Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Urbanization as state building : a case study of governance reform in the Guangzhou Luogang District,… Wong, Siu Wai 2013

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Notice for Google Chrome users:
If you are having trouble viewing or searching the PDF with Google Chrome, please download it here instead.

Item Metadata


24-ubc_2014_spring_wong_siuwai.pdf [ 3.73MB ]
JSON: 24-1.0165669.json
JSON-LD: 24-1.0165669-ld.json
RDF/XML (Pretty): 24-1.0165669-rdf.xml
RDF/JSON: 24-1.0165669-rdf.json
Turtle: 24-1.0165669-turtle.txt
N-Triples: 24-1.0165669-rdf-ntriples.txt
Original Record: 24-1.0165669-source.json
Full Text

Full Text

 URBANIZATION AS STATE-BUILDING: A CASE STUDY OF GOVERNANCE REFORM IN THE GUANGZHOU LUOGANG DISTRICT, CHINA  by    SIU WAI WONG   B.Sc., Hong Kong Polytechnic University, 1998 Master of DevPlan, University of Queensland, 2004     A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  DOCTOR OF PHILOSOPHY  in  THE FACULTY OF GRADUATE AND POSTDOCTORAL STUDIES  (Planning)   THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver)     November 2013   ? Siu Wai Wong, 2013   ii  Abstract   Over the past three decades, China?s urban population has increased by more than 400 million. How could this rapid, unprecedented process of urban transformation come about without causing widespread social upheavals in socialist China? With an in-depth case study of the Guangzhou Luogang District, this dissertation takes a micro-historical approach to understand this swift urbanization process and its impact on the transformation of rural society as well as the restructuring of local governance. By examining how the local state, village organizations, villagers and other non-state actors have dealt with their land-based conflicts and struggles arising from state-led urbanization, this study argues that the requisition of rural land for urban expansion was not merely a process of land grabbing by corrupt and unscrupulous local state to facilitate economic growth. Rather, it should also be interpreted as a process of state building, whereby the local government reshaped its governance strategy so as to mitigate potential social unrest and to strengthen its legitimacy in the eyes of villagers. The findings of this study will shed new light on how and why socialist China has been able to maintain such unparalleled and unprecedented rates of urban growth despite incessant land disputes and numerous social tensions at different localities. It is also hoped that this research will arouse broader intellectual discourse on governance restructuring and local government, specifically about the relationships between power decentralization, public-private partnership, citizen empowerment and civic engagement in governance strategies that attempt to realize collective interests under the present epoch of economic globalization and neo-liberalization.   iii  Preface   This dissertation is an original intellectual product of the author, Siu Wai Wong.  The field research undertaken was covered by the UBC Ethics Certificate number H12-01320.    iv  Table of Contents   Abstract???????????????????????????????????ii Preface ????????.. ......................................................................................................... iii Table of Contents .......................................................................................................................... iv List of Tables ................................................................................................................................. vi List of Figures ............................................................................................................................... vii Notes on Weight, Measures and Currency ............................................................................... viii List of Characters .......................................................................................................................... ix Acknowledgements ........................................................................................................................ xi 1. Introduction ................................................................................................................................ 1 The Puzzle of State Power Decentralization in Urbanizing China .............................................. 1 Research Question ........................................................................................................................ 2 Reflections on Theoretical and Methodological Governance Research ...................................... 5 Governance as an Analytical Approach ..................................................................................... 13 Conceptual Framework of Governance Analysis....................................................................... 15 Governance Analysis in the Chinese Context ............................................................................ 17 Case Study: the Guangzhou Luogang District ........................................................................... 23 Dissertation Chapter Outline ...................................................................................................... 25 2. Case Study of Guangzhou Luogang District:  Research Design and Field Research ........ 28 GLD: A New Urban District of Guangzhou City ...................................................................... 28 Studying Urbanizing Villages (New Urban Neighbourhoods) as a Method of Inquiry ............. 31 Conceptual Framework for the Neighbourhood Governance Analysis ..................................... 34 Analytical Framework for the Analysis of Local Governance in the GLD ............................... 37 Fieldwork: Stage One ................................................................................................................. 41 Fieldwork: Stage Two ................................................................................................................ 49 3. Guangzhou Luogang District?s Rural-urban Transition ..................................................... 59 Historical Origin of the GLD ..................................................................................................... 59 Village Society and the Urban-rural Continuum........................................................................ 64 Agriculture Collectivization in the Maoist Era .......................................................................... 70 Collectivization and Rural-Urban Dichotomy ........................................................................... 75 De-collectivization and Re-collectivization in the Reform Era ................................................. 79 From Development Zone to Urban District ............................................................................... 82 4. Land Requisitions, Farmer Resettlement and Local Land Politics ..................................... 90 Hukou as Part of Compensation in the 1980s ............................................................................ 91 ?Reserved Commercial Land? as a Policy of Compensation in the 1990s ................................ 94 Villagers? Resistance .................................................................................................................. 98 Compromise with Villagers ..................................................................................................... 102 Retreat of the Local State and Village Autonomy ................................................................... 105 State versus Village Collectives, Collectives versus Villagers ................................................ 110 v  5. Shareholding Reforms and State-village Power Restructuring ......................................... 112 Purposes of Shareholding Reforms .......................................................................................... 112 Background of Shareholding Reforms in the 1980s ................................................................ 114 Formalities versus Informalities: Implementing Shareholding Reforms in Villages ............... 120 Problems of Shareholding Reforms in the 1990s ..................................................................... 124 ?Fixing Share Rights? in 2003 ................................................................................................. 127 Shareholding Reforms and Social Control ............................................................................... 129 Shareholding Reforms and Governing Capacity Building ....................................................... 132 6. Grassroots Elections and Social Participation..................................................................... 136 The Reach of the State under ?Two Committees, One Association? ....................................... 137 Transparency and Autonomy in Neighbourhood Elections ..................................................... 141 Grassroots Democracy ............................................................................................................. 144 Exclusion of Migrant Workers ................................................................................................. 146 Landlord Villagers versus Tenant Migrant Workers ................................................................ 148 Inclusion of Migrant Workers .................................................................................................. 150 The Increasing Influence of Migrant Workers ......................................................................... 154 7. Conclusions: Building the Local State .................................................................................. 158 Regulatory and Conciliatory Local State Building .................................................................. 158 Central-Local State Relationships and State-Society Separation in Reform China ................. 161 Local Autonomy and Centralism ............................................................................................. 168 Rural-urban (In)equality .......................................................................................................... 171 Privatization of Collective Land: a Solution or a Problem? ..................................................... 172 State-village (shequ) Power Restructuring for and against Self-governance ........................... 174 Urban Prospects of Urbanizing Villages .................................................................................. 177 Community Planning of Urbanizing Villages .......................................................................... 179 Some Reflections on Urban Studies in China .......................................................................... 181 Bibliography? ........................................................................................................................... 185 Appendix I   Some Statistics about the GLD ........................................................................... 208 Appendix II  An Overview of Informants ................................................................................ 210 Appendix III   Interview Guide................................................................................................. 211    vi  List of Tables   Table 1.1   Definitions of Good Governance ................................................................................. 11 Table 2.1   A Overview of New Urban Neighbourhoods (Urbanizing Villages) ........................... 33 Table 3.1   Collectivization and Collective Organizations ............................................................. 73 Table 3.2   Important Regulations on Rural-urban Migration: 1953-1958 .................................... 75 Table 3.3   National Development Zones in China ........................................................................ 83 Table 4.1   Snapshot of Land Requisitions and Compensation in the GLD: 1984-1988 ............... 92 Table 4.2   Snapshot of Land Requisition and Land Compensation: 1991-2000 ........................... 95 Table 5.1   Organizational Features of Shareholding Cooperatives in the GLD .......................... 123 Table 6.1   Neighbourhood Size and Party Organization ............................................................. 138 Table 6.2   Neighborhood Size and the Resident Committee ...................................................... 138 Table 6.3   The GLD Party Organizations (at and below the Street Level) ................................. 140    vii   List of Figures   Figure 1.1   Hierarchy of Urban-rural Governance ............................................................................. 3 Figure 1.2   Conceptual Framework for Governance Analysis ......................................................... 16 Figure 1.3   Guangzhou City and Its Districts ................................................................................... 27 Figure 2.1   Local Governance Hierarchy of the GLD ...................................................................... 32 Figure 2.2   Local Governance Structure in GLD: Key Actors and Institutions ............................... 35 Figure 2.3   Analytical Framework for Local Governance Structure and Practice in GLD .............. 38 Figure 2.4   Map of GLD .................................................................................................................. 54 Figure 2.5   New Urban Neighbourhoods in the GLD ...................................................................... 55 Figure 2.6   ?Kissing? and ?Shaking-hand?  Buldings ...................................................................... 56 Figure 2.7   Development of ?Reserved Commercial Land? ............................................................ 57 Figure 2.8   Job opportunities for Villagers ...................................................................................... 57 Figure 2.9   Residents? Committees .................................................................................................. 58 Figure 3.1   Traditional Landscape Features of Urbanizing Villages ............................................... 88 Figure 3.2   Guangzhou Science City ................................................................................................ 89 Figure 3.3   Knowledge City ............................................................................................................. 89 Figure 4.1   Land Requisitions Procedures in the GLD during the 1990s ........................................ 97 Figure 4.2   Collective Ownership and Governance Bureaucracy in Rural China .......................... 106 Figure 4.3   Rescaling of Collective Ownership in the GLD .......................................................... 108 Figure 5.1   Local State-society Relationships in Periurban China: Regional Diversity ................ 119 Figure 6.1   ?Two committees, one Association? (Liangwei yishe) ................................................ 138 Figure 7.1   Governance Transformation in China since 1949 ....................................................... 184    viii  Notes on Weight, Measures and Currency    1 catty = 500 grams or 1.1 pound  1 mu =  1/15 hectare or 666.67 square meters   1 yuan =  US$0.41- 0.44 from 1969 to 1973   US$0.50 - 0.60 from 1974 to 1978 US$0.64 - 0.50 from 1979 to 1983 US$0.43 - 0.21 from 1991 to 1990 US$0.19 - 0.17 from 1991 to 1993 US$0.12 from 1996 to 2005 US$0.13- 0.16 from 2006 to 2011  (Sources: exchange rate data from State Administration of Foreign Exchange of the People?s Republic of China available at, accessed in July 2013)     ix   List of Characters  Buzhi       ?? bumen    ?? Cangwu    ??   Chaiqian fei   ??? Chang?an     ?? chengshi zhichi nongcun     ?????? chengxiang jiehebu   ????? chengzhi tanfu cunguan    ?????? Chucheng     ?? Chuji Hezuoshe    ????? Cishi    ?? cungaiju shequ ????? cunzhang zhenguan    ???? dangnei tuijian   ???? Dayu    ?? Dupang    ??  feifa pidizui    ????? gongliang   ?? gongye fanbu nongye    ?????? guhua guquan     ???? Gaoji Hezuoshe    ????? Guilin Jun    ??? Hucundui      ??? Huzhuzu    ??? hezuohua yundong    ????? jiachuqu de nver    ??????    pochuqude shui     ????? jiewen lou  ??? Jinshi     ?? jingji tequ    ???? jiti   ?? Junjianyushi     ????  Junshou      ??  Junwei    ?? kaifaqu    ??? Liangtui yixuan   ???? litu bu lixiang    ?????       jinchang bu jincheng  ????? liangwei yishe   ???? Lifang     ?? lingdao manyi, tongshi manyi     ????, ????     cunmin manyi     ???? Lingnan     ??  Longcuan   ?? Luogang     ?? manyizhidi   ???? meili cunzhuang      ???? Mengzhu    ??  Nanhai Jun    ??? Nanyuewang     ??? nongdi zhuanhuan zhibiao  ?????? Nongtian shuili fei   ????? peitaofei    ??? Qingmiao buchang fei  ????? qingzhongyang jiujiu Wukan renmin    ????????? Qitian   ?? qunzhongshijian     ???? qunzhong tuijian   ???? Renxiao    ?? ruzhui  ??  sanjisuoyou, duiweijichu   ????, ???? sannong wenti    ???? san jiu gaizao   ???? Shaoguan     ??  x  Shaozhou qujiang     ????  shedui qiye    ???? sheng bu zeng, si bu jian   ???, ???     chu bu jian,  jin bu zeng  ???, ???  shehui   ?? shehuizhuyi Xinnongcun   ??????? shequ ?? shequ Guanliju    ????? shequxing gufen hezuoshe   ???????? sheyuan   ?? shezhang cunguan  ???? shiboshi   ??? siqing   ?? Songtaizu   ??? Tangxuanzong    ??? tongshou tongzhi   ???? tonggou tongxiao    ???? tudi buchang fei   ????? Tusui   ??  weishang, weizhong, weixia    ??, ??, ?? woshou lou   ??? Wuchang    ?? Wuzhou    ??  Xiang Jun     ?? Xinminzhu zhuyi    ????? xinxing chengshihua fazhan     ??????? Xijiao shangtian     ???? Xue   ? xuexi taolun  ???? Yang   ? Yingyuan    ?? Yuecheng    ?? Zeng    ? Zengcheng   ??  Zhaijidizheng  ???? Zhang Jiu-ling    ??? Zhaotuo   ?? zhibaodui   ??? zhichi Gongchandang    ????? Zhong Feng-qing    ??? Zhong Yu-yan     ??? Zhong Zi-huan    ??? Zhujixiang       ??? ziliudi   ??? ziliu jingji yongdi    ?????? zizhu jingying, dulihesuan  ????, ????     zifu yingkui , mingzhu guanli   ????, ????  zunzhong nongmin yiyuan   ??????      weihu nongmin liyi   ??????     zengjin nongmin fuzhi   ??????    xi  Acknowledgements   I feel indebted to many people who have contributed to this dissertation. First and foremost, I am deeply indebted to the members of my supervisory committee ?John Friedmann, Michael Leaf and Terry McGee ? for their invaluable guidance and advice. I am particularly grateful to John Friedmann for guiding me throughout the research process and for the extraordinarily generous manner in which he guided me in the writing and the development of my ideas. His directed reading course on urban development in imperial China was an inspiration for adopting a micro-historical approach to investigate the fascinating complexity of China?s urban transformation.   I feel immense gratitude to all those interviewees I had the privilege of meeting in my field research. They generously donated much time and energy to share their stories and knowledge.  I have not mentioned their names as the ethical review approval for this research requires that their identities must be kept anonymous, but to them, I wish to send my heartfelt thanks. In addition, I am especially grateful to Shen Kui and Chen Yong Pin. Without their great support and friendship, my field research could not have been completed.   I would like to gratefully acknowledge the financial support from the Four Year Fellowship of the University of British Columbia, the Brahm Wiesman Memorial Scholarship, and the Bottom Billion Fieldwork Fund.   My debt of gratitude is also to Bo-sin Tang, George Z. Hong, Gang Li, Randall W. Lau, Emily Li, Li-yin Shen, David Ip and the late Basil van Horen. Their support and encouragement at various stages of my research life made this study possible. I want to acknowledge the Liu Institute for Global Issues. Being a Liu Scholar, I have benefited from its great support network which has given me many opportunities to collaborate with other outstanding scholars and extend my research focus from China to the global context. I have also benefited from my time as a member of St. John?s College over the past four years. I am very blessed to have enjoyed the warm friendship of many residents, colleagues and visiting scholars who accompanied me throughout this study. Finally, and above all, I am indebted to my family and friends who kept me on track along the way.  1  Chapter 1  Introduction       The Puzzle of State Power Decentralization in Urbanizing China  From 1980 to 2011, China?s urban population more than tripled, from 191 million to 690 million. This was achieved through an unprecedented process of urbanization, which was driven by the Chinese government?s deliberate attempts to unleash market forces, support decentralized decision making, and promote local economic development. Privatization and decentralization have led to the revival of a thriving business sector and the emergence of relatively autonomous territorial governing organizations (Pearson, 1997). After two successive fiscal reforms promoted by the central state in the early 1980s and 1990s, local governments in China have become relatively independent fiscal entities, which have increased autonomy in revenue generation and allocation for local public infrastructure provision. However, in the absence of stable revenues generated by alternative means, such as property tax and goods and services tax, local governments have mostly relied on land leasing for revenue generation (Tang and Wong, 2011).  It is estimated that more than 40 million Chinese farmers have lost their land to urbanization over the past decade. This is also described as a predatory form of urbanization, under which millions of Chinese farmers have been dispossessed by the frenetic land grabbing orchestrated by corrupt, self-serving local officials to facilitate property-led local economic development (Zhang, 2001; Sargeson, 2002; Zhang et al., 2003; Hsing, 2010).  Alongside the incessant land disputes and protests in the periurbran areas, the failure of the local state to deliver public goods and services in urbanizing areas has been thought of as having two possible, separate causes. First, the fixation on economic progress gave rise to the ?entrepreneurial state? whose ever-mounting entrepreneurial incentives and authority motivated the pursuit of adventurous development projects and land-based industrial investment which often delivered a mix of economic progress, confrontation and resistance in different localities (Oi, 1992 and 1995; Parris, 1993; Solinger, 1999; Liu, 2004).  Second, the ?hollow state? generally became unable to deliver essential public goods and services, as the work of local officials was constantly interrupted by frequent bureaucratic restructurings and performance 2  assessments, the need to organize and coordinate grassroots elections, compounded by the reduction in their agricultural tax income (Cai, 2003; Tsai, 2007; Kung et al., 2009; Yao, 2009; Smith, 2010).  These studies highlight the severe problems and challenges confronting local states, implying that the governability and legitimacy of the socialist state have been undermined by its rapacious, corrupt and incompetent local agencies.  Nevertheless, what remains puzzling is that the Chinese government can stand strong and its local states remain strongly committed to advancing the urbanization level in the country, despite the potential consequences for causing widespread land disputes and social rebellions. According to the official statistics, China?s urbanization level still recorded tremendous growth from 36 percent in 2000 to 51 percent in 2010. Yet the widespread land disputes and mounting social tensions instigated by this massive population of dispossessed farmers ? as extensively reported in the scholarly literature, domestic reports and international media ? did not cripple the legitimacy of the Chinese government and its determination to advance urbanization. Clearly, local states must have done something ?right? in coping with the problems and challenges that arose from urbanization; this has been largely ignored by the mainstream literature. Therefore, I suggest that there is a need to develop a new perspective on the strategies of local states in managing the process of rapid urbanization, especially in relation to the aspects of building governing capacity, settling displaced villagers and establishing social control over rural unrest.   Research Question  To fill this research gap, I focus my study on one primary question: how has the local state managed to deal with land conflicts and the various problems arising from the rapid urbanization in reform China? To address this question, I must first clarify the definition of the local state in my study. In China?s administrative system, local governments refer to all lower level governments below the central state. Although they include all governments at the provincial, municipal, county, township, and district levels, the term local state used in this study is restricted to the governments at the township and district level.1 As Li (2009a:1) suggests, governments at this level are ?the lowest echelons of the state organization where the majority of                                                           1 Under the urban-rural dualist system, each locality is governed by a form of local government, which is termed ?township? government in the rural context and ?district? government in the urban context respectively. 3  the people have their first-hand experience with the state, and where decisions made ?high-up? in the state hierarchy are translated into processes in society.? The purpose of my study is to understand how local land conflicts and the various urbanization problems have been dealt with in real life situations. I, therefore, focus on governments at the lowest level which have had the direct interaction with villagers.   Figure 1.1 1   Hierarchy of Urban-rural Governance   Using land requisitions and disputes as my puzzles to understand the transformation of the local state in urbanizing China, I take a departure from Hsing?s (2010:7) concept of ?urbanization of the local state,? which views space as an active rather than passive force for stimulating urban transformation. Hsing adopts a perspective of interdependency to understand how local state power shapes and is in turn shaped by the process of urbanization. Through a spatial framework which divides the metropolitan region into three zones, she explores land battles between the central state, municipal governments and private homeowners in the inner cities, the land 4  exploitation by municipal governments for urban expansion in the urban edge, and the land exploitation by county and township governments for project development in the rural fringe. Based on the ?thick description? of land-centered local politics in these three types of place, she characterizes the Chinese pattern of urbanization as a process of land appropriation by local state leaders for property development. Although this study shares Hsing?s emphasis on the importance of land-centered conflicts for our understanding of local governance restructuring in urbanizing China, it adopts a different perspective to understand the interplay between urban transformation and local governance restructuring by focusing on the local state?s involvements in not only land requisitions but also the provision of social welfare for the affected villagers.  To this end, I do not use a dichotomous approach to simply define the role of the local state as either ?extractive? or ?protective.? Rather, I explore how local governments have combined both approaches to shape their strategy for governance intervention. In forming my own perspective, I have drawn insights from the recent scholarship which suggests that the local state in reform China is neither simply ?predatory? nor ?conciliatory?, but exhibits the characteristics of both in varying degrees at different points in history (see Remick, 2004; Hillman, 2010; Schubert and Ahlers, 2012). This idea of an ambivalent local state is made based on the observation that local states receive both top-down and bottom-up pressures. As the lowest echelons of the state apparatus, it falls upon local states to implement national and state interest. At the same time, they must serve local demands, because they are subject to the scrutiny of local residents, who have been increasingly empowered by democratic elections. The idea of an ambivalent local state is equally important for understanding local land politics in urbanizing China, given that land requisitions force local governments to get directly involved in the provision of social welfare for farmers.   Land in rural China is never merely a factor of agricultural production; it also constitutes an essential part of the autarkic social welfare system of rural farmers. China?s urban-rural dualist system distinguished rural farmers from urban residents by forfeiting their entitlements to urban residence, state subsidized education, medical care and retirement pensions. Without access to these state services, rural farmers had to rely on the village collectives for their public goods and services. Farmers? reliance on land for social welfare can be better understood by examining how farmers have struggled to make their self-sustaining system work in practice. For example, Brigades (now Villagers? Committee) and Production Teams (now Villagers? Groups) usually 5  mobilized villagers to construct and maintain village infrastructure, such as irrigation facilities and roads during non-farming seasons (Lin, 1990). Another example was the self-sufficient medical services in village areas. The medical system in each village was financed by a collective fund maintained by the village collective and villagers, through which a village doctor (or barefoot doctor) could be appointed to provide basic health care services for villagers, such as immunizations and curing simple ailments. In the absence of top-down financial support, most resources required for the provision of rural infrastructure and services relied on the meager income generated from the land collectively owned by villagers.   This special relationship between land and farmers? social welfare is exactly what makes farmer-resettlement as a ?protective? strategy so important for understanding why local state building was effective or ineffective in maintaining social stability and governability in the process of rapid urbanization. Land requisition in rural China was not only a process of land grabbing for local development. Rather, it was also a process of building governing capacity, as land requisitions pressured the local state to get directly involved in the building a  new social security system for the affected villagers in order to reduce possible social unrest in the process of urban transformation. As Baum and Shevchenko (1999) suggest, local officials have increasingly realized that failing to address local interests may result in a flood of complaints and collective protests which will eventually affect their career and promotion prospects.  In this regard, my research addresses not only the acute conflicts between the local state and villagers arising from land requisitions, but also the question of how the local state has dealt with farmer resettlement and regained its legitimacy in the eyes of local people.  Reflections on Theoretical and Methodological Governance Research    Taking a departure from local land politics to understand local governance restructuring in periurban China, I first draw theoretical and methodological insight from the recent discourse on governance. Governance theory in the Western sense is primarily concerned with the transformation of state-society relationships in the context of neoliberalization and globalization. Premised on the notion that the state?s traditional powers were waning due to the promotion of deregulation and privatization, governance studies tended to assume that the state had been hollowed out, and started to pay attention to non-state actors (such as business elites and NGOs) 6  who were playing increasingly important roles in the governing process. In this way, what makes governance an interesting field of research is the varied forms of discourse linked to it, such as administrative decentralization, private-public partnership, public participation, power sharing, networking, collective decision making and so on. In what follows, I will briefly explore how governance debates have been conceptually and methodologically constructed over time. By doing so, I aim to understand what we can gain by using a governance perspective to frame discussions on state-society relationships, and through such an understanding to formulate my analytical framework for generating new knowledge out of the Chinese experience for the ongoing social dialogue on governance discourse.  From ?Government? to ?Governance?  The Western concept of governance has its own specific history and intellectual origin. Before the 19th century, Western Europe was dominated by a monarchic system, under which the state was solely the embodiment of the King?s power. When the question of governance was largely placed on how to stabilize the ruling power of the monarchy over a defined territory, the distinction between the state and society was by no means clear-cut. According to Brook (1997: 19), the state-society distinction in the West has grown out of the process of three interacting social forces: ?elite challenge to the power of sovereigns (in the process of constitutionalism), the ascendancy of market principles (in the form of capitalism), and the pressure of public opinion on state decisions (in the realm of the public sphere).? This transformation process emerged in the 19th century at a time when the rising urban bourgeoisie had gained control over capitalist production and strived for a voice in state affairs. Their desire to access political power and policy making, as Brook suggests, stimulated the growth of a mechanism for representation and the related forms of civic life that are now referred to as ?civil society.? Against this historical background, the landscape of governance in Western Europe was dominated by institutional arrangements under which the state (often referred to as the government) remained separate from society and gained its authority and legitimacy through the periodic election of political representatives.       With the growth of society, the central question of governance was now about what organized practices (such as specific rationalities, and techniques) could be deployed by the state to render a society governable. From the late 1970s, the unbridled forces of neo-liberalization and 7  globalization profoundly restructured the state?s rationale and strategy for governance. Instead of relying on coercive instruments to impose its will on society, the state sought a wider range of collaboration between the multiple actors from market and civil society in an attempt to reach mutually acceptable decisions. Writing in the 1970s, the French philosopher Michel Foucault initially proposed the concept of ?governmentality? for understanding knowledge, power and governable subjects based on the analysis of different rationales and mentalti?s of government. Foucault?s analysis of governmentality, as Rabinow and Rose (2003: x) suggest, attempts to understand ?the characteristics of liberalism as a mentality of government, which presupposes that society existed externally to the state, and with this the state constrained itself by limiting the space of legitimate political power and constantly required to justify itself.? In addressing the transformation of governance arising from the neo-liberalization and globalization movements, governance theorists also address the question of governmentality, but tend to locate the debate in a set of interrelated concepts such as power sharing, public participation and administrative decentralization.     The development of governance as a theory owes its origin largely to urban politics and public administration. During the 1950s and the 1960s, urban politics, which addressed authoritative decision-making at the sub-national level, was consolidated as a sub-discipline of political science in the study of urban bureaucracy and public administration (John, 2006). Conventionally, urban governance in the welfare states such as Britain was centered on a hierarchical administrative structure, a separation between politics and administration as well as a clear demarcation between the public and private sectors (Kjaer, 1995). Under this old model of governance, the government is supreme. The state played a role in steering, regulating and controlling development. The state became embedded in a hierarchical urban bureaucracy under which civil servants were recruited according to the principle of meritocracy and driven to faithfully implement government policy by predictable career prospects. In this context, urban politics was primarily concerned with the study of city government, which focused on examining ?semi-constitutional questions and administrative doctrines about the appropriateness of local autonomy or how best politicians could hold bureaucrats to account without undermining the distinctive contribution of each? (Stoker, 2000: 91)  In the 1970s, two social changes began to shift the study of state transformation away from the institutional analysis of governmental organizations and structure. First, the state-centric mode 8  of governance practices in many welfare countries confronted the problems of bureaucratic rigidity, which ultimately undermined the accountability and efficiency of public authority. Second, the challenges to the conventional way of governance came from the change in the state?s external environment. Since the 1980s, the rapid expansion of neo-liberalism, which views the state as a major source of problems rather than a solution, advocated reducing government intervention and maximizing the role of the private sector in determining economic and political priorities (Savoie, 1994). The ideological triumph of neoliberalism championed the market allocation of spatial, political, and financial resources in ways that would favour economic growth. The swift expansion of market forces caused employment, land-use, leisure and welfare to increasingly devolve into the private sector. Consequently, the local state faced increasing difficulties in controlling and regulating economic development.  In responding to these challenges, a new array of public management concepts and tools were introduced in the Reagan and Thatcher era through the promotion of public-private partnering, contracting out public services, corporatization of public organizations, and introducing performance measures to provide incentives for civil servants (Kjaer, 2004). The profound impact of privatization was the creation of a so-called minimalist state, which had to rely on a negotiated pattern of public-private cooperation, rather than command and control, in making and implementing policy (see Hood, 1991; Rhodes, 1996; Peters and Savoie, 1998). In addressing this wave of public management reforms, which is termed by political science scholars as ?new public management? (see Rhodes, 1997; Pierre, 2005), urban politics research was steered towards the effects of marketization and globalization. Premised on the notion that political landscapes are shaped by a division of labour between state and market, urban politics focused on explaining how the government and private sector interacted in decision making (Smith, 1993).  This new approach, however, quickly lost ground. After a decade of economic liberalism, there was an appeal for re-regulating the unrestrained market forces, which were not considered able to solve the mounting social and environmental problems (Lowndes and Sullivan, 2004; Stoker, 2005). Yet, this turn of events did not revitalize the power of the state to regulate and control development for two reasons. First, state capability was questioned when the state was ?deprived of many of its traditional sources of power, policy capacity, institutional capacity and legitimacy? (Pierre, 2000: 2). Second, the changing state-market relationship under neo-liberalism in the 9  1980s reinforced the link between business interests and political elites. This close public-private coalition created worries that the state was being eroded by corporate interests, corrupt politicians and unaccountable bureaucracies. Under such circumstances, rather than extending state power, public participation was becoming an increasingly important part of decision making in urban development. The rapid development of social democratic agendas promoted citizen empowerment and political inclusion, through which local communities could get more and more involved in policy making and implementation as a counterweight to both state authority and corporate power (Kjaer, 2004). In this way, democracy constitutes an essential element of governance in that it creates a mechanism for involving multiples actors, under which governance can become a set of legitimate processes for containing conflicts (Hirst, 2000).  These changing social and political arenas resulted in what Gerber and Kollman (2004) call ?authority migration.? With a shift of power from the central state towards not only the local state but also local non-state actors, the policy agenda is no longer steered and implemented by the local government, even though decentralization enables the local state to enjoy a higher degree of autonomy and flexibility in urban management. Under devolution, non-government actors, such as business representatives, NGOs, pressure group organizations, the media and so forth, all have important roles to play in policy innovation and implementation processes. Moreover, alongside the advent of globalization, the scope of non-government actors has moved beyond the local business sector and civil society to the global societies, such as supra-national units, transnational enterprises and international NGOs. The local state has to work more closely with all these non-government sectors both locally and internationally to sustain co-ordination and cooperation in promoting the collective welfare of society.   As a consequence of ?authority migration,? governance in the advanced capitalist world has become a form of collective decision-making by a plurality of actors or organizations in a setting ?where no control system could dictate the relationship between these actors and organizations? (Chhotray and Stoker, 2009: 3). This gave birth to a new paradigm of public management reforms, which emphasized decentralization, public-private partnering, and networking in the pursuit of collective interests. Premised on the notion that the state?s traditional powers were waning due to the promotion of deregulation and privatization, governance studies tended to assume that the state was being hollowed out, and started to pay attention to the non-state actors (such as business elites and NGOs) who have played increasingly important roles in the 10  governing processes. To frame the discussion on the changing power relations between state and society under neo-liberalism and globalization, ?governance? emerged as an alternative concept to ?government? by focusing on theorizing the state?s ability to collaborate with non-state actors, in place of its traditional power to steer and control. This makes governance a research discipline distinct from its use as a general term as well as the related conventional approaches for the analysis of state-society relationships.  Governance Debate in Developing Countries   Originated in the West, the ideas of governance quickly spread to other parts of the world. This has been done mainly through the adoption of ?good governance? criteria advocated by Western donors as an explicit condition of foreign aid. In the late 1980s, Eastern Europe experienced the collapse of communist regimes, producing an opportunity for the West to legitimize its model of democratization and public administration in the rest of the world. As Leftwich (1994) suggests, the collapse of communist regimes made Western democracy more convincingly worthy of pursuit by discrediting the old political systems of Eastern Europe as ?bureaucratically sclerotic,? ?non-democratic? and generally ?insufficient.?  Under these circumstances, the international donor community, with the World Bank in the lead, started to move away from a narrow focus on the market to engage more broadly in various forms of political and institutional reform in aid-dependent countries. In the 1989 publication of a report titled Sub-Saharan Africa: From Crisis to Sustainable Growth, the World Bank concluded that the future development of African countries was in urgent need of institutional capacity building to cope with the ?crisis of governance.? Taking refuge in a good governance approach to direct development in developing nations, the international community sought to get involved in broader matters than market institutions to include other institutions without having to explicitly talk about politics (William and Young, 1994; Martinussen, 1998; Hyden, et al., 2004).   The World Bank?s position on governance is made clearer in another report titled ?Governance and Development? which was published in 1992. In this report, the World Bank formulated its vivid articulation of what it regards as ?good governance? on the basis of an interrelated set of principles such as the rule of law, political accountability and transparency. Other active international development agencies also developed their own definitions of good governance. As shown in Table 1, these definitions were divergent according to which principles are prioritized. 11  Nearly all of them, however, have clustered around the discourse of participation and rights-based development, taking an explicit interest in a fair mix of emphasis on the transparency and accountability of government administration, the provision of a sound macro-economic environment as well as broader issues related to democratic elections and human rights. Implicit in the notion of these definitions is the hope that a version of the societal architecture of neoliberalism, which is constituted by ?a limited state and a largely self-regulating civil society and market economy,? will be eventually created in the non-western developing countries (Hirst, 2000: 15).  Table 1.11  Definitions of Good Governance  Organizations  Definitions of Good Governance  Source Document   World Bank  Accountability, legal framework for development, and transparency   Governance and Development (1992)  Asian Development Bank Accountability, participation, predictability, and transparency.  Elements of Government (2008)  International Monetary Fund (IMF) Transparency of government accounts, effectiveness of public resources management, the stability and transparency of the economic regulatory environment for private investment   Good Governance: The IMF Role (1997) Swedish International Development (SIDA) Democratic governance with emphasis on democracy, human rights, participation and rule of law  About the Division of Democratic Governance (2007)  Canadian International Development Agency (CIDA)  Effective, honest and accountable exercise of government power with the common values such as justice, equity, and participation  Human Rights and Good Governance: Policy and Objectives (1996)   Data Source: Chhotray and Stoker (2009)  Alongside the good governance agenda, the discussion of governance issues in developing countries is controversial. From the perspective of ?good governance? proponents, the poor institutional capacity in developing nations can be reinvigorated only through the implementation of democratic and collaborative governance, based on the application of interrelated principles of decentralized power, private-public partnership and social participation (Richardson, 1993; Rondinelli, 1993; Rakodi, 2001). While squatter settlements constituted a substantial part of chaotic urban development in many Third World countries, the central question of this proposition becomes how informalities can eventually be integrated into a modern, legal and manageable system. One of the representative works of this era is De Soto?s (2000) call for empowerment of the poor by titling their assets in the informal sector, assuming 12  that the establishment of property rights system for the poor to trade their assets in formal markets will automatically lead to the elimination of poverty. In this way, the state is required to confine its role to legalization (or more specifically, title offering) by leaving the market as an invisible hand to shape the contours of cities (Gilbert, 2002).   Nevertheless, many studies have shown that the tides of formalization through privatization and legalization during the 1990s created more problems than they solved. First, formalization led to greater conflicts and inequities. Roy?s (2003) work on Calcutta shows that the process of sorting out title claims often stirred up violent and brutal competition among residents. Moreover, formalization weakened the security of female household members, as it consolidated gender divisions and hierarchies by vesting property rights in the male head of the household. Second, formalization failed to improve residents? access to formal finance. In studies of Bogota and Lima, residents with land title continued to face credit risks in the face of limited employment opportunities (Kagawa and Tukstra, 2002). The third problem lies in the fact that formalization displaced the most vulnerable residents by making land markets unaffordable to the poorest. Taking the case of Manila as an example, Berner (2000) suggests that one-third of the squatters who were offered loans by the Community Mortgage Program to buy the land they occupied were eventually displaced due to their inability to make regular mortgage payments.   Apart from the dilemma generated by formalization, the promises of democratization were equally unfulfilled. The good governance agenda is critiqued for ignoring the situations of local politics in developing counties. As Cranenburg (1998) suggests, efforts to promote multiple-partyism are likely to fail in developing countries where the dominant party usually holds all the trumps. Lacking social pluralism, the result of transplanting Western forms of multi-party politics is only the creation of a kind of ?form over substance? democracy in developing countries. Moreover, having over-emphasized the development of property rights regimes and electoral democracy, the promotion of democracy has indeed eroded the role of the state in the provision of health, education and social welfare services (Moore, 1999). These critiques have progressed towards rejecting the notion of good governance, by arguing that good governance was merely an instrument deployed by the west to mask its interference into the internal affairs of developing countries (Chhotray and Stoker, 2009).   13  Governance as an Analytical Approach  The highly controversial effects of good governance have stimulated some researchers to pursue a new paradigm of research, by which they appeal for scholarly attention to how the specific experiences of the developing world can contribute to theory production in the fields of development and governance studies. As Roy (2005: 147) precisely suggests:   The study of cities is today marked by a paradox: much of the urban growth of the 21st century is taking place in the developing world, but many of the theories of how cities function are rooted in the developed world. ? Beyond this mundane fact of urban growth is also the pressing issue of what might be learned by studying urban transformation of the developing world. ? I am interested in what it means to locate the production of theory and policy in the cities of the developing world.  At the same time, some scholars were beginning to question the assumption of the ?hollowing-out? state. For example, Hirst (2000: 21) argues that ?the state is just fragmented and merged with non-state sectors but not so much reduced in scale and scope.? Similarly, Pierre (2000) suggests that the emergency of new governance denotes the state?s strategic response to the global changes rather than the decline of state. Addressing that fact that state transformation under neoliberalism has created diverse outcomes in different countries, Wollmann (2001) argues state restructuring has not implied the decline of local governance institutions in countries (such as France and Germany) which have the strong state traditions. By contrast, in the United States, with its weaker state traditions, state restructuring has shown reduction in the power and scope of the state at all scales, allowing market forces increasingly to steer urban development (Kleinberg, 1995).   