UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Restless landscapes: spatial economic restructuring in China’s lower Yangzi delta Marton, Andrew Mark 1996

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

Item Metadata

Download

Media
831-ubc_1996-147932.pdf [ 32.99MB ]
Metadata
JSON: 831-1.0087834.json
JSON-LD: 831-1.0087834-ld.json
RDF/XML (Pretty): 831-1.0087834-rdf.xml
RDF/JSON: 831-1.0087834-rdf.json
Turtle: 831-1.0087834-turtle.txt
N-Triples: 831-1.0087834-rdf-ntriples.txt
Original Record: 831-1.0087834-source.json
Full Text
831-1.0087834-fulltext.txt
Citation
831-1.0087834.ris

Full Text

RESTLESS LANDSCAPES: SPATIAL ECONOMIC RESTRUCTURING IN CHINA'S L O W E R YANGZI  by ANDREW MARK MARTON B.Sc, The University of British Columbia, 1983 M.A., The University of Victoria, 1990 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE STUDIES (Department of Geography) We accept this thesis as conforming to the required standard  THE UNIVERSITY OF BRITISH COLUMBIA September 1996 ©  Andrew Mark Marton, 1996  DELTA  In  presenting this  degree at the  thesis in  University of  partial  fulfilment  of  the  requirements  British Columbia, I agree that the  for  an advanced  Library shall make it  freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholariy purposes may be granted by the head of department  or  by  his  or  her  representatives.  It  is  understood  that  copying  my or  publication of this thesis for financial gain shall not be allowed without my written permission.  Department of The University of British Columbia Vancouver, Canada  Date _ — .  DE-6 (2/88)  .  —  _  ,  ;  —  11 ABSTRACT  The development of market socialism in China has contributed to a spatial economic transformation characterized, among other things, by the apparent capacity to rapidly industrialize without transferring large numbers of people into big cities. The most striking element of this transformation has been the phenomenal growth and spatial proliferation of industries in particular areas of the Chinese countryside. The conventional wisdom of existing theories of development, industrialization, and urbanization does not adequately explain the emergence of these relatively productive regions. This thesis examines the key patterns and underlying processes and mechanisms which must be accommodated in a new analytical and conceptual framework for understanding rural transformation and the wider spatial economic restructuring in China's lower Yangzi delta. The overall objective is to explore the theoretical implications of the local character of regional change through an evaluation of a hypothetical model of mega-urbanization. The model situates the emergence and specific patterns of industrial production within a complex network of interactions and interrelationships embedded in overlapping administrative and institutional structures which are themselves largely tied to the circumstances of particular places. The resulting investigations are based upon an analysis of regional and local level statistical and other documentary sources, numerous interviews, field observations, and a survey questionnaire of rural enterprises which was part of a detailed case study of one county level area in the lower Yangzi delta. Two central findings are revealed. First, the patterns and underlying processes and mechanisms of regional development in the delta are fundamentally linked to intensely localized exigencies and opportunities within the wider Chinese space economy. Second, external economies, the dynamics of agglomeration, and the role of large cities and other exogenous forces, while significant,  Ill  were less important in the delta than were endogenous forces. The details of these findings are incorporated into a revised model of megaurbanization which highlights the critical processes and mechanisms which underlie the patterns observed, what establishes these processes and mechanisms, and what stabilizes and reproduces them. The thesis concludes by suggesting an agenda for the creation of appropriate planning and management responses for the lower Yangzi delta region.  iv 1  TABLE OF CONTENTS  Abstract  ii  Table of Contents  iv  List of Tables  vii  List of Figures  viii  List of Plates  xi  Romanization  xii  Acknowledgements  xiii  Dedication PARTI  xv  INTRODUCTION A N D T H E O R E T I C A L F R A M E W O R K  Chapter One  INTRODUCTION  1 2  1.1 Rationale, Scope, and Objectives  4  1.2 Redefining a New Critical Regional Geography 1.2.1 Chinese landscapes of transformation and the representation of place 1.2.2 Regional geography: As method and as theory 1.3 The Lower Yangzi Delta Region and Kunshan 1.4 Research Questions 1.5 Methodology 1.6 Organization of the Dissertation Notes  7  Chapter Two  2.1 2.2 2.3 2.4 2.5 2.6 2.7  REGIONAL D E V E L O P M E N T A N D INDUSTRIALIZATION: TOWARDS MEGA-URBANIZATION  Development Theory in Crisis: Beyond the Impasse Confronting the Post-Modern Void: Taking Diversity Seriously Linkages and the Transactional Revolution A New Geography of Production: Making Space for Place Cities, Towns, and Rural Transformation: The Chinese Development Debate Rural and Urban in China's Regional Development: Seeking a Middle Ground Mega-Urbanization in the Lower Yangzi Delta: Enterprise Location and the Reconstitution of Local Space Notes  7 10 13 20 21 27 29  32 33 40 43 45 53 62 66 71  V  PART II  T H E LOWER YANGZI D E L T A : HISTORY A N D PRESENT CONDITIONS  81  Chapter Three HISTORICAL G E O G R A P H Y A N D C O N T E M P O R A R Y PATTERNS O F C H A N G E  82  3.1  3.2 3.3  3.4 3.5  The Lower Yangzi Delta From the Late Imperial Period 3.1.1 Natural environment and spatial economic structure 3.1.2 Rural-urban relations and the urban penumbra Pre-Reform Political Economy Reforms and Transformation in the Rural Economy 3.3.1 Changes in agriculture 3.3.2 Township and village enterprises: New shapes, old patterns Spatial Economic Patterns in the Lower Yangzi Delta At the Edge of Shanghai: Kunshan to the Fore Notes  PART III  RESTLESS LANDSCAPES: RESTRUCTURING T H E KUNSHAN COUNTRYSIDE  Chapter Four  4.1  4.2 4.3  4.4  STRUCTURE OF LOCAL GOVERNMENT AND RELATIONSHIP T O ENTERPRISES  Bifurcation of the Functions of Local Government 4.1.1 Community administration 4.1.2 Ownership and management of enterprises Spatial Proliferation of Non-Agricultural Activities Formalizing Local Institutional Structures in a Partially Reformed Command Economy 4.3.1 Capitalism with Chinese characteristics 4.3.2 Individual interactions and interrelationships 4.3.3 Horizontal and vertical linkages 4.3.4 Economic Cooperation Commission Socialist New Rural Area With Chinese Characteristics Notes  82 83 86 90 93 94 102 109 118 123  129  130 131 131 138 144 151 151 153 155 157 162 164  vi Chapter Five  5.1 5.2 5.3 5.4  Transportation Dianshanhu Town Tongxin Village Specialized Development Zones Notes  Chapter Six  6.1  6.2 6.3 6.4  CASE STUDIES: TRANSPORTATION; DIANSHANHU TOWN; TONGXIN V I L L A G E ; A N D SPECIAL D E V E L O P M E N T ZONES  LINKAGES A N D T H E LOCATION OF NON-AGRICULTURAL PRODUCTION  Investment in Local Development 6.1.1 Capital formation 6.1.2 Reproduction of capital Ownership, Land, and Labour Enterprise Procurement, Marketing, and Management Transactional Environment Notes  170 170 183 188 194 200  204 205 205 207 212 219 231 233  PART IV CONCLUSIONS  237  Chapter Seven MEGA-URBANIZATION IN T H E LOWER Y A N G Z I D E L T A : T H E O R E T I C A L A N D POLICY C H A L L E N G E S A N D RESPONSES  238  7.1 7.2 7.3  Negotiating and Managing the Regional Transformation: Institutional Parameters and Rural-Agglomeration Theoretical Implications: Urban Transition or Regional Resilience? Planning and Management Agenda: Chinese Solutions for Chinese Problems Notes  239 247 255 263  BIBLIOGRAPHY  266  APPENDICES  288  Appendix 1 Appendix 2 Appendix 3 Appendix 4 Appendix 5  Timetable of Interviews Notes on Chinese Statistical Sources Enterprise Survey Questionnaire Structure of Chinese Administrative Divisions Abbreviations  288 294 298 315 319  Vll  LIST OF TABLES  Table: 1.1  The Lower Yangzi Delta in China, 1994  17  4.1  Structure and Development of Community Administration in Kunshan, 1978 - 1992  135  4.2  Town Level Public Finances in Kunshan  140  5.1  Transportation in Kunshan: Passenger and Freight Volumes  175  6.1  Residential Savings and Investment in Fixed Assets in Kunshan  208  6.2  Distribution of Markets by Product Category and Distribution of Product Categories by Market: Kunshan, 1991  224  vm LIST O F FIGURES  Figure: Ll  East China, the Lower Yangzi Delta, and Kunshan, 1991  14  2.1  Rural Industrialization and the Development of Rural Urbanization  64  2.2  Post-Reform