Class organized by party schools throughout the system. Another 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 the most talented officials to provide long tenures of service to the party. Analyzing the effect of training 65

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 et al. 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 schools and promoted, and these confounding factors confuse the true effect of party school training on cadre promotion. To determine the effect of party schools, the counterfactual question to answer is, 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 “control” group to those “treated” with party school training by first matching survey respondents along a set of observable characteristics. This study employs propensity score matching, a two-stage technique to reduce selection bias by first predicting via a probit regression and a set of observed predictors the probability that an individual will be in a treatment versus control group. First, pretreatment characteristics are summarized in a propensity score for matching purposes (Boyd et al. 2010; Rosenbaum and Rubin 1983). Second, 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 (Rubin 1973). 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 (in this case, promotion to some administrative rank) across matched individuals from the two groups.16

16 To compare the ﬁndings from this matching process versus a probit regression, Appendix E presents estimates from a probit regression in which the dependent variable is promotion to administrative rank and the key explanatory variable is selection for training at a party school. 66 Managing the managers

Propensity score matching reduces, but does not eliminate, bias entirely. Selection 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 observedvariablesincludedintheﬁrst-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. 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.17 A total of 589 respondents within this survey reported careers subject to the party and government ranking system. Table 3.1 summarizes characteristics of the cadre subpopulation within the general population surveyed. Within this cadre population, 156 (26.5 percent) were female and 408 (69.2 percent) were CCP members. This compares with a CCP membership of 18.6 percent among the sampled population, of which 51.9 percent were

Table 3.1 Descriptive summary of cadre and general population

Cadres General population N=589 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

Note: Mean comparison tests show that differences in each of these areas are statistically signiﬁcant at the 0.01 level. Source: 2003 China GSS.

17 This survey is modeled after the General Social Survey conducted in the United States. A report of the CGSS sampling design and survey methodology is available at 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 regression estimates are presented with weights to compensate for this. Analyzing the effect of training 67

400

350

300

250 241 No party school 200 Attended party school 150 Total individuals Total 100 100 63 50 108 9 27 35 0 6 Below section Section Department Above department Rank

Figure 3.1 Frequency of administrative rank, 2003 China GSS female.18 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. The outcome of interest is whether, conditional on party school training, a cadre is more likely to be promoted to a higher administrative rank. A series of dummy outcome variables were 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). Figure 3.1 summarizes the distribution of administrative ranks over the 589 surveyed cadres and whether or not they reported attending a party school. A dummy variable that captures whether an individual attended a non-diploma training program at a party school is hypothesized to have a signiﬁcant positive correlation

18 The national numbers are lower: according to 2011 ﬁgures, total CCP membership stood at 6.1 percent of the total population. For CCP membership levels, see www.chinatoday. com/org/cpc/, accessed December 2, 2012. National population ﬁgures are from the PRC National Bureau of Statistics, available online at www.stats.gov.cn/english/newsand comingevents/t20120120_402780233.htm, accessed December 2, 2012. 68 Managing the managers 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.

Who is selected for party school training? The first stage of this matching procedure provides insights on which individual-level characteristics affect the likelihood of selection to a party school training program. A first-stage probit regression model estimates individual-level propensity scores for the treatment (party school training) and the effect of respondent-reported variables on selection for party school training.19 This analysis controls for several demographic characteristics, including an individual’s age, gender, monthly income, total years in the work force, and total years of schooling. The survey data also captured information on respondents’ educational backgrounds. Within these educational histories, individuals reported their university majors, which are coded here as a set of dummy variables: sciences and engineering, law and humanities, economics, and management. To control for spatial variation, individuals’ home provinces were also coded as a set of dummies. Several political control variables were also included. Most prominent among these were an individual’s party membership and that of his parents. Respondents also reported whether or not they served in the military. Furthermore, the regression included a dummy variable coded for whether an individual reported frequent interaction with her superiors.20 This variable is an attempt to capture the possibility of a strong relationship between subordinates and their managers at higher levels in the bureaucracy. Table 3.2 reports probit regression estimates, where the dependent variable is whether an individual attended a party school training program. Several variables were statistically significant, including those that were more overtly political in nature. Party membership was key and positively correlated with selection for party school training. The marginal effect of this dummy variable was 0.32 (Table 3.3), meaning that an

19 This model was estimated using the pscore command in Stata. Descriptive statistics are reported in Appendix E. 20 The nature of these interactions and the actual frequency were not given in the data. Individuals were asked, “In your job, the frequency of contact with your upper-level leaders is which of the following?” and could respond “never,”“rarely,”“sometimes,” and “often.” Admittedly, this is an imperfect way to measure the strength of respondent ties to her superior; response categories are subjective and no anchoring measure is available. The analysis here collapses the latter two and former two response categories to create a dummy variable. Analyzing the effect of training 69

Table 3.2 First-stage probit regression results for propensity score estimation

DV: party school training (dummy) Coefﬁcient Robust SE

Constant −1.89 0.84* Age 0.00 0.01 Female −0.07 0.18 Monthly income 0.00 0.00 Total years in the workforce −0.28 0.02 Education (total years) −0.28 0.02 Major: science and engineering 0.28 0.18 Major: law and humanities 0.54 0.21** Major: economics −0.01 0.20 Major: management 0.25 0.21 Political control variables (dummies) CCP membership 1.19 0.21** Parents’ CCP membership 0.03 0.18 Military experience −0.37 0.18* Frequent interaction with 0.60 0.24** superiors Provincial-level dummies Beijing 0.32 0.58 Tianjin 0.71 0.57 Hebei 0.71 0.63 Shanxi −0.06 0.81 Neimenggu 1.17 0.70 Liaoning 0.21 0.64 Jilin 1.22 0.89 Heilongjiang 0.18 0.82 Shanghai 1.03 0.59 Jiangsu 0.86 0.61 Zhejiang 0.81 0.82 Anhui −0.18 0.60 Fujian 0.65 0.67 Jiangxi 0.51 0.68 Shandong 0.39 0.66 Henan −0.18 0.61 Hubei 1.53 0.58** Hunan 1.36 0.58* Guangdong 0.76 0.59 Guangxi 0.38 0.63 Chongqing 1.78 0.84* Sichuan 0.78 0.71 Guizhou −0.76 0.73 Yunnan 0.60 0.61 Shaanxi −0.37 0.75 Gansu 1.40 0.71* Iteration 0: log pseudolikelihood −338.86 70 Managing the managers

Table 3.2 (cont.)

DV: party school training (dummy) Coefﬁcient Robust SE

Iteration 1: log pseudolikelihood −264.57 Iteration 2: log pseudolikelihood −259.70 Iteration 3: log pseudolikelihood −259.56 Iteration 4: log pseudolikelihood −259.56 N 550 Wald chi2(39) 115.20 Prob > chi2 0 Pseudo R2 0.23

Notes: Signiﬁcance codes: ** for p<0.01, * for p<0.05. 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 category: Xinjiang regional dummy. Source: China GSS 2003.

individual who is a CCP member is 32 percent more likely to attend a party school than one who is not a member, holding all else constant. That this marginal effect is not higher suggests that party school training programs, particularly for the entry-level cadres that comprise the bulk of survey respondents, may also be pipelines for general party recruitment and not only promotion to higher party office.21 Findings also indicate that party schools present an institutional pathway for the nonhereditary selection of officials. While party membership was an important characteristic among those selected for party school training, parents’ party membership was not. This latter finding cuts against arguments that parental connections place individuals in a relatively advanta- geous position for attaining party office, at least at the more general national level captured by the survey data. One implication of this finding is that individual-level characteristics and qualities, rather than family connections and the kinship-based patronage ties therein, are more salient in determining career paths. It appears the party has maintained the organizational bases for such elite turnover and attenuated elite

21 The total number of cadres in China numbered 40.2 million in 1997, of which 24.9 million (61.9 percent) were non-CCP members (Heilmann and Kirchberger June 2000; Central Organization Department 1999). Party membership rates increase the higher the administrative rank. Analyzing the effect of training 71

Table 3.3 Marginal effect for signiﬁcant control variables

Variable Marginal effect

CCP membership 0.32 Frequent interaction with superior 0.16 Military experience −0.10 Major: law and humanities 0.14 Hubei province 0.41 Hunan province 0.36 Chongqing municipality 0.47 Gansu province 0.37

Note: Other control variables were held constant at their medians. reproduction through hereditary means (Parkin 1971).22 While this does not negate the importance of hereditary ties in attaining party membership (Walder 2003), the empirical analysis presented here speaks to the possibility of elite circulation or “the renewal of the existing elite by the accession of individuals from [other] classes of society” (Bottomore 1964: 49). An important caveat is in order, however. Elite selection (at the leading cadre level, above the department or county level) is not captured in significant numbers in the survey data; studies have shown that “princeling” status is consequential at these highest levels of political power. Patron–client ties are a further consideration. Relationships with higher-level officials appeared to matter in selection for party school training programs. Those respondents who reported frequent interaction with superiors were 16 percent more likely to be selected for party schools (Table 3.3), which supports the common belief that connections and patron–client relationships pervade promotion to political office in China. Connections may matter, but, as will be discussed below, party schools also present an additional impersonal organizational strategy to screen candidates for promotion. A positive relationship with a boss is helpful, but the analysis below suggests that it is one step in a multilayered selection process.

22 Parkin focuses on the social, economic, and political privileges that accrue to the hereditary-based transferal of property and education (pp. 152–4). Efforts to modernize and apply merit-based criteria for selection have also taken place in China’s civil service reforms. Burns observes, “The speciﬁc goals of the Party’s civil service reform were ﬁrst, to improve the quality of the cadre corps so that it could meet the new challenges of economic development, and second, to increase the perceived legitimacy of the selection of cadres which had been tarnished during the Cultural Revolution by such doctrines as the ‘blood line theory’ and since then by favoritism and patronage (‘going through the back door’)” (Burns 2004: 40). 72 Managing the managers

Those with military backgrounds were less likely to undergo party school training. An individual who possesses military experience is 10 percent less likely to attend a party school than an individual without military experience, holding all other characteristics constant. 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 China’s PLA, there has been increased emphasis on the recruitment and training of military officers at military academies (Zheng 1997: 228–31). Certain educational backgrounds were also statistically significant predictors of selection for party school. The emphasis on educational credentials was part of the reform-era push for a more professional bureaucratic corps. Among those with university degrees, individuals who majored in law or the humanities were more likely to attend party schools. A background in these fields, which had a marginal effect of 14 percent, has been on the rise in the most recent generations of party leaders (Li 2007b, 2008a, 2008b). Given earlier emphasis on technical degrees, particularly in engineering and the hard sciences, the variables for economics and engineering majors were expected to be significant (Lee 1991;Li2001b). However, it is likely the case that individuals with these more technical credentials were channeled into technocratic career tracks of the state bureaucracy rather than the political offices that party schools feed into (Li 2001a). Some provincial-level dummy variables were also significant, both substantively and statistically, but there does not exist a strong theoretical explanation for why particular locales might correlate positively with selection for a party school training class. There may be a spatial dimension to cadre selection, which has been borne out by studies that find overrepresentation of western provinces in cadre ranks. Allocating relatively more bureaucratic posts to China’s western regions may be a way to extend the reach of the state to the most distant areas of the country and/ or create patronage links to areas with reputations for restive minority populations (Ang 2007).23

Party schools and promotion 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. Table 3.4 illustrates

23 This logic may explain the ﬁnding for Gansu province, but not for the other three signiﬁcant provincial-level units. Analyzing the effect of training 73

Table 3.4 Total cadres in treated and control groups, by propensity score block

Propensity score block Control Treated Total

0 177 14 191 0.1 79 13 92 0.2 72 46 118 0.4 60 46 106 0.6 24 42 66 0.8 1 15 16 Total 413 176 589

the distribution of estimated propensity scores for individuals by propensity score block. A higher propensity score indicates a higher likelihood that an individual will be selected for the treatment, though not all individuals with high propensity scores actually attended party school. There is an 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 only one comparison in the control group (Table 3.4). Because of the broad overlap in propensity scores across treated and untreated groups, with the exception of those in the highest (0.8 to 1) propensity score block, it is possible to carry out differences-of- means tests across the two groups and determine the average treatment effect on the treated (ATT). To test the effectiveness of the matching procedure, the kernel density graphs (Figures 3.2 and 3.3) demonstrate the distribution of propensity scores across the treatment and control groups both before and after matching. While the kernel densities do not overlap perfectly in the postmatch graph, the distance between the treatment and control groups has decreased after the matching process. This indicates that the matching procedure has identified and sorted individuals into more comparable groups, relative to the prematch groupings. 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 E. Postmatch, there were no statistically significant differences in the means of control variables across the treatment and control groups.24

24 Additional robustness tests, provided in Appendix E, include estimations of the ATT using different matching algorithms. 74 Managing the managers

1 Kernel density of respondents Kernel

0 0 0.2 0.4 0.6 0.8 1 Propensity score Treated prematch Control prematch

Figure 3.2 Prematch propensity score distribution

1.5

0.5 Kernel density of respondents Kernel

0 0 0.2 0.4 0.6 0.8 1 Propensity score Treated postmatch Control postmatch

Figure 3.3 Postmatch propensity score distribution

Using nearest neighbor matching with replacement, over the common support, the matching process yields the ﬁnding that individuals selected for party schools are more likely to be promoted over the course of their careers than individuals with similar qualiﬁcations and characteristics, Analyzing the effect of training 75

Table 3.5 Estimation of average treatment effect on the treated (ATT) enrollment in a non-degree party school training class

Treatment: enrollment in a non-degree party school training class

Dependent variable Treated (N) Control (N) ATT SE t-statistic

Section rank (dummy) 167 77 0.165 0.082 2.016 Department rank (dummy) 167 90 0.126 0.050 2.536 Any rank (dummy) 167 90 0.174 0.068 2.541

Notes: Estimates obtained over the common support using nearest neighbor matching (random draw version). Bootstrapped standard errors are reported (200 replications).

save for the party school training experience. Table 3.5 gives the estimation of the average treatment effect on the treated, which offers a measure of the effect of party school training on an individual’s likelihood of promotion. This table presents the ATT on several binary outcomes of interest, that is, whether a cadre was promoted to a section, department, or any rank during his career. These results are statistically significant. For the section and department ranks, 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 16.5 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 12.6 percent for attaining the rank of department chief. Those who attended party school training were 17.4 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 includes all manner of party school training programs, from core to auxiliary to vocational classes, so the effect of training on promotion might be even higher for certain key-point classes. Because of data limitations, it is not possible to determine the particular type of training class attended (other than whether it is for a degree). In comparison, separate tests to determine the ATT of party school degree programs found no statistically significant effect (Table 3.6). This suggests that individuals who enroll in party schools for a degree are not more likely to be promoted, 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 76 Managing the managers

Table 3.6 Estimation of average treatment effect on the treated (ATT) party school degree

Treatment: party school degree (dummy)

Dependent variable Treated (N) Control (N) ATT SE t-statistic

Section rank (dummy) 51 25 0.140 0.124 1.133 Department rank (dummy) 51 37 −0.020 0.093 −0.210 Any rank (dummy) 51 37 0.098 0.077 1.267

Notes: In order to satisfy the balancing property: (1) the Hainan, rather than Xinjiang, dummy was dropped from the ﬁrst-stage probit regression and (2) a dummy for parents’ CCP membership was substituted for the father and mother’s CCP membership dummies. Estimates obtained over the common support using nearest neighbor matching (random draw version). Bootstrapped standard errors are reported (200 replications).

Table 3.7 Estimation of average treatment effect on the treated (ATT) university degree

Treatment: university degree (dummy)

Dependent variable Treated (N) Control (N) ATT SE t-statistic

Section rank (dummy) 115 60 −0.134 0.062 −2.138 Department rank (dummy) 115 69 0.252 0.083 3.038 Any rank (dummy) 115 69 −0.052 0.066 −0.796

Notes: In order to satisfy the balancing property: (1) the Anhui, Liaoning, and Jiangxi, rather than Xinjiang, dummies were dropped from the ﬁrst-stage probit regression and (2) the “total years of education” variable was dropped. Estimates obtained over the common support using nearest neighbor matching (random draw version). Bootstrapped standard errors are reported (200 replications).

appear to yield payoffs in terms of climbing the career ladder, at least for lower ranks. On the other hand, individuals who attain a university degree stand better chances of promotion to a leading cadre position (Table 3.7). While a university education does not have a positive effect on an individual’s promotion to the lowest section rank, this changes dramatically for promotion to a department, or leading cadre position. Individuals with a university degree are 25.2 percent more likely to be promoted to the department rank than peers without a degree, and this is a statistically significant result. Overall, the findings are somewhat uneven: a low or unranked official with a university degree might not expect higher Testing mechanisms 77

Table 3.8 Estimation of average treatment effect on the treated (ATT) party school training and university degree

Treatment: party school training and university degree (dummy)

Dependent Variable Treated (N) Control (N) ATT SE t-statistic

Section rank (dummy) 38 23 0.090 0.117 0.766 Department rank (dummy) 38 27 0.289 0.110 2.631 Any rank (dummy) 38 27 0.105 0.075 1.395

Notes: In order to satisfy the balancing property: (1) the Shanghai, Liaoning, Gansu, Guangxi, and Xinjiang dummies were dropped from the ﬁrst-stage probit regression, (2) the “total years of education” variable was dropped, and (3) the “SOE level: lower-middle” was substituted for the “SOE level: low” dummy. Estimates obtained over the common support using nearest neighbor matching (random draw version). Bootstrapped standard errors are reported (200 replications).

chances of promotion, but a degree appears to have a strong effect on those in consideration for a leading cadre position. 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 opens 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 conﬁrm that individuals with a university degree and party school experience are the most likely to become leading cadres (Table 3.8). The effect of these combined credentials is to increase the chances of promotion to a department rank by 28.9 percent. The joint effect of party school training and higher education on the likelihood of promotion to a leading cadre position is greater than for either of these credentials alone. There was no effect on the likelihood of promotion to the section rank or a higher rank generally. These ﬁndings are somewhat inconclusive, however, due to dwindling numbers of survey respondents. Only 38 individuals in the sample held both a university degree and have attended at least one party school training program.

Testing mechanisms While the analysis presented here suggests that party school training has an effect on a cadre’s likelihood of promotion, the next step is to consider 78 Managing the managers the mechanisms driving this result. At least two mechanisms may be at play, both of which assist party ofﬁcials with making selection decisions. First, it might be the case that individuals attend these training classes in order to signal their intentions to pursue careers as party functionaries, and they might lobby their superiors or personnel ofﬁcials in order to attain requisite credentials. This “cadre-driven” process involves two key players: the cadre who seeks promotion and the superior who makes the decision to recommend a subordinate for a party school program. In this scenario, the likelihood of patronage ties driving selection is higher. Second, it may also be the case that party schools serve a more impersonal screening function for higher party authorities. If so, these schools constitute a party-driven – as opposed to cadre-driven – organizational strategy for solving the party’s selection problem. These are not mutually exclusive mechanisms and, as discussed below, both are plausible explanations for why party school training increases the likelihood of promotion. Importantly, evidence supports the argument that schools, as sites for screening cadres, are an organizational means to reinforce selection based on party-driven criteria.

Cadre-based signaling In assisting personnel officials with hiring, promotion, and other selection decisions, party schools may be used instrumentally by cadres to signal career intentions. 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, that is, those with strong political ambitions and who have patrons that support their aspirations. The matching strategy employed in the previous section shows that party school training may be used by cadres to signal ambition, but these schools also serve additional functions. If signaling were the only explanation for the effect of party school training on promotion, then it would necessarily be the case that only high ambition (and well- connected) types attend these training programs. In this analysis, 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. Individuals who might signal political ambitions through party membership and frequent interaction with superiors are matched to those who also possess these characteristics, but those selected for party school training are, in the second stage of the empirical analysis, more likely to be promoted. This supports the prospect that party school training accomplishes an Testing mechanisms 79 additional important function for the party; that is, the process of training itself is a means for party authorities to observe and screen for appropriate candidates. Whereas the “signaling only” interpretation of party schools implies that selection for training is the only outcome of consequence, the matching procedure confirms that an additional process is at play: the experience of undergoing training generates useful information about individuals for party officials.

Party-driven screening Selection for party school correlates positively with upward career mobility because training provides party officials with information regarding a particular bureaucrat’s fitness for higher office. 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. Evidence drawn from two sources, field interviews at party schools and examination of post-training career paths, test the plausibility of this mechanism. Interviews and official documents reveal that there is careful evaluation of cadres during party school training programs. Cadres’ performance in training programs are recorded in their professional dossiers, and these training records are reviewed during regular evaluations.25 Organization departments will send monitors to observe and record cadres’ performance during training.26 Such observation mitigates the patronage that may have helped get a cadre into party school, as monitors are supposed to be outside observers. Overall student performance in training, including appearance (biaoxian) throughout discussion sessions, field investigations, and on-site visits, are submitted to each cadre’s dossier. In short, there are organizational links between training programs, the party’s

25 In an interview with a county-level organization department cadre in May 2008, he described how a certain number of points are allocated to training in evaluation rubrics for party cadres (Interview 212). According to this official, individual cadres have a “study and training” (xuexi peixun) component to their evaluation package, of which training performance is worth up to 30 percent. In terms of official regulations, 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). 26 Only one interviewee, at a department (chu)-level training at the Central Party School, 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 hand- written notes. 80 Managing the managers personnel bureaucracy, a cadre’s training performance, and his career prospects within the party and government. A look at post-training career paths also suggests the impact of party school performance on career trajectories. If cadre training itself, and not the selection process for training, serves a screening function, some cadres may experience stalled careers in the period following party school training. Analysis of bureaucrats’ careers captured in the survey data and the official biographies of Central Party School alumni reveals that not all bureaucrats selected for training move to higher administrative ranks. The possibility remains that individuals are not promoted due to factors other than their performance in a training course. However, in combination with interview data, this analysis lends support to the idea that party schools are sites for screening out unsuitable candidates. Within the population of bureaucrats in the survey, 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 Mid-Career Cadre Training Classes at the Central Party School offer additional evidence (Table 3.9). 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 – for example, knowledge of party policies and 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 the one- and two-year training process. Table 3.9 on post-training promotion paths suggests that some screening does occur. Those individuals whose difference in rank, that is, their rank in 2009 minus their rank at the time of training, is zero have stalled in their careers. Based on the available data, approximately 17 percent of trainees

Table 3.9 Promotion patterns of Central Party School trainees, 1995 and 2000 classes

Difference in rank (rank in 2009 – rank at the time of training) N %

0 45 16.73 1 145 55.13 2 64 24.33 3 10 3.80 Total 264 100.00

Sources: CPS yearbooks, ofﬁcial biographies (Xinhua). Central Party School Mid-Career Cadre Training classes 81

Table 3.10 Career tracks of CPS trainees in stalled careers, 1995 and 2000 classes

Career track N %

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

are not promoted at all following their enrollment in the elite CPS Mid- Career Cadre Training Class. While similar to the “stall rate” in the general cadre population sample, this promotion rate must be considered in light of the expectations and resources accompanying enrollment in such a high-level training program. Cadres who are identified for this program are provincial and central leaders of high status and ostensibly high potential; if training was more ritual and signaling, the stall rate should be quite low for this group. Table 3.10 lists the career tracks of CPS-trained bureaucrats who have stalled in their careers despite the investment of one year of high-level training. A total of 45 individuals were not promoted following their enrollment in the CPS Mid-Career Cadre Training Class. The largest share (17 officials) was on a propaganda career track. About one-quarter of those who stalled were in economic management positions, while the remainder were evenly distributed across science and engineering, security, and law-related occupations. To sum, evidence supports the line of reasoning that party school training serves a screening function. Party officials glean useful information about cadres during the process of a party school training program, above and beyond the strategic use of these organizations by ambitious cadres. This is in addition to 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 serve both functions. Further tests might also establish the criteria used to screen cadres within these training classes.

The select few: Central Party School Mid-Career Cadre Training classes While analysis of CGSS data demonstrated that party school training programs have an independent, positive effect on the likelihood of 82 Managing the managers promotion to higher office, this section will consider the effect of party school enrollment on the rate of promotion. It will examine whether party school training has an effect on the timing of promotions within an individual’s career. Career development in the Chinese bureaucracy is a long and complex process, and those who are identified early in their careers stand a stronger chance of achieving high office.27 This section will test the following hypothesis: 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. This hypothesis was tested using an original dataset of the education and career histories of Central Party School alumni. CGSS data are insufficient for at least two reasons. 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 Mid-Career Cadre Training Classes. In principle these are convened in all party schools, at all administrative levels, and they are a core (zhuti) training class for all party schools.28 Second, CGSS 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 Mid-Career 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 assess 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 of individuals who enrolled in CPS

27 This is compounded by such factors as the enforcement of retirement laws under Deng. See Manion 1993. 28 Inerview 216, Central Party School professor, May 2008. Central Party School Mid-Career Cadre Training classes 83

Mid-Career Cadre Training 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 prior to their enrollment. However, personnel data of this nature is not 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 CPS yearbooks (which were also published irregularly), but names for students of the Mid- Career 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 three most recent training classes held in 2000–01 (N=243), 1995–96 (N=160), and 1995–97 (N=39). Focusing on these classes was due in part to the completeness of officially published career histories of current officials. In compiling the educational and professional backgrounds of these individuals, I relied most heavily on official biographies published online. Information was obtained from government web pages whenever possible, such as ministry web pages, local government homepages, and Xinhua news service articles. A descriptive summary of the trainees from these Mid-Career Cadre Training Classes is available in Appendix F. The majority of trainees were men (81.9 percent) and those 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 2000.29 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 in official biographies 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 in the more recent class. This is consistent with

29 This includes those who were part-time graduate students (zaizhi yanjiu sheng), a common educational choice among Chinese officials. In such programs, an individual can retain her office and continue to carry out official duties while completing requirements for an educational credential. 84 Managing the managers

Table 3.11 Administrative rank, CPS Mid-Career Cadre Training Class alumni, 2010

2000 1995 Combined

N % N % N %

Department level 0 0 1 0.5 1 0.2 Vice-bureau level 25 10.3 9 4.5 34 7.7 Bureau level 109 44.9 37 18.6 146 33.0 Vice-minister level 75 30.9 96 48.2 171 38.7 Minister level 10 4.1 35 17.6 45 10.2 Vice-premier level 0 0 2 1.0 2 0.5 Dismissed 2 0.8 3 1.5 5 1.1 NA 22 9.1 16 8.0 38 8.6 Total 243 100.0 199 100.0 442 100.0

the preceding analysis and studies of Chinese political elite that have observed a move toward diverse educational backgrounds and away from more narrow technical training (Li 2007b, 2008b). Almost 10 percent of the 2000 class were law majors, compared to 7 percent listing engineering majors. This compares with 7 and 5 percent, respectively, for the 1995 class. Table 3.11 lists the administrative rank of offices occupied by alumni in 2010. This table summarizes the promotion patterns of 1995 and 2000 CPS Mid-Career Cadre Training Class alumni. Data were collected in 2010, which provides a 10- and 15-year lag between cadres’ training at CPS and their subsequent career trajectories. In terms of administrative rank, the largest share of 2000 class alumni held bureau- level (sitingju ji) positions (44.9 percent) ten years after attending party school. 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 levels is 131, or 65.8 percent of the total class. Also of interest is whether these cadres hold generalist positions in the party-state or are channeled to more specialized careers. It appears that these CPS alumni are scattered throughout party and government bureaucracies. Table 3.12 summarizes the present occupation of these classes. Although it is common practice for officials to hold multiple titles – for example, a city mayor is often a vice-party secretary for that city – Table 3.12 tabulates results based on the first occupation listed for each individual in official biographies. Central Party School Mid-Career Cadre Training classes 85

Table 3.12 Occupation category, CPS Mid-Career Cadre Training Class alumni, 2010

Combined 2000 1995

N % N % N %

Party secretary 38 8.6 18 7.4 20 10.1 Government head 42 9.5 20 8.2 22 11.1 Party administrative assistant (mishu) 14 3.2 9 3.7 5 2.5 Economic development and planning 69 15.6 40 16.5 29 14.6 Security 9 2.0 4 1.6 5 2.5 Foreign affairs 10 2.3 8 3.3 2 1.0 Media 26 5.9 12 4.9 14 7.0 Personnel 16 3.6 8 3.3 8 4.0 Discipline and inspection 12 2.7 2 0.8 10 5.0 Propaganda and party history 20 4.5 13 5.3 7 3.5 Courts 19 4.3 13 5.3 6 3.0 People’s Congress 24 5.4 5 2.1 19 9.5 CPPCC 18 4.1 7 2.9 11 5.5 Other mass organization 9 2.0 6 2.5 3 1.5 United Front 4 0.9 3 1.2 1 0.5 Party school 3 0.7 3 1.2 0 0.0 Education 26 5.9 18 7.4 8 4.0 Social/cultural 24 5.4 15 6.2 9 4.5 Scientiﬁc 21 4.8 16 6.6 5 2.5 Dismissed/retired 5 1.1 2 0.8 3 1.5 NA 33 7.5 21 8.6 12 6.0 Total 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 (fagai- wei) and State Council State-Owned Assets Supervision and Administration Commissions (guoyou zichan jiandu guanli weiyuanhui). Provincial and city party secretaries and government heads were also common positions, totaling 15.6 and 21.1 percent of 2000 and 1995 alumni, respectively.30 Those in sensitive occupations, such as the

30 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. 86 Managing the managers

Table 3.13 Comparing average ages, by rank

1995 CPS class 1995 national averagea

Average age N Average age Two-sample t-test

Department 91 34.8 407,207 48.0 * level Bureau level 124 43.5 35,620 53.2 * Ministry level 26 54.7 2,459 58.9 *

a 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. First, 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 1 percent level. Third, 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. Note: Mean comparison test significance codes: *p<0.01 Sources: Author’s dataset and COD 1999.

security and personnel systems, were less common. This may be due to separate training systems for cadres in these fields.31 To test the hypothesis that invitation to and enrollment in a party school training program increase the rate of promotion, cadre career histories were coded in terms of the age at which an individual assumed office at a particular rank. Table 3.13 summarizes the average age, by rank, for cadres of the CPS Mid-Career Cadre Training Class of 1995, as compared to national averages.32 The 1995 class was used for comparison because it was the closest in time to available national data. Findings support the hypothesis. No cadres attend training at CPS if they are ranked below the department level, hence the impossibility of evaluating whether cadre age at the time of promotion is significantly

31 For example, the Beijing Institute for International Studies, located across from the Central Party School, 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. 32 National data for cadres in 2000 were not available for comparison. Conclusion 87 lower for nonleading ranks. CPS trainees were, on average, 13 years younger than their counterparts at the time they obtained a department- ranked position. At the bureau level, this difference was ten years, while CPS-trained cadres were, on average, four years younger than their 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. A convergence in age across CPS- and non-CPS trained officials appears to occur the higher the administrative rank. 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, support the argument that party school training places cadres on a ‘fast track’ to promotion.