These contested debates on the ?weak or strong state? point to the fact that state restructuring is a complex and dynamic process rather than a static phenomenon. In an attempt to explain the divergent forms of statehood or governance outcome, some scholars engage with the theoretical insights of New Institutional Economics (NIE) to explain the diverse development and governance outcomes of different places (see Jessop, 2002; Wilson, 2004; Brenner and Theodore, 2005). With a particular emphasis on bounded rationality and path dependence, the NIE explanation is that institutional changes are always thwarted by cumulative past experiences, 14  opposing organizations and prevailing beliefs in society (North, 1997 and 2001). From the NIE perspective, the key question does not lie in whether Western concepts are appropriate to the needs of non-western countries. It lies in how and why the present institutional arrangements in developing countries have persisted and constrained these countries from choosing the Western model as a path of development. Addressing the question of ?why it is? rather than ?what it should be,? the institutional perspective helps in breaking with the conventional normative approach. Also, it rediscovers the importance of studying urban development in developing countries from historical and institutional perspectives.   In more recent years, governance research has moved towards a more empirical agenda, as evidenced by a growing body of literature which develop new analytical approaches to researching how multiple actors intertwine to shape the processes and outcome of governance. Many of these analytical approaches, such as the ?integrated model? of DiGaetano and Strom (2003) and Healey?s (2007) ?institutional approach,? tend to emphasize that governance practices must be understood in institutional and cultural contexts. DiGaetano and Strom (2003) view governance as a process of coordinating political decision, which is constructed by a series of intermediations across the cultural, structural, and agency levels of governance. Using comparisons of public-private partnering in four countries including Britain, France, Germany and the United States, they demonstrate the capability of their integrated model in explaining why urban governance outcomes varied dramatically, even though these countries share democratic traditions in their political and social systems.   Similarly, Healey (2007) puts forward an institutional approach by looking at three levels of power dynamics in which governance activities are conducted. They include: specific episodes involving multiple actors to interact in a range of institutional sites; governance processes in which discourses and practices co-aligned and drift apart to shape networks and coalitions; and cultural assumptions which determine how legitimacy is defined and what modes of governance are accepted. Healey also contains in her institutionalist approach a fair mix of emphasis on actors, institutions and culture. She, however, has a few critical points of difference with DiGaeteno and Strom especially in the way that these terms are defined. For example, instead of assuming the presence of rational actors, Healey (2007: 32) views interactions between actors through the lens of an episode, which is ?a period when a particular effort is being made to articulate a strategic response to urban area development?. Moreover, she has a broader 15  definition of culture. By ?governance culture,? Healey means a new paradigm of strategy-making for urban development in contemporary Western Europe, which must respond to a wider range of concerns beyond economic growth, to include also environmental protection, lifestyles and cultural values.  These analytical frameworks shed light on the study of governance phenomena in the different contexts because they address a number of dimensions which are indeed relevant to all governance systems. The first dimension is concerned with multiple actors and their interdependencies. As an analytical lens, governance seeks to explain the achieved economic and social outcomes through the lens of collective contributions from multiple actors in promoting public purposes. In this context, using a governance perspective to understand the empirical phenomena of governance practices firstly demands an effort to identify the key actors from both government and non-government sectors involved. Multiple actors raise questions regarding multiple sources of legitimacy and the specific ways they are linked in the course of collective actions. The second dimension is related to the institutional milieu, through which the relations and interactions among different actors are shaped by both formal and informal processes. Closely related to these questions, the third dimension is about formality and informality. To understand how governing processes are rendered legitimate, governance theory suggests that the analysis of governance practices must go beyond the formal hierarchical organizations to the networks of informal ties within formal institutions.  Conceptual Framework of Governance Analysis    Reflecting on the recent theoretical and methodological movement in governance research, this study takes governance as a method rather than a set of normative criteria to guide governance practices. In this regard, I consider four essential dimensions as constituting an analytical framework of governance: processes, actors, institutions and cultures. As shown in Figure 1.2, the core of a governance analytical framework is a process analysis which seeks to understand how multiple actors actually interact to create and solve problems in daily practice. To facilitate the analysis of processes, we need to first identify the key actors from both the government and non-government sectors. Then, we need to understand how formal and informal institutions have intertwined to shape the processes by providing incentives and constraints for the interaction of multiple actors. Last but not least, we need to examine governance cultures, which are usually 16  historically derived, in order to understand how institutions have evolved to influence governance practices.   Figure 1.22   Conceptual Framework for Governance Analysis    The application of the analytical framework relies on an in-depth case study to understand the power dynamics between state and society which are context-dependent events embedded in everyday governing practices. State restructuring and transformation of state-society relationship are a series of dynamic and iterative governance restructuring processes constructed through the interaction of cultural, economic, social, political and institutional factors. How these factors interact in the real world is highly complex and cannot be unfolded without using an in-depth and intensive approach. The study of governance phenomena requires producing pragmatically governed interpretations of contestable events in the context of power. Case study analysis, which is essentially a monographic approach striving to highlight the features and attributes of the object of study through in-depth observation, reconstruction and analysis (Hamel et al., 1993), is therefore considered appropriate for governance research. The methodological underpinnings relevant to the application of case study approach for governance analysis are drawn from Flyvbjerg?s ?phronetic planning research.? According to Flyvbjerg (2004:283),  Processes: Interactionof multiple actors Actors: State and non-state actors Institutions:  Formal and informal rules  Cultures: State traditions, political ideologies 17  ?[P]hronetic planning research is to provide concrete examples and detailed narratives of the ways in which power and values work in planning and with what consequences to whom, and to suggest how relations of power and values could be changed to work with other consequences. Insofar as planning situations become clear, they are clarified by detailed stories of who is doing what to whom. Clarifications of that kind are a principal concern for phronetic planning research and provide the main link to praxis?  As revealed by this definition, phronetic planning research contains two essential arguments. The first one is about the importance of the study of power. It emphasizes that conflict and power ought to be established as major considerations when conducting planning research. The second one is concerned with the value of experience from particular cases. It emphasizes that contextual analysis of situational practices is necessary to understand the value and interests of specific groups in specific power relations. Drawing from the ideas of phronetic approach, I would emphasize, the purpose of using a case study as a method for governance analysis is not to create a set of consistent, coherent and carefully bounded criteria which can somewhat be carried out from place to place to test against local circumstances. Rather, it aims to provide a starting point for researchers to conceptualize the meaning of governance and its pertinent concepts (such as decentralization, formality and informality, public participation and so on) out of the locally-specific experience under their study. This means that researchers are expected to contextualize the issues underlying the institutions and cultures of governance and their relationships with practices in the local context.  Governance Analysis in the Chinese Context  Using governance as an analytical tool to conceptualize local governance restructuring in urbanizing China, I start with an inquiry into the intertwining of the state and society in the context of China?s urban-rural dualist system, which has imposed a major institutional constraint on the country?s economic and social changes in the reform period. Established in the 1950s, China?s dualist urban-rural system grew out of the Chinese Communist Party (CCP)?s strategy of building a sovereign socialist country through agricultural collectivization and national industrialization. By geographically confining farmers to the countryside and mobilizing them to produce food and raw materials for supporting urban-based industrialization, the Chinese government maintained a clear-cut segregation between urban and rural areas. Such segregation 18  was implemented by means of a set of interconnected mechanisms, including the hukou system (as a measure for demographic control), the collective land system (as the foundation of property rights system) and the highly centralized administrative system, namely the Commune-Brigade-Production Team system (as a means of state power extension), to ensure the state?s absolute controls over agricultural production as well as rural-urban migration throughout the Maoist era. These interlocking mechanisms institutionalize an exploitative relationship between urban and rural areas. Rural China, as Eyferth et al. (2003) note, became a workhorse that subsidized the urban sector with cheap grains and other material inputs, and enabled the socialist state to realize its industrialization and urbanization objectives.   As a result of the urban-rural dualist system, China went through a development of ?industrialization without excessive urbanization? during the 1950s and the 1970s (Chen,1992). Unlike many Western countries, urbanization in the Maoist era did not happen along with the process of industrialization. Between 1952 and 1978, the proportion of the population living in cities was maintained at a very low level that ranged from 12.5 percent to 17.9 percent. Producing an extremely low rate of urbanization, the dualist system provided room for the stunning pace of urbanization that followed in the succeeding reform era. On the other hand, it led China to an unavoidable historical legacy: more than 750 million farmers, who had been excluded from the state welfare, had put all their life chances and collective well-being on their land. This prevented massive rural-urban migration. As a result, urbanization during the 1980s was driven by another force derived from the small town development policy.    In the early 1980s, the Chinese government initiated its spectacular economic reforms in the countryside by introducing the Household Responsibility System, which contracted out land to households and tallied rewards to individual performance in agricultural production. The implementation of the Household Responsibility System released millions of surplus labour from agricultural production. Yet, in view of the weak fiscal and administrative capacity of cities, the Chinese government remained convinced that rural-urban migration should not be allowed (Wu, 2003). In practice, it opted for a policy of small-town development - a proposal made by Fei Xiaotong, a renowned sociologist in contemporary China - as the key strategy for 19  urbanization in the early 1980s (Li, 2004).2 Premised on this strategy, the government encouraged farmers to engage in non-agricultural activities in nearby towns, while retaining their household registration and land in their home villages.3 Alongside the rapid growth of TVEs, small towns in rural areas appeared like ?bamboo shoots after a spring rain? in the 1980s. These small towns, as Fei (1986 and 1991) suggests, differed from rural villages in terms of size, economic and occupational structure but kept a close connection with the surrounding rural communities. Ma (1992: 138) explains such a close relationship in greater detail:   (1) a certain amount of peasants need to work the field seasonally; since their family will need them occasionally, it is better to work near home; (2) gaining employment in a rural enterprise entails some risk, and many Chinese peasants will not lightly give up all their ties to their land; (3) town living is more expensive than village residence, so it makes sense to earn money in town and spend it in the village; and (4) many peasants prefer a rural lifestyle. Land ownership in villages provides food, a place to build houses and to feed pigs, hens and cows.   The rapid development of small towns gave rise to a complex mix of rural and urban phenomena in the countryside of coastal regions: many farmers alternated between farming in villages and working in the nearby factories in towns. It also triggered scholarly debates on whether a desakota process (see McGee, 1991) was beginning to rise in China. Guldin (1997: 62) suggests that the process of China?s townization is ?desakota with Chinese characteristics? which is powered by ?rural grassroots dynamism? rather than metropolitan expansion. Holding a similar view, Lin (1997) and Li (2004) explain that the development of TVEs and small towns has led China to bypass the experience of other developing countries such as Mexico, in which urbanization is driven by the expansion of large cities through rural-urban migration. In the 1980s, the huge urban-rural disparities created incentives for rural-urban migration but the government continued to strictly enforce restrictions on farmers? movement to cities. Under these circumstances, small towns offered alternative destinations for farmers to pursue an                                                           2The Chinese government began with an urban development policy of ?controlling the growth of large cities, developing middle-sized cities at rational speed, and develop small towns energetically? in the early 1980s. In 1989, the urban policy was slightly revised to be ?control the growth of large cities, and develop middle-sized cities and small towns energetically?. In the late 1990s, it was further amended to ?the coordinated development of large cities, middle-sized cities and small towns.?  For more details, see the Urban Planning Act. 3The Chinese saying ?leaving the land but not the rural area, entering the factory but not the city (litu bu lixiang, jinchang bu jincheng)? aptly describes these circumstances. 20  urban way of life without moving to cities. In this way, small towns absorbed a large share of rural migrants and contributed to regulating the growth of the large coastal cities in the 1980s.  Alongside implementing its small town development strategy, the Chinese government quickly shifted its reform focus from the countryside to cities in the mid-1980s. One of the compelling reasons was attributed to the historical interdependence between farmers? social security and collective land ownership. After the introduction of the Household Responsibility System, further attempts to deepen economic success in the rural areas necessitated more radical reforms of the collective land tenure system. Ambiguity in the ownership rights of rural land inevitably became a key question to be resolved. As it was practically impossible for the state to establish a welfare system for 750 million farmers, the Chinese government came to believe that privatization of rural land would ultimately endanger social stability. Under an urban-bias development strategy, both state and private investments were highly concentrated in the urban areas. State investments in rural areas had been residual and confined to disaster relief only. Consequently, China?s rural-urban disparity, as the UN-Habitat report for 2008/2009 suggests, has seen ?one of the widest income gaps between rural and urban areas of any country in the world? (UN Habitat, 2008: 77). Farmers contributed substantially to create the economic prosperity of today?s urban China, but were disallowed to share the fruits of economic and social growth. Lacking economic and social powers, farmers (indulging migrant workers working in cities) have been rendered extremely vulnerable in the process of urbanization.  In search of better living conditions, many young farmers streamed into large cities to find work, leaving their old parents and small children in the villages. Since the late 1980s, the central government has gradually released its restrictions on farmers? movement to cities (Zweig, 1997). However, the vast majority of these rural migrants usually worked in low-paying jobs and lived in poor quality housing. They were discriminated against by local residents and became a marginalized group in urban China (Ma and Xiang, 1998; Hong, 2005; Chan and Buckingham, 2008). Their hardships illustrated not only the widening class divide and soaring social tensions in the urban areas, but also the continuously spreading poverty in rural areas. With an increasing number of young famers floating into cities, their old parents and small children remaining in the villages were not capable of cultivating their arable lands to full potential. Consequently, rural China experienced a severe decline in agricultural production and the degradation of rural 21  development during the late 1990s. Calling for action from the state to rectify a situation which had nearly reached a point of crisis, Li Chang Ping, who is a researcher investigating rural development in the northern parts of China, raised the ?three agricultural problems (Sannong Wenti)? in his letter to Premier Wen Jia Bao in 2001. By Sannong Wenti, Li (2002) refers to three interlocked rural issues: the grief of farmers, the poverty of rural villages and the decline of agricultural production. Sannong Wenti quickly attracted extensive attention from policy makers, scholars as well as the general public, presenting a critical turning point for policy making at the national level.   In response to the mounting concerns from the public about rural problems, the Chinese government attempted to revitalize the village sector through not only rural tax cuts, but also the allocation of more resources from the central state budget to the construction of infrastructure and the provision of public services in rural areas. At the 16th National People?s Congress in 2005, the central state formally mapped out its new strategy, ?Construction of Socialist New Countryside?. According to the official lexicon, the socialist new countryside (shehuizhuyi Xinnongcun) is defined as modern villages which should be well equipped with ?advanced production, affluent living standards, civilized society, clean and tidy environment and democratic management.? Stated plainly, if the objectives of the New Countryside Construction policy were to be properly met, millions of Chinese peasants, who were still residing in poor and shabby rural areas, would be able to live in conditions comparable to urban standards.   Over recent decades, rural planning in developed countries has been mainly concerned with how to integrate agricultural restructuring with social and environmental objectives while coping with the increasing market competition under globalization (Johnsen, 2003; Clout, 2006; Woods, 2007). ?Asian style? experiences of countryside development, mainly derived from the ?new village movement? campaigns in both Japan and Korea, put emphasis on the promotion of self-help and cooperation among rural communities (Long et al., 2009).Associating China?s new countryside construction with these experiences may provide insights for understanding the Chinese methods of deliberative practices. This, however, is insufficient to embrace the complexities of this Chinese policy, which has been positioned as a national strategy to modify the exploitative relationship between cities and countryside.   22  China?s new countryside building was an expression of an underlying principle: ?cities support countryside (chengshi zhichi nongcun), industries nurture agricultural production (gongye fangbu nongye).? Based on this principle, it proclaims the motto of ?Respecting farmers? wills (zunzhong nongmin yiyuan), safeguarding farmers? interests (weihu nongmin liyi), enhancing farmers? wellbeing (zengjin nongmin fuzhi).? In order to translate these principles into practice, the central state has enhanced its financial input in the construction of infrastructure, leading to improved transportation and living environment in many rural areas. Moreover, in 2008, the Chinese government promulgated the Urban-rural Planning Act (Chengxiang Guihuafa) to indicate its commitment to integrated urban-rural development. Compared to the Urban Planning Act published in 1989, this new Act puts much greater emphasis on the promotion of public participation (especially from villagers), environmental protection, and conservation of rural culture and heritage in development planning. Turning its attention to the countryside, the Chinese government has recognized that market forces must be reined in by socialist state control to cope with the challenges arising from the widespread rural poverty (Ong and Zhang, 2008).        Based on a historical review of China?s urban-rural dualism and its impact on urbanization, I argue that the rationale of and constraints behind the urban-rural dualist system, invented by the authoritarian party-state during the Maoist era, have determined what the state wants to do and can do in the reform period.  In this sense, a central research issue is how a powerful party-state has transformed, adjusted and adapted its structure, policy and strategy, under the opportunities and constraints given by the urban-rural dualist system, to accomplish a rapid urbanization process, through which more than 400 million Chinese farmers became urban residents within a time span of less than 30 years.   Undoubtedly, the ability of the socialist state to accomplish this has been deeply affected by the uneven capacities of local administrations in transmitting higher-level policies, managing public resistance and providing public goods and services. Associating the discussion of local governance with the urban-rural dualist system will lead towards a cohesive framework for examining how local practices have interacted with a set of interlinked institutions, namely the hukou system, the collective land tenure system, the rural social welfare system, the village electoral system and so on, to reconfigure the institutional milieu for the interaction among local states, village collectives and villagers.  23   Given that local governance is context-dependent, I focus on how local practices have interacted with national policies and the institutional legacies of the urban-rural dualist system to shape specific governance processes and outcome in different localities. In this regard, I rely on an in-depth case study as a method to better understand ?the Chinese experience? ? indeed an umbrella term ? which captures a whole range of locally-specific governance structures and practices that are neither static over time nor homogeneous across different regions.   Case Study: the Guangzhou Luogang District  For this study, I draw insights from the concrete experiences of local state building in the Guangzhou Luogang District (GLD) which is one of the most rapidly urbanizing regions in China. In 2005, the Guangzhou municipal government combined the Guangzhou Development District (GDD) with the Luogang Town, to form an amalgamated entity that is known today as the GLD. As a new administrative district of Guangzhou City, the GLD had an area of about 393.22 square kilometres and a population of 373, 700 in 2010. Geographically, the GLD is located in the eastern part of Guangzhou City (see Figure 1.3). It is bordered in the north by the Conghua District, in the east by Baiyun and Tianhe, in the West by Zengcheng, and in the south by the Huangpu District and Dongguan City. Historically, the GLD was a village area owing its origin to the pre-Song dynasties.   My selection of the GLD as a case study is based on four major considerations. First, the GLD has undergone a very rapid process of urban transformation driven by the local state?s land requisitions for industrial and urban development. With its extensive experience in dealing with land requisitions and farmer resettlement, the GLD provides a productive site for understanding how the local state has worked with villagers, village cadres and other non-state actors in dealing with land conflicts and various problems arising from rural-urban transition. Second, its urban growth is coeval with China?s urbanization in the reform era. Urban transformation in the GLD claims a 30-year history that goes back to the early 1980s, during which the GDD was established as one of the first state-level development zones for experimenting with China?s market reform and urban development policies. In this context, the local state building and the evolution of the state-society relations in the GLD can be seen as a microcosm of local 24  governance restructuring more generally under China?s trajectory of urban transformation. Third, the GLD local state is a district government, which has transformed from an industrial zone administrator to a local bureaucrat. Its experience in local state building will add new insights to the current debate, which have mainly focused on rural county and township governments. Last but not least, the history of the GLD reached the pre-Song dynasties. The GLD not only provides a vivid case of rapid urbanization and governance transformation under a set of powerful institutions derived from the juxtaposition of socialism and market economy. It also provides a rich and reflexive example of how a village-based society with an ancient agrarian civilization has undergone a stunning urbanization journey via socialist market reforms.   The GLD is a rapidly urbanizing area which comprises what Hsing (2010) calls the ?urban fringe? and the ?rural fringe.? Hsing?s analysis selects cases from different parts of periurban China to represent the countrywide urban core (from Beijing, Shanghai and Changsha), the urban edge (from the villages of the Guangzhou), and the rural fringe (from the villages of Hebei and Anhui). Using an in-depth case study approach, I treat ?inner city ? urban edge ? rural fringe? as a spatial and temporal continuum to depict their phenomena in the same regional context. In this regard, I start with a historical review (see Chapter 3) to understand how the GLD has transformed from a rural village area, to an industrial development zone, and finally a new urban district of Guangzhou city. The core of this study examines how the local state, villagers, and other actors (such as investors and migrant-workers) have dealt with various conflicts and problems in the midst of this rural-urban transformation. Based on intensive field research, my research goes beyond the analysis of top-down policy reforms by constructing a direct observation of local state building and governance restructuring in the urbanizing villages.  This study serves as a starting point for my pursuit of a micro-historical approach from an endogenous perspective to understand the dynamics and complexities of urban transformation and governance restructuring. With its huge population, prolonged development history and all-embracing governing institutions of state and party, China has undoubtedly emerged as one of the most productive sites for understanding the interplay between urbanization and governance transformation. For researchers, the real challenge of producing insightful analyses from this dynamic and iterative process of urbanization for theory production primarily came from the vast temporal and regional variations in local practices throughout the country. As Baum and Shevchenko (1999: 327) describe, 25   In some ways the problems is redolent of the parable of the blind men and the elephant: analysts probing different parts of China?s reforming political autonomy often produce substantially dissimilar sketches of the body politics. Even in cases where the same (or similar) part of the elephant?s autonomy has been touched, there are often significant differences of interpretation (Baum and Shevchenko, 1999: 327).  This challenge has become increasingly prominent over the past decade with the continuity of rapid urbanization. In this regard, Li (2009a: 1) urges researchers to be ?modest in the conclusion drawn from substantive observations obtained at a specified time-space,? given ?the fluidity and contingency of existing state practice and image.? Working with this challenge, I conduct a close examination of the interactions of local actors in daily life to understand how economic and social changes have been enabled and constrained in the process of urbanization. The purpose of this study is not to provide a generalized account of the Chinese pattern of urbanization and governance, but rather, by taking GLD as a specific case, to explore some problems and pitfalls which have been commonly observed in China?s urban development and local governance. Through an in-depth analysis of the specific conditions under which the GLD has been able to achieve its current governance outcome, I aim to explore in what ways its locally specific experiences can inform governance reforms in other periurban areas.  Dissertation Chapter Outline  I divide the remainder of my dissertation into six chapters. Chapter 2 provides a detailed account of the methods and procedures I have used for my case study. Chapter 3 is a historical analysis of the GLD?s rural-urban transition. In this Chapter, I examine the development trajectory of the GLD with a particular focus on the unique expression of the urban-rural dualist system in the area in response to urban transformation and its subsequent impact on local governance restructuring. Chapter 4 examines the process of land requisitions and its impact on local governance restructuring in the GLD. It explains how and why the land requisitions during 1990s led the local government to cede its power to urbanizing villages, enabling a rapid growth of autonomous villages in the region. Chapter 5 and Chapter 6 explore how the GLD government have engaged in shareholding reforms and grassroots elections to re-penetrate its 26  power into the urbanizing villages. Chapter 7 concludes my research. Returning to my research question, I elucidate how the GLD government consolidated its power in the urbanizing villages through integrating regulatory and conciliatory strategies with its development policies to maintain regime legitimacy and regain controls over local land disputes. Drawing upon the theoretical and methodological issues that have emerged from this research, this concluding chapter also explores in what ways the locally-specific experience of the GLD can inform the current debate about local governance restructuring in periurban China.    27      Figure 1.3 3   Guangzhou City and Its Districts    Zengcheng   Panyu  GLD   Conghua  Huadu  Baiyun  Nansha  Liwan Yuexiu Shenzhen City   Huizhou City   Fuoshan City   Dongguan City   Qingyuan City   N   Tianhe Pearl River     Haizhu  Pearl River    Urban centre   Legend: Pearl River  Huangpu Pearl River    28  Chapter 2   Case Study of Guangzhou Luogang District:  Research Design and Field Research         The primary purpose of my case study is to understand how the local government has interacted with village cadres, villagers and other non-state actors in dealing with land conflicts and the various problems arising from rapid urbanization. In this chapter, I will present a detailed account of my methodological considerations, analytical framework and field research process, through which I have endeavoured to understand the complexity of the urban transformation and the local governance restructuring in the GLD.  GLD: A New Urban District of Guangzhou City    In 2005, the Guangzhou Municipal Government merged the GDD with the Luogang Town to form the present GLD.4 As a result of this merger, the GLD became a new urban district, which comprises four development zones, five street wards, and one small town (see Figure 2.4).5 The GLD has a total geographical area of 393.22 square kilometres. By 2010, its population stood at approximately 373,700, among which about 170,000 were local residents. Alongside this state-led urbanization process, the urban transformation of the GLD has been primarily driven by the local state?s massive requisition of arable land for industrial development, leading to an in-situ urbanization process through which villagers become urban residents without having to move to cities.  Under these circumstances, land conflicts came to stand at the core of local governance in the GLD. Moreover, how to facilitate the transition of villagers from a rural to an urban way of life presented a significant challenge to local governance.                                                              4 Luogang Town was previously a rural area under the administrative purview of the Baiyun District. 5 Four development zones under the administrative purview of the GDD include the Guangzhou Economic and Technological Development District, the Guangzhou Hi-Tech Industrial Zone (i.e. Guangzhou Science City), Yonghe Economic Zones, the Guangzhou Free Trade Zone, and thte Guangzhou Export Processing Zone. Five street wards include Luogang, Lianhe, Yonghe, Xiagang and Dongqu. One town refers to Jiulong Town.   29  One year after the establishment of the GLD, rural surplus labour in the district had reached already 40,000 persons and about 49,000 villagers urgently needed to shift employment from the agricultural to the non-agricultural sector. How to ensure adequate social welfare provision for villagers was another pressing challenge for the new GLD government. Under the urban-rural system, rural farmers relied heavily on the village collective organizations to obtain public goods and services, such as health care, education and so forth, as well as on collective land for their subsistence. The GLD?s villagers were not an exception. Pension schemes were not available to them. Moreover, like many other rural areas in China, village medical services in the GLD were not integrated with the urban system. Village medical services in the GLD were minimal and villagers had great difficulties in getting affordable and proper treatment when they fell ill. The lack of an affordable medical system has become a major cause of rural poverty as expensive hospitalization fees can swallow up the lifetime savings of a rural family. In addition to the challenges arising from employment and social welfare provision, another problem was the management of land compensation fess and collective assets.   Nevertheless, it seems the local government has championed local state building in the face of these widespread problems and challenges. A previous study on a land requisition program, conducted by the GDD authority for the construction of the ?Guangzhou Science City?, reveals that 705 households from three affected villages were relocated to new apartment buildings provided by the government without causing severe conflicts (Tang et al., 2008). Since 2004, more than 50,000 villagers in the district have changed their hukou status from ?agricultural? to ?urban?. Moreover, in most neighbourhoods where these villagers now live, many new facilities (such as libraries and sport centres) have been made available (see Figure 2.5). Between 2003 and 2009, several major programs were carried out to deal with land disputes and farmer resettlement with a particular focus on the following areas:   ? Creation of non-agricultural employment opportunities. In 2006, the district government introduced a ?Non-walled Factory Scheme,? under which factories within the GDD were encouraged to contract out assignments to local households so that villagers could work at home rather than within the factories. Within the same year, an ?Entrepreneurship Scheme? was launched with the aim of helping villagers to start their 30  own businesses.6 Under this scheme, a special fund of 20 million yuan was established in 2007 to provide subsidies for jobless villagers who had started their own business and had operated them for at least 6 months. Moreover, a vocational centre was established in collaboration with local industrial institutions to provide more than 8,000 training opportunities for ex-villagers.  ? Provision of medical system and pension scheme. The district government introduced a new cooperative medical system in 2005.7 This new system was run like an insurance program. Villagers participated in the system on a voluntary basis. As of 2009, more than 90 percent of the villagers had enrolled in the new scheme. At the same time, more than 90 percent of the villagers have been covered by the government?s pension scheme. A steadily increasing rate of participation provided a solid foundation for the government to carry its medical reform forward. In 2007, the district government engaged in an upgrading program to establish a three-tier system for the delivery of medical care services.8 The first tier was implemented at the neighbourhood level; a clinic network was established through the construction of new neighbourhood clinics and the revitalization of the existing village clinics. The second tier was at the street level. For each street ward, a medical service center of about 2,000 square meters is provided. To overcome the lack of local qualified medical professionals, the management and operation of these medical centers is contracted out to county or municipal hospitals. The third tier established the link between community medical centers and large hospitals with agreements to refer seriously ill patients to the large hospitals for further treatment.   ? Shareholding reforms. Shareholding reforms for the management of collectively owned village assets in the GLD region were first introduced by the municipal government in the mid-1990s. In 2003, the district government implemented a new shareholding reform                                                           6 This fund provided guarantee services for the eligible business owners when they needed to borrow money from banks. It also provided one-off subsidies for successful applicants to pay tax, expanses for their participation in business management training, and so on. To integrate this scheme with local economic development, priorities were given to applicants engaging with business related to wholesale and marketing of local farming products, convenient stores, repairs of household appliances, bookstands so on. See the policy document ?Implementation Methods of Self-employment Programs? which was issued by the GLD government in 2007 for details.  7 See policy document ?Implementation Methods for the Development of Rural Medical Services? issued in 2005.  8 See policy document ?Implementation Methods for Accelerating the Development of Rural Medical Services? issued in 2007.  31  program called ?fixing share rights (guhua guquan ????)? to provide better delineation of ownership and revenue rights of individual villagers to collective assets.  These emerging phenomena in the GLD?s local governance restructuring raise a range of interesting questions: how have land conflicts and disputes been dealt with? How and why does the existing district government substantially engage with social welfare provision? What are the impacts of shareholding reforms on reshaping the institutional arenas for the interaction among the district government, villagers, village organizations and other non-state actors? Is the collaboration between the local state and villagers real or merely apparent? How have village traditions, which are extensively encompassed by kinship ties, gender role and family relations, influenced daily village governance? How can the local state expand its power, influence and functions in governance processes? Does the GLD operation present a new type of state-society relationship? These questions drove me to seek a micro-historical approach to understand how local actors have contested with each other to shape the existing governance outcomes.   Studying Urbanizing Villages (New Urban Neighbourhoods) as a Method of Inquiry  In selecting a micro-historical approach for this study, I have also been deeply influenced by the recent epistemological shifts in planning endeavours. Friedmann (1973) advocated the approach of ?transactional planning,? or ?mutual leaning,? which linked local knowledge with technical expert knowledge in specific planning endeavour. His work has gradually led methodological discussions on planning research to make a shift from the emphasis on objectivity to the study of lived community experience and invisible power relations embedded in the flow of daily life (see Forester, 1989; Healey, 1996; Sandercock, 1998; Innes and Booher, 2000; Flyvbjerg, 2002 and 2004). In response to these epistemological shifts, I focused on elucidating the lived experience of the multiple local actors to understand the process of governance restructuring that took place during the swift urban transition.   Going beyond mere policy analysis, I started my study with urbanizing villages, where local officials, village cadres, villagers and other non-state actors (such as migrant workers and property developers) have contended with the realities on the ground to shape the narrative of local governance. What I mean by urbanizing villages in this study refers to ?new urban 32  neighbourhoods (cungaiju shequ),? which were recently converted from administrative villages. Within these new urban neighbourhoods, Villagers? Committees had been re-organized as Residents? Committees. Moreover, all villagers became urban residents as the local government changed their hukou from ?agricultural? to ?non-agricultural? status. In these transitional neighbourhoods, urban institutions and practices are gradually adopted and integrated with daily life. At the same time, informal norms and traditions indigenous to rural communities, such as villagers? sense of collective identity (for the purposes of this discussion, better understood as perhaps the interdependence between a villager and his or her respective village and subsidiary organizations, such as shareholding cooperatives), their shared assumptions about kinship relations and gender role as well as their perception about authority and legitimacy, continued to inform the daily interactions among villagers.   Figure 2.14  Local Governance Hierarchy of the GLD    New Residents? Committees Residents? Committees Guangzhou Municipal People?s Government Luogang District Government (GDD Administrative Committee and Luogang Administrative Centre)  GDD  (4 Development Zones)  5 Street Offices Jiulong Town Villagers? Committees  New Residents? Committees  Residents? Groups / Shareholding Cooperatives  Residents? Groups  / Shareholding Cooperatives   Villagers ?Group/ Shareholding Cooperatives   Guangdong Provincial People?s Government The Central People?s Government (State Council) 33  In these urbanizing villages, local residents still referred to themselves as villagers, even though they had converted their hukou status into urban, and their village had been officially recognized as an urban neighbourhood (shequ). Thus, I continue to use ?villager? and ?village? in this dissertation, despite the fact that their legal status suggests that they were already urban. An in-depth investigation of some newly emerging problems and phenomena about the governance of these transitional neighbourhoods will provide windows to explore the fascinating complexity of power dynamics between the local state and society in the GLD. As Leaf (2007:170) suggests, the analysis of the phenomenon of urbanizing village in China can ?lead us into questions of community building and identity, consideration of urban housing challenge in China?s evolving market economy, and more broadly still, issues of property and development rights and what they imply for power contestation and other processes of local governance.?    Table 2.12  A Overview of New Urban Neighbourhoods (Urbanizing Villages)    Name    Area (km2)  No. of Natural Villages  No. of Shareholding Cooperatives  No. of party Group (Members)  Indigenous Population    No. of Household  Major Family Names  N-1 9.24 9 12 3 (169) 7435 2257 Zhong, Luo, Han N-2 12.92 8  15 3 (178) 8081 2489 Zhong, Luo, Li N-3 13.35 6  11 3 (160) 5154 1569 Zhong, Yang, Deng, Qu, Zeng N-4 13.61 9 13 3 (190) 6202 1860 Zhong, Liu, Lai, Chan, Xiao, Mo N-5 13.4 7 23 4 (181) 7775 2363 Zhong, Ng, Gong, Li, Lai, Sun N-6 5.4 1 9 4 (108) 3750 1042 Kong, Wong, He N-7 10 9 9 4 (62) 2192 595 Zhang, Zhou, Wu, Zhong, Wen, N-8 6.8 1 4 3 (48) 1301 383 Zhu, Zhang N-9 10.2 5 6 6 (107) 2117 607 Wong, Sun, Zhang, Zhou N-10 9.6 6 12 6 (123) 4620 1263 Long, Zhu, Zhang, Kong N-11 11.5 7 19 3 (200) 8100 2460 Zhu, Xu, Lu N-12 0.5 1 3 1 (47) 1023 285 Long, Liang, Tang  Data Source: field research and Website of the GLD Government9      Until 2009, there were a total of 27 new urban neighbourhoods managed by 27 Residents? Committees within the jurisdiction of the GLD district government. In my study, I paid attention to the ?older? of the new urban neighbourhoods, which were reorganized from Villagers?                                                           9 Available at (Accessed in June 2012). 34  Committees on or before 2005 (see Table 2.1).10 I focused on the older ones because they permit a close examination of the effects of urban transformation on the local community, which come about through accumulative changes over relatively lengthy periods of time. The neighbourhoods sampled for this study allowed me to observe such changes over a longer period of time; that is, of more than seven years. After a primary investigation, I categorized them into three groups based on their varied experience of losing land to urbanization: (1) neighbourhoods that lost their arable land completely, but had some collective land returned by the state for non-agricultural uses after land requisition; (2) neighbourhoods that lost their arable land in part, held some non-agricultural land returned by the state, and retained some arable land, which itself might be requisitioned at a later stage of urbanization; (3) neighbourhoods that still were holding a large amount of arable land and which remained relatively untouched by urbanization. While expanding the scope of my investigation to as many neighbourhoods as possible, I placed a particular emphasis on six of them, which were selected from different groups for in-depth study.   Conceptual Framework for the Neighbourhood Governance Analysis    Taking governance as an analytical tool to examine the GLD case in depth, I first devoted efforts to identify the key actors and institutions which may have an effect on the processes and outcomes of changing local governance. As shown in Figure 2.2, there is no question that villagers, village cadres and district government officers are the main actors in each village. Here I want to highlight other important interest groups. The first group includes industrial investors who are operating their businesses in the GDD. The second group refers to the property developers who have actively sought development opportunities in the urbanizing villages by setting sight on the ?reserved commercial land? with great potential for commercial and real estate development. The third group includes mainly migrant workers who are residing in the village. Although not entitled to the benefits of the village-based social welfare system and shareholding, they work for the industrial enterprises in the GDD and lease houses from villagers.      At the village level, these key actors usually cooperate and contest with each other in two village organizations, namely the Villagers? Committee (now Residential Committee) and the                                                           10 All neighbourhoods are pseudonyms in order to protect the confidentiality of research participants. 35  shareholding cooperatives (gufenzhi jingji hezuoshe). Under China?s urban-rural dualist governance system, Villagers? Committees are set up in rural villages whilst Residents? Committees are set up in urban neighbourhoods. Both of them are known as ?mass organizations of self-governance,? in which residents / villagers manage their own affairs, educate themselves and serve their own needs. Under the leadership of Residents? Committees, shareholding cooperatives are self-governing village organizations responsible for the management of assets collectively owned by villagers. Despite the fact that their members are chosen by popular elections, neither Residents?/ Villagers? Committee nor shareholding cooperative should be thought of as a level of government. The operations of all these village organizations are primarily embedded within an institutional milieu constructed by four sets of formal and informal institutions, which take their roots in the urban-rural dualist system.  Figure 2.25   Local Governance Structure in GLD: Key Actors and Institutions    The first is the collective land tenure system. Established in the 1950s, the collective land ownership system has remained anchored in rural China. Under the urban and rural dualist system, land in rural China constituted an essential part of farmers? self-sufficient welfare system. In this way, dealing with land conflicts and disputes is not merely a matter of property rights. What is at stake is whether displaced farmers, who were historically sealed off from state social welfare, can gain access to adequate social services such as health care and old-aged pensions. For by removing the farmer?s livelihood, by taking his land, the government must Local State (District Government and Street Offices)Market (Inverstors /property developers)Neigbouhood Organizations (Residents'/Villagers' Committee and shareholding sooperative)Community (villagers/ Migrant workers)Clan-based village norms & customs (informal)  Electoral System (formal)   Collective land System  (formal) Shareholding System  (formal) 36  make certain that the social welfare system can fill this gap of need; otherwise, failure to do so risks the threat of widespread social unrest, protest, and perhaps uprising. The GLD government has demonstrated significant progress in this respect. A detailed examination of why and how the district government has actively responded to the needs for better land compensation and social welfare services will provide a vantage point for understanding how it has been able to reduce famers? resistance to land requisitions. After land requisition, ?reserved commercial land? constitutes a major source of income for ex-villages. How this ?reserved commercial land? is developed and managed for income generation was also explored.   The second is the shareholding system. Shareholding Cooperatives were formerly known as Production Teams in the Maoist era. In seeking a better mechanism for distributing the increasing incomes derived from collectively owned land and village assets, like their counterparts in the Pearl River Delta, village collectives in the GLD introduced a shareholding system into their management by translating villagers? rights to collective income into shares. Consequently, villagers became shareholders and were entitled to dividends each year. During the 1990s, the Shareholding Cooperatives in the GLD developed their own regimes for defining shareholders? interests and determining dividend policy. This allowed the Shareholding Cooperatives to exercise autonomy in managing their collective assets, but created loopholes that village cadres could use to reallocate shares. The result has been that some villagers engaged in open disputes with village cadres, who have manipulated the operations of asset management and failed to fulfill their deep-seated desire for a fair distribution of collective income. In this way, a more detailed analysis of the operation of the shareholding system led me to understand how  the tension between village cadres and villagers have created space for the district government to reshape its authority in managing instability in urbanizing villages through shareholding reforms.   