Rural Transformation and Mega-Urbanization in the Lower Yangzi Delta  68  3.1  Prefectures, Counties, and Cities: Jiangsu and Shanghai, 1991  84  3.2  Total Agricultural Output by Sector: Jiangsu, Selected Years  96  3.3  Agricultural Output Value Per Mu of Arable Land: Jiangsu and Shanghai, 1991  99  3.4  Population Density: Jiangsu and Shanghai, 1991  100  3.5  Rural Industrial Development: Jiangsu, Selected Years  103  3.6  Total Social Product and Labour Force by Sector: 105  3.7  Rural Jiangsu, Selected Years Non-Agricultural Employment as a Proportion of the Rural Labour Force: Jiangsu and Shanghai, 1991  107  3.8  Rural Per-Capita Net Income: Jiangsu and Shanghai, 1991  110  3.9  Rural Industrial Output: Jiangsu and Shanghai, 1991  112  3.10  Gross Value of Industrial Output: Jiangsu and Shanghai, 1991  114  3.11  Industrial Output Per-Capita Increase, 1985 - 1991: Jiangsu and Shanghai Total Social Product and Sectoral Distribution: Kunshan and China, Selected Years  116  3.12  Kunshan: Administrative Divisions, 1993  133  4.1  119  ix  4.2  4.3  Distribution of Ownership and Gross Value of Output by Ownership of Industrial Enterprises: Kunshan and China, 1991  146  Sectoral Distribution of Enterprises, Employees, and Output: Kunshan, 1991  148  5.1  Kunshan Transportation: Roads and Railway, 1993  172  5.2  Most Important Methods of Transportation for Kunshan Enterprises  175  5.3  Kunshan Transportation: Canals and Waterways, 1993  180  5.4  Dianshanhu Town, Tongxin Village, and Special  6.1 6.2  Development Zones: Kunshan, 1992 Source Distribution of Enterprise Investment: Kunshan Disbursement of Net Earnings and the Most Important Reasons for Establishing Enterprises: Kunshan  6.3  185 206 210  Ownership Distribution and Gross Output by Ownership of Enterprises in Kunshan, 1991  213  6.4  Kunshan: Location of Enterprises Surveyed, 1992  214  6.5  Costs of Land as a Proportion of Initial Enterprise  6.6  Investment: Kunshan Most Important Reasons for Enterprise Location: Kunshan  6.7  Source Distribution of Enterprise Labour Force:  216 216  Kunshan, 1992  218  6.8  Procurement Methods of Enterprise Inputs: Kunshan, 1991  220  6.9  Distribution Channels for Sales of Enterprise Production:  6.10  Kunshan, 1991 Source Distribution of the Procurement of Enterprise Inputs: Kunshan, 1991  220 222  X  6.11  Distribution of Markets for Enterprise Production: Kunshan, 1991  222  6.12  Disadvantages of Enterprise Location: Kunshan  227  6.13  Most Important Methods of Solving Technological and Management Problems in Enterprises: Kunshan, 1990 - 1992  227  >  6.14  7.1  Type, Location of Training, and Source of Expertise for Enterprise Employees: Kunshan, 1990 - 1991  229  Rural-Agglomeration and Mega-Urbanization in the Lower Yangzi Delta  249  xi LIST O F PLATES  Plate: 1.1  Crop Production in the Lower Yangzi Delta  15  1.2  Rural Industrialization in the Lower Yangzi Delta  15  3.1  Household Sideline Production  97  3.2  Office Farmers  97  4.1  Rural Enterprises  145  4.2  Surgical Glove Factory  145  4.3  Restless Landscapes  150  4.4  Agricultural Land Protection Zone  150  5.1  Town Road  173  5.2  Village Road  173  5.3  Main Canal  18.1  5.4  Small Canal  181  5.5  Fields of Canola  191  5.6  Fields of Factories  191  Xll  ROMANIZATION  The People's Republic of China uses the pinyin system of romanization. Therefore, with only very minor and infrequent exceptions, I have used pinyin throughout this thesis. As a result, certain Chinese names, perhaps better known by their older more conventional spellings, have been slightly altered. Thus, Mao Tse-tung becomes Mao Zedong, Nanking and Peiking become Nanjing and Beijing respectively, and Yangtze becomes Yangzi. The name "Yangzi" traditionally referred only to the lower reaches of the great river beginning near the city of Nanjing and running downstream to the sea. Although this name is now rarely used by the Chinese, for the sake of convenience and familiarity I will refer to the river as the Yangzi. The more common Chinese name for the Yangzi is Changjiang ~ literally Long River. In Chinese, surnames are written first (usually one character) followed by the given name (usually two characters). Chinese names which appear in the text or footnotes will also adhere to this style. I have cited books and articles which included older forms of romanization by their exact title. However, when quoting from translations, texts, or articles, I have taken the liberty of substituting pinyin for the original romanization in which some words may have appeared.  Xlll  ACKNOWLEDGEMENTS  I wish to express my deepest gratitude to my supervisor, Professor Terry McGee, for his patience and encouragement and for providing me with the opportunity to fulfill my dream of living and working in China. In the streets of several Asian cities and in Empress lounges around the Pacific, from his lectern, and during Sunday afternoon "dissertation strategy meetings" in the "seminar room" on Dunbar Street, his counsel and intellectual support have been inestimable. I have also benefited from the guidance and expertise of the other members of my dissertation committee: Dr. David Edgington undertook meticulous editing of earlier drafts which added to the clarity and readability of the thesis; Dr. Graham Johnson, who introduced a sceptic to the many delights of Guangdong cuisine, provided the critical sinological perspective; and Dr. Robert North, with whom I frequently compared notes about a restructuring China in the context of the former Soviet Union, enhanced my understanding of the wider socialist world in transition. I must also record a special note of thanks to Dr. Claude Comtois of the Department of Geography at the University of Montreal, for his camaraderie and support. His vigorous probing of my early speculations, mostly while mutually engaged in intensive fieldwork in China, challenged me to refine and focus the investigations. These investigations were facilitated by research collaborators, colleagues, and friends in China to whom I express my profound appreciation. To my teacher and mentor, Liu Junde, Professor of Geography at East China Normal University in Shanghai and Director of the Research Centre for Administrative Divisions of China, I am especially grateful. Professor Liu embodies all of the finest qualities of a truly remarkable scholar and gentleman. His wisdom and energy in the field (his students call him the "fieldwork tiger"), combined with a subtle, unpretentious manner, opened China to me in a way that revealed the essence of her underlying qualities and complexities. The greatest first-hand assistance in the field was provided by my friend and confidant, Zhang Ming, now pursuing his Ph.D. at the University of California, Berkeley. Research collaborator and fieldworker extraordinaire, he endured with immense intellectual and intestinal fortitude and good humour my persistent prying and determination to get at the whole story. I also wish to express my deep gratitude to certain individuals in the Civil Affairs Bureau in Kunshan who shall remain nameless. Two in particular, who love to fish in the streams and lakes of Kunshan, frequently reminded me of their philosophy regarding the size of the fish they caught... Back in Canada, a special debt is owed to Jeanne Yang who transformed the written manuscript into beautiful type, and Catherine Griffiths who turned the maps and charts in my head into illustrations in the text. I also wish to acknowledge J. F. Proulx, wherever he may be, for preparing earlier versions of some of the maps. He was part of a small coterie of like-minded graduate students who afforded the  xiv stimulating environment so vital to my intellectual development. Among them, Charles Greenberg, now variously of the University of Sydney and Capilano College, must be singled out for special mention. From the foot of glaciers in the coast mountains, to the khlongs and golf courses of Bangkok, and the back alleys of Ho Chi Minn City, he has provided the most enduring scholarly sustenance and friendship. The research for this study would not have been possible without the generosity of a number of organizations. A doctoral fellowship from the Social Sciences and Humanities Research Council of Canada provided crucial financial support for my Ph.D. program. A Young Canadian Researchers Award from the International Development Research Centre, Ottawa, and a Graduate Research Scholarship from the Canadian International Development Agency funded the fieldwork in China. A Canada/China Scholar Exchange Fellowship, jointly from the Canadian Department of Foreign Affairs and International Trade and the Chinese State Education