Chapter 1: Introduction
1.4. Research Design
國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
remove political enemies and establish his unchallengeable status. (Huang, 2014;
Sternberg, 2014; Tiezz, 2014; X. Wang, 2014) But this simple theory of “fighting enemies” – that Xi Jinping purges his rival factions and replaces them with king’s men – falls short in explaining why so many high-ranking cadres outside any of the factional groups are also taken down at all the political cost for the party-state. The ongoing anticorruption campaign is far more than directly taking down enemies. It creates a climate where Xi’s colleagues dare not oppose his power centralization and all regime elites get in line with Xi’s agenda. The factional groups that are cracked down are not only presumed to be in Xi’s way but are also reminders for the audience of other regime elites of what shall happen to those who dare to oppose.
To sum up, Xi Jinping’s anticorruption campaign is a costly yet necessary cause. The prevalent corruption in the party-state erodes legitimacy and can potentially doom the regime, which compels Xi to take decisive measures to respond to public discontent.
The scale and complexity of corruption in CCP ask Xi to make the full use of his institutional power, and it is impossible to push the campaign through without consolidating his status within the party first. Further, Xi Jinping’s ambition of steering the country to realize the “Chinese Dream” (中國夢, zhongguo meng) asks him to adopt massive reform agenda. This cannot happen against the vested interest of regime elites if Xi is confined within the institutionalized power-sharing arrangements in the post-Deng era. Therefore, he needs the unprecedented anticorruption campaign to create a climate in favor of his establishing personal authority so that he can push through his ambitious reform agenda.
1.4. Research Design
This thesis applies quantitative and network strategies to study the early stage of Xi Jinping’s anticorruption campaign. It examines temporal and geographical/functional
‧
successors fill the vacant positions, and draws a colleague network of tigers in order to discuss factional dynamics, if any, in Xi Jinping’s combat against corruption.1.4.1. Unit of Analysis
The unit of analysis of this thesis is individual cadre defined as “tiger”: the corrupt cadres at or above vice-provincial level in the administrative ranking system of CCP11. Starting from December 2012, CCDI puts online the corruption case announcements of the cadres under central management (中管幹部, zhong guan ganbu) and some lower-ranking cadres.12 We compile the list of civilian tigers according to CCDI. As for the military tigers, PLA makes official announcements of corruption cases on its website from time to time13, from which we compile the list of military tigers.
A typical vice-provincial-level position includes but is not limited to:
- Standing member of provincial party committee, provincial vice-governor, vice-chairman of provincial Standing Committee of People’s Congress and CPPCC of the 31 provinces14;
- Party chief, mayor, chairman of municipal Standing Committee of People’s Congress and CPPCC of the 15 vice-provincial-level municipalities;
- Vice-minister, deputy director and deputy head of full-ministerial-level units directly under the State Council and Central Committee;
- Director of vice-ministerial-level units directly under the State Council and Central Committee;
11 For an explanation of relevant administrative rank levels in CCP, see Appendix 4.
12 The announcements are in the section of “Jilv shencha (紀律審查, Disciplinary Inspection)” on the Zhongyang jiwei jianchabu wangzhan ( 中 央 紀 委 監 察 部 網 站 , the website of CCDI/Ministry of Supervision), at http://www.ccdi.gov.cn/jlsc/
13 See Zhongguo junwang (中國軍網, 81.cn), at http://www.81.cn/
14 A “province” in this thesis, unless specified otherwise, refers to a provincial-level local unit that can either be a province, an autonomous region or a municipality directly under the State Council. The People’s Republic of China claims 34 provincial units, but 3 of them are not included in this thesis. The two Special Administrative Regions, Hong Kong and Macau, are outside the direct governance of CCP. The alleged province of Taiwan is not included either for similar reasons.
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
- Deputy director of full-ministerial-level offices of leading or coordination groups directly under the State Council and Central Committee;
- Deputy person-in-charge of full-ministerial-level mass organizations; and
- Deputy commander and commissar of field armies and provincial military districts;
etc.15
Cadres of these positions and even higher-ranking cadres shall be counted as tigers if they are put under disciplinary inspection.
1.4.2. Time Scope
This thesis studies the early stage of the anticorruption campaign launched by Xi Jinping. The study covers the period from November 2012 to March 2015. November 2012 is the beginning of Xi’s rule when he was elected General Secretary of the party and Chairman of CMC. It is self-explanatory to choose this month to be the starting point. The selection of March 2015 as the cutting point, however, deserves a brief discussion.