The third is the electoral system. Since the late 1980s, the ongoing promotion of self-governance and competitive elections in villages has significantly transformed local power structures in rural China. While many studies suggest that village elections restrained arbitrary local officials and held them accountable, some argue that village elections were just flawed processes manipulated by rapacious local officials to benefit their own purses at the expense of the majority interest of the villagers. To understand the electoral process and its impact on social participation in the case of the GLD, it is worthwhile to explore: (1) how elections have been 37  carried out in the villages; (2) what procedures empowered/disempowered villagers in the process of election; and (3) in what ways the local state has influenced the process and outcome of elections.  The fourth is clan-based village norms and traditions. Clan-based social relations are a crucial factor affecting many aspects of village governance, even though the existing literature has offered relatively few studies on them. The collective land system in rural China has not been enforced by legal procedures with which we are familiar in today?s urban areas. As Ho (2001) suggests, individualized, standardized and notarized contracts for land leasing were only made available nearly 10 years after the introduction of the Household Responsibility System. Despite an explicit policy movement to get land contracts clearly defined, its turns out that many documents are missing and fragmented due to the poor administrative capacities of localities (Wang, 2001). In the absence of a well-established legal system, a village-based social organization supporting exchanges on a non-market basis has demonstrated its critical role in enforcing the collective land system. Moreover, social relations in rural villages were traditionally governed by intimate personal ties rather than formal instrumental relations and rule of law. As such, clan-based customs and traditions have played a critical role in village elections and domestic conflict mediations. The villages in the GLD are undergoing an in-situ urbanization process, through which clan-based social organization, which atrophied in the Maoist era, have been revived in the current stage of the rural-urban transition. In this context, I considered village norms and traditions as an important dimension of an analytical framework for understanding local power structure and governance process.   Analytical Framework for the Analysis of Local Governance in the GLD  In order to articulate the study of local state governance restructuring within these four sets of interconnected institutions, I formulate an analytical framework by which I examine the real-life interaction and contestation among local actors in different urbanizing villages. My analytical framework consists of three levels (See Figure 2.3). Starting with the micro-level, I address specific episodes (such as land requisitions, elections, medical services provision and so on) involving the local state, village collectives and villagers in a range of formal and informal 38  institutions.11  At the meso-level, I address the governance processes through which governments and their agencies at higher levels (provincial and municipal) have managed to translate relevant national policies into practices. At the macro-level, with the aim of connecting these local findings to a wider spectrum of discussion, I explore ?political cultures? and ?structural contexts? with a particular attention to some prominent policy issues, such as national land use control policy, New Countryside Construction and so on.    Figure 2.3 6  Analytical Framework for Local Governance Structure and Practice in GLD Dimensions    Study Components and Specific Research Questions       Structural Context and Governance Culture    Exploring how the macro structural and cultural factors have intervened to determine the governing logic and the range of accepted modes of local governance in China.  ? Economic globalization  ? Discourse on urbanization and effective urban governance  ? Party-state traditions , state restructuring and political objectives ? Integrated urban-rural development ? Fiscal reforms and local government finance         Historical and Institutional Legacies of the Urban-rural Dualist System    Studying how the flowing issues have intervened to provide constraints and opportunities for urban transformation and interaction between the state and society at localities:    ? Hukou system  ? Rural collective land system  ? Rural administrative hierarchy  ? Separation of urban-rural development planning          Local  Governance Institutions     Investigating how local practices have interacted with these interconnected institutions:   Collective land tenure System   ? How are collective land and assets defined? ?  What is the composition of these collective assets?  ? What are the procedures and standards for land requisition and compensation? ? How has collective land been developed and managed for income generation?  Shareholding System     ? How are villagers? rights to the collective land and assets translated into shares? ? How are shares classified, allocated and managed?  ? What are the rights and obligations of villagers as shareholders? ? How have shareholding reforms been deployed as a measure of social control and social transformation? ? How have the reforms over the shareholding system given rise to a series of new institutions from which the government can exert greater influence on the operation of neighbourhood organizations?                                                            11 In these new urban neighbourhoods, the formal institutions usually refer to the collective land system and the neighbourhood electoral system, while informal institutions refer to village norms and traditions. 39  ? How have these reforms given rise to a new mechanism in which ex-villagers? can participate to influence the distribution of income derived from the collective land assets?   Electoral System (Election of CCP & non-CCP members of Residents? Committee and Election of Directors of Shareholding Cooperatives)   ? Who are eligible for voting and being voted?  ? How are villagers? representatives selected?  ? How do villagers? representatives go for final election of Residents? Committee members? ? How is the election process being managed and monitored? ? What are the role and functions played by the district government in electoral processes? ? How have these reforms been managed to resolve the previous conflicts between village collective units (now RCs or SCs) and villagers by plugging the loopholes in which local cadres can manipulate the management of collective assets?  Clan-based Traditional village custom and norms    ? How are the clan-based relations constituted?   ?  What functions have village norms and traditions played in conflict mediation?  ? In what way do clan-based blocs influence village management and elections? ? How do local village traditions and customs affect local residents? attitude to migrant workers?       Neighbourhood Governance Practices and Processes   Investigating how the local state, ex-village collectives, ex-villagers and other actors have collaborated and contested  in governing processes   Specific episodes   ? Land requisition and compensation  ? Shareholding reforms   ? Provision of health care services and pension scheme    ? Creation of non-agricultural job opportunities  Organizational structure   Government Departments  ? The district government (and its party organizations) ? Street office (and its party organizations)  Ex-village Organizations ? Villagers? Committee (now Residents? Committee) ? Shareholding Cooperative  ? What are the organizational structures of these state and non-state organizations? ? What degree of autonomy does RC/SC have in its management and operation? Are SCs (which have no party presence in formal terms) truly free of party influence? ? How can the district government secure the cooperation of RC and SC in the implementation of its policy and programs? ? How can ex-villagers (who are now neighbourhood residents as well as shareholders of SCs) monitor the operation and performance of RCs and SCs?  ? What is the mechanism for conflict resolution? ? What institutional incentives and constraints have been provided for the district government to engage in the social dimensions of development? ? How does the district government create incentives and constraints to secure local cadres? dedication to policy implementations?  ? How has the district worked with the upper lever government (municipal and provincial) in making and implementing policies?  ? How is the role of the party as the glue that holds the connections between the central state and the district government at the lowest level of the administrative pyramid?  Decision-making Processes   ? Who are the key actors? Who are the decision makers? ? What percentage of the compensation fee and the resettlement subsidies are retained by the village collectives? ? What is the mode of local financing? What is the role of revenues generated from tax, land 40  lease, collective assets and other revenue instruments in financing neighbourhood facilities and infrastructure? ? What are the key problems, disputes and conflicts? ? How can consensus be reached? ? What percentage of residents are temporary workers? Do they have sufficient access to participate in neighbourhood development planning and governance?  ? What are the formal and informal rules governing the cooperation among various actors?  ? In what ways can residents participate in and influence decisions making for new neighbourhood development? ? How can migrant workers get their voice heard, given that they are excluded from the shareholding and electoral systems?  ? What are the key problems and disputes?       Local Power Structure and Dynamics        Explaining the transformation of the relationships between the local state, ex-village organizations, ex-villagers and other non-state actors    ? Local state: a facilitator or a commander?  ? RC and Shareholding Cooperatives: a representative of residents, a subsidiary to the district government, or an integral and defining element of the dynamics of the local state? ? Villagers: active or passive participants? ? Migrant workers: consumer or contributor to the neighbourhood  ? State-society relationship: partnering, coalition, confrontations, networking or others?      In order to connect the micro-level analysis with the macro-level analysis, a historical review and a comparative approach constitute essential parts of my research. Through an historical study which focuses on interpreting the past in the light of the present, I explore how various historical factors and forces (at both the local and national levels) have interacted to shape the current development pattern in the GLD. With a comparative approach, I draw upon the existing literature to understand urban transformation and local governance practices in other urban frontiers. The main purpose of comparative analysis in this study is not to produce generalized concepts that can travel across the globe. Rather, it aims to understand under what specific conditions the GLD has been able to achieve its current governance outcome. As Robinson (2011: 19) suggests, comparison is ?neither a parochial universalism nor a uniform global analytical field but a rich and fragmented array of ongoing conversation across the world of cities.? Through examining the GLD experience comparatively, I seek to identify the critical factors that underpin the special experience of the GLD, and to examine the implications of the GLD experience for local governance practices in other urbanizing areas.     41  Fieldwork: Stage One     Intensive fieldwork, interviews and observations constitute the core of my research undertaken for this study. To enable a back-and-forth process between my data collection and analysis, I divided my field research into two phases. During my initial field research in July 2012, I started with my interviews in the urbanizing villages, aiming to understand the daily practices of neighbourhood governance with a particular focus on the impact of land requisitions on collective asset management and community life.12 Many researchers who have had intensive experience in doing fieldwork in China suggested that it was extremely difficult to identify and contact local cadres and villagers for interviews if there was no formal introduction and referral provided by the local government. The same was true in the GLD. In this regard, I first sought support from the Policy Research Office of the GLD district government.   My relationship with the GLD Policy Research Office came from my master?s studies at the University of Queensland, during which I focused my research on examining the urban management problems arising from the transformation of the GDD from an industrial zone to an urban area. After my graduation in 2004, I was given an opportunity to participate in a research project conducted by the Hong Kong Polytechnic University in collaboration with the Policy Research Office. The main theme of this project was about developing a set of measurable evaluative criteria for assessing urban management performance in the GDD. Working as a co-investigator for this project, I conducted field work in the GDD in 2005. After the completion of the project, I have maintained contact with some colleagues of the office over the years, sharing with them my progress and experience in doing research on Hong Kong?s land use planning and China?s property taxation. In order to seek support for this research, I paid a visit to the Policy                                                           12 My study mainly involved five groups of interview participants: village leaders from Residents? Committees/shareholding cooperatives, villagers, migrant workers, investors, and public officers from both the municipal and the district level. The background of interview participants is listed in Appendix II. I adopted the method of open-ended interviews. I prepared a set of questions for highlighting the information that I intended to collect (See Appendix III). Starting with these questions, I encouraged participants to raise questions to me and to address what they considered important. Interviews were conducted in the places that were convenient to the participants. I used written notes rather than a tape recorder to record interviews. To ensure confidentiality, the name of participants and the places of doing interviews were anonymous in my notes; they were only recorded in my interview journal which was encrypted and kept in a separate location. Moreover, any presentations of the results in this dissertation report are anonymous to ensure the proper protection of individual confidentiality. The principle of confidentiality was also applied to my field observations. I observed the behaviours and interactions of multiple local actors in difference places. I kept notes of my observation but did not involve identification of any subject.    42  Research Office and made a presentation about my research proposal in the winter of 2011. After the presentation, I returned to UBC to continue my studies. In May 2012, when I came up with a solid schedule for my field research, the Policy Research Office formally issued me a letter, indicating their willingness to provide all necessary support for me to conduct independent fieldwork in the GLD for my doctoral dissertation.    The Policy Research Office gave me introductions and referrals to the target interviewees, but always left me to decide which neighbourhoods I wanted to go to for research. To gain as wide a scope of daily governance practices as possible, my strategy was to visit as many villages as I had time for. As a result, I paid visits to more than 12 Residents? Committees and their related shareholding cooperatives during my initial fieldwork. In each Residents? Committee, those local cadres who participated in my research included the party secretary, committee director, committee members, and shareholding cooperative leaders. Local cadres mostly took initiatives to greet me with ?Nihao? and ?Huangyingni.? They knew that I could understand Chinese, but did not expect I could understand their dialect well. When I responded friendly to them in Cantonese, they felt relieved and happy. They told me that they struggled with speaking Putonghua; they were not very good at it even though they were required to use it in formal meetings with officials and visitors. Sharing a common dialect of Cantonese (even though we had different accents) helped us find a common starting point for our dialogue. I began by asking questions about their organizational structure and their everyday work. They usually explained to me how they implemented their duties in accordance to the guidance of the Article of Association, which was the formal guidelines neighbourhood governance. Their answers enabled me to understand the division of labour amongst them as well as the organizational relationships between the Residents? Committees and shareholding cooperatives. They also told me about their plans for village development, which was one of the more prominent tasks that they were engaged in.   After interviews with local cadres, I asked for their help in introducing me to individual villagers. Having considered my interests in their village life in the pre-reform period, local cadres usually referred me to the elderly, who had rich lived experience of the collectivization movement in the Maoist era. Elderly villagers loved sharing stories, especially when their experiences were valued. They generously shared with me some interesting stories about their village in the imperial period, which they had learnt from their parents and grandparents long ago. Some of 43  them took me to visit their ancestral temples. Reading the couplets and inscriptions on tablets in these temples, I found a way to understand some of the rituals that had shaped villagers? identities in their clan-based community. Elderly villagers also talked to me about their suffering as well as their understanding of state-village relationships during the Maoist era. Their oral narratives offered me tremendous insights into villagers? daily life during the collective era and how the tolerance and sacrifice of villagers contributed to socialist China?s industrial and urban development. They shared with me not only their own lived experiences but also some historical documentation which they had carefully retained. One elderly villager I met was a retired village cadre who was in his early seventies, but extraordinarily friendly and full of vitality. He kindly showed me two volumes of original land use contracts and told me that he prepared these documents during the early 1980s when he acted as the production team leader in his village. Written in black ink, the first volume collected the contracts for rice fields, and the other collected the contracts for orchards. They were kept in very good conditions, even though the papers were beginning to yellow.   I found that all the land use contracts collected by these documents were made between the production team and individual farming households. The formal lessee was usually the male head of each household. Each contract had two pages. The contract period for rice fields was from the 1st January 1986 to the 1st January 2000. The contract term for orchards was a little bit longer, governing a period from the 1st January 1986 to the 1st January 2005, taking into account the fact that the fruit harvest usually took longer than the rice harvest.  In these contracts, there are four key provisions governing the use of land.  They are: (1) land was owned by the state and the village collective. The Lessee was not allowed to build any building or grave on the contracted land and convert it from agriculture into other uses; (2) the State had the right to requisition the contracted land when necessary; (3) the contracted land should be solely used by the lessee. If the lessee wanted to transfer its land use rights to another, approval was required from the Production Team; and (4) the lessee was required to meet the production quota (gongliang) set by the state each year and pay agricultural taxes. I noted that the amount of land allocated to each household varied. The village leader explained that the amount of land allocated under the Household Responsibility system was determined on the basis of production output per capita. The basic principle was to ensure that each villager could have similar production output from the contracted land. Given that the fertility of each piece of land varied, the actual amount of land allocated to individual households was different from one another.  In 44  terms of land area, each villager was allocated with about one mu of rice field on average. He was kind enough to give me permission to take pictures of all these historical documents for my records and further investigation. It turned out that these historical land documents helped me to understand the operation of the rural collective land system and its relationship to today?s shareholding cooperative system in the GLD.  While searching for a historical perspective to understand local governance restructuring, I also spent time with middle-age villagers discussing the recent changes caused by land requisition and effects of urbanization on their life. Once I started my interviews, contacting more villagers for research was never a difficult task in these transitional neighbourhoods, where kinship ties and family relationships still constituted an essential part of their social organization. After my self-introduction to villagers, we usually started our chat form their curiosity in my background as a Hong Kong person and my study in Canada. Most villagers generously shared with me their family life, their income, and their employment conditions. In general, villagers encountered significant difficulties in looking for jobs in the non-agricultural sector. After losing arable land to urbanization, many female villagers worked as housekeepers in hotels, part-time housemaid in families or janitors in offices, while male villagers usually worked as security guards in the village or gardeners (see Figure 2.8). I asked them why they were not working in the nearby factories. They gave me two reasons. First, they did not want to work in the factories doing repetitive tasks with non-flexible working hours. Second, they didn?t have the skills required by the local enterprises, which mostly hired experienced technical workers. Despite their difficulties in job hunting, their income witnessed significant increases after land requisition. This was because they could secure income from two other sources: share dividends and rental income.    Like their counterparts in the Pearl River Delta region, the villagers affected by land requisition in the GLD are not landless farmers displaced by urbanization; they have become urban villagers who co-own (through the Shareholding Cooperative which was formally known as Production Team) a certain amount of land for commercial uses as a result of the state?s land requisition for industrialization and urbanization. Based on ethnographic research on the ?village-in-the-city? of the Shenzhen Special Economic Zone, Bach (2010: 423-424) describes the life of villagers after land requisitions for urban development as follows:   45  Thinking back to the early years of the zone, the wife of a former village head express gratitude to Deng, for she had been working since she was ten years old, rising at 3 or 4 a.m. and getting to bed at midnight, eating bad food and working until the zone, as the clich? goes, changed everything. . . .  ?Now?, she said with satisfaction, ?we have nothing to do but play mah-jong and the stock market.? They meet in the park every day for exercises at 7:30 a.m., have tea (yum cha) at 8:30 a.m., and go home at 10 a.m. to ?do chores.? The wives were preparing for a shopping trip to Japan, and they travel abroad once or twice a year: They have traveled to six countries in Europe, to South Africa, southeast Asia and Taiwan, and all over China, including five trips to the Yellow River. On their first trip to Europe the village head?s wife spent RMB40, 000 (about $5,800) on gifts; the cost of the trips are covered by the village, which also pays for the village children?s private education, including that of her son who studied in England, returned to the village, and now works in a foreign investment company Bach. (2010: 423-424).   The villagers I met in the urbanizing villages in the GLD were not as wealthy as their counterparts (mentioned in the above passage) in Shenzhen. However, most villagers from the neighborhoods, whose farmland have been requisitioned, agreed that they have been able to sustain a remarkable improvement in their living conditions in the recent years, primarily relying on two major sources of income. One was the dividends from the village collective, which were usually sufficient for their subsistence needs. After land requisitions for industrial and urban development, many villages benefited from the change due to their proximity to industrial and urban areas. The constantly improved accessibility and infrastructural facilities in the area resulted in an increase in the rental value of the ?reserved commercial land.? This led villagers to obtain more dividends at the end of each year. My examination of the operation of the shareholding cooperatives shows that each villager received annual dividends of about 10,000 to 14,000 yuan. Another was the income generated from leasing houses to migrant workers. Villagers could earn more private income as their houses were appealing to migrant workers working in the manufacturing and service sectors in the surrounding areas. In some villages I visited, the ratio between villagers and migrant workers was one to 1.5. Many villagers in these villages have one to two housing premises for leasing; they could enrich their living by generating an additional income of 2,000 to 5,000 yuan per month.     46  However, the villagers in those neighbourhoods which still held a large amount of arable land remained relatively untouched by urbanization did not have these sources of income. Although they had established their own shareholding cooperatives, their cooperatives were somewhat empty shells without having any income from either land compensation or ?reserved commercial land.? Moreover, these villages were mostly relatively far away from the industrial areas so that very few outsiders came to lease houses from villagers. Lacking these two sources of income, many villagers were forced to look for jobs outside the village. I asked why they had given up agricultural production. They said that farming for livelihood was hard. They grew rice twice a year on their small plots. With an ox to help in plowing and raking the earth, other farm works including sowing, watering, pest controlling and harvesting were all man-powered. Moreover, orchards needed intensive care during the harvest season, requiring villagers to work with intolerably long hours everyday to get these works done.  They really did not want to re-live the old days, in which they needed to get up at two o?clock in the morning to harvest lychee. More importantly, their hard work could not bring them fair rewards.   In the late 1980s and early 1990s, it was very difficult for villagers to profit from agricultural production, even though villagers? livelihood improved remarkably due to the implementation of the Household Responsibility System. Throughout the 1990s, the Chinese government adopted an urban-biased development policy and weakened its policy, capital and technological inputs to rural areas. The harvest was good in the GLD. However, farmers had to sell their products in the markets at a very low price. Villagers indicated that their net income from selling vegetables and fruits constantly dropped since the mid-1990s. One villager told me that the price of fertilizer substantially increased from 0.8 yuan per catty to two yuan per catty between the late 1980s and the early 1990s. The income after deducting the cost of fertilizer was not enough for the farming households to make ends meet. They also said that their problems got worse after China?s entry into the WTO, given that local fruits lost competitive advantage when the markets were increasingly dominated by foreign imports of fruits.   In this context, villagers complained that the government?s recent decision to slow down its land requisitions was the main cause to their current conditions of poverty and underdevelopment. They told me that they did not consider themselves urban residents, even though they had been provided with better village infrastructures and social welfare services. They did not view themselves as rural farmers either; they could not enjoy the favorable policies for agricultural 47  production under the recent New Countryside Building project as their hukou status was officially changed from ?agricultural? to ?non-agricultural.? Compared to their counterparts in the villages where arable land had been requisitioned, they apparently felt dissatisfied and aggrieved.  Starting with interviews, I found no difficulties in contacting migrant workers who were interested in my research. Compared with villagers, my interviews with migrant workers were shorter, mainly because most of them usually worked very long hours and thus could not afford time for a longer conversation. Migrant workers living in the villages came from different parts of China, such as Guangxi, Hunan, Henan, Anhui and Qinghai. Most of them I met were working in nearby factories and companies, earning a net monthly income of about 1,500 to 5,000 yuan. Some migrant workers not only lived in villages but also operated small business such as restaurants and grocery shops in the villages. I observed that many neighbourhood restaurants were operated by migrant workers, providing a variety of foods, such as ?Lanzhou Lamian (noodles),? ?Hunan spicy cuisine,? ?Guilin rice noodles,?  ?north-eastern dumpling? and so on. To save money, some migrant workers lived in the restaurant where they worked, even though they knew this was officially prohibited for fire safety reason. Some of them shared with me a main reason why they were given opportunities to do business in villages. They explained villagers showed interests in these small businesses, which demanded intensive work but could not earn a lot of money. Struggling for survival in the GLD, they had different plans for their future. Some of them, especially young people, indicated that they would leave once they could get a better employment somewhere else. Most migrant workers coming with their families wanted to stay but found it very difficult to find schooling opportunities for their children.  Having gained a preliminary understanding of daily governance practices at the grassroots level, I turned to bureaucratic officials for answers to questions about policy making and implementation at the government level. I visited a range of governmental departments at the district level and interviewed the officials who were familiar with policy making and implementation in the areas of land requisitions, social welfare provision for villagers, village collective asset management. Moreover, I interviewed officers from Street Offices who were the frontline staff in converting government policies into daily practice at the grassroots level.   48  Local officials told me about a significant change in what they considered important in relation to local governance. During my fieldwork in the GDD in 2005, local officials only shared with me their experience in dealing with foreign investors. Nowadays, they mostly talked about how to respond to the demands of villagers. They explained that the national policy of New Countryside Construction provided a major spur for them to move in this direction. Following the central state?s call for integrated rural-urban development, the GLD started from 2005 to map out its comprehensive policies for farmer resettlement. Thereafter, they had been preoccupied with policymaking in relation to rural medical reforms, shareholding reforms and the like. From their curiosity in my previous experience as a civil servant in Hong Kong government and my current study in Canada, I learned that they were interested in searching for a perspective from outside their bureaucracy to review their own work. I therefore shared with them my practical experience in Hong Kong?s land use planning and what I had learnt from my school about community planning in Vancouver. After interviews, they referred me to read a range of document materials, such as government working papers, government policy notices, leader speeches, past development layout plans and aerial photographs. Although most of them had been made for public, it was hardly possible for me to locate these materials without their referral.13 It turned out that these documents provide ample data for me to understand how central policy had been interpreted and implemented at the district level. Some data, such as compensation rates and standards, which I expected to be highly sensitive could also be found in these documents.    When I shared my insights and experience into their daily struggles and triumphs, I found an extraordinary willingness on the part of local actors, not only villagers but also village cadres and local officials, to share with me their experience and knowledge. Given the fact that most villagers did not resist land requisitions, I found a need to understand the local governance process with a particular focus on what exactly constituted the conflicts and contestation among local actors. Carrying more than 30 kilograms of notes, books and documentary materials collected from the field, I came back to UBC to analyze the data. Keeping myself away from the GLD for three months, I concentrated on the analysis of villagers? narratives as well a variety of local data, such as local gazetteers, yearbooks, policy papers, and literature written by local authors, through which I sought to deepen my contextual understanding of how development and                                                           13 A list of these documentations is provided in Bibliographies and Citations. 49  governance problems were understood and discussed by local actors. This led me to a more in-depth and interactive dialogue with local people during my refocused field research from November to December in 2012.  Fieldwork: Stage Two   During this Stage Two fieldwork, I focused on six urbanizing villages: four of which had lost all their arable land to urbanization, one of which had only partially lost its land, and one of which remained untouched by urbanization. I found that the problems and conflicts in the first four villages were no less than those faced by the other two villages, even though the population of the first four villages earned a higher level of income. Interviews with villagers suggested that two main issues were at the core of the significant conflicts in the wealthier neighbourhoods. The first issue was related to the allocation of shares. In the early 2000s, many shareholding cooperatives in the GLD started to implement the ?fixing share rights? policy promulgated by the district government. After the implementation of this policy, share re-allocation was disallowed irrespective of the changes in the village population, leading some villagers to be permanently excluded from shareholding benefits. The affected villagers mainly included: villagers who were born after the ?fixing share rights? reforms, female villagers who married indigenous villagers after the reforms, villagers who had violated the one-child policy during the reforms, and villagers who moved out from the village before the reforms. Seeing that the dividends from the village collectives had kept increasing over the past several years, they strove for equal entitlement with other indigenous villagers in sharing the income derived from the land compensation fees and the development of ?reserved commercial land.?   The second issue concerned the so-called illegal structures owned by villagers. Since the 1980s, villagers had built as many houses as they were able to, for both self-occupation and investment purposes. The result was the widespread of ?kissing buildings (jiewen lou)? or ?shaking-hand buildings (woshou lou)?. The distance between the buildings was usually less than a few centimetres (see Figure 2.6). As Chung (2009: 256) describes: ?one can shake hands with, or kiss, neighbours in the next block through window.? Some of them were defined as unauthorised as they were built at the expenses of public space or with a height greater than the 3.5-storey restriction imposed by the government. Over the past decade, the district government had taken 50  severe enforcement actions through the Villagers? Committees against these buildings. This triggered strong resistance because more and more villagers had increasingly relied on the income generated from leasing these buildings for improving their livelihood after they lost their arable land to urbanization.     These conflicts and problems were validated in my interviews with local village cadres. During our conversations, most village cadres repeatedly emphasized that their work was really hard. What made their work different from that of the past, they explained, stemmed from its intermediary role between the government and villagers. On the one hand, they had to adhere to the district government?s supervisions and instructions, by implementing state policies in a range of areas, including but not limited to birth control, building control, social security, conscriptions, and conflict mediations. On the other hand, they were monitored and examined by their fellow villagers who could vote them out in the next grassroots election if they did not act in the interests of villagers. While airing out their grievances, some village cadres shared with me their management philosophies of ?weishang, weizhong, weixia,? which means ?safeguarding the interests of higher level governments, the interests of his colleagues in the Residents? Committees and shareholding cooperatives, and the interest of villagers. Another similar principle was ?lingdao manyi, tongshi manyi, cunmin manyi,? which means ?satisfying the three parties including local leaders in the government, their colleagues as well as their fellow villagers.?  In practice, however, these principles sometimes could not help local cadres find a way out of their dilemma. A few female villagers had filed lawsuits against the village collectives in relation to share allocation disputes. Moreover, control of unauthorized buildings had led the cadres to directly confront villagers. One party secretary I met indicated that he felt extremely frustrated, as his relatives treated him as the enemy after he became involved in demolishing their ?illegal houses.? To demonstrate his impartiality in implementing building control measures, he chose to start enforcement actions against the buildings owned by his sister and his mother?s brother. Consequently, most members in his clan condemned him and refused to maintain contact with his family. Some village cadres complained that villagers did not treat them with much trust and were suspicious of their intention. In recent years, many shareholding cooperatives, especially those holding a large amount of land compensation and ?reserve commercial land,? actively engaged with real estate development projects with the aim of 51  creating more collective income (see Figure 2.5). Some villagers, however, feared that the involvement of local cadres in these projects were to benefit their own purses at the expense of the majority.  Many existing studies tend to emphasize the decisive effect of state-village conflicts in shaping the chaotic complexity of local governance restructuring in periurban China. Based on this field research, I suggest that the internal conflicts between the village collectives and individual villagers constituted equally important dimensions which help explain the transformation of state-society relationships at the local level. When the village cadres and villagers did not co-align with each other, the district government became the major external authority for arbitration and mediation. This saddled the district government with mounting complaints and petitions raised by villagers. Local officials suggested that they were very cautious about these petitions and made every effort to prevent them from escalating into protests or what they called ?mass events (qunzhong shijian).? One local officer shared with me his experience in dealing with a dispute between the village collective and villagers. The cause of the dispute was that some village cadres abused the land compensation fees for private purposes. Villagers consequently vented their anger at the village collective and burned the office of the Villagers? Committee. In view of this severe clash, the district government, while seeking to impose penalties on the cadres concerned, quickly injected money into the account of the village collective and assured the villagers that their collective assets were not affected by the misconduct of village cadres. In dealing with these conflicts, one of the important considerations on the part of the government was that maintaining a socially and politically stable environment for retaining and attracting investment should always be given utmost priority.   These emerging conflicts and disputes were brought to the attention of the municipal government, which always keeps an eye on the social unrest in all districts under its judiciary purview. In recent years, more serious confrontations between village collectives and individual villagers had occurred in many other urbanizing areas such as Baiyun and Zengcheng. In the face of such soaring social tension in the rural sector, the municipal government found a pressing need to explore a new strategy for urban development. In 2012, the municipal government formally put forward its policy of ?New Model of Urban Development (xinxing chengshihua fazhan),? by which the government set out to achieve its intended outcomes in urban development without causing irreversible environmental and social consequences. One of the 52  main programs promoted under this policy was the development of ?Beautiful Villages (meili cunzhuang).? The primary purpose of this program, according to the municipal policy makers, was to reconnect villagers to their land and their place within nature. Liantang village, which was located in the Jiulong Town of the GLD, had been selected as one of the pilot projects for experimenting with this program.   Through connections provided by the GLD Policy Research Office, I was given the opportunity to meet the senior officials from the Planning Bureau and the Policy Research Office at the municipal level and discussed with them these new policies and programs. During our meetings, the municipal officers explained that land requisition was no longer a driving force for urbanization. Recognizing the soaring social tension arising from land requisitions and the fact that more and more traditional villages of historical value had been swallowed up by urbanization, they decided to offer substantial financial input to preserving the village landscape and meeting the welfare needs of villagers without having to acquire village land. I raised two questions in relation to policy and program implementation.   My first question was about the financial feasibility of these programs. They admitted that, subject to fiscal constraints, the government could only start with these policies in a limited number of pilot villages. Moreover, various forms of partnerships with investors from the private sector would be explored to make more projects financially feasible. My second question was about how local planners, who had tended to understand urban development as a construction project rather than a process of social production of space, could help translate these new policies into practice. The officials realized that the education and professional training of local planners were largely grounded in technical skills in relation to urban design and spatial land use planning. In this regard, local planners had been required to receive course training on village culture and traditions. Moreover, local planners were mandated to attend workshops with village cadres and villagers to understand their problems and needs before participating in planning ?beautiful villages.? Taking these meeting opportunities, I also approached more officials for answers to how a range of rural policies relating to land requisitions, shareholding reforms, and other issues had been made at the municipal and provincial levels.   After interviewing the officials, I returned to the villages to meet more local actors and to explore various means towards understanding daily practice at the grassroots level. During my 53  repeated visits to the villages, I spent quite a lot of time reading the bulletins posted on the board outside the Residents? Committee (see Figure 2.9). These bulletins provided data about the daily governance management of the collective assets, such as the collective?s revenue and expenditure, official documents about government policies and regulations, the activities and the decisions of neighbourhood party organisations, and the personal particulars of party member candidates. Local cadres explained that Residents? Committees were mandated to prepare these monthly and quarterly reports and make them available for public inspection. All reports had been scrutinized and endorsed by the Street Offices before they were posted for villagers? inspection.  Moreover, after interviewing and chatting with the villagers, I usually walked around the villages with an open mind to unanticipated events and incidents. Good spots for conducting observations included small restaurants operated by migrant workers, supermarkets, and community recreational centres. During my observations, I paid particular attention to the migrant-workers, who have been playing an increasingly vital role in the local scene but were excluded from shareholding benefits and neighbourhood elections. I sought to understand how they had been integrated into and/or separated from the local village community. Moreover, I tried to understand how they were able to have their voices heard by the village cadres and the district administrators in spite of very limited participation in the formal decision-making process. All these observations together with my intensive interviews provided useful insights for me to understand the lived experience of local governance at the grassroots level.  Getting close to the urbanizing villages in the GLD to understand their inner working from the inside was one of the most fruitful times in my life. From these transitional neighbourhoods, I have learnt much more than I expected. What I have learnt is not only about the problems and conflicts arising from land requisitions for urban development. More importantly, I have learned that the local people have endured and survived these conflicts and difficulties to improve their quality of life. Their daily struggles and triumphs to overcome the constraints derived from the urban-rural dualist system have informed me to pursue an endogenous perspective to examine how the GLD government has dealt with land requisitions for urban development and struggled to enhance its legitimacy through capacity buildings for the attraction of foreign investment, the management of public resistance and the provision of social welfare services.   54            Figure 2.47  Map of GLD     Luogang  Street Dongqu Street    . Bigang   Luogang   .  Huo .  Yushu  . Liu  .  Huangdeng    . Huangma . Xiagang Street  Jiulong Town Yonghe Street Lianhe Street  Hefeng   .  Knowledge City (under construction) Science City   Yonghe Economic Zone  Economic & Technological Development Zone (West Section) and Free Trade Zone     . Xiangang  . Luofeng  . Changping   . Shuixi Huangbei . Tianlihu . Subway  (under construction)         . New Urban Neighbourhood nstruction)  Rural Village  Tianhe Baiyune Haizhuhe Panyu Zengcheng      Tianhe Zengcheng      Tianhe Highway  N Economic & Technological Development Zone (East Section) and Export Processing Zone   55  Figure 2.58   New Urban Neighbourhoods in the GLD      New apartment buildings for villagers                                              New houses for villagers                    Neighbourhood football field                                                        Neighbourhood library                    Neighbourhood market place                                              Villagers? houses for leasing      56  Figure 2.69   ?Kissing? and ?Shaking-hand?  Buldings                          57  Figure 2.710  Development of ?Reserved Commercial Land?         Figure 2.811  Job opportunities for Villagers           58  Figure 2.912   Residents? Committees                                Office buildings                                                                   Office building            Monthly/Quartely finaical reports posted on the boards outside the Residents? Committee               59  Chapter 3     Guangzhou Luogang District?s Rural-urban Transition        Based on a micro-historical approach, I begin my study with a comprehensive review of GLD?s urban transformation over the years. The history of the GLD reaches back to the pre-Song dynasties. This chapter endeavours to understand the process of how the GLD has embarked on its transformation from a village-based society to an urban area as a result of the local state?s massive land requisitions for industrial development. A vivid portrait of the GLD?s development over time enables me to understand the historically derived institutions which have constrained and enabled urban transformation. Such an understanding sets the foundation for me to explore the deep causes of local land conflicts among local actors in the process of urbanization.    Historical Origin of the GLD  The recent archaeological excavations near Huo village and Xiangang village suggest that village settlements first appeared about 3,000 years ago in the GLD. However, reliable information in the Luogang Town gazetteer only traces the history of the GLD back to the Song dynasty (960 -1297), during which the Zhong family moved to build Lougang village in the area. To better understand the origin of the GLD, I draw insights from the historical trajectory of Guangzhou City.14 During the period from 203 BCE to 111 BCE, Guangzhou was the capital city of an independent kingdom, which was called Nanyue Nation. Geographically, Nanyue Nation was located in the Lingnan Region, which we toady know as Guangdong Province and Guangxi Province. Situated in the southern part of China, Lingnan was separated from the central plain by five high mountains, named, Dayu, Qitian, Dupang, Mengzhu and Yuecheng. Before the Qin dynasty (221 ? 206 BCE), a spectacular civilization had been created in China?s Central Plain China along the lower reaches of Yellow River. Lingnan, however, was still considered by the Central Plain rulers as the land of ?barbarians (manyizhidi).? After taking                                                           14 I pieced together this historical survey after spending several afternoons touring the generous exhibits of the Guangzhou City Museum ?Guangzhoubowuguan,? which is located in Zhenhai Tower, and familiarizing myself with the development history of the city.  60  control over the Central Plain area, Emperor Qin-shi-huang sent a troop of 500,000 soldiers to march south with the aim of expanding his territory to the Lingnan Region.  River transport was the only option at the time because the difficult terrain of the five high mountains made land transportation impossible. When the Qin army arrived at Lingnan through Xiang River and Xi River, Tusiu, the commander of the troop, died in a battle with the indigenous tribes. Renxiao, succeeding Tusiu as commander, led the Qin army forward and finally conquered the region. In 214 BCE, Qin-shi-huang established three prefectures in Lingnan: Nanhai Prefecture (Nanhai Jun), Guilin Prefecture (Guilin Jun) and Xiang Prefecture (Xiang Jun). Due to his success in conquering the region, Renxiao was granted the power to rule all prefectures, even though he was only nominally the magistrate of Nanhai Prefecture. In the Qin Dynasty, the governance of the prefecture was divided among three posts: the Junshou which was responsible for local policy making and administration, the Junwei which was responsible for local military affairs, and the Junjianyushi which was responsible for monitoring the performance of the local administration. However, in the three prefectures of the Lingnan Region, only the post of Junwei was created.  