Commission, furnished two return airfares to Shanghai and two years of critically important official status in China as a Senior Advanced Student. East China Normal University provided office space, access to other university facilities, and a small monthly stipend. I wish to record my sincere gratitude to all of these organizations. The sustained intensity of the work necessary to successfully complete my Ph.D. has also placed enormous demands upon my family. These demands were especially onerous as they coincided with the birth of our twin daughters, Vanessa and Suzanne. Now nearly three, they have never quite understood why baba always had to go back to work after supper and on weekends, although they have remained largely cooperative throughout. Their cheerfulness and exuberant curiosity are a constant source of delight and spiritual nourishment. Finally, and far above all else, I wish to acknowledge the prodigious forbearance and patience exhibited by my wife, Wei, whose love and understanding eased my endeavours in the research and writing of this dissertation. Not only has she toiled tirelessly behind the scenes, helping with the fieldwork in China and clarifying my clumsy translations, she has simultaneously undertaken most of the parenting of our children, as well as the academic and research demands of her own graduate degree. Yet through all this, we both agreed recently, perhaps assuaged by the many pleasures of living in university family housing, the experience has been thoroughly and entirely enjoyable and fulfilling. This is deeply reassuring since the completion of this dissertation and my Ph.D. signals not the end of a prolonged and arduous project, but rather the beginning of a much longer journey.  A. M. M. Vancouver, Canada August 1,1996  XV  To the memory of Joseph, Robert, and Ian.  1  PARTI  INTRODUCTION A N DTHEORETICAL  FRAMEWORK  The first two chapters of this thesis comprise Part I of four parts. The introduction (Chapter One) which follows positions the scope and objectives of this study in the context of contemporary reforms in the People's Republic of China. It will elaborate in some detail the particular spatial and methodological focus of the research. Chapter Two will review and interrogate the relevant theoretical literature. Chapter Two will conclude Part I of the thesis by proposing a theoretical framework that informs the analysis and subsequent deliberations in the remaining five chapters.  2 Chapter One INTRODUCTION  June 26,1992 Penglang is a small lower Yangzi delta town in the southeastern fringe ofJiangsu Province. Located near the historic city ofSuzhou and at the edge of the great metropolis of Shanghai, the town and its 24 villages were simultaneously in the midst of these large conurbations while somehow not quite a part of them. The expanse of bright yellow canola and luxuriant green wheat crops were frequently interrupted by small settlements, many with factories nearby or sometimes alone among the fields. In the distance, an old man in faded blue tunic and trousers tending a gaggle of goslings, stumbles off a dirt track as two fastidiously attired young women swoosh by on scooters. Down the main road a throng has gathered to watch two drivers, in the vicinity of two overturned trucks, loudly arguing among a shambles of spilt running shoes and bamboo shoots. At one end of the scene a man pushing a large black bicycle, hopelessly overloaded with hundreds of compact discs to which were lashed several fresh bamboo shoots and a new pair of running shoes, weaves recklessly through the distracted crowd. There were many such prominent and paradoxical elements of the countryside here which did not conform to the sharp boundaries between city and rural so clearly indicated on the map. Touched by these ambiguities, I retreated later that evening with a colleague, as I always did after arriving in a new field site, into the lanes and back alleys of Penglang's oldtown. Though I had never before been there, the ambience was familiar to me. Wandering from the government offices and the new two lane bridge over the main canal which established the administrative and economic centre of Penglang, it took about an hour to find what I was looking for. Clinging to the end of a row of whitewashed two-storey buildings, beneath the slanted tile roofs blackened by the shading of coal dust and the smoky air that always hung over such towns in the still of the early evening, was the New Joy Noodle and Wonton Shop. Proprietor Jiang, in scruffy white t-shirt and boxer shorts which belied his heavy gold chains, new watch, and rings, beckoned to us from the now deserted establishment. Through the door to the back was the cooperatively owned hatchery that he managed and which supplied this shop and others as far away as Shanghai and Zhejiang.  3  Several hours later, satiated with bowls of noodles and salubrious tea-boiled fertilized goose eggs, and relaxed from the libations, Mr. Jiang proclaimed that if anyone could understand what was happening between his shop and the primary school some 400 metres or so to the south, they could understand China. That is: down the narrow cobbled lane, past the lodgings of his neighbours and the tiny store that sold us the beer — bottled at a brewery a few kilometers down the road; across the stone bridge that arched over the pitch black water of the canal filled with barges laden with construction materials, cabbages, and large ceramic jars of pickled vegetables -- a famous local specialty food; past the gates of the garment factory and around behind the chemical plant, still audibly wheezing even at this late hour; along the path next to the new house with colour television blaring from the second floor veranda, with two shiny motorcycles parked in front, and a white picket fence enclosing a rose garden that faced fields of ripening winter wheat extending across the delta to a village some distance away; to the now quiet four-storey primary school with the silvergrey dome of its astrophysical observatory perched incongruously on the roof. Embedded in those 400 metres was the epitome of all the social, political, and economic character of the region, and all the human and physical geographies which defined this particular place. A place signifying the tantalizing prospects for change and its enormous problems and complexities, yet also a place that has somehow compelled loyalty and in many ways defied change. Surrounded and absorbed by the subject of my inquiries, I became conscious of an underlying tension that I would need to confront more directly during the remainder of the fieldwork. It was a tension, when resolved, that cautioned against a spatial and cultural voyeurism in my approach and attitude towards the investigations, in favour of something profoundly more sensitive and exquisite in its experience and flavour. Later, as we groped along the labyrinthine alleyways of old Penglang on our way back to the guesthouse, arms extended sideways touching the walls through the darkness, and eyes cast skyward to follow the ribbon of stars between the curled roof-tops, I knew Mr. Jiang's simple, but deeply perceptive words about that place, indeed about China, would stay with me for a long time.  4 1.1 Rationale, Scope and Objectives This thesis is about understanding the nature of change in places like Penglang as part of the wider transformation occurring across the region and throughout China. The degree to which modern China has been transformed and the way in which this transformation has been expressed are at once fascinating, encouraging, baffling, and bothersome. Since "Liberation" in 1949, the development of a centrally controlled economy with shifting policies towards industry and agriculture, and cities and countryside, has been arduous and at times painful. By the late 1970s, however, a fundamental reorientation of the strategies for development in China began to emerge. The major turning point was the institution of reform experiments in the countryside authorized by resolutions promulgated at the Third Plenary Session of the Eleventh Central Committee of the Chinese Communist Party (CCP) in December 1978. These reforms had, by the early 1980s, culminated in the household production responsibility system in agriculture and other policies which gave a more prominent and flexible role to the development of non-agricultural activities in the countryside. Other macroeconomic and industrial reforms were generally characterized by the gradual abandoning of the command economy structures and parallel efforts embedded in the "open door" policy designed to encourage the input of foreign capital, technology, and expertise. Notwithstanding very short-lived attempts to roll back economic reforms and recentralize the economy, the now widely accepted goal of transition to a "socialist market economy with Chinese characteristics" was officially proclaimed in 1992.  