Conclusion This chapter has centered on how the ruling CCP has maintained control over the selection of bureaucrats during its recent period of economic liberalization. Multiple bureaus participate in personnel control in China, and the analytical strategy here has been to focus on one set of organizations and their role in elite production. This chapter has shown how party-managed training academies present a pipeline to higher office within the party-state. As one official summarized, “Training is critical to promotion because we are a one-party system, and the party needs training to control human resources.”33 Investigating the linkages between party school training and promotion reveals the combination of patronage and meritocratic factors at play in building China’s political elite. Findings draw attention to the organizational and political logic behind the persistence of Leninist organizations with revolutionary origins. Analysis of survey data finds that party training schools have an independent and significant effect on cadre promotion rates. This overturns the

33 Interview 34, retired central-level ministry cadre, October 2007. 88 Managing the managers conventional wisdom that officials may be selected for party-sponsored training academies because these officials are already slated for promotion. The matching methodology employed in this chapter reveals that if two identical cadres are in consideration for promotion, and the key observable difference between them is that one went to party school and the other did not, the former is more likely to be promoted. Evidence further suggests that at least two processes are unfolding in these schools: high-ambition cadres may seek to distinguish themselves by signaling their abilities with a party school training credential, but what happens during training also impacts a cadre’s career prospects. The “party school pipeline” identified here mitigates the selection problem facing the ruling party. Single-party systems such as China’s are capable of projecting the party’s authority over large populations and territories. One problem of governance, obtaining critical information regarding a given agent’s talents (and her subsequent ability to serve the party well), must be addressed in order to maintain the party’s ruling capacity. For the CCP, a solution has been continued investment in a set of political organizations that assist with promoting the most promising party officials. Over the past three decades of economic and social reform, the party has maintained formal strategies for managing the managers of the party and government. These strategies, 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).Atthesametime,thesefindings are one piece of a multifaceted process. The content of what is taught in the schools 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 three chapters, which unpack the various processes that have motivated dramatic changes within the party schools themselves. 4 Fusing party and market Introducing market-based incentives to the party school system

Contemporary China’s rise to global prominence over the past three decades presents the puzzle of how its political institutions, crafted in a revolutionary context and for centralizing purposes, 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’scenter.1 With the initiation of market reforms in the late 1970s, individuals at various levels of the party’s organizational framework have increasingly shifted 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, political leaders have begun reforming sites of elite training and education situated within the party itself. Party schools would seem an ideal site for centralized dissemination of updated information on the party line and unifying beliefs throughout the system. This chapter will demonstrate that central authorities, since the early

1 Under different leaders, there have been varying degrees of collective input into policy. Hu Jintao’s administration is believed to have embraced collective decision making more than Xi Jinping’s, for example. With respect to cadre training, policies and directives emanate from three 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 May 3, 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 (7–9 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. The COD, and its local counterparts, also draws up lists of potential students for training classes. 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 party school system.

89 90 Fusing party and market

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 were limitations to this local autonomy. Local schools were 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 included the creation by central authorities of redundancies in training. Other organizations now compete with party schools for training contracts. The result of these reforms is that local party schools possess greater 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 by the disciplining effect of competition: local actors must now weigh whether to be responsive to local, national, and international factors. 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. Competition also pits party organizations against each other for scarce resources, but it is constrained by a combination of top-down policy decisions and local conditions. All of these changes have generated new opportunities and anxieties for schoolteachers and leaders. When asked about the marketization of cadre training, one city party school professor embraced the change, enthusing, “the market is where money is made!”2 A school principal at an inland

2 Interview 131, April 2008. Bringing in market forces 91 school was more inclined to worry: “Now we have to think of ways to raise funds, we can’t depend on the transfer from the government ﬁnance department. But central schools are taking away our business.”3 A mix of opportunity and risk accompanies party school reforms, and school leaders have had to respond in creative ways. This chapter will explore the central and local processes that have brought together party schools and two markets of the reform period. 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 economies 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 solutions to problems both ﬁscal and relating to their core training function.

Bringing in market forces 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 creation of enterprises within government organs illustrates this process. Scholars have identified three waves of administrative enterprise-building (Gore 1998; Lin 2001), beginning with income-generating organizations such as internal government enterprises in 1985. As these ballooned in number, central officials launched a consolidation campaign in 1986 and reduced enterprises by over 40 percent. Controls loosened again in response to official policies permitting “sideline earnings” by government agencies. The underlying rationale was that these earnings would reduce dependence on government subsidies and integrate government organs with the ongoing market transition. In response to perceptions of rampant official misuse of sideline businesses, a second wave, a clean-up campaign, commenced in 1988, and nearly 90 percent of government enterprises shut down by the end of 1989. The third wave of administrative enterprises followed Deng Xiaoping’s 1992 Southern Tour, contributing to a peak of 900,000 enterprises by 1993. Following these waves of enterprise development in the government, the resultant “bureaucratic entrepreneurialism” has entailed “government officials engag[ing] 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).4

3 Interview 167, Communist Youth League party school principal, April 2008. 4 A discussion of state bureaus’“backyard proﬁt centers” can be found in Lin (2001: Chapter 4). 92 Fusing party and market

The process of bringing market incentives into government and party organs was uneven and, as in other policy realms in China, subject to waves of loosened and tightened controls. The party school system was no exception to these more general processes. Driving this story of marketization was a combination of deliberate choices made at many levels throughout the system, for example, new policies originating from central party authorities and initiatives pursued by local party school leaders. Key changes accompanied party school marketization, including the harden- ing of previously soft budget constraints and the relaxation of controls over self-raised revenues by various party and state units. 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.5 As the party matured, decentralization posed less of a threat to organizational unity. In his study of Russian Bolshevik tactics, Selznick observed as follows: “One of the advantages of a firmly established organizational character is the possibility of increasing decentralization without sacrifi- cing unity of policy or stability of command” (Selznick 1960: 65). Economic decentralization has political implications, and the political authority of central authorities has waxed and waned with the devolution of authority and economic resources to localities (Huang 1996;Oi1992, 1999; Yang 2006).6 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). Decentralization also affects the personnel control policies of the party-state. Reforms leading to administrative decentralization have been studied most extensively with respect to personnel appointments and monitoring (Burns 1993; Huang 1995; Landry 2007; Manion 1985). In a reforming personnel management system where decisions are

5 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 et al. 1995; Qian 2003; Qian and Weingast 1997) and those who find no such relationship (Cai and Treisman 2006; Yang 2006). 6 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 recentralize 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 (Oi et al. 2012). Bringing in market forces 93 made “one level down,” party authorities still appear capable of asserting system-wide party policies and rotating officials through posts, in the process limiting the formation of too-powerful local kingpins. Despite efforts by the center to rein in localities and especially the provinces, China remains a remarkably decentralized system: nearly 70 percent of government revenues is generated by localities, a level that places China among the most decentralized of countries tracked by the International Monetary Fund (Landry 2007: Chapter 1; World Bank 2002). 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, subnational units gain the 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. Party schools have been swept up in these decentralizing trends. 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: 114–18). Accordingly, within China’s cadre training schools, local party schools possess limited autonomy in several areas: financial management, curriculum matters, and organizational arrangements.

Limited local autonomy in financial matters As discussed in Chapter 2, funding for party schools is localized. Rather than allocate funds at the central level and then transfer financial support downward through the school system, party schools from the provincial to the township levels receive funding via a horizontal transfer from their respective local government finance departments. This arrangement has been in place since at least 1983.7 In the pre-reform period, revenues and

7 See Appendix 3: Main Points of the National Cadre Training Plan in the Notice of the Organization Department of the Central Committee, October 5, 1983, in Burns (1989b, 94 Fusing party and market expenditures were centralized in a mode common to command economies of the Soviet model. Subnational administrative levels sent up revenues and received budgetary allocations from central authorities (Ang 2009: Chapter 2). One implication of this more recent local fiscal independence is a harder budget constraint, and schools are left to flour- ish or flounder as their local fiscal situation allows. Given the 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 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.8 This difference is even more striking when considering the higher number of leading cadres per capita in Province B compared to Province A.9 Such variation is the result of whether a local economy is booming or sluggish; no significant central-level transfers reach party schools to equalize differences. These horizontal disparities are also reflected vertically within the system. Schools at the lowest, township level are often 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 pressuring 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). These upgrades must be done with the provision of funds from local finance bureaus, and each locale’s planning bureau should allocate capital

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. 73–4). 8 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. 9 For the period from 1981 to 1998, Province B’s total number of leading cadres as a percentage of the province’s population was 0.028, compared to 0.020 for Province A. Source: COD (1999). Bringing in market forces 95 funding in support of these training endeavors (Ibid., Article 24). Such top-down pressures do little to relieve disparities across locales given the distance between the center’s calls to update and upgrade cadre training and the varying realities on the ground.

Local school control over training content There exists curricular autonomy in the party school system. Higher-level schools cannot impose their training content on schools at lower administrative levels. This is illustrated by the difference between the availability of texts for party school courses and actual adoption by individual schools. The Central Party School publishes textbooks on socialist theory, leadership, and party history, among other topics, ostensibly to guide curricula at subnational levels. Local schools decide whether to adopt these materials. When asked about these CPS texts, 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.”10 Similarly, a provincial party school head declared, “The CPS and National School of Administration each have their set of training materials, but ...these don’t match our training content. The content of trainings changes too fast. For example, I used to teach public policy analysis, and now I teach public services provision. Are we not a service- oriented government now?”11 Examining party school syllabi from the past few years offers a glimpse of the diversity in teaching materials used by local schools. Beyond the standard texts on leaders’ selected writings and central policy documents, some schools include more current books on management and world affairs. The Hubei party school, for example, assigned popular books such as The China Model and Chinese-Style Management in its reading list for 2011 training classes.12 Other schools have opted to emphasize ofﬁcial publications. In 2010 and 2011, the Shanghai party school assigned a mix of government documents, CPS-published books, and publications from party presses such as the Central Propaganda Department’s.13 Some schools rely heavily on CPS texts: at the Jiangxi provincial party school, nearly three-quarters of assigned texts in fall 2008 were CPS publications.

10 Interview 75, December 2007. 11 Interview 66, November 2007. 12 See Zheng Yongnian, The China Model (zhongguo moshi) (Zhejiang People’s Press, 2010) and Zeng Shiqiang, Chinese-Style Management (zhongguo shi guanli) (China Social Sciences Press, 2005). 13 Books from the Central Propaganda Department’s press, Study Press, included a reader on “scientiﬁc development” and other theoretical issues. 96 Fusing party and market

Yet 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 depths. Party school ofﬁcials have reported in interviews that they do turn to other party schools for guidance in curriculum matters. One provincial ofﬁcial characterized the relationship as one where schools will “imitate each other” (huxiang mofang).14 This is in part because party schools transmit their training syllabi up and down the party school system as part of their advisory relationship. Provincial schools will look to the CPS and other provincial party schools for ideas on updating course content and pedagogy. Some teachers have welcomed non-party school training providers as yet another source for guidance on updating curricula.15 This diversity lends dynamism to the system. As a former county party secretary pointed out, “the party school head has leeway to structure the school’s curriculum and teaching methods, so there is some unpredictability.”16

Local organizational experiments Separate from but related to curricular autonomy, there is some decentralization in local organizational matters. Local party schools possess the authority to test novel organizational arrangements. In Province A, for example, the party school of the provincial capital is merging two county schools and six district schools under one roof, presumably with the provincial capital determining training content. This organizational shuf- fling was conceived by a city vice-party secretary and party school vice- principal in order to “improve teacher quality” and “consolidate resources.”17 Interviewees reported that this decision reflects concerns with the financial viability of and staff qualifications at lower level schools, so city leaders decided to take matters into their own hands. The idea is to combine programming and management for several schools under one school committee. The city vice-party secretary and party school vice-principal came up with this plan on a flight they were sharing. There will be a new four-star hotel, with

14 Interview 182, provincial party committee policy research ofﬁce director, April 2008. 15 Interviews 143 and 144, provincial party school professors, April 2008. During this interview, one professor showed me a Powerpoint presentation on Pudong’s Cadre Executive Leadership Academy training content and declared, “Within the next six to eight years, the gap between cadre executive leadership academies and local party schools will be reduced to nothing. Local party schools will transition into the CELA model.” 16 Interview 182, April 2008. 17 Interviews 207 and 208, provincial and city party school teachers, May 2008. Bringing in market forces 97 capacity to house 1,600 students, an entertainment center ...Each school will still be responsible for organizing their own core classes, but non-core classes can be held anywhere. The old district school heads will be on a new school committee, and a new “second teaching district” will be in charge of the combined classes of the old district schools. Some district party school leaders don’t agree with this plan, they are afraid they won’t be able to claim the training money in their districts. But the party schools are too scattered. Basically the quality of grassroots party schools is too low.18 This consolidation plan appears to beneﬁt the city party school the most, given its higher status and ability to grab auxiliary (outside-the-plan) classes from the district schools. 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 have the leeway to pursue new organizational arrangements.

Official incentives to “dive into the sea” of market activity The dependence of party schools on funding from local finance bureaus affects 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 the planned economy 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). 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 local finance bureaus, but increased revenues would provide party school employees with bonus income. While national and local party school budget information is not publicly available, some financial information comes to light during field interviews. Party schools are considered fully funded public service units (quan’e bokuan shiye danwei), and in theory local finance bureaus transfer enough funds to schools to cover operating expenses.19 In reality, transfers of public funds leave virtually no room for organizational slack.

18 Ibid. 19 Ang (2008) offers a lucid description of the distinction between public service units (or extrabureaucracies) and party or government bureaus. 98 Fusing party and market

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 School-generated revenue Training expenses (e.g., teaching materials, site visits, studying in other institutions) If: Training is inside-the-plan Local finance bureau and trainees’ work units If: Training is outside-the-plan Trainees’ work units Salaries Local finance bureau Employee benefits (e.g. mobile phone School-generated revenue service allowance, meal allowance)

This 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 activities. There also appears to be 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 been reduced, leaving schools severely constrained (Pieke 2009: 132– 40). Provincial-level training institutions in both Provinces A and B did not report the sort of declines experienced by their western counterparts.20 In the 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, would have to be covered through auxiliary (jihua wai, or “outside-the-plan”) training classes, degree programs, and other revenue schemes. Table 4.1 offers a breakdown of general expense categories and expected income sources for local party schools. There are five major

20 Interview 95, provincial socialism institute principal, December 2007; Interview 177, provincial capital party school teacher, April 2008. Bringing in market forces 99 categories of school expenditures. First, general operating expenses are covered by transfers from local finance bureaus; however, schools must bridge shortfalls with school-generated revenue. Second, capital construction costs, which are part of the party’s larger building boom, are often borne by local development bureaus via the local finance bureau. Revenues from the sale of campus lands can be part of the capital construction fundraising process. Some infrastructural items, such as equipment and facilities upgrades, may also be supported by party school-raised revenues.21 A third major expense for party schools, training costs, are offset by various income streams. Training classes organized as part of a locale’s annual training plan are accompanied by transfers from the local finance bureau, though training costs may be split with the work units of trainees.22 If a training class is not part of an annual plan, then the trainees’ work unit is responsible for training expenses.23 In these cases, training rates are subject to negotiation between the party school and sending work units. Fourth and finally, schools must manage personnel costs. Employee- related benefits constitute another party school expense category, as well as an important office perk. 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 certain 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.24 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 1,000 and 1,500 yuan per

21 Interview with a county party school vice-principal, July 2008. This vice-principal mentioned air-conditioning units, computers, and classroom projectors as items purchased by the school from school-raised revenues. 22 Interview 104, cadre in a city business department January 2008. This interviewee and others estimated training costs at between 50 and 100 yuan per day per trainee for city and county party schools (Interviews 21 and 22, county-level party school vice-principal, December 2006; Interview 104, county-level party school vice-principal, January 2008). 23 In interviews with city- and provincial-level party school teachers in Province B, they claimed that work units may dedicate anywhere from 2.9 to 10 percent of their budgets to training expenses. This lower threshold of 2.9 percent was recently increased from 1.5 percent. Interviews 143, 144, 176, and 177, April 2008. Under Xi Jinping, however, training budgets have been cut by as much as half. 24 A detailed discussion of the components of cadre wages and benefits can be found in Whiting (2006). 100 Fusing party and market employee per month.25 Supplements can include direct transfers to employees to cover expenses associated with mobile phones, transporta- tion, and hosting meals, among other categories.26 In one wealthy county school, the school vice-principal reportedly receives an 1,800 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.27 In summary, central authorities’ conscious decision to decentralize the management of party schools created new conditions for these organizations. 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 transfers from local finance bureaus. This discretion extended to other areas of operation. In the reform period, party school leaders could also initiate organizational changes and reforms in school programming and curricula. While new ventures or initiatives pursued by party schools carried certain risks, these risks were compounded by another development in the organizational environment within which these schools operated. Beginning in the early Deng Xiaoping period, party schools were exposed to a central aspect of the market: competition.

Motivations to marketize training Party schools once enjoyed a near monopoly on the training of party and government leaders.28 Over the course of diversifying activities in a new market context, the leaders of party schools began to face pressure from other suppliers of training services. Figure 4.1 depicts the variety of central-level cadre training organizations now in existence, many of which were established from the 1980s to 2000s. Over these three

25 Interview 177, April 2008. This cadre went on to say that leaders can expect supplements upward of 3,000 yuan a month. 26 Interviews 162 and 163, two provincial party school teachers, April 2008. 27 Interview 191, county party school vice-principal and teacher, May 2008. 28 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 (xitong) and SOEs are considered subsidiary schools (fenxiao) of the party school system (see, for example, the Beijing party school Web page for a listing of its subsidiary schools in ministries and SOEs and Beijing district party schools: www.bac.gov.cn/web/swdx/about/Link.aspx?NodeID=1&Link=区县站点导航 &Tag=3#, accessed July 3, 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). Motivations to marketize training 101

Central Committee State Council of the CCP

Central United National School Government Central Party Organization Front of ministry School Department Department Administration training centers

Four branch Cadre executive United Front schools and leadership China Socialism cooperative academies: Institute profit-sharing Pudong, programs with Jinggangshan, local party Yan’an schools

Figure 4.1 Central-level organizations of cadre training decades, new and existing institutions were allowed, often by central decree, to plan and implement training programs for cadres. These institutions include universities, administration schools, new training centers opened by other party units, and international competitors. New organizational players were given official sanction to claim training market share from party schools, and they have done so with varying degrees of success. Motivations to diversify cadre training ranged from the desire to expand the system’s capacity 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 et al. 2000), Deng’s generation of leaders nonetheless perceived a shortage of managerial talent across these bureaucracies. 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). Nearly two decades later, in the 2001–05 102 Fusing party and market

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 2,000 for the five-year period. This was combined with the stipulation that all cadres at that level should have three months of training within a five-year period. Since cadres ranked at the provincial level numbered over 2,500 by 1998, branch campuses (fenxiao) would have to make up the difference (Yang 2002: 286–95).29 This was logistically challenging and provided motivation to expand the system. Accordingly, the training plan called for “speeding up the construction of cadre education training bases and reforming each administrative level’s party school education as the main channel for rotating party and government leading cadres through training.”30 By 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.”31 The numbers are revealing. Shambaugh observes that the CPS could train up to 5,000 cadres a year by the early 2000s, so circulating tens of thousands of central-level cadres through this one institution would take well beyond the five-year cycle laid down in official training plans (Shambaugh 2008a: 143).32 This is not a recent problem, as capacity issues plagued the national cadre training system during the early reform period, when there was a big push in the 1980s to train new party and government managers (Paltiel 1990: 593). 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 result was 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.33

29 Exact numbers of cadres at each rank are difﬁcult to locate after 1998 (see COD 1999). 30 Section 4, 2001–05 National Cadre Education and Training Plan. 31 Interview 66, November 2007. 32 According to COD (1999), there were 41,698 central-level cadres in 1998. 33 Quoted in Liu (2001), p. 67. Motivations to marketize training 103

Concerns with the quality of party school training were addressed more directly in the 2001–05 National Cadre Education Training Plan, which sought organizational diversity as the means to “raise training quality.”34 The plan stipulated that party schools “should 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 recom- mended 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: 142–5). In official training plans, cadre training was to include multiple actors, beyond traditional party schools. Cadres were another source of the reform push. In evaluations of party school training programs, participating cadres complained about various aspects of the experience, from old-fashioned teaching methods to the lack of relevance of training content for cadre work on the ground.35 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).36 A party school teacher summarized these bottom-up pressures and resulting changes: Cadre training has been opened (fangkai) to all institutions. Why? Because in the past, party schools overemphasized political education. They discovered, through student evaluations, that cadres were not happy. Cadres thought party school training was empty, without practical content. These evaluations were not initiated by the top; they were initiated by teachers. So training opened up to competition to deal with this problem.37 In another interview, a cadre complained, “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!”38 Such critiques were 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

34 Article 2 on “Fundamental Principles” (jiben yuanze). 35 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). 36 Interviews 207 and 208, provincial party school teachers, May 2008. 37 Ibid. 38 Interview 104, city government cadre, January 2008. 104 Fusing party and market improving training; they are just interested in their own job security and benefits.”39 In Province B, a similar realization was dawning: the party school system was at risk of becoming ossified in the reform era and increasingly disconnected from problems that cadres were facing.40 To be fair, there were calls for reform as early as 1987 by individuals within the Central Party School, though it is unclear how much momentum these calls gained within the school or system wide (Paltiel 1990: 598).41 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. Official documents suggest that central leaders considered competition more capable of shifting resources quickly to match new developments. The 2006 Cadre Education and Training Trial Regulations called for “a system of orderly competition” (jingzheng youxu de jigou tixi) between training institutions to “optimize” (peihua) resource allocation.42 This was echoed in the 2010–20 Program for Cadre Education and Training Reform, which asserted that competition across central-level training centers would “better allocate ... resources” and prompt more efficient organizational innovation in a way that top-down planning, with its informational disadvantages, could not.43

Increasing national training capacity Early in the reform period, various key actors within the party and government were calling for reforms in personnel management, speciﬁcally improvements in cadre training. Zhao Ziyang’s March 1982 administrative reform package proposed the creation of training-in- rotation programs (Burns 1983). The Central Organization Department advocated for the construction of new cadre colleges and the integration and expansion of cadre training programs in universities and other educational institutions.44 Ofﬁcial declarations spelled out

39 Interview 212, May 2008. 40 Interviews 143 and 144, provincial party school teachers, April 2008. 41 At CPS, a formal appraisal system was put in place in 2007. Students rated lectures on a 10-point scale, and below-average teachers could be suspended from teaching duties. See Eisenman (2012:6–7). 42 Chapter 5, Articles 27 and 28. 43 The 2010–20 Cadre Education and Training Reform Plan is available online at http:// politics.people.com.cn/GB/1026/12467912.html, accessed September 27, 2012. 44 This was not an entirely foreign proposition. Since the early 1950s, delegations of cadres were sent to the Soviet Union and Eastern Bloc, and small numbers of cadres were also selected for university education. See Liu (2001), p. 98. Increasing national training capacity 105 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, key-point colleges and universities were to create advanced studies and training programs for cadres.45 There was also a call to increase construction of cadre colleges, whose main task was vocational training.46 The universe of training institutions in the early reform period thus included party schools and a variety of other schools charged with the modernization of China’s cadre corps. After the turn of the century, central-level regulations again declared that other institutions such as universities and research institutes “take on the task of cadre education and training.”47 Deepening this initiative in 2009, the COD designated 13 national-level universities as cadre training bases.48 Party schools responded in several ways. Universities and cadre colleges were to boost technical skills, not stand in as authorities of party doctrine. Party schools still retained their role as a pipeline for promotion; only party schools had long-term training classes for those bureaucrats tapped for top leadership positions.49 University party committees also wielded their considerable inﬂuence to draw boundaries around the kind of cadre training organized by their institutions. At universities ranging from Xiamen University to Renmin University (in Beijing), party secretaries could shape the direction of management institutes

45 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. 74–5). 46 Literacy and vocational courses were to constitute 70–80 percent of total credit hours, with Marxist theory filling the rest. Ibid., p. 73. 47 See the 2006 Cadre Education and Training Regulations (Trial) and 2008 Cadre Education and Training Regulations (Trial), Chapter 5, Art. 27. 48 These included the following universities: Peking, Tsinghua, Renmin, Beijing Normal, Fudan, Xi’an Communications, Harbin Industrial, Zhejiang, Nanjing, Sichuan, Nankai, Wuhan, and Zhongshan. See Yu (2014). Pieke (2009) notes how three of the most prestigious universities in China, all based in Beijing, have profited from the opportunity to train cadres in management and non-ideological coursework (p. 127). These are Peking University, Tsinghua University, and Renmin University. In terms of training volume, the School of Continuing Education at Tsinghua trained 107,065 individuals in 2010, divided over 1,735 training classes. These figures were 87,763 and 1,884, respectively, in 2009. On Tsinghua’s SCE, see www.tsinghua.edu.cn/publish/jxjy/e_edition/ about_sce.htm, accessed January 12, 2013. 49 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. 106 Fusing party and market that were formally responsible for secular public administration training (Williams 1993). While effective, these efforts could not prevent the entry of universities as partial competitors to party schools in an emerging national training market.

Building government capacity: schools of economic management and administration During Deng’s early economic reforms, Beijing became the site for a new type of training facility: economic management colleges for cadres (jingji guanli ganbu xueyuan). Such colleges were opened throughout the country, and a handful of them have since merged with administration schools.50 Others have been absorbed by or combined with other schools.51 The first such cadre college, founded in 1979, was created to train cadres in related disciplines such as finance, commerce, and information technology, in addition to market economics. Schools were established throughout the country in the 1980s, and today the enrollment for their degree programs is open to the general public. These early entrants into the cadre training market were largely intended to increase system capacity for building the professional skills of cadres. This impulse was deepened in the 1980s with the creation of a more extensive network of administration schools. Beginning in 1987, the movement to build a system of public administration schools gained momentum with a resolution passed at the Thirteenth National Party Congress.52 Shortly after, construction of the Beijing campus of the National School of Administration (NSA) began, and, due to delays wrought by the Tiananmen protests, in 1994 the organization officially opened its doors (Huang 1993;Li1993).53 This central-level school was part of Zhao Ziyang’s public administration reform agenda; the NSA is modeled on the French Ecole Nationale d’Administration, which

50 At least 22 out of 31 provincial-level units were homes to these colleges. At the provincial level, mergers between these cadre colleges and administration schools have occurred in Anhui, Fujian, Shandong, Shanghai, and Tianjin. 51 This is the case for the Zhejiang cadre economic management college, which has since become part of the Zhejiang University of Technology (Zhejiang gongye daxue). In Xinjiang, the cadre economic management college merged with the provincial finance institute and tax school to become the Xinjiang Finance Institute (Xinjiang caijing xueyuan) in 2000. 52 “NSA: Cradle of China’s Public Servants,” People’sDaily(English service), July 10, 2000, http://english.people.com.cn/english/200007/10/eng20000710_45138.html, accessed February 21, 2006. 53 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), p. 100. Increasing national training capacity 107 serves as a pipeline for French civil servants (Paltiel 1990: 595). Organizationally, the NSA falls under the guidance of the State Council, State Education Commission, and the Ministry of Personnel (Huang 1993).54 Since then, a web of administration schools has spread to the provinces and subprovincial levels. Officially, administration institutes (xingzheng xueyuan) 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 as blurred as the lines between party and state in China’s political system. From the provincial to county levels, party schools have combined campuses with administration institutes. Table 4.2 lists whether party schools and administration institutes at the provincial level share a campus under the “one set of classes, two signs” (yi tao banzi, liang kuai paizi) joint school arrangement. Combining schools on one campus has not been exercised uniformly throughout the system; 6 out of 31 provincial-level units have separate party schools and administration institutes. Most administration schools share campuses with party schools, and it is not uncommon for single school administrators to hold parallel, dual titles across both organizations. There are multiple reasons for merging these organizations. One rationale given for this system of “two signs, one set of personnel” (liang kuai paizi, yi tao renyuan) is to reap the benefits from overlap in organizational goals and resources. In these arrangements, the party school retains higher status and authority.55 The end result is a consolidation of training resources, which can compensate for limited capacity at subnational levels. Most crucially, merging these schools has the benefit of enhancing party control over the training of government officials.

Intraparty competition: organization departments join the training market Despite their mandates to engage in cadre training, universities and administration institutes both lacked the stamp of party authority that party schools enjoyed. A more recent development presents a clearer challenge. The newest additions to the universe of cadre training schools are a variety of organization department–sponsored training schools. At the national

54 This school was renamed the Chinese Academy of Governance in 2009. 55 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 institute 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). 108 Fusing party and market

Table 4.2 Provincial-level party schools and administration institutes

Year party school Year administration Place Merged campus? established institute established

Anhui No 1951 1991 Beijing (in 1993) 1950 1988 Chongqing Yes 1997 1997 Fujian Yes 1950 1995 Gansu No 1952 1989 Guangdong (in 2001) 1950 1990 Guangxi (in 1994) 1950 1994 Guizhou Yes 1950 1997 Hainan Yes 1951 1988 Hebei No 1987 Heilongjiang (in 2006) 1948 Henan Yes 1949 1996 Hubei (in 1993) 1950 1993 Hunan (in 2003) 1951 1953 Jiangsu (in 1992) 1953 1992 Jiangxi (in 2001) 1950 1981 Jilin (in 2006) 1948 Liaoning No 1954 1990 Neimenggu (in 1995) 1948 1995 Ningxia Yes 1958 1996 Qinghai Yes 1951 1991 Shaanxi No 1934 Shandong No 1938 Shanghai (in 1989) 1949 1986 Shanxi Yes 1949 1986 Sichuan (in 2001) 1952 1997 Tianjin (in 2001) 1953 2001 Xinjiang (in 2000) 1956 2000 Xizang (in 1991) 1961 1991 Yunnan Yes Zhejiang Yes 1949 1988

Source: Author’s dataset. level is a trio of cadre leadership academies. These academies are funded and managed by the Central Organization Department, 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.56

56 The ﬁgure given during an interview with a Pudong professor was 900 million yuan, or approximately US$112.5 million. Interview 187, April 2008. Increasing national training capacity 109

Earlier efforts to expand the country’s organizational capacity for cadre training differ from this initiative. The decision to pull universities and administration institutes into the domain of cadre training was aimed more directly at 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. In the case of administration institutes, merging campuses with party schools had the effect of folding state resources into party organizations. Later efforts by the COD to create a new system of cadre training academies more explicitly challenged the dominant position of party schools and established an alternative training model within the party. The 16th party congress in 2002 saw the introduction of this intraparty 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).57 These academies, which would remain distinct and separate from the party school system, were to train the population of elite cadres once channeled exclusively to the Central Party School.58 The lofty goals articulated at their founding overlapped considerably with those of party schools. In a 2005 letter to commemorate the academies’ opening, President Hu Jintao proclaimed, CELAP, CELAJ, and CELAY ...are expected to serve as valuable platforms in leadership training. They play an important role in ensuring that we uphold the Party’s ﬁne traditions and values, improve the leaders’ capacities in building [a] modern socialist country, strengthen the Party’s leadership competency of governing for the people, and preserve the Party’s vanguard nature ... [By] innovating training curricula, modernizing training methodologies, integrating

57 See the October 18, 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 June 29, 2009. The locations chosen are representative of the party’s bases of legitimacy: in a revolutionary past and bright economic future. Each school was to have a particular focus, i.e., Pudong to train cadres on the economic reform process, Jiangxi on the party’s revolutionary history, and Yan’an on problems in “backward” regions. The challenge facing each school has been developing permanent faculty and staff with expertise in these respective areas. I thank an anonymous reviewer for pointing this out. 58 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 classes. Interview 184, CELA director, April 2008. 110 Fusing party and market training resources, and optimizing training teams, the three institutions are expected to make significant improvements in overall training quality.59 These academies would find new means to meet old goals. 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 they will be eliminated (taotai).”60 Cadre executive leadership academies bring new training experiences to China’s cadres via diverse programming, much of it taking place outside of conventional classrooms. Experiential learning is one selling point for the Jinggangshan and Yan’an campuses, where cadres can re-enact moments in CCP history. In some cases cadres don uniforms and carry heavy loads on journeys made by earlier CCP revolutionaries; the goal is a deeper connection with the party’s roots and “serve the people” ethos. The Pudong campus features several “labs” where cadres can study international financial transactions, urban planning, and crisis management, among other topics. Many of these labs operate with strategic partners such as the Shanghai Futures Exchange, Xinhua, and IBM.61 Such new approaches to training have challenged the “speeches in lecture halls” model that was dominant in party schools. Another difference lies in the emphasis on leadership development at the CELA. Organization department officials, who favored a shift toward “leadership science” and management in cadre training, saw the party school system as an obstacle to realizing these objectives. Negotiating the creation of a new system of high-level training academies was one consequence of the jockeying between these two bureaucracies (Chin 2011). At CELA, “the goal was to identify groups of trainees who could act as managerial innovation entrepreneurs, first-adopters and diffusion agents ... those best positioned to promote leadership innovation and managerial reform in their respective institutional settings” (Ibid., 29). These new academies, more than their party school counterparts, would disseminate new ideas about leadership.62

59 CELAP English language web page, “Congratulatory Letter by President Hu Jintao,” http://en.celap.cn/art/2013/9/2/art_2061_29660.html, accessed December 7, 2008. 60 Interview 184, CELA director, April 2008. 61 A summary of CELAP’s labs is available on their English language Web page, http://en. celap.cn/col/col2092/index.html, accessed June 10, 2014. 62 This emphasis on leadership is apparent in a CELAP publication, Trends in Leadership Education (lingdao jiaoyu dongtai), which contains practical articles on topics such as using new technologies and public-speaking techniques; theory pieces are few compared to CPS publications such as Study Times. Increasing national training capacity 111

During the period between the 16th and 17th party congresses (2002– 07), local organization departments also began establishing their own cadre training schools. 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).63 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).64 These new training bases divert cadre-trainees, and the training funds that accompany them, from party schools to other party organs. One provincial party school professor gave her impressions of the motivations underlying these changes: This was to ensure (baozhang) competition. In China, spending a lot of money, investing in new buildings, is a way to show that this project is important. Perhaps they were built because of vested interests. But central authorities for sure are emphasizing cadre training ...Maybe there will be a separation of labor based on market forces. We are in an exploratory period.65 An additional motivation for organization departments to build new training bases rests in concerns about cadres’ loss of respect for party school training. In the words of this organization department cadre, [Organization-department training centers] turn training into something serious again [and] 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 with party schools]. This is unlike the cadres who come to the organization department’s training center; they are very well-behaved (guaiguai).66 Opening new competing academies 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

63 The organization department had been organizing training classes for grassroots cadres since 2002 but decided in 2006 to open a formal training center. Interview 212, city organization department cadre, May 2008. 64 The plans to build the academy were announced in 2004. Total construction costs are reported to be 65 million yuan. 65 Interview 211, May 2008. 66 Interview 212, city-level organization department cadre, May 2008. 112 Fusing party and market organization department–approved training providers. The emphasis in these new programs has been on flexibility in training content and cadre choice. One prominent example is Shenzhen, which has been at the forefront of training reforms in the country. Since October 2003, city civil servants and party cadres have participated in a “self-selection” (zixuan) training program in which they may select training sessions from a course catalog. By law, bureaucrats must fulfill five days of training each year, and in Shenzhen they may meet this requirement by choosing their own training courses. Three training days consist of mandatory theory and party-building classes and the remaining two training days may be fulfilled with elective courses.67 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 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.68 Whereas the city party school focused on classes such as “Research and questions in ‘Three Represents’ important thought” and “Research on leadership capacity,” the city university featured courses from its Master of Public Administration (MPA) and Master of Business Administration (MBA) programs.69 In this training experiment, the party school has become one of several providers of elective classes (zixuan ban) for local bureaucrats. 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 that 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 full-time residential courses organized exclusively by party schools, this “self-selection” training model has put party schools in more direct competition with other

67 Shenzhen’s “self selection” training guidelines are available at www.szps.gov.cn/zxpx/ detail.aspx?NewsID=556, accessed August 9, 2009. 68 “Shenzhen Cadre ‘Self Selection’ Training will Start, Six Training Institutions will take up Positions,” Shenzhen Economic Daily, October 8, 2003, available online at www. southcn.com/news/dishi/shenzhen/shizheng/200310080733.htm, accessed August 8, 2009. 69 Ibid. Increasing national training capacity 113 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.70 Whether this model will spread to other locales remains unclear.