Consequently, Renxiao as a Junwei of Nanhai Prefecture became the de-facto supreme leader, having a monopoly of the administrative and military power over all three prefectures.       After settling Lingnan, Renxiao began to build the capital city of Nanhai Prefecture in Guangzhou, which was known as Panyu at that time. According to the archaeological discovery, the city built by Renxiao had a total area of about 0.09 square kilometres. Renxiao?s site selection was itself a puzzling decision: why did he not build up the capital city somewhere else in Guilin Prefecture which had been much more developed than Guangzhou at that time? During the Qin dynasty, the economic and political center of Lingnan was indeed located in Guilin Prefecture, which is now known as Guangxi Province. In a close proximity with Chu Nation, which was one of the Central Plain vassal states during the Spring-Autumn and Warring Periods, Guilin Prefecture was the main hub for north-south transportation. Locating at the intersection between Li River and Xi River, Guilin was the main hub for north-south transportation. Moreover, in a close proximity with Chu Nation, which was one of the Central Plain vassal states during the Spring-Autumn and Warring Periods, Guilin Prefecture achieved better economic development than other parts of the Lingnan Region through frequent economic and social interactions with the Central Plain.   61   Comparatively, then, Guangzhou was located in a desolate place. Its development as a city was apparently subject to two physical constraints. One was the shortage of fresh water resources. The other was the lack of space for city expansion. Guangzhou was situated at the mouth of the Pearl River where it flows into the South China Sea. Two thousand years ago when marine transportation was not yet developed, locating Guangzhou at the mouth of the estuary did not provide the city with a geographical advantage for development. Conversely, the location caused a problem of drinkable water shortage at the mouth of the Pearl River where fresh and salty water mixed.  When land filling was technically not a possible option, Guangzhou?s expansion was severely restricted by the extensive coverage of water bodies in its vicinity.     With the passage of time, Renxiao?s rationale for choosing such a seemingly inhospitable place revealed itself. Guangzhou was a good location to Renxiao as he orchestrated to build his own kingdom rather than a capital city of Nanhai Prefecture. In this regard, what Renxiao really wanted was not a place ideal for urban development, but a strong foothold which was far away from the sway of the Central Plain power, making it easy for him to defend and difficult for invaders to attack. As described by a Chinese proverb, ?mountains are high and the emperor is far away.?  Separated from the central plain power, Renxiao had great autonomy in local governance, by which he could manage to realize his plan of seeking independence of the emperor. In his last will, Renxiao entrusted his subordinate Zhaotuo to implement his plans of building an independent kingdom in Lingnan. Zhaotuo was the former deputy commander of the Qin army. When Nanhai Prefecture was founded, he was appointed as the magistrate of Longcuan County of Hainan Prefecture. In succeeding Renxiao, Zhaotuo built up his power and territory, partially through alliances with the native Yue nobility and chieftains. In 204 BCE, Zhaotuo appointed himself as the King of Nanyue Kingdom (Nanyuewang) and continued the unfinished undertakings of capital city construction left by Renxiao.  To conceal his rebellious intentions, Renxiao deliberately kept his capital city small. The city built by Renxiao had an area of less than one square kilometre and is referred to as the ?Renxiao City? in many historical achieves. Extending from the base of Renxiao City, the city built by Zhaotuo had an area of 32 square kilometres. To overcome the problem of fresh water shortage, construction of surface wells constituted a core element of Zhaotuo?s city building.  It is said that Yuewang well, which is located in Yingyuan Road of today?s Guangzhou City, was built more than 2,000 years ago for Zhaotuo?s palace. 62  In 202 BCE, the Han succeeded the Qin. In the early years of the Han Dynasty, the Han emperor sent diplomatic agents to persuade Zhaotuo to surrender to the Han. In 196 BCE, Zhaotuo yielded to the call from the Han emperor and revoked his own title of emperor. Nanyue became a vassal state of the Han, but these changes were somewhat superficial as Zhaotuo retained the autonomy of his Kingdom until he passed away in 137 BCE at the age of ninety-three. After the death of Zhaotuo, the Han emperor sent a troop of 100,000 soldiers to conquer Nanyue in 111 BCE. This battle brought ruin to the Nanyue Kingdom as well as Guangzhou City. To reinforce the central control over the region of Lingnan, the Han emperor abolished Nanhai, Guilin and Xiang Prefectures and restructured the region into nine prefectures, headed by a Cishi, who was directly appointed by the central government. Under the direct supervision of Cishi, magistrates of prefectures were deprived of all military power and were granted with administrative authority only. With this administrative restructuring, Guangzhou was placed under the jurisdiction of Cangwu Prefecture. In order to eliminate the residual power of Nanyue Kingdom, the Han government chose another city named Wuzhou to replace Guangzhou City as the capital city of Cangwu Prefecture.    Following this political change, Guangzhou?s development was suspended for nearly 300 years until Buzhi, the Cishi of Lingnan region, decided to move the capital city from Wuzhou back to Guangzhou in 217.15 When the city?s rebellious history was gradually forgotten and marine trades began to grow, Guangzhou came to resume its importance on the national stage, taking advantage of its location at the delta of the Pearl River. In the late Han dynasty, China began to send fleets of ships to the Indian Ocean via the South China Sea with the aim of exploring trading opportunities. The subsequent development of marine transportation and trades shaped Guangzhou to become a very important port in southern China. During the Tang dynasty, the construction of the Meiguan Road provided another opportunity for Guangzhou to further evolve into an international commercial centre in the succeeding centuries to come. The completion of the Meiguan Road provided a passageway through the five high mountains, improving land transportation between the Central Plain and the Lingnan region.16 After the completion of the Meiguan Road, many imported and exported goods stopped over in Guangzhou before they were                                                           15 According to the archaeological records, the city rebuilt by Buzhi was about 460,000 square meters. 16 The construction of the Meiguan was launched by Zhang Jiuling (673-740), who was served as the chancellor during the reign of Emperor Tangxuanzong. Zhang was originally from Shaozhou qujiang (today?s Shaoguan City of Guangdong Province) and thus realized how the five high mountains had created the major obstacle to the communication between Lingnan and the Central Plain. After retirement, he returned to his hometown and initiated the Meiguan project. 63  moved on to other destinations. As revealed by Pan?s (2010) research, more than 120, 000 foreigners resided in Guangzhou for business during the Tang dynasty, constituting about one third of the city?s total population. The commercial prosperity of Guangzhou city in the Tang dynasty was also reflected in the fact that the central government set up the first customs and excise department to be headed by shiboshi to levy taxes on the prosperous marine trading activities.   Such specific local contexts shaped Guangzhou as the most open Chinese city in history. During the Tang dynasty (618 ? 907), China reached the apex of its imperial era. Under a centralized powerful dynasty, not surprisingly, the city was stringently regulated and controlled. For example, Chang?an, Tang?s capital city, was divided into walled wards and commercial activities were heavily restricted, in both time and space, to the official markets. As Heng describes,  Siu-Tang Chang?an was extremely controlled and divided into large enclosed wards by extraordinarily wide streets. These closely patrolled streets become vast ?no-man? lands at night. The people of Changan who live in wards were subject to stringent supervision and forbidden to leave the wards during curfew hours. Access to the streets after dark was strictly regulated. Guards stationed in police posts located at the junctions of avenues at the corners of the fang enforced compliance. The main streets were also devoid of commercial activities, which were restricted to the fortress-like East and West Markets in the City. There, trading was permitted only during certain hours of the day (Heng, 1999: 205).      While Chang?an was implementing this rigid system during the Tang dynasty, Guangzhou was adopting a much freer urban layout in which trade and commerce were allowed anywhere within the city (Zeng, 1991; Pan, 2010). Political culture rooted in the Nanyue Kingdom?s history of rebellion, together with the mentality of profit-making derived from being a commercial centre during the Han and Tang dynasties, influenced Guangzhou?s governance culture to be more liberal and accepting of dissent than other parts of China. Moreover, the commercial prosperity of Guangzhou city during the Han and Tang dynasties supported the early emergence of village settlements in the GLD area. Located only about 30 kilometres away from the urban center of Guangzhou City, it was more likely that the present-day villages in the GLD had already been 64  the hinterland of Guangzhou City during the pre-Song dynasties, even though local historical publications, such as the Luogang Town Gazetteer and some village gazetteers, traced the history of these villages to the Song dynasty, which was the peak of immigration into the area from elsewhere in China.   Village Society and the Urban-rural Continuum  In the South Song dynasty years, widespread warfare in northern China forced many people to move to the South. Many migrants reached the Lingnan Region through the Meiguan Road. Most of them first stopped over in Zhujixiang, which is located in today?s Nanxiong City, and then moved to the other parts of today?s Guangdong Province. The GLD area was no doubt one of the ideal destinations for these migrants. Located just south of the Tropic of Cancer, it has a humid subtropical climate influenced by the Asian monsoon: the summer months (July and September) are humid with high temperatures, and the winter (December ? February) is moderately warm and dry. A humid subtropical climate, together with the good soil conditions, make the region ideal for growing a range of fruits, such as oranges, lychees (i.e. Litchi chinensis), olives, plums and pineapples. As some villagers suggested, they have a long history of growing these fruits.   According to the Luogang Town Gazetteer, the Zhong family, which was one of the biggest clans, immigrated there in 1163. Gradually, more families with different surnames, such as Liu, Kong, Huang, He, Luo and so forth, arrived to build their so-called ?natural villages.?  Maintaining neighbourhood watch and helping to defend each other became practical strategies for these new migrants to settle and flourish in a new place. In this regard, it was a trend for different families with different surnames to come together in forming a village. Taking Xiangang Village as an example, three big families including Kong, Huang, and He have lived amicably together here for centuries. Villagers from each clan family built their houses alongside ponds. Each family inherited their position around the pond?s edge from their ancestors. Even without the force of law, generations respected this layout and ensured that the ponds were never encroached upon. As a result, ponds were equally shared amongst villagers, occupying a central position in village life. Villagers shared not only the water resources for watering plants, washing cloths, putting off a blaze in case of fire, and so on. They also collectively viewed the 65  ponds as anchoring their families together and giving them a sense of belonging to the place. Such ?pond-centric? layout was common to the villages in not only the GLD but also other parts of Guangdong Province (see Figure 3.1).   Because of multiple clans, a number of temples were usually built in one village to worship different ancestors and gods. Each clan had their own traditions and norms governing weddings, funerals and conflict resolution. Villagers from the same family, usually collectively, held ancestor worship every three years to pray for good fortunate (Zeng, 2010). ?Pork Delivery Ceremony? usually constituted an important part of ancestor cults. After the collective worships, the pig, which was used as oblation, was divided into pieces and delivered to villagers. Nowadays, some villages in the region still follow this tradition. However, only male villagers could receive pork pieces. Female villagers were not entitled to such benefits, reflective of the male-dominated clan relations in the Chinese patriarchal tradition. These clan activities, on the one hand, expressed tribute to the ancestors and suggested the importance of filial piety in maintaining social order. On the other hand, they represented a symbol of tribal unity and maintained cohesion and sense of belonging among clan-based families. Ancestral trusts provided the financial costs for these ancestral cults and religious ceremonies. Most of them consisted of both land and buildings collectively owned by villagers. These collective properties were leased to individual village households to generate regular incomes for maintaining the operation of the trusts.   Different families and clans maintained their own identities. Yet, they established firm social bonds by inter-marriages and residing in close proximity to each other to create a sense of community. Community ties were therefore developed based on kinship and neighbourhood, conditioning the behaviours and social interactions of villagers. As kinship networks were organized along ancestral lines and family surnames, being born within an indigenous residence itself constituted the eligibility of participation and entitlement to the collective assets of the village. When kin relationships expressed commitments to a long-term reciprocity, a kind of primitive collectivism was developed, by which social exchange was mostly carried out on the basis of mutual trust rather than market value calculation (Hebel, 2004). Community recognition legitimized local governance power to exist and operate according to village norms and traditions. In these villages, those who could gain the recognition from villagers were mostly 66  local elites from big families, which had members holding academic degrees and/or bureaucratic offices in urban areas.      In imperial China, it was not uncommon for people, both rich and poor, to work in cities and later return to their villages (Skinner, 1977). This was especially the case for the bureaucratic elite class. Many of them were from rural areas, and then gained academic degrees and bureaucratic office in the city with the financial and spiritual support from their rural family and lineage. They were always an honour to their native place, and thus usually hoped to return to their place of origin in the autumn of life. This was also true in the GLD region. As revealed by the genealogical book of the Liu family, Liu Liang (1030 - 1112), the primitive ancestor of the Liu family, served as the Defense Minister of the Songhui Reign of the Song dynasty and eventually retired in Liu village. In the Zhong family, similar stories were not only recorded in the family genealogical book, but also posted on the walls of the current Luogang Residents? Committee office. In 1205, Zhong Yu-yan got the degree of Jinshi in the palace examination and then was appointed as the Magistrate of Wuchang Prefecture. After retirement, he returned to his home village in 1219. During the Yuan dynasty, Zhong Zi-huan, another member of the Zhong family, held a professorship in Suzhou University and eventually retired in his clan village. These retired bureaucrats and scholarly elites not only underpinned the continuum between rural villages and urban areas. They also played a critical role in village governance, devoting their efforts to build temples and schools. Earning rapport and esteem from their fellow villagers, they and their descendants became the local leaders, or what we now call ?local gentry.? Like other local elites in rural areas, the local gentry in the GLD region played a crucial role in mediating between their self-governing villages and a centralized empire throughout the late imperial era.    The Song dynasty paved a solid foundation for the development of a village society in the GLD region. It was also a critical turning point for China?s dynastic era, when the empire gradually experienced a significant decline in its power to effectively govern local affairs. Building upon the war-torn Five Dynasties (907-979), the Song dynasty was no longer as powerful as the Han or Tang, even though commercial activities were allowed to develop under a pluralistic and mercantile society managed by pragmatic bureaucrats. The Song?s first ruler (Songtaizu), who was a commander of a battle group in the Late Zhou, proclaimed himself emperor through mutiny. To prevent another coup by military leaders overthrowing his regime, Songtaizu did not wish to see the growth of the army. The early Song, in general, saw efforts made by the Song 67  court to weaken the authority of military leaders in the governance structure (Qian, 2002). With the national diminution in military power, what logically followed were relentless political instability and incessant invasions by the Jurchen tribes from the northern border. As a result, the Song dynastic power to control local development was largely eroded. This was evidenced by a number of significant changes, such as the appearance of commercial activities outside the markets, the tearing down of walls surrounding wards, the encroachment of commercial uses into street space, and so forth. The weakening of controls over urban life gave birth to a new setting in which open urban forms were able to grow. This transition was termed the ?medieval revolution in urbanization? by Mark Elvin and the salient features of the revolution are well documented in Skinner?s The City in Late Imperial China (Skinner, 1977: 23-24):   (1) a relaxation of the requirement that each county could allow only one market; (2) the breakdown and eventual collapse of the marketing organization; (3) the disappearance of the enclosed marketplace, along with the warded-wall system, and their replacement by ?a much freer street plan in which trade and commerce could conducted anywhere within the city or its outlying suburbs?; (4) the rapid expansion of particular walled cities and the growth of commercial suburbs outside their gates; (5) the emergence of ?great numbers of small and intermediate-sized towns? with important economic functions.    In line with the declining state power in the late imperial period, the local gentry in the GLD area played an increasingly important role in directing village affairs. They not only shouldered ritual activities, conflict resolutions, charity works and public security. They also actively got involved in the construction and management of local markets.  In Chinese agrarian society, local markets were crucial for local development as they knit farming households into a massive economic network. China has a long history of agrarian civilization for more than 2,000 years. The foundation of this prolonged civilization was an intensification of a farming system sustained by numerous diligent households working fields from generation to generation. Hsu (1980) provides two reasons for the formation and the development of this farming system in imperial China. First, with a huge population on a relatively limited land area, what the society could afford was labour-intensive agricultural production. The second reason was related to how the peasant class was positioned in the Han Dynasty society from which the foundation of China?s agrarian civilization was formally established. As he suggests,  68   There are always two options regarding farm sizes: either individual farming households working on small plots, or large-scale farming operations under manorial, plantation, village commune, cooperatives or other arrangements. The Han government felt it initially necessary to maintain farmers as the foundation of the empire and to allow no part of the private sector to control the human resources. The attitude (which I think had its roots in both Confucian and the Fa-chia ideology) prevented the appearance of manorial plantations under the control private landlord (Hsu, 1980:150).    Dynastic changes occurred from time to time. The family farm system, however, remained as the basic institution of rural China. As farming households could not produce everything as they needed and had to meet their subsistence through exchange, farm trading could be fairly intensive in Chinese agrarian society (Hsu, 1980). In the late imperial era, the growing demands for trading of not only agricultural surplus but also sideline products produced by farmers gave rise to the rapid growth of market towns throughout the country. During the Ming and Qing dynasties, rapid population growth created surplus labour in rural areas, especially in the Lower Yangzi Region and Southern China (Ho, 1994). These people, who were released from direct farming activities, gradually shifted to focus on non-agricultural production such as weaving and textile, pottery, and so forth. The booming of non-agricultural production created demand for sales networks and thus led to the emergence of many market towns.  Luogang market town was just one of many. It was built by the Zhong family in 1431. This town served the local villages within a 20 kilometre radius. During markets days, villagers came to exchange their agricultural products, such as poultry, fish, reptiles, pigs, fresh fruits and vegetables. The markets were opened for six days each month during the Ming dynasty. If villagers missed the market days during the first two weeks of a month, they had to wait for the next three-day, which were set for the 3rd, the 6th, the 9th day of the last two weeks. In the Qing dynasty, the market was further expanded. As described by Liang and Cheng (2010), there were restaurants, tea houses, clinics, porcelain shops, groceries shops, coffin shops and so on in the Luogang market town, through which villagers could exchange their farming surpluses for goods from other parts of China. What logically followed the thriving power of local gentry were their efforts to renovate the primitive ancestral halls. In the late Qing Dynasty, local gentry from the Zhong family invested a huge sum of money in renovating their ancestral hall, which was 69  symbolic of the rising local authority deeply rooted in their lineage organizations. In 1840, a ceremony was held to celebrate the completion of the renovation. Zhong Feng-qing, who was an influential local leader, inscribed words on the tablet to commemorate this great moment in the clan family.   In 1911, the Qing dynasty was overthrown and the republic was proclaimed. These dramatic changes did not immediately bring an end to the power of the local gentry, but foreshowed its steady decline in the subsequent decades. After the collapse of the Qing dynasty, the nationalist government of the republic was unable to effectively extend its sphere of control from its urban bases to the countryside, allowing the local gentry to continue their dominant role in managing village affairs. However, the power of the local gentry to effectively govern villages was unlikely to endure, when the entire country was locked in turmoil due to feuding warlords, the Japanese invasion, and the civil war that arose after the collapse of the Qing Dynasty. Between 1911 and 1949, the villages in the GLD region witnessed harassment from various military forces fighting for the control of Guangzhou City (Zhong, 2010; Zhu, 2010). As revealed by the Luogang Town Gazetteer, banditry was rampant between 1927 and 1928, and more than 80 villagers were kidnapped for ransom. In 1931, a smallpox outbreak caused more than 100 deaths. During 1939 and 1940, more than 200 adult male villages were drafted by force for military service. In 1941, the Japanese army destroyed many houses of villagers and the light railway connecting the GLD region with Guangzhou City. In 1942, some villagers were forced to grow opium poppies to make a living. Amid this widespread warfare and social unrest, the local gentry lost their ability to maintain social order and to protect their fellow villagers. Most of them just sought self-protection rather than striving for the collective interests of their follow villagers. They imposed heavily on arable land in order to secure their own wealth. As a result, local gentry, who used to win support and respect from the local populace, became greedy landlords who exploited the hard toil of villagers.  At the end of the 1940s, the CCP?s working groups entered the GLD region to establish and expand Peasant Associations (Nongmin Xiehui) in each village.17 Association members were mostly selected among poor farmers. Local knowledge of these selected villagers helped to                                                           17 According to the General Regulations on Peasant Associations of 1950, Peasant Associations were self-governing mass organizations. However, they were delegated with the legitimate executive authority by the Land Reform Act of PRC to implement the CCP?s land reform policies and programs in village areas.     70  determine the class status of each village household according to their landed properties and to mobilize villagers to participate in ?struggle meetings? to condemn landlords for their misdeeds. Compared with villages in northern China, land reforms in the GLD region were on the whole less violent. As suggested by some old villagers, one reason was that quite a large portion of land had previously been owned by ancestral trusts rather than private landlords. Land reform therefore mostly involved privatizing those publicly owned assets rather than taking away private properties held by individuals. Under the leadership of Peasant Associations in the region, land reforms were completed in 1952. The CCP township government and the village administrative structures were subsequently established. Members of Peasant Associations were appointed as village cadres, who were going to take up the role of local gentry in village governance.   Agriculture Collectivization in the Maoist Era     After land reforms, private land, farming tools and other properties confiscated from all landlord households and some rich peasant households were redistributed to poor villagers. Farming on their private plot, most villagers were provided with incentives to improve agricultural productivity. Moreover, villagers were encouraged to open up new farmland on hill slopes to grow olives, plums and pineapples with the aim of generating extra income. At the same time, many farming households formed a mutual-aid group (usually including five to six households) by pooling their labour, farm tools and draft animals together for the growing and harvest seasons. As some villagers recalled, after land reforms, most farming households had enough food for own consumption and even had some excess agricultural products sell on the market. However, such good times did not last, as the central government started to promote collectivization in the early 1950s.  In what follows, I will explore how the national collectivization movement gave rise to the urban and rural dualist system in the region and its impact on the daily life of villagers.   After the founding of the People?s Republic of China (PRC) in 1949, the CCP strived to achieve a rapid process of heavy industrialization with the aim of building a powerful, modern country. Cities in China were therefore ideologically positioned as the industrial bases for socialist industrialization. To develop heavy industries such as steel works from a war-torn economy 71  would be an extremely difficult task. The CCP?s strategy for the development of heavy industries, however, was unshakable. It laid out such a strategy in the First Five-Year Plan in 1951,18 arguing that the goal of national industrialization could still be achieved by relying on the very limited domestic capital and resources. Lin et al. (2000) define this as a ?catch-up development model,? which struggled for speedy industrialization to catch up with the development status of Western countries irrespective of domestic resource constraints. In 1953, the central government formally launched the collectivization campaign, aiming at establishing a centrally controlled economic system, with which it could decide not only the goals of all production activities, but also how to work towards these goals and under what kind of rules and systems.   As an important measure for the promotion of collectivization, the central state first adopted a ?tongshou tongzhi? policy to nationalize the banking system through unified collection and allocation of banking deposits in 1950. Under this policy, all deposits gathered by all levels of banking had to be finally transmitted to the central bank (i.e. the People?s Bank of China) and then all loan and credit businesses were centrally managed by the central bank. Under this system, the state could decide who would receive loans and how much they would get. Simultaneously, the central state imposed direct restrictions on foreign exchange to secure its power in controlling import and export activities. Under the foreign exchange control policy, the conversion of foreign currencies required the approval of the state administration of foreign exchange. Having seized control over the financial system, the central government started to engage in sweeping collectivization of private enterprises. One important step taken by the state was to monopolize the supply of raw materials and mining resources. All organizations had to apply for a license or permission from the government if it wanted to use these resources. Another important step taken by the state was to directly intervene in the labour market by imposing severe mandatory control over the pay structure, the pay scale, and the pay adjustment mechanism (Lin et al., 2000). Through these polices and measures, the Chinese government found a way to turn nearly all private firms into state enterprises by the end of 1953. These state enterprises, which were called work-units (danwei), acted as agents to operate a centrally controlled economy to ensure that all possible resources would be channelled to the development                                                           18 The implementation period of the plan was 1953-1957. Revisions were made during the implementation, and thus its final version was not available until 1955.   72  of heavy industries. Thereafter, most urban residents lived their life under work units over the next several decades.   While collectivization was widespread in urban areas, rural China was undergoing the same movement. The main purpose of the collectivization movement was to establish an institution with which the state could extract rural surplus through the levy of agricultural taxes and the requisition of low-priced farming products for industrial development in urban areas. The first stage of collectivization took the form of ?Elementary Cooperative? (Chuji Hezuoshe). The size of an Elementary Cooperative was about 30 households. Cooperation under the Elementary Cooperative had certain noteworthy features. First, when farming households joined a cooperative, they brought in their means of production, such as labour, land, farming tools and the like. These resources were then centrally managed by the cooperative in the production process, but they were still under the ownership of individual farming households. Cooperative members indeed had a dualist identity: employees and shareholders. They were organized by the cooperative to work together and then received incomes according to their shares and their hours of labour. In other words, they were able to benefit individually and collectively from agricultural production and other economic development undertaken by their cooperative as a legal entity. Closely associated with this institutional arrangement, the second feature was about income distribution. The income attributable to cooperative members referred to net income, which was calculated by taking revenues and adjusting the costs of running the cooperative, the agricultural tax, the food supply quota required by the state, the funds for pensions and social security expenses. In practice, the distribution of net incomes was carried out in two ways: (1) dividend payments for land, draft animals, and farm tools contributed by each farm household and (2) remuneration for work performed by each individual member. Moreover, the principle of equal remuneration for women and men was adopted to promote gender equality so as to win the support of women (Ngo, 2009). Last but not least, participation in an Elementary Cooperative was voluntary. After joining the cooperative, participant farmers still had the right to opt out and get back their assets if they decided to withdraw.   The voluntary nature of this cooperative campaign constituted a key contributing factor for its success at the initial stage. In order to benefit from the economy of scale in production, more and more farming households were willing to become cooperative members. In 1955, there were about 633,000 Elementary Cooperatives throughout the country (Li, 2004). Moreover, given that 73  participation was on a voluntary basis and members had the right to opt out, local cadres were compelled to perform effectively, with the aims of retaining their members and convincing other farming households to join. Last but not least, the organizational setting of the Elementary Cooperative basically followed that of the traditional village. Conventionally, cooperation and collaboration among farming households in agricultural production were mostly confined within a village, in which the social structures were organized along kinship and clan lines (Myers, 1970 and 1975). When production activities took place in such a society, participants tended to exercise self-discipline, as they knew each other very well (Ye, 2009). This mechanism was of great importance to ensure effective individual performance in a small-scale family farming system, under which farming work was not easily quantified and monitored (Lin et al., 2000).  Table 3.13    Collectivization and Collective Organizations   Voluntary Collectivization   Compulsory Collectivization    Elementary Cooperatives (1953-1956) Advanced Cooperatives (1956-1957) People?s Communes (1958-1960) Commune-brigade-team  (1961-1978)  Organizational Size (No. of households) 20-30  150-170  about 5,400  Commune: 1,500-1,800  Brigade: 150-170  Production Team: 20-30  Form of Participation   Voluntary  Voluntary   Compulsory Compulsory Land and Resources Ownership   Private   Collective ownership with limited private  ownership   Collective ownership only Collective ownership with  limited private ownership  Farmer?s Identity  Cooperative member, employee, shareholder   Cooperative member, employee, shareholder  Commune member/employee  Commune member /employee Income Distribution System  Dividends payment and Work Points  Work Points Average distribution of daily necessities and Work Points   Work Points  Organizational Boundary  Within village boundary  Beyond village boundary  Beyond village boundary   Within village boundary   Having been encouraged by the initial success in promoting the cooperative campaign, the Chinese government quickly moved towards the second stage of the cooperative campaign ? the ?Advanced Cooperative? (Gaoji Hezuoshe). In 1956, the central government issued the ?Model Bylaws for Advanced Agricultural Production Cooperatives.? According to these bylaws, all assets of cooperatives, which were originally owned by cooperative members, were collectivized. This meant that individual households after becoming cooperative members could no longer 74  retain private ownership of land and other production resources. In terms of organizational scale, an Advanced Cooperative was about seven times the size of an Elementary Cooperative (see Table 3.1). Under the Advanced Cooperatives, dividend payment for lands was repealed as all lands had been collectivized. Income distribution was therefore merely based on work points (Zweig, 1997; Unger, 2002; Ye, 2009)19. Legally speaking, farmers still had the right to opt out. In practice, however, they were usually prevented by the local officials from doing so especially at the later stage of establishing Advanced Cooperatives.   The ongoing attempt to enhance the scale of rural cooperative organizations from ?elementary? to ?advanced? was in essence a transitional arrangement leading to compulsory collectivization in the communal era. From 1949 to 1952, the total grain output of the country increased. The food stores of the state, however, decreased as farmers were not willing to sell their products to the state at a low price. With private land ownership which was derived from the CCP?s land reforms, farmers enjoyed a high degree of autonomy to make decisions on production and marketing of their products (Wu, 2003; Li, 2004). Due to economic consideration, they preferred to sell their output on the private market rather than to the government which could only afford relatively low prices. To overcome this dilemma, the central state introduced the ?centralized purchase and sale (tonggou tongxiao)? policy to monopolize the procurement and marketing of agricultural products in the countryside. The state monopoly was initially applied to grain, cotton and oil only. Under this policy, farmers had to sell their farming outputs to the central state at prescribed prices, as the private market for farmers to exchange their agricultural produces was almost entirely replaced by the official market.20 In 1955, it was quickly extended to cover nearly all agricultural products. At the same time, the central government started to impose restrictions on rural-urban migration through a number of policies and institutions to restrict farmers? physical and social access to cities. As shown in Table 3.2, legislation was introduced in 1953 to enable the state monopoly of food markets in urban areas. When farmers were not                                                           19 Work points were primarily calculated on the basis of the time committed by an individual member without address the actual work output that had been accomplished.  20 In 1953, a speech given by Chen Yun further clarified the intentions of this monopoly policy: ?the government would be beggared if it continues to rely on the free market to acquire agricultural products. We had considered other alternatives, but we could not come up with a better choice than requisition. What can we do? If we use our very limited foreign exchange reserves to import food and raw materials, we will not be able to import machines and other equipment for industrial development. Our national industrialization will be hopeless? This speech was collected in the Selected Literature of Chen Yun: 1949-1956 (Chen Yun Wenxuan: 1949-1956) published by the Renmin Publishing House in 1984. 75  able to buy food from the private market, it was difficult for them to stay in cities. Thereafter, more and more restrictions were imposed on rural-urban migration.   The ?tonggou tongxiao? policy enabled the central government to control the supply of food and raw materials. Nevertheless, the government quickly discovered that it was extremely costly to implement requisitions which involved dealing with more than 750,000 rural cooperatives throughout the country (Lin, 1990). To reduce these transaction costs, the government was in urgent need of a more centralized system, through which it could seize absolute power over not only the production, but also over the distribution of agricultural products in the countryside. This was one of the key reasons why the Chinese government went for a communal system at the end of the 1950s (Li, 2004). Apart from food requisition, as Lin et al. (2000) suggests, the compulsory collectivization movement was rooted in another important notion of better mobilizing rural labour for the construction of large infrastructure projects, such as irrigation system and dams.   Table 3.24   Important Regulations on Rural-urban Migration: 1953-1958  Year  Ordinance / Regulations  Key Provisions  1953   Provisional Management Methods for Food Market (Liangshi Shichang Guanli Zanxing Banfa) All private sectors were not allowed to engage in food-related business after the introduction of the ?tongxiao tonggou? policy.?   1953  Revised Labor Insurance Ordinance (Liaodong Baoxian Tiaoli)  Only urban residents were entitled to state-subsidized health care services and old age pension.   1955  Provision Measures for Food Supply in Cities and Towns (Shizhen Liangshi Dinglian Gongying Zanxing Banfa)  The state was responsible for the supply of grain and oil for urban residents only.   1957  Provisional Regulations on Work-units? Recruitment of Temporary Workers from Rural Areas (Guanyu Ge Danwei Cong Nongcun Zhong Zhaoyong Linshigong de Zanxin Guiding)  All work-units should not recruit temporary workers from rural areas or farmers who have streamed into cities; all state organizations in rural areas should not recommend farmers to work in cities or mining areas.   1958 Household Registration Ordinance (Hukou Dengji Tiao Li)  Only holders of employment letters issued by (1) the Labour Department, (2) admission letters issued by educational institutions or (3) permission issued by the Hukou Registration Department could move into cities.    Collectivization and Rural-Urban Dichotomy   In 1958, the household registration system was formally established. As a result, villagers became peasants with an ?agriculture? hukou, that denied them access to cities and state welfare 76  such as subsidized housing, education, health services, and so forth. The unequal hukou status was hereditary. It was so enmeshed with the life of rural farmers that the inequality between urbanites and farmers was passed on from generation to generation. This strong emphasis on differentiating the legal identity of urban residents from that of rural farmers also made socialist China?s hukou system distinctive from its counterparts in imperial China and other countries. As Zhou (2009) suggests, something like household registration in China originated in the Zhou Dynasty, developing into a fairly mature institution during the Tang Dynasty for the purposes of social welfare provision and social control. Household registration, however, was never intended as an administrative means to distinguish the legal identity of urbanites from that of rural residents. Compared to the household registration of the Soviet Union from which the modern concept of hukou was derived, socialist China?s control over farmers by the hukou system was even more rigid (Lu, 2005). Under a rigid hukou system, ?farmers? in the Chinese context represented a social identity rather than an occupation. In this sense, the Chinese peasant farmers bore no comparison with European and American farmers working in industrialized agriculture, not only because of their traditional way of small-scale family farming, but also because they represented an inferior social group as defined by law.   Following this compulsory collectivization movement, the Luogang People?s Commune system was established in September 1958. Private family-based farming was strictly prohibited. Villagers were forced to join the People?s Communes and all their arable land and farming tools were confiscated. The People?s Commune made plans, designed work procedures and then assigned duties to villagers through broadcast every day. Without sufficient knowledge, communal officials always made incorrect decisions to guide practices, leading to destructive effects on agricultural production. In responding to Mao?s call for steel production during the Great Leap Forward period, Communal leaders organized villagers to construct small blast furnaces in the backyards of each village and herd young villagers into mountains to look for iron ores. In one village, a villager recalled his memory that more than three hundred young villagers (about one-third the village?s total population) were organized to work for backyard steel production. After one month of hard labour, only about 250 catties of low quality pig iron were produced. The diversion of labour for steel production meant that much of the agricultural harvest was left standing. Many pineapples and olives were left to rot in the fields. Moreover, to fuel the backyard furnaces, many lychee trees with a history of more than one hundred years were chopped down, causing a huge loss to villagers. 77  The Great Leap Forward ended in disaster. Although actual harvests substantially dropped, local officials competed with each other to announce increasingly exaggerated results, as they were under political pressure from the upper level authorities to record harvests in response to Mao?s radical Great Leap Forward policy. Production team leaders were pressured into falsely reporting ever-higher grain production figures to their Commune leaders. For example, the annual grain output in the Luogang area was normally 300 to 400 catties per mu on average.  Such an actual production output was often inflated up to 10, 000 catties. To ?substantiate? their false reports, Production Team leaders organized their team members to pull the crops from several fields and put them all into one piece of field, creating an ever-high unit output (Zhang, 2010). These false reports were used as a basis for determining the amount of grain to be taken by the State, and thus eventually depleted all village food supplies to fulfil the state?s tax levy. Villagers were unable to secure enough food and had to rely on wild vegetables for subsistence. As described by Kong (2010), many villagers suffered from oedema and some young female villagers suffered from amenorrhea due to starvation. Before the Great Leap Forward, there were about 40 new-born babies in Hong?s village each year. During the period from 1960 to 1962, only 10 babies were born and about 30 villagers died in this village which at the time had a total population of about 1,600.   To restore productivity and social order from the devastating crisis caused by the Great Leap Forward, the central government made a decision to restructure the communal system to a less centralized ?Commune-brigade-team? system. In 1962, the People?s Communes in the GLD were reorganized into the ?Commune-brigade-team? system. Under the restructured system, all confiscated land and farming tools were returned to Production Teams. Moreover, each household was allocated a tiny plot, namely ?reserved land? or ziliudi, with which villagers could grow some vegetables and livestock to supplement the very limited amount of grain allocated by the Production Team. According to some villagers aged over 45, the area of reserved land for each villager was about 10-13 square meters and such allocations were adjusted annually with the change in village population. Villagers usually grew onions, Choy Sum, lettuce, and other vegetables on their ?private? plot. They also raised some pigs, chickens and ducks for meat.  Under the ?Commune-brigade-team? system, villagers still had to collectively work under the direct supervision of the production teams to meet the grain production quota set by the state. 78  Villagers had to apply for leave from the production team if they wanted to leave their villages even just for one or two days. Under the ?work points? system, the working hours contributed by villagers to collective production were mostly directly exchanged for grains and food rather than for currency. As some villagers aged over 50 claimed, during the 1960s and 1970s, they worked 30 days a month and could only get a monthly salary of 3.6 yuan. With little cash on hand, farmers could not travel to or stay in cities very often, if at all, even though their villages were just about 30 kilometres away from the urban area of Guangzhou City. Consequently, the Work Point system combined with other institutions such as hukou system effectively prohibited farmers from residing in cities.   As elsewhere in rural China, villagers in the GLD area were excluded from the state welfare system which applied only to urban residents. The majority of their agricultural output was taken by the state to support industrial development in cities. This left villagers with barely enough food for survival. The deliberate intention of the socialist state to organize rural areas to serve national and urban interests drained the GLD region?s development potential. From mid-1962 onwards, the commune organized agricultural research institutions to improve farming techniques for growing rice and oranges. In 1965, the research findings of these institutions were first used the Tangdong Production Team of Luofeng Village. Other production teams had to send a member to attend the training courses organized by the commune to learn these new techniques and skills and then teach the villagers how to apply them in real production.  These research efforts helped improve overall productivity, leading to a gradual increase in the total output of grain and fruit from the late 1960s onward (Appendix I). However, most outputs were requisitioned by the state at very low prices to support industrial development in urban areas under the urban-rural dualist system.  In addition to the poor living conditions, villagers suffered from the torture caused by a series of political campaigns including the Four Cleanups (siqing) Campaign and the Cultural Revolution.21 Villagers were forced to participate in these political campaigns as failure to participate might result in detention and the suffering by the entire family. These collectivization movement and political campaigns atrophied traditional norms and customs, weakening the                                                           21 The Four Cleanups Campaign constituted part of the Socialist Education Movement initiated by Mao in 1965. It focused on rural areas and involved the cleaning of four aspects, including politics, economy, organization and ideology. This campaign eventually turned to political class struggle, marking the starting point of the Cultural Revolution.   79  kinship ties in the villages.  Consequently, villages were no longer self-governing societies. They became the lowest part of the ?Commune-Brigade-Production Team? system, which extended the power spectrum of the state from the central government down to family households. Under such an elaborate bureaucratic network, which was primarily designed to serve the needs of the state for national industrialization rather the interests of villagers, villagers lived at a marginal level until the implementation of the Household Responsibility System in the early 1980s.  De-collectivization and Re-collectivization in the Reform Era    Most villages in the GLD area started to implement the Household Responsibility System in 1981, when the Chinese government formally recognized such a system as a key strategy for reforms in rural China.22 At the beginning, the contract system adopted by some villages in the GLD area addressed production quotas rather than land; it divided the production quota imposed by the state into sections and contracted them out to individual farming households. In the following three years, this contract method was found unhelpful in improving the overall grain output, as villagers were not given incentives to produce more after meeting the contracted quota. From 1984 onward, more and more villages adopted an alternative land contract system, by which production teams divided collective land into sections and allocated them to farming households. In return, the farming household was required to sign a contract with the village collective (Production Team), in which farmers promised to fulfill a production quota set by the state as a land premium.   