1  The inherent ideological ambiguities and contradictions of such proclamations were captured in bizarre theoretical concoctions which rationalized elements of the reforms which did not neatly conform to socialist orthodoxy. But there are other 2  reasons why development in China deserves attention. The sheer magnitude, rapidity, and spatial extent of changes which have affected one-fifth of mankind simply cannot be ignored. Furthermore, contemporary "socialist transition" in China has differed from  5 the reforms implemented in Eastern Europe in the 1960s which merely tinkered with the existing system to rationalize and improve the command economy structures. It 3  has also differed from the "big bang" reforms implemented in the 1990s in the former Soviet Union which aimed to accomplish a quick transformation to a market economy. The more gradual, sometimes halting approach to economic reforms in China, on the other hand, should be viewed more as a pragmatic attempt to avoid the trauma and large costs of a sudden transformation, rather than as ambivalence about the objectives of the reform measures. Perhaps the most important reason for paying attention to the socialist transition in China is the truly intriguing and unique patterns and processes of growth and development which have emerged during the reform period. Chief among these, was the apparent capacity to rapidly industrialize without transferring large numbers of people into big cities. One of the most striking elements of this transition has been the phenomenal growth and spatial proliferation of industries in the Chinese countryside. Through a combination of circumstances, sometimes deliberate, but often not, China has fallen upon a means of dealing with the worst problems of massive uncontrolled migration to large cities. That is not to say that cities in China were not undergoing tremendous stress, nor do I wish to suggest that they were unimportant in the development process. With perhaps 850 to 890 million people residing outside urban agglomerations, however, the patterns and processes of development in the Chinese countryside were clearly extremely important. The opening up of markets and the increasing autonomy of enterprises with reforms, also entailed an erosion of the divisions between rural and urban, reflected in the emergence of particular administrative and institutional responses. The old unitary hierarchy of command relations, which relied upon distinct administrative and economic boundaries between rural and urban, has yielded to modes of interaction which more vigorously utilized a diverse array of manipulation, concessions, coercion, and partnerships. Changes in the structure of ownership also meant that increasingly independent economic actors in a  6  restructuring space-economy were negotiating and managing their own mutually beneficial interactions and interrelationships. What were the spatial consequences of this transition in China and how, if at all, did they resemble the patterns and processes suggested by the conventional expectations? The scope of this study is framed by two assertions. First, that development of market socialism in China has contributed to spatial economic restructuring most clearly demonstrated by the rapid industrialization of particular areas of the countryside. This thesis will assess the transformation in the countryside of the lower Changjiang (Yangzi River) delta region since 1978. It will focus in particular upon the growth and spatial proliferation of industrial activities in areas hitherto considered "rural". The dramatic changes seen in the lower Yangzi delta, as with other similarly transforming regions in China, were imbued with a host of ideas about industrial prosperity and urban life as well as a body of political and socio-economic values which have confronted the integrity of the modern socialist system. Viewed from afar, as they most often were, such changes seemed "wondrous". Up close, as this study 4  will show, the story was "messier and more complicated". The patterns and processes 5  of development in the lower Yangzi delta, in fact, conformed to few of the conventional expectations. This brings us to the second assertion that the conventional wisdom of existing theories of development, industrialization, and urbanization does not adequately explain the emergence of these relatively productive regions. What were the characteristics of spatial economic transformation in the lower Yangzi delta which made it different from the outcomes predicted in models based upon advanced, mostly Western, free-market economies? The objectives of this study are to identify the key patterns of this transformation and to determine the crucial processes and mechanisms which drive local and regional development in the delta. Two further objectives include an exploration of the theoretical implications of this transformation and the resulting challenges for planning and policy formation.  7 To address these objectives the research undertaken in this study was contingent upon two fundamental assumptions. First, modernization in China is profoundly more complex than is commonly appreciated. Second, this greater complexity is revealed within deeply Chinese circumstances and conditions. In other words, trajectories of development in China are more a product of Chineseness than of anything else. My hunch was that there were latent patterns and trends sine qua non, built into Chinese socio-economic and institutional sub-structures, which were emerging as more prominent under the reforms. To adequately grasp the essence of spatial economic restructuring in the lower Yangzi delta, therefore, will require setting aside most of the conventional theoretical and methodological assumptions. 1.2 Redefining a New Critical Regional Geography Bai wen buruyijian (It is better to see once than to hear a hundred times from others) Chinese Proverb Geography based on books is mediocre. With maps one can do a little better. But the only way to do it well is in the field. Paul Vidal de La Blache  1.2.1  6  Chinese landscapes of transformation and the representation of place Pondering the mood of this study I could not help recalling the famous painting  Qingming Shanghe Tu (Qingming Festival by the River). Painted sometime between 7  1111 and 1126 by the virtually unknown court artist Zhang Zeduan, the 5.25 metre long horizontal handscroll depicts scenes in and near the Northern Song capital of Bianliang (or Bianjing ~ modern Kaifeng in Henan Province). One of the great masterpieces of landscape painting, it captured the transition from countryside and farmyards to the busy streets of a traditional Chinese city. The poignancy and relevance of the painting for this study lie in its two main features. The first is the exquisite detail and accuracy  8 of the landscapes captured in the original, and how the tone of the composition is informed by a delicate realism and a magnificent sense of space and perspective. The second is the relative inconsistencies, inaccuracies, and artistic attributes of the dozens of later copies, and the nature of the subsequent debates about their authenticity, which have emerged in the nine centuries since the original. Taken together, these features 8  highlight important issues about the transformation of landscapes and place in China and the ways in which this transformation is experienced, understood, interpreted, and represented. Perusal of the scroll itself reveals from right to left three mutually interdependent landscapes each flowing gracefully into the next. It begins in the flat countryside and gives a clear view of the distant trees and a group of nearly deserted farm buildings. Families and other groups are seen coming towards the viewer and proceeding in a quickening pace towards the city, while others are reluctantly leaving. Perusing further, great barges moored along the river come into view in the foreground with a few street activities and shops behind and to the right. This section climaxes where one of the barges tracks into the oncoming current to pass under Rainbow Bridge and into town. Leaving the river, the scene then follows a street as it crosses a moat outside the city walls and enters the capital through one of its imperious gates. The busy vibrancy of the city streets then reach a peak in this section before gradually tapering-off towards the end of the painting. Throughout its length, the scroll is replete with meticulous detail, revealing