Competition goes global Beyond domestic competition and experimentation, party schools must also contend with international training providers.71 Training activities have been initiated via centralized efforts and at the behest of local organization departments. Centrally, the China Development Research Foundation (zhongguo fazhan yanjiu jijinhui) was created in 1997 by order of the State Council, and it conducts several high-level training programs.72 The Central Organization Department oversees and selects participants for these courses. Most prominent is an annual training program for cadres ranked at the bureau level and above, who attend classes at Tsinghua University’s public management school and Harvard University’s Kennedy School of Government. Beginning in 2002, around 60 ofﬁcials were sent each year to these schools for short-term training programs in public administration.73 An Executive Leadership Program, initiated in 2005, offers groups of approximately 30 managers of large SOEs to study at Cambridge University. Topics in this month-long course include such mainstays of the international business world as global supply chain management, risk management, and mergers and acquisitions.74 These high-proﬁle programs offer China’s political and economic elites access to cutting-edge ideas, but there has been a tightening of overseas

70 “Report on Guangdong Study Tours,” www.hnredstar.gov.cn/yueyang123/gbjy/gj_gjdt/ t20060321_52030.htm,and“Several Insights from Shanghai, Shenzhen, and Ningbo Party School Inspections,” www.swdx.dl.gov.cn/yxlt/Document/5063/5063.html, accessed August 8, 2009. 71 International exchange has been an important component of cadre training since the early days of the CCP. By 1921, the Soviet Union had become the “principal foreign training ground for Chinese revolutionaries” (Price 1976: 30). In 1980, Dalian was the home of the National Center for Industrial Science and Technology Management, which hosted US business school instructors (Paltiel 1990: 591). These programs differ, however, from the competitive bidding process detailed here. 72 See www.cdrf.org.cn for an overview of the foundation’s history, purpose, and training and research programs (accessed December 19, 2012). 73 “China’s Central Party School Trains 50,000 Officials in 30 Years,” Xinhua, October 2, 2007. On the CDRF’s English website, this program is translated as China’s Leaders in Development Executive Program (gonggong guanli gaoji peixun ban). 74 A brief summary of each Executive Leadership Program class is available at www.cdrf. org.cn/plus/list.php?tid=21, accessed December 19, 2012. 114 Fusing party and market training since the changeover to Xi Jinping’s administration. In the case of the public management program organized by CDRF, Tsinghua, and Harvard, class size has been cut in half, from around 60 officials in early cohorts to 30.75 The duration of these classes has also been reduced, from three months to just over one. Recent restrictions on overseas study may be a response to early efforts by the Xi administration to demonstrate that it is serious about eliminating corruption, since travel abroad is widely perceived as vacation rather than work time.76 Another motivation may be to strengthen the role of Chinese training providers in elite education. There has also been a less centrally driven process of bringing international providers into China’s cadre training market. Since at least 2003, local party committees, through their organization departments, have been issuing public calls for training proposals.77 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 competition between training providers for local contracts.78 In addition to bids from Chinese universities, party schools, and other training centers, these calls attracted applications from universities located abroad. Introducing international competition to the allocation of training contracts is, as with domestic competition, a cost-saving measure. Prices for training courses are dictated by market supply rather than mandated by the plan. Furthermore, tendering bids is a means to introduce new ideas to the realm of cadre training. Table 4.3 lists various local party

75 I thank an anonymous reviewer for drawing my attention to these restrictions in overseas training. It appears that the class reduction began in 2013, though class sizes are not given for every year. For an overview of the 2013 China’s Leaders in Development Executive Program, see www.cdrf.org.cn/plus/view.php?aid=1088, accessed July 1, 2014. Reports of these training classes, dating from 2002, are available at www.cdrf.org.cn/plus/list. php?tid=20&TotalResult=11&PageNo=1. 76 See Xinhua’s “Central Organization Department: Public Funds may not be used during Cadre Training for Entertainment,” March 18, 2013, for one example of anticorruption regulations passed for cadre training programs. 77 Yang (2004) notes how experiments with market-based government procurement began in major cities such as Beijing and Chongqing by 1997, eventually spreading nationwide and to many government goods and services by 2000. See Yang (2004), pp. 198–207. 78 Ofﬁcial declarations of the motivations underlying the government procurement system for cadre training programs are available via www.chinabidding.org, a website authorized by the State Development Planning Commission to post bidding notices. See www. chinabidding.com/zbjg-detailTwo-2534138.html and www.chinabidding.com/zbjg- detailTwo-2534139.html, accessed October 2009. The Guangdong provincial organization department’s web page also refers to how a bidding process, which includes party schools, administration schools, cadre academies, universities, and research institutes will “improve training performance.” See www.gdzz.cn/javaoa/article/articleview_simple.jsp? act=view&articleID=999884555479ab276880f95f70f56215&catalogID=8cf83a421ad83 bf7f8dfcf1c0374f696&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,accessedOctober 2009. Increasing national training capacity 115

Table 4.3 A selection of training bids, by locale and awardee

Training contract Year Locale awarded to Training topic(s)

2003 Wuhan, Hubei Tsinghua University Public Administration 2004 Mianyang, Sichuan Southwest Issues for female University of cadres Science and Technology 2006 Beijing >20 Chinese and Leadership and overseas management universities (such Urban planning as Peking Law University, Economics University of Toronto, and Georgetown University) 2006 Shenyang, Liaoning Korean university Rural development (not speciﬁed) 2006 Sichuan National training E-governance center on information technology 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 Leadership School of Education and Training 2009 Heilongjiang 5 institutions: Mid-Career Cadre Harbin Institute of Training Class Technology Heilongjiang University Peking University Marxism Dept Jilin University School of Administration Tianjin University

Source: Author’s dataset. 116 Fusing party and market 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 the University of California, Los Angeles 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 training classes for mid-career 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 Mid-Career Cadre Training Classes. There are similarities and differences between the intraparty competition discussed here and previous scholarly work on administrative bargaining and intrabureaucratic bargaining. Earlier work has emphasized inter- and intrabureaucracy 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). Yet there are differences depending on the policy area. In Lampton’s(1987) case studies of water projects, the bargaining process was more consultative than what occurs in the emerging cadre training market. The emphasis on fairness that Lampton finds is not, in this area of political activity, a component of the competition for training contracts. Party schools with marketable resources to draw upon, such as model projects, more robust local economies, or access to better teaching talent, are often able to edge out less attractive schools for outside-the-plan training contracts. In short, certain key aspects of markets now structure organizational behavior in cadre training. Since the early 1980s, various waves of training providers have increased the number and types of options available to consumers of training services. Supply has increased in diversity and volume. This process, in turn, entails bargaining over prices rather than the fixed rates once dictated by planning authorities. There is also some freedom of market exit and entry: non-party school training providers may participate in China’s cadre training market in accordance with shifts in market demand and supply, contributing to a process that is both more dynamic and demanding than what existed before. Shifts in party school market share 117

Shifts in party school market share In the face of such domestic, intraparty, and global competition, party school market share of cadre training has decreased. An examination of the allocation of training contracts in one locale, City Z in coastal Province A, illustrates the impact of competition on one party school’s hold over the local training market (Figure 4.2). The pie chart summarizes the 2008 allocation of training classes in City Z. While the city’s party school still retains a majority share of local training contracts, it has had to cede ground to various competitors that have emerged during the reform period. 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 and target enrollment. In 2008, the city’s party committee approved funding for 40 training classes, which would include 3,785 local

Other training centers 16%

University abroad 5%

Chinese university 16% City party school 61%

Central party school 2%

Figure 4.2 Share of total training, by school type Notes: Training share calculated as a percentage of total trainees planned for 2008. “Other training centers” includes: Shenzhen Socialism Institute, the City Z Socialism Institute, Shanghai Socialism Institute, City Z police academy, the cadre school at the provincial construction ministry, and the Central Discipline and Inspection Training Center. 118 Fusing party and market cadres. Classes ranged from large courses on the recently convened Seventeenth Party Congress to more general “advanced” and Mid- Career 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.79 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 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 the awarding of training contracts; the city party school was assigned both general and more specialized classes (e.g., a public security section chief training class and 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,” while 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 D. Data on the party school’s share of training during the pre-reform period is not available. It is notable that 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 captures enrollment levels, by year, based 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. 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

79 Partners included the Central Party School (1 class), a provincial university (2 classes), and the city personnel training and testing center (1 class). Shifts in party school market share 119 25,000 20,000 15,000 Trainee-months 10,000 5,000 0

1975 1980 1985 1990 1995 2000 Ye a r

Figure 4.3 Central Party School trainee volume, 1977–2000 Source: CPS Yearbooks, 1985–2001. 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 volume. 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 2007.80 For its part, the Yan’an CELA reported training over 7,200 cadres from its opening in 2005 until 2007 and over 13,400 from the end of 2008 until mid-2011.81 This translates to an annual average of nearly 2,500 cadre-trainees at each school (and over 5,300 per year for the 2008–11 period at the Yan’an CELA), which exceeds the CPS annual average of 1,509 trainees for the period 1990–2000.

80 On the Jinggangshan total, see “China’s Jinggangshan Cadre Academy Trains Nearly 10,000,” People’s Daily, September 27, 2007. 81 On the Yan’an numbers for 2005–07, see “China’s Yan’an Cadre Academy Reignites Fire in a Sacred Revolutionary Place,” China’s Personnel, September 7, 2007; for the 2008–11 numbers, see “Class Set-up” (banci shezhi)atwww.celay.org.cn/index.php? id=5, accessed August 2, 2011. 120 Fusing party and market

Market gains and losses These reforms have 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. All renovations were supported in part by revenues generated through local party school ventures. A subset of party schools has been more vulnerable to competitive pressures. Given the localization of finance, schools at the lowest administrative levels often languish from lack of resources.82 One party school leader laid out his reasoning for some schools’ lagging performance: Some party schools are not working out (bu xing). There are two types of places where party schools have done well: where the economic situation is good and where local leaders are strong. When the economy is very developed, change is fast, and people realize they need training. And in places where government is strong, the party school has a high status. Shandong is a good example of this. It’s not like Guangdong, where society is strong and government is weak.83 In other interviews, party school leaders acknowledged the potential risk of a grassroots crisis in the party school system.84 One township-level official noted, “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.”85 Higher-level party schools are aware of a potential crisis in grassroots training programs, but as a general rule, schools at different 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.

Conclusion The process of administrative reform calls for “organizing attention” around the need for change (Laegreid and Roness 1999). To generate

82 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, 2009). 83 Interview 195, district-level party school vice-principal, May 2008. 84 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. 85 Interview 180, former township vice-party secretary, April 2008. Conclusion 121

1979 1985 1987 2002

Economic Management Chinese Administration Cadre Leadership Colleges Universities Institutes Academies

Nationwide network of By central decree, At 13th Party Congress, At 16th Party Congress, cadre colleges created to allowed to dedicate up to nationwide network of three key-point academies increase system capacity 10 percent of annual schools created to train founded to address the to carry out economic enrollment to cadre government civil servants revolutionary spirit and reforms training global outlook of cadres

Figure 4.4 Timeline of domestic training providers momentum behind the updating of certain political institutions, China’s leaders saw an opportunity in the integration of market incentives and party organizations. This dovetailed with the push for broad-based modernizing reforms in the PRC, which provided the impetus for more targeted reforms in CCP personnel management. The processes detailed in this chapter have demonstrated that market forces directly shape the conditions and decisions facing party school leaders. School administrators must now adjust to a certain degree of financial insecurity and pressure to raise income beyond official budgetary allocations. Party schools throughout China have had to respond to new market pressures: tighter budgets, competition from various training providers, and negotiated training contracts. This process of marketizing cadre training accompanied China’s broader reform process and began when the COD welcomed universities and other organizations to train cadres (Figure 4.4). 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 resources to these various training providers. There is 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. 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. In the current period the provision of cadre training has fragmented, creating a “many-headed” (duo tou) training landscape. A district-level 122 Fusing party and market party school leader lamented these changes: “There are too many bodies organizing trainings. Comprehensive training planning is lacking. This many-headed training is too messy.”86 Messy or not, this is the reality facing party schools today. Attempts by Xi Jinping’s administration to curb some international programs and domestic degree programs can do little to moderate the many organizational choices that have emerged over decades. 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. Party schools now exist in a competitive institutional environment. In response to various market-based incentives, school leaders must 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).

86 Interview 195, May 2008. 5 The entrepreneurial party school Party school responses to reforms

In 2006, when I first visited the county-level party school in a coastal province’s County Y,1 the school campus exuded all the allure of socialist-style concrete drab: gray dormitory buildings, cold concrete- floored hallways, and 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, and there were new projectors in the classrooms, new air-conditioning units, and computers for all the schoolteachers.2 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 on the school campus. 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 wei zhu), along with a center for retired cadres and a training center for the people’s militia (minbing xunlian jidi).”3 He

1 This site was not located in Province A. In the province where County Y is located, I visited schools from the county levels down. Province-wide, I was unable to conduct research vertically through the party school system, from the provincial to township levels, and hence field notes from this province are not as comprehensive as those for Provinces A and B. 2 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. 3 Interview 232, July 2008.

123 124 The entrepreneurial party school 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 competition in cadre training 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” sweeping across the country modified the internal processes and external environment within which schools operated. Party school leaders confronted new challenges, including greater volatility in funding matters and competition for training contracts. This chapter will detail how, in response to these developments, schools have instituted a variety of organizational changes. 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. In managing school finances, all variety of school employees – principals, faculty, and administrators – have 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. The results of various strategies are mixed. All of these activities served, in multiple ways, to increase the likelihood of party school survival in an increasingly crowded arena. In some ventures, school leaders attempted to position party schools within the training market as globally connected but politically authoritative training centers. As part of this reorientation, they also adjusted the content of what was actually taught to cadres in an effort to remain at the forefront of the drive to reshape China’s administrative and political leaders. Examining these changes reveals an organizational adaptiveness and dynamism that is not immediately apparent when one focuses on the seemingly static nature of party schools’ formal position within CCP organization. Officially, party schools continue to reside where they have always resided in the CCP’s bureaucratic structure, but to stop there would overlook significant organizational responses to new Staying afloat through alternative income streams 125 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. As Mertha (2006) notes, “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” (p. 301). While other training systems had to begin from scratch, party schools had the advantage of decades of experience to draw from. The party school system is not unique, however, in its responses to new incentives. Schools’ adjustments to market reforms have parallels in another domain of party control, the media. This suggests how political organizations of the party have moved away from the old command-andfive-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 embracing the market, party authorities and local party entrepreneurs have potentially hit upon a strategy that renews the relevance and vitality of communist party organizations created nearly a century ago.

Staying afloat through alternative income streams Due to changes in budgetary allocations, financial shortfalls have become a reality for many party schools. Pressures to “self-raise” funds have prompted the search for income-generating ventures. 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).”4 Party schools from the county level up are obliged to develop some entrepreneurial spirit, with the success of any given scheme dependent on a range of factors, from local economic conditions to the perceived prestige of a given party school. 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. Officially, this school is a county-level unit. Along with its counterparts in other provinces, it is responsible for training 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

4 Interview 207, Province A capital party school teacher, May 2008. 126 The entrepreneurial party school

Table 5.1 2001 Income and Expense Report, county-level party school in Province B

(in yuan)

Income from government transfers 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 Rental income 380,000 Total self-raised income 1,230,000 Net balance 20,000

Source: Internal school accounting document. the provincial finance department to the school were reduced significantly beginning in the 1990s. This was due to a policy shift that required work units sending trainees to cover a greater share of the training and living expenses for cadre-students.5 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 five classes for 320 students by 2007.6 With such a decline in enrollment, school leaders sought to plug up budget deficits through alternative means. The 2001 school budget is provided in Table 5.1. In 2001, the provincial finance bureau transferred 1.13 million yuan to the school, enough to cover approximately 44 percent of total expenses. The finance bureau added two additional supplements, totaling 220,000

5 Interview 180, Provincial CYL school vice-principal, April 2008. He asserted that after this 1991 policy shift, work units were reluctant to send employees to training programs at his school. 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 boarding charges. 6 Interview 167, CYL school vice-principal, April 2008. Staying aﬂoat through alternative income streams 127 yuan, to this initial transfer. Taken together, the three ﬁnance bureau transfers covered 53 percent of the school’s annual expenses.7 Three additional sources of income made up for the budget shortfall: tuition from degree programs, outside-the-plan training classes, and noneducational entrepreneurial activities.

Income from degree programs The first alternative income source was tuition collected for degree programs organized by the school. This was a significant revenue source, equivalent to 65 percent of total self-raised income and offsetting 31 percent of total expenses. 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 (da zhuan xueli), which are granted to students who may be party cadres or noncadres. In 1990, the campus also became the home of a provincial “youth political school,” an educational institution that enrolled secondary and postsecondary students in basic coursework. However, from 1996 onward, the school decided to narrow its target audience and switch its focus 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 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 coexist alongside the CYL school’s training programs, with faculty and staff engaged in both endeavors. Income from vocational programs provides a bulwark against ever- tightening government transfers.8 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

7 This is compared to a finance transfer covering one-third of total income for the Beijing city party school, reported in Pieke (2009: 129). 8 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. This particular CYL school’sofficial 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. 128 The entrepreneurial party school this movement in 1985 with the creation of distance learning degree programs that eventually involved provincial-, city-, and county-level schools. Lower-level schools joined the endeavor for a share of the profits.9 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.10 Additional benefits stemmed from party schools’ various degree programs. They expanded the reach of party school education. By 2002, CPS distance degree courses were being taught in 2,635 branches (including fen yuan, xue qu, and fudao zhan) throughout the country.11 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. There are several motivations for enrolling in a party school degree program. Many students are officials being proactive about their careers and hoping to have higher education degrees in hand when personnel authorities review dossiers for promotion candidates.12 Importantly, party school degree programs are priced competitively and have the added convenience of a national network of local sites for exam taking and evening classes. Gaining access to elite networks is a less central motivation, as distance degree programs are the favored option among officials. Graduate degrees, offered first at the Central Party School in 1985, are now options at provincial and city party schools. Enrollees include officials and members of the general public. Motivations for

9 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. 10 Interview 111, February 2008. The training curriculum for one of the Central Party School’s earliest combined training and degree classes in 1984, for example, lists course topics such as basic sciences, math, logic, writing, and language arts (CPS 1985 Yearbook: 224–25). 11 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. 12 In the 2003 China GSS data analyzed in Chapter 3, holding a party school degree was not a significant predictor of promotion. Staying afloat through alternative income streams 129 pursuing these degrees – often in fields such as CCP history and Marxism – vary widely, from career-building to knowledge generation. The previous discussion of the CYL school’s various degree offerings, in which programs and partnerships come and go, reflects the deliberate creation of opportunities for school leaders to cope with insufficiencies in government funding. School leaders can now participate in the broader national (and international) markets created for the provision of 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, all physically located on the party school’s campus, is part of the school’s official history. Each venture 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 outside of formal training plans.

“Outside-the-plan” training classes As detailed in Chapter 2, party schools now dedicate a sizable share of total training classes to courses and topics not officially prescribed in annual training plans. These outside-the-plan classes are important spaces for schools to exercise curricular autonomy and generate income. They are not degree-granting courses of study but rather auxiliary classes often organized by a school in cooperation with requesting work units. They may also originate from the school itself. School leaders can 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. Accordingly, another source of income for the CYL school consists of outside-the-plan training programs. In this case, these classes were a very small share of total self-raised income. Out of 1.23 million yuan in self- raised income, only 50,000 yuan, or 4 percent, was raised through outside-the-plan classes. This may be due to the relative uncompetitive- ness of lower-level training schools. For a CYL school, outside-the-plan training classes are in direct competition with those of other training organizations, such as provincial-level party schools and universities, and the CYL school often loses 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 130 The entrepreneurial party school programs for cadres who once were the target audience for the CYL school.13 The Central Party School has dipped into the cadre population previously reserved for lower-ranking party schools. One example is the CPS’s first county party secretary training class (xianshi wei shuji jinxiu ban) convened in 2000. Given the convention 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, 5,000 county-level party secretaries and magistrates were sent to central training schools in a five-year series of classes.14 There appear to be varying motivations for the organization of this training class. A county-level training class was one facet of efforts begun during the Jiang Zemin administration to increase the control of central party bodies over subnational units. This special training class was also a means to disseminate policy priorities bound up in building a “new socialist countryside,” a policy wind aimed at the grassroots and developed in the late 1990s.

Rental income A third and final source of income consists of rent collected from various tenants located on school grounds. In the CYL case, the school enjoyed an income stream from occupants of the school’s on-campus dorms.15 Rental income in 2001 constituted 31 percent of total self-raised income and was sufficient to cover 15 percent of total school expenses. This is a relatively straightforward and common practice in many party schools.16 In Province B, for example, the provincial party school rented out rooms in on-campus housing units to student exam takers every year.17 In the case of a county-level school, the decision to rent rooms 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.18

13 Interview 180, April 2008. 14 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 www.nanfangdaily.com.cn/zm/20070104/xw/ szxw1/200701040006.asp, accessed February 10, 2008. 15 Interview 180, April 2008. 16 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, party school professor, April 2008), a city party school (Interview 201, party school professor, May 2008), and a county party school (Interview 232, school vice- principal, July 2008). 17 Interview 66, provincial party school vice-principal, November 2007. 18 Interview 232, county party school vice-principal, July 2008. Staying afloat through alternative income streams 131

The arrangement for renting out property varies across schools. In one coastal city, the management of the party school’s hotel and restaurant halls was contracted out to a corporation.19 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.”20 This has the added benefit of giving school leaders more bargaining power when requesting budget outlays from the local finance department. Beyond renting out residential units, party schools across inland and coastal field sites were engaged in leasing out property for other uses and opening their facilities to a broad range of clients. On a field visit to a city party school in Province A, I encountered a breakfast service crowded with young people in sharp company uniforms: a local private insurance company had rented out the school dorms, cafeteria, and classrooms for a new hire orientation program.21 Another school was crowded with schoolchildren and parents on the day of my visit: one newly renovated county party school had rented out its hotel conference facilities for a local children’s chess tournament.22 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.23 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, established around 2003, was proudly announced to me by the vice-principal of the county party school as part of the school’s fund-raising efforts to “improve the teaching environment.”24 The English school for kindergarteners was established in 2003 on the property of a different county party school, a successful enterprise begun by an unemployed factory worker-turned-entrepreneur.25 One provincial party school vice- principal expressed to me that they were “lucky to be located near a national highway,” which ostensibly brought in customers looking for

19 Interview 133, April 2008. I was unable to obtain the name of this corporation and hence do not know its ownership structure. 20 Interview 133, April 2008. 21 City party school visit, May 2008. 22 County party school visit, May 2008. 23 Interview 207, city party school teacher, May 2008. 24 This initiative was noted in the vignette at the beginning of the chapter. Interview 232, county party school vice-principal, July 2008. 25 Interview 192, city party school teacher, May 2008. 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). 132 The entrepreneurial party school

Table 5.2 Local party school entrepreneurial activities

Purpose Partnership level

Update Activity Examples Income content Local National

Degree programs and Distance learning programs, XXX new schools night school, vocational programs, graduate programs “Outside-the-plan” Crisis management seminar X X training programs Renting property Dormitory rooms, shopping XX mall units, conference facilities shopping options. School property in this case was rented out to various entrepreneurs who opened storefronts on a shopping strip on campus.26 Based on these and other interviews, Table 5.2 summarizes the range of entrepreneurial activities observed on party school campuses and indicates whether they exist primarily to generate income or promote the updating of party school training course content. These entrepreneurial ventures are similar to revenue-generating measures adopted by institutions of higher education throughout China. Beginning in the early 1980s, schools sought to supplement state transfers with a variety of programs (Lo 1993). These initiatives include supplemen- tary training courses for the employees of government work units and private enterprises, arranged through methods similar to those detailed by party school officials – that is, initiated by schools (in response to perceived needs or available in-house skills) or crafted by schools in collaboration with requesting work units. Institutions of higher education have also created income-earning enterprises, such as electronics factories and consulting services, sometimes in partnership with local enterprises. The variety and scope of these ventures speak to the loosening of party and state controls over organizations once dependent on party and government transfers of funds for survival. 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. The various school ventures observed at field sites, described in interviews, and documented in official

26 Interview 66, November 2007. Staying afloat through alternative income streams 133 reports all illustrate the 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 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 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 their various partners. Income generated from ventures decreases party school dependence on government transfers as well as increases tax revenues. In the relationship between party organs and the holder of public purse strings, all parties benefit. Matters are more complicated with respect to central party authorities. 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, in 2007, to end party schools’ distance degree programs.27 This policy change, emanating from the Central Committee, was de facto rather than publicly announced in official documents.28 The cue comes in part from the omission of “correspondence degree programs” in official party

27 This move is not necessarily because income from these degree programs was 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). 28 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, October 9, 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 web page, www.news.haedu.cn/GNYW/633609825171310000.html,accessed July 19, 2009. 134 The entrepreneurial party school school work regulations.29 CPS officials recently declared that they would cease to accept new correspondence degree applicants for undergraduate degree programs beginning in 2008.30 While the Central Party School took the lead in implementing this policy, it was 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. 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 cadres. As one party school teacher pointed out, degree programs are not central to party schools’ organizational purpose.31 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 the sale of degrees for profit.32 One party school leader admitted that “degree programs have a moneymaking flavor (zhuan qian de weidao)” to them.33 The real reasons for the cancellation lie 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 CCP Central Committee has the wherewithal to interfere in party schools’ profit-generating ventures if it perceives a threat to the reputation of the party school system and, by implication, the party itself.

29 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 www.news.xinhuanet.com/newscenter/2008–10/29/content_10275191.htm,accessed July 19, 2009. 30 Zhu Hongjun, “Central Party School Calls for Stopping Correspondence Degrees,” Southern Weekend, November 29, 2007. Interviews 238 and 239, county party school vice-principal and township organization department ofﬁcial, November 2006. 31 Interview 17, central party school teacher, November 2006. 32 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 web page conﬁrms that “there were some problems associated with the school’s management of its independent degree program” (www.dx.hainan.gov.cn/html/intro.asp, accessed June 2, 2007). See also Chapter 3 for the case of Huang Zhijia’s lawsuits against the Hubei and central party schools for making fraudulent claims about their correspondence degrees. 33 Interview 195, district-level party school vice-principal, May 2008. Coping with competition in the training market 135

In many ways, the developments detailed here mirror the “state entrepreneurialism” (Duckett 2001)and“bureau-contracting” (Ang 2009)thathas 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 extrabudgetary revenue. However, there are some key differences in reforms 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 services in exchange for some share of the revenue. 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. While 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, party school ventures are not “semi-legitimate” (Duckett 2001) andareinsteadpartofofficial 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, 1998b, 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 examined in this study did not take out loans to finance new ventures, nor was there a transfer of resources either between schools or with local agencies to support new entrepreneurial ventures.

Coping with competition in the training market Beyond fund-raising activities, party schools have responded in several ways to the emergence of organizational competition for “training market 136 The entrepreneurial party school 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. Regular schools lack the capacity to instruct the special breed of students sent to party schools, one argument goes. Second, based on their long history as “insider” party institutions, party schools have also extended training programs to new non-party audiences. They have sought to augment their program offerings through partnerships with secular 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.

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 for partybuilding and the study of party 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.”34 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 central authorities will always emphasize cadre training in party schools ... There will be a division of labor based on market forces (shichang fenniu).”35 A county-level organization department official concurred, “You can contract out training in public management, finance, etc. to universities, but there is some core work that party schools will always have. This [core content] is political theory. But there are lots of institutions to contract out to. The party school specialty is theory, but the pressure facing them has increased.”36 There is also the argument that party schools are equipped to train a certain kind of student. As a Guangdong provincial party school vice-

34 Introduction to the CPS, www.ccps.gov.cn/dxjj/index.jsp, accessed June 29, 2008. 35 Interview 211, May 2008. 36 Interview 206, cadre training department head, May 2008. Coping with competition in the training market 137 principal asserted, “Universities face young people without much experience, [while] party schools’ students probably understand more from having faced real problems, they have a richer mastery of policies – students have implemented policies, [thus] how can university teachers exceed their students?”37 University teachers, by this logic, are incapable of offering adequate instruction to student-practitioners. Certain routine training programs, such as those following a party congress, continue to be organized and staged by the party school system. In the wake of 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 each administrative level. Party school teachers were expected to hold study sessions for cadres and other party school teachers below them. Party schools’ reputation as bastions of party authority also holds new and 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.38 Entrepreneurs who gain entry to CPS training programs mingle with ofﬁcials, sometimes with the hope of securing public contracts or making connections that may pay off in the future (McGregor 2010). Local schools are also organizing training programs for private enterprises interested in establishing their own party committees or strengthening party ties.39 Bringing private entrepreneurs and managers into party schools fulﬁlls the mandate in the Three Represents for the party to absorb the “most productive”–code for entrepreneurial – segments of society.40 Importantly, targeting the private sector is one way for the party “to ensure that the private sector remains committed to the party, and short courses at the party schools are seen as one means of cultivating support” (Dickson 2008: 44). Within this mutually reinforcing relationship, party schools are perceived by key

37 Quoted in Southern Weekend (nanfang zhoumou), “Rely on Ofﬁcial Documents to Protect You, Not the Road Taken by a Provincial Party School” (kao hongtou wenjian bao ni, bushi zhen benshi yige shengji dangxiao de ‘zaizao’ zhilu), November 4, 2010. 38 Private entrepreneurs are ﬂocking to the CPS for six-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 (in the mid-2000s) is 6,800 yuan. See He Huifeng, “Taking Care of Business at Central Party School,” South China Morning Post, April 24, 2006. One of the earliest CPS training classes for entrepreneurs took place in April 2000, for 70 businesspeople from Wenzhou (Dickson 2008: 43). This was followed in 2001 by a training class for a group of entrepreneurs from Shanghai. See Li Weiping, “Shanghai Entrepreneurs Attend Study Program at Central Party School in Beijing,” Hong Kong Wen Wei Po, January 10, 2003. 39 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. 40 The call was for “the Party [to] represent the developmental needs of China’s advanced productive capacity.” 138 The entrepreneurial party school groups in society as the organizations where one can gain exposure to and a foothold on the party’s inner workings.