Under this new system, villagers no longer needed to work under the supervision of the production team. Like other farmers in rural China, villagers in the GLD region had much greater freedom in deciding what to grow and how much to grow once they had fulfilled the production quota set by the state. They were therefore provided with an incentive to maximize the output of land in response to market demands. As the prices of fruit and vegetable were much higher than rice, villagers diverted efforts from growing grains to growing vegetables and fruits                                                           22 In the late 1970s, a group of local cadres and farmers in a small village, which is called Xiaogang in Anhui Province, first initiated a contract responsibility arrangement allowing farmer households to lease land from Production Teams and to manage agricultural activities on their own. At that time, the proponents of this new system had to take the risk of getting jailed as contracting collective farmland to individual farmer households could be viewed as an illegal undertaking to violate the collective regime of socialist China. The contract responsibility system quickly proved conducive to productivity improvement and was subsequently reduplicated by villages in other provinces. For more details, see Li (2009). 80  (such as oranges, lychee, and plums) in the mid-1980s, leading to a dramatic increase in both farming output and income.23  As the statistics provided by the Luogang Town Gazetteer demonstrate, the annual output of oranges increased from 453,900 catties to 41,008,000 catties during the period from 1976 to 1990 (see Appendix I). During the pre-reform era, lychees were requisitioned by the state at extremely low prices of usually 0.25 to 0.33 yuan per catty. But in the 1980s, farmers could sell them on the market at about 5 yuan per catty. In the early 1990s, the price even increased further to 10 yuan per catty. Some farming households in the GLD could thus earn an annual income of approximately 4,000 yuan, mainly generated from selling lychees, vegetables and oranges to the nearby cities and towns in Panyu, Shenzhen and Hong Kong. In addition to this market-oriented strategy, another important reason that explained the rapid increase in local output was related to the improvement in farming techniques. The better use of farming machines, chemical fertilizers and weed killers helped villagers increase their productivity. Between 1976 and 1990, the unit output of grain increased from 984 catties to 1,222 catties per mu.   While villagers diverted their labour from the rice field to the orchards, more and more off-farm employment opportunities were available for villagers. One of the off-farm jobs was transportation services. During the late 1980s and the early 1990s, the public transportation system of the Pearl River Delta region lagged behind its economic growth, providing opportunities for villagers to engage with intra and inter-city transportation services. A villager told me that he had invested about 5,000 yuan on buying a motorcycle. Using the motorcycle for transporting goods and carrying people between the GLD and Guangzhou City, he could earn a monthly income of about 400-600 yuan. His ?success? was envied by his brother who only earned a monthly income of about 160 yuan as a secondary school teacher. During the late 1980s, more and more of his relatives and friends in the village switched to non-farm activities, leaving behind the rice fields to their families. As a result, some families required help from the non-local farmers, especially during the fruit harvest seasons. This explained the presence of non-local farmers in the GLD villages throughout the 1980s. In the early 1980s, some farmers from Guangxi Province moved to the GLD and leased arable land from local farmers for growing rice and vegetables. These non-local farmers might be the earliest ?migrant workers? in the GLD.                                                           23 According to the Luogang Town Gazetteer, about 70 percent of the agricultural land was used for growing vegetable and fruits at the end of 1990s.  81  They were forced to leave their hometowns to look for new opportunities because farming in their own villages did not provide them with an adequate livelihood.  The implementation of the Household Responsibility System restored private family farming, leading to the de-collectivization of agriculture production in rural China. When farmers were preoccupied with cultivation on their private plots, village collectives, which once intervened into their day-to-day production and life during the Maoist era, were limited to performing their administrative roles in collecting agricultural taxes and miscellaneous fees. Moreover, from the late 1980s onwards, the central government gradually relaxed its control over rural-urban migration, many young farmers left their homes to look for jobs in cities. During my fieldwork in the GLD, villagers told me that some migrant-workers living in their villages in the early 1990s were local cadres from the villages of other inland provinces, such as Hunan and Hebei. When even village cadres had to leave their own villages and look for urban employment, it was not surprising that the collective systems of these villages had experienced a severe decline.   Elsewhere in some remote villages, what followed the de-collectivization of agricultural production was a severe decline and even collapse of collective organizations. In the case of the GLD, the effect of agricultural de-collectivization on the transformation of collective village organization was different. Unlike rural farmers in inland provinces, most villagers in the GLD remained in their villages as they could earn increasing income by selling their vegetables and fruits to nearby cities. The improvement of villagers? living conditions was also evident by the building boom in the late 1980s. As recorded by the Luogang Town Gazetteer, 1,996 new village houses with a total gross floor 146,700 square meters were built between 1984 and 1990. Moreover, more than 1,000 old houses were upgraded to provide an additional floor area of 55,100 square meters. Some of these handsome village houses (which were usually 2-3 storeys in height) were still there at the time of my fieldwork, even though most of them had been turned into modern houses of 4-5 storeys in the 1990s. After making improvements in their income and housing conditions, villagers desired to upgrade their village facilities and services. However, they had to be self-reliant, given that the state continued to implement the urban-rural dualist system, providing only minimal state welfare to the countryside. Villagers therefore tended to join with each other for mutual benefits in the provision of public goods and services, strengthening the village?s collective organization. In one such instance, local cadres told me 82  that their village collective had invested as much as 90,000 yuan in building a tap-water system for every household in the late 1980s.   While villagers were struggling with this new albeit voluntary round of collectivization, the Guangzhou Municipal Government was constructing an industrial zone - the Guangzhou Development District (GDD). The massive land requisition programs for this mega project foreshowed the region?s urban future and reshaped the ongoing process of village re-collectivization. Established in 1984, the GDD was one of the first state-level development zones in China. To understand the development of the GDD and its impact on the urban transformation of the GLD, I begin with an examination of how and why the construction of development zones was adopted by the central government as a strategy for urban development.  From Development Zone to Urban District   From the mid-1980s, the Chinese government shifted its reform focus from the countryside to cities. Such a policy shift was clearly evident by 1984, when the central government designated 14 coastal cities to experiment with the ?open door? policy. It is worth noting that development zones are different from special economic zones (Jingji tequ).24 The decision to establish Shenzhen, Zhuhai, Shantou and Xiamen as Special Economic Zones was made as early as 1980. To achieve the objective of testing the applicability of the western capitalist system to socialist China, Special Economic Zones were given a high degree of autonomy and flexibility in local administration, especially in giving preferential treatment25 to foreign investors. Drawing from this experience, the government decided in 1984 to set up Development Zones in 14 coastal cities with the aim of expanding its open door policy. Development Zones were allowed to adopt similar preferential policies for foreign investors. However, in terms of the scale in their development, Special Economic Zones and Development Zones are different. The former                                                           24 Development Zones in China can be generally classified into four types: Economic and Technological Development Zones (ETDZs), Hi-tech Development Zones (HIDZs), Free Trade Zones and Export Processing Zones (EPZs). ETDZ is the first generation of development zone, which mostly accommodate labour-intensive manufacturing industries. In the late 1980s, the Chinese government started to set up HIDZs in which excellent infrastructural facilities were provided to support the development of technological industries. In recent years, more and more HIDZs have been merged with ETDZs. To improve the investment environment of ETDZs and HIDZs, the government approved the creation of free trade zones and export processing zones within ETDZs and HIDZs to provide all-purpose facilities and services for investors in the field of trading, bonded storage, material logistics as well as processing and manufacturing.    25 Favourable treatment includes tax concessions, access to credit and raw materials, and so on. 83  designation covers the entire city, while Development Zones are often built as delimited industrial zones and managed as part of the city by the municipalities concerned.    Many development zones, especially those established in the early 1980s, were located in what were then remote rural areas. In other words, at an early stage of their development, they were physically discrete from their parent city. He (1995) explains that both economic and political considerations determine this locational pattern. For economic reasons, locating development zones in old urban areas was not cost-effective as renewing or upgrading urban infrastructure to meet the needs of foreign investors required huge investments. For political reasons, the central government was still uncertain about embracing capitalist enterprises in the early 1980s. It located development zones in remote rural areas in part at least to keep a buffer distance between its people and foreign investors.      Table 3.35   National Development Zones in China Time Established  Name of Development Zones Total Areas  1984 ? 1988 Dalian, Qinhuangdao, Tianjin, Yantai, Qingdo, Lianyungang, Nantong, Minhang, Hongqiao, Caohejing, Ningbo, Fuchou, Guangzhou and Zhanjiang. 121.15 Km2 1992 ? 1993 Yingkou, Changchun, Shengyang, Harbin, Weihai, Kunshan, Hangzhou, Xiaoshan, Wenzhou, Rongqiao, Dongshan, Guangzhou Nansha, Huizhou Dayawan, Wuhu, Wuhan, Chongxing, Beijing and Urumichi.  206.9 Km2 2000 ? 2002 Hefei, Zhengzhou, Xian, Changsha, Chengdu, Kunming, Gunyang, Nanchang, Shihezi, Xining, Huhhot, Taiyuan, Yinchuan, Lanzhou and Lasa  162.96 Km2 2003 Suzhou Industrial Park, Hainan Yangpu ETDZ, Shanghai Jinqiao EPZ, Ningbo ETDZ and Xiamen Haicang Investment zone 209.47 Km2  Data Source: Li et al. (2011)  To understand the impact of development zones on China?s urbanization, we also need to distinguish between national development zones from those at the provincial and municipal levels. In the 1990s, the fiscal decentralization promoted by the central state led many Chinese cities to experience a wave of ?zone fever?26 However, the establishment of Development Zones at the local level did not follow good planning practice and sometimes was just a result of some                                                           26 According to the statistics of the Ministry of Land Resources, the total number of development zones at all levels in China reached more than 3,800 at the end of the 1990s. 84  local officials? personal ideas (Tang, 2000; Deng and Huang, 2004). These local governments acquired a large amount of agricultural land from farmers and then invested heavily in providing public services and infrastructural facilities such as site preparation, road connection, water, sewage, and electricity provision. In order to attract investments, they tended to offer unreasonably preferential policies for investors to compete with other already well-developed zones. Consequently, many development zones during this period suffered great losses, even though they were successful in bringing in new investors. Moreover, a large proportion of available land remained vacant or was used for real estate projects (Deng and Huang, 2004). The massive conversion of farmland into development zones created not only serious environmental problems but also conflicts between the local state and villagers.   In view of these problems, the central government required local governments to stop approving any new Development Zones from 2003 onwards and urged local governments to ?clean up? their development zones. Under such circumstances, many development zones throughout the country were forced to cease operations; those that continued were mostly national development zones directly ratified by the State Council. According to the Year Book of Development Zones, there are there are now 53 national development zones located in 35 cities. Altogether, they consume a huge amount of arable land, and are differentiated from local zones by a number of features. First, national development zones are both geographically and politically superior to those at other levels. They are usually located in well-developed cities with good accessibility and infrastructure networks.27 Second, their establishment requires the direct approval by the State Council. To obtain approval, the municipalities concerned are required to demonstrate their capacity to build and manage development zones. Subject to approval, their performance in a number of related areas, such as building quality infrastructure according to international standards, utilizing foreign investments, and enforcing laws and regulations to protect foreign investors? legitimate rights and interests, is closely monitored by the State Council. Third, their governance systems are overwhelmingly dominated by the single-minded pursuit of industrial expansion. Such governance cultures, however, have served as a foundation that allowed the local authorities of national development zones to develop impressive financial and                                                           27 For example, Guangzhou ETDZ is near Guangzhou City which is the capital city of Guangdong Province as well as the hub of the Pearl River Delta; Dalian ETDZ is near Dalian City which is situated on the estuary of Yellow River and a major business centre in northern China. 85  administrative capacities (Wei et al., 2007). With these comparative advantages, most national development zones have performed well in industrial and economic development.   Having been planned, designed, built and managed by the state, National Development Zones can be seen as a direct outcome of China?s state-led development model. The basic idea of these zones resembles the ?growth pole? strategy, which was fashionable in developing countries during the 1960s. The Chinese government expects that its development zones could play a role as the ?engines of growth? and eventually spread their effects to drive economic and urban growth in less-developed regions (Wong and Tang, 2005). Driven by these zone development policies at the national level, the Guangzhou Municipal Government established the GDD in 1984. Here I try to divide the GDD?s urban development trajectory into three stages.   The first stage covers the 1980s when the GDD was designated as one of the state-level development zones to carry out the experiment and promotion of the state?s open door policy. During this stage, the GDD had a total area of less than 7 square kilometres. Developing from a banana plantation located in the southern part of the existing GLD, the GDD was essentially an isolated zone for attracting and accommodating labour-intensive manufacturing industries with capital mainly coming from Hong Kong, Taiwan, Japan, South Korea and the United States. Like other state-level development zones in China, the GLD was given a lot of flexibility in making its own preferential tax policy to increase attraction for inbound foreign investment. As a result, the GDD of the 1980s was a purely industrial area, in which non-industrial properties, such as residential apartments, office, and shopping malls, were rarely found.  The second stage refers to the fast growing period in the 1990s. Having accumulated some successful experience in managing FDI in the preceding period, the GDD embarked on a period of accelerated growth and expansion. During 1992 and 1998, three new economic and high-tech industrial zones, including Yonghe Economic Zone, Guangzhou Free Trade Zone and Hi-tech Industrial Development Zone, were set up and merged with the GDD. As a result, the total area of the GDD substantially increased to more than 100 square kilometres by the late 1990s. Bureaucratic support coupled with the local authority?s continual effort to build high-standard infrastructural facilities provided the GDD with the most favorable environment for investors. During this decade, about 800 enterprises from more than 50 countries and regions set up their companies in GDD. According to the GDD Gazetteer, the actual utilization of foreign capital in 86  GDD increased to US$21.5 million in 1994, which accounted for 17 percent of the total foreign investment of Guangzhou city.  Apart from these expansions in physical area and industrial output, the development of the GDD during this stage was characterized by its repeated attempts to promote industrial transformation and urban development. In the mid-1990s, Guangzhou Municipality initiated the development of a Science City with a total planning area of 45 square kilometres in the GDD. Unlike other industrial zones of the 1980s, Science City was planned to provide not only world class infrastructure for hi-tech industrial development but also a liveable environment for its residents (see Figure 3.2). Moving beyond the function of spreading hi-tech industrialization28, the Science City allowed more land for office buildings, hotels, shopping malls and residential development, giving fresh impetus to the transformation of the GDD from a purely industrial zone to an urban region.   The third stage began in the early years of the new millennium when the GDD accelerated its pace in urban transformation as a response to the policy changes at national and municipal levels. Alongside China?s accession to the World Trade Organization (WTO) in 2001, the central government adjusted its national policies for development zones. To comply with the free trade and fair competition requirements of the WTO, development zones were required to gradually remove preferential taxes to equalize the treatment of foreign and domestic investors. With this policy shift, development zones in China were expected to switch their traditional function from simply attracting FDI to playing a greater role in leading regional development. Another important force for this transformation was the change in the spatial strategy at the municipal level. In the late 1990s, the Guangzhou Municipal Government upgraded Panyu and Huadu from ?county-level cities? to ?districts? of Guangzhou City. Meanwhile, the municipal government pursued a new spatial development strategy described as ?exploring the south, optimizing the north, extending in the east and coordinating in the west.? Among these development directions, the most influential one on the GDD was the ?extending in the east,? which suggested developing the east of Guangzhou (where the GDD is located) into a new urban centre of within the city region. To enhance its accessibility, the municipal government is constructing a subway to connect it with the city core. The construction project is expected to be completed in 2016 (see Figure 2.4). Due to these policy changes, the GDD accelerated its                                                           28 Science City comprises 4 zones for hi-tech industries: Micro-electronics Information Science Park, Biotech Industrial Science Park, New Energy & New Material Industrial Science Park and Science Park for Other High technology. 87  urbanization pace. Over the past several years, the GDD kept on expanding its urban development to the neighbouring rural area via the construction of a Knowledge City (with a planning area of 123 square kilometres) in the northern part of the region (see Figure 2.4 and Figure 3.3). In 2005, the GLD was formally established as a new urban district of the Guangzhou Municipality, amalgamating the GDD and the Luogang Town.    Based on this historical review, I would like to highlight a primary question of urbanization and local governance restructuring in GLD in the reform period. How did a segregated, village-based society, which had embraced a mixture of institutions combining socialist collectivism and an ancient lineage tradition, escape its historical constraints to establish entirely new institutions that reflect the urban market-oriented society of today?s China. In the reminder of this dissertation, I will provide detailed narrative and concrete examples of how local actors have worked together to deal with land conflicts and the various problems arising from rapid urbanization under the conditions of the urban-rural dualist system. In the next chapter, I will start by exploring the implementations of land requisitions and compensation and their impact on the life of villagers and local governance restructuring.    88  Figure 3.1 13   Traditional Landscape Features of Urbanizing Villages            Village pond                                                                             Village pond              Temple                                                                                     Temple           Ancestoral hall                                                                            Ancestoral hall   89  Figure  3.214   Guangzhou Science City             Figure 3.315  Knowledge City         Scale model of Knowledge City                                        Construction sites in Knowledge City           90  Chapter 4      Land Requisitions, Farmer Resettlement and Local Land Politics      Conventional approaches tend to explain land disputes in periurban China as being merely an outcome of an exploitative local state ferociously evicting villagers from their land and converting rural land into lucrative property development. Viewed in this light, however, two important consequences caused by land requisitions tend to be overlooked. The first is related to the actual impact of land requisitions on the multiple levels of political authority in rural China. Land requisitions inevitably restructured the bureaucratic system on which the socialist state had relied to maintain its effective control over rural villages for nearly half a century. The collective land system and rural bureaucracy were inextricably linked, because the ?commune-brigade-team? system was not only about land ownership but also about power structure. As the collective land system has continued to persist in the reform period, this elaborate bureaucracy, albeit in modified form and with reduced effectiveness, has continued to impose its control over rural villages. Land requisitions, on the one hand, enabled the local state to turn rural land into urban land for economic development. On the other hand, when more and more collective land was acquired for sale in the urban land market, land requisitions led to the decline of this elaborate bureaucratic system by which the local state could once pursue what Ho (2001 and 2005) called ?intentional institutional ambiguities? as a strategy of governance.   The second set of consequences is concerned with the ambivalent effects brought about by land requisitions on the life of farmers. Under the urban-rural dualist system, farmers were severely discriminated against and most of them were living on the verge of extreme poverty. During the early 1980s, the implementation of the Household Responsibility System enabled farmers to achieve a remarkable improvement in their living conditions. However, subsequent urban-biased development polices and the lack of infrastructure investment and social welfare in the countryside in the 1990s began to change the livelihoods of many farmers from being manageable to again becoming vulnerable. Given this situation, land requisitions by the local state generated not only threats but also created new opportunities for farmers. Although land 91  requisitions for urban development dismantled the self-reliant life support system in rural society, they also forced the local state to face the necessity of having to provide social security and urban services to the affected villagers so as to mitigate the risk of social unrests developing out of the process of urban transformation. These dual effects have shaped the ambivalent attitude of villagers towards land requisition, which in turn have influenced their strategy of resistance and accommodation to the policies and actions taken by the local state. With a particular focus on these two issues, the following sections will explore how land requisitions have transformed the interactions and power relationships among the GLD government, village collectives, and villagers.   Hukou as Part of Compensation in the 1980s  Throughout the 1980s and the 1990s, the responsible authority of the GDD, namely, the GDD Administrative Committees, requisitioned massive plots of land for industrial expansion. According to the statistics provided by the GDD Gazetteer 1984-1990, the GDD Administrative Committee, requisitioned 6,837 mu (equivalent to 4,558,000 square meters) of arable land from villagers for the development of industrial zones between 1984 and 1989. As shown in Table 6.1, all land requisition and development programs from 1984 to 1988 involved an expenditure of approximately 470 million yuan, including 210 million yuan for compensation payments. In the early years of the reform period, there were few specific policies and legislative provisions to regulate land requisitions and compensation. In determining appropriate policies for compensation, the GDD Administrative Committee mainly made reference to the ?Implementation Methods Regarding Land Requisition for State Construction in Guangdong Province? promulgated by the Guangdong Provincial Government in 1983. Based on these guidelines, the GDD Administrative Committee offered compensation based on four categories of claims: (1) land compensation (Tudi buchang fei), or the value of the arable land requisitioned; (2) compensation for agricultural losses (Qingmiao buchang fei), or the value of any agricultural products grown on the requisitioned land; (3) compensation for irrigation facilities (Nongtian shuili fei), or the amount of damage or loss to irrigation facilities constructed by villagers; and (4) housing compensation and relocation expenses (Chaiqian fei), or the value of the building occupying the requisitioned land, and the amount of any expenses reasonably incurred by villagers? relocation.  92   In spite of these reasonable-sounding categories, the villagers affected by land requisitions received only a very small amount of cash compensation. As explained by the local officials, compensation fees for land value were not directly allocated to individual households because land was not individual property but collectively owned. As a result, the affected villagers only received Qingmiao Buchan fei, amounting to 10 times the annual agricultural output of their arable land. If the requisition involved relocating villagers? houses, the affected villagers could receive Chaiqian fei for the costs of their village houses and the relevant relocation expenses. In 1988, the compensation rate for arable land was set at 40,000 yuan per mu. But, as revealed by the records of the GDD Yearbook 1984-1990, the affected villagers only received about 2,500 yuan in cash (i.e. 6 percent of the total compensation fees) as Qingmiao Buchang Fei for their agricultural outputs. The remaining fees were divided into two portions: 12,000 yuan (about 30 percent) were deposited in banks in the name of the village collective while 25,500 yuan (about 64 percent) were retained in trust by the township government.  Table 4.16  Snapshot of Land Requisitions and Compensation in the GLD: 1984-1988 Year Requisitioned  Areas (mu) Compensation  Unit Rate Total Compensation (yuan)   (yuan per mu) Land Value   Relocation Fee Irrigation Facilities   Agricultural Outputs & Others  1984 3743  20,000 73,205,798  2,629,913  492,555  2,249,195  1985 357 20,000 6,022,995  106,938  48,297  220,543  1986 352 28,000 10,033,246  620,419  96,210  804,947  1987 212 28,000  5,671,932  43,220  Nil 402,274  1988 2164 40,000 86,888,036  31,83,371  1,173,628  6,763,114  Data Source: GDD Gazetteer Editorial Board (1993)   These compensation arrangements, however, should not lead us to conclude that villagers were rendered economically vulnerable as a consequence of land requisitions. Quite the opposite: the affected villagers were able to make a remarkable improvement in their economic conditions. First of all, the interest generated from the bank deposits was enough for them to live on after land requisitions. Recalling their experience in the 1980s, some villagers suggested that each affected villager received about 100 yuan of interest monthly. With this income, they were better 93  off than danwei workers, who usually earned 36 to 60 yuan in the mid-1980s. Moreover, employment opportunities associated with the construction of development zones provided additional sources of income. As the first phase of the GDD was located in a stretch of marsh, land reclamation was required to protect the area from anticipated flooding. In order to gain the support from villagers for land requisitions, the GDD Authority made it a policy of granting job priority to villagers who wanted to participate in land reclamation projects. As a result, many villagers collected raw materials (such as sand and soil) from nearby rivers and hills and made handsome profits from selling them to the project contractors. Villagers were also granted a non-agricultural hukou, which gave them access to the same social security services as those enjoyed by urban residents.   According to the official statistics, a total of 6,622 villagers affected by land requisitions were granted ?non-agricultural? hukou between 1984 and 1989. It is important to note that at the time the central government continued to impose tight restrictions on the growth of cities by limiting the quota of urban hukou for rural farmers. Under such circumstances, the local township government in the GLD had to deal with the food and employment problems at its own expense if it wanted to provide non-agricultural hukou for the affected villagers. With more than 60 percent of land compensation fees, local township government official had sufficient money to purchase grains from the central government for distribution to the affected villagers. Moreover, with the same funds, the township government was able to develop Township and Village Enterprises (TVEs) that could provide jobs for local villagers. When the municipal government of Guangzhou City made the development of TVEs a matter of policy in order to create jobs for those who had lost land to urbanization during the 1980s, local township governments and village collectives were given autonomy to pursue non-agricultural production. Being loosely administered and free of environmental controls, TVEs were mostly built on the foothills near villages and sometime encroached on arable land. These TVEs engaged in a range of industrial production, such as processing local fruits, making furniture, making electroplating products, producing construction materials and so on. As revealed by the statistics compiled by the Luogang Town Gazetteer, the total industrial output of TVEs in the area increased from 5,697,800 yuan in 1983 to 26,154,000 yuan in 1990. These industrial enterprises not only provided local employment opportunities but also improved the financial abilities of village collectives, enabling their greater involvement in social welfare provision for villagers. Some 94  village collectives provided public goods, such as tap-water facilities, kindergartens, primary schools and elderly nursing centres for villagers.  As discussed in the previous chapters, under the urban-rural dualist system institutionalized by the hukou institution, rural farmers with an ?agricultural? hukou were exploited in order to support national industrialization and had long been excluded from state welfare benefits. When land requisitions resulted in a remarkable improvement in their living conditions, it is not surprising that villagers tended to welcome rather than resist land requisitions. In the 1980s, land requisitions in the GLD only affected a handful of villagers. Seeing that farmers in the nearby villages were still engaged in subsistence farming, villagers felt grateful when land requisitions provided them with a once-in-a-lifetime opportunity to escape the cycle of toil and poverty. In the words of villagers, land requisitions provided them with the opportunity to ?wash feet and leave the field (xijiao shangtian).? They emphasized ?feet? rather than ?boots? as they always walked barefoot and got their feet muddy in the field. One local resident, who was granted a ?non-agricultural? hukou as a result of land requisitions in the early 1980s, suggested that he had felt like the winner of the first prize in a lottery.   ?Reserved Commercial Land? as a Policy of Compensation in the 1990s    In the early 1990s, the central government imposed greater control over land requisitions for the construction of development zones. This was partly because the proliferation of development zones had not only caused environmental problems but also triggered mounting tensions between local governments and villagers. The GDD, therefore, had to comply with more rigid procedures in implementing land requisitions. As shown in Figure 4.1, all plans for land requisitions had to be further endorsed by the Guangdong provincial government and the State Council after getting approval from the municipal level. Moreover, before implementing land requisitions, the local authority was required to reach agreement with villagers as regards the compensation arrangements. However, these more stringent controls did not slow down the pace of GDD Administrative Committee in promoting industrial development. Between 1991 and 2000, the total amount of land requisitioned by the GDD Administrative Committee increased to 47,700 mu (see Table 4.2), which was nearly seven times that of the 1980s. In carrying out these land requisition programs, the GLD Administrative Committee, however, was pressured to 95  revise its land compensation policies mainly because the offer of urban hukou as part of the compensation package was no longer sufficiently attractive to villagers. In what follows, I will briefly explain how and why urban hukou lost its attractiveness to villagers.    Table 4.27  Snapshot of Land Requisition and Land Compensation: 1991-2000 Location  Area (mu)  Compensation Fees (yuan) West Section 54.50 N/A TPZ  1,925.48 100,245,400 Southern Part of East Section 4,497.80 397,396,900 Bigang 320 320.26 29,928,400 Northern Part of East Section 5,600.52 450,641,000 Bigang 2000 2,000.00 177,767,000 Luoyong Highway 330.10 34,236,900 Donghui Plaza 506.59 107,835,100 Foreign Trade Center 216.30 5,942,500 Detention Centre 40.00 4,277,160 Yonghe Section 20,902.17 387,280,780 Science City  11,377.24 1,495,915,300  Data source: GDD Gazetteer Editorial Board (2004) Notes: the compensation rate for Yonghe Section was exclusive of the compensation of old lychee trees and graves which were calculated on another set of criterion agreed with villagers.   After an initial burst of expansion, the TVEs in the GLD quickly experienced a decline in the 1990s. Past studies highlight two major reasons for this decline. First, the Chinese government deliberately suppressed TVEs by limiting their access to bank loans in the formal financial market, as TVEs not only directly competed with state-owned enterprises but were also creating serious pollution problems (Huang, 2008; Zhao and Wong, 2009). Second, many TVEs were subject to the constraints of their ambiguous ownership structure, limiting their abilities to pursue growth in the 1990s (Ho, 1994; Naughton, 1994; Shen and Ma, 2005).  The GLD experience suggests one more reason in explaining why TVEs were shrinking fast in the 1990s. With the rapid growth of FDI in the area, the villagers in the GLD quickly found that direct land leasing was more profitable than running their own TVE business. More and more village collectives, therefore, just closed down their factories and leased their land to outside investors. Their target tenants were mostly investors from Hong Kong who set up some small factories as well as the companies that provided support services for the larger FDI-invested companies in 96  the GDD. The shrinkage of TVEs did not harm the overall income of village collectives, but it did significantly reduce local employment. The newly established companies and factories in the GDD provided some new job opportunities, but over time, the GDD began to update its industrial structure by gradually replacing many labour-intensive manufacturing and highly polluting industries into hi-tech and high value-added production. As a result, local employment opportunities were not altogether useful to the villagers affected by land requisitions because their knowledge and skills did not meet the requirements of the new enterprises.  During the 1990s, the decline of TVEs in the GLD occurred in conjunction with the central government?s ongoing attempts to promote the privatization of state-owned enterprises. As a result of the reforms of state-owned enterprises, many danwei workers were being laid off and most welfare housing previously owned by danwei become privatized. These changes invalidated urban hukou as a guarantee to public housing, education, health care, employment and other social services in urban areas. As a result, many villagers preferred to keep their hukou status as ?agricultural,? which freed them, at least to some extent, from the one-child policy restriction and gave them hope for having an additional child if the first was a girl. This change was also evidenced in the declining auction price of urban hukou in the villages. During the 1980s when the central government still imposed rigid restrictions on urban growth, only a few urban hukou quotas were allocated to each village per year. In distributing these hukou quotas, preference was usually given to army veterans and village cadres, while the rest was usually sold to villagers by way of public auctions.   Some villagers remembered that in the late 1990s the auction price of each hukou quota once reached 20,000 yuan. A decade later, the price substantially dropped to a mere 4,000 yuan. In villages where most of the land was requisitioned, the value of hukou even became negative. Urban hukou, which was once one of the most valuable policies in dealing with village collectives, gradually became increasingly burdensome in the 1990s. According to a village cadre, his village was assigned with a quota of 10 conversions of agricultural hukou to urban ones. To meet this quota, the village collective had to provide a subsidy of about 5,000 yuan to each villager who was willing to convert their hukou status from ?agriculture? to ?urban.? It turned out that those who were willing to accept this offer were mostly old villagers for whom raising a family was no longer a concern. Very few young villagers agreed to convert their hukou status, unless they wanted to further their study in urban areas.  97   Figure 4.116 Land Requisitions Procedures in the GLD during the 1990s   Data source: GDD Gazetteer Editorial Board (2004)  When urban hukou was no longer able to sustain the livelihood of villagers after land requisitions, the GDD Administrative Committee decided to adopt a special policy of compensation, under which the government returned a certain portion of the requisitioned land (about ten percent) to the village collectives as part of their overall compensation. All ?returned lands? were called ?reserved commercial land (ziliu jingji yongdi)? and legally allowed to be 2. Submission of Site Selection Plan and Preliminary Land Use Plan  3. Approvals from Planning Bureau, Land Surveying Bureau, Land Resources Bureau at the municipal level   4. Submission of Land Use Plan, Land Requisition Plan, Hukou Conversion Plan and Land Supply Plans 5. Approvals from by Planning Bureau and Land Resources Bureau at the provincial level  6. Approval from State Council 7. Payment of Land Premium to Land Resources Bureau at the municipal level    8. Announcement of Land Requisition by the Guangzhou Municipal Government   9. Formulation of Land Compensation Plans  10. Agreements with Village collectives on Land Compensation    11. Promulgation of Land Compensation Standards by the Land Resources Bureau at the municipal level   12. Implementation of Land Requisition    1.Land Requisition Proposal by the local government    98  leased for non-farming uses such as hotels, factories and storage buildings. This policy aimed at generating recurrent income for the affected villagers and pay for future social welfare services once their productive land was lost to urbanization. As a result of this policy, the local government was no longer directly involved in the creation of employment opportunities and the provision of social welfare, and was thus able to turn its attention to attracting FDI and the single-minded pursuit of economic development.   The implementation of this policy contributed to the rapid industrial growth of the region in the 1990s, allowing villagers to remarkably improve their living conditions. However, it did not provide a perfect solution to conflicts arising from land requisitions. The policy failed to address villagers? demands for higher compensation not only for arable land but also lost homes. Furthermore, the management and development of ?reserved commercial land? created conflicts between local cadres and villagers, which in turn conditioned the local government?s intervention in the process of local governance.  In what follows, I will explain the deep causes of these problems and how they affected the interaction among local government, village collectives and villagers in the process of local governance.   Villagers? Resistance     Villagers? resistance to land requisitions was not unanimous, but did strive for a more equitable compensation. During the 1990s, land requisitions in many villages were divided into several phases. For example, in one village I visited, four land requisitions were conducted: 400 mu for the construction of Guangzhou-Shenzhen Highway in the late 1980s, 4,000 mu for the construction of the Economic and Technological Zone in the early 1990s, 700 mu for the East Erhuang Highway in the mid-1990s, and 4,000 mu for the construction of the Science City. In these ongoing requisition processes, villagers learned how to adapt their strategies for dealing with local government. In the 1980s, villagers voluntarily dedicated some arable land to the local government for the construction of highways; they expected to improve the local transportation conditions so that they could more efficiently deliver their fruits and vegetables to sell in nearby cities. Without compensation, the village collective usually re-allocated any remaining land to those villagers who had lost their land so that they might continue farming. In the end, however, villagers longer had any land left for cultivation and started demanding higher levels of 99  compensation. In addition, they became increasingly aware that the local government was making enormous profits from leasing the requisitioned land for industrial development and could thus afford to pay them higher compensation.   Another significant conflict was related to the compensation for village housing that served not only as shelter but also a major source of rental income. Before the mid-1990s, land requisitions mostly involved arable land and rarely touched villagers? houses. The jarring juxtaposition of the GDD?s well-developed industrial zones intermingled with traditional village settlements was perceived as a ?chaotic landscape.? In view of this, the local authority decided to relocate some villages near the outskirts of the Science City in the late 1990s. Requisitions of villagers? houses, however, triggered unexpectedly strong resistance. One important reason for resistance was that unclear property titles of some village houses caused numerous disputes about compensation.  In rural China, ?zhaijidizheng? was the sole legal document to prove a villager?s ownership rights (i.e. the user rights) over their self-occupied houses. As mentioned in Chapter 3, many villages in the GLD experienced a housing boom during the 1980s and the 1990s. The reasons for why farmers were willing to sacrifice their scarce arable land for housing construction have been extensively explored by Sargeson (2002) in her research on housing construction in some Zhejiang villages. Here, I will focus instead on explaining why some houses in the GLD did not get notarized ?zhaijidizheng? and how this problem created conflicts between the local government and villagers.   Before the mid-1990s, the state did not strictly enforce building codes in villages. The village collectives were therefore given room to promote housing construction in order to achieve two objectives:  responding to villagers? demands for new housing and earning income from ?selling? housing lots to other villagers. As some suggested, it was not too difficult to get a site for housing construction as long as one was willing to pay. So long as housing construction was not strictly regulated, the applications for title documents did not involve complex procedures. Nevertheless, it turned out that many houses built during this period did not have formal title documents. One important reason, I suggest, was that the absence of a housing market in rural China did little to encourage villagers to establish formal title ownership of their homes. Under the urban-rural system, farmers? housing was individually owned, but not allowed to be sold in the open market. Thus, the continued secure access of the villagers to their home was seen as a sufficient proof of their ownership. In other words, the traditional concept of ?ownership? in rural 100  China was largely based on use rights rather than the right to sell or mortgage their houses. Legal certificates, which were primarily essential for market transactions, were therefore redundant to villagers. A villager told me that he submitted tender for a housing site from the Villagers? Committee in 1990s. He paid about 60,000 yuan to the village collective as a ?land premium.? After the completion of his house, like many villagers, he did not apply to the township government for a formal ?zhaijidizheng? certificate as the application for this involved an extra sum of money. The administrative fee for a ?zhaijidizheng? was set at 70 yuan per square meters. In this case, he would have to pay 6,300 yuan for the title application. The administrative fee was only one consideration. A more important reason was that he simply did not feel that the missing title document was a problem. As he said, all people in the village agreed that he was the owner of the house and naturally he was free to use his house for self-occupation or lease it to migrant workers if he wished. Like many villagers who did not apply for the formal title deeds, he was upset when his ownership rights were questioned by the local government under the compensation assessment.    The assessments of some houses with a building height exceeding 3.5 storeys were also often subject to disputes. Under the prevailing laws, ?zhaijidizheng? only permitted a maximum building height of 3.5 storeys for each housing lot. However, during the late 1990s, the increasing income generated from leasing houses to migrant workers encouraged more and more villagers to rebuild their houses into five to seven storeys. The GLD government defined these houses as ?unauthorized? structures as the additional building works had been carried out without prior approval of the government. Another category of houses subject to disputes were those erected after the announcement of village relocation plans. After learning that their village would soon be relocated, many villagers mobilized all of their resources to add new storeys to their existing houses and to erect new houses in the open space as much as they could. According to some officials from the Land Requisition Office, the floor space of a village could nearly double, within a month of announcing the programs. The local government defined these houses as ?unauthorized? or ?illegal? structures and refused to pay compensation.  Last but not least, the development of ?reserved commercial land? constituted a major source of conflict between local government and villagers. During the early 1990s, village collectives were required to pay a sum of money called ?charge for supporting services (peitaofei)? to the government to cover the costs of infrastructural facilities, such as roads, water supply, electricity 101  supply and so on. However, many village collectives could not afford such a charge. A village cadre provided this example: between 1994 and 1995, the compensation for each mu of land was about 80,000 yuan. After paying qingmiao buchang fei to the affected villagers and reserving 20 percent for social welfare provision, the shareholding cooperative could only get 40,000 yuan, which was not enough for it to pay the ?peitaofei.? In that case, the village collective was forced to sell the ?commercial land? back to the government at a price of about 67,000 yuan per mu. In response to villagers? resistance, the GLD government terminated the levy of ?peitaofei? from the late 1990s. Over the recent decade, more conflicts between the government and the village collectives were associated with the planning control on ?reserved commercial land.?  Many village collectives complained that it was very difficult to get a planning permit for the development of their reserved commercial land. A key to understanding these conflicts was related to how national land policy affected land management at the local level.     Concerned with environmental sustainability and food security, the central government further tightened its control policy on the conversion of arable land for urban development. In 1997, the State Council issued the Notice on Further Regulation of Land Management and Protection of Arable Land, highlighting arable land preservation as an important national strategy. The target of the arable land preservation was incorporated into the state?s subsequent five-year development plans. According to the eleventh five-year plan for 2006-2010, the total area of arable land of the country had to be maintained at the level of no less than 1.8 billon mu. To balance the needs between development and preservation, this target was to be achieved through regulating the supply of development land and arable land. To this end, the central government set an annual quota for the conversion of arable land for construction (nongdi zhuanhuan zhibiao) in each province. Under this land conversion quota, whenever arable land was used for construction purposes, an equal amount had to be provided somewhere else to ensure the overall arable land area of the province would not been reduced. To ensure effective local implementation of this policy, the PRC?s Criminal Law was amended in 1998 to include specific provisions to enforce arable land protection. As a result of this amendment, local officials risked criminal consequences if they granted approvals for the conversion of arable land (feifa pidizui) into development without complying with the regulations of the Notice.   As a consequence of these land control measures, the GDD Administrative Committee encountered growing difficulties in acquiring the necessary planning permits from the provincial 102  and municipal governments for development. During the 1990s and the early 2000s, when arable land was converted for industrial purposes in the GDD, an equal amount of arable land was created in other parts of the province under the administrative coordination of the provincial and the municipal governments.  