keen powers of observation and a deep familiarity with the scenes portrayed. The painter knew the atmosphere of the hustle and bustle of street activities, the configurations of flowing water and the navigation of barges, the subtleties of architecture and construction and how to render them, and the ambience of the countryside. Rather than merely depicting the many types of individuals and activities, moreover, the painter was secure enough to characterize them ~ sometimes humorous, often excited ~ but always truthfully. The painting embodies a vivid sense of space and various  9 perspectives, by suggesting at every point a continuation beyond or inside that which is actually portrayed. Zhang Zeduan's style and technique were deeply linked to the specificity and character of the subject. In his knowing and loving representation of countryside, people, and the built environment, objects and their character and form coincide. The honesty of the painting's full and accurate description also revealed a thorough indifference to the conventional formalistic concerns. There were, of course, other examples of such realism among the established elite through Chinese history. Excited by the dramatic landscapes of Mount Hua, the fourteenth century doctor Wang Lu felt unable to convey his true experiences filtered through the traditional stylistic norms. Rather, his paintings, poetry, and travel writings reacted directly to the 9  particular forms and characteristics of the landscapes which had so moved him, fundamentally questioning the accepted narratives. Similarly, the great traveller Xu Xiake conveyed with unmediated vision, meticulous detail, and concise style his experiences of a direct encounter with the landscape. His prose captured a strong sense of the reality of travel in seventeenth century China, revealing insights into the nature of the environment and places as a "fascinating texture of interacting phenomena".  10  Zhang Zeduan's Qingming Shanghe Tu, however, seems to have emerged as the supreme instance of a unique and gentle realism in Chinese landscape painting.  11  The work was frequently copied, but the integrity and veracity of the original scroll were never duplicated. Seen from a distance these copies retained the basic layout of countryside, river scenes, and city-scape. Upon closer examination the subtleties and clarity of arrangement in the original are lost. While there are dozens of examples, the most obvious contrast is with the barges depicted in the original which lie according to the current.  12  In several copies, the barges are moored with their bows  converging on Rainbow Bridge with the sterns of some facing upstream, an impossible situation in the presence of a current. Such copies were apparently based only on literary references to the original, written records of the capital, or in rare cases, access  10 to the original scroll itself. Later painters of Qingming Shanghe Tu, as was common with historical chroniclers and subsequent commentators in China, clearly had a greater desire to render works of aesthetic and erudite merit than to produce an accurate representation of the subject. The tendency to manipulate key elements and to crowd the painting resulted in elegant visual images, organically implausible and difficult to read in terms of continuity, space, and form. The techniques and style employed by Zhang Zeduan, characterized primarily by ingenious changes in perspective to suggest unseen activities within as well as continuing further space beyond the scenes depicted, are in sharp contrast to the later versions. Yet it is the relative attributes of these copies which also serves to emphasize that what was true in the elaboration of the original composition ~ its consummate detail and subtle transitions ~ was also true of the complex landscapes it so uniquely and eloquently portrayed. 1.2.2  Regional geography: As method and as theory I have indulged in this reverie on Qingming Shanghe Tu as a metaphor for the  two underlying perspectives of this study which are linked to my assumptions about the complexity of the transformation that has occurred in China and the "Chineseness" of its inherent processes and mechanisms. The first and more explicit aspect of the metaphor has to do with the nature of how investigations into complex socio-spatial phenomena are undertaken. Here I wish to stress a major goal of the "new" regional geography in examining how the specificity of place is at once preserved and modified within the wider patterns of regional change.  13  Such change may emerge from within the region,  penetrate from outside it, or more likely, arise through some combination of and articulation between the two. The latter phenomenon is generally perceived in spatial terms as the local-global dialectic.  14  Whatever the case, the approach advocated here  necessarily focuses on the linkages between the circumstances and characteristics of locality and place, and the processes through which the social and economic relations of production are established and modified within regions.  15  This approach may be  11 operationalized by creating a set of analytical categories to "interrogate historical geographical situations" to provide "empirical substance".  16  In terms of this study, this  implies specific methods of analysis which include the selection and detailed exploration of particular interactions and interrelationships over space, the resulting structures of economic production, and the historical, cultural, and political circumstances in which they were embedded. Furthermore, the results of these explorations will suggest configurations and connotations "consistent with the questions and regional setting under consideration".  17  More will be said about the finer methodological implications i n the next chapter. For now, I merely wish to emphasize that it is the landscapes and places themselves, as they were for Zhang Zeduan, Wang L u , and X u Xiake, that should always be the subject of our first and most direct attention. This approach is not preoccupied, at least not initially, with the etiquette and social critique of those unable or unwilling to conduct meaningful fieldwork. It is impossible really, to fully understand spatial economic transformation in China say, unless you have actually been there to feel it. Prolonged field investigations, and the physical and intellectual discipline they compel, are essential for the larger project of refining our theoretical understanding of these Chinese landscapes of transformation. They are, moreover, fundamental to what we do as geographers. The second aspect of the metaphor I wish to infer from Qingming Shanghe Tu has to do with the conceptualization of space and place. Like the new critical regional geography alluded to above, Zhang Zeduan has moved well beyond mere description and representation of the landscapes. H e has also touched upon the essence of their underlying character with a style, technique, and set of perspectives which evoke the functional dynamics and mutual interactions among the constituent elements. By synthesizing his keen observations of the finest details of these elements and their wider spatial context, Zhang Zeduan has highlighted the relationships which at once produced and linked different landscapes. The original portrayal of Qingming Shanghe Tu  12 fundamentally captured more than the mere sum of its parts, giving it the presence of a masterpiece and distinguishing it from all subsequent copies. Thus, the new critical regional geography I wish to invoke here, collapses methodology and the conceptualization of spatial differentiation and change, into a theoretical framework concerned with the particular intersection of relations (social, economic, cultural, political, and historical) as they were manifested in and as they constituted place. This notion of place includes more than just the sense of a local setting that is derived solely from the connections which individuals and institutions attach to it, as it is most often defined.  18  It would be useful now to clarify some of the  other terms which will appear throughout this study. The term locale is seen as the physical setting across and through which interactions and interrelationships occur over time.  19  Unlike the conventional definition of locale, however, I will not limit such  encounters only to the social. The term region will refer to spatial segments, at various scales and frequently ill-defined, which incorporate some "formal, functional, or perceptual significance".  