Forging partnerships Reputation may not be enough to attract students, and party schools have also strengthened their market position through partnerships. Schools have embraced programs in which they are not the sole issuer of degrees and certificates. Partnership-based degree programs have become common, for example, between two or more party schools or between party schools and local universities, to generate slush funds (xiao jin ku). Partnerships do not necessarily entail degree offerings, however. One county-level party school in Province A was a study site for adult education classes offered by a city-level university. The party school received an income stream for providing this service, which was not sent along to the local finance bureau and instead remained in the school’s coffers.41 Early in the reform period, when the Central Organization 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 captures the variety of domestic partnerships established between provincial-level party schools and universities. It is likely that only those partnerships that might elevate the status of a party school are broadcast to the general public, for example, 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. Three decades ago, when the training landscape began shifting, the CPS was the only party school with international connections (Liu 1989: 110–14).42 The CPS took a leading role early on with a research partnership, initiated

41 Interview 192, county-level party school professor May 2008. This interviewee noted that enrollment had been dwindling; in 2008, only about 40 students were attending the city university’s adult education classes at the party school. 42 Chinese leaders were already seeking partnerships with Western management experts and academics by 1981, when a Sino-US partnership was forged at the National Center for Industrial Science and Technology Management Development at Dalian, Liaoning. Faculty from US institutions such as the University of California, Berkeley, and the Massachusetts Institute of Technology introduced state managers to Western management techniques in six-month training classes. After the initial three-year partnership period, the program boasted 700 graduates. Students were required to have English language skills, be less than 50 years of age, and have factory management experience. Coping with competition in the training market 139

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 NA Beijing Foreign Language University School of Foreign Diplomacy Beijing Capital Normal School Chongqing Peking University 2009 Beijing Foreign Language University School of Foreign Diplomacy Beijing Capital Normal School Fujian Fujian Normal University 2008 Hubei Zhongnan Economics and Finance School 1993 Hubei University Chinese University of Geosciences, Wuhan 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 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. in 1995, between its Center for International Strategic Research and Harvard’s Fairbank Center for East Asian Research (Kirby et al. 2005: ix–x). In other partnerships, the CPS received a delegation in December

Pedagogies included case study and simulations. An example of the latter required teams of five students to decide how to allocate resources within a firm, at the risk of bankruptcy. Given the early reform climate in which these classes were organized, reports of the partnership were quick to point out that “profits, which American texts tell of as the sole goal and reason for being of [sic] an enterprise, will never play the same role in Chinese enterprises ...Chinese managers also have a lot to learn about cutting costs and raising productivity, but the common Western method of firing workers to cut costs will never be an option in socialist China” (Fan 1984: 60). A student asserted, “Of course I didn’t learn how to be a capitalist” (Ibid., p. 60). 140 The entrepreneurial party school

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.43 The purpose was to educate party school scholars on democratic institutions in the USA, which in turn would enhance teachers’ ability to teach the structure of foreign political systems to cadres. The rationale for such 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.”44 Local schools have followed the example set by the CPS. The Wuhan city party school lists on its web page that it has exchange programs with the Moscow National University, Far East Research Institute at the Russian Academy of Social Sciences, Hanoi Institute of Politics, and Lyon Institute of Public Administration.45 Some schools will also channel partnerships through their administration institutes. To give a sense of the scope and volume of these activities, Table 5.4 lists the partnerships and exchange programs that the Shanghai city party school and administration school 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 central and provincial-level party schools is available in Appendix G. 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 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).Ingeneral, 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 theentireworld’s best cases. As Deng said, it doesn’tmatterifit’sa

43 Interview 42, CPS professor. 44 Interview 34, retired central-level cadre, October 2007. 45 Wuhan party school introduction, available at www.whdx.gov.cn/xyjj.asp, accessed June 1, 2009. Coping with competition in the training market 141

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 at least 2006 Administrative Sciences Speyer Kennedy School of Harvard at least 2007 University Maxwell School of Syracuse at least 2000 University Nagoya University at least 2008 Sciences Po University of Alberta at least 2004 University of Georgia at least 2004 Shanghai administration Civil Service College in Singapore at least 2004 institute 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 Potsdam University at least 2004 Sciences Po at least 2004 University of Alberta at least 2004 University of Georgia

Source: Author’s dataset.

black cat or a white cat, so long as the knowledge and models we adopt are the best.”46 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 inter- nationally oriented universities and organization department schools, partnerships are instrumental for lending additional cache and depth to the curriculum. Furthermore, these partnerships increase the number and types of institutions involved in cadre training, which serve the

46 Interview 184, CELA director, April 2008. 142 The entrepreneurial party school party’s interest in creating new content and expanded problem-solving capacity.

Outmaneuvering competition In a less constructive vein, party schools have also dealt with competitors through preemptive maneuvers. At least 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 “ideological training wouldn’t be preserved at the university.”47 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].”48 Such a gambit may limit the efforts of universities, which are not party organs, but this strategy has limited leverage with intraparty competition. 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.”49 In summary, party schools have pursued several strategies to cope with new competitive pressures. 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. Yet the decision to draw on market mechanisms to create new incentives for political organizations has not been unique to the party school system. There has been a similar marketization of China’s media, which has generated a combination of dynamic, market–oriented, and appropriate political behavior across media providers.

A comparison within the party: creating a commercialized media Over the past two decades, the Chinese media has adjusted to the introduction of market-based incentives to produce new content and ﬁnd new

47 Interview 211, provincial party school teacher, May 2008. 48 Ibid. 49 Interview 212, provincial capital organization department cadre, May 2008. Creating a commercialized media 143 sources of non-state revenue. As with party school reforms, these changes have been top-down in nature but have also imposed self-regulating market discipline on key political actors. While media providers in China were once dominated by the economic logic of the centrally planned economy, they must now heed market demand and public appetites for more sophisticated media content. There are many similarities across the two cases, such as more entrepreneurial behavior, increased responsiveness to end users, and updating of content, but there are also key differences. Party controls over the media are broader and more severe. Relatively lax controls over party school activity exist as part of a deal with party school authorities: so long as these schools fulﬁll baseline party-mandated responsibilities, they are free to experiment with market opportunities. In this sense, a dual-track system – in which party and market channels coexist – simultaneously constrains and offers space for innovation.50 Media providers are almost entirely dependent on the market channel for income and are less subject to the discipline implied by plan quotas. Higher-level propaganda authorities thus invest in a greater range of tools for curbing the release of sensitive information and opinions. The unfunded and uncertain nature of political controls over China’s marketized media places providers in a more precarious position within the universe of party organizations. Party schools, on the other hand, have more space in which to experiment, though the nature of those experiments may have less potential to threaten system stability.

From party subsidies to commercialization Since 1978, the propaganda mouthpieces of the party – that is, the media – have been buffeted by a variety of forces, including commercialization, privatization of funding sources, and new information technologies. Party insiders and the Chinese public have stepped up pressures to reform this highly guarded arena of party rule. Beginning in the late 1970s, party authorities began commercializing media outlets, driving these

50 One of the key components of China’s economic transition strategy was the deployment of a dual-track system. This entailed the coexistence of plan and market whereby the production plans of the command economy would persist alongside a market channel in determining allocation decisions. This dual-track system included two channels: (1) a planned economy channel, which imposed quotas on producers to sell a specified quantity of a given output at state prices, and (2) a market channel for output generated in excess of planned quotas to be sold at (higher) market prices. As Qian Yingyi (2003) noted about the dual-track system and other “transitional institutions,” the dual track was a way to ease into a price system based on market prices and protect those who would be disadvantaged during the transition. This “reform without losers” was key for negotiating the market transition. Beyond economic institutions, this coexistence of plan and market incentives also pervades certain political bureaucracies of the party. 144 The entrepreneurial party school organizations to seek new and alternative sources of funding.51 The shift toward advertising revenues and away from government subsidies was underway by April 1979, when the People’sDailybegan publishing adver- tisements in its pages (Li 2001c: 3). In broadcast media, radio stations were the first targets of commercial reforms, followed by television stations. During these early reforms, the first hints of liberalization of content were discernible in central declarations such as the CCTV director’sannounce- ment that local radio and television outlets should “walk on their own two feet” and devise news content that went beyond repeating central news stories (Lynch 1999: 31). When Deng Xiaoping embarked on his famous Southern Tour in 1992, that year was also notable for several major reforms in the propaganda system. At the 1992 National Working Conference on Press Management, leaders announced a series of changes to mass communications on China. Not only would local and national papers cease receiving subsidies, they would begin competing for advertising and other sources of funding. By 1994, only major central outlets such as People’s Daily, Economic Daily, Red Flag, and Seeking Truth would continue to receive subsidies (Chan 1993: 4).52 Across different media, there is variation in levels of state subsidies.53 After the initial phase of commercialization in the 1990s and deepening in the post–World Trade Organization accession period, there is now pushback against these commercial pressures, with some local outlets opting to experiment with noncommercial funding schemes. In 2011 Chongqing Satellite TV was converted to a public interest (gongyi) station and ceased relying on advertising revenues (Lu 2011).54

51 I refer to commercialization here as a subset of activities within the process of market transition. Daniel Lynch deﬁnes media commercialization as the “increasing tendency of mass media to create products that appeal to the assumed tastes of target audiences in society, as opposed to the tastes of the central party-state‘s propaganda cadres” (p. 6). Commercialization is thus greater responsiveness to broader audiences, with one objective to produce higher revenues. In this sense, simple market principles of supply in response to demand are operative, though here there is also a shift away from state- centered and state-generated demand and a reorientation toward societal demands. 52 Seeking Truth is a publication of the Central Party School. 53 Lynch observes that laggards in radio and some central-level papers continue to receive subsidies (pp. 72–5). Zhao (1998) further notes that broadcast stations generally receive some government funding, though at low levels. CCTV remains under the direct control of the Central Propaganda Department. Even this powerful media player has become dependent on commercially generated revenues: by 2001 government funding accounted for only 0.5 percent of CCTV revenues (30 million out of 5.5 billion yuan). 54 Part of the shortfall would be made up by government subsidies (150 million yuan), which would only make up for about half of the lost advertising revenues. This move is part of a larger Chongqing-based initiative to transform the role of the state and market in local governance. In this “Chongqing model,” there is particular reliance on government- owned corporations to plow proﬁts into public services. See Huang (2011). Creating a commercialized media 145

Similar to the party school system, the decision to commercialize media was part of Deng Xiaoping’s broader modernization initiative. Central leaders saw commercialization as walking down the reform path toward economic modernization. Rather than framing advertising as a problematic practice of capitalist society, the resumption of this practice in 1979 was meant to update and heighten the appeal of media outputs. Underlying this was a critique by post-Mao propaganda theorists of the “simplistic, saturation-style propaganda of the Mao era” and greater borrowing from Western mass communications practices (Brady 2008:71–3). Financial motivations and the decentralization of fiscal authority also prompted the decision to commercialize. As detailed in Chapter 4, decentralization of fiscal authority to subnational governments generated harder budget constraints for all party organs, including party schools and media providers. With the deepening of Deng-era reforms, these units could no longer be reliant on subsidies issued by higher-level authorities.55 This also implied a loosening of the top-down control mechanisms that accompanied financial dependence on the state. Second, reductions in central revenues throughout the 1980s and early 1990s led to funding cuts throughout the system. The “shrinking purse” of the state was felt across different media providers; during the 1980s, government funding fell from full subsidies to between 50 and 70 percent of operating expenses for state television channels and only 9 percent of technological upgrades in state newspapers (Zhao 1998: 53). Such system-wide shifts in resource allocation called for greater fiscal discipline on the part of local party organs but also created conditions for adaptive organization-level responses. In this new environment, decentralization generated incentives for local organizations to come up with creative responses to tight budget constraints: media outlets, like party schools, became more responsive to the needs of end users and created new content in order to attract broader audiences, which could then be par- layed into sales and advertising revenues. While some organizations merely sought to stay afloat in a world of harder budget constraints, others sought to expand their portfolios of revenue-generating activities, given the possibility of retained surplus revenues. Government shortfalls in funding to state media outlets were also due to broader changes in Chinese society. As Chinese citizens grew more prosperous during the reform era, their demand for mass media increased. The 3.4 million television sets in China in 1978 had jumped

55 With the exception of a brief experiment with advertising in the 1950s, Chinese media outlets were completely state subsidized until 1978. See Zhao 1998: 52. 146 The entrepreneurial party school to 230 million by 1992 (Ibid.). A ballooning national viewership spurred the creation of more channels and programming options, and media providers moved to expand supply to meet this growing public demand. This expansion of demand compares with the situation facing party schools, which faced increased demand for their services due to the post-Mao initiative to boost the professional credentials of cadres. However, in the case of the party schools, this was a party-led, rather than a society-driven, process. As in the party school system, media providers searched for revenue through two types of activities, those consistent with their core mission and nonrelated commercial endeavors. Advertising presented one early revenue source. At the time of Deng’s political takeover in 1978, fewer than 20 advertising work units existed in China, but this climbed quickly to over 43,000 by the end of 1994 (Lynch 1999:55).The ideological strictures on advertising as a vehicle for capitalist exploita- tion loosened, and advertising was reframed as a neutral tool for revenue generation, subjecting ﬁrms to market discipline.56 Media outlets have also explored different ways to drive up audience numbers to raise revenues. New media content ranges from softer human interest pieces to more hard-hitting investigative journalism. Many newspapers have introduced expanded weekend lifestyle sections, and investigative journalists are carving out a space for critical reporting in China (Bandurski and Hala 2010). While advertising and more audience-friendly programming have been two changes adopted in the pursuit of revenues, media outlets have also engaged in income-generating ventures less obviously linked to mass communications. Some have expanded into industries as diverse as meat processing and energy.57 These pursuits are variations on a familiar theme: that of entrepreneurial behavior pursued by party and state actors once restrictions on market activity are loosened. As noted previously, it is commonplace for party schools to invest in noneducational business

56 There is further variation in the kinds of advertising revenue that media outlets have pursued. Beyond straightforward product or service ads, there is advertising through “paid news,” where reporters are paid to publish favorable reports about some item. Another category is “soft advertising,” where news outlets, rather than reporters, are paid to generate content supporting a particular good or service. Attempts by central party officials to crack down on “paid news,” which involves bribing reporters, have largely failed. See Lynch 1999:61–72. 57 Chan (1993) notes that “by 1992, the Beijing Movie Studio had already opened 22 subsidiaries engaged in a wide range of businesses, including gas retailing, meat processing and video production” (p. 7). He also reports that People’s Daily has expanded into real estate and data processing. In the decades since, it is likely that media conglomerates have reduced these nonmedia investments, given the profitability of media activities. I thank Jonathan Hassid for pointing this out. Creating a commercialized media 147 pursuits, ranging from rental properties to tourism services. During the early boom years of the reform period, the People’s Liberation Army also branched into commercial enterprises such as shipping and steel production.58 In a famous corruption case in the late 1990s, party authorities cracked down on a lucrative smuggling ring operated by the PLA and ordered, with limited success, the gradual divestiture of the PLA from business management.59 Beyond the search for revenue, other processes have accompanied the marketization of the media. As in the party school system, there is international competition. WTO accession in 2001 brought foreign players to China’s mass communications market. Some post-WTO changes have been gradual. In advertising, for example, foreign firms were not initially allowed majority shares in joint venture advertising companies. Wholly foreign-owned firms were phased in by 2005. Despite the gradual nature of these changes, China’s entry into the WTO initiated irreversible changes, ushering in joint ventures and foreign providers in media services ranging from advertising to film distribution.60 The combination of these global forces, the emergence of private revenue sources within China, and the creation of large media conglomerates have led to greater diversity in media outlets and content.61 Citizens can tune in to central or international media providers for news about business, economics, and world affairs, while a vast selection of global and domestic providers offer media content which “amuse[s] the masses and mediate[s] their

58 See Cheung (2001). Shambaugh notes that “PLA Inc.” included over 15,000 companies during the peak of military commercialization in the 1990s (Shambaugh 2003: 200). 59 See Yang (2004), Chapter 4, for a detailed account of the rise of smuggling practices within the PLA and the flawed implementation of the 1998 divestiture order. 60 WTO agreements did not establish a date for liberalization of television or radio markets, however. Guo (2003) notes that officially, neither foreign nor private capital was allowed in radio and television “stations, channels, or frequencies,” but regulations have allowed for private and foreign capital in “television production, the film industry, broadcasting and press distribution, news websites, [and] broadcasting networks” (p. 14–15). Since 2002, some global satellite channels (such as Star TV and Time-Warner) have been allowed to operate in southern China. See also Hu (2003: 23), for a table of legal reforms in media industries stipulated in China’s WTO accession documents. 61 By the late 1990s, a process of media conglomeration had begun, complementing reform- era changes to the funding structure of media outlets. At present, the largest of these conglomerates, CCTV, boasts over 22 channels and a growing international presence through various foreign language outlets. Following CCTV in size are other regional media conglomerates such as the Shanghai Media and Entertainment Group, Hunan Satellite TV, Jiangsu Satellite TV, and Phoenix TV, which round out the top five conglomerates. For a fuller listing, see the USC Annenberg School for Communication and Journalism posting on media in China (available online at http://ascportfolios.org/ chinaandmedia/2011/03/08/broadcast-and-online-media-in-china, accessed October 21, 2011). Hu (2003) notes that the creation of media conglomerates in China differed from processes observed in Western countries because, unsurprisingly, in China the process was led by the government. 148 The entrepreneurial party school multifaceted lived experiences with popular entertainment” (Zhao 2008: 195). These changes, that is, greater diversity of providers and outputs and the entrance of international competition, parallel changes to the party school system. These reforms do not amount to privatization of media outlets. While some government and party bureaus will contract out the production of periodicals to private actors, these contractors have little editorial control over content.62 Furthermore, the state continues to mandate a portion of media output. Several prominent national news providers such as People’s Daily and Economic Daily are managed directly by the Central Propaganda Department.63 The party thus continues to shape outputs within Chinese media. For the bulk of media providers, the market now determines advertising revenues, readership, and other revenue- generating opportunities. Still, there is scant evidence that a liberal media has emerged in China, one in which the state is limited to enforcing rules against “libel, slander, sedition, and treason” and “circulat[ing] public service messages” (Lynch 1999: 4).64 In China’s commercialized media, the guiding hand of the state remains strong.

Similarities in market processes The presence of both plan and market has generated similar outcomes in the party school system and media. First, there is greater diversity in outputs. In commercialized media, this is evident in the explosion of new programming ranging from talent shows with national audience voting to investigative reporting on local economic and development issues.65 Similarly, party schools now pursue a wide range of activities

62 See Chan (1993) for a description of these contracting arrangements. This is most common in print media. The production of a periodical is often contracted out to management by the official publishers (and license holders). In exchange for paying a fixed sum, the management (or contractor) will print and distribute a periodical, with rights over revenue streams. 63 The CPD has appointment authority over state bureaus such as the State Administration of Radio, Film and Television (which, in turn, controls China National Radio, China Radio International, and CCTV) and the General Administration of Press and Publishing, among others. See Brady (2008: Chapter 2), for an overview of propaganda organization and controls over government agencies. 64 See Liebman (2006) for an analysis of defamation cases brought against the Chinese media. While these suits are instruments used by the party-state to restrain the media, Liebman’s findings also suggest that such litigation encourages both court protection of citizen rights and citizen rights awareness. 65 A prominent example of newer, flashier entertainment programming includes the highly popular talent show, “Super Girl,” aired by Hunan Satellite TV beginning in 2005. It was banned for three years, from 2007 to 2009, and banned again in 2011, on charges that the show ran overtime (some shows were as long as three hours) and did not promote “moral Creating a commercialized media 149 to take advantage of China’s booming market in cadre training and rise in public demand for secondary and tertiary education. Second, underlying this diversity is the greater autonomy enjoyed by these marketized actors. Editors have benefited from the relaxation of ideological controls as much as party school principals have been released from the party’s earlier emphasis on ideological training. Third, there is also a shift toward deepening ties with local economic actors; party schools rent property and facilities to local entrepreneurs and businesses, while media providers have offered content to attract the interest of local audiences and advertisers. Probing this diversity more deeply also reveals certain similarities in how these party organizations have responded to their respective markets. Three general types of market-based activities have emerged. First are those activities that are consistent with the core political work of each organization. These include expanded coverage in commercialized newspapers and the offering of auxiliary cadre training programs in party schools. Second, actors have branched out to revenue-generating activities completely unrelated to their organizational mission. As noted previously, these can include transactions in real estate, food and service industries, and other pursuits with no direct link to either educational or mass communications work. Third are murkier pursuits that leverage organizational resources but skate closer to the boundaries of illegal activity. One way for media outlets to earn income in the information marketplace, for example, is to trade positive press for “red envelopes” of cash (Chan 1993:6–7). Within the party school system, there have been complaints of schools “selling degrees” on false promises that certain degrees are sufficient credentials for promotion within party and government career tracks. While these activities benefit those with the savvy to tap into the “sea” of market opportunity, there has been an uneven distribution of gains across organizations. Actors with relatively thinner resources or weaker access to market opportunities are falling behind. In the party school system, schools with more prestigious assets – usually those of higher administrative rank or in booming coastal regions – are relatively attractive to cadre-students and the general population. Schools located in prosperous locales often enjoy more demand for their rental properties,

ethics and public safety.” Speculation has abounded that the show was cancelled by the State Administration of Radio, Film, and Television (SARFT) due to its popularity in comparison to more staid CCTV programming. In the realm of investigative reporting, local media have allied with issue-based groups, such as environmental NGOs, to expose industrial wrongdoing. Media have also reported on cadre corruption, examined the rulings of courts, and looked critically at police activity. See Baum (2008: 176–9). 150 The entrepreneurial party school degree programs, and so forth. It is the party schools in poorer regions, inland provinces, and without the cache of a high title that are falling increasingly behind. There is a similar variation in outcomes across media providers. For example, rural papers have experienced precipitous declines in readership: circulation of the Liaoning Peasant News (Liaoning nongmin ribao) dropped from a peak of 1.3 million to 100,000 by 1993 (Zhao 1998: 69). National-level rural papers have experienced similar problems. Unsurprisingly, newspapers targeting peasant readers have difﬁculty attracting advertising revenues – Internet penetration notwithstanding – and can turn to few alternative commercial ventures. In China’s contemporary media marketplace, the most lucrative ventures cater to urban consumers and their appetites for particular kinds of media content. This has the end result of exacerbating perhaps the most vexing problem facing contemporary Chinese leaders: the sharp increase in social and economic inequality between town and country.

Variation in control regimes From the perspective of party leaders, one risky development to these reforms is that of greater organization-level autonomy. The non-state revenue streams that party schools and media outlets enjoy opens up the possibility of independence from party controls. Party authorities have guarded against the risk of too much organizational autonomy by retaining and exercising the power to repeal market access. Even in China’s “decentralized authoritarian” context, central party organs hold ultimate authority to censure subparty organizations. In an effort to keep at bay the centrifugal pressures resulting from greater organizational autonomy, party authorities have put in place different monitoring and control regimes to maintain party discipline. The most notable difference across these two bureaucracies is the extensiveness of party controls over media outputs. Importantly, these media controls also have a spatial dimension, which affects the circulation of information throughout the political system. As detailed in Chapter 2, party control over the party school system is exercised in three ways: via a combination of controls over personnel, school activities, and funding. As is the case throughout the party apparatus, control over party school personnel appointments ensures loyalty to party mandates. Overlapping appointments reinforce this; a party school principal is also the vice-party secretary for a locale. Party control is also exerted through the deployment of inspection teams and imposi- tion of reporting requirements. Training plans issued at the national and Creating a commercialized media 151 local levels generate party mandates for local schools, and party-ordered activities are accompanied by funding from the local finance bureau. Market forces such as competition and hard budget constraints have increasingly shaped the supply of media content in China, but the party has retained control over various aspects of media life. In mass communications, the party employs a broader range of tools to constrain media providers (Esarey 2005, 2006). All media outlets are subordinate to a sponsoring party or government body, and the Central Propaganda Department maintains personnel control over all major state agencies charged with media work. The heads of media conglomerates, for example, are party appointees. Locally, party leaders select the managers of media organizations. In addition to personnel appointments, the party has continued to subsidize and control the content of major national media outlets. These outlets serve to signal shifts in the party line and also indicate to observers the official position on any given issue. Finally, media outlets remain subject to editorial guidelines issued by propaganda bureaus, conveyed either through written or online directives or regular meetings of editors and propaganda cadres.66 At the prepublication or prebroadcast stage, party control is exerted through internally circulated editorial regulations (also known as propaganda circulars), editorial meetings, and ad hoc communications. Even prior to this filtering process, party-issued licenses for media providers serve a gatekeeping function. Party authorities ultimately retain the authority to curtail unsatisfactory activities by party schools and media providers. In the party school system, this is evident in limitations imposed on certain income- generating activities such as degree programs. Similarly, certain news items are “off the table” for news reporting, even though the public demand for investigative reporting on sensitive subjects would arguably drive up revenues for media outlets.67 It is also not uncommon for officials to pressure the editors and reporters of critical media outlets to resign.68 These crackdowns are ex post in nature, which has the effect of

66 The New York Times published one leaked and translated version of these guidelines. See “What Chinese Censors Don’t Want You to Know,” New York Times, March 21, 2010. Since around 2007, an internal CPD website disseminates these directives. 67 See Zhou (2000) for a discussion of the various ways that party authorities have defanged investigative journalists and used “media watchdogs” in the service of the state. De facto and de jure restrictions on investigative reporting of high level ofﬁcials and bureaucratic opacity have all limited the emergence of a liberal press. 68 Hassid (2008) details this tactic in a case study of Southern Metropolis Daily (nanfang dushibao). The newspaper’s editors were asked to resign after reporting on two sensitive events. One was the 2003 murder of the migrant worker Sun Zhigang at the hands of local police and the second was the paper’s 2003 reporting on the state’s attempt to hide evidence during the SARS crisis. 152 The entrepreneurial party school generating uncertainty over whether any given market activity might be permitted or restricted by party minders. Such a “regime of uncertainty” promotes self-censorship and favors conservative decision making across party organizations (Hassid 2008).69 There are differences in the depth of party controls over these two bureaucracies. Across both systems, party authorities retain control over personnel, but media providers are subject to an additional level of punishment: the shutdown of outlets that have crossed ofﬁcial boundaries or informal rules of the game.70 Such punishment, or even the threat of such punishment, does not exist for party schools, which are more deeply entrenched in the party structure. Party authorities may not shut down party schools the way they have unsystematically closed media outlets in the past. Structural constraints are in operation here: where there is a party committee, there is a party school. In this sense, the party school system is much more rigid in organization than the media. This organizational reality limits the types of party punishment meted out to party schools. Differences in “control regimes” are also attributable to differences in the missions that motivate the two bureaucracies. While party schools contribute to “internal quality control” of party personnel, media providers are focused on “external image control” and the shaping of public perceptions of the party. In the interactions between the media and its party minders, minders must push against media actors advocating for genuine media liberalization. The party has a strong interest in keeping reporters, who are more likely to be system outsiders, from collecting and disseminating potentially subversive information throughout the system. The media and its minders are engaged in a sometimes collaborative, sometimes contentious relationship over how to depict China’s political reality to the public. Journalists inhabit a gray zone: at times they are at the forefront of building proregime public sentiment, but at other times they engage more actively in the activities of a critical civil society. Adding to the party’s suspicion, journalists are less fully vetted contributors to the party’s state-building endeavors. In contrast, party schools are given relatively free rein to explore new techniques in building the modern global managers that are at the heart of

69 While Hassid focuses on uncertainty in media censorship rules, it is possible to apply this principle to the party school system. 70 The party has shown its willingness to shut down provocative publications known for edgy or muckraking tendencies. For example, popular blogger Han Han’s literary journal, Party, was shut down after the publication of the ﬁrst issue. Another prominent example includes the closure of the Communist Youth League weekly magazine supplement, Freezing Point (Bingdian), in 2006 for controversy over historical interpretations of the Qing Dynasty. See Bo (2006). Conclusion 153 recent CCP reforms. Individuals within the party school system have weaker incentives to undermine the political system that provides them with a steady stream of beneﬁts.71 Cadre training at party schools exists to empower cadres, who are system insiders, and perpetuate the circulation of governance-related information throughout the political system. This is consistent with the physical circulation of cadres throughout the political system. Party schools are allowed to explore new ideas in a partially marketized context because their entrepreneurial ventures serve as demonstration sites – even illustrative case studies – for the cadres who study them and who study in them.

Conclusion Organizations may undergo adaptive change along multiple dimensions: in their goals, the boundaries that demarcate an organization’s reach, and the activities in which they are engaged (Aldrich 1999: Chapter 7). Party schools experienced transformation in all three of these. Schools modiﬁed their goals to move away from the revolutionary mandate of the past and toward the administrative modernization of the post-Mao period. In order to do so, they expanded organizational boundaries to forge partnerships and exchanges with a variety of party and non-party actors. This process entailed the reformulation of party schools, from closed-off enclaves of party study to more open sites of teaching and commerce. Market pressures have induced schools to adjust their activity portfolios and maximize revenues while balancing their pre-existing mandate to train the party’s political elite. These transformations were, furthermore, adaptive in the sense that they moved party schools closer to the changed context in which these organizations found themselves. In response to reforms in cadre training, party school leaders have implemented a variety of organizational changes. These include the diversiﬁcation of revenue-generating schemes and the initiation of new programs and partnerships to cope with competition. A nationwide party school degree system, begun in 1985, was the beginning of a stream of income-generating activities devised by party schools. School leaders have leveraged their particular organizational resources by providing facilities and human resources to entrepreneurs in search of everything from well-located rental property to event managers. These strategies address the various challenges presented by market competition and less stable income streams. The entrepreneurial activities detailed in

71 See Geddes (1999a) on the strong corporate incentives within parties, which may explain the relative durability of single-party authoritarian systems. 154 The entrepreneurial party school this chapter reflect a new spirit pervading party schools. It appears that nearly century-old party organizations can learn new tricks. Party organizations have made 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. Above all, party schools now have strong incentives to search actively for solutions to ever-shifting circumstances. New entrepreneurial ventures serve important functions for the school leaders who are responsible for the allocation of resources between official (training) and market activity: in maximizing revenues through new ventures, schools can offer increased benefits to employees and generate the resources for reinvestment in the school. These, in turn, enhance a party school’s competitiveness. Another related outcome of the introduction of market forces to party organizations has been an increasing integration of party schools in local economies. 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 enterprises in search of facilities, entrepreneurs seeking out party school training to build professional networks, entrepreneurs 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, and 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 wholesale redefining of the role of party schools, but rather by retaining the place of these schools within party organization and grasping new opportunities. New functions have been layered atop the old. Local embeddedness also pays off within the national training market. Party schools in model regions may serve as interlocutors for governments located elsewhere in China, which are interested in learning the secrets behind a local success story (often linked to cadre promotion incentives). So long as the local party school is able to maintain a convincing research profile or knowledge of conditions in its locale, interested parties will inquire. With this in mind, party schools often have research centers for investigating topics specific to their locale. 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 Conclusion 155 experiences on different economic development models.72 Local research conducted by schoolteachers and groups of cadre-trainees are then dis- seminated and stored in locally distributed volumes of published field reports.73 Market-based reforms have extended beyond party schools, to other areas of political importance such as the media. The incentives that now shape behavior in these arenas were initially driven by the push to modernize the party and society in the post-Mao period. Decentralization of China’s fiscal system meant a turn toward alternative market-based sources of revenue. Together, these changes reflected a search, on the part of party authorities, for incentives that would motivate party actors to engage in adaptive behavior. Economic reforms required party organizations to shift out of the rigidities of a centrally planned political and economic system,buttodosogradually.The resultant organizational dynamism has served central leaders’ desire for more modern and savvier political outputs. However, this marketization has the potential to push the boundaries of the political system in new directions, beyond the economic liberalization observed thus far. Greater organizational autonomy, weaker ideological controls, and greater end-user orientation all stand in tension with traditional top- down party authority structures. Party controls over a marketized and globally connected media have been sufficient to contain pressures for a truly free press. Furthermore, there is as yet no evidence that party schools are shifting their training content to incline cadres toward more liberal modes of governance. It appears that party control over cadre management (dang guan ganbu) and media management (dang guan meiti)74 have been maintained despitesomeliberalizingreforms across these party bureaucracies. The current outcome serves the party well on multiple fronts. Party authorities can point to promarket reforms within these bureaucracies as evidence of a liberal thaw, which has the effect of appeasing some domestic and international critics. At the same time, the party has continued to refine and utilize a variety of carefully calibrated controls to hedge against the risk of unanticipated and system-threatening organizational change.