However, in recent years, the GLD could not secure the land quota as easily as before because provincial and municipal governments tended to cater more to the development needs of other less developed areas. This change not only slowed down local industrial and urban development, it also affected the development of ?reserved commercial land.? With the restricted land development quota, the construction of Knowledge City had to be suspended. Local planners suggested that construction projects would not resume until a new land quota was approved by higher level governments. They further explained that the higher-level governments regularly took aerial photos to strictly monitor land-use in the area. As a result, all senior officials at the municipal and provincial levels were now very serious about the enforcement of the land quota. Any contravention might not only risk their career prospects but also make them liable for criminal prosecutions. In this situation, one can see how difficult it was for the village collectives to obtain planning approval for their ?reserved commercial land.? This was particularly the case in those villages where most of their arable land had been requisitioned after the late 1990s.    Compromise with Villagers     These conflicts created increasing social tensions, and the local government realized that its pragmatic pursuit of industrial expansion required it to reach a compromise with villagers. Local government did not want to leave villagers an impression that they had ?bullied farmers?, because this might lead to stronger resistance in the next round of land requisitions. Therefore, local officials deliberately pacified villagers by proactively responding to their demand for better compensation. In the mid-1990s, with the construction of the Guangzhou Science City underway, the GDD Administrative Committee more than doubled the land compensation fee for arable land from the unit rate of 40,000 yuan per mu adopted in 1988 to 80,000 ? 100,000 yuan per mu.  In the early years of the new millennium, land compensation rate was further adjusted upwards to 120,000-130,000 yuan per mu. To address the conflicts arising from housing compensation, the local government offered 60 percent of full compensation for those ?unauthorized? housing structures built between January 1987 and December 2000. Moreover, the government promoted 103  a transparent mechanism for determining the compensation, by making public the valuation reports of all housing acquired for villagers? inspection.  However, for ?unauthorized? constructions built after January 2001, the government firmly refused to pay any compensation. Some local officials explained that the local government had stopped granting housing sites and issuing ?zhaijidizheng? in the late 1990s. Housing built after January 2001 was thus considered illegal from the perspective of the government. To further placate villagers, the GLD government carried out a ?home ownership scheme,? by which villagers were provided not only compensation for their old village houses, but were also given opportunities to buy dwellings at a subsidized price from the government. The average price of resettlement housing was set at a level of 1,000 yuan per square meter, which was 40% lower than the cost of construction. Moreover, for any households which had financial difficulties in taking advantages of the ?home ownership scheme?, the government would offer subsidies to ensure that they could obtain a unit with a total area of no less than 90 square meters.   Regarding the development of ?reserved commercial land,? the GLD government could not overcome the land quota hurdle as the real power of its allocation was vested in the authority of higher-level governments. To get around this problem, it decided to lease or buy back villagers? ?reserved commercial land.? In a village I visited, 1,100 mu of ?reserved commercial land? had been rented to the district government at an annual rent of about 3,000 yuan per mu, generating a total rental income of more than 3.3 million yuan each year. In another village, the village collective had just sold all their ?reserved commercial land? back to the government at a price of 380,000 yuan per mu. Without a land conversion quota, local government could not use this land for development in the near future. The purpose of leasing and buying it back was not simply about calming the emotions of villagers. More importantly, it was to ensure that villagers could continue to access the provision of social welfare which had become increasingly dependent on the income from ?reserved commercial lands.?   At the same time, the GLD government encouraged village collectives to carry out village redevelopment. As such redevelopment did not involve the conversion of arable land, villagers did not need to wait for the land quota to be allocated by higher-level governments. In 2009, the Land Resource Ministry of the central government issued a development policy called san jiu gaizao, which meant the redevelopment of three categories of old areas, namely, old towns, old industrial buildings and old villages. Guangdong Province was the first to experiment with san 104  jiu gaizao. Under this policy, the strategy was to rebuild existing village sites into high-rise residential and commercial properties, in order to improve the living environments and increase collective incomes. San jiu gaizao provided two major policy incentives for villagers. First, after paying a symbolic penalty fee of two yuan per square meter, villagers were allowed to legalize those of their ?unauthorized? houses built between 1987 and 2007. Second, each house owner was guaranteed a total re-housing area of 240 to 280 square meters in the new development. If he owned real properties with a total area of more than 280 square meters, he would be entitled to additional cash compensation for the difference between the area of his existing houses and the re-housing areas at a rate of 1,000 yuan per square meter. After resettling the villagers, the remaining floor areas were to be sold or leased in the open market so that the village collectives would generate more income to improve community facilities and services. In recent years, some villages have launched their own redevelopment projects in collaboration with private property developers searching for real estate sites.  The real effect of this village redevelopment scheme remains to be seen. However, the lack of consensus amongst villagers has been an emerging obstacle in the implementation of village redevelopment. It was the mandatory requirement that any redevelopment project under the san jiu gaizao policy had to be supported by at least 90 percent of house owners in the village. As some local cadres suggested, it was extremely difficult to meet this requirement in practice because each household reacted differently to the proposal. Some households fully supported village redevelopment as the existing ?kiss buildings? and ?shaking-hand buildings? had created a poor living environment and increased fire hazards. Other households opposed redevelopment for three main reasons. First, they were highly pleased with the handsome rental income that they were earning from leasing their houses to migrant workers and were not interested to make a change. Moreover, some of these rich ?landlords? had already moved out of their villages and were now living in better housing elsewhere, even though they maintained their household registrations with the village. In this context, improvement of the living environment was not their major concern. Second, some preferred to live in low-rise housing rather than new apartment buildings, and therefore objected to giving up their former properties. Last but not least, some villagers distrusted village cadres, arguing that redevelopment was merely a tactic adopted by them to benefit their own purse.  105  Although many problems and conflicts remained unresolved, the actual process of land requisitions in the GLD suggests that villagers were no longer powerless. They were increasingly aware of their rights and thus resisted unpopular policies that jeopardized their interests. In the face of villagers? rightful resistance, local officials became more responsive and constantly adjusted their tactics in dealing with the needs of villagers to avoid widespread conflicts. What really made this possible, I would suggest, is the implementation of the ?reserved commercial land? policy that caused many villages to become increasingly autonomous and financially independent entities.    Retreat of the Local State and Village Autonomy     Much of the existing literature tends to suggest that the collective land system in rural China offers ambiguous ownership and thus puts farmers in a weak position to resist the local state?s land requisitions for urban development (see Ma and Xiang, 1998; Gou, 1999; Zhang, 2001; Cai, 2003; Soto, 2006; Hsing, 2010). However, few have paid serious attention to the question of how land requisitions themselves affect the transformation of the collective ownership system and how this transformation consequently affects local governance processes.   The collective system in rural China took root in the ?commune-brigade-production team? system during the Maoist period.  In the official lexicon, the ?Commune-brigade-team? system was defined as ?three-tier ownership with production teams as the basic accounting units (sanjisuoyou, duiweijichu).? Under this structure, the ownership of rural land was collectively shared by three levels of authority, namely, commune, brigade and production team (see Figure 4.2). At the same time, the commune played a role akin to that of the local state. Brigades could be viewed as the extension of the Communes to supervise and manage the activities of production teams. Under the direct supervision of the commune-brigade, production teams organized farmers to work in daily agricultural production. Moreover, administratively, production teams were the basic accounting units of the rural collective system responsible for the management of the collective land and other assets, and assuming all profits and losses of 106  production activities of their team members.29 In this context, the ?Commune-brigade-team? system signified both the structure of ownership and administrative hierarchy.    Figure 4.217  Collective Ownership and Governance Bureaucracy in Rural China    This complex authoritative organization reached down to village neighbourhoods and even individual households. It allowed the political power of the local state to intervene in rural villages not only as an administrator but also as a co-owner of the collective land system. In 1984, the commune, brigades and production teams were replaced by the township government, Villagers? Committees and Villagers? Groups respectively. In line with this administrative restructuring, all enterprises and economic organizations operated by brigades and productions teams were reorganized as Association of Economic Cooperatives (Jingji Lianshe) and Economic Cooperatives (Jingji Hezuoshe) respectively. During the 1980s, these organizational                                                           29 People?s Communes were required to return all collective lands and assets to Production Teams (not individual farming households) after the organizational restructuring in the early 1960s.  Central State Commune Brigade  (Advanced Cooperative) Farming Households  Main Tenant Rights to Land  Administrative Hierarchy           Maoist Period                           1980s Production Team (Elementary Cooperative) Rights to manage and Sub-lease Rights to use Primary Owner  Ownership and land policy making    Composition of Collective Ownership  Sub-tenant  (Users)  Township Government  Village Committee / Association of Cooperative Villager Group/  Rural Economic Cooperative Farming Households   Local State intervention Authority to determine criterion for land allocation, land disposal and land use  107  changes just poured old wine into new bottles, because the local state remained as the local authority and the co-owner of collective land in the process of land requisition. This is also reflected in the fact that under this multiple-tier ownership system the township government was still allowed to hold in custody a large portion of land compensation fees.  What really transformed this multiple-tier ownership structure, I would suggest, was the implementation of the ?reserved commercial land? policy. Having introduced the ?reserved commercial land? to replace direct provision of employment opportunities and social welfare to the villagers affected by land requisitions, township government found no excuse to continue to hold the lion?s share of the land compensation fees. From the early 1990s onward, all land compensation fees were directly channelled to the village collectives. This marked the fact that the local government gave up its role as a co-owner in the collective land system and left the ownership rights to the lower level of organizations, namely, the former brigades (now Association of Shareholding Cooperatives) and production teams (now shareholding cooperatives). The conventional three-tier ownership system was thus transformed into a two-tier one (see Figure 4.3). The end result of the ?reserved commercial land? policy in the GLD, in a nutshell, was the retreat of the local state from the collective ownership system, leaving the authority of managing collective land and other assets to the self-governing grassroots organizations. However, this was no sign that the collective system in the GLD would be dismantled. Instead, a more self-contained, collective village-based society was now in the making as a result of ownership restructuring.    After the retreat of the local state, village collectives began to hold in custody more than 70 percent of the land compensation fees and oversee the management of the ?reserved commercial land? after land requisitions. Under the collective land system, all compensation for land value belonged legally to the membership of the Villagers? Group. Therefore, after paying qingmiao buchang fei to the affected farming households, the remaining portion of land compensation fees and all incomes derived from the ?reserved commercial land? were placed under the management of the shareholding cooperatives. This mandatory arrangement was undoubtedly a key factor sustaining the continuity of the collective system in the area. However, such an institutional arrangement was unlikely to succeed if it could not effectively meet villagers? principal needs. Interviews suggest that most villagers were willing to accept the collective management of their land compensation fees chiefly for two major reasons.  108   Figure 4.318  Rescaling of Collective Ownership in the GLD      First, villagers were aware that the total amount of land compensation fees would not be sufficient for them to sustain their livelihood once they had lost their arable land to urbanization. As mentioned earlier, the compensation fee in the late 1990s was upwards of about 100,000 yuan per mu. In the villages I visited, land per capita ranged from 0.8 mu to 1.45 mu. This means that each villager could receive at most a total amount of no more than 145,000 yuan, even if all compensation fees were to be directly distributed to them. In these situations, only a few villagers who had been able to firmly establish their income sources from elsewhere and working in urban areas desperately wanted to get back all cash compensation from the village collectives. Most of them did not oppose the central pooling and management of their money and assets by the village collective, expecting that a higher rate of monetary return and the provision of better community facilities and social welfare services would be achieved.  Second, the collective organization had constituted part of the established system and lifestyle of villagers for more than half a century. In the Maoist period, they were dependent on their village collectives for coordinating agricultural production and all social welfare provisions. In the 1980s, the state still offered very limited financial contributions for rural infrastructures and                                                 Brigades Production Teams  Government Villagers?/ Residents? Committee Villagers? Committee / Association of Economic Cooperatives  Economic Cooperative  Association of Shareholding Cooperatives Shareholding Cooperative  Commune  Township Government  Township Government  Three-tier Ownership in the Maoist Era  Three-tier Ownership in the 1980s  Two-tier Ownership since the 1990s 109  social services. Villagers therefore still relied on their collective organizations for most public goods and services, even though they no longer had to work under the direct supervision of the production teams. Given this long dependent relationship with the village collective, they did not feel an urgent need to abandon it, especially when they could see a steady improvement in their incomes and social welfare services during the 1990s. Villagers? willingness to accept the central management of land compensation fees led to a rapid expansion of their collective assets. For example, in a village which had lost all arable land to urbanization by 2002, the village collective now held nearly 100 million yuan of compensation fees.   The growing collective incomes continuously enhanced the financial prowess of village collectives to engage with the provision of village facilities and social welfare services. The development of village-based welfare systems in the GLD witnessed two major breakthroughs in the reform era: in the 1980s, village collectives could only afford to construct village facilities such as village roads and water supply systems. In the 1990s, with the expansion of income derived from land compensation fees and the use of ?reserved commercial land,? social welfare provision by village collectives were extended to a much wider scope of services. In some of these villages, various social facilities and services, such as kindergartens, elementary schools, medical clinics, nursing homes for the elderly, and community recreational centres which comprise a football field, a basketball court, a table tennis room and a mini library, were usually provided free of charge for village members. As a result, some former villagers who had moved to cities in the early 1980s even tried to buy back their agricultural hukou in order to enjoy these social welfare and shareholding benefits. In addition to the provision of better community infrastructure and services, villagers established self-organizing security teams (zhibaodui or hucundui) to maintain local order. Expanding populations of migrant workers in urbanizing villages have been seen as a major source of hygiene problems and social security threats, presenting a great challenge to local governance in chengxiang jiehebu (i.e. periurban areas) (see Liu, et al., 1998; Liu et al., 2004). Indeed, this phenomenon was also the experience of the GLD. Historically, villages received little assistance from police forces and trust was largely built on networks familial ties and personal relationships. The influx of migrant workers, which nearly doubled the village population, saw immediate strain develop in the village?s social fabric. As villagers saw theft rates and rates of public disturbance climb steadily, they became increasingly fearful of the threat their new neighbours posed. In response, they mobilized to self-organize security teams to manage this perceived threat. During my visits to some villages, local village 110  cadres told me that their security teams comprised more 100 members who were mostly villagers. The annual operational costs of their security teams required about 1,500,000 yuan and all these expenses were shouldered by the village collective.  State versus Village Collectives, Collectives versus Villagers  Obviously, the results of the ?reserved commercial land? policy were not only what the local policy makers had originally intended, that is, providing social security for villagers who lost arable land to urbanization. It was the unexpected growth of territorial and financial autonomy, which had empowered villagers in their political and economic relationships with the local state. After land requisitions, the collective organization in the urbanizing villages was constantly strengthened rather than weakened. This ?collectivization?, however, was apparently different from that in the Maoist era. Between the 1960s and the 1970s, the collectivization movement had been compulsory and was deployed by the state as a tactic to ensure its extraction of surplus from the countryside for socialist industrialization. In the new round of collectivization during the reform period, the village collectives were no longer merely the agents of the state to implement state policies. Now, self-organized village collectives gave better social services for villagers and provided a platform for villagers by which they combined to strengthen their collective bargaining power in dealing with the local state. Hsing (2010) describes this type of village autonomy as a form of ?village corporatism,? by which villagers self-organized to resist land requisitions from the extractive local state. This conclusion, however, only addresses one side of the story, overlooking the internal conflicts between the village collectives and individual villagers. The realities in the GLD suggest that the contestations between the local state and the increasingly autonomous villages in the process of land requisitions were more complex.   Villagers did not single-mindedly resist land requisitions. Under the ?reserved commercial land? policy, the more arable land was requisitioned, the more commercial land the villagers would get back in return. In the GLD, the fact is that villagers who had lost their arable land were richer than those who had not. The former villagers in the GLD, like their counterparts in the Pearl River Delta region, were not landless farmers displaced by the state?s land requisitions; they were able collectively to own a certain portion of commercial land after land requisitions, with which they could generate much higher income than keeping their arable land in agricultural production. In this context, villagers remained hostile to state-led land requisitions not because 111  they fought for keeping their tiny fields to sustain family farming, which could only allow them to live barely above subsistence. Rather, taking land requisition as an opportunity, villagers adopted a resistance strategy with a view to removing all extractive measures imposed by the socialist state through the urban-rural dualist system and safeguarding their legitimate shares of the windfalls generated in the new rounds of urban transformation. In this sense, the concept of ?fair compensation? that villagers perceive is by no means merely the so-called open market value of their land and housing or the relevant displacement costs. Villagers perceived that the fair compensation should at least include two more essential elements. One was ensuring their access to adequate social welfare services, both physical and social, of which they had long been deprived by the state. Another was about ensuring adequate and sustainable income sources for villagers who had lost their land to urbanization.    There is no question that the growth of village autonomy has strengthened the collective power of villagers to defend their interests in dealing with the state. However, this empowerment has occurred in conjunction with the increasing conflicts between the village collective and individual villagers. One of the key conflicts was derived from the fact that the rapidly expanding collective wealth after land requisitions created loopholes for local cadres to manipulate the management of collective assets, leading to increasing internal tensions between the village collective and individual villagers. These internal tensions provided opportunities for the local state to re-assert its power in the urbanizing villages. As a result, local state building is no longer merely a result of land grabs for local accumulation. It is also a process of policy adjustment to consolidate the local state?s power in villages with the aim of regaining control over urbanization. In what ways the GLD government has been able to strengthen its control in the urbanizing villages will be further explained in the next three chapters.  112  Chapter 5      Shareholding Reforms and State-village Power Restructuring     The shareholding cooperative system was once a popular research topic for scholars interested in studying market reforms in rural China. Most of the existing literature treated the shareholding cooperative as a transitional institution under socialist China?s emerging property rights regime. The principal debate was about how far shareholding cooperatives would drive socialist China away from its command economy to a genuine market economy (see Vermeer, 1999; Lin and Chen, 1999; Whiting, 1999; Jefferson and Su, 2006). In my research, however, the GLD experience suggests that the development of shareholding cooperatives was not simply a process of market building or property rights reforms. Instead, it constituted a consolidation of state power rather than a transfer of state power to the market. As shown in the previous chapter, after the implementation of the ?reserved commercial land? policy, the GLD government largely reduced its presence in the urbanizing villages. With the retreat of the state as well as the huge collective wealth generated from the land compensation fees and the ?reserved commercial land,? many urbanizing villages in the GLD were gradually developing into self-governing and fiscally independent communities. In this chapter, I will explore how the local government has used shareholding reforms as an instrument to establish its regulatory power over the urbanizing villages, and how shareholding reforms have reshaped the power dynamics between the local state, the village collective and villagers.    Purposes of Shareholding Reforms     Shareholding reforms in the GLD area began in the mid-1990s. Over the past two decades, shareholding reforms have been managed by the local government to achieve three main objectives. The first was to delineate villagers? rights to collective assets through share classification and allocation. The second was to encourage compliance amongst villagers by linking villagers? entitlement to dividends to compliance with state policies and laws. The third 113  one was to promote the accountability of village cadres in the management of collective assets through organizational restructuring. These objectives underscored two conflicting strategies adopted by the local state for shareholding reforms.    One strategy was empowering villagers to oversee village cadres in the management of collective assets. As discussed earlier, village cadres are now responsible for the management of all collective assets including land compensation fees and the rental income derived from ?reserved commercial land,? which constitutes an essential source of funding for the villagers? medical services and pension schemes created after land requisitions. The local government decided to put village cadres under local control to avoid the potential mismanagement of collective assets by incompetent or corrupt village cadres that would put the local social welfare system at risk. According to the GLD Statistical Yearbook 2009, there are 517 shareholding cooperatives which are under the administrative purview of 52 Residents? (or Villagers?) Committees in the GLD. In attempting to regulate these cooperatives, the government quickly discovered that it was nearly impossible to effectively monitor their operation at a distance. Hence, empowering villagers to detect more effectively any malpractices on the part of village cadres became a practical strategy.   Another strategy was dampening the loyalty of village cadres to villagers in order to limit the opportunity for collusion between them. After villagers were empowered in the management of collective assets, village cadres were likely to act in the interests of fellow villagers rather than the state in hopes of securing villagers? votes in subsequent elections. This dynamic encouraged village cadres to co-align with villagers in resisting the state. As discussed in chapter 4, land requisitions gave rise to new complexities in the relationships between village cadres and ordinary villagers. The major duties of village cadres shifted from merely implementing state policies to serving the needs of their fellow villagers through the improvement social welfare and village facilities. Despite the internal conflicts in the management of collective assets, village cadres and villagers gradually cultivated their joint interests in dealing with the state. Moreover, the revival of kinship power enhanced villagers? social cohesion of the villagers as well as their power to stand up to the state. This movement was reflected not only in the fact that local cadres co-aligned with villagers to bargain with the local government for higher land compensation in land requisitions. It was also evidenced by fact that local cadres tended to identify themselves more with villagers than the state when they dealt with village housing 114  development in the 1990s. Under such circumstances, it is not surprising that the local government sought to strengthen its direct supervision of local cadres in order to regain control over the increasingly autonomous villages.  These two strategies, which often conflicted with each other, combined to shape the outcomes of shareholding reforms in today?s GLD. In the following sections, I will explore the process of shareholding reforms in the GLD and their impact on the interactions among the local government, village cadres and villagers. Much of the contemporary Chinese literature tends to view the emergence of shareholding cooperatives as a spontaneous process driven by the initiatives of rural farmers (e.g. Tang and Chen, 2003; Li, 2004). However, shareholding reforms in the GLD took place in the mid-1990s, when many policies and laws governing shareholding cooperatives were first being developed. The actual process of shareholding reforms in the GLD, therefore, was top-down rather than bottom up. In this regard, it is helpful to look back into the history of nationwide shareholding reforms in rural China a decade earlier, before moving on to discuss how local practices interacted with the state polices at the upper level governments in the implementation of shareholding reforms in the specific case of the GLD.   Background of Shareholding Reforms in the 1980s  It is true that the shareholding cooperative as a specific form of organizational structure in rural China had come into being before its legal framework was formally established. Shareholding cooperatives first appeared in Shenzhen between 1981 and 1983, when rural villages there were undergoing massive waves of land requisition and urbanization. To help villagers make a living after land requisitions, the Shenzhen Municipal Government offered land to the affected villagers as part of their compensation:  a residential site of 150 square meters for each household and 15 square meters of non-agricultural land for each person for self-development (Tang and Chen, 2003). Drawing on the ?Elementary Cooperative? that they had experienced in the early 1950s, villagers unified the management of their land to form a so-called shareholding cooperative. The land was then centrally managed by the shareholding cooperatives for development and leasing; farmers became shareholders and were entitled to dividends at the end of each year. At the same time, many TVEs in Wenzhou and the Sunan Region adopted the shareholding system to create economic incentives for their employees. The formal legal 115  framework governing the operation of these cooperatives was absent until 1985 when the CCP?s Decree No. 1 was issued.30  In Article 8 of the document, the central government commented positively on the grassroots nature of shareholding cooperative and encouraged local officials to adopt it as a model of incorporation for rural enterprises.   In attempting to formalize the emerging shareholding cooperatives, the central government did not impose stringent restrictions on their operation. Assaad (1996) suggests that formalizing the informal may lead to only a partial formalization and a greater tolerance of ?informality? on part of regulatory authorities. China?s shareholding reforms during the 1980s had similar outcomes.  By clarifying the rights of individual farmers? to their collective assets, shareholding reforms transformed many rural enterprises from ?collective? organizations to ?private? companies. The central government acknowledged that the shareholding cooperative system would benefit rural industrial development. Yet simultaneously, it saw this transformation as a major challenge to the established national ideology, which proclaimed collective institutions as the bedrock of socialism. In its attempts to overcome this contradiction, the central government employed the deliberately ambiguous term ?shareholding cooperative? to downplay the presence of partial privatization in the management of the shareholding cooperatives. Disturbed by the presence of certain market mechanisms (such as the form of a limited shareholding company) that had been built into the collective system to ensure a fairer distribution of interests among stakeholders in rural enterprises, the central government sought to defend the notion that a collective regime was still a fundamental pillar of its socialist economy. As some scholars (Guo, 1999; Vermeer, 1999; Tang and Chen, 2003) suggest, the primary purpose of the central government?s decision to officially recognize shareholding cooperatives was to steer the development of rural enterprises, which were exhibiting some signs of de facto privatization as a result of shareholding reforms, closer towards to the path of the collective economy.   The shareholding cooperative was often criticized for being a self-contradictory, hybrid organization, which lacked most of the key attributes of both a shareholding company and an ordinary cooperative. Such a conflicting institutional arrangement, however, did not jeopardize its capabilities in meeting local needs. Local officials were happy with this direction of shareholding reform advocated by the central state. The vaguely-defined legal status of                                                           30 The title of the document was ?Ten Policies for Further Revitalizing Rural Economy.? 116  shareholding cooperative itself helped avoid rigid control by the central state in the formalization process. It allowed rural enterprises to legitimately use the title of ?shareholding cooperative? to support various ownership patterns of the collective assets based on local specificities (Guo, 1999; Whiting, 1999). As a result, the local application of this loosely-defined and officially-recognized system gave rise to many different forms of shareholding cooperative across rural China, which in turn generated dissimilar local governance processes resulting in the varying relationships between the local state and society.   In the Pearl River Delta region, the development of shareholding cooperatives was closely associated with direct foreign investment. As the first area to implement the open door policy, the Pearl River Delta region pioneered economic reforms and urbanization through a progressive engagement of attracting foreign investment and promoting export production. Apart from FDI investment, property investments from overseas Chinese also contributed to the swift development of the region. In the mid-1980s, many municipal governments, such as Guangzhou and Shenzhen, started to get involved in real estate developments with the aim of generating revenue for the construction of urban infrastructure. Given the fact that the purchasing power of local residents at that time was still very weak, the target customers of their projects were those overseas Chinese who had close relatives in villages. To attract them, two to four urban hukou were usually attached to the residential apartment that they brought, so that they could help their rural relatives move into the city. These overseas buyers, as Chan (1992) suggests, paid for much of the ?urban infrastructure cost?, which further accelerated urban development in the Pearl River Delta region.   In these situations, like the local state of the GLD, many local governments in the region have devoted their main efforts to make land available through land requisitions for overseas investors rather than directly assist local village collectives and villagers in developing rural enterprises. As discussed in Chapter 4, the villagers in the Pearl River Delta region received some ?reserved commercial land? as part of their compensation package by the local state for taking their farmland. Such land could be used for limited commercial and urban uses by the village collectives to generate a recurrent source of income for villagers. Seeking a better way for the distribution of income, some village collectives introduced a shareholding system under which the villagers? rights to collective assets were transformed into shares. The emergence of this new system has been widely viewed as a process of ?spontaneous re-collectivization.?, It was so 117  termed because a previous period of Maoist collectivization had been undermined by the Household Responsibility System, which had tended to strengthen individualized, self-interested farming activities at the expense of co-operation among households at the rural collective level. In this new wave of collectivization, however, shareholding cooperatives re-activated the economic potential of pooling together collectively owned assets and strengthening household co-operation. However, the new shareholding cooperatives no longer organized villagers to work in agricultural production as in the past. In the Pearl River Delta region, they chiefly rely on the leasing of landed properties to generate income for village collectives.   Shareholding cooperatives in the Pearl River Delta region developed three main characteristics. First, their core business was concerned with land and property leasing (Tang and Chan, 2003; Li; 2004; Po, 2008). Second, their organizational structure was inherited from the former Production Team (now Villagers?/Residents? Group), which was officially recognized as the legal holder of collective land and assets. Third, they envisaged their purpose of not only to produce dividends but also to generate stable income to support village facilities and social welfare provisioning (Guo and Zhou, 1997; Zhou and Zhang, 1997). Addressing these organizational features, policy makers and scholars in China have tended to classify the shareholding cooperatives in the Pearl River Delta region as ?Neighbourhood Shareholding Cooperatives (shequxing gufen hezuoshe)? in order to differentiate them from their counterparts in the Wenzhou and Sunan regions.    Wenzhou and Sunan regions represented a different form of power relationships between the local state and villages, given that their economic development in the 1980s relied mainly on local capital rather than foreign direct investment. In the Wenzhou area, many rural enterprises operating in the form of shareholding cooperative were mostly initiated by private and family-based capital, but had very close relationships with the township governments. As Parris (1993) suggests, local governments colluded with local stakeholders to pursue rapid industrial development by assisting rural enterprises in obtaining raw materials, technology and loans, whose supplies and allocation were still monopolized by the central state at that time. The development of TVEs in Sunan region also relied on local capital. However, compared with 118  those in Wenzhou area, they were much more ?collective? in the sense of the former commune-brigade enterprises (shedui qiye) in the Maoist era.31   Sunan was one of the traditional centers of artisan and handicraft activities in imperial China. Sunan farmers had long worked on non-agricultural sideline productions such as embroidering, sewing and making handicrafts of bamboo and grass, leading to the prevalence of rural industries in the region before the establishment of the PRC. During the early 1950s, the CCP once introduced a series of policy measures to support the development of agricultural sideline industries in rural China. These included providing governmental subsidies and loans, organizing the Rural Supply and Marketing Cooperatives (Nongcun Gongxiao Hezuoshe) to develop markets for farmers to trade their sideline products, and establishing Rural Cooperative Banks (Nongcun Xinyong Hezuoshe) to offer a source of finance for their development. These measures helped the post-war recovery of rural industries in Sunan region and encouraged local farmers to form cooperatives for their sideline production. Later in the communal era, all these non-agricultural cooperatives were collectivized and were subsequently reorganized as commune-brigade enterprises after the Great Leap Forward. During the 1960s and the 1970s, these organizations were denied access to the necessary raw materials and other resources, which were largely reserved for the state enterprises (Unger, 2002). Having relied on the very limited domestic resources, these collective organizations mostly produced chemical fertilizer and farming machinery to support agricultural development rather than channeling agricultural production to other activities (Ho, 1994). In the reform era, commune-brigade enterprises were granted tax holidays and were permitted greater flexibility in the operation of industries in textile, food and beverage, construction materials, chemicals, transport and so on. Moreover, these enterprises could sell their products to the end-users directly without having to await state approval. With the reorganization of communes and brigades into townships and administrative villages, commune-brigade enterprises were reorganized as TVEs.   Various forms of shareholding cooperatives yielded a great variety of governance restructuring at the local level (See Figure 5.1). In Sunan and Wenzhou areas, urban and TVE development resulted in an upward power restructuring which combined ?commune? and ?brigade? of the Maoist bureaucracy into a single entity, by which the local governments had direct control over                                                           31 Geographically, Sunan region generally refers to the southern part of Jiangsu Province, bordered in the north by the Yang Tze River and in the south by Shanghai and Zhejiang Province. 119  local production. In contrast, the neighbourhood shareholding cooperatives in the Pearl River Delta Region saw urban development occur in conjunction with state power decentralization, by which local governments withdrew their direct involvement in the operation of rural enterprises. These varied forms of power structure shaped different strategies of local governance and the relationship between local officials with villagers. In Sunan and Wenzhou areas, the more developed TVEs were, the more powerful the local state became. Oi (1992) describes this interdependent relationship of the township governments with the rural enterprises as ?local state corporatism.? Under this system, the local state treated rural enterprises as parts of a larger corporate structure and participated in the decision making of the development and management of these township enterprises. In the Pearl River Delta region, local governments did not establish corporate forms of economic growth and dominated economic decision-making, as they did not get directly involved in the operation of shareholding cooperatives. They acted less like a dominant corporate authority and more like a regulator and provider of administrative services.   Figure 5.119  Local State-society Relationships in Periurban China: Regional Diversity   Communities (Individual households and villagers)    Communities (Brigade + production teams and villagers) Commune  Brigade Production Team  Wenzhou/Sunan Model Pearl River Delta Model Local State (Township Government + brigade and production teams) Local State (Township Government)  TVE and Shareholding Cooperatives in the 1980s Maoist Rural Bureaucracy  120  Formalities versus Informalities: Implementing Shareholding Reforms in Villages   The power dynamics between the local state and villages in the GLD, which had been shaped by the local state?s land requisitions for foreign investment, shared striking similarities with their counterparts in the Pearl River Delta region as a whole. However, shareholding reforms in the GLD began in the mid-1990s, when a regulatory framework for shareholding reforms had already been formulated.32 Unlike those in Shenzhen which were mainly driven by local initiatives, shareholding reforms in the GLD were a state-led rather than a bottom-up process. From the early 1990s onward, the central government began to tighten control over shareholding cooperatives by setting up standardized rules and procedures for their operation. Following national legislation in February 1990, the Provincial Government introduced the Organic Law on Rural Neighbourhood Economic Cooperatives three months later in May of the same year. In Guangzhou City, the municipal government started in 1987 to experiment with shareholding reforms in two villages, Yangji and Dengfeng, which had experienced rapid expansion in their collective assets and income as a result of urbanization. The ongoing attempts by the higher level governments to strengthen shareholding cooperatives as a manageable system during the 1990s provided policy support for the municipal government to promote shareholding reforms in other local villages. Shareholding reforms in the GLD, therefore, were driven by the government?s attempts to regulate the management of the ever growing village collective wealth arising from land requisitions.    In my interviews, some village cadres, who had experienced the shareholding reforms of the 1990s, admitted that they themselves did not understand what shareholding cooperatives were about before they were asked to implement them. After the introduction of the Regulations Concerning Audit on Village Collective Assets at the municipal level in 1995, the Luogang                                                           32 In February 1990, Provisional Regulations on Farmers? Shareholding Cooperatives was promulgated. By these regulations, the Chinese government defines ?farmers? shareholding cooperative? as an organization which has the following institutional features: (1) comprising three or more farming households which voluntarily pool their assets (such as capital, technology or labor) for non-agricultural production; (2) accepting the guidance of state plan and policy; (3) distributing profits and benefits according to shares and individual performance; (4) accumulating funds for public goods and services provision; (5) implementing democratic management and being able to bear civil liabilities; and (6) requiring approval from the government as a prerequisite to establishment. In December 1992, the Chinese government extended its regulations to TVEs by introducing a Notice Regarding the Promotion and Improvement of the Shareholding Cooperative system in TEVs.  In this Notice, ?farming households? are replaced by ?workers? / ?investors? in order to reflect the fact that the main actors in many rural enterprises were not limited to rural farmers only.   121  township government required the Villagers? Committees to verify and report all their collective assets (including land, landed properties and cash) held by their cooperatives. Villagers? Committees were then instructed to reorganize their economic cooperatives into shareholding cooperatives by converting villagers? entitlements to these assets into two types of shares: collective (jiti) shares and individual (sheyuan) shares (see Figure 5.1). Collective shares, which were equal to 51 percent of the total shares, were to be held by the shareholding cooperatives. The remaining shares were distributed to individual villagers. Above the shareholding cooperatives in each administrative neighbourhood (village or shequ) was the Association of Shareholding Cooperatives (jingji lianshe), which was formerly known as ?brigade? in the Maoist era. In each Association of Shareholding Cooperatives, a Board of Directors was formed to implement the decisions of the Shareholders? Meeting. The number of Directors varied from three to nine, depending on the size of the Cooperative. Simultaneously, a Board of Supervisors, comprising three to four elected villagers? representatives, was formed to oversee the performance of the Directors and monitor the financial conditions of the shareholding cooperative. To avoid conflict of interests, no members of the Board of Supervisors could be a relative of the members of the Board of Directors. Under this organizational structure, the operation of shareholding cooperatives was governed by a set of rules and procedures derived from the so-called Articles of Association.  My field interviews together with a study of the Articles of Association issued by the local shareholding cooperatives in the 1990s suggest that two major factors had affected the allocation of individual shares in shareholding cooperatives. First, indigenous identity and hukou registration were instrumental in determining a villager?s entitlement to his/her shareholding benefits. Only indigenous villagers who had registered their hukou with the village could enjoy full benefits from their shareholdings. Indigenous villagers generally meant those resident villagers who were descended through a male line of a clan within the village. If an indigenous villager relocated his household registration away from his village, he would only have a partial entitlement to his shareholding benefits. The same principle was also applied to non-indigenous villagers who had registered their hukou with the village. Most villagers in this group belonged the ?sent-down youth (zhiqing)? of the Maoist era, who had returned to the city but had retained their household registrations in the village. These ?ex-villagers? and ?non-indigenous villagers? received partial shareholding entitlements because they were considered to have once made some contribution to the village. But what, one may wonder, is ?a contribution? to the village? 122  Cadres explained that contributions might include the participation in collective agricultural production during the Maoist era or participation in the Household Responsibility System during the 1980s. In the latter period, villagers were still required to pay rural taxes and meet the grain production quota imposed by the state, and in a departure from the Maoist era, the System allowed them to keep the remaining harvest for own consumption. Therefore, the Household Responsibility System was not only about the household rights of leasing land from the village collective for the purpose of family farming, but also the family obligations to cooperatively assist the village collectives in meeting agricultural tax levies and state production quotas.   The age of a villager was another important factor that determined his/her share allocation. As shown in Table 5.1, a villager?s share entitlement usually varied according to his or her age in three different phases of the life cycle: it increased in direct proportion to one's age in Phase 1 (between 1 and 29), reached the height in Phrase 2 (between 30 and 60), and slightly decreased in Phase 3 (above 61). Under this formula, villagers were able to earn more shares as they grew older. This arrangement inherited some similar principles of the work point system at the Maoist period. But it took into consideration the economic needs of the individual at different stages of his/her life cycle, rather than his/her labour contribution under collective agricultural production. In the Maoist period, the principle of equal remunerations for male and female villagers was adopted to promote gender equality so as to win the support of women (Ngo, 2009). The same principle of equal treatment continued to be adopted in shareholding reforms, allowing female villagers to obtain the same shares as male villagers. To accommodate the changes in age and village population, the shareholding cooperative would re-adjust the share entitlements of all eligible individuals every three years.  State policies set the formal rules in governing the management of collective assets, but they could not possibly prescribe all the implementation details of shareholding reforms at the local scene. Shareholding reforms at the GLD were also substantially influenced by the informal village rules and covert practices embedded in village norms and traditions. A case in point is the way share entitlements of female villagers were calculated. As discussed in Chapter 3, ancestral lines and kinship networks constituted the foundation in organizing village society in the GLD. Villagers acquired their eligibility to participate in village affairs as well as their entitlement to village assets mainly through birth or marriage. These village traditions, informal norms and rules, which had been suppressed by the state during the Maoist period, began to 123  revive and influence village governance during the reform period. One vivid example was related to gender inequality in traditional villages.   