20  While conventionally perceived as bounded in some way, I  am less concerned at this point with defining a particular region than with establishing a conceptual focus on the critical interactions and interrelationships which it encompasses. Thus, the relevance of these terms for this study is how they draw attention to local circumstances and context. Rather than treating the region as a convenient neutral backdrop for spatial change, there needs to be a greater commitment to probing the dynamics of the processes of regional formation as they articulate with those changes.  21  Ultimately, I wish to argue that understanding spatial  economic restructuring in the lower Yangzi delta requires conceptualizing the underlying processes and mechanisms in the context of the local geographical and historical circumstances which shape and reflect the character of the region.  13  1.3  The Lower Yangzi Delta Region and Kunshan This study will focus on an area of the lower Yangzi delta defined initially to  include the 7 prefectural regions of Suzhou, Wuxi, Changzhou, Zhenjiang, Nanjing, Yangzhou, and Nantong adjacent to the Yangzi River in southern Jiangsu, and the Shanghai Municipal region.  22  Figure 1.1 illustrates the 9 major cities and 45 county  level administrative units that comprise the lower Yangzi delta region, and the position of Jiangsu Province and Shanghai in East China as of early 1992. Some studies include all of Jiangsu, Anhui, and Zhejiang Provinces in addition to Shanghai as part of the lower Yangzi delta, while others focus on southern Jiangsu, Shanghai and northeast coastal Zhejiang.  23  The particular areal focus of this study is somewhat narrower for  two reasons. Cursory examination of the statistical evidence, and extensive field observations, suggest that the spatial similarities and contiguity of the regional transformation between southern Jiangsu and Shanghai were much stronger than that between Shanghai and northeast Zhejiang. A number of geographical and historical circumstances elaborated upon in subsequent chapters help to explain this phenomenon. The other reason was a more arbitrary methodological decision to exclude any of the region of Zhejiang Province to avoid the burden of having to consult a third largely separate set of statistical sources. Acknowledging that many of the specific findings of this study almost certainly also apply to the region of Zhejiang Province to the south of Shanghai, it is the southern Jiangsu region of the lower Yangzi delta wherein the spatial economic transformation was most dramatically demonstrated. The clearest indication of this transformation arose when passing through the delta's countryside. Infused into the scenes of agricultural production among the dense clusters of rural settlements were a variety of non-agricultural activities. Plates 1.1 and 1.2 provide a glimpse of the spectacular results of crop production in the lower Yangzi delta and just one example of the way tens of thousands of industrial enterprises have  14  Figure 1.1 EAST CHINA, THE LOWER YANGZI DELTA, A N D K U N S H A N , 1991  Yangzi River Source: Adapted from: ZGXZHC 1992, pp. 27-29.  Plate 1J2  RURAL INDUSTRIALIZATION IN THE LOWER YANGZI DELTA Transplanting rice seedlings near a large textiles enterprise Shipai Town, Kunshan  16 insinuated themselves into the agricultural landscape. The development of industry in rural communities such as villages and small towns I will now label rural industrialization.  24  This is to be distinguished from  processes of industrial decentralization from large urban centres to secondary cities or into the countryside  2 5  The term rural is defined initially as areas that were  administratively classified below the county level including towns (zhen), townships (pdang), and villages (curi). County level administrative seats, usually large towns (xianshu zhen) or small cities (xian cheng) were excluded. Rural industry will refer to most enterprises owned and operated at or below the level of towns, townships, and villages. Most state run "rural" enterprises owned and operated at or above the county level were in fact located in county level seats or areas they administered, and should not be considered rural. Similarly, most other county level enterprises were also located in county cities.  Agriculture refers to the five standard Chinese statistical  26  categories of cultivation, animal husbandry, fisheries including aquaculture, forestry, and peasant household sidelines.  27  Thus, for the lower Yangzi delta, rural non-  agricultural activities will refer to industry, construction, transportation, commerce, and other tertiary level services located in rural areas. Many of these designations here defined become rather more complicated when issues of ownership and spatial administrative classifications are further elaborated. The resulting complexities will be discussed throughout this study as they arise. Some of the reasons which justify a focused and sustained examination of recent changes in the lower Yangzi delta are highlighted in Table 1.1. The table positions the 54,645 square kilometers of the delta in a national context as the most important economic region in C h i n a .  28  More than 51 million people ~ 4.28 percent of the  national total ~ live here, on only 0.57 percent of China's territory, making the delta one of the most densely populated contiguous concentrations of people in Asia. It generated 11.37 percent of China's gross domestic product in 1994, and 5.54 percent of total agricultural output.  29  However, the delta's most significant contributions to the  17 Table 1.1 T H E L O W E R Y A N G Z I D E L T A I N C H I N A , 1994  A r e a (km ) Share of national area (%)  54645 0.57  Population (millions) Share of national population (%) Population density (people/km )  51.30 4.28 939  2  2  Gross domestic product ( G D P ) (billion R M B ) Share of national G D P (.%)  511.80 11.37  1  Gross value of agricultural output ( G V A O ) (billion R M B ) Share of national G V A O (%) 1  Gross value of industrial output (GVTO) (billion R M B ) Share of national G V I O (%) 1  Rural G V I O (billion R M B ) Share of national G V I O (%)  87.19 5.54 1266.63 16.47 675.45 20.89  1  1979 - 1994 Average annual growth in G V A O  2  (%)  6.6  1979- 1994 Average annual growth in G V I O (%)  20.7  1979 - 1994 Average annual growth in rural G V I O (%)  33.7  2  2  Notes: 1.  $US 1 = RMB 8.7 in December 1994.  2.  Figures here are for Jiangsu Province.  Sources:  HDDTNJ1993, pp. 230-233; JSTJNJ1995, pp. 13,143, 275-292; SHTJNJ1995, pp. 20-21, 26, 48,141; ZGTJNJ1995, pp. 20-23, 365.  18 national economy were in terms of industrial production. Nearly 16.5 percent of China's industrial output was concentrated here in 1994. The relative importance to China of industrial output in the lower Yangzi delta was almost 4 times its population and 29 times its area. Even more noteworthy was that 53.3 percent of industrial output in the delta was generated by rural enterprises, accounting for more than one-fifth of the nation's total rural industrial output (see Table 1.1). The 1979-1994 average annual growth rates in agricultural output (6.6 percent), industrial output (20.7 percent), and rural industrial output (33.7 percent) shown at the bottom of Table 1.1 were also substantial. These figures refer to all of Jiangsu Province excluding Shanghai, and should be considered the minimum average values for the region of the lower Yangzi delta within southern Jiangsu (Sunan).  30  By contrast, the growth rates in Shanghai over  the same period were 3.2 percent for agriculture and 9.7 percent for industry, while for China the growth rates in agriculture and industry were 6.2 percent and 14.9 percent respectively  3 1  The relative prominence of these values for output and growth of rural  industry in the lower Yangzi delta provides the initial empirical evidence and rationale for an in depth study of spatial economic restructuring in this region. The finer elements of this restructuring will be explored in a detailed case study of Kunshan, a county level city (xianji shi) located in southern Jiangsu Province adjacent to the Shanghai Municipal region (see Figure 1.1). The centre of Kunshan is located 55 kilometers from downtown Shanghai and 36 kilometers from the city of Suzhou. Comprised of 20 towns and 466 villages, Kunshan covers an area of 865 square kilometers, 63.9 percent of which was arable land in 1994, with another 22.3 percent containing lakes, rivers, and canals.  32  A t the end of 1994 the population was 578,269.  33  The average annual growth of industrial output in Kunshan between 1980 and 1992 was over 35 percent.  