72 In one coastal, city-level party school, an in-house research center focuses on the development of the local private sector. Interview 97, city party school teacher, Province A, December 2007. 73 When visiting one city-level party school, I was given 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). 74 See Chan (2010: 12). 156 The entrepreneurial party school

The mix of party controls and market discipline that shape outputs across party schools and media outlets indicates the party’s careful balancing act. These two sets of actors, while still deeply embedded within the political work of the party, now inhabit a middle ground between traditional state structures and the new market order. 6 Adaptation 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. As one party school teacher observed, Marketization (shichang hua) has meant that party schools need to improve the teaching quality. Party schools still have a lot of strengths, in party theory and practical knowledge, while universities have better foreign language and specialized knowledge. Party schools have a traditional status as the place for cadre training. But with more institutions offering training, we have become flexible and are adjusting. For example, we’ve shifted to more specialized (zhuanti) training classes.1 Examining the content of party school training classes reveals how party schools have “kept pace with the times”2 and, in so doing, embody the organizational adaptability that has in part defined the CCP for the past three decades. By examining longitudinal trends in party school curricula, it is possible to pinpoint how party schools reflect the changing priorities of central party leaders and corresponding shifts in curricula and pedagogies. This chapter will map changes in training content and ascertain whether they indicate organizational adaptiveness on the part of the party school system. Given the indeterminate nature of organizational change – that is, change may move an organization toward paralysis or revitalization – here the adaptiveness of organizational change will be evaluated based on two factors. First, this chapter will assess whether observed changes move the organization closer to the context in which the organization is embedded. The underlying assumption is that change that

1 Interview 143, provincial party school assistant professor, April 2008. 2 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, May 15, 2007).

157 158 Adaptation measured parallels, rather than conﬂicts with, an organization’s environment abets organizational survival. There is a temporal dimension to measuring this change, as lags are expected in the adaptive process. Two caveats apply to this approach. First is the reactive, rather than anticipatory, nature of the adaptation that such an empirical strategy will unearth. Furthermore, since this is a study of process, the endpoint is indeterminate.3 Of interest instead is whether there is movement toward convergence rather than divergence of organization with environment, with the former implying adaptive capacity. A second indicator of adaptiveness considers whether the incentives driving this organizational change are capable of motivating sustained adaptive behavior or whether these incentives contain within them the possibility for deviation from system-wide goals. This addresses the durability of a given change and whether it is system-reinforcing in nature. Organizations may make a one-time adaptive switch to accommodate environmental change, but the longer-term issue is whether those organizations have incentives to maintain adaptive behavior in the face of future change. In the complex multilayered bureaucracies that comprise the CCP, central-local discipline is particularly important. Some local deviation from central mandates is expected, but this can become system challenging if divergence is too great. To measure whether and how party schools responded to new political goals and realities, one logical place to look is in the classroom. Without minimizing the importance of non-classroom activities such as school- wide speeches by prominent party leaders and scholars, ﬁeld trips, and networking among the student-cadres, course content constitutes the most direct means to assess the changing priorities of party school decision-makers. Party school faculty have noted certain shifts over time: In the 1970s, before changes to the aims of party schools, ideology was the entire content of training. Now, about a quarter of teaching time is on ideology; a quarter on general knowledge such as public management, leadership, and law; a quarter on cadres’ world view, like the Taiwan question, democracy in other countries; and a quarter on the provincial situation.4 To test more rigorously these assertions, this chapter analyzes syllabi for classes organized at the Central Party School. Complete syllabi are available in irregularly published yearbooks, which contain the teaching plans for training programs organized over the period 1983–2000. Additional training syllabi, for 2007, were collected during site visits to the Central

3 As March (1994) notes, “there is no guarantee that convergence will have been achieved by any particular time” (p. 42). 4 Interviews 93 and 94, provincial party school professors, December 2007. Adaptation measured 159

Table 6.1 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

Party School in 2007 and 2008. Table 6.1 offers a summary of the syllabi collected, by year. Content analysis was conducted by coding, within each syllabus, each course unit (dan yuan) and subunit (usually class/lecture topics) into the following categories: partybuilding, management, national and international briefings, policy and law, and other course content. The categories used for coding syllabi are given in Appendix H. Each category was measured in terms of percentages of total training class time for a given training class. Partybuilding content encompasses the traditional work of party schools, that is, coursework dedicated to orthodox party theory such as Marxism-Leninism, Mao Zedong Thought, and the ideological contributions of subsequent party leaders. Partybuilding content also includes the study of party history, the party line, and party documents focused on the strengthening of party spirit and/or organization. Figure 6.1 maps change in this category of coursework, as a percent of total training time for a given year. The trendline suggests a rise and fall in emphasis on partybuilding course content over time, from highs of 69 and 68 percent of total training time in 1990 and 1994, respectively, to a low of 23 percent in 2007. The increase in the early 1990s may be due to heightened efforts to impose party discipline in the wake of the 1989 Tiananmen Square protests. While consistent with the professors’ assertions, this decrease in partybuilding content is somewhat surprising because party schools in many ways embody and shape the party line of the day. Students study articles written by CCP and CPS leaders 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 leadership. Yu Yunyao, vice president of the CPS, has written at length about bundling together this concept of social harmony with “scientific development,” two weighty terms in official declarations.5

5 “PRC Central Party School Vice President on Building Socialist Harmonious Society,” Guangming Ribao, May 18, 2005. 160 Adaptation measured 80 60 40 20 0 Percent of class time dedicated to partybuilding Percent 1980 1985 1990 1995 2000 2005 2010 Ye a r

Partybuilding Fitted values

Figure 6.1 Percent of CPS training class time dedicated to partybuilding content, 1983–2007

Shen Baoxiang, also a CPS professor, has contributed to the party’s ideological and developmental agenda by noting 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 building an innovation-oriented nation.”6 Declarations emanating from party schools thus serve as a bellwether for shifts in party ideology, and it would make sense for cadres to attend party schools to learn about these cutting-edge developments. This downward trend may be due to crowding out by the more secular topics (explained below) that have demanded increasing shares of training time, topics that are compatible with 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 “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,

6 Shen Baoxiang, “Reform and Opening Up Is Still the Idea that Decides China’s Destiny,” Study Times, March 20, 2006. Adaptation measured 161

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.7 Such complaints conflict with the official story that party schools and organization department training academies have a mutually supportive (huxiang buchong) relationship.8 They also hint at an emerging division of labor, where party schools remain sites for the serious study of party leaders’ theoretical development (i.e., grooming trainees for leadership in party theory), while the organization department handles matters of party esprit de corps. Disaggregating this partybuilding category, however, reveals declines in the amount of training time dedicated to party theory. This decrease in emphasis on ideological content has been uneven over time, but a look at course syllabi reveals gradual change in what is accorded emphasis within the partybuilding portion of course content. In the 1983 Central Organization Department’s eight-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” (e.g., CPS First Short-term Training Class in 1985). Central also is the study of Mao Zedong Thought, with “the Party cadre expected to know Mao’s theories in relation to general principles, especially in relation to his own particular environment” (Tang 1961: 26). By 1995, however, it is possible to detect a shift in the canon. The first section in CPS training classes directed students to the study of Deng

7 Interview 114, CPS teacher of party history, February 2008. A listing of course topics on the Jinggangshan CELA web page includes some related to Jinggangshan’s revolutionary history, but the full listing also includes topics that overlap with party school syllabi, such as the leadership psychology and national economic and social development. See www. celaj.gov.cn/html/jxsj/kcsz/index.html, accessed July 2, 2011. 8 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 www.rensb. com/showarticle.php?articleID=10758, accessed March 9, 2010. This narrative, however, is undercut by ofﬁcial documents that discuss competition and the mixed opinions given by the interviewees. 162 Adaptation measured

Xiaoping’s writings (1995 CPS Yearbook: 172–92), dropping the works of Mao, Marx, and Lenin. In the 2010–20 Cadre Education and Training Reform Plan, newer leaders’ theoretical contributions (i.e., Deng Xiaoping Theory, Jiang Zemin’s “Three Represents,” and Hu Jintao’s scientific development) are all mentioned as the guiding thought (zhidao sixiang) for party school education, with the notable omission of Mao Zedong Thought.9 Marxism is referenced in the context of the CCP as a “Marxist learning party” (makesi zhuyi xuexi xing zhengdang).10 Compared to these official plans, recent CPS syllabi have circled back and frame the general goals of training classes in terms of the long pedigree of thinkers shaping the party’s ideology today. The 2007–08 Mid-Career 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 system- atically study fundamental questions of Marxism-Leninism, Mao Zedong Thought, Deng Xiaoping Theory, ‘Three Represents’ Important Thought, and contemporary world economy, world technology, international law, global military affairs, global ideational trends, world religions, etc.”11 Figure 6.2 illustrates the share of total training time dedicated to foundational orthodox theories of the party in collected CPS training syllabi. Orthodox theory is defined as course content dedicated to Marxism-Leninism and Mao Zedong Thought. 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. 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. 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

9 2010–20 Cadre Education and Training Reform Plan (2010–20 nian ganbu jiaoyu peixun gaige gangyao), Section 2, Article 4. Mao Zedong Thought is mentioned only once in this document, in Section 5: Reform in content and methods. 10 Ibid., opening sentence. 11 Teaching plan, CPS 23rd One-year Mid-Career Cadre Training Class. Adaptation measured 163 30 20 10 Percent of class time Percent 0

1980 1985 1990 1995 2000 2005 2010 Ye a r Orthodox theories Fitted values

Figure 6.2 Percent of CPS training time dedicated to orthodox theory, 1983–2007 old subjects”; some want issues in economic development and related theories to be central.12 It may be that the decreasing emphasis on foundational thinkers has given way to studying 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 Hu Jintao (Figure 6.3). Classes spent the most time on Deng’s “reform and opening up” in the mid-1990s, but the overall amount of time devoted to studying the guiding words of top leaders has decreased from a peak of 20.6 percent in 1995 to 4.5 percent in 2007. In contrast to these trends in partybuilding and party theory, two categories of training content have gained prominence over time: management and brieﬁngs on national and international events. Increasing emphasis on management and global awareness are consistent with environmental change at three levels: with a new set of modernization goals articulated from Deng Xiaoping’s rule onward; with China’s transition from plan to market, which required a different public management skill

12 Interview 114, CPS professor of party history, February 2008.The “ﬁve old subjects” are: Marxist philosophy, Marxist economics, Marxist scientiﬁc socialism, CCP history, and partybuilding. 164 Adaptation measured 25 20 15 10 Percent of class time Percent 5 0

1988 1990 1992 1994 1996 1998 2000 2002 2004 2006 2008 Ye a r Reform-era leaders’ theories Fitted values

Figure 6.3 Percent of CPS training time dedicated to the theories of reform-era leaders, 1983–2007 regime; and with China’s reintegration with the global economy and society. Management-related content, in contrast to the other categories discussed thus far, abets these shifts. This category includes the following topics: leadership theory or “the art of leadership,” crisis management, speechmaking and communication, media relations, principles of service- based government, public administration, and strategic thinking.13 Figure 6.4 gives a summary of changes in the teaching of management, as a share of total training time in a given year. The trendline shows an increase in emphasis on these skills, reaching a high of 23 percent of total training time in 2007. The increase in 2000 was signiﬁcant also because this was 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.”14

13 While management-related content is now different, with a greater emphasis on the market economy and ideas adopted from Western administration and management schools, management was an area of training emphasis even during the Mao period. See Tang (1961: 10). 14 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,” January 4, 2007. A second on-the-job training class (renzhi peixun) for county party secretaries began in 2010, also spearheaded by the CPS. Adaptation measured 165 25 20 15 10 5 0 Percent of class time dedicated to management Percent 1980 1985 1990 1995 2000 2005 2010 Ye a r Management skills Fitted values

Figure 6.4 Percent of CPS training class time dedicated to management content, 1983–2007

Separate but related to the building of new management capacity in the CCP is creating a nationally and globally aware cadre corps. In this regard, cadre-students have spent an increasing amount of training time listening to briefings by local and national officials, studying the experiences of Asian Tigers, understanding international affairs, and viewing internally circulated films on the downfall of communist parties in Europe and the Soviet Union.15 The outward focus of training classes is evident in topics such as, “Political parties of the contemporary world and governance law” (dangdai shijie zhengdang ji zhizheng guilv)and“Research on the governance of political parties around the world during global change” (quanqiu bianju zhong de shijie zhengdang zhengzhi yanjiu), taught at provincial-level party schools from 2007 to 2011.16 Figure 6.5 provides a summary of training time dedicated

15 In 2007 training class syllabi, the Shenzhen Party School scheduled screenings of the “internal reference ﬁlm” (neibu ziliao pian) Revelations on the Violent Change in the USSR and Eastern Europe (sulian dong’ou jubian qishi lu). 16 These courses were listed in Hubei provincial party school syllabi of 2009 and 2011, for advanced training classes; Hunan provincial party school syllabi of 2010; and Yunnan provincial party school syllabi of 2007, including that year’s mid-career cadre training classes. 166 Adaptation measured 25 20 15 10 5 Percent of class time dedicated to briefings Percent 0

1980 1985 1990 1995 2000 2005 2010 Ye a r Briefings Fitted values

Figure 6.5 Percent of CPS training class time dedicated to briefings, 1983–2007 to briefings, which spans a low of 5 percent in 1985 and a high of 23 percent in 2007. The final category of training content, policy and law, includes the study of five-year economic plans, economic policies for specific sectors or regions (e.g., agriculture, western provinces), and social policies. Figure 6.6 suggests that there has been a decrease in the amount of training time dedicated to this category, from an average of 28.5 percent in the mid-1980s to 5.6 percent by 2007. One explanation for this downward trend may be a shift in reliance on other preexisting party channels for communicating policies such as the party’s documentary and meeting systems (Oksenberg 1974).

Competition and new approaches to cadre training An additional set of changes in training content, separate from subject matter, has been the introduction of new pedagogies. One of the earliest changes was allowing more discussion between students and instructors. “There are reforms in cadre training all the time. We now encourage discussion in class instead of lecture ...I’ll lecture for two hours followed by 90 minutes of discussion. It can be very lively (renao). I like this a lot, Competition and new approaches to cadre training 167 30 20 10 Percent of class time dedicated to policy Percent 0

1980 1985 1990 1995 2000 2005 2010 Ye a r Policy Fitted values

Figure 6.6 Percent of CPS training class time dedicated to policy and law content, 1983–2007 but teachers must have a higher level of education. Implementing this reform is a matter of the quality of teachers,” commented one Central Party School professor.17 A variation on this theme of greater student participation is the addition of simulations in training classes. “These are to build problemsolving skills. Simulations will be solved by small teams. For example, we have a team ﬁgure out how to deal with an environmental disaster in a major city. Or simulations can be national in scope,” described one city ofﬁcial responsible for training and education.18 In another common simulation, cadre-trainees must “meet the press” and hone their public relations skills in mock press conferences. “A recent [CPS] class learned how to deal with media relations. We invited CCTV leaders to come give a talk. Then we had mock reporters ask tough questions so cadres could practice answering,” noted a Central Party School professor.19 Cadres also study and discuss case studies of leadership culled from throughout the country. Content analysis of a combined dataset of central and local party school training syllabi (N = 247) over the period 2001–11 revealed that, on average, case studies constituted 12 percent of training

17 Interview 208, Februrary 2008. 18 Interview 89, December 2007. 19 Interview 208, February 2008. 168 Adaptation measured time. According to the 2013–17 Cadre Training Plan, case studies were to constitute 30 percent of total training time at the provincial and central levels. One of the first schools to embrace “leadership science” and the case study method, Heilongjiang’s provincial party school, compiled in the 1980s a text containing “over 180 cases from China and abroad” (Paltiel 1990: 599). 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 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 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.20 The subject matter and tone of this case study contrasts with the less ambiguous “model cadre” figures detailed in earlier teaching materials. In a case study from the Heilongjiang provincial party school’s 1987 text- book, a state-owned enterprise manager is cast in a distinctly positive frame (Paltiel 1990: 601–2). In this lesson, the manager, while lacking technical skills and higher education, is nonetheless adroit in deflecting requests by subordinates to place relatives in choice jobs. The leadership demonstrated by this individual is clear, and the takeaway message for students is to exercise similarly sound judgment in their management decisions. This contrasts with the more controversial tactics deployed by the cadre in the newer CPS case study, which students are invited to debate. These newer case studies seem to reflect a shift toward critical evaluation. They illustrate the participatory shift in training and the move away from unidirectional, doctrinal learning. Another change has been the introduction of greater choice within training classes. The so-called “menu style” (caidan shi) training allows students to choose from an array of lectures and classes for a given portion of their training. Elective lectures are a means to match students’ interests

20 Teaching case studies in the model of Western business schools requires new skills on the part of party school teachers, and teacher training courses have been initiated with schools such as Harvard University’s Kennedy School of Government. Competition and new approaches to cadre training 169 with those of the experts on hand in a party school, be they professors or practitioners. Such granting of choice is an acknowledgment that corral- ling students through the same regimen of classes may be a disservice to students’ diverse interests and professional needs. More generally, such offerings are consistent with the greater end-user orientation of contemporary training classes. In 2010, the COD and other party and government committees gave department- and bureau-level cadres the opportunity to participate in a trial elective program.21 The Yunnan Provincial Party School’s elective offerings in 2007 provide a glimpse of what topics are relatively attractive to students as well as what one school’s education department is able and willing to teach.22 The most popular lectures were those on “The art of leading cadres’ speechmaking” (lingdao ganbu yanjiang yishu), with 109 enrollees out of 222 (49 percent), and “Raising the management abilities of leading cadres in a harmonious society” (tisheng hexie shehui lingdao ganbu jiayu nengli), with 81 enrollees out of 218 (37 percent). These fared much better than those elective lectures that attracted 12 or fewer students (and were subsequently cancelled). Lectures in this unpopular category included “The development, evolution, and adjustments (in theory and policies) of European Social Democratic parties” (Ouzhou shehui minzhu- dang de fazhan yanbian yu lilun zhengce tiaozheng) and “Research on Yunnan’s population and coordinated economic and social development” (Yunnan sheng renkou yu jingji shehui xietiao fazhan yanjiu). It would appear, from this particular case, that cadres were more interested in self-improvement and less in those offerings related to local and global issues.23 Taken together, these trends reflect an interest in heightening cadre participation and tailoring training experiences to specific needs. These shifts make sense given the combination of unpredictable, rapid local development and the party’s aim of forging leaders capable of making autonomous decisions that are consistent with the center’s developmental goals. Changes in training content and teaching methods suggest that changes observed in cadre training are adaptive in nature, reflecting an

21 See the COD Institute of Party Building, n.d., “The Selection, Appointment, Education and Training of Cadres,” English-language brochure from “The ABCs of the Communist Party of China” series. 22 These elective lectures were offered for all core (zhuti) classes in the spring semester of 2007. Lectures were held every Tuesday morning. Students were to choose one out of five or six options. For a full listing of electives, see www.ynce.gov.cn/ynce/site/school/ article001.jsp?ArticleID-18126, accessed July 7, 2011. 23 The popularity of teachers may have mattered, along with the scheduling of classes, but this information was not reported. 170 Adaptation measured effort by party schools to update their focus and heed the calls for modernization and (partial) liberalization begun in 1978. One important question arises as to whether these adaptive changes are linked to the creation of a competitive market in cadre training. It is possible that these changes came about through the top-down channels still extant in the party, such as the training plans that are a holdover from traditional authority relations. But central authorities have relied on market forces, allowing diversity in training providers and drawing on forces of supply and demand to determine the prices and content of training outputs. The expectation was that these market-based pressures would be drivers of change. Conversely, bottom-up demands from cadres for party schools to update teaching content could be another pathway to change. While these feedback channels gained greater prominence in the 1990s and are a real consideration for party school leaders and teachers, student complaints were not organized enough to drive reforms across the party school system. Field interviews and official documents indicate that competition was the key mechanism. One party school teacher mused, The party school system was [at risk of being] outdated and too traditional ...and not innovative. The traditional training system was becoming insufficient. Given all these problems, the solution was to create a new institution that fit the party’s needs and create competition between party institutions.24 In response to this competition, party schools have sought to raise the skill levels of party school teachers, either by hiring instructors with more education or sending existing teachers to obtain more training.25 From the other side, one CELA administrator asserted, “We have stirred up the market here. Just like universities in the U.S., there’s competition and elimination. No one has a monopoly on education. It’s not as simple as before. Local party schools have to deal with market changes. If they want to survive, they have to change. This is just like SOEs, which have had to adjust to market competition.”26 A CELA professor struck a conciliatory note in her musings on the relationship between central-level training organizations, but she ultimately con- cluded that party schools must contend with new training providers, which are at the cutting edge of cadre training.

24 Interview 144, provincial-level party school professor, April 10, 2008. 25 Field interviews, professors at city- and county-level party schools, May 2008. See also the 2010–20 Cadre Education and Training Reform Plan, Section 6, Articles 20–22, which call for improved recruitment, training, and assessment of party school teachers. 26 Interview 184, CELA director, April 2008. Sustainability of changes in the party school system 171

With the Central Party School, there are teacher exchanges, and the Central Organization Department will invite teachers from all three schools [i.e., CPS, CELA, NSA] to conferences and meetings. Some training classes will study at both CPS and CELAP. But administration schools have copied CELAP’s press simulation exercise; CELAP was the first to do it. We can’t copy the CPS and NSA; we have to present a new model of cadre training. We have to constantly innovate as a result.27 In response to the notion that other central-level training academies are gaining prominence, a CPS professor exclaimed, “Yes, the [CELA] have been reforming and coming up with new training styles, but visiting heads of state will still come to the CPS, not a cadre executive leadership academy!”28 These sentiments, that competition prompted changes in party schools, are supported by official documents that link competition for training resources to continued innovation. Party documents refer to the creation of a market in cadre training to spur on change. Recent central-level documents have called for “competition to select the fittest” (jingzheng zeyou) as the guiding thought and a fundamental principle of reform (gaige de zhidao sixiang, jiben yuanze)andasa way for party schools to “liberate thought” and increase their creative capabilities.29

Sustainability of changes in the party school system A deeper assessment of the cadre training market that has emerged in China must grapple with the sustainability of the adaptive changes observed. To draw a conclusion on this issue is to get at the viability of the choices made by central party authorities when they embarked on this reform process. The underlying logic of competitive redundancy belies an ideational shift, one that grants market forces a greater role in the allocation of state resources. Now, schools must respond to competitors’ innovations, the needs of cadre-students, and local authorities who hold the purse strings for training contracts. The competition introduced to cadre training would seem to have generated sustainable outcomes, insofar as it has spurred adaptive change over the nearly three decades since this system has taken form.

27 Interview 187, April 2008. 28 Interview, May 2008, CPS professor. 29 Section 1, Article 3 and Section 2, Article 5 of the 2010–20 Cadre Education and Training Reform Plan, available online at http://politics.people.com.cn/GB/1026/ 12467912.html, accessed September 27, 2012. See also the 2008 Cadre education and training provisional work regulations for a discussion of the “orderly system of competition” that would characterize cadre training. 172 Adaptation measured

Table 6.2 Shanghai party school pedagogies, by percentage of total class time

Mid-career cadre training class

Type 2010 2011 Average

Lecture (jiangshou)423538 Site visit (tiyan)161917 Field investigation (diaoyan)14 18 16 Self-study (ziwo jiaoyu)105 7 Report/brieﬁng (baogao)5 7 6 Elective (xuanxiu)465 Case study (anli)534 Simulation (moni)354 Student forum (xueyuan luntan)1 2 2

Notes: Percentages may not sum to 100 due to rounding. Source: Shanghai city party school teaching plans, 2010 and 2011.

Changes observed have been systemic in nature, and the party school bureaucracy has retained a high degree of coherence in the face of decentralizing reforms. This coherence is evidenced by the overlap in course content for a class regularly organized by party schools from the county to central levels. Content analysis was conducted on 14 syllabi from 2007 and 2008, all for the highly selective and prestigious Mid-Career Cadre Training Class (zhongqing nian ganbu peixun ban). These training classes are core (zhuti), planned (jihua nei) courses that party schools organize by order of local party committees; their purpose is to identify high- performing cadres within a jurisdiction and gather together local political talent for further training and observation by party officials. Classes tend to be small, with approximately 40 students selected for classes at local party schools. Pedagogies employed in these classes are quite diverse, ranging from lectures to student forums. In the case of the Shanghai city party school, about one-third of the course content was conveyed via lecture, and another third of the course content involved field visits and investigations.30 The remaining third of class time comprised case studies, simulations, elective (“menu-style”) classes, briefings, and self- study. Table 6.2 gives a summary of types of teaching strategies employed in the 2010 and 2011 Shanghai city party school mid-career cadre training classes.

30 According to Shanghai party school teaching plans, the ratio of party school instructors to external instructors was 65:35, and this shifted slightly in favor of in-house instruction to 70:30 in 2011. Sustainability of changes in the party school system 173

Table 6.3 Description of mid-career cadre training classes, 2007–08

Central Provincial City County Overall

N (syllabi) 2 8 2 2 14 Of which: Coastal n/a 5 1 2 8 Average duration (months) 7.69 3.20 3.88 1.63 3.71 Average number of trainees n/a* 40 46 40 43

* In 2000, the most recent year for which ofﬁcial enrollment data are available, the CPS Mid-Career Cadre Training class (10.5 months long) included 243 cadre-students.

Of the 12 syllabi collected at the subnational level (provincial, city, and county) for 2007 and 2008, there is representation of training classes across central and coastal regions at the provincial and city levels. Syllabi for western provinces were unavailable for analysis. Table 6.3 gives a summary of the classes analyzed. The actual content of these classes is, as with other training classes, at the discretion of local and school party officials. In analyzing the content of classes convened in 2007 and 2008, there is consistency in training content across key curricular areas. It also appears that subnational party schools remain a channel for party authorities to build discipline within party ranks; this is clearly evident in the increasing amounts of training time dedicated to party policies and laws the further down the administrative ladder one reaches, from 8.4 percent at the CPS to 27.5 percent at county party schools. Finally, there is variation in non-critical curricular areas such as electives and vocational coursework. Figure 6.7 gives a summary of training content for 2007–08 Mid-Career Cadre Training classes from the central to county levels, offering a “core sample” of work carried out throughout the party school system. There is uniformity in the amount of training time dedicated to the party’s ideological foundations. In particular, this includes the “partybuilding” work that still figures prominently in party school training content: theoretical foundations, major political documents, and activities intended to build “party spirit” among cadres. Interestingly, there was also the study of capitalism and democracy at the central and provincial levels, which is included in this theory category.31 Six of the 14 syllabi also detailed visits to sites of historic significance such as Yan’an and Jinggangshan.

31 At the central and provincial party schools, 3.5 and 5.3 percent of training time were dedicated to the study of theories of democracy and capitalism, respectively. 174 Adaptation measured

100 90 32.12 32.65 33.75 80 34.83 70 Partybuilding 60 17.37 25.04 20.4 22.5 Management 50 0 Briefings 40 15.28 17.7 17.08 Policies 30 27.5 Other 22.58 20 8.35 20.99 Percent of training class time of training Percent 10 16.79 12.12 16.25 6.7 0 CPS Provinces Cities Counties

Figure 6.7 Mid-career cadre training class training content, 2007–08 Note: “Other” category includes electives and general interest classes in basic disciplines such as math, literature, social sciences, and the like.

The newest curricular area, management, commanded a sizable share of training content across administrative levels. Approximately one-fifth to one-fourth of training time was dedicated to developing leadership abilities. Unlike the policy realm, there is no clear trend when moving from center to county; it would appear that counties are more aggressively promoting leadership skills than the intermediate city and provincial levels. The overall finding on management content, however, is sugges- tive of the lag time required for new content to seep to lower levels of the administrative hierarchy. There is some curricular consistency in training time dedicated to briefings on the “national situation” (guoqing) and foreign affairs. This consistency held across the central, provincial, and city levels only, dropping to zero at the county party schools. One reason for this may be the relative importance of understanding national and global affairs for cadres at higher administrative levels. Another rationale may be the importance of global awareness among leading cadres, who would be trained at city- level party schools and higher. Syllabi revealed less system-wide consistency in the study of party policies. This category includes the study of official documents related to economic and social development at the local and central levels. Sustainability of changes in the party school system 175

Study of five-year plans (central and local) and specific national laws are also included in this category. The dominant trend across administrative levels is of a steady increase in the amount of training time dedicated to the study of policies generally. This suggests greater reliance on local party schools as forums for the dissemination and study of specific economic and social policies the further a cadre is from the center. To summarize, party schools from the central to county levels dedicate similar amounts of training time in their key point mid-career cadre classes to core curricular topics such as partybuilding, issue briefings, and management. While there is some variation in the amount of training time dedicated to management skills, on average all administrative levels dedicated 20.9 percent, or approximately one-fifth, of the total training time to developing managerial and leadership skills in the party’s rising stars. There is a steady increase in training time dedicated to party policies the further one is from Beijing, suggesting the use of party schools to build knowledge of policy directives and relevant laws as proximity to “the emperor” wanes. These findings are tentative, however, given the very few syllabi obtained for analysis. 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. It was also not possible to obtain comparable syllabi for other classes held across party schools, such as advanced training classes or classes for specific areas of expertise. The evidence presented here indicates that local schools teach content that is generally consistent with the CPS in key areas, but these local schools also enjoy some autonomy with respect to specific training content. Viewed in conjunction with the longitudinal analysis of cadre training class content at the CPS, this vertical examination across party schools demonstrates that there remains a high degree of curricular discipline within the party school system. Even in the absence of strong coercive authority relations, which exist in other xitong, the party school system exhibits remarkable uniformity. This speaks to the organizational unity the party has enjoyed in the reform era, a unity that persists in the face of decentralizing pressures in the cadre training market. Together, these findings demonstrate how reforms in cadre training have seen professio- nalizing and modernizing shifts in training content, but these changes have not come at the expense of compromising system coherence. This combination of adaptability and consistency in one set of party organizations suggests the foundations of the party’s institutional resilience during the contemporary period. 176 Adaptation measured

Conclusion Content analysis of CPS training syllabi provides one measure of adaptiveness in the party school system. In these syllabi, the mandate to craft a professional bureaucratic corps is evident in the increased emphasis on 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 party bureaucrats. This diversity represents a locally adaptive response to incentives generated by central party authorities. 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. In contemporary reforming China, leadership and management skills consume an increasing share of training time. In contrast, training time dedicated to the theories of core party leaders – party orthodoxy – is on the decline, though these ideas still command a significant share of training content. Such changes dovetail with the new environment that party schools, and the ruling CCP more broadly, have found themselves immersed in since the 1980s. 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 that it is more logical for promising leaders to be groomed in the broad arenas deemed most relevant to contemporary leadership. Party leaders appear well aware of the importance of cadres’ “ability to synthesize, to adopt global views and to make global decisions, [which are] considered the hallmark of the elite” (Suleiman 1977: 142). What is taught in party schools also offers an indication of 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.32 This was followed by a gradual process of replacing revolutionaries with “semi-bureaucrats” and, more recently, professional administrators (Lee 1984, 1991;Li1998; 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. These new emphases, however partial, accord with the view that a more

32 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 p. 1 of the Beijing Daily (Beijing ribao) dated November 11, 1979. Conclusion 177 effective bureaucracy should comprise (somewhat) independent-minded mandarins rather than bureaucrats subject to the mandates of all- controlling executives (Aberbach and Rockman 1988). What appears to be taking place with the reform of cadre training and transformation of party schools is, in postsocialist China, movement toward the technocrat- filled bureaucracy that was so difficult to achieve in the highly politicized climate of Maoist China. Market processes in the party school system have 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, requires a new sort of cadre leader.33 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. 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 2009: 93) and reflects a concerted attempt to deflect calls for more dramatic political liberalization. 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). Even more, it may be the case that party schools have an interest in broadening the range of ideas introduced in their classrooms so as to retain their position as party-sanctioned authorities on new models in governance. Party schools are in a prime position to demonstrate and moderate the CCP’s responsiveness to global waves of democratization in the past four decades. These organizations are vetted spaces for the contemplation of risky political ideas and, ultimately, can abet the party’s desire to control the global diffusion of ideas, particularly in the realm of political liberalization.