Table 5.18   Organizational Features of Shareholding Cooperatives in the GLD    1990s Since 2003  Classification of Shares : 1. Collectives Shares (51%) 2. Individual (sheyuan) shares (49%)  1. Collective shares (51%) 2. Individual Shares (49%) ? Shequ Shares  ? Shehui Shares  Eligibility of Shareholders :   By indigenous identity and hukou registration  Shequ Shares: by indigenous identity, hukou and inheritance Shehui Shares:  by past contribution to the village collective Allocation of Shares :    Age 0 ? 3 4 ? 6 7 ? 9 10 ? 12 13 ? 15 16 ? 18 19 ? 20 21 ? 23 24 ? 26 27 ? 29 30 ? 60 Above 61 No. of Shares 1 2 3 4 6 8 10 13 16 19 23 22  Type of shareholder              No. of Shares  Shequ                                     200 Shehui                                    60     Re-allocation:  Every three years  Fixed shares (no readjustment) Participation and Withdraw :   Voluntary Participation is voluntary but withdrawal is disallowed after participation.  Initial Price of Each Share :  2 yuan  3 yuan  Distribution of Share Dividends:  By shares and by population  By shares only Restrictions on Shares : All shares were non-transferrable. Moreover, they cannot be used for mortgage purpose.   All shares are non-transferrable. They cannot be used for mortgage purpose. Shequ shares are inheritable. Shehui shares are non-inheritable; they are forfeited by the collective after the holder passes away.   Rights of Shareholders : ? Right to vote directors nominated by the board (one man one vote) ? Right to be nominated for directorships  ? Right to receive dividends  ? Right to participate in the operation of the shareholding cooperative ? Right to monitor the financial status of  the shareholding cooperative  ? Right to monitor the performance of village cadre and report their malpractice to the government  ? Right to share all benefits and welfare services provided by the shareholding cooperatives  (Notes: Shehui shares pay the same dividends as Shequ shares. However, the holders of Shehui shares do not have right to vote or to be voted.)  Obligations of Shareholders :   ? Implementation of government policies and regulations (such as birth control, conscriptions etc.) ? Implementation of the resolutions made by the General Meeting of Shareholders ? Implementation of village regulations and customary rules    Data Sources: Articles of Association from various shareholding cooperatives and field interviews.  124   The long-standing dominance of men over women in village affairs made the actual process of share distribution much more complex than what they were stated to be in their Articles of Association.  When a village man married a woman from another village, his family acquired additional shareholding entitlements to the collective income from that village. But, when an indigenous female villager married a man outside the village and moved her hukou registration out of her own village, her original entitlements and rights were forfeited. This rule originated from the traditional village norm that a married daughter no longer belonged to her village. A Chinese saying that ?a married daughter is just like water that has been thrown out (jiachuqu de nver, pochuqu de shui)? aptly described such a traditional way of thinking. Another practice is that an indigenous female villager did not move out from the village after getting married; instead, she invited her husband to move to her village. In traditional Chinese village society, this was called ?uxorilocal marriage (ruzhui).? The bridegroom was asked to reside with the bride?s clan because her family did not have any male descendants. This served the purpose of maintaining the family line and taking care of the elders in the bride?s family. This son-in-law, however, was not eligible to enjoy shareholding benefits even if he adopted the surname of his wife?s clan and registered his hukou with the local village. Application of local village traditions to the shareholding reforms thus resulted in inequitable gender discrimination and aroused the discontents of some villagers.  Problems of Shareholding Reforms in the 1990s  Shareholding reforms in the 1990s generated a range of undesirable consequences, which ultimately led to widespread tensions between village collectives and villagers. Dissatisfaction of some female villagers was one issue. More prominent problems were associated with the management of collective assets. One problem, for example, was that a periodic re-adjustment of share entitlements among villagers created loopholes for village cadres to manipulate the allocation of shares. Village cadres had great discretionary power in dealing with share allocations, especially for female villagers and non-indigenous villagers whose entitlements were not clearly stipulated under the formal state policy. For example, some village cadres were found to have given a favourable treatment to those who were their own family members or close relatives in share re-allocation. Another major problem was that the dividend policy did 125  not reward the village cadres for good management of collective assets. During the 1990s, village cadres were not paid dividends during their terms of appointment. The initial objective of this policy was to avoid possible conflicts of interest. However, it turned out that village cadres had little incentive to enhance the operational efficiency of the shareholding cooperatives because their remunerations were not dependent on the business performance of the collectives. Even worse was that some cadres abused their power in the negotiation and deal-making with external partners in order to line their own pockets. The leasing out of collective landed properties was usually made through the personal networks of local cadres rather than public tender. As a result, the ?reserved commercial land? of the villages was usually contracted out at extremely low rents, compromising the financial well-being of the village collectives.   Villagers? disinterest in shareholding reforms was another major reason for the increasing malpractice of village cadres in the management of collective assets. Villagers? reactions to shareholding reforms in the 1990s were subject to the influence of two factors.  First, villagers were generally disinterested in the shareholding system of the 1990s primarily because it generated only tiny, unstable dividend income. Under the collective land system, compensation for land requisition by the local state was legally claimed by all members of the Villagers? Group. After deducting the payment of qingmiao buchang fei to the affected farming households, the remaining portion of land compensation fees and all income derived from the ?reserved commercial land? were shared by all eligible villagers, including not only those whose farmland was requisitioned but also those other villagers whose land remained untouched by land requisition. Throughout the 1990s, land requisitions in many villages were conducted in several phases. At the very beginning, when only a small portion of village land was requisitioned, the interest income generated from compensation fees and the rental income from the ?reserved commercial land? were both very limited. When this small amount of income was shared by all villagers, it was not surprising that the annual dividend payment to each villager was low. Some villagers recalled that they had received only 300 to 400 yuan of annual dividends at that time. In addition, as the village population kept growing, dividends per capita kept decreasing in some villages. Under these circumstances, not many villagers were enthusiastic about the operation of shareholding cooperatives. As some village cadres explained, only those villagers whose arable land had been requisitioned were interested to know about shareholding reforms. After having lost their farmland and in turn, their livelihood, the distribution of dividends became a matter of extreme importance, as they became more and more dependent on such income for their 126  livelihood. Those villagers who could still make a living out of cultivating their farmland did not actively participate in the meeting of shareholders. They had a general perception that the results of such meetings, whatever they were, had little bearing on their lives. The fact that most villagers paid scant attention to monitoring the operation of shareholding cooperatives created ideal conditions for some corrupt village cadres to easily pursue their own self-interested undertakings in managing collective assets.  Second, traditionally, villagers tended to be acquiescent to village cadres. Both lived close together in the same village. They also inherited strong inter-familial ties cultivated by their ancestors who had made their home in the same place for generations. It was natural that villagers and cadres shared various kinds of intimate relationships with each other (such as being close relatives or neighbours). As a result, villagers usually tried to avoid direct clashes with cadres, unless they felt that their wrongdoings had negatively affected their well-being. During the 1980s and the early 1990s, village collectives contributed to the provision of village infrastructure and social welfare services, and villagers earned their main income from farming and fruit selling to nearby cities. Although the performance of village cadres in managing the collective assets had some impacts on the overall quality of village infrastructure and welfare, it did not directly affect the principal earnings of villagers. As long as villagers could maintain their subsistence from agricultural production in their own farmland, most of them did not bother with the management and distribution of the collective assets. Moreover, most villagers found that village infrastructure and services in general had significantly improved in the 1990s. As a result, villagers tended to be tolerant and acquiescent, even though they were aware of the malpractices of some village cadres, such as receiving kickbacks from contractors involved in village construction projects.  However, large scale land requisitions by the local state completely altered the villagers? attitudes and approaches in dealing with village cadres. This is because land requisitions increased the reliance of villagers on collective income, elevating them to an unprecedented level of enthusiasm for participating in the management of collective assets. When their lifelong assets were in the form of cash (land compensation fees) and this huge sum of money was in the hands of village cadres, villagers found that their well-being was no longer under their sole control. Many started to worry that they might lose all of what they owned if corrupt cadres, for example, were to run away with their money. As mentioned in Chapter 2, some village cadres 127  used the collective funds for their own purposes and even gambled with villagers? land compensation fees. Seeing that land compensation fees might disappear overnight, villagers now kept a close surveillance on the village cadres in managing their collective funds and their collective income. They started to question how their money had been used and how much village cadres had really received from operating the collective assets. Participation in formal elections for the Villagers? Committee and the shareholding cooperatives provided a means for villagers to express their support to or dissatisfaction with villager cadres.   ?Fixing Share Rights? in 2003  In the late 1990s, as more and more arable land was requisitioned by local governments and the amount of ?reserved commercial land? under the management of village collectives expanded enormously, more and more problems in relation to the management of collective assets were brought to the surface. As mentioned in Chapter 2, some villagers had even burned the offices of the Residents? Committee when they discovered that their collective assets had been mismanaged. Because the GLD government recognized that any failure to address these grievances could eventually erupt into a more profound political crisis, it launched a second round of reforms for the shareholding cooperatives in 2003.   The first important step was to require all Shareholding Cooperatives to reclassify their shares into two categories: shequ shares for indigenous villagers and shehui shares for villagers who had previously made contributions to the village but had re-located their hukou registration elsewhere. In defining the share entitlements, the GLD government sought to change the old system by imposing a new regulatory framework to formalize prevailing practices. As mentioned earlier, under the shareholding system of the 1990s, indigenous villagers who had moved their hukou elsewhere and non-indigenous villagers who had registered their hukou with the village where they now resided received only ?partial entitlements? to shareholding benefits. However, the number of shares they received depended greatly on the discretion of the local village cadres. After the shareholding reforms initiated by the district government in 2003, both groups of villagers were formally designated as shehui shareholders and their rights to the shareholding system were put under legal protection. To better balance the interests of indigenous and non-indigenous villagers, however, the government differentiated these two 128  types of shares by limiting shehui shares to receive only dividends while withholding their rights to vote in the Shareholders? Meetings. In other words, shehui shares received the same dividends as shequ shares, but their holders were not entitled to the voting rights that were held by shequ shareholders only. Furthermore, shehui shares were not inheritable, and thus reverted automatically to the corresponding shareholding cooperatives after the death of the shareholders.   The second step taken by the GLD government was to introduce a new regulatory framework, which was called ?fixing the share rights (guhua guquan),? for the management of shequ shares. Shequ shares could now be inherited by family members, but the total amount of shares no longer fluctuated with respect to age (sheng bu zeng), the growth in village population (jin bu zeng), or changes in identity status (chu bu jian).  Moreover, villagers could designate their shares to be inherited by any recipient of their choosing upon death, whereas in the previous system, shares had to be permanently returned to the shareholding cooperatives. This principle was called ?si bu jian? in Chinese.   The implementation of these four principles for share allocation and management required amendments to the Articles of Association of every Shareholding Cooperative. In this regard, the GLD government issued a new standard template of the Articles of Association. According to this template, each Association of Shareholding Cooperatives was required to revise its own Articles of Association to satisfy the local contexts of the village that it belonged to. To ensure legitimacy of the Articles, the draft had to be endorsed by the Meeting of the Shareholders? Representatives and then submitted for formal registration at the Community Management Bureau (Shequ Guanliju). Once the Articles were registered, any further amendments and revisions had to repeat the aforesaid procedures. Moreover, if the amendments and the revisions would result in changes in the appropriation of shares, they had to be supported by open ballots of more than 95 percent of qualified voters in the shareholders? meeting.  This new institutional arrangement disallowed share re-allocation and thus plugged the loopholes for cadre abuse in periodical share adjustments. The clear composition and stable structure of share rights encouraged transparency and accountability in the management of collective assets. Moreover, all shareholding cooperatives were required to contribute at least 20 percent of their total share income to the provision of community infrastructure and social welfare services such as medical care and old age pension.   129  After reviewing these two major rounds of shareholding reforms which were carried out between the mid-1990s and 2003, I will move on to discuss their impact on transforming the interactions among the local government, villager cadres and villagers. In this regard, I will focus on two questions. The first is about how shareholding reforms have created a link between dividend benefits of villagers and their compliance with state policies and regulations. The second is about how shareholdings have created a regulatory framework to provide new incentives as well as constraints for village cadres in the management of collective assets.      Shareholding Reforms and Social Control  Alongside these reforms, villagers in the GLD showed their growing interest in the management of collective assets. This is evidenced by the fact that many villages achieved a voter turnover of more than 80 percent in the last grassroots elections for the shareholding cooperatives and Residents? Committees. One possible reason behind villagers? active participation was that the shareholding reforms had resulted in a clearer delineation of villagers? ownership rights to their collective assets. Ownership rights increased their confidence in formal institutions and thus pushed them to increasingly turn to formal channels to articulate and defend their interests. Shareholding reforms empowered villagers to participate in the management of collective assets.  This, however, occurred in conjunction with increasing control from the state. Taking the opportunities of shareholding reforms, the local government attached compliance conditions to the villagers? entitlement to dividends.   Since its inception in the 1990s, the shareholding cooperative system has been adopted by the GLD government as a means to secure the villagers? support in the implementation of unpopular state policies. For instance, from the 1980s onwards, the central government began to require its local agents to strictly enforce the one-child policy. Birth control was adopted as a key criterion for assessing the administrative performance of local officials. Situated at the lowest echelon of the formal bureaucracy, local officials at the township level were obliged to carry out this unpopular policy, but were given very few policy instruments with which to implement it. Under these circumstances, what the local state usually did was to impose non-negotiable birth control targets on village cadres. To comply with this mandatory directive, some village cadres relied on 130  the use of coercive measures. For example, once an unplanned pregnancy was detected, the cadres could compulsorily send the woman to have a surgical abortion.   These harsh measures were not only unpopular with villagers, but were also demanding on local officials and village cadres. Introduction of shareholding cooperatives provided local officials an amicable mechanism to carry out this challenging task. In practice, this was done by establishing the link between compliance of villagers and their shareholding benefits. During the 1990s, for instance, villagers with an agricultural hukou were allowed to a second child. But, if a couple gave birth to a third child, their dividend payments could be suspended for three years as a penalty. And if a couple gave birth to a fourth one, dividend payments would be suspended for another six years. Similar punishments were also imposed on villagers serving a prison sentence. On the other hand, credits were granted to those villagers who were law-abiding and performed well in carrying out national policies and serving the village collective. For example, a couple who voluntarily received a ligation surgery after giving birth to their first child could receive extra dividends as bonuses. Active participation of villagers in military services and state affairs might also be counted as ?good? behaviour worthy of additional monetary rewards from shareholding cooperatives.                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                            During the implementation of the ?fixing share rights? program, this penalty and reward system was further strengthened in many of the shareholding cooperatives. There were obvious reasons why the local state had decided to strengthen them as a measure of social control since the early 2000s. After many rural villages converted into urban neighbourhoods, the implementation of birth control encountered more and more difficulties. Once their hukou status was changed to ?urban?, villagers had to strictly comply with the one-child policy as did all urban residents. Many villagers whose first child was a girl, therefore, tried to use all sorts of methods to escape from the surveillance of village cadres in order to have a second child. The government had substantially increased the fines in order to deter potential violators. But as some villagers suggested, while the one-off fine was undoubtedly a heavy burden, it was not insurmountable. Many could still pay it off with financial support from family networks. Faced with the increasing number of violations, the local state continued to make use of the shareholding cooperative system as a means of social control to secure compliance. It turned out that the increasing dividend income proved to be an extremely effective incentive to villagers. As a female villager suggested to me, though she desperately wanted to have another child, she could 131  not afford to do so because this would cost her family lose a dividend income amounting over 40,000 yuan per year.               Since more and more villagers were becoming urban residents, the GLD government became preoccupied with the many new administrative tasks developing in relation to employment creation, medical services, pension schemes, and so on. Realizing that shareholding cooperatives were an effective mechanism treasured by villagers, the District government began to attach new compliance conditions to villagers? dividend entitlements in its second round of shareholding reforms. Linking villagers? employment with dividend payment was one of the examples. After their land was taken away, many villagers lacked incentives to look for jobs. One reason was that it was difficult for villagers to find new jobs in the non-farming sectors. Another reason was that they were able to receive dividends and rental income from leasing their houses to migrant workers. Although the unemployment problem did not cause any immediate burden on the current social welfare and security system, the local government perceived it as a potential source of social problems because there was a trend that more and more jobless villagers were becoming addicted to gambling and drinking. In order to encourage jobless villagers to join the job market, the government stipulated that their share dividends would be reduced if they refused to take up the government job referrals more than three times in a given year. A local female villager told me that this was the key reason why she was still working as a janitor in the hotel, even though she had been able to receive a significant rental income from leasing premises to migrant workers.   A direct consequence of these shareholding reforms was that villagers needed to comply with the laws, regulations and state policies at both national and local levels in return for their share dividends and voting rights. Therefore, villagers were indeed exchanging their compliant behaviour for their rights to share in the management and distribution of collective assets. To enforce this contract, however, required the GLD government to exercise an increasingly restrictive level of control over village cadres, who were acting in the front-lines to monitor and evaluate the villager?s compliance.  In what follows, I will explain how the local government has succeeded in effectively restricting the activities of local village cadres in the shareholding reforms.  132  Shareholding Reforms and Governing Capacity Building     Building a new performance appraisal system to motivate and restrain village cadres was another major endeavour of the local government in the early 2000s. Unlike local officials, village cadres were not part of the state bureaucracy. They were neither paid a salary by the government nor entitled to receive the pension guaranteed to all civil servants upon retirement. As a result of the shareholding reforms starting from 2003, although village cadres were still not salaried officials, the local state set a policy that divided the remuneration of each village cadres into two parts. One was the basic subsidies provided by the government. Another was a performance bonus provided by the village collective. The bonus payment to village cadres was usually equivalent to two to five times the average dividends allocated to each shareholder, depending on not only the profitability of their shareholding cooperatives, but also their actual performance in community management. As a village party secretary explained, he received a monthly subsidy of about 4,100 yuan from the district government. Over the past year, he and his colleagues in the Resident Committee and the shareholding cooperatives received the highest rating for their performance in the areas of birth control, social security and building a ?civilized neighbourhood.?  As a result, he and his colleagues received a handsome year-end bonus.  It seemed that the incentives for village cadres to perform effectively were derived from not only increased remuneration but also improved working conditions. One village cadre who was in his late fifties told me a remarkable story of the changes he had personally experienced since the 1970s. During the Maoist Period, he had served as the production team leader. Every month, he had to attend a working meeting that was more than two days journey away. Lacking transport and proper roads, he took to the mountain paths carrying enough food and water to last the journey, and a simple reed mat to sleep at a temple which marked the midway point. This was in the 1970s. In the early 1980s, the Household Responsibility System was starting to take hold, and the boost in agricultural productivity allowed him to save enough money to buy a bicycle. By this time, there was now a simple dirt path winding round the surrounding mountains so he could ride his bicycle to meeting. By the 1990s, the land compensation fees and leasing income afforded him a motorcycle. He paused for a moment and then gestured to a car across the street.  In the early 2000s, he had been granted a company car by the village collective. Now, he made the journey on a newly paved road and in the air-conditioned comfort of his official vehicle. As I have mentioned earlier, the duties of a village cadre such as effecting birth control policy, 133  enforcing building control, managing collective assets, were demanding and often placed them at odds with those they are ostensibly serving. Despite these hardships, this cadre spoke with gratitude of the role that he was fortunate enough to have held for more than 30 years. He had comforts and incentives that 30 years ago, he would never have dreamt could be possible for a man of his station.  He felt that this reward reaffirmed the pattern of service he had practiced, and he would continue to serve both government and the villagers with dedication.    Alongside these changes, a new mechanism for checks and balances between the collective and individual villagers began to emerge. On the one hand, local cadres were monitored by their shareholders because they now needed to act for the welfare and interests of all shareholders if they wanted to take up a supervisory position in the collective. On the other hand, local cadres needed to adhere to the local state?s supervisions and instructions, given that the remuneration and promotion systems of senior management staff (such as members of the Board of Directors) were subject to stringent regulations by the Street Office. These relations highlighted the intermediary role of village cadres as a link between local state and villagers, allowing the local government to secure the involvement of local village cadres and villagers in the implementation of policy and to originate solutions that were based on compromise and cooperation.  After the establishment of a new appraisal system for village cadres, the District government moved to develop a new accounting system in order to improve the transparency and accountability in the management of collective assets. To this end, the municipal government put an emphasis on promoting a number of principles in the early 2000s, including: autonomous operation (zizhu jingying), independent accounting (duli hesuan), self-finance (zifu yingkui), and democratic management (mingzhu guanli). In line with these principles, the district government first required all village collectives to set up separate financial revenue and expenditure accounts for the Residents? Committee and the Association of Shareholding Cooperatives respectively. Before this separation exercise took place, the financial matters of these two organizations in many villages during the 1990s had been collapsed together into a single account. This ambiguity allowed some village cadres to abuse the use and allocation of land compensation fees by administrative fiat. To monitor the operation of these separate accounts, the district government further implemented a two-tier accounting system to reduce the opportunity of misconduct in financial management. The first tier was shezhang cunguan which meant that all the accounts of the shareholding cooperatives had to be audited by the Residents? Committee. 134  The second tier was cunzhang zhenguan, which meant that all the accounts of the Residents? Committee had to be audited by the Street Office. Moreover, both monthly and quarterly auditor reports had to be made public, allowing villagers to inspect the actual amount of expenses and transactions involved (see Figure 2.9). Through these reforms, the local government was able to impose its direct control over the management of village financial affairs. One can expect that these so-called spontaneous village collective organizations, which are supposed to be the self-governing units of the villager, can never remain wholly autonomous from the local political institutions and the supervision by the local state.  To conclude, shareholding reforms in the GLD have been primarily a state-led process through which the local government has made ongoing attempts to consolidate its power in the management of collective assets with the aim of ensuring its control over the potential social unrest arising from rapid urbanization. As a result of shareholding reforms, villagers were empowered in the management and distribution of the collective income. The promotion of a shareholding system to manage the collective land/assets and the subsequent arrangement of share/dividend distributions among eligible villagers provided a better delineation of ownership and revenue rights for individual villagers in these collective assets. These institutions encouraged villagers to support continued urbanization, which promised to enhance the values and revenues of their collective land holdings.   At the same time, shareholding reforms followed a highly restrictive agenda, because the local government saw that exercising control over village cadres was a necessary intermediary step towards achieving the accountability of village cadres in the management of collective assets and village affairs. As a result, local cadres can now enjoy better remuneration and the legitimacy derived from formal elections, but at the expense of more constraints imposed by not only the prescribed rules and procedures of the Articles of Association, but also the direct supervision of the district government. In the policy paper ?Opinions on Improving Rural Shareholding Cooperatives? which was issued by the GLD government in 2003, it was officially announced that henceforth all shareholding cooperatives were placed under the administrative purview of the district government. In addition, they are required to accept guidance from whichever the Party Committee of their respective village. Such a hierarchical structure allows the local government to secure local village cadres? and villagers? participation in policy matters and to find appropriate solutions to problems through on negotiation and cooperation. This has 135  been one of the important factors explaining why the enduring problems and conflicts have not yet evolved into severe confrontations and clashes between villagers and the local state. Alongside the shareholding reforms, the local state of the GLD has also adjusted its own interventions in the election process. How it has managed to win support from villagers while strengthening its control over the urbanizing villages through grassroots elections will be the focus of the next chapter.    136  Chapter 6       Grassroots Elections and Social Participation      Most literature suggests that grassroots elections implemented by Chinese authorities at the village level during the 1980s were held for the purposes of warding off political unrest, reinvigorating self-governance and halting further deterioration of village cadre-mass relations. However, to what extent these grassroots elections had led China to democracy remains highly controversial in the current scholarly debates. Some scholars have focused on the relationship between village elections and the level of economic development. For example, Oi and Rozelle (1996) and Hu (2005) suggest that competitive elections are more likely to occur in poor villages as the enthusiasm of villagers for political participation in wealthier villages tends to be lower. O?Brien (1996), however, holds a different view, arguing that it is easier to achieve democratic elections in rich villages which run profitable collective enterprises. Other scholars highlight the problems of the electoral process and outcome, arguing that grassroots elections hold little hope for democratic development in rural China. Judging from the Chinese official debates on village elections, Kelliher (1997) argues that the ?feigned compliance? of local officials in promoting self-governance created a significant obstacle for the development of village democracy. Similarly, Zhong and Chen (2002) suggest that local cadres? manipulation impedes real free elections in rural villages. Yao (2009) argues that, while some village cadres did not have the ability to manipulate elections, free elections still failed to ensure fairness in the distribution of village collective assets. These contrasting findings illustrate that the actual impacts of grassroots elections may vary from time to time and place to place, and suggest that a micro approach to understand the complexity of elections in the local context is necessary.    The key to understanding the interplay between grassroots elections and local governance in the GLD, I shall argue, has to include at least three factors which have generally been overlooked by the existing literature. The first is about the local state?s conflicting strategies for promoting elections in urbanizing villages. There are three types of grassroots elections in the GLD: the election of the Party Committee, the election of the Residents?/Villager?s Committee members, 137  and the election of the shareholding cooperative. In fact, the GLD government has adopted different strategies for different types of election, by which it not only exercises political control over the community but also cultivates rapport in the eyes of villagers. The second factor is related to the impact of kinship and family ties on elections. In the urbanizing villages, what influences villagers? participation in grassroots elections are village norms and traditions, in addition to the imposed state policy. The third factor refers to the participation of migrant workers, who have been playing an increasingly important role in local governance but are still for the most part excluded from neighbourhood elections.  My aim in this chapter is to better comprehend the role of elections in local power dynamics through an analysis of how multiple actors have engaged with elections and other institutions to reshape the landscapes of local governance. Land requisitions have equipped the GLD villagers with wealth and a high degree of autonomy in financial and social management. I will focus on two main questions. First, in what ways has the local government maintained its political grasp on an increasingly autonomous rural community? Second, how and to what extent did migrant workers, given their very limited participation in formal decision making processes, influence local governance practices and outcomes?   The Reach of the State under ?Two Committees, One Association?     As shown in Figure 6.1, most villages in the GLD adopted a governance institution called ?Two Committees, One Association (liangwei yishe).?  The two committees in question are the Party Committee and the Residents? (Villagers?) Committee, which were the common institutions in rural China. One association, namely the Association of Shareholding Cooperatives, exists in most villages of the Pearl River Delta region, in which the policy of ?reserved commercial land? has been implemented. Under this organizational structure, there are three types of elections, including the election of the Party Committee, the election of the Villagers? Committee, and the election of the shareholding cooperative leadership. These elections are held every three years and together configure the institutional arenas for the exercise of political power at the lowest level of governance.      138  Figure 6.120 ?Two Committees, one Association? (Liangwei yishe)         Data source: field research         Table 6.19 Neighbourhood Size and Party Organization  No. of Party Members Party Committee  CCP Committee Members      Party Secretary  Deputy Party Secretary 7 ? 50 3 1 1 50 ? 100 3 - 5 1 1 - 2 > 100 5 - 7 1 2 - 3           Date Source: field research        Table 6.210  Neighborhood Size and the Resident Committee   No. of Households Residents? Committee   No. of Professional Staff (through open recruitment) RC Committee Members Director   Deputy Director  < 1000 5 1 1 1 1000 - 2000 7 1 1 - 2  2 > 2000 9 1 2 - 3 2          Date Source: field research  Residents? Committee Party Committee Party  Branches  Association of Shareholding Cooperatives Party  Groups  Shareholding Cooperatives Administrative Offices (Public Security, Mediation, Birth Control, Medical Services, Housing Management and Accounting) 139  In each round of grassroots elections, the election of the Party Committee was conducted before the elections of the Residents? Committee members and shareholding cooperative directors. The election of the Party Committee in the GLD was a process of what village cadres called ?two nominations and one election (liangtui yisuan).? ?Two nominations? meant that candidates were nominated in two ways: internal nomination by party members (dangnei tuijian) and external nomination by villagers? representatives (qunzhong tuijian). The number of candidates was typically greater than the number of elected positions by 20 percent. The party branch first convened all party members in the neighbourhood/village to nominate candidates. The nomination was valid as long as it was supported by two-thirds of all party members. At the same time, candidates were also nominated by a committee composed of non-party members from villagers? representatives, heads of Residents? Groups and former members of the Villagers?/Residents? Committees. The list of nominees together with a description of their experiences and qualifications was then submitted to the Party Branch at the Street Office for approval. After finalizing the candidates, the party branch called a general meeting to elect the secretary and the members of the Party Committee. This was what they call ?one election? after ?two nominations?. During the election meeting, voters? choices were anonymous. The relevant policy document emphasized that the attendance at the election meeting had to be over 80 percent of the party members who had registered their memberships with the neighbourhood. However, it did not specify the rules for determining the winning candidates, even though candidates with the highest number of votes were mostly declared elected. This left some room for the party organization at the Street level to have a final say, when the results of the election meeting were submitted for final approval and endorsement.   By these procedures, the party organization at government level (i.e. the Street level) could influence not just the nomination of candidates. It also retained a great deal of control over determining which nominees actually appeared on the winning list. There were obvious reasons why the GLD government focused its intervention on the election of the Party Committee. As shown in Figure 6.1, the Party Committee was indeed the highest level of authority in neighbourhood governance. As the head of this highest level of authority at the neighbourhood level, the Secretary of the Party Committee also acted as the Director of the Association of the Shareholding Cooperatives. He or she therefore had the authority not only to handle community affairs but also to manage all collective assets held under the name of shareholding cooperatives. Given this power structure, the Residents? Committee and the shareholding cooperatives all 140  remained under the domination and direct control of the Party Committee. This also means that the state?s dominance over the Party Committee was crucial in securing its arm?s length control over the day-to-day operation of the villages.  In other words, if the state could gain control over the Party Committee, its political influence over local administration and the management of collective assets could be ring-fenced.   In the election of Party Committee, the local government wanted to ensure that the elected party secretary and the committee members would meet two essential requirements. First, they must be loyal to the party-state and devoted to the implementation of all policies and duties assigned by the state. Second, they must have a high degree of political credibility and popularity among villagers. Undoubtedly, loyalty to the Chinese Communist Party is the top priority in all considerations. However, the reputation of party secretary and the committee members in the eyes of villagers was no less important. Without popular support, the secretary and the committee members were unlikely to be able to effectively implement state policy at the neighbourhood level. To enhance the legitimacy of the party in the eyes of villagers, the party had to appear to be open to their opinions. Thus, it actively recruited new, capable villagers from the neighbourhoods to become party members. As revealed by the official statistics, the number of local party members at and below the street level increased from 4,665 to 5,150 between 2007 and 2008 (see Table 6.3).  Table 6.311  The GLD Party Organizations (at and below the Street Level)  Street Office No. of Party Committee No. of Party Headquarter No. of Party Branches No. of Party Members (2007)                  (2008) Xiagang 1 1 16 319 342 East Section 4 3 38 583 633 Luogang 2 3 24 863 893 Lianhe 2 1 21 534 628 Yonghe 1 0 8 384 482 Jiulong Town 1 5 41 1982 2172 Total 11 13 148 4665 5150   Data Source: GLD Yearbook 2007 and 2008    The state?s deliberate attempts to dominate the election of the Party Committee and to expand the party organization at the neighbourhood level, however, should not lead us to simply 141  conclude that other grassroots elections in the GLD villages were equally manipulated by the state. Quite the opposite is the case: it allowed villagers to decide on the operations of two other elections. In contrast to its firm grip on the electoral process for Party Committees, the local government adopted a different strategy to deal with elections of the Residents? Committee and the shareholding cooperatives. In the following section, I will explore how the local government has tried to enhance its political image and legitimacy through promoting transparency and giving a free hand in these elections.   Transparency and Autonomy in Neighbourhood Elections   Village grassroots elections in the GLD began in 1988 when the Chinese government introduced the Organic Law Regarding the Organization of Villagers? Committees to formally establish Villagers? Committees as the mass organizations of self-government. Villagers held direct election every three years for the Villagers? Committee. They selected committee members to fill positions in the areas of accounting, public security, birth control, conflict mediation, and conscriptions. As mentioned in Chapter 2, all villages under this study reorganized their Villagers? Committees as Residents? Committees in the early 2000s. These reorganization exercises did not bring fundamental changes to the procedures of grassroots elections, but allowed the villagers more flexibility in selecting the method of election. According to the Organic Law on the Organization of Residents? Committees, elections of Residents? Committee are carried out through direct election by all villagers (now residents) or election by villagers? representatives. In the case of direct election, the quorum was set at 50 percent and the candidate had to win a majority vote in order to win the seat. In the case of election by villagers? representatives, the quorum was set at a higher level: at least two-thirds of the representatives were required to attend the election and the winning candidate had to win more than half of the votes.   Which method should be adopted was the decision of the Electoral Affairs Commission, which was composed of the party secretary, the existing committee members and the elected villagers? representatives. If the Electoral Affairs Commission  decided to continue with the method of direct election, the members of the Residents? Committee would be elected by all registered villagers aged 18 or above. If the Electoral Affairs Commission opted for indirect election, 142  committee members would not be elected directly by all qualified villagers who had the right to vote. Instead, they were elected indirectly by voters? representatives who won the elections at the Residents? Group level. In recent years, while some villages still continued with direct elections, some had switched to adopt indirect elections by residents? representatives. Interviews with local village cadres and villagers suggested that several considerations had influenced their choice of electoral method.   First, indirect elections involved less complex procedures and thus demanded lower costs in administration. As a major part of the expenditures for elections were financed by the collective income and resources, villagers therefore tended to support a method which could help lower administrative costs. Second, the election method also affected the decision-making process in day-to-day community management. This is because the same election rules would be used in determining the collection of fees, the allocation of collective incomes, the construction of a road or other public projects, or the provision of welfare programs in the villages. Therefore, indirect elections generally met with more welcome from villagers than direct elections, because the latter was considered too cumbersome and time-consuming in the decision making process.    However, indirect elections had a major shortcoming: it made it easier for some powerful clans to dominate elections. Under indirect election, villagers? representatives were first elected to the Residents? Group, in which every 20 or 30 households selected one villager to represent them. These elected representatives then went for the election of the Residents? Committee on behalf of their voters.  The total number of households in a GLD village usually ranged from 400 to 800. If indirect election was adopted, only about 20-50 representatives would be selected for the final election of the Residents Committee. In this case, candidates could easily bribe the villagers? representatives or influence their decisions in the process of election. By comparison, in the case of direct election, it was difficult for candidates to influence the outcome of elections as they could not possibly influence hundreds of villagers who were entitled to vote. Given the pros and cons of these methods, it is found that villages in which village cadres and villagers have established a high degree of mutual trust tended to adopt indirect election. In contrast, those villages where tensions between village cadres and villagers have long existed continued to adopt direct election for selecting members of the Residents? Committee. For example, one village initially had an electoral system based on implicit trust between cadres and villagers.  However, it came to light that village cadres had been abusing the collective income for private 143  use through personal embezzlement. In response, the villagers not only set fire to the cadre offices, but reverted to a system of one-man-one-vote direct election.    Regardless of whether direct or indirect election was selected, they were subject to standardized government procedures. In 2003, the GLD government introduced a series of policy directives to guide elections of the Residents? Committee and the shareholding cooperative. Under these policy instructions, grassroots elections in all neighbourhoods have generally involved five major steps. The first was to form an Electoral Affairs Commission. Afterward, the Electoral Affairs Commission worked on the registration of qualified voters. It needed to prepare a list of registered voters available for villagers? inspection at least 20 days before the formal election. The third was the nomination of candidates by villagers to stand for elections. Under the multi-candidate electoral system, two candidates are usually nominated to compete for the position of director or deputy director. For the election of committee members, the number of candidates is usually two or three more than the number of positions. The fourth step was the selection of three to seven poll watchers from villagers? representatives to be observers of the election and responsible for monitoring the entire voting process and scrutinizing ballot-counting operations. The final step was the process of voting itself by all qualified voters. All ballots had to be counted and the results of the elections were declared within the same day.   Through this process, the local state could no longer manipulate the nomination of candidates. Although the Party Secretary usually continued to be one of the members in the Electoral Affairs Commission, he was not able to control the nominations directly because the opinions of other members from villagers? representatives were also decisive in the committee. Moreover, the election results did not require the approval or endorsement from the Street Office. In fact the Street Office? involvement was limited to recording the results only. Some local officials from the Street Office suggested that they deliberately were absent from the election stations in order to avoid interfering villagers in the process of election. This strategy, as suggested by local officials, allowed the district government to build up its reputation for impartiality. In turn, they were increasingly seen as credible judges in dealing with the election disputes that might arise between village cadres and villagers.   144  In the views of villagers, they felt that these elections were more ?democratic? than before. One reason was that villagers could recommend and vote for candidates in the Residents? Committee elections. Moreover, during the final election, villagers could either vote for the candidates already nominated on the list, or they could write directly on the ballot additional candidates whom they supported. Another reason was related to the actual performance of village cadres in community management. For the villagers, democracy was a high-sounding principle, and had to be backed by material results. Whether something is democratic or not, is not solely related to the fairness of the electoral process and procedure. It is also judged from its outcomes in the daily management of the community. Some villagers observed that the elected cadres were more dedicated to improving the welfare facilities of villagers. In a village that I visited, some villagers told me that they were satisfied with the elected members of the existing Residents? Committee as the village collective had just renovated a building for a neighbourhood clinic so that villagers could have more convenient and better medical services.  