As this thesis is to study the early stage of Xi Jinping’s rule when new patterns of power dynamics are being set for the entire 10-year tenure, we need to find a cutting point that marks the end of the initial power-consolidating stage. There are two possible ways to identify an appropriate cutting point. First, we list all the significant positions that Xi Jinping occupies that grant him outstanding institutional powers and make him the paramount leader he is. The cutting point of the early stage of his tenure should be no earlier than the moment when he takes the last significant position.
Second, the proclamation of his pivotal programs can be seen as the end of his early tenure. It takes time for a new autocrat to get accustomed of ruling the country and explore the direction where he would like to lead the people. When he comes up with a blueprint of fundamental programs, he ends the period of uncertainties and
15 See Appendix 2 for the provincial-level institutions relevant to the tigers in this thesis.
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
proclaims himself in full command of the governance apparatus.
It turns out that March 2015 is a significant juncture that appropriately marks the end of Xi’s early tenure. First, by this time Xi had assumed offices of all the major positions that grant him autocratic powers. He was elected General Secretary of CCP and Chairman of CMC in November 2012, became the President of the People’s Republic of China in March 2013, found and chaired the Central Leading Group for Comprehensively Deepening Reform and the National Security Commission in January 2014, and found and chaired the CMC Leading Group for Deepening National Defense and Military Reform in March 2014. The position of General Secretary of the party, chairmanship of CMC and presidency of the state compose the trinity of supreme power in post-Deng China, and the three newly founded coordinating institutions are critical to Xi’s furthering his autocracy beyond Jiang Zemin and Hu Jintao.
Second, the blueprint of Xi’s governance, summarized as the “Four Comprehensives”
(“Renmin ribao”, 2015), was proclaimed in February 2015. The “Four Comprehensives” are 1) to comprehensively build a moderately prosperous society (全面建成小康社會); 2) to comprehensively deepen reform (全面深化改革); 3) to comprehensively govern the nation according to law (全面推進依法治國); and 4) to comprehensively strictly govern the Party ( 全 面 從 嚴 治 黨 ). Each of the four programs had been propagandized in previous campaigns separately, but in early 2015 people were told for the first time that these four programs together formed an organic whole and pointed to the direction of the “Chinese Dream”. The “Four Comprehensives” unsurprisingly set the tone for the national Two Sessions16 in the following March.
16 The “Two Sessions” refer to the annual meetings of People’s Congress and CPPCC. These two bodies are the institutions of legislation and political consultation in China and serve functions similar to those of the two houses of parliament in other countries.
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
Therefore, the national Two Sessions in March 2015 are the first ones after Xi assumed all his significant institutional powers and proclaimed his pivotal programs.
It is a time when uncertainties ended and equilibrium under the new autocrat emerged.
Xi had consolidated his power and set the new normal for intra-party power dynamics by that time. We find it a theoretically significant cutting point to close the time scope of this study.
1.4.3. Tiger Distribution and Succession
This thesis illustrates and analyzes temporal distribution, geographical/functional distribution, and succession of tigers. The discussion of these frequencies clarifies the correlation between tiger hunt and Xi’s power consolidation. Temporal distribution studies how many tigers are hunted in each month during the time scope. In this section, we shall mark down the key events of Xi’s power consolidation in the timeline and compare the trend of tiger frequency with the distribution of these events.
The key events include Xi’s taking on significant positions and proclaiming each of his pivotal programs.
Geographical/functional distribution sees the career tracks of the fallen tigers, i.e. in what provinces, central ministries, state-owned enterprises (SOEs), etc. a corrupt senior cadre has served. The fighting corruption interpretation of Xi Jinping’s anticorruption campaign leads us to expect that the tigers hunted down should be scattered widely across geographical and functional units of the governing system.
That said, some provinces and institutions may have disproportionately more tigers than others because they are the so-called “nest cases”, i.e. the majority of its governing members are compromised (“Jinnian fubai”, n.d.).17 A simple version of fighting enemies interpretation, on the contrary, should predict that the tiger hunt does
17 For analyses of nest cases, see Sun (2004), Tu (2012) and Yang (2013).
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
not cover many provinces and institutions because Xi would seek to minimize the cost of anticorruption and lock down his rival factions only. The thesis pays special attention to the position a tiger holds when he is brought down for disciplinary inspection, the last position he serves before his retirement, or the province/institution where he most significantly builds up his career.
The succession of tigers refers to who fill in the vacancies, especially whether the successor is promoted or transferred from within the given province/institution or as an outsider – cadres who do not work in the given province/institution before he is moved there. If the anticorruption campaign is designed to directly strike down Xi’s enemies and replace them with his own men, we would expect to see that outsider successors are significantly more than insiders. Xi would make sure that his loyalists fill in the vacancies left by tigers, so loyalty to Xi prevails working experience in the given province/institution. Xi is a new autocrat who has just risen to supremacy lately and cannot easily find successors he trusts within all given provinces/institutions. He would find someone from outside.