33 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). 7 Conclusion Risks and limits to party school reforms

What is remarkable about the communist party-led regimes of the twentieth century is the overall longevity of their rule. One-party autocracies have lasted for, on average, 25.5 years,1 while collapsed single-party communist regimes endured for an average of 48.3 years.2 While they lasted, these communist party systems beneﬁted from a certain institutional stability. China under CCP rule is fascinating not only because it endured through the ideological collapse and tumult of the 1980s and 1990s that 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. The party’s survival, which presents a constant outcome over time, suggests an analytical turn, that is, examining the intraparty processes underlying its organizational longevity.3 This resilience is a compound outcome affected by the actions and decisions made by the many constituent parts of the party, and this book sought to disentangle a key factor, and set of processes, underlying party survival. While the focus was on the more modest story of how one inner-party bureaucracy has adapted and survived, a single bureaucracy nonetheless has the capacity to shape the shifting social, economic, and political reality facing contemporary China. This book 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 thrive 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?

1 This is for the period 1960–2003, according to the Hadenius and Teorell dataset, totdur1ny and regime1ny variables (Hadenius and Teorell 2007). 2 See Dimitrov 2010 (N = 10). 3 For a summary of China-related scholarship on “authoritarian resilience,” see Cheng Li (2012).

178 Conclusion 179

In focusing on party schools to gain traction on broader questions of party change, I have asserted that party schools are a valuable window into how CCP organizations are capable of adaptive learning. This study has also suggested that organizational changes are responses to dramatic shifts in the party’s external environment as well as internally driven.4 First, I sought to identify whether party schools contribute to solving a particular internal problem of single-party rule. In addition to the ideological authority that they wield, these schools 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 subparty organizations of elite selection is, in some ways, an investment in both: political elites are the agents who populate and manage the party’s repressive apparatus(es) and distribute the policy goods that placate the general population. 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. 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 demonstrate how party organizations still play a unifying role in Chinese governance. Party schools lend structure and regularity to the party’s vital need to 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. Furthermore, authoritarian rulers with a long time horizon seek to institutionalize their rule and invest in human capital development. Jiang Zemin, China’s third generation leader after Deng Xiaoping and Mao Zedong, emphasized that “whether a party and a state can develop excellent leaders to a large extent determines the survival of the party and

4 Similar to Harmel’s(2002) approach to understanding why parties change, this project supports the idea that parties do not change primarily in response to internal versus external stimuli, but rather due to a combination of both. In the Chinese case, party leaders initiated “reform and opening up” in a concerted effort to modernize a country that had fallen behind regional neighbors, but these reforms were also based in an internal desire to overcome the dysfunctions of Mao-style governance. 180 Conclusion the state.”5 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 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 determine strategies to remain competitive. Throughout the reform period, party schools have adapted to changing conditions. Party schools have responded to the erosion of their monopoly position in cadre training and the introduction of competition. Competition, by deﬁnition, introduces redundancy to the system. One beneﬁt to a redundant system, and the motivation cited by central party authorities in China, has been the desire to generate incentives for organizational adaptation. A key force for change has been the creation by central party authorities of a training market whereby party and non- party, domestic and international organizations are allowed to compete with party schools for training contracts. 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 new training content and organizational directions. Many have become more outward-looking, even global, organizations. All of this has led to a system of competing and collaborating training organizations in which there is more innovation overall. The opening of markets for goods and services, which was 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 the 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 facilities to partnering with local organizations. They 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, for example, 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.

5 People’s Daily, June 9, 2002. Conclusion 181

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. As a proponent of bureaucratic redundancy would predict, party schools have invested in updating training content. On the other hand, schools have also engaged in activities of a decidedly non-training stripe. These investments, while seemingly “off mission,” are school responses to the financial 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, there is variation in who profits. Schools at the county levels and above, across inland and more prosperous coastal provinces, exhibited a remarkable similarity in their embrace of market opportunities. Townships and second-tier party schools, such as those for Communist Youth League cadres, have fallen behind. One aspect of party resilience explored in this study is the tension between top-down, system-wide control and local organizational autonomy. Party schools would seem an ideal site for centralized dissemination of updated information on the party line and unifying beliefs throughout the system. Yet field observations at schools from the provincial to county levels revealed considerable nuance in 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 bureaucrats. This diversity pushes against but does not swamp out the overall coherence in training content at various levels in the system. Central authorities demand, on the one hand, that subnational party organizations impart content that is consistent with central political goals but appropriate for local conditions. On the other hand, they have reduced 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. These choices reflect a broader and significant 182 Conclusion 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. 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 a partial layering of new organizations and actors rather than significant restructuring of the party school system itself. There has been an enlargement of the organizational space within which a set of tasks is carried out. Party school relationships with these new organizations are collaborative, competitive, and at times hierarchical. 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. The function of party schools within the party apparatus remains unchanged, though the content of the political messages and information that these schools convey, and their resource bases, have 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. Some have framed the CCP as an increasingly cloistered organization without the capacity to manage new social and economic pressures.6 This study suggests otherwise. Party authorities have retained and reshaped organizational channels for selecting leaders of the party and government. Existing party organizations have also been reformed to promote new bureaucratic priorities, namely the importance of market-oriented, globally minded public managers. As party school syllabi have shown, cadres passing through training programs receive strong signals that updated managerial skills are now a key criterion for success. A party-led focus on building “administrative civilization” (Pieke 2009: 121) is in full swing. The construction of an administrative civilization contains, beyond ideological updating, organizational responses to new economic incentives and maintenance of key party institutions. The party has altered some core features of bureaucratic life under Mao, such as recruitment based on political credentials and mass campaigns, but these changes have not led to a discernable decrease in the internal stability of the party-state. With the ascendance of the

6 Zheng argues that “the CCP has become the major obstacle to state-building in post-1949 China” (Zheng 1997: 255). Creating the context for change 183

“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 efforts to rationalize the bureaucracy were ephemeral or enduring. The verdict, based on the 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 promarket, pragmatic form that party adaptation has taken.

Creating the context for change The emergence of party schools’ entrepreneurialism has depended on several enabling conditions. Decentralization has been an important part of the story. The devolution 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 both for pecuniary gain and to maintain party schools’ standing in a competitive training market. At the same time, central authorities have driven these processes of change. As with previous realignments of institutional and individual incentives, central party authorities led the charge.7 Beyond this combination of decentralization and central direction 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. If party authorities were instead focused more exclusively on fidelity to the ideological goals of the Mao period rather than economic growth by way of heterodox market strategies, 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 skills to manage a particular set of economic priorities arose along with a longer- term growth-enhancing outlook on the part 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

7 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. 184 Conclusion 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 sought to reform their bureaucracies (Campos and Root 1996). Crucially, China’s leaders were inclined to pursue growth strategies through state-led promarket reforms, which reflects a normative shift that has moved 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 unheard of under 30 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 was critical for enlarging and shifting the boundaries of feasible reform strategies. Accordingly, 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 market outcomes. At the same time, bringing market principles into party organization is part of the long, uneven reform process of legitimizing market practices. 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. This study accords with Shirk’s(1993) analysis that a political logic constrained but lent sufficient incentives for (possibly less efficient) economic reform under CCP rule. But reform of the party school system differed from more general economic reform in important ways: rather than determine policy through elite, consensus-driven processes, leaders allowed a more contentious, competition-driven process to determine outcomes in cadre training.

Risks Marketization of cadre training has led to reforms in the operation – 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 Risks 185 potential to unleash a train of events that might prove difficult to restrain. Unsurprisingly, policy redirection has led to unanticipated outcomes and subsequent central retrenchment. For example, the rapid expansion of local, collectively owned enterprises in the 1980s and 1990s led to explo- sive but unbridled growth and high levels of local debt. Central authorities sought to check these developments through dramatic fiscal centralization policies by the mid-1990s. High local debt burdens remain. 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 progrowth incentives. The decision to devolve authority to local party schools thus speaks to this enduring tension between central control and local interests.8 The main risk to central authorities is whether the market, and its decentralizing tendencies, 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 monitoring policies may ensure control over local party school training content. Second, the partial nature of marketization entails some risk that party school failure would hinder the 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, schools must provide information on cadres’ potential for promotion 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 and inside- and outside-the-plan training classes attract income and, to varying degrees, fulfill schools’ training mandate. Yet even these developments present challenges for school leaders. As two vice-principals at a county party school noted,

8 See for instance the historical overview and thematic essays in Jia and Lin (1994). 186 Conclusion

It’s hard to balance between the government and the market. From the market perspective, it’s more important to emphasize training that raises abilities, but the government also wants more thought-type training. So we have to balance between these but it’s difficult.9 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 training goal. While schools must strive to fulfill both, the issue is one of relative resource allocation. 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, but another problem has arisen in its place. Market opportunities present more immediate rewards to school leaders. School staffs may choose to comply minimally with the less profitable, but more collectively important, goal (selection and training) and shift resources toward activities that serve the more locally beneficial goal (non-training 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 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 as long as party schools possess some discretion over local organizational decisions. Whether this discretion can be taken away in a recentralization effort 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 a possibility. As long as local party schools are minimally compliant with central goals, however, the interests

9 Interviews 202 and 203, vice-principals in charge of ﬁnance and operations and teaching and research, May 2008. Risks 187 of center and locality will both continue to drive, to varying degrees, party schools’ organizational choices.

Local embeddedness eroding central mandates 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. The party now has a stronger street-level presence. Party schools are deeply embedded within the party apparatus, but they are simultaneously building a locally relevant presence. 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). 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 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. Studies of organizational behavior and examinations of central–local bureaucratic relations in the USA, for example, have found that the more dependent a 188 Conclusion 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 et al. 1991; Whitford 2002). 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 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. 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 2000).

The risk of failure Exposing party schools to market forces has created winners and losers. While some schools, through a combination of leadership, local economic conditions, and strategic partnerships, have prospered, 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 eradi- cation of entire ministries.10 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. 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 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:

10 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. Risks 189 organization departments might become an even larger player in party management and repeat, though on an even bigger scale, the descent into stagnation that was feared so much by 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 2009: 142). Among these two possible substitutes, there is the risk of greater con- centration 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 preclude the assessment of longer-term trends. Even without comparative data on these alternatives, the party school system at present remains the most comprehensive unified system for cadre training. As such, it is positioned to continue serving as an institution of elite selection and training for the party. 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, 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 significant portion of the total training classes organized in a locale. I have not yet encountered the case of a party school receiving no planned training contracts. This is further buttressed by the center’s prerogative 190 Conclusion to allocate training contracts to party schools in general. One illustration of this is the centrally mandated rollout of 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 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 that 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 that, for the time being, political authorities have been able to skirt. I stress that the marketization of cadre training is a dynamic process, one that in theory must maintain some stability in process and function in order to remain effective. Competition, and the adaptation it promotes, requires multiple players, and the system is 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 is 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 USA, 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: 238–41).

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 Limits to party reforms 191 marketization of cadre training 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 those mechanisms that ensure that the behavior of organizations and individuals may be constrained and checked by the public and/ or other governmental bodies.11 While party schools are 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, party school reforms do not speak to the ability of party authorities to reduce the bureaucratic corruption that is drawing domestic ire and that, scholars argue, 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). Whether it is a growing pain or more serious terminal condition, current organizational reforms have fallen woefully short of addressing this particular ill. Subjecting party schools to competition sends a signal that schools must strengthen their training efforts. 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 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 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. 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

11 Rather than focus on the vertical and horizontal forms of accountability that are the conventional emphasis 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. 192 Conclusion government apparatus. It is unclear whether party entrepreneurialism can address a more encompassing vision of sound governance, beyond the updating of cadres’ managerial skills. 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 and scholarship, but the true extent of this practice is unknown.12 While party schools 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. This critique presumes the ability to measure these costs and benefits when in fact they may be difficult to compare either quantitatively or qualitatively. Perhapsthisconcernisalsolessrelevantinanauthoritariancontext 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 capacity. In the end, efficiency considerations are difficult to evaluate and it is doubtful that a comparison of the party school system before and after

12 Newspaper reports of this practice have detailed, for example, the sale in 2006 of eight leading cadre offices (three at the city level (tingji)andfive at the department level (chuji)) in Hunan and 110 county-level offices in Anhui. See “Loudi City, Hunan, Party Standing Committee’s Xie Wensheng is Arrested for Selling Office,” 21st Century Business Herald, April 22, 2008, and “Anhui County Party Secretary Receives Largest Bribe Amount from Selling Office,” China Youth Daily, December 10, 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,October20,2008,http://news.sina.com.cn/c/2008- 10-20/182816489732.shtml, accessed February 22, 2010. Zhu (2008) unravels the many systemic causes for the sale of dozens of offices in Heilongjiang, while Li (2012) has noted attempts to purchase offices as high as vice-premier in the State Council. Comparative cases 193 market reforms, notwithstanding data limitations, is likely to yield a clear verdict on this issue.

Comparative cases At best, the introduction of competitive redundancies to reforming bureaucracies can improve the state’s capacity for skillful and effective public management. While most studies of redundancy have been limited to democratic systems, there is no theoretical reason barring its application in authoritarian contexts. That the PRC has seen the creation of a competitive system in cadre training begs investigation of developments in other reforming systems. Party schools’ change in status from being the only significant provider of bureaucratic training and education to one 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. Furthermore, the party retains, to some degree, the ability to rein in party school activities. In Southeast Asia, two communist party-led countries, Vietnam and Laos, embarked on reforms similar to those in China: disarticulating political from economic reforms, with state-led efforts to advance liberalization in the latter. In 1986, doi moi policies in Vietnam and resolutions emanating from the Fourth Congress of the Lao People’s Revolutionary Party initiated the building of “socialist market economies” in both countries. As in the Chinese case, economic reforms cannot occur in isolation, and these countries have also embarked on administrative reforms of varying depths. In Laos, there has been a push to rationalize and professionalize the bureaucracy, evident in organizational changes at the central level. In cadre training, political and administrative education came together with the 1995 merger of the National School of Administration and Management and the School for Higher Studies in Political Theory. The resultant National Organisation for the Study of Politics and Administration has been engaged in training senior state managers. The activity profile for these schools is similar to Chinese counterparts in several respects. In 1992, these schools began offering four-year bachelor’s degrees, diversifying beyond training programs, and they have also partnered with peer institutions in Vietnam (UN Division for Public Administration and Development Management 2005). Degree programs pit these Lao institutions against global competitors such as the Japanese 194 Conclusion universities enrolling state and party officials in degree programs.13 What remains unclear is whether there exists a market in cadre training beyond degree programs and whether Laotian party schools face financial incentives similar to those in China. A similar story has unfolded in Vietnam, where a new emphasis on Western-style public administration has been added to the political education of officials. In 2007, a Communist Party of Vietnam Politburo decision mandated the merging of the National Academy of Public Administration with the Ho Chi Minh Political Academy. The resultant organization, named the Ho Chi Minh National Academy of Politics and Public Administration (HCMA), retains its position as the “national center for personnel training and scientific research for the Communist Party of Vietnam and the State of Vietnam.”14 Beginning in the 1990s, before the merger, the academy underwent an expansion of activities, organizing more training classes and identifying international partners. At present, the academy trains approximately 14,000–15,000 officials per year, with an additional 200–350 graduate degree students.15 By the mid- 1990s, the academy had begun collaborations with German nonprofits such as Friedrich Ebert Stiftung, to explore, among other topics, “socially and economically balanced city development ... basic values of democracy, democratic socialism and the representation of social group’s interests.”16 An overview of HCMA international collaborations in 2009 lists exchanges with US schools such as Temple University and Portland State University; in 2012, the academy signed a memorandum of understanding with Rutgers University, to continue a series of training courses for officials.17 Reforms in Vietnam’s party school system have been studied at the central level (Painter 2003), along with processes such as the privatization and modernization of education and training generally (Moock 1999). Less understood are the changes taking place

13 Since 1999, the Japan International Cooperation Center has been providing scholarships (under the Japanese Grant Aid for Human Resource Development Scholarship program) to Lao ofﬁcials; to date, over 314 fellows have been selected for degree programs in eight participating Japanese universities. Degrees are in ﬁelds such as economics, public policy, and comparative law. For an overview of this program, see http://jds-scholarship.org/ country/laos/, accessed March 10, 2015. 14 Ho Chi Minh Academy Website, “Historical Information,” www.hcma.vn/english/ HistoricalInformation/tabid/281/Default.aspx, accessed December 6, 2012. 15 Ibid. 16 See www.fesvietnam.org/en/partners-44/npaa-39.html, accessed December 6, 2012. 17 2009 international activities are described at www.hcma.vn/english/HOME/tabid/238/ctl/ Details/mid/816/ItemID/1125/Default.aspx, accessed December 6, 2012. On the Rutgers partnership, see http://global.rutgers.edu/news/734-rutgers-signs-memorandum-of-under standing-with-the-ho-chi-minh-national-academy-of-politics-a, accessed December 6, 2012. Comparative cases 195 throughout the party school system and the degree that market incentives, beyond competition for students, now shape organizational decisions.

Elite schools, elite networks There is also comparative work on the role of schools in 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, due to the network creation 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, or “privileged relationships,” are forged (Appleton 1994; 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. Japan’s most celebrated public university, Tokyo University, has been a pipeline to top positions in government and business since the nineteenth century (Rohlen 1988). A majority of Japanese civil servants are Tokyo University graduates, particularly from Todai’sfacultyof law; in 1954, 72.7 percent of civil servants were alumni (Koh 1989: 140).18 The socialization and cliques (gakubatsu) that form at this elite institution, and others like it around the world, suggest comparative processes in elite formation. China is no exception, and scholars have mapped the rise of a “Qinghua clique,” that is, graduates of the prestigious Tsinghua University in Beijing, among top party leaders (Li 2001b: Chapter 4). Party-managed institutions of cadre training would also be a prime site for the creation of both administrative and personal networks. One 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.”19 This may hold true for local party

18 Data are for civil servants ranked section chief and higher. See Koh (1989), Chapters 5 and 6. 19 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.) 196 Conclusion school training classes, while CPS courses can help build the nationwide networks critical for high officials. Party schools may serve this networking function, which is vital in China’s bureaucratic and business worlds, but evidence from field interviews remains mixed. Some interviewees cited the social, rather than professional, value of networks formed during party school training. One provincial-level official who attended several training classes at the Central Party School could not recall any professional problems that he actually solved by accessing networks from his party school training classes.20 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.”21 Another party school teacher corroborated this and pointed out how “cadres are seeking broad networks, so they want to attend training classes as high up as possible. They want more than local connections.”22 On the other hand, a former party cadre noted that attending training classes was increasingly about building political capital among the students (zhengzhi ziben).23 This was echoed by a county-level official who was sent to a city-level Mid-Career Cadre Training Class. He acknowledged that his classmates have been valuable in helping him resolve local governance issues. “Many of my classmates [at the Mid-Career 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.”24 He noted that “these networks form not after the short, one week training classes – only after training classes of longer duration.”25 Perhaps a popular saying among cadres sums it up best: party schools are a time for “studying, making connections, resting, and eating.”26 As elsewhere, the success of connections most likely depends on the type of training class a cadre is sent to, the level of the party school, and her individual initiative. Party schools offer the enabling conditions for building crucial connections, but what happens next is idiosyncratic. One party school trainee, who was sent to CPS for a one-month in-residence training class, understood this unevenness of outcomes. “Some students develop very close relationships, but for me they were nothing special

20 Interview 182, April 2008. 21 Interviews 202 and 203, May 2008. 22 Interview 215, former city-level party school teacher, May 2008. 23 Interview 15, November 2006. 24 Interview 88, December 2007. 25 Ibid. A vice-principal of a provincial socialism school echoed this sentiment, pointing out that “Usually for classes longer than two weeks, the concept [that we’re] classmates (tongxue gainian) will develop” (Interview 95, December 2007). 26 In pinyin and adopted Japanese, “Xuexi xuexi, lianxi lianxi, xiuxi xiuxi, mishi mishi.” Whither the party? 197

(yi ban de). For me, networks were not very important in the class that I enrolled in, but for others this can be an important aspect of a training class.”27 Until more systematic data are available to test the utility of these school networks, anecdotal comments from ﬁeld interviews paint at best a murky picture.

Whither the party? 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 there is an uneven withdrawal of the state from social and economic realms in exchange for the party’s unrivaled political hegemony. In formulating and pursuing this new developmental path, the CCP is committed to pragmatism in outlook and policy experimentation in practice. In this policy context, local party organs have strong incentives to be drivers of change, derive benefits from market opportunities, and open up to new influences. The boundaries around these activities are drawn at the point where they do not undermine in any overt way the legitimacy of the ruling party. Such organization-level changes show how the party is acting upon its resolution to become a “learning party.”28 In an

27 Interview 65, CPS trainee, now a county-level university official, November 2007. His training class at CPS was for cadres and administrators in universities. 28 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 www.people.com.cn/GB/40531/40746/2994977. html, accessed September 13, 2009. This initiative has spilled into the work of the 17th Party Congress. See, for example, the official interpretation, dated November 17, 2009, of the decision passed down at the Fourth Plenary Session of the 17th Party Congress on “How to Build a Learning Party,” available at www.gov.cn/jrzg/2009-11/17/con tent_1466399.htm, accessed February 12, 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 document, dated November 1, 2009, of the decision passed down at the Fourth Plenary Session of the 17th Party Congress on “Party Building Must Persist in Reform and Innovation,” available at www.gov.cn/jrzg/2009-11/01/content_1453837.htm, accessed February 12, 2010. One academic book, published in China, which elaborates upon the priorities of this “learning party” is Xie Chunhong’s Research on Contemporary CCP’sBuildingofa Learning Party-State (2009), which argues that “not to study is to retreat or die” (bu xue ze tui, bu xue ze wang). 198 Conclusion assessment of the party’s trajectory over time, party authorities and scholars have observed “two transformations”: the first shift entailed the CCP transforming from a revolutionary to a ruling party, and in the second, more recent, shift 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 demanded a capacity to manage the complexities of a market economy and a more globally integrated society (Shi 2006:92–3). In order to carry out this second transformation, Chinese leaders have chosen to position the party as more willing to learn from different models of public management. Understanding party adaptation 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 the 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 but CCP investment in new forms of cadre training reflects a real commitment of resources. It is via the bureaucratic class that the learning capacity of the party becomes manifest, in policy decisions and implementation on the ground. Party- led bureaucratic modernization is foundational among the activities that a learning party might engage in, and this modernization ensures party control over the reinvention of the Chinese state. Examining the record of cadre training over the past three decades reveals how the party is seeking new pathways for building its governing capacity. 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 management and 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 SOE restructuring, national market creation, and global integration are the imperatives of the times.29 Investing in broad-based human capital development within the party is one way to forestall the dangers of stretching scarce administrative talent too thinly over many reform fronts (Rawski 1994: 274–5). The administrative capacity that is under construction, however, is not necessarily one that will seep evenly throughout the political system. As this study has

29 This transformation was observed in the late 1980s by Brzezinski, 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). Whither the party? 199 indicated, marketization, while generating strong pressures and incentives, lacks the tidiness of a more centralized effort to engage in state 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. Throughout the CCP’s evolution, party authorities have striven to avoid the mistakes of other single-party regimes and also learn from domestic policy stumbles. In its reflective moments, the party is capable of acknowledging errors of leadership (Central Committee 1981). More generally, 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 Communist Party of the Soviet Union 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. This study of CCP-led reforms also refines those arguments 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 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 internal changes that ultimately support the construction of an enduring party-state. The CCP’s resilience is a consequence of deliberate organizational proliferation, in certain realms, 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 cadre training 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. 200 Conclusion

These considerations speak to broader issues in comparative studies of authoritarian party resilience and breakdown. Rather than focus on the array of internal party factions and cleavages that may affect party cohe- sion, this study has dwelled on organization-level determinants.30 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 (Punyaratabandhu-Bhakdi 1983). While a more 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 preferences and court politics. Maintenance of single-party rule has been the core concern driving all central decision making in China. In this sense, a political logic 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 into de facto irrelevance attests to the resilience of the party’s political project more generally. At the same time, the adoption of a “learning party” identity is as yet in its early stages. The building of market-embracing party-state structures is progressing in the now- familiar manner of crossing a river of uncertain waters, one step at a time. Because the emphasis is on the preserving party rule, rather than efficiency, reforms may be rescinded by central authorities at will. Party authorities retain the prerogative to tighten or relax their control over local organizations as they see fit. In closing, I return to the underlying logic of organizational change observed within the party. 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. Counterintuitively, central party authorities have reinforced control over political elites by introducing 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. Local experimentation and innovation across party

30 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 a 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. Whither the party? 201 schools demonstrate 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 2008a: 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. The variation in training content and organization now observed across China may be the intended outcome. In that sense, diversity is here to stay. Appendix A Number of party schools, by locale and national share of leading cadres

Number Total party Share of total leading Share of total leading Administrative of party schools in the cadres in the country, cadres in the country, unit schools country (%) 1985 (%) 1998 (%)

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 90 3.27 3.05 2.43 Mongolia 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 2,753 100.00 100.00 100.00

Sources: COD 1999; 1985 DXNJ. Note: Numbers include provincial, municipal/prefectural, county, and central ministry party schools.

202 Appendix B Note on sources and research methods

This book draws on three major sources: interviews, documents, and survey data. Over the course of 14 months of fieldwork in China from 2006 to 2011, 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 date. 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 conducted interviews at the provincial capitals of Provinces A and B, 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 of their general and leading cadre populations (Figures 1 to 6). 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, through library searches in China, and from

203 204 Note on sources and research methods

Table B.1 Administrative level of interviewees

Administrative level N %

Township 12 5.08 County 38 16.10 City 58 24.58 Province 60 25.42 Central 68 28.81 Total 236 100

Table B.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 nonexclusive categories. For example, some interviewees were both trainees and students at some point, or some party school teachers were also administrators. Table B.3 Interviewee gender

N %

Male 166 70.34 Female 70 29.66 Total 236 100 government Internet sites. When visiting party schools, for example, I asked interviewees for copies of training syllabi and training materials. Additional data were 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. Province A 11,000 10,000 9,000 8,000 GDP per capita (yuan) 7,000 6,000 0.0006 0.0008 0.001 0.0012 0.0014 0.0016 Number of cadres per cap

Figure B.1 Cadre population and GDP per capita, coastal region Source: Data on cadre populations were obtained from COD 1999; all other data from provincial-level statistical yearbooks, China Data Online, 1998. 6,500 Province B 6,000 5,500 5,000 GDP per capita (yuan) 4,500 4,000

0.0006 0.0008 0.001 0.0012 Number of cadres per cap

Figure B.2 Cadre population and GDP per capita, central region Source: Data on cadre populations were obtained from COD 1999; all other data from provincial-level statistical yearbooks, China Data Online, 1998. Province A 11000 10000 9,000 8,000 GDP per capita (yuan) 7,000 6,000 0.00025 0.0003 0.00035 0.0004 0.00045 0.0005 Number of leading cadres per cap

Figure B.3 Leading cadre population and GDP per capita, coastal region Source: Data on cadre populations were obtained from COD 1999; all other data from provincial-level statistical yearbooks, China Data Online, 1998. 6,500 Province B 6,000 5,500 5,000 GDP per capita (yuan) 4,500 4,000

0.0002 0.0003 0.0004 0.0005 Number of leading cadres per cap

Figure B.4 Leading cadre population and GDP per capita, central region Source: Data on cadre populations were obtained from COD 1999; all other data from provincial-level statistical yearbooks, China Data Online, 1998. Note on sources and research methods 207

30,000

20,000

Province A 10,000

GDP per capita (yuan) Province B

0 0 0.0005 0.001 0.0015 0.002 Number of leading cadres per cap

Figure B.5 Leading cadre population and GDP per capita, all regions Source: Data on cadre populations were obtained from COD 1999; all other data from provincial-level statistical yearbooks, China Data Online, 1998.

30,000

20,000

Province A 10,000

GDP per capita (yuan) Province B

0 0 0.001 0.002 0.003 0.004 Number of cadres per cap

Figure B.6 Cadre population and GDP per capita, all regions Source: Data on cadre populations were obtained from COD 1999; all other data from provincial-level statistical yearbooks, China Data Online, 1998. Appendix C Central Party School organization

Office/department Party committee (机关党委) General office (办公厅) Education department (教务部) Research department (科研部) Academic departments: Marxism department (马克思主义理论教研部) Philosophy department (哲学教研部) Economics department (经济学教研部) Scientific socialism department (科学社会主义教研部) Politics and law department (政法教研部) Party history department (中共党史教研部) Literature and history department (文史教研部) International Institute for Strategic Studies (国际战略研究所) Research office (研究室) Advanced training department (进修部) Training department (培训部) School for graduate students (研究生院) Library (图书馆) Organization department (组织部) Finance office (财务行政管理局) Retired cadre bureau (离退休干部局) Periodicals (报刊社) Cadre education school (干部教育学院) Services office (机关服务局) Information technology center (信息中心) Publishing house (出版社)

Source: CPS webpage, www.ccps.gov.cn/organization/, accessed December 20, 2012.