34  There were several reasons why Kunshan was selected for the most intensive analysis undertaken in this study. In addition to the conspicuousness and rapid pace of its spatial economic change, Kunshan's physical and administrative position in the delta  19 were appropriate for an evaluation of a range of potential forces which might have influenced the patterns and processes of this change. When I initially entered the field specifically to conduct the research for this study, I envisaged an analysis that would focus on the linkages between rural areas in Kunshan and nearby urban centres such as the cities of Suzhou and Shanghai. I considered it important in this context to select an area that was not administratively subordinate to a large city. This ruled out areas such as W u County around Suzhou City or Wuxi County around Wuxi City and all of Shanghai's suburban counties. Kunshan's administrative status was essentially equal to that of the city of Suzhou.  35  Kunshan was between and adjacent to two large urban  centres, but also administratively distinct and more independent than other areas lower in the administrative hierarchy. It also seemed reasonable to expect that proximity to China's largest urban industrial complex would provide ample analytical fodder. It turns out, for reasons that will become clear later in this study, that external forces from Shanghai and elsewhere did not influence local development in places like Kunshan i n the ways and to the degree initially anticipated. Thus, Kunshan's location at the edge of Shanghai and near Suzhou City would preclude distance from urban areas as a variable in rural restructuring. It might have been easier, for example, to demonstrate that the influence of large cities was less important in an equally well developed area further away from Shanghai. Kunshan was also topographically fairly uniform and, as well as straddling major regional and national transportation corridors, had its own well developed internal transportation network. For all these reasons, Kunshan provides a quintessential example of the local character of regional change. Like the unique confluence of bio-geo-climatic circumstances which allow the exquisite qionghua flower to blossom between A p r i l 20 and May 10 only in Kunshan, it is necessary to understand specific local processes and mechanisms, as they intersect and articulate with exogenous forces in particular ways over space to produce patterns of regional development in the lower Yangzi delta.  36  20 Lastly and in more practical terms, Kunshan was an area that I could easily penetrate from my research home in Shanghai. It was also accessible in terms of contacts and networks of informants arranged through colleagues and friends in Shanghai. I am also convinced that I was able to establish a deeper rapport and trust with my hosts in Kunshan since I could converse, at least to a very limited degree, in the local dialect.  1.4  Research Questions In the context of economic reforms and a particular focus on industrial  development in the countryside, satisfactory elaboration of the multidimensional nature of spatial economic restructuring in the lower Yangzi delta will need to consider a number of complex interrelated issues. H o w do the wider forces of economic reform interact with particular local circumstances to condition local and regional outcomes? What is the scope of influence of community based individuals or agencies to significantly determine such outcomes? What are the relationships between these economic and political actors at the local, regional, national, and international levels? How do local actors reproduce, reconstitute, or reject the wider forces of change? What are the relationships among individuals and agencies operating at the local level? Answering these questions will require an analysis of the regional transformation from a local perspective since understanding the complexities and interactions at this level is crucial to realistic analysis of the wider patterns of change in the lower Yangzi delta. However, localities must also be treated holistically, as suggested in the new regional geography discussed earlier, by seriously considering the role of space and place in such questions, and not merely juxtaposed against some conveniently abstracted spatial construct. These issues are to be addressed in the subsequent theoretical and empirical explorations undertaken in this study which are framed by three fundamental research questions: 1) What is the nature of the spatial economic restructuring that has occurred  21 in the lower Yangzi delta region since the beginning of reforms in 1978?; 2) What are the key underlying processes and mechanisms which determine the character of this spatial economic restructuring?; and 3) What are the theoretical implications of this restructuring in the delta for conceptualizing the forces which drive regional development? These research questions will be answered through investigations in the lower Yangzi delta along five overlapping modes of analysis. The first will survey the geographical and historical pre-conditions and circumstances of contemporary change. The second will elaborate the spatial patterns of economic restructuring. The third will explore the structural parameters of interactions and interrelationships among individuals and agencies. The fourth will examine the demographic dimensions of population and labour movements. The fifth will concentrate on the specific morphology of industrialization in the delta. These modes of analysis will rely upon a wide range of available data. Methods of acquiring, cataloguing, and verifying these data are introduced in the next section. 1.5  Methodology M y first entry into the region of the lower Yangzi delta occurred in May and  June of 1983 during a field trip with other geographers from the University of British Columbia. After spending time in and around Nanjing and Yangzhou we travelled the length of the delta to the city of Shanghai. M y initial views of China's largest metropolis were thus framed by experiences in places like Suzhou and the countryside of Kunshan. Like the landscapes portrayed in Qingming Shanghe Tu, the resulting impressions remained vivid and would set the tone for all my subsequent encounters with the region. Two years later I returned to Shanghai to undertake a two year research residency at East China Normal University ( E C N U ) during which my frequent travels outside of the city piqued my interest in the region. It was also during this period that I would establish enduring personal and scholarly relationships that would  22 prove fundamental to the investigations undertaken for this study. However, it was not until my seventh trip to Shanghai in 1989 that I began to consider more seriously the possibility of advanced research in this region of China that I had come to know and love so well. That two month sojourn in 1989 culminated with my first extended visit to Kunshan and the tumultuous events of June 4th of that year. Never before, nor since, has my departure from China been so charged with emotion and imbued with such an intense desire to return. The opportunity came the following summer when I attended the International Geographical Union meeting in Beijing and travelled to Shanghai to initiate the procedures through E C N U that would allow me to conduct fieldwork in Kunshan and elsewhere in the lower Yangzi delta. With formal letters of approval from China's State Education Commission and E C N U , I returned to Shanghai in October 1991 to begin the research. Along with two further extended sojourns during 1992 and two shorter follow-up trips in December 1992 and in May and June of 1993,1 spent just over 20 months in China conducting the fieldwork for this study. The large number of visits over a long period of time, their duration, and the close personal and professional contacts they engendered provided me with unprecedented access to the region and to crucial sources of information through key informants and research collaborators. This study is thus based upon a variety of data collected over a long period of time from a range of sources at several scales of analysis. First hand data sources included information collected from observations, interviews, and a survey questionnaire. Observations were documented through hundreds of slides and a detailed photo-journal, dozens of sketches, and extensive notes written in the field, at my research home in Shanghai, or during countless long rides on trains and buses. I also audio taped some observations which were later transcribed. I visited every prefectural level administrative unit in Sunan except Nantong, in addition to all of northeast Zhejiang Province. I also spent time in every urban and county level administrative unit in Wuxi, Suzhou, and