Kinship was another important factor influencing villagers? decisions in elections, which in turn shaped the democratic development in villages. To be sure, a villager would first pay attention to the candidate?s credibility, leadership and administrative ability, particularly his integrity and past performance in the management of collective assets. This, however, does not mean that kinship played little role in their voting decision. In general, villagers still favoured a candidate from their own lineage. Descent groups considered that it was essential to have one of their members to represent them in the management of the community. A village cadre indicated that he devoted his best effort to win because his loss in the election would shame not only himself but also the entire lineage. In this context, a candidate?s appeal to voters was usually not only because of his personal ability but also the traditional notions about villagers? desire to sustain the power of their clan in the village community. These clan-based effects on grassroots elections are evidenced in the fact that elected cadres in two committees and on associations were mostly the members from the big clan families in the neighbourhood.  Grassroots Democracy     Land requisitions led villagers to become more economically dependent on the collective, providing economic incentives for villagers to actively participate in grassroots elections with the aim of ensuring proper management and fair distribution of their collective assets. In their 145  active participation in formal elections, villagers have come to combine their economic demands with political appeals in their struggles over the challenges arising from land requisitions for more and more urbanization. However, whether these small-scale initiatives can build over time into transformative changes in grassroots democracy remains uncertain.   One possible obstacle is that villagers? participations have been mainly motivated by economic considerations, leading villagers to be highly pragmatic about their short-term gains from grassroots elections. An example of this point was their different modes of participation in different elections. As some officials from the Street Office described, during the elections of the Resident Committee and Shareholding Cooperative, most villagers not only came to cast the votes but also waited in the polling place to hear the announcement of the results of the election. Villager cadres thus were cautious about each procedure in order to survive villagers? scrutiny and the government?s cross-examination. In the election of the Representatives of People?s Congress, villagers were enthusiastic about participating too. However, few of them stayed after casting their votes. Some village cadres explained that most villagers just came to vote because they received 50 yuan of ?Wugongfei.?33 Quite frankly, some villagers also indicated that this was the main incentive for them to participate in the election of the People?s Congress; they assumed that the election outcome of the Congress Representatives would not directly affect their financial position.  Another possible obstacle relates to the clan-based power bloc. Taking their roots in a village society firmly established since the Song dynasty, lineage networks have remained paramount in many aspects of village life for generations in the GLD villages. With the penetration of the powerful party-state into rural politics, some lineage networks and their powers atrophied during the Maoist era. They have revived in the reform period, however, especially after the introduction of the ?reserved commercial land? which made villages financially independent, self-governing entities. As a result, when involved in elections, the political behaviour of villagers was affected by not only the external, formal policies, but also the informal institutions, such as kinships ties, family power relationships, gender roles and so forth. Although villagers                                                           33 In most GLD neighbourhoods, it has been the common practice that each villager received 50 yuan after casting their vote. Such a payment was called ?Wugongfei? by local people. ?Wugongfei? also paid for voting for the elections of Residents? Committees. It primary purpose was to provide an economic incentive for villagers to vote in elections.  146  have been provided with economic incentives to participate in elections, the influence of lineage networks and their power has remained strong.    The influence of clan-based power has become a major concern of the GLD government. One measure of the local government to reduce the political influence of big clans on elections was through its direct intervention in Party Committee elections. Through involvement in the nomination procedures, the state could eliminate those candidates which were incapable of balancing the interests of their own clan family and those of others. Another important strategy that the GLD government introduced in 2005 was to openly recruit young degree holders and then assigned them to work in the Residents? Committee. The primary purpose of this program was to train these young talents to participate in grassroots elections so as to undermine the influence of traditional networks on local governance. However, this plan failed to achieve its objective, given that only one of them could successfully become an elected member of the Residents? Committee over the past decade. This result suggests that villagers were still reluctant to accept a non-indigenous leader to direct community affairs.   Exclusion of Migrant Workers   While villagers were showing their unprecedented enthusiasm for electing community leaders, migrant workers lacked access to both shareholding cooperative elections and Resident Committee elections. They were not eligible for the shareholding cooperative elections as they were not shareholders. In addition, they were not qualified to register for the elections of the Residents? Committee. Under the Organic Law of Urban Residents? Committee, all committee members are selected by local residents who have registered their hukou with the neighbourhood. According to the policy paper ?Some Suggestions on the Election for Residents? Committee? issued by the GLD government in 2005, the residents who did not register their hukou with the neighbourhood could vote in elections only if they fulfill two requirements: proven ownership of real estate in the neighbourhood and residency in the neighbourhood for more than half a year. However, the fact is villagers? houses were not allowed to be legally sold on the open market, and this policy thus implicitly barred migrant workers from participation in elections.    147  To understand why village houses were not transferable, it is necessary to trace the policy back to its origin in the collectivization movement during the Maoist era. After the establishment of the People?s Commune system in the late 1950s, all farmland in rural China was collectivized, but farmers? housing was not. This meant that farmers always owned the houses in which they lived. However, farmers were prohibited from selling their houses and were legally restricted from using their village housing as collateral for loans. The primary intent of such restrictions was to keep farmers in the countryside to ensure steady production of what the state needed for socialist industrialization (Li, 2004; Wen, 2005). These restrictions have remained in effect during the reform period. The official hope was that the countryside should continue to serve a function of what Friedman (2012) called a ?safety valve,? which would allow rural farmers to secure a subsistence living in case they lost their jobs in cities. If land and houses were allowed for mortgage financing, farmers would incur the risks of becoming homeless and dispossessed in the event they defaulted on their loans. In the absence of a well-established state welfare system for rural farmers, the government believed that the potential population of homeless farmers would ultimately endanger social stability.   Given these considerations, the local government continued to enforce the restrictions on the sale of villagers? houses, despite that more and more villages in the GLD have been officially recognized as urban neighbourhoods. As a result, it was nearly impossible for a migrant worker to own real estate in urbanizing villages. Moreover, without owning real property in the neighbourhood, a migrant worker could not qualify for the vote unless they could accomplish the near impossible feat of securing a hukou as an outsider and registering it within the neighbourhood. For traditionally, unless one was born in or married into the village, the issuing of a new hukou was exceptionally rare. Indeed, it became much more difficult for an outsider to register his or her hukou within the neighbourhoods after the implementation of the shareholding system. The issuing of a new hukou to an applicant such as a migrant worker would change the number of shares held by the existing owners. This change might eventually decrease everyone?s share of voting power and dividends. Naturally, in the hopes of guarding villagers? interests against outsiders, the registration of a new hukou with the neighbourhood became increasingly scarce as a matter of self-preservation.    Despite the fact that migrant workers were barred from neighbourhood elections and in turn lacked access to the village?s formal decision making process, one should not simply conclude 148  that they were powerless. In fact, quite the opposite was true. Over the past decade, migrant workers have been playing an increasingly important role in local governance arrangements. As I have mentioned in Chapter 2, migrant workers were concentrated in those villages where most of their land had been claimed by urbanization. In these wealthy villages, the number of migrant-workers was greater than that of indigenous villagers. Migrant workers leased accommodation from villagers in the neighbourhoods, enabling the indigenous population to generate substantial rental income. In addition to contributing to the personal income of villagers, migrant workers played a crucial role in providing community services for villagers. Some migrant workers operated restaurants and groceries shops in the urbanizing villages. Some worked in the small companies owned by villagers. A migrant worker from Hunan told me that she had been working as a manager in a hostel owned by a villager for more than five years. She not only was responsible for the daily operation of the hostel, but also acted as an agent to help villagers look for part-time jobs as domestic helpers in migrant workers? families. Slowly but surely, the arrival of migrant workers forced villagers, who had been accustomed to living in a traditional and exclusive village society, to build a more inclusive community in their day-to-day interactions with non-indigenous residents.  In this study, I aim to understand how migrant workers, who were excluded from grassroots elections, have been able to influence the process of local governance. By doing so, I take a departure from the ?landlord and tenant? relationships to understand how migrant workers have contested with villagers to shape their existing interactions in the neighbourhood.   Landlord Villagers versus Tenant Migrant Workers   Interviews with villagers reveal that their acute conflicts with migrant workers took place in the 1990s, when the rapid expansion of the GDD was attracting many migrant workers to search for job opportunities in the area. As discussed in Chapter 3, the first batch of migrant workers who were farmers from Guangxi came to help villagers in agricultural production in the 1980s. While farming on the land leased from the villagers, these non-local farmers built their shacks near the fields. They usually did not demand any local resources. During the 1990s when more and more arable land was requisitioned by the government for industrial development, most of these non-local farmers gradually returned to their home villages or switched to work in nearby factories. At the same time, more and more new migrant workers from different parts of China moved to 149  the GLD. Some migrant workers were living in dormitories provided by the factory owners. Some of them rented housing from the villagers. The arrival of migrant workers, on the one hand, brought rental incomes to villagers. On the other hand, it created numerous tensions with the local society. Recalling their experiences in the early 1990s, villagers admitted that they felt upset about the expanding population of migrant workers in the village and tended to view migrant workers as ?dangerous intruders.? In the GLD villages, lineage-based social networks rooted in imperial China had firmly established a society of acquaintance, under which villagers tended to view all those who were not descendants from the line of the ancient families within the villages, as outsiders. The villagers? exclusionary attitude, together with the communication problems caused by language barriers and ethno-cultural differences, at times triggered heated, violent clashes between villagers and migrant-workers.   A village cadre told me that his daily work at the time was spent with dealing with the disputes between villagers and migrant workers. Villagers fought with migrant workers just because they could not stand the odor of spicy food spreading out of the homes of migrant workers. Another source of conflict between villagers and migrant workers was related to tenancy contracts, which were usually oral rather than written agreements. During the late 1980s and the early 1990s, many migrant workers were short-term tenants. They usually chose to break the contracts when they needed to move elsewhere for better jobs. This caused much displeasure among villagers, leading to major disputes. In some extreme cases, villagers were ?kidnapped? when they went to argue with migrant workers who refused to make rental payment. In these situations, migrant workers were seen as creating not only hygiene problems but also posing social security threat to the community. Many village collectives, therefore, organized teams of security guards to maintain local order.   In addition, villagers complained that migrant workers were consuming portions of village resources which did not belong to them. As mentioned in Chapter 4, the village collective held more than 70 percent of villagers? land compensation fees and managed their ?reserved commercial land.? All the basic facilities, such as village roads, community clinics and creational centers, were funded by the village collectives, and the money actually came from villagers? land compensation fees and the incomes of their ?reserved commercial land.?  From villagers? perspective, these village facilities and services were not public property, but their private holdings which had been built with the sacrifice of their land for urbanization. Villagers, 150  therefore, did not agree that it was obligatory for them to share them with migrant workers.  To recover their costs, some village collectives imposed various levies on migrant workers. As a village cadre described, his village charged a so-called monthly management fee of four yuan on each migrant worker. Such a fee put a heavy financial burden on most migrant workers who could only earn an income of 50 to 80 yuan per month at that time. Lacking knowledge and confidence to redress their hardships through proper legal action, many migrant workers chose to express their angers to the villagers. All these problems only served to aggravate the confrontation between villagers and migrant workers. Inclusion of Migrant Workers   According to village informants, these conflicts were significantly reduced in the early 2000s. The main reason for this, I would suggest, is that the rapidly increasing rental income generated from leasing property provide incentives for villagers to seek improvement in their relationships with migrant workers. As mentioned in Chapter 2, alongside the industrial expansion of the GDD, the population of migrant workers doubled that of indigenous villagers in some villages and the rental value of villagers? house saw a significant increase. With the additional private income derived from property leasing, villagers were able to enjoy a reasonably good quality of life. For example, they could spend more on travelling for leisure. Many villagers valued such opportunities, as travel had been a rare luxury during the 1980s and 1990s. In their words, ?travelling to the city (chucheng)? was a significant life event at the time, as most of them were given very limited opportunities to leave their villages during the Maoist era. During the 1980s and 1990s, the government gradually relaxed restrictions on their movement between their village and the city. However, they still usually only travel to Guangzhou City or other cities within the Pearl River Delta region once a month to visit their relatives, or to shop. Before the land requisitions, fruit selling had been their principal means of income generation. Being preoccupied by the farming work for their orchards, leisure time was still hard to come by.    In recent years, although few villagers have travelled abroad so far, most of them have travelled many times to other Chinese cities, such as Beijing, Hangzhou, Shanghai, and so on. After learning of my Hong Kong background, many villagers were particularly interested in sharing with me their travel experiences to Hong Kong. A village cadre told me that he had had many opportunities to travel to the city for business during the 1990s because their village collective 151  had been selected as a franchising agent to supply livestock and poultry to their Hong Kong wholesalers. At that time, all villagers were jealous of his work as travelling to Hong Kong was for many only a dream. Nowadays, villagers no longer envied him as many of them traveled to Hong Kong a number of times over the past years. One villager described to me her experience as follows, ?I have a cousin who migrated to Hong Kong 20 years ago. Two weeks ago, I just went to see him and his family. He was working as an Assistant Manager in a Chinese restaurant. Working for more than 10 hours everyday, he earned about HK$16,000 each month, with which he could only afford to rent a very small apartment with a floor area of about 300 square feet for his family (including his wife and two sons). Compared with my cousin, I felt fortunate that my family is living in a house with total area of 170 square meters and we can receive stable dividend income from the village collective and rental income from property leasing.? Another villager shared similar views, ?Hong Kong is a very beautiful and dynamic city. To be sure, I will visit it again in the near future. But I still prefer to live in our village. The government is constructing a subway. One of the stations will be located beside our village. Upon the completion of the project, it will take us less than 30 minutes to travel between our village and the urban core of Guangzhou City.?    Rental incomes generated from leasing housing to migrant workers also enabled villagers to provide better education opportunities for their children. Most middle-aged villagers in the GLD only had primary education or no schooling at all. Villagers did not want their next generation to remain poorly educated and thus were very willing to devote village resources to education. A case in point is the construction of kindergartens and primary schools in the villages. Due to financial constraints, these village schools were able to meet some basic schooling purposes but failed to ensure teaching and learning quality at a decent level. As a result, many village students could not progress for further study in good secondary schools in urban areas. In recent years, the district government has gradually taken over the management of these village schools at the cost of the village collective. This not only relieved the financial responsibility of the village collective, but also smoothed the progression of villagers? children to higher education. Moreover, with the continual growth in family income, more and more villagers can afford to send their children for tertiary education. A villager told me that she received about 3,000 yuan of rental income each month, which was enough to cover the tuition fees and living costs of a daughter enrolled in the social work program at the Guangzhou Zhongshan University.   152  Moreover, it is worth noting that migrant workers also made a significant contribution to the collective income of villagers because the tenants of those apartments built on villagers? ?reserved commercial land? were mainly migrant workers. Since villagers increasingly rely on income contributed by migrant workers, it is not surprising that the former have taken initiatives to improve their relationships with the latter. During the field research, I found that villagers had been increasingly willing to share with migrant workers the use of village facilities, such as clinics, primary schools, recreational facilities, and so on. In the community clinics, for example, villagers no longer insist on priority service for themselves over migrant workers. Instead, they have come to accept the rule of ?first come, first served? when waiting for medical treatments. This is a comfort to the migrant workers, although they were still paying a higher price than villagers for medical services. In many but not all village schools, scholarships which were originally set up for the children of indigenous villagers are also accepting applications from the children of migrant workers who are living in the same neighbourhood. Moreover, community recreational centers and school bus services, which were previously provided for the exclusive use of villagers, have become accessible to migrant workers as well. In handing local conflicts involving migrant workers, Residents? Committees now invited migrant workers to participate in their mediation. For example, local cadres are more willing to pay visits to migrant workers rather than ask them to meet at the office of the Residents? Committee during office hours. Moreover, the time and place for meetings are mostly scheduled to the convenience of the migrant workers. In some neighbourhoods, if migrant workers were required to come to the office to participate in dispute mediation, the Residents? Committee would pay them a small compensation for wages foregone in having attended the meeting. This interaction was common in the villages where villagers stood to gain enormous profits from leasing buildings to migrant workers.   Another important factor contributing to the change in the relationship between villagers and migrant workers is the industrial upgrade and restructuring of the GDD. During the 1980s and the 1990s, the GDD underwent a rapid industrialization process which was mainly driven by labour-intensive and export-oriented production. As revealed by the official statistics, light industries such as weaving, food processing, and production of aluminum cans for carbonated drinks accounted for nearly 80 percent of total local industrial output. From the late 1990s onward, the GDD has gradually adjusted its industrial structure through driving out labour-intensive and polluting factories to make way for a higher value-added form of production. In an 153  attempt to achieve this objective, the GDD Administrative Committee has gradually imposed more and more stringent environmental and labour protection laws in its development zones (Zhuo, 2005; Zhang, 2005). Moreover, the GDD Administrative Committee has launched the Science City project in order to build a new zone for the development of hi-tech industries. Alongside this industrial restructuring, more and more international companies, which specialize in producing liquid crystal displays, automobiles, telecommunication equipments and so on, have come to operate in the GDD. At the same time, some creative industries and research and development companies selected the GDD as a base for their research and business development. These new enterprises have generated new employment opportunities for migrant workers.   During my field research, the migrant workers I met in the urbanizing villages worked in a wide range of professions. They included not only factory workers, but also salesmen, technicians, software programmers as well as doctors and nurses working in the neighbourhood clinics. The higher income and job security of migrant workers enabled them to afford higher rents and commit to a longer period of tenancy agreement. Although many migrant workers were still uncertain about whether they would reside in the GLD permanently, they wanted to improve their living conditions in the neighbourhood. When villagers and migrant workers came to understand that they all have a stake in the neighbourhood, a sense of shared responsibilities for the community began to develop. All in all, the interests and concerns of the migrant population have gradually risen to become one of the foremost issues of village politics, despite the fact that they remain excluded from neighbourhood elections. At first, most migrant workers were employed in low-skilled occupations and composed the bulk of the labour in local industries. Their relative poverty and ignorance of the village rules and etiquette inspired much prejudice from prospective landlords and downright hostility from the rest of the village. Over time, however, the migrant worker community would grow and consolidate. Indeed, such was the eventual size of the migrant population that their collective output and consumption became a powerful determinant of the village?s economic health and prospects relative to the indigenous population.    154  The Increasing Influence of Migrant Workers  Some scholars have documented the marginalized living experiences by migrant workers with a particular focus on the underlying policy issues and the relationships between migrant workers and urban enterprises (Chai and Chai, 1997; Knight et al., 1999; Feng et al., 2002; Wong et al., 2007). Moreover recent studies suggest that migrant workers have been increasingly aware of their economic and social rights, enabling them to enhance their positions in society. The new generation of migrant workers came to pursue a dream of dignified and fulfilling life in the city (Friedman, 2012). They were more prepared to fight for better incomes. For example, after seeing that the company was making huge profits but was unresponsive to their requests for better pay, they had the courage to protest through strikes. Unlike the older generation of migrant workers, who had to endure abuse, discrimination and violation of their labour rights, the present generation of migrant workers not only developed a better awareness of social equality but also built up the capacity to safeguard their rights and interests (Shan, 2012). They have kept strong ties with each other and thus were able to mobilize workers on social action. Moreover, they knew how to turn to professional legal consultants for help and support (Chen, 2012). We could see this similar pattern of empowerment replicated in the GLD. With their growing contributions to the neighbourhood, migrant workers also demonstrated their increasing ability to exact policy reforms from the government.  During the late 1980s and the early 1990s, some migrant workers in the GDD spent long working hours in the factories. Some workers were forced to continuously work for 72 hours on the production line, and their movements, behaviour and even their ?moral conduct,? were tightly controlled by their employers, who could terminate their employment without any justification (Li, 2010). Interviews with villagers suggest that migrant workers in their villages mostly lived with them for a rather short period of time from several months to two years during the early 1990s. Villagers further explained that many migrants changed their jobs very frequently because most local enterprises offered low job security for their employees. Once they changed their jobs, they moved out of the village and looked for new accommodations which were closer to their working locations. Moreover, most of them planned to finally return to their hometowns after retirement. This older generation of migrant workers worked for their families in the countryside at the cost of their own quality of life. They maintained a minimal standard of living in the workplace but remitted a major portion of their incomes back to their 155  families in the countryside. Due to the huge differences in the costs of living between cities and the countryside, the remittances, though tiny by the urban standards, could still enable their families to substantially improve their well-being in their home villages.    This was indeed a microcosm of the experiences by Chinese migrant workers more generally in the 1980s and the 1990s. As discussed in Chapter 1, from the 1980s when the Chinese government has gradually relaxed its control over farmers? movement, more and more farmers have streamed into cities to look for jobs. Similar tides of rural-urban migration were observed in some Latin American countries such as Brazil and Mexico. In these counties, many rural farmers sold all their land and houses in the countryside in order to buy a ticket into the city. After arriving in the city, they could not afford to rent accommodation and thus were forced to build their shelters on land in or near the urban areas (see Turner, 1968 and 1969; Aguilar, 2008; Perlman, 2010). Unlike these urban squatters, migrant workers in China, however, did not give up their land and houses in their home villages, because the Chinese government continued its stringent control over the sale of rural housing and arable land in the open market. As a result, while squatters in Latin America strived for ?the power of the ballot box as leverage with which to bargain with urban politicians? (Hsing, 2010:143), most Chinese migrant workers of the 1980s and the 1990s were much more concerned about how to earn sufficient money, so that they could return to their home villages, where they still had homes, land and family (Cai, 2003; Wen, 2005).    Interviews with migrant workers suggest that the current generation works for realizing their own ambitions rather than working for their families in the countryside. A migrant worker, who was in his early twenties, told me that he did not have to remit too much money to support his family as his parents in Henan could earn fair income from growing and selling watermelons. Like many of his friends in the village, he came to the city for work as they did not want to stay in the villages like their parents, and expected their material and cultural needs would be better met in the urban areas. He had been working in the GLD for three years. However, after the completion of his current employment contract with an electronic company, he was planning to look for a job elsewhere; he had a feeling of isolation as the GLD provided very few places for entertainment after working hours. Another migrant worker from Sichuan told me that she just got a job offer in a factory in Chongqing. She decided to take up the new job as she wanted to get closer to her friends and parents who were living in the rural villages near Chengdu. Another 156  migrant worker and her family were planning to move out from the GLD because she hoped that her children would be able to have better schooling opportunities in the urban core after their study in elementary school.    From the late 1990s onward, with the industrial adjustment and upgrading of the GDD, migrant workers have more opportunities to work in new enterprises, which were more willing to improve working hours and attend to the issues of workers? working environment. These changes, however, did not alleviate the conflict and tensions between employees and employers. According to the official statistics, between 1995 and 2000, 25 labour strikes were recorded in the GDD. After a small decline in the subsequent five years, the number of strikes rebounded again between 2007 and 2010, to a total of 29 strikes recorded in GDD. Li (2011) investigated these strike cases and identified a significant change in the demand of migrant workers over the past three decades: migrant workers of the 1980s and the 1990s mainly protested against wage offences and unfavourable employment contracts that significantly violated their labour rights, whereas migrant workers in today?s GDD were fighting for higher pay and participation in company management.   Moreover, migrant workers in the GLD have increasingly participated in the election of the People?s Congress at the district level. Interviews with migrant workers suggest that neighbourhood elections were not their major concern for a number of reasons. First, they had their voting rights in their home villages or neighbourhoods. Second, they assumed that participation in grassroots elections would not make a change in their life, as the existing elections centred on the issues related to the management of villagers? collective assets. Last but not least, they were not certain about how long they would stay in the neighbourhood. When many migrant workers did not expect that they will stay permanently, it was not surprising that they were not eager to strive for access to grassroots elections. It seemed that migrant workers were more interested in participating in the elections of the People?s Congress Representatives at the district level. During my field research, a village cadre told me that a migrant worker living in his neighbourhood was just elected as the Representative in the forgoing district People?s Congress election. From the perspective of migrant workers, the elections of the People? Congress is perhaps more meaningful than grassroots elections. Unlike villagers, a migrant worker has no strong ties to the land in the GLD, his stay may be interrupted by a work stoppage and he would uproot himself if better opportunities arose. As a result, it is not surprising that the 157  migrant workers paid little attention to grassroots elections, which administered merely local concerns. Instead, they were interested in the elections of the People?s Congress, which determined matters (such as labour rights) of more immediate relevance to their lives.   The migrant workers? ongoing strive for empowerment has pressured the local state to readdress labour rights and to mount a direct response to the social needs of migrant workers. The GLD government has closely monitored the performance of local enterprises in dealing with labour rights. In this regard, a new centre was set up in the 1990s to receive and deal with complaints from migrant workers. Also, recognizing the importance of labour supply for local industrial development, the GLD government was prepared to provide affordable housing and better social services to retain migrant workers in the local industrial enterprises. Moreover, it was increasingly recognized that migrant workers, who constituted more than 50 percent of the total local population, will continue to be a major driving force for the social and economic development of the area. Interviews with officials suggest that the integration of migrant workers with the local community has risen in their policy priorities. How policy initiatives will be translated into practice is central to the ongoing economic and social transformation of GLD.  To recapitulate, what shaped the effect of grassroots election on local governance was a multiplicity of formal and informal institutions operating in the local communities, which had motivated and constrained not only the villagers but also migrant workers in their pursuit of personal gains and participation. Faced with the struggles of villagers and migrant workers for empowerment, it is not surprising that the local government also sought to strengthen its control over the urbanizing villages. However, it is worth noting that the local state has no longer relied on the conventional coercive approach to establish its power in villages. Despite its ongoing attempts to ensure its regulatory power over shareholding reforms and the election of the Party Committee, the GLD government has pursued conciliatory policies to gain greater legitimacy. This was done through promoting social welfare as well as transparency in the management of community affairs. This paradoxical reconfiguration of state power will become clear when I move to the discussion of local state building in the next chapter.       158  Chapter 7     Conclusions: Building the Local State      This chapter returns to the main research question I raised at the beginning of this dissertation: how has the local state managed to deal with land conflicts and the various problems arising from the rapid urbanization in reform China? The case of the GLD shows that the local government has constantly adjusted its strategies through combining both regulatory and conciliatory measures in dealing with land disputes and farmer resettlement in the process of urbanization. Drawing from the insights of the GLD experience, my discussion in this concluding chapter will focus on two issues: one concerns the specific conditions which have allowed the GLD to achieve its current governance outcomes. Another is the locally-specific experience of the GLD, specifically the implications this experience has for our understanding of local governance restructuring in urbanizing China.   Regulatory and Conciliatory Local State Building  Rapid urbanization in the GLD was primarily driven by local state-led industrial development made possible by the extensive requisition and conversion of farmland. These massive land requisitions have put the local GLD government in direct confrontation with the village collectives and villagers about issues of land compensation. However, the fact is that these land conflicts and disputes did not undermine the governance capacities of the local government. Two main factors explain how the local state was able to maintain its control over the restless urbanizing villages and mitigate conflicts arising from land requisitions for urban development.      First, one undeniable fact is that land requisitions for industrial and urban development enabled many villagers to make remarkable improvements in their living conditions. The villagers affected by land requisitions were able to earn stable and generous dividends from their shareholding cooperative, which derived their revenues from a combination of land compensation fees and rental income from the ?reserved commercial land.? At the same time, villagers could earn a handsome personal income from leasing their properties to migrant 159  workers in nearby enterprises. More importantly, the villagers, who under the urban-rural dualist system had had their rights to state welfare services withheld from them, were now given access to the educational system, medical services and pension schemes subsidized by the government. The GLD experience suggests that land requisitions for urban development, on the one hand, dismantled the basis of the indigenous social support system, by destroying their historical ties between the villagers and their land, with which they had long mingled their practical and spiritual well-being for more than half a century. On the other hand, land requisitions created a climate in which new income generation opportunities could flourish and in which local state officials were pressured to build a new social security system for the affected villagers, inducing villagers to give up their land for industrial development. The popular narratives in the media coverage of the time cast the villagers as a sort of passive dispossessed victim of land requisitions, but this cannot be further from the truth in the case of the GLD. One must understand that arable land yielded to the villagers very meagre profits at the cost of much toil under China?s urban-rural dualist system. The land requisition policy provided the villagers an opportunity to escape this way of life. Thus, contrary to the dominant narrative in existing studies, the villagers were amenable to the idea of land requisition. However, it was the compensation structure proposed by the government which would prove to be a major sticking point with the villagers. As I argue in Chapter 4, most villagers offered resistance not because they strove to keep their tiny plots of land for farming. Rather, they strove to achieve long held hopes of equal footing with urban residents and to share in the windfall of industrial developments that were engulfing their arable land.  Second, land requisitions for industrial and urban transformation in the GLD were not merely a process of ?land grabbing? by the local state for local development. They were also a process of local state building for the local government to strengthen its governing capacities and to enhance its legitimacy in the eyes of villagers. As discussed in Chapter 5 and Chapter 6, the local government of the GLD made ongoing attempts to establish its regulatory power over shareholding cooperatives and grassroots elections through the promotion of transparency and accountability in the management of collective assets and village affairs. If these measures presented a ?hard? approach to the mediation of conflicts arising from the rural-urban transition, the local government?s engagement with social welfare provision can be seen as a ?soft? strategy to regain its legitimacy and popularity amongst villagers. As mentioned in Chapter 2, from the early 2000s onwards, the GLD government started to get directly involved in the development of 160  education, medical care services and pension schemes for villagers. Although social welfare provision was tasked with improving the livelihood of villagers, it was also an effort on the part of the local state to strengthen its governing capacity so as to regain social control and legitimacy in the eyes of villagers. These new development trends led to the emergence of a new phenomenon of local governance which to date has not been adequately addressed in the current debates on Chinese local governance.    In the eyes of the Western observers, social welfare provision as a governance strategy may not seem a ground-breaking notion. However, under China?s urban-rural dualist system, the introduction of state administered welfare programs in villages surely represented a monumental break with the past. Under the urban-rural dualist system, social welfare in rural China was largely self-organized as farmers were excluded from the state social welfare system. In the 1990s, the income derived from land compensation fees and the ?reserved commercial land? enabled many villages to establish their own community clinics. However, at that time village medical services in the GLD were not integrated with the urban system. These clinics could only address the villagers? needs for the most basic health care services and were wholly inadequate for the treatment of serious ailments. Under such circumstances, it is not surprising that the state?s direct involvement in social welfare provision rekindled villagers? hope for rural reforms widely held to be long overdue. This not only eased villagers? resistance to land requisition, but also led them to increasingly accept the penetration of state power.   The GLD experience is an example of a local state managing the problems arising from the rural-urban transition through a deliberate engagement with village collectives and villagers in the management of collective assets and social welfare provision. The restructuring of the local government in the GLD embodies a shift of local governance strategy from the pursuit of industrial efficiency as the single, ultimate goal of urbanizing rural areas, to a comprehensive, yet pragmatic provision of public, employment and welfare services to the villagers. The mixture of both regulatory and conciliatory measures, I shall suggest, is the single most important factor explaining why land requisitions as swift and massive as they were, did not evolve into widespread clashes between the local state and villagers.   There is no question that the GLD experience is embedded in its local contexts. Strong local financial capabilities enabled the local state to undertake a greater role in social welfare matters. 161  During 2006 to 2008, the GLD recorded a significant increase in its tax revenues, both national and local (see Appendix I). The steady growth in tax revenue enabled the district government to provide direct subsidies in many aspects of social welfare development. Moreover, the GLD government combined the GDD Administrative Committee with the Luogang Township Government. The former was a manager of economic zones which had accumulated a rich experience in the management of business investments and the pursuit of administrative efficiency. The latter was a township government with extensive experience in the management of its local affairs. Combining these two entities has enabled the new district government to adopt a relatively simple, lean and efficient governmental structure,34 ideal for the adaptation and implementation of new policy according to local conditions.   But upon closer investigation, there existed one more critical factor. That is, the policy of ?reserved commercial land? has profoundly restructured the power balance between the state and village society. As discussed in Chapter 4, implementation of the ?reserved commercial land? policy led to the growth of autonomous villages and the ?retreat? of the local state. In this context, the stronger the village collectives grew, the weaker the local state became. This power restructuring drove the GLD government to carefully re-concentrate its power in urbanizing villages. As the villagers strengthened their collective power in dealing with the state in the matters concerning land requisitions, villagers developed an increasing awareness of not only their economic rights but also of equality in their social and political relations. In this context, the GLD government had to use pacifying rather than coercive measures. In hope of regaining legitimacy in the eyes of villagers, it got directly involved in social welfare provisioning as well as in mediating the conflicts between village organizations and villagers in collective asset management.  Central-Local State Relationships and State-Society Separation in Reform China  The locally-specific governance processes and outcomes in the GLD, which were different from those in other parts of periurban China, suggest that fiscal decentralization in the reform period did not produce a uniform pattern of local governmental participation in socio-economic                                                           34 Interviews with local officials suggest that, compared with other district governments in Guangzhou City, the total numbers of departments and civil servants of the GLD are less by about 33 percent and 50 percent respectively.    162  development. Some scholars viewed the disparate governance outcomes at the local level as the emergence of a fragmented bureaucracy, which represented the declining power of the central state over local governments (Lampton, 1987; Lieberthal, 1992). When I first began my research in the GLD, I held this assumption too. I assumed that the GLD had a relatively liberal governance culture because having grown out of a development zone, it had frequent and direct contact with foreign investors and enjoyed a high degree of autonomy in managing its local economic development. However, during my field research, two emerging phenomena inspired me to rethink the relationships between the GLD government and the central state.   First, as discussed in Chapter 4, many construction projects in the Knowledge City were suspended due to the lack of planning permits from higher level governments. Apparently, the land control policy from the central government was still decisive on local development. The increasingly strong control of the central state over the conversion of arable land into construction land at the localities not only affected the existing development projects in the GLD but also re-directed its long-term future. In recent years, the district government was forced to decelerate the pace of land requisition.  Recall that the district government needed to ensure building permits before it could develop requisitioned land. Without them, the government would have been forced to re-direct its efforts in developing the existing built-up areas of the region. As a consequence, the district government has recently switched to encourage village redevelopment. When the central government still retained such powerful tools to shape local destinies, it is hard to believe that its power to supervise its local agents has been substantially undermined by fiscal and administrative decentralization.   Second, the impact of the 18th Congress of the CCP at the top quickly rippled down to the GLD government. Between the 8th and 14th November 2012, the 18th Congress of the Chinese Communist Party was held in Beijing, announcing a wholesale replacement of the old with a new Party leadership. I am particularly reminded of the month after the 18th Congress, when I needed to reschedule many of my interviews as many local officials had to attend meetings to learn and discuss (xuexi tiaolun) the main themes of the 18th Congress. Having had some of my schedule cleared, I took to the streets, and the interviews I conducted with villagers, though unplanned, were nonetheless illuminating. Everywhere I looked, whether there were radios in a roadside barber shop, television sets inside restaurants, or the ubiquitous print media being hawked at every corner, I and surely every resident in the city were inundated with constant 163  updates from the 18th congress meetings.  Indeed, the media presentation was overwhelming. It appeared that the bureaucratic machine seemed to be operating with the same alacrity as before. In particular, I can recall one lady who shared her views about a visit in last December by the new president of China, Xi Jin Ping, who had come to tour Guangzhou himself in December. It was the manner of his entrance that struck a cord with her. President Xi entered the city neither aboard a state limousine, nor escorted by the customary motorcade, but rather, aboard a simple chartered bus. It was clear the juxtaposition of his tremendous power and his humble entrance was a striking image. The relative simplicity and lack of ostentatiousness was felt as a distinct departure from the presidents of the past. She felt certain that such a president, willing to shed the trappings of his office, would surely usher in policies that would benefit local people. Most of all, she was struck by the speed of his visit; that is, it had been scarcely a month, since he was installed as the new president. She took the humility and speed of his visit to be signs that the presidency, and by extension the government apparatus, was strong and willing to help people.  I couldn?t help but recall the hundreds of interviews conducted with villagers, whose displeasure with the village cadres and some local policies were always palpable. This drove me to rethink the relationships between the GLD government and the central state through a historical review of state-society power transformation in reform China.   Governance restructuring in reform China has sought to move along what Friedmann (2005: 117) calls ?a sustainable path of transformation between two poles of ever-present danger: the lapse into anarchy and the reimposition of a totalitarian rule leading to stasis.? Moving along this path of transformation, on the one hand, the socialist state was eager to promote market reforms and to improve people?s livelihood so as to claim greater legitimacy. On the other hand, the state strove to regain its power over society, worrying that the growing market and community sectors might eventually break away from its sway. With this dialectical approach, the Chinese government has adopted an experimental approach to implement its decentralization strategies. Many important policies were first tested in selected pilot projects and then the central government decided when and where to selectively or wholly transplant these polices into other places. The result of this experimental approach has been a non-linear decentralization process, during which the central government has to keep revising and adjusting its major plans and policies, which gave rise to the regional and temporal disparities in economic and social development throughout the country.   164  Alongside this iterative process of power decentralization, governance transformation in reform China has been essentially a process of state-society separation steered by the state. In this state-led process, even though administrative and fiscal decentraliza