208 Appendix D City Z training allocations, 2008

Class name Duration (days) Enrollment Location(s)

Mid-Career Cadre training 45 60 City party school, class Central Party School Spirit of the 17th Party 3 360 City party school Congress class City-level cadre advanced 45 60 City party school, class provincial university City-level cadre advanced 45 60 City party school, class provincial university Promoting urban and rural 3 74 City party school development and building a new socialist countryside advanced research class Building a civilized city and 3 49 City party school cultural production development research and discussion class Environmentally conscious 3 52 City party school city and civilization building research and discussion class Innovation and rapid 3 70 City party school transitioning economy research and discussion class “City of innovation” research 15 35 Hong Kong and discussion class University Social work management 15 35 Singapore University research and discussion class Enterprise manager advanced 7 120 Tsinghua University, research class Fudan University Enterprise manager advanced 7 120 Tsinghua University, research class Fudan University

209 210 City Z training allocations, 2008

Class name Duration (days) Enrollment Location(s)

Agricultural technologies 7 50 City party school training class Democratic parties, 15 45 Shenzhen Socialism Association of Industry and Institute Commerce, and non-party representative training class Industrial economic 7 50 City party school management leader training class Personnel allocation training 6 50 Fudan University class Private enterprise party 3 40 City party school secretary demonstration class Newly promoted section-level 15 60 City party school cadre training class City party representative 2 100 City party school training class Party spirit education training 3 City party school class High-level crisis management 6 70 City party school research class Overseas Chinese Ofﬁce 7 50 Fujian university cadre training class Family planning leading 2 200 City party school cadre training class Family planning leading 2 130 City party school cadre training class Grassroots letters and 3 120 City party school complaints leader training class Non-CCP Mid-Career Cadre 30 55 City socialism training class institute, Shanghai Socialism Institute Retired cadre training class 2 100 City senior citizen university CCP activist training class 4 200 City party school Discipline and inspection 7 50 City party school cadre training class The city’s “Low income rural 7 80 City party school household wellness project” cadre training The city’s “Low income rural 7 80 City party school household wellness project” cadre training Public security section chief 7 480 City police academy training class City Z training allocations, 2008 211

Class name Duration (days) Enrollment Location(s)

Retired cadre bureau chief 6 20 City party school training class Communist Youth League 3 130 Another city cadre training class Party affairs cadre training 3 150 City party school class City building leading cadre 15 30 Provincial training class construction ministry, Cadre school City military affairs civilian 45 City party school, cadre training class City personnel training and testing center Party committee ofﬁce affairs 3 120 City party school training class Theoretical reserve cadre 3 200 City party school training class Discipline and inspection 15 30 Central discipline party secretary training and inspection class training center

Source: Internal school document. Appendix E Descriptive statistics and robustness tests of PSM presented in Chapter 3

212 Table E.1 Pairwise correlations between independent and control variables

Science or CCP Total work Total years engineer- Economics Management Law Monthly member years of education ing major major major major income Age Female Frq. Upper Military

CCP member 1.00 Total work yrs 0.01 1.00 Total yrs educ −0.04 0.20 1.00 Sci/eng major −0.06 0.15 0.21 1.00 Econ major 0.01 −0.01 0.21 −0.22 1.00 Manage major 0.11 −0.10 0.16 −0.13 −0.02 1.00 Law major 0.07 0.00 0.11 −0.14 −0.09 −0.05 1.00 Income/month −0.09 0.03 0.21 0.08 0.17 0.06 0.01 1.00 Age 0.18 0.06 −0.32 −0.01 −0.18 −0.16 −0.10 −0.20 1.00 Female −0.21 −0.04 0.09 0.01 0.13 −0.04 −0.01 0.06 −0.11 1.00 Freq upper 0.15 −0.04 −0.03 −0.01 −0.01 0.06 0.03 −0.02 0.02 −0.02 1.00 Military 0.21 0.03 −0.17 −0.13 −0.07 −0.03 0.01 −0.07 0.16 −0.23 −0.01 1.00 214 Descriptive statistics and robustness tests

Table E.2 Descriptive summary of variables

Number of Variable observations Mean SD Min Max

CCP member 586 0.70 0.46 0 1 Total work years 586 20.48 3.97 12 42 Total years of education 589 6.82 4.59 0 70 Science/engineering 589 0.30 0.46 0 1 major Economics major 589 0.16 0.37 0 1 Management major 589 0.12 0.33 0 1 Law major 589 0.14 0.35 0 1 Income/month 564 1,378.56 1,308.71 0 20,000 Age 589 50.78 12.10 20 72 Female 589 0.26 0.44 0 1 Frequency of interaction 578 0.86 0.35 0 1 with superiors Military experience 588 0.20 0.40 0 1 Parent CCP member 589 0.33 0.47 0 1

Table E.3 T-tests for equality of means across treatment and control groups, before and after matching

Mean Mean Variable Sample treated control t-test p>t

CCP member Unmatched 0.61 84.40 8.16 0.00 Matched 0.93 0.95 −0.71 0.48 Total work Unmatched 20.01 20.68 −1.81 0.07 years Matched 20.01 19.79 0.61 0.54 Total years of Unmatched 6.65 6.89 −0.55 0.59 education Matched 6.65 6.27 0.94 0.35 Science/ Unmatched 0.32 0.30 0.38 0.70 engineering Matched 0.32 0.33 −0.32 0.75 major Economics Unmatched 0.15 0.17 −0.58 0.57 major Matched 0.15 0.12 0.88 0.38 Management Unmatched 0.17 0.10 2.46 0.01 major Matched 0.17 0.14 0.84 0.40 Law major Unmatched 0.22 0.12 3.17 0.00 Matched 0.22 0.20 0.50 0.62 Income/ Unmatched 1,301.90 1,406.10 −0.85 0.40 month Matched 1,301.90 1,302.70 −0.01 0.99 Age Unmatched 51.49 50.83 0.58 0.56 Matched 51.49 52.28 −0.62 0.54 Female Unmatched 0.22 0.29 −1.76 0.08 Matched 0.22 0.21 0.27 0.79 Descriptive statistics and robustness tests 215

Table E.3 (cont.)

Mean Mean Variable Sample treated control t-test p>t

Frequency of Unmatched 0.90 0.83 1.92 0.06 interaction Matched 0.90 0.89 0.22 0.82 with superiors Military Unmatched 0.18 0.21 −1.00 0.32 experience Matched 0.18 0.20 −0.49 0.62 Parent CCP Unmatched 0.36 0.32 0.87 0.39 member Matched 0.36 0.33 0.43 0.67 Beijing Unmatched 0.07 0.11 −1.55 0.12 Matched 0.07 0.05 0.64 0.52 Tianjin Unmatched 0.10 0.09 0.35 0.73 Matched 0.10 0.09 0.28 0.78 Hebei Unmatched 0.03 0.03 −0.05 0.96 Matched 0.03 0.02 0.43 0.67 Shanxi Unmatched 0.01 0.02 −0.91 0.36 Matched 0.01 0.01 −0.45 0.65 Inner Unmatched 0.02 0.01 0.75 0.46 Mongolia Matched 0.02 0.00 1.17 0.24 Liaoning Unmatched 0.02 0.03 −0.59 0.56 Matched 0.02 0.01 0.82 0.41 Jilin Unmatched 0.01 0.01 0.88 0.38 Matched 0.01 0.03 −1.22 0.22 Heilongjiang Unmatched 0.01 0.02 −1.25 0.21 Matched 0.01 0.01 −0.17 0.87 Shanghai Unmatched 0.07 0.06 0.69 0.49 Matched 0.07 0.05 0.67 0.50 Jiangsu Unmatched 0.06 0.04 1.12 0.27 Matched 0.06 0.09 −0.99 0.32 Zhejiang Unmatched 0.01 0.01 −0.72 0.48 Matched 0.01 0.01 −0.17 0.87 Anhui Unmatched 0.02 0.08 −2.32 0.02 Matched 0.02 0.02 0.09 0.93 Fujian Unmatched 0.02 0.03 −0.44 0.66 Matched 0.02 0.02 0.48 0.63 Jiangxi Unmatched 0.02 0.02 0.00 1.00 Matched 0.02 0.03 −0.63 0.53 Shandong Unmatched 0.04 0.05 −0.79 0.43 Matched 0.04 0.06 −0.85 0.40 Henan Unmatched 0.02 0.06 −1.67 0.10 Matched 0.02 0.02 0.18 0.85 Hubei Unmatched 0.15 0.03 5.07 0.00 Matched 0.15 0.11 0.99 0.32 Hunan Unmatched 0.11 0.05 2.57 0.01 Matched 0.11 0.12 −0.34 0.73 216 Descriptive statistics and robustness tests

Table E.3 (cont.)

Mean Mean Variable Sample treated control t-test p>t

Guangdong Unmatched 0.05 0.07 −0.95 0.34 Matched 0.05 0.05 0.13 0.90 Guangxi Unmatched 0.04 0.03 0.15 0.88 Matched 0.04 0.04 −0.07 0.94 Chongqing Unmatched 0.01 0.00 1.39 0.17 Matched 0.01 0.00 1.17 0.24 Sichuan Unmatched 0.02 0.02 −0.20 0.84 Matched 0.02 0.03 −0.94 0.35 Guizhou Unmatched 0.01 0.05 −2.49 0.01 Matched 0.01 0.02 −0.80 0.42 Yunnan Unmatched 0.05 0.03 0.82 0.41 Matched 0.05 0.06 −0.43 0.67 Shaanxi Unmatched 0.01 0.02 −1.25 0.21 Matched 0.01 0.00 0.19 0.85 Gansu Unmatched 0.02 0.01 1.24 0.21 Matched 0.02 0.03 −0.26 0.80 Frequency 0 50 100 150

0 0.02 0.04 0.06 0.08 Absolute difference in propensity score

Figure E.1 Absolute distance between the propensity scores of each treatment observation and nearest neighbor match Descriptive statistics and robustness tests 217

Tables E.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

Treated Control Method (N) (N) ATT SE t-statistic

Nearest neighbor, 167 76 0.108 0.098 1.097 random draw logit (attnd) Nearest−neighbor, 167 77 0.165 0.071 2.325 equal weight (attnw) Radius (attr) NA –––– Stratiﬁcation (atts) 167 347 0.145 0.067 2.160 Kernel (attk) 167 322 0.085 0.060 1.416

DV: Department rank (dummy)

Treatment: Enrollment in non-degree party school training class Method Treated (N) Control (N) ATT SE t-statistic

Nearest neighbor, 167 92 0.060 0.073 0.825 random draw logit (attnd) Nearest-neighbor, 167 90 0.126 0.052 2.414 equal weight (attnw) Radius (attr) 135 243 0.087 0.069 1.254 Stratiﬁcation (atts) 167 347 0.076 0.038 2.030 Kernel (attk) 167 322 0.072 0.046 1.565 218 Descriptive statistics and robustness tests

DV: Any rank (dummy)

Treatment: Enrollment in non-degree party school training class

Method Treated (N) Control (N) ATT SE t-statistic

Nearest 167 92 0.102 0.072 1.420 neighbor, random draw logit (attnd) Nearest- 167 90 0.174 0.080 2.179 neighbor, equal weight (attnw) Radius (attr) 135 243 −0.015 0.053 −0.273 Stratiﬁcation 167 347 0.146 0.058 2.523 (atts) Kernel (attk) 167 322 0.094 0.056 1.680

Note: Bootstrapped standard errors are reported (200 replications).

Table E.7 Probit regression results, effect of party school training on promotion

DV: Administrative rank (dummy) Coefﬁcient Robust SE

Constant −3.25 0.77** Age 0.04 0.01** Female −0.57 0.20** Monthly income 0.00 0.00** Total years in the work force −0.01 0.02 Education (total years) 0.05 0.03 Major: Science and engineering 0.46 0.22* Major: Law and humanities 0.19 0.23 Major: Economics 0.70 0.26** Major: Management 0.31 0.27 Political control variables (dummies) Party school 0.43 0.05* CCP membership 0.49 0.18** Parents’ CCP membership −0.12 0.17 Military experience 0.07 0.21 Frequent interaction with superiors 0.29 0.23 Provincial-level dummies Beijing 1.27 0.49** Tianjin 1.54 0.53 Hebei 1.05 0.56 Descriptive statistics and robustness tests 219

Table E.7 (cont.)

DV: Administrative rank (dummy) Coefﬁcient Robust SE

Shanxi 1.60 0.66* Neimenggu 0.56 0.63 Jilin 1.12 0.79 Heilongjiang 1.48 0.74** Shanghai 0.05 0.49 Jiangsu 1.32 0.59* Zhejiang 1.32 0.63* Anhui 0.95 0.49* Fujian 0.63 0.55 Jiangxi 1.58 0.63* Shandong 1.04 0.54 Henan 1.02 0.51* Hubei 0.52 0.50 Hunan 0.20 0.50 Guangdong 0.50 0.50 Guangxi 1.43 0.58** Chongqing −0.84 0.79 Sichuan 1.32 0.61* Guizhou 1.71 0.57** Yunnan 1.37 0.54** Gansu 0.34 0.63 Iteration 0: log pseudolikelihood −277.56 Iteration 1: log pseudolikelihood −210.16 Iteration 2: log pseudolikelihood −204.41 Iteration 3: log pseudolikelihood −204.02 Iteration 4: log pseudolikelihood −204.02 N 524 Wald chi2(39) 116.67 Prob > chi2 0 Pseudo R2 0.27

Notes: Signiﬁcance codes: ** for p<0.01,* for p<0.05. Dropped regional dummies: Xinjiang, Shaanxi, and Liaoning. Table E.8 Marginal effect for signiﬁcant control variables

Variable Marginal effect

Party school 0.17 CCP membership 0.20 Age 0.02 Female −0.23 Monthly income 0.0001 Major: science and engineering 0.19 220 Descriptive statistics and robustness tests

Table E.8 (cont.)

Variable Marginal effect

Major: economics 0.28 Beijing 0.51 Shanxi province 0.64 Heilongjiang province 0.59 Guangxi province 0.57 Sichuan province 0.53 Guizhou province 0.68 Yunnan province 0.55

Notes: Other control variables were held constant at their medians. Appendix F Central Party School Mid-Career Cadre Training Classes descriptive data

Table F.1 Age at the time of training

Combined 2000 1995

N % N % N %

30–39 43 9.7 13 5.3 30 15.1 40–49 268 60.6 152 62.6 116 58.3 50–55 38 8.6 24 9.9 14 7.0 NA 93 21.0 54 22.2 39 19.6 Total 442 100.0 243 100.0 199 100.0

Table F.2 Gender

Combined 2000 1995

N % N % N %

Male 362 81.9 200 82.3 162 81.4 Female 45 10.2 27 11.1 18 9.0 NA 35 7.9 16 6.6 19 9.5 Total 442 100.0 243 100.0 199 100.0

Table F.3 Home province at the time of training

Combined 2000 1995

N % N % N %

Beijing 12 2.7 4 1.6 8 4.0 Tianjin 9 2.0 5 2.1 4 2.0 Hebei 19 4.3 9 3.7 10 5.0

221 222 Training Classes descriptive data

Table F.3 (cont.)

Combined 2000 1995

N % N % N %

Shanxi 12 2.7 6 2.5 6 3.0 Inner Mongolia 9 2.0 5 2.1 4 2.0 Liaoning 18 4.1 10 4.1 8 4.0 Jilin 15 3.4 9 3.7 6 3.0 Heilongjiang 9 2.0 3 1.2 6 3.0 Shanghai 4 0.9 0 0.0 4 2.0 Jiangsu 21 4.8 13 5.3 8 4.0 Zhejiang 22 5.0 13 5.3 9 4.5 Anhui 15 3.4 11 4.5 4 2.0 Fujian 8 1.8 3 1.2 5 2.5 Jiangxi 11 2.5 4 1.6 7 3.5 Shandong 29 6.6 20 8.2 9 4.5 Henan 16 3.6 10 4.1 6 3.0 Hubei 13 2.9 4 1.6 9 4.5 Hunan 13 2.9 8 3.3 5 2.5 Guangdong 11 2.5 8 3.3 3 1.5 Guangxi 8 1.8 3 1.2 5 2.5 Hainan 7 1.6 3 1.2 4 2.0 Chongqing 5 1.1 3 1.2 2 1.0 Sichuan 12 2.7 6 2.5 6 3.0 Guizhou 6 1.4 2 0.8 4 2.0 Yunnan 8 1.8 4 1.6 4 2.0 Shaanxi 12 2.7 6 2.5 6 3.0 Gansu 8 1.8 4 1.6 4 2.0 Qinghai 8 1.8 4 1.6 4 2.0 Ningxia 4 0.9 0 0.0 4 2.0 Xinjiang 8 1.8 2 0.8 6 3.0 Tibet 7 1.6 3 1.2 4 2.0 NA 83 18.8 58 23.9 25 12.6 Total 442 100.0 243 100.0 199 100.0

Table F.4 Educational attainment

Combined 2000 1995

N % N % N %

Professional school 1 0.2 0 0.0 1 0.5 Part-time specialized college 5 1.1 5 2.1 0 0.0 Full-time specialized college 21 4.8 12 4.9 9 4.5 University 71 16.1 36 14.8 35 17.6 Training Classes descriptive data 223

Table F.4 (cont.)

Combined 2000 1995

N % N % N %

Graduate school 133 30.1 84 34.6 49 24.6 Training apprentice 1 0.2 1 0.4 0 0.0 Other 112 25.3 47 19.3 65 32.7 NA 98 22.2 58 23.9 40 20.1 Total 442 100.0 243 100.0 199 100.0

Note: The “other” category includes party school graduate degrees.

Table F.5 Percent with party school degrees*

Combined 2000 1995

N % N % N %

Undergraduate 10 2.3 4 1.6 6 3.0 Graduate 114 25.8 48 19.8 66 33.2 Total 124 52 72

* Note: 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 master’s degrees obtained these through the CPS (Li 2008b).

Table F.6 University major

Combined 2000 1995

N % N % N %

Science 12 2.7 5 2.1 7 3.5 Engineering 27 6.1 17 7.0 10 5.0 Computer application and software 3 0.7 2 0.8 1 0.5 Medicine and pharmacology 2 0.5 2 0.8 0 0.0 Agriculture, forestry, ﬁsheries, animal husbandry 10 2.3 7 2.9 3 1.5 Finance and economics 47 10.6 26 10.7 21 10.6 Management and administration 34 7.7 18 7.4 16 8.0 Law 38 8.6 24 9.9 14 7.0 Social sciences 6 1.4 1 0.4 5 2.5 Humanities 42 9.5 17 7.0 25 12.6 Foreign languages 3 0.7 2 0.8 1 0.5 Education and information systems 3 0.7 0.0 3 1.5 NA 215 48.6 122 50.2 93 46.7 Total 442 100 243 100 199 100 Appendix G International partnerships, central and provincial-level party schools

Year partnership School Partner began

Central Party Sozialdemokratische Partei Deutschlands (Germany) School 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

224 International partnerships, central, provincial-level party schools 225

Year partnership School Partner began

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 Shaanxi 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 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 at least 2004 Administration National Academy of Public Administration (Vietnam) at least 2004 226 International partnerships, central, provincial-level party schools

Year partnership School Partner began

National Political and Administrative Academy Ho Chi at least 2004 Minh (Vietnam) Korean Research Institute for Local Administration 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) Appendix H Categories for coding training syllabi

Partybuilding Central party policy – party building (including united front, unions, etc.) Party history Party constitution Party line/party building Marxist-Leninist theory Mao Zedong theory Deng Xiaoping theory – this includes “Socialism with Chinese Characteristics” Jiang Zemin theory – this includes “Three Represents” Hu Jintao theory – this includes “Harmonious Society” and “Scientiﬁc Development” Capitalist theory/non-Marxist western economic theory Liberal democratic theory General theories of socialism (no clear attribution) – e.g., socialist development, socialist economy Minorities in a socialist system Management Leadership theory/”art of leadership” Management (including crisis management) Speechmaking and communication Media relations Principles of service government Public administration Strategic thinking National and international brieﬁngs International – USA International – Asia International – Western Europe International – USSR and Eastern Europe International – general cases/lessons from abroad International – international organizations Domestic – building new countryside Domestic – coastal economic development Domestic – inland economic development

227 228 Categories for coding training syllabi

Policy and law Central party policy – economic development (e.g., state-owned enterprises, agriculture, 5-year plans) Central party policy – social reform 16th Party Congress 17th Party Congress 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”) General law Administrative law Business/ﬁnance law Criminal law Labor law Property rights Other Basic social science Basic humanities Chinese history and culture Basic sciences Investigation and basic writing (reports and composition) Budgeting and accounting Military training Foreign language Computer class Ethics General interest lectures Elective course – via TV or Internet 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):606–12. Aldrich, Howard E. 1999. Organizations Evolving. London: Sage Publications. Alpermann, Bjorn. 2001. “The Post-Election Administration of Chinese Villages.” The China Journal 46:45–67. Ang, Yuen Yuen. 2007. “Public Employment as Redistribution in China: Fiscal Games, Mass Clientelism, or Risk Management?” Midwest Political Science Association, Chicago. Ang, Yuen Yuen. 2008. “State, Market, and Bureau-Contracting: Organizing Rents of Power in Reform China.” Ph.D. Thesis, Department of Political Science, Stanford University, Stanford. Ang, Yuen Yuen. 2009. “State, Market, and Bureau-Contracting in Reform China.” Political Science, Stanford University, Stanford, CA. Ang, Yuen Yuen. 2012. “Counting Cadres: A Comparative View of the Size of China’s Public Employment.” China Quarterly 211:676–96. Appleton, Andrew. 1994. “The Formal versus Informal Rules of French Political Parties.” In How Political Parties Work: Perspectives from Within, ed. K. Lawson. Westport: Praeger, pp. 23–54. 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. Bandurski, David, and Martin Hala. 2010. Investigative Journalism in China: Eight Cases in Chinese Watchdog Journalism. Hong Kong: Hong Kong University Press. Bardhan, Pranab. 2000. “Local Governance and Delivery of Public Goods.” In Governance, Decentralization and Reform in China, India and Russia, ed. J.-J. Dethier. Boston and London: Kluwer Academic Publishers, pp. 179–87. 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, NJ: Princeton University Press. Baum, Richard. 2008. “Political Implications of China’s Information Revolution: The Media, the Minders, and Their Message.” In China’s Changing Political Landscape: Prospects for Democracy, ed. C. Li. Washington, DC: Brookings Institution Press, pp. 161–84.

229 230 References

Becker, Sascha O., and Andrea Ichino. 2002. “Estimation of Average Treatment Effects Based on Propensity Scores.” The Stata Journal 2 (4):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, Princeton, New Jersey. Blaydes, Lisa. 2011. Elections and Distributive Politics in Mubarak’s Egypt. Cambridge, UK: Cambridge University Press. Bo, Zhiyue. 2002. Chinese Provincial Leaders: Economic Performance and Political Mobility Since 1949. Armonk, NY: M.E. Sharpe. Bo, Zhiyue. 2006. “The ‘Freezing Point (Bingdian)’ Saga in China: Did the Chinese Communist Party’s Propaganda Department Make a Mistake?” EAI Background Brief (275):12. Bottomore, T. B. 1964. Elites and Society. New York: Basic Books, Inc. Boyd, Christina L., Lee Epstein, and Andrew D. Martin. 2010. “Untangling the Causal Effects of Sex on Judging.” American Journal of Political Science 54 (2):389–411. Brady, Anne-Marie. 2008. Marketing Dictatorship. Lanham: Rowman and Littleﬁeld. 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, UK: Cambridge University Press. Brownlee, Jason. 2011. “Executive Elections in the Arab World: When and How Do They Matter?” Comparative Political Studies 44 (7):807–28. Brunsson, Nils. 1999. “Standardization as Organization.” In Organizing Political Institutions: Essays for Johan P. Olsen, ed. M. Egeberg and P. Laegreid. Oslo: Scandinavian University Press, pp. 109–28. Brzezinski, Zbigniew. 1989. The Grand Failure: The Birth and Death of Communism in the Twentieth Century. New York: Charles Scribner’s Sons. Burnham, James. 1962. The Managerial Revolution. Bloomington: Indiana University Press. Burns, John P. 1983. “Reforming China’s Bureaucracy, 1979–82.” Asian Survey 23 (6):692–722. Burns, John P. 1989a. “Introduction: The Chinese Nomenklatura System.” In The Chinese Communist Party’s Nomenklatura System: A Documentary Study of Party Control of Leadership Selection, 1979–1984, ed. J. P. Burns. Armonk, NY and London: M.E. Sharpe, Inc., pp. ix–xlii. Burns, John P., ed. 1989b. The Chinese Communist Party’s Nomenklatura System: A Documentary Study of Party Control of Leadership Selection, 1979–1984. Armonk, NY 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. References 231

Burns, John P. 1994. “Strengthening Central CCP Control of Leadership Selection: The 1990 Nomenklatura.” China Quarterly 138:458–91. Burns, John P. 2003. “‘Downsizing’ the Chinese State: Government Retrenchment in the 1990s.” China Quarterly 175:775–802. Burns, John P. 2004. “Governance and Civil Service Reform.” In Governance in China, ed. J. Howell. Lanham: Rowman & Littleﬁeld Publishers, Inc., pp. 37–57. Burns, John P. 2006. “The Chinese Communist Party’s Nomenklatura System as a Leadership Selection Mechanism.” In The Chinese Communist Party in Reform, ed. K. E. Brodsgaard and Z. Yongnian. London and New York: Routledge, pp. 33–58. Burns, John P., and Bidhya Bowornwathana, eds. 2001. Civil Service Systems in Asia. Cheltenham, UK and Northampton, MA: MPG Books Ltd. Burns, John P., and Wang Xiaoqi. 2010. “Civil Service Reform in China: Impacts on Civil Servants’ Behaviour.” China Quarterly 201:58–78. Cai, Hongbin, and Daniel Treisman. 2006. “Did Government Decentralization Cause China’s Economic Miracle?” World Politics 58 (4):505–35. 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’sNomenklatura System: A Documentary Study of Party Control of Leadership Selection, 1979–1984, ed. J. P. Burns. Armonk, NY and London: M.E. Sharpe, Inc., pp. 50–83. 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. Central Party School Yearbook Editorial Committee, ed. 1985. Central Party School Yearbook 1984. Beijing: Central Party School Publishing House. Chan, Joseph Man. 1993. “Commercialization without Independence: Trends and Tensions of Media Development in China.” Chapter 25. In China Review 1993, ed. J. C. Yu-shek and M. Brosseau. Hong Kong: Chinese University Press. 232 References

Chan, Ying. 2010. “Introduction: The Journalism Tradition.” In Investigative Journalism in China: Eight Cases in Chinese Watchdog Journalism,ed. D. Bandurski and M. Hala. Hong Kong: Hong Kong University Press, pp. 1–17. Chang, Gordon G. 2001. The Coming Collapse of China. New York: Random House. Chen, Guisheng. 2007. China’s Cadre Education, 1927–1949 (zhongguo gan bu jiao yu (1927–1949)). Shanghai: Huadong Normal University Press. Cheung, Tai Ming. 2001. China’s Entrepreneurial Army. Oxford: Oxford University Press. Chin, Gregory T. 2011. “Innovation and Preservation: Remaking China’s National Leadership Training System.” China Quarterly 205:18–39. 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 54:359–77. Chow, King W. 1993. “The Politics of Performance Appraisal.” In Public Administration in China, ed. M. K. Mills and S. S. Nagel. Westport and London: Greenwood Press, pp. 359–77. Cox, Gary W. 2009. “Authoritarian Elections and Leadership Succession, 1975– 2004.” Paper presented at the 2009 American Political Science Association Annual Meeting. Toronto, Canada. Available at SSRN: http://ssrn.com/ abstract=1449034. CPS, ed. 2004. Research on Certain Questions Concerning the Building of a Governing 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), ed. T. Shi. Beijing: Central Party School Publishing House, pp. 314–20. 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):1053–62. 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.” Available at SSRN: http://ssrn.com/abstract=1153510; http://dx. doi.org/10.2139/ssrn.1153510. Unpublished manuscript. Dickson, Bruce J. 2003. Red Capitalists in China: The Party, Private Entrepreneurs, and Prospects for Political Change.Cambridge,UK:Cambridge University Press. Dickson, Bruce J. 2008. Wealth into Power: The Communist Party’s Embrace of China’s Private Sector. Cambridge, UK: Cambridge University Press. References 233

Dimitrov, Martin. 2010. “Popular Accountability and Regime Resilience in Contemporary China: Evidence from Citizen Complaints.” Dartmouth College, Hanover, New Hampshire. Ding, X. L. 1994. The Decline of Communism in China: Legitimacy Crisis, 1977–1989. Cambridge, UK: 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. 1967. Inside Bureaucracy. Boston: Little, Brown and Company. 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. Eastman, Lloyd E. 1974. The Abortive Revolution: China Under Nationalist Rule, 1927–37. Cambridge, MA: Harvard University Press. Eisenman, Joshua. 2012. “China’s Central Party School: A Primer.” American Foreign Policy Council China Brief. Eisinger, Peter K. 1988. The Rise of the Entrepreneurial State: State and Local Economic Development Policy in the United States. Madison: University of Wisconsin Press. 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, ed. US Congress, 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, ed. D. Braun and F. Gilardi. London and New York: Routledge, pp. 77–98. Esarey, Ashley. 2005. “Cornering the Market: State Strategies for Controlling China’s Commercial Media.” Asian Perspective 29 (4):37–83. Esarey, Ashley. 2006. “Speak No Evil: Mass Media Control in Contemporary China.” Freedom at Issue: A Freedom House special Report. Fan, Zhilong. 1984. “Can China Learn Anything from Western Techniques?” China Reconstructs (January):60–3. 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):247–52. 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, ed. D. M. Finkelstein and M. Kivlehan. Armonk, NY and London: M.E. Sharpe, pp. 152–64. 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. Washington, DC: International Republican Institute. 234 References

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, UK: Cambridge University Press. Gandhi, Jennifer, and Ellen Lust-Okar. 2009. “Elections Under Authoritarianism.” Annual Review of Political Science 12 (1):403–22. Gandhi, Jennifer, and Adam Przeworski. 2007. “Authoritarian Institutions and the Survival of Autocrats.” Comparative Political Studies 40 (11):1279–301. Geddes, Barbara. 1994. Politician’s Dilemma: Building State Capacity in Latin America. Berkeley: University of California Press. Geddes, Barbara. 1999a. “Authoritarian Breakdown: Empirical Test of a Game Theoretic Argument.” In Annual Meeting of the American Political Science Association. Atlanta. Geddes, Barbara. 1999b. “What Do We Know About Democratization After Twenty Years?” Annual Review of Political Science 2:115–44. Geddes, Barbara. 2003. 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, Princeton, New Jersey. Gehlbach, Scott, and Philip Keefer. 2008. “Investment without Democracy: Ruling-Party Institutionalization and Credible Commitment in Autocracies.” Journal of Comparative Economics 39: 123–39. General View of the Nation’s Party Schools (quanguo dangxiao gailan). 1996. Beijing: Red Flag Publishing House. 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):105–9. Gore, Lance L. P. 1998. Market Communism: The Institutional Foundation of China’s Post-Mao Hyper-Growth. Hong Kong: Oxford University Press. Gore, Lance L. P. 2011. The Chinese Communist Party and China’sCapitalist Revolution: The Political Impact of the Market. London and New York: Routledge. 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, Singapore. Guo, Zhenzhi. 2003. “Playing the Game by the rules? Television Regulation Around China’s Entry into WTO.” The Public 10 (4):5–18. Hadenius, Axel, and Jan Teorell. 2007. “Pathways from Authoritarianism.” Journal of Democracy 18 (1):143–56. Hall, Roger I. 1976. “A System Pathology of an Organization: The Rise and Fall of the Old Saturday Evening Post.” Administrative Science Quarterly 21 (2):185–211. References 235

Hannan, Michael T., and John Freeman. 1984. “Structural Inertia and Organizational Change.” American Sociological Review 49 (2):149–64. Hanson, Stephen E. 1995. “The Leninist Legacy and Institutional Change.” Comparative Political Studies 28 (2):306–14. Harding, Harry. 1981. Organizing China: The Problem of Bureaucracy, 1949–1976. Stanford: Stanford University Press. Harmel, Robert. 2002. “Party Organizational Change: Competing Explanations?” In Political Parties in the New Europe: Political and Analytical Challenges,ed. K. R. Luther and F. Muller-Rommel. Oxford: Oxford University Press, pp. 119–42. Hassid, Jonathan. 2008. “Controlling the Chinese Media: An Uncertain Business.” Asian Survey 48 (3):414–30. Heilmann, Sebastian. 2008a. “From Local Experiments to National Policy: The Origins of China’s Distinctive Policy Process.” China Journal 59:1–31. Heilmann, Sebastian. 2008b. “Policy Experimentation in China’s Economic Rise.” Studies in Comparative International Development 43 (1):1–26. 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). Heilmann, Sebastian, and Elizabeth J. Perry, eds. 2011. Mao’s Invisible Hand: The Political Foundations of Adaptive Governance in China. Cambridge, MA: Harvard University Press. Heimer, Maria. 2006. “The Cadre Responsibility System and the Changing Needs of the Party.” In The Chinese Communist Party in Reform, ed. K. E. Brodsgaard and Z. Yongnian. London and New York: Routledge, pp. 122–38. Hermet, Guy, Richard Rose, and Alain Rouquie. 1978. Elections Without Choice. New York: John Wiley and Sons. Hu, Zhengrong. 2003. “The Post-WTO Restructuring of the Chinese Media Industries and the Consequences of Capitalisation.” The Public 10 (4):19–36. Huang, Da-Qiang. 1993. “Promoting the Quality of the Chinese Government.” In Public Administration in China, ed. M. K. Mills and S. S. Nagel. Westport and London: Greenwood Press, pp. 97–104. Huang, Philip C. C. 2011. “Chongqing: Equitable Development Driven by a ‘Third Hand’?” Modern China 37 (6):569–622. 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):828–43. Huang, Yasheng. 1996. “Central-local relations in China during the reform era: The economic and institutional dimensions.” World Development 24 (4):655–72. Huang, Yasheng. 2002b. “Managing Chinese Bureaucrats: An Institutional Economics Perspective.” Political Studies 50 (1):61–79. Huber, John D., and Charles R. Shipan. 2006. “Politics, Delegation, and Bureaucracy.” In The Oxford Handbook of Political Economy, ed. B. R. Weingast and D. A. Wittman. Oxford: Oxford University Press, pp. 256–72. 236 References