Shanghai, and managed to visit all 20 towns  23 and several dozen villages in Kunshan, as well as other towns and villages throughout the delta. Some of the resulting observations are specifically referred to in the text, figures, and 14 plates. Perhaps the most valuable first hand data were collected during 7 months of intensive interviewing in 1992 and 1993. Ninety-one interviews involving 98 individuals were conducted. Sixty-five of these interviews were in Kunshan. Informants ranged from village, town, city, and county level government and C C P officials, to enterprise managers, engineers, and entrepreneurs. The interviews were largely unstructured and open ended, although they tended to follow a consistent format in terms of protocol, the questions asked, and opportunities for follow-up and elaboration. Prior to a few of the early interviews in Kunshan, informants were provided with a brief statement regarding the research and copies of formal letters authorizing the fieldwork. These were quickly dispensed with in later interviews as my hosts became more familiar with the objectives of my inquiries. Interviews averaged 2 hours in length, ranging from as short as an hour in some cases to all day affairs in others. There were 3 key informants in Kunshan with whom I had repeated exchanges over the duration of the fieldwork, one of whom I have regularly corresponded with since. Typically for the interviews in Kunshan, I was escorted by one of my hosts (by bicycle, or air conditioned government sedan), to meet officials in different bureaucracies or in various towns and villages. Outside of Kunshan, I relied more on my Chinese research collaborators for the introductions. Once the rhythm of interviewing was established (and the driver sufficiently induced) we were conveyed smoothly from official to official across the delta. These interview data were supplemented by dozens of informal discussions with other local officials and residents, and my Chinese academic colleagues and research collaborators. Much of the interview work also included elaborate banquets involving copious quantities of food and alcohol. In keeping with the spirit of this research, the interviews focused on studying the strategies and networks of these local actors within their wider economic and political  24 contexts. Several officials were, of course, wary at the start about divulging too much, providing little more than the ubiquitous "brief introduction". I was usually able to penetrate the issues further by asking leading questions which revealed that I already knew at least part of the whole story. Most eventually provided detailed insights which, taken together, helped to create a fuller account of the relevant issues. Interview notes were handwritten during the sessions themselves, and were reviewed and expanded upon as soon as possible thereafter. These notes, along with those from the field observations amounted to just over 600 loose-leaf pages. References i n the text to specific comments and views recorded in the interview notes are made only according to the date of the interview. These dates, the organizations, and the title(s) of each interviewee are listed in the Timetable of Interviews provided in Appendix 1. Occasionally, if the information discussed was somewhat sensitive, I have chosen to protect the anonymity of my informants, by making reference only to "informants" in the text. The third set of primary data was collected via a questionnaire administered in Kunshan mostly during June and July 1992. The eight page survey was developed i n close consultation with research collaborators at E C N U and officials in the Economic Research Centre in Kunshan. The objective was to elicit a range of information from mostly industrial enterprises across several towns. In addition to basic information about size, ownership, labour force, and production, the questionnaire focused on the details of procurement and marketing activities of enterprises and a variety of issues regarding their location. A copy of the questionnaire as it was administered and an English translation are shown in Appendix 3. A l l but a few of the surveys were completed at locally arranged meetings with enterprise managers or accountants i n 7 towns. In preliminary discussions with town officials I would request that a representative cross-section of town and village enterprises be selected. Heads of local industrial organizations would then summon managers from the enterprises selected to attend a meeting. Accordingly, with minimal preamble, we would lead between 15 and  25 50 enterprise representatives step-by-step through the questionnaire. I also ensured that each respondent was well stocked with cigarettes and tea. Some questionnaires were completed during visits to particular enterprises and others were left with enterprise officials who forwarded them by mail to my research office in Shanghai. In all, some 190 questionnaires were distributed of which 129 were returned. Twenty-five of the questionnaires were deemed too illegible or incomplete to be included, leaving a final tally of 104 usable surveys. A l l four of the surveys returned by mail were included in this final tally. I will make no claims regarding the  statistical  significance or otherwise of the  findings from this survey. The respondents were not randomly selected from the total population of approximately 4000 industrial enterprises in Kunshan at the time. I am confident, however, that when considered in the context of findings from other sources, the results of this questionnaire can reasonably be viewed as representative. Basic ownership and output data from the enterprises surveyed, for example, were consistent with published county level statistics. Other trends revealed in the survey also exhibited internal consistency. Data from the 104 questionnaires used were tabulated in China on a D - B A S E III System. Problems with formatting and data configuration, however, meant that most of the subsequent calculations had to be performed long-hand. Apart from the technical glitches, I was more concerned about highlighting important subtleties i n the survey results which would not have otherwise been revealed. O f course, with the benefit of hindsight, and perhaps more time and money, I might have designed the questionnaire a little differently. But for the purposes of this study it was more than adequate. In addition to these first hand data, a large number of secondary Chinese statistical and documentary sources were also consulted. Statistical sources were largely of the official published yearbook variety. I was also able to locate several supplementary statistical materials in the latest gazetteers, sometimes in local newspapers, but most often in limited circulation internal (nei bu) statistical bulletins  26 and other documents. Though rarely seen by most researchers, I was able to obtain comprehensive statistics at the village level, for example. However, one must exercise great caution when utilizing such Chinese statistical sources. Issues related to the use of such published Chinese statistical sources were of sufficient concern to be included in a separate appendix, (see Appendix 2). The final set of data was gathered from a wide array of secondary documentary sources. These included local, regional, and national newspapers and magazines, local gazetteers, maps and atlases, widely circulated government documents, and other materials provided by informants such as investment brochures and regulations, and information about local trade and labour fairs. Also collected were dozens of books, academic articles published in scholarly journals, along with several more obscure research monographs. The breadth and depth of these data permitted frequent cross-checks and triangulation. Specific figures provided by informants in interviews, for example, were substantiated by utilizing published sources or other informants. The same also applied to the detailed accounts and viewpoints about local developments provided by interviewees. Obviously, most officials spoke from the perspective of their own organizations. However, the versions of issues and events revealed by informants were confirmed and reiterated by other accounts, and were consistent with the larger and rather complex story of local and regional transformation that began to emerge. Findings from the survey questionnaire were similarly corroborated with evidence gathered during field visits and interviews conducted at fourteen enterprises in Kunshan. Thus, while no single element of these data was sufficient on its own to accept or reject particular findings, taken together they provided coherent and