1.4.4. Colleague Network of Tigers
This thesis draws a network of tigers in order to identify if there are any factional groups among the senior corrupt cadres. We use colleague relations between cadres to establish ties in the sociogram, following a pioneer research of network analysis that studies CCP elites by Shilin Jia (2014). A tie between two tigers is coded only if they have worked or are working in the same province/institution at the same time. During their colleagueship, there can be no more than two administrative rank levels between their positions. Detailed coding methods are presented in Chapter 3.
There are two reasons why we use colleague relations. The first is theoretical. Faction is classically defined on clientelist ties that should be dyadic and unequal. (Nathan,
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
1973) Colleague relations with identifiable rank levels fit this requirement. It is to be noted, though, that Andrew Nathan (1973) describes clientelist ties as “trellis” on formal bureaucracy but he does not equate colleague relations with clientelist ties. If the alleged factional follower is also a bureaucratic subordinate, we cannot observably tell the difference between a client following his patron’s directives and a bureaucratic subordinate following his superordinate’s orders.
Tang Tsou (1976) criticizes Nathan for underestimating the role of formal bureaucracy.
Tsou argues that colleagueship itself often constitutes factional ties. We follow this thread of discussion initiated by Nathan and Tsou and continued by many, e.g. Lowell Dittmer (1995) and Lucian Pye (1995), and define factional ties on colleagueship. If two cadres have an unequal colleague relation in the past, the tie is no longer bureaucratically directive for the subordinate but this past relation can be the basis of a clientelist tie. If two cadres were working together in the same province/institution when put under inspection, they are legitimate colleagues who fit in Tsou’s (1976) conception of “informal group”. If two cadres are equal colleagues in a province/institution, they can be factional allies and thus be coded with a potential factional tie. Many scholars simply adopt various types of shared identity as factional relationship. (See, for example, Shih, 2004; Li, 1994; 2004; 2013; Bo, 2006; Kou, 2001; 2007; Li & Pye, 1992; Tsai & Dean, 2015) This thesis does not take this approach for the theoretical reasons laid above and the operational reason as follows.
The second reason is operational. Factional ties, if loosely defined, can be messy. The types of guanxi (關係, relationship) are beyond counting, but most of them do not have accessible and reliable data. Colleague relations, however, are reliably recorded in the resumes of cadres that are readily accessible from CCP official sources.
It is to be noted, of course, that colleague relations do not necessarily make factional
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
ties. In a strict sense, they only provide cadres with the chance of frequent and meaningful contact. Colleagues may as well develop a relationship of rivalry and confrontation as one of exchange and alliance. Whether a circle of cadres tied by their working experience is an actual faction, as a functional action group, begs further examination with caution.
After we code colleague relations, centrality measurement assists us with the identification of factional groups in the network. The classical definition of faction requires the patron to be linked with all other members and the subordinates are barely linked with each other. (Nathan, 1973) Such a theoretical requirement is difficult to be fulfilled in reality, but statistics that derive from network analysis can help us determine whether a cadre is truly a patron, and thus whether a group of cadres makes a faction. The centrality indices (de Nooy, Mrvar & Batagelj, 2005) of the presumed patron, the highest-ranking tiger in a circle, should be higher than those of other circle members. The index at use in this thesis is PageRank, a powerful algorithm developed by Google engineers to calculate relative importance of web pages (Brin & Page, 1998). The theoretical implications of centrality and the algorithm of PageRank are discussed in detail in Chapter 3.
1.4.5. Source of Data
The data of this thesis are collected from a variety of public sources of CCP, the Chinese government, official news agencies, leading newspapers and a number of new media both in China and abroad. For personal backgrounds and career tracks of corrupt officials, this thesis relies on the “Chinese Political Elites Database” run by Chien-wen Kou at the National Chengchi University18 and the “Leading Cadres of CPC and Chinese Government Database” maintained by the official website of CCP19.
18 See “Zhonggong zhengzhi jingying ziliao ku (中共政治菁英資料庫, Chinese Political Elites Database)”, at http://cped.nccu.edu.tw
19 See “Zhongguo dangzheng lingdao ganbu ziliao ku (中國黨政領導幹部資料庫, Leading Cadres of CPC and
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
The database created by Victor Shih, Wei Shan, and Mingxing Liu20 is for reference as well as others.