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):325–48. 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, ed. J. Rodden, G. S. Eskeland, and J. I. Litvack. Cambridge, MA: MIT Press, pp. 289–324. 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, MA: Harvard University Press. Klugman, Jeffry. 1989. The New Soviet Elite: How They Think and What They Want. New York and Westport: Praeger. Koh, Byung Chul. 1989. Japan’s Administrative Elite. Berkeley: University of California Press. Kornai, Janos. 1992. The Socialist System: The Political Economy of Communism. Princeton, NJ: Princeton University Press. 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. Laegreid, Per, and Paul G. Roness. 1999. “Administrative Reform as Organized Attention.” In Organizing Political Institutions: Essays for Johan P. Olsen, ed. M. Egeberg and P. Laegreid. Oslo: Scandinavian University Press, pp. 301–29. 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, ed. K. G. Lieberthal and D. M. Lampton. Berkeley: University of California Press, pp. 33–58. 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):535–7. References 237

Landau, Martin. 1969. “Redundancy, Rationality, and the Problem of Duplication and Overlap.” Public Administration Review 29 (4):346–58. 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. Landry, Pierre F. 2007. Decentralized Authoritarianism in China: The Communist Party’s Control of Local Elites in the Post-Mao Era. Cambridge, NY: Cambridge University Press. Lazarev, Valery. 2005. “Economics of One-Party State: Promotion Incentives and Support for the Soviet Regime.” Comparative Economic Studies 47 (2):346–63. Lazarev, Valery. 2007. “Political Labor Market, Government Policy, and Stability of a Non-democratic Regime.” Journal of Comparative Economics 35 (3):546–63. Lee, Hong Yung. 1983. “China’s 12th Central Committee: Rehabilitated Cadres and Technocrats.” Asian Survey 23 (6):673–91. 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]. Collected Works. Translated by J. Fineberg and G. Hanna. 4th ed. Vol. 5, May 1901–February 1902. 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. 2001a. “Manufacturing Meritocracy: Adult Education, Career Mobility, and Elite Transformation in Socialist China.” Sociology, Stanford University, 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):1371–408. Li, Cheng. 2001b. China’s Leaders: The New Generation. New York and Oxford: Rowman and Littleﬁeld. 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):24–8. Li, Cheng. 2007b. “China’s Leadership, Fifth Generation.” Brookings Institution, www.brookings.edu/articles/2007/12_china_li.aspx (accessed March 11, 2015). 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, ed. C. Li. Washington, DC: Brookings Institution Press, pp. 98–117. Li, Cheng. 2012. “The End of the CCP’s Resilient Authoritarianism? A Tripartite Assessment of Shifting Power in China.” China Quarterly 211:595–623. 238 References

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):371–99. Li, David D. 1998. “Changing Incentives of the Chinese Bureaucracy.” American Economic Review 88 (2):393–7. Li, Kangtai. 1993. “Administrative Law in China.” In Public Administration in China, ed. M. K. Mills and S. S. Nagel. Westport and London: Greenwood Press, pp. 81–93. Li, Wei. 1994. The Chinese Staff System: A Mechanism for Bureaucratic Control and Integration. Berkeley: Center for Chinese Studies. Li, Xiaoping. 2001c. “Signiﬁcant Changes in the Chinese Television Industry and Their Impact in the PRC: An Insider’s Perspective.” Center for Northeast Asian Policy Studies, The Brookings Institution, Washington, DC. 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, NJ: Princeton University Press. Liebman, Benjamin L. 2006. “Innovation Through Intimidation: An Empirical Account of Defamation Litigation in China.” Harvard International Law Journal 47 (1):33. Lin, Yi-min. 2001. Between Politics and Markets: Firms, Competition, and Institutional Change in Post-Mao China. Edited by M. Granovetter. Cambridge, UK: Cambridge 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):255–79. 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. Lo, Leslie Nai-kwai. 1993. “The Changing Educational System: Dilemma of Disparity.” Chapter 22. In China Review 1993, ed. J. C. Yu-shek and M. Brosseau. Hong Kong: Chinese University Press. Lorentzen, Peter L. 2013. “Regularized Rioting: Permitting Public Protest in an Authoritarian Regime.” Quarterly Journal of Political Science 8 (2):127–58. Lu, Xiaobo. 2000. Cadres and Corruption: The Organizational Involution of the Chinese Communist Party. Stanford: Stanford University Press. Lu, Xinyu. 2011. “Government Subsidies, Market Socialism, and the ‘Public’ Character of Chinese Television.” Modern China 37 (6):661–71. 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. References 239

Lust-Oker, Ellen. 2005. Structuring Conﬂict in the Arab World: Incumbents, Opponents, and Institutions. Cambridge, UK: Cambridge University Press. Lust-Okar, Ellen. 2006. “Elections under Authoritarianism: Preliminary Lessons from Jordan.” Democratization 13 (3):456–71. Lynch, Daniel C. 1999. After the Propaganda State: Media, Politics, and “Thought Work” in Reformed China. Stanford: Stanford University Press. MacFarquhar, Roderick. 1971. “On Photographs.” China Quarterly 46:289–307. Magaloni, Beatriz. 2008. “Credible Power-Sharing and the Longevity of Authoritarian Rule.” Comparative Political Studies 41 (4–5):715–41. Magaloni, Beatriz, and Ruth Kricheli. 2010. “Political Order and One-Party Rule.” Annual Review of Political Science 13:123–43. Manion, Melanie. 1985a. “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):736–48. March, James G. 1994. “The Evolution of Evolution.” In Evolutionary Dynamics of Organizations, ed. J. A. C. Baum and J. V. Singh. New York and Oxford: Oxford University Press, pp. 39–49. Massey, Andrew. 1993. Managing the Public Sector: A Comparative Analysis of the United Kingdom and the United States: Aldershot, UK: 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):165–79. McGregor, Richard. 2010. The Party: The Secret World of China’s Communist Rulers. New York: Harper Collins. 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, Frank S. 1961. The Moulding of Communists: The Training of the Communist Cadre. New York: Harcourt, Brace and Company. Milgrom, Paul, and John Roberts. 1992. Economics, Organization, and Management. Englewood Cliffs, NJ: Prentice Hall. Miller, Alice L. 2008. “Institutionalization and the Changing Dynamics of Chinese Leadership Politics.” In China’s Changing Political Landscape: Prospects for Democracy, ed. C. Li. Washington, DC: Brookings Institution Press, pp. 61–79. Miller, Gary J. 1992. Managerial Dilemmas: The Political Economy of Hierarchy. Cambridge, UK: Cambridge University Press. 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. 240 References

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. Moock, Peter R. 1999. “Improving Education in a Transforming Country.” In Market Reform in Vietnam: Building Institutions for Development, ed. J. I. Litvack and D. A. Rondinelli. Westport, CT and London: Quorum Books, pp. 95–111. Morgan, Maria Chan. 1981. “Controlling the Bureaucracy in Post-Mao China.” Asian Survey 21 (12):1223–36. Munro, Donald J. 1971. “The Malleability of Man in Chinese Marxism.” China Quarterly 48:609–40. 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, ed. K. G. Lieberthal and D. M. Lampton. Berkeley: University of California Press, pp. 245–79. Naughton, Barry. 2007. The Chinese Economy: Transitions and Growth. Cambridge, MA 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, UK: 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):253–96. 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, NJ and Oxford: Princeton University Press. O’Brien, Kevin J. 1990. Reform Without Liberalization: China’s National People’s Congress and the Politics of Institutional Change. Cambridge, UK: Cambridge University Press. O’Brien, Kevin J., and Lianjiang Li. 1999. “Campaign Nostalgia in the Chinese Countryside.” Asian Survey 39 (3):375–93. 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. References 241

Oi, Jean C. 1998a. “The Collective Foundation for Rapid Rural Industrialization.” In Cooperative and Collective in China’s Rural Development, ed. E. B. Vermeer, F. N. Pieke, and W. L. Chong. Armonk, NY: M.E. Sharpe, pp. 93–109. Oi, Jean C. 1998b. “The Evolution of Local State Corporatism.” In Zouping in Transition: The Process of Reform in Rural North China, ed. A. G. Walder. Cambridge, MA and London: Harvard University Press, pp. 35–61. 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, ed. W. C. Kirby. Stanford: Stanford University Press, pp. 264–84. Oi, Jean C., Kim Singer Babiarz, Linxiu Zhang, Renfu Luo, and Scott Rozelle. 2012. “Shifting Fiscal Control to Limit Cadre Power in China’s Townships and Villages.” China Quarterly 211:649–75. Oi, Jean C., and Scott Rozelle. 2000. “Elections and Power: The Locus of Decision-Making in Chinese Villages.” China Quarterly 162:513–39. Oksenberg, Michel. 1974. “Methods of Communication within the Chinese Bureaucracy.” China Quarterly 57:1–39. Oliver, Christine. 1991. “Strategic Responses to Institutional Processes.” The Academy of Management Journal 16 (1):145–79. Olson, Mancur. 1993. “Dictatorship, Democracy, and Development.” The American Political Science Review 87 (3):567–76. Painter, Martin. 2003. “Public Administration Reform in Vietnam: Problems and Prospects.” Public Administration and Development 23:259–71. Paltiel, Jeremy. 1990. “Administrative Reform and the Politics of Management Training in the People’s Republic of China.” Canadian Public Administration 33 (4):584–604. Parkin, Frank. 1971. Class Inequality and Political Order: Social Stratiﬁcation in Capitalist and Communist Societies. New York and Washington, DC: Praeger Publishers. Pearl, Judea. 2000. Causality: Models, Reasoning, and Inference. Cambridge, UK 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, ed. C. Li. Washington, DC: Brookings Institution Press, pp. 229–50. Pfeffer, Jeffrey, and Gerald Salancik. 1978. The External Control of Organizations. New York: Harper & Row. Pieke, Frank N. 2006. “The Changing Mode of Reproduction of the Chinese Party-State: Party Schools and Cadre Training in Contemporary China.” In 105th American Anthropological Association Annual Meeting. San Jose, CA. Pieke, Frank N. 2008. “Marketization and Centralization of Cadre Education and Training.” In Reinventing the Chinese Party-state. Shenyang Normal University. Pieke, Frank N. 2009. The Good Communist: Elite Training and State Building in Today’s China. Cambridge, UK: Cambridge University Press. 242 References

Price, Jane L. 1976. Cadres, Commanders, and Commissars: The Training of the Chinese Communist Leadership, 1920–45. Boulder: Westview Press. 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, UK: 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):510–9. Putnam, Robert D. 1976. The Comparative Study of Political Elites. Englewood Cliffs, NJ: Prentice-Hall. Qian, Yingyi. 2003. “How Reform Worked in China.” In In Search of Prosperity: Analytic Narratives on Economic Growth, ed. D. Rodrik. Princeton, NJ and Oxford: Princeton University Press, pp. 297–333. 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.” American Economic Review 84 (2):271–5. Rodden, Jonathan. 2004. “Comparative Federalism and Decentralization: On Meaning and Measurement.” Comparative Politics 36 (4):481–500. Rohlen, Thomas P. 1988. “Education in Japanese Society.” In Inside the Japanese System: Readings on Contemporary Society and Political Economy, ed. D. I. Okimoto and T. P. Rohlen. Stanford: Stanford University Press, pp. 25–31. Rosenbaum, Paul R., and Donald B. Rubin. 1983. “The Central Role of the Propensity Score in Observational Studies for Causal Effects.” Biometrika 70 (1):41–55. Rostow, W. W. 1960. The Stages of Economic Growth: A Non-Communist Manifesto. Cambridge, UK: Cambridge University Press. Rowen, Henry S. 2011. “China: Big Changes Coming Soon.” Policy Review (170). Rubin, Donald B. 1973. “Matching to Remove Bias in Observational Studies.” Biometrics 29 (1):159–83. Schoenhals, Michael. 1991. “The 1978 Truth Criterion Controversy.” China Quarterly 126:243–68. Scholz, John T., Jim Twombly, and Barbara Headrick. 1991. “Street-Level Political Controls Over Federal Bureaucracy.” American Political Science Review 85 (3):829–50. 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. 2003. Modernizing China’s Military: Progress, Problems, and Prospects. Berkeley: University of California Press. Shambaugh, David. 2008a. China’s Communist Party: Atrophy and Adaptation. Washington, DC: Woodrow Wilson Center Press and University of California Press. References 243

Shambaugh, David. 2008b. “Learning from Abroad to Reinvent Itself: External Influences on Internal CCP Reforms.” In China’s Changing Political Landscape: Prospects for Democracy, ed. C. Li. Washington, DC: Brookings Institution Press, pp. 283–301. Shambaugh, David. 2008c. “Training China’s Political Elite: The Party School System.” China Quarterly 196 (1):827–44. Shi, Li. 2000. “Efficiency and Redistribution in China’s Revenue Sharing System.” In Governance, Decentralization and Reform in China, India and Russia,ed. J.-J. Dethier. Boston and London: Kluwer Academic Publishers, pp. 103–22. 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. Shih, Victor, Christopher Adolph, and Mingxing Liu. 2012. “Getting Ahead in the Communist Party: Explaining the Advancement of Central Committee Members in China.” American Political Science Review 106 (1):166–87. Shirk, Susan L. 1993. The Political Logic of Economic Reform. Berkeley: University of California Press. Shirk, Susan L. 2007. China: Fragile Superpower. New York and Oxford: Oxford University Press. Simon, Herbert A. 1979. “Rational Decision Making in Business Organizations.” 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. Smith, Peter H. 1979. Labyrinths of Power: Political Recruitment in Twentieth- Century Mexico. Princeton, NJ: Princeton University Press. Solnick, Steven L. 1996. “The Breakdown of Hierarchies in the Soviet Union and China: A Neoinstitutional Perspective.” World Politics 48 (2):209–38. Solnick, Steven L. 1998. Stealing the State: Control and Collapse in Soviet Institutions. Cambridge, MA: 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):355–74. Stiglitz, Joseph E. 1975. “The Theory of ‘Screening,’ Education, and the Distribution of Income.” American Economic Review 65 (3):283–300. 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):137–58. Tang, Peter S. H. 1961. The Training of Party Cadres in Communist China, Vol. 3. Washington, DC: Research Institute on the Sino-Soviet Bloc Studies. 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, ed. E. Lust-Okar and S. Zerhouni. Boulder and London: Lynne Rienner, pp. 51–74. 244 References

Ting, Michael M. 2003. “A Strategic Theory of Bureaucratic Redundancy.” American Journal of Political Science 47 (2):274–92. 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. Tsai, Kellee S. 2006. “Adaptive Informal Institutions and Endogenous Institutional Change in China.” World Politics 59 (1):116–41. 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. UN Division for Public Administration and Development Management, Department of Economic and Social Affairs. 2005. “Lao People’s Democratic Republic: Public Administration Country Profile.” Vogel, Ezra F. 1967. “From Revolutionary to Semi-Bureaucrat.” China Quarterly 29:36–60. Walder, Andrew G. 1995. “Career Mobility and the Communist Political Order.” American Sociological Review 60 (3):309–28. Walder, Andrew G. 2003. “Elite Opportunity in Transitional Economies.” American Sociological Review 68 (6):899–916. Walder, Andrew G. 2004. “The Party Elite and China’s Trajectory of Change.” China: An International Journal 2:189–209. 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. 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: gua- nyu dangdai shijie zhongda wenti), ed. 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), ed. T. Shi. Beijing: Central Party School Publishing House, pp. 96–118. Wang, Qun. 2010. “Administrative Reform in China: Past, Present, and Future.” Southeast Review of Asian Studies 32:100–19. 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. References 245

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, ed. J. C. Oi, S. Rozelle, and X. Zhou. Stanford: Walter H. Shorenstein Asia-Paciﬁc Research Center, pp. 117–42. 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):167–93. 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, ed. B. J. Naughton and D. L. Yang. New York: Cambridge University Press, pp. 101–19. 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, ed. K. E. Brodsgaard and Z. Yongnian. London and New York: Routledge, pp. 139–56. Wilbur, C. Martin, and Julie Lien-ying How, eds. 1972. Documents on Communism, Nationalism, and Soviet Advisers in China, 1918–1927: Papers Seized in the 1927 Peking Raid. New York: Octagon Books. Williams, Oliver. 1993. “An Outsider’s Perspective.” In Public Administration in China, ed. M. K. Mills and S. S. Nagel. Westport and London: Greenwood Press, pp. 145–53. Williamson, Oliver E. 1989. “Transaction Cost Economics.” In Handbook of Industrial Organization, ed. R. Schmalensee and R. D. Willig. Amsterdam: North-Holland, pp. 136–82. 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. Wright, Joseph. 2008. “Do Authoritarian Institutions Constrain? How Legislatures Affect Economic Growth and Investment.” American Journal of Political Science 52 (2):322–43. 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: gua- nyu dangdai shijie zhongda wenti), ed. 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. 246 References

Yang, Chungui. 2002. On Party School Education (dangxiao jiaoxue lun). Beijing: Central Party School Publishing House. Yang, Dali L. 2001. “Rationalizing the Chinese State: The Political Economy of Government Reform.” In Remaking the Chinese State: Strategies, Society, and Security, ed. C.-m. Chao and B. J. Dickson. London and New York: Routledge, pp. 19–45. Yang, Dali L. 2004. Remaking the Chinese Leviathan: Market Transition and the Politics of Governance in China. Stanford: Stanford University Press. Yang, Dali L. 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. Yongnian, Zheng. 2004. Globalization and State Transformation in China. Cambridge, UK: Cambridge University Press. Yu, Keping. 2014. “CCP Cadre Education and Governance.” Zhejiang Provincial Party School Journal (3). Zammuto, Raymond F., and Kim S. Cameron. 1985. “Environmental Decline and Organizational Response.” In Research in Organizational Behavior, ed. L. L. Cummings and B. M. Staw. Greenwich and London: JAI Press, Inc., pp. 223–62. 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 157:115–41. 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):186–99. Zhao, Yuezhi. 1998. Media, Market, and Democracy in China: Between the Party Line and the Bottom Line. Urbana and Chicago: University of Illinois Press. Zhao, Yuezhi. 2008. Communication in China: Political Economy, Power, and Conflict. Lanham: Rowman and Littlefield. Zheng, Shiping. 1997. Party vs. State in Post-1949 China: The Institutional Dilemma. Cambridge, UK and New York: Cambridge University Press. Zheng, Yongnian. 2006. “The Party, Class, and Democracy in China.” In The Chinese Communist Party in Reform, ed. K. E. Brodsgaard and Z. Yongnian. London and New York: Routledge, pp. 231–60. Zheng, Yongnian. 2010. The Chinese Communist Party as Organizational Emperor: Culture, Reproduction and Transformation. London and New York: Routledge. Zhou, Yuezhi. 2000. “Watchdogs on Party Leashes? Contexts and Implications of Investigative Journalism in Post-Deng China.” Journalism Studies 1(2):577–97. Zhu, Jiangnan. 2008. “Why Are Offices for Sale in China? A Case Study of the Office-Selling Chain in Heilongjiang.” Asian Survey 48 (4):558–79. Index

2001–2005 National Cadre Education and Bureaucrat, 6, 7, 10, 12, 13, 34, 46, 52, 58, Training Plan, 102, 103 59, 79, 80, 81, 87, 101, 105, 107, 112, 2003 China General Social Survey, 12, 60, 140, 176, 177, 179, 180, 181, 191, 64, 65, 66–71, 82, 204 195, 199, See Cadre 2003 Regulations on the Selection of Party Bureaucratic accountability, 192 and Government Ofﬁcials, 47 Bureaucratic entrepreneurialism, 91 2004 CCP Party School Work Trial Bureaucratic transformation, 10, 11, Regulations, 15 60, 183 2006 Cadre Education and Training Trial Regulations, 48, 63, 104 Cadre, 3, 6, 8, 10, 12, 14, 44, 45, 47, 63, 64, 2006–2010 National Cadre Education and 75, 77, 78, 79, 81, 86, 87, 101, 118, Training Plan, 48 154, 165, 168, 175, 176–177, 179, 2010–2020 Cadre Education and Training 180, 189, 191 Reform Plan, 104, 162 career histories, 25, 56, 57, 64, 80, 82, 2013–2017 National Cadre Education and 83, 86, 87, 204 Training Plan, 48 education levels, 31 evaluation, 61 Adaptation ranks, 67 deﬁned, 18 total number in China, 8 Adaptive capacity, 16, 24, 158, 182, 192, Cadre colleges, 46, 104, 109 198 economic management colleges, 106 Adaptive change, 16, 18, 153, 170, 171, Cadre Executive Leadership Academies, 201 46, 108, 109–110, 161, 171, See Adaptiveness, 2, 16, 17, 20, 124, 157, 158, Jinggangshan, Pudong, and Yan’an 176 enrollment, 119 Administration schools, 46, 101, Cadre training, 14, 15, 22, 23, 24, 25, 103, 106–107, 109, 121, 140, 26, 28 171, 203 history, 29–35 Anhui, 39, 192 international providers, 113–115 Asian Tigers, 165 marketization, 25, 26, 90, 104–116, 190, Authoritarian resilience, 4, 200 See Party schools marketization Autocracy, 6, 178 marketization preconditions, 183–184 motivations to diversify, 101–104, 111 Beijing, 29, 105, 106, 108, 118, 175 reform era trends, 46–51 Beijing Institute for International standardization, 34 Studies, 86 system capacity, 101–102 Bolshevik, 92 Cambridge University, 113 Bolsheviks, 5 Capitalism, 11, 30, 173 Building a new socialist countryside, 39, Capitalist, 16, 18, 145, 146 160 CCP, 1, 6, 7, 8, 9, 11, 12, 13, 15, 18, 19, 22, Bureaucracy 24, 27, 28, 29, 30, 31, 46, 47, 50, 52, as organizational weapon, 4, 14 61, 62, 70, 87, 88, 89, 109, 110, 121,

247 248 Index

124, 129, 134, 140, 153, 157, 158, Content analysis, 26 159, 165, 176, 177, 178, 179, 182, Corruption, 3, 49, 53, 114, 147, 191 183, 184, 189, 191, 193, 197, 198, 199 CPS, 42, 43, 83, 84, 136, 137, 161, adaptation, 27, 182, 183, 198–199 162, See Central Party School as learning party, 47, 48, 162, 197–198, Institute for International Strategic 200 Studies, 29 CCTV, 144, 167 Credible commitment, 2 CELA. See Cadre Executive Leadership Cultural Revolution, 10, 29, 32, 52, 53, 59, Academies 71, 118, 128 Central Commission for Discipline and Czech Republic, 11 Inspection, 192 Central Committee, 8, 28, 46, 89, 101, 133, dang’an, 61, 79 134 Decentralization, 8, 26, 90, 92–93, 96, 100, Central Organization Department, 8, 44, 121, 145, 180, 183 46, 48, 86, 89, 104, 108, 113, 138, Deng Xiaoping, 10, 46, 53, 90, 94, 97, 161, 171, 204 100, 124, 136, 144, 145, 162, 163, Central Party School, 25, 28, 29, 31, 33, 34, 180, 183 38, 41, 42, 45, 47, 50, 55, 60, 63, 64, Development and Reform Commission, 85 80, 81, 82, 86, 87, 89, 95, 96, 102, 104, Dual track system, 143 108, 109, 118, 119, 120, 127, 128, 130, 134, 136, 137, 138, 142, 158, Economic Daily, 144, 148 159, 160, 161, 162, 167, 168, 171, e-governance, 116 173, 175, 176, 196, 204, See CPS Elections, 61–62 county magistrate training class, 130, 164 Elite circulation, 71 distance degree program, 128, See Party Embeddedness, 26, 154 schools distance degree programs ex ante political control, 53 enrollment, 118–119 Experiments, 200 Mid-Career Cadre Training Class, 83–87 cadre training, 111–113 training content, 157–167 local, 90 Central Propaganda Department, 95, 144, local policy, 17 148, 151 organizational, 96–97 Changsha, 30 Extrabudgetary revenues, 12 China Development Research Foundation, 113 Five old topics, 33 Chinese bureaucracy Fujian, 118 ranks, 8 reforms under Deng, 9–10 Gansu, 72 size, 8 Governance, 1, 3, 6, 18, 19, 20, 46, 47, 48, turnover, 11–12 52, 62, 88, 153, 155, 165, 177, 179, Chinese Communist Party. See CCP 183, 192, 196, 198 Chinese party-state, 3, 13, 16, 18, 23, 24, Growing out of the plan, 10 32, 46, 54, 79, 84, 87, 92, 135, 182, Guangdong, 39, 56, 113, 120, 136 191, 197, 199, 200 Guomindang, 30, 31, 58 Chongqing, 144 COD. See Central Organization Hainan, 134 Department Harmonious socialism, 29 Command economy, 5, 10, 143 Harmonious society, 28, 29, 159, Communist University of the National 160, 169 Minorities of the West, 30 Harvard University, 113, 114 Communist University of the Toilers of the Heilongjiang, 116, 168 East, 30 Hong Kong, 66, 118 Communist Youth League, 14, 30, 125, Hu Jintao, 29, 33, 47, 109, 159, 162, 152, 181, 203 163, 198 Competition, 19, 180 Hubei, 55, 56, 95, 165 generated by redundancy, 19–22 Hunan, 30, 165, 192 Index 249

Incentives, 3, 12, 14, 15, 16, 17, 18, 19, 20, Modernization, 1, 15, 33, 46, 145, 163, 170 23, 24, 35, 38, 54, 57, 59, 61, 65, 90, bureaucratic, 33, 105, 153, 198 100, 104, 122, 125, 135, 142, 145, economic, 10, 145 153, 154, 155, 158, 176, 180, 182, Monitoring, See Party schools monitoring 183, 184, 185, 186, 187, 191, 192, 194, 197, 199 National Judicial College, 86 Institutions, 1–2 National School of Administration, 29, 95, authoritarian, 1–4 102, 106, 171 Networks, 3, 42, 43, 50, 62, 71, 128, 154, Jiang Zemin, 12, 28, 29, 130, 162, 179 195 Jiangsu, 55 Nomenklatura, 8 Jiangxi, 31, 95, 109 Jinggangshan, 109, 110, 119, 161, 173 Organization department, 38, 39, 41, 42, 103, 107, 111, 112, 116, 125, 136, kaohe, 41, 42 141, 142, 160, 161, 188, 189, 190 cadre training schools, 107–111 Laos, 193 partnerships, 141 cadre training reforms, 193–194 Organizational change, 15, 17, 20, 23, 24, Leading cadres, 7, 8, 44, 45, 50, 60, 61, 71, 25, 52, 122, 155, 157, 158, 200, 203 76, 77, 94, 101, 102, 162, 168, 169, 174, 203 Party congress, 29, 41, 137, 190 Legitimacy, 10, 12, 71, 135, 197 Eighth, 53 Leninist party, 4–7, 11, 18 Seventeenth, 29, 47, 111, 118, 137 Liaoning, 113, 138 Sixteenth, 109, 111 Local state corporatism, 10, 92, 135 Thirteenth, 102, 106 Long March, 31 Party entrepreneurialism, 24, 133, 192, 201 Party organization, 4 Mao Zedong, 13, 30, 52, 92, 136, 145, 159, Party School Work Regulations, 38 161, 162, Party schools, 14–15, 46, 178–180 Market cancellation of degree programs, 133–134 incentives, 27, 91, 92, 121, 143, 155, 195 comparative cases, 193–195 processes, 22–23 comparison to extrabureaucracies, 135 risks, 24 comparison with Chinese media, 148–153 Market transition, 10, 143 comparison with political campaigns, party and state agencies, 91 51–54 Marketization, 16, 22, 92, 121, 124, 142, competition, 104–106, 170–171 147, 153, 155, 185, 189, 191, 199, curriculum, 95–96 200, See Cadre training marketization, degree programs, 75–76, 127–129, 133, Party schools marketization 153 Marxism, 161 distance degree programs, 42, 43, 55, Marxism-Leninism, 13, 30, 34, 159 127, 133, 185 May Fourth Movement, 30 effect on cadre promotion, 62–77, 81–87 Media effects of competition, 120, 170–171 advertising, 144, 145, 146, 147, 148, 150 enrollment, 44–46, 117–118, 126 commercialization, 142–148 entrepreneurial activities, 123–132 investigative journalism, 146, 151 evaluations, 103 Mid-Career Cadre Training Class, 45, funding, 93–95, 97–100, 125–127 50–51, 63, 64, 80, 81, 82, 83, 84, 116, history, 29–35 118, 162, 196 ideological function, 28–29 content, 172–175 inside-the-plan training classes, 43, 185 Military, 3, 20, 31, 68, 72, 162 international competition, 113–115 Ministry of Education, 46, 55, 134 local embeddedness, 187–188 Ministry of Finance, 8, 49 marketization, 15, 22, 90, 185–187, See Ministry of Foreign Affairs, 32 Cadre training marketization Ministry of Personnel, 46, 107 monitoring, 35–42 250 Index

Party schools (cont.) Reforms networks, 195–197 sequencing, 184 non-training income, 130–132 Regime durability, 1, 3, 4, See Authoritarian organization department competition, resilience 107–111 Republican period, 30 outside-the-plan training classes, 43, 97, Resolution on the CCP’s strengthening and 127, 129–130, 185 improving party school work in the partnerships, 138–142 twenty-first century, 15 reputation, 136–138 Russia, 5, 11, 193, See Soviet Union risk of failure, 188–190 training content, 157–170 Screening, 59, 78, 79–81 training content compared, 171–175 Seeking Truth, 144 types of training programs, 42–44 Selection bias, 25, 64, 65 Party state Selection problem, 6, 25, 59, 60, 78, 88, relations, 6 179 Patronage, 3, 61, 62, 71, 72, 78, 79, 87 Self-Education College, 30 kinship-based, 70 Shaanxi, 109 Patron-client ties, 71, See Patronage Shandong, 97, 120 People’s Daily, 144, 148 Shanghai, 95, 109, 110, 118, 140, 172 People’s Liberation Army, 72, 147 Shenzhen, 112, 113, 165 Personnel management, 54, 62, 92, 104, Signaling, 59, 78–79 121, 179, 185, 188 Singapore, 118, 140 Policy winds, 28, 52 Single party regime Political campaigns, 51–54 longevity, 178 Power elite, 7 Single party system, 2–4 Practice as a sole criterion of truth, 28 Socialism schools, 14, 203 Princeling, 62, 71 Southern Tour, 91, 144 Principal-agent, 6, 25, 57–58 Soviet Union, 1, 5, 9, 11, 18, 30, 31, 165, Probit regression, 65, 68–71 184, 193, 199 Profit-sharing contracts, 12 State capacity, 199 Promotion, 8, 15, 25, 34, 44, 56, 57, 58, 60, State Council, 13, 46, 107, 113, 192 62, 63, 64, 65, 70, 71, 77, 78, 80, 81, State Education Commission, 107 82, 86, 87, 88, 105, 128, 149, 154, State-Owned Assets Supervision and 177, 182, 189 Administration Commission, 85 Propaganda, 5, 30, 33, 81, 143, 144, 145, State-owned enterprises, 14, 56, 90, 100, 151, 125, 168 Propensity score matching, 25, 64, 65–66, Study Times, 39, 160 68, 72–75, 78, 79, 88 average treatment effect on the treated Taiwan, 158 (ATT), 65, 75 Three Represents, 12, 28, 33, 112, 133, nearest neighbor matching with 136, 137, 154, 162 replacement, 74 Tiananmen, 106, 159 propensity score block, 73 Training plans, 23, 32, 38, 99, 102, 103, Public office 129, 161, 170 sale of, 192 Tsinghua University, 39, 113, 114, 195 Pudong, 108, 109, 110, 171 Two whatevers, 28

Red Flag, 144 United Front, 85 Redundancy, 15, 19, 21, 22, 24, 171, 180, Universities, 13, 15, 23, 34, 101, 103, 104, 181, 188, 190, 192, 193 105, 109, 118, 119, 121, 128, 129, conditions, 21 136, 138, 141, 142, 157, 170, 189, efﬁciency of, 21, 192–193 190, 194, 203, 204 generating adaptive capacity, 20 networks, 195 reliability, 19 providers of cadre training, 104–106 Reform and opening up, 163, 198 University degree, 76 Index 251

Vietnam, 193, 240, 241 Yan’an, 109, 110, 119, cadre training reforms, 194–195 173 Yunnan, 98, 165, 169 Wang Zhen, 34, 89 WTO, 144, 147, 234, 235 Zeng Qinghong, 47, 157 Zhao Ziyang, 102, 104, Xi Jinping, 29, 48, 53, 114 106 Xinhua, 47, 49, 50, 80, 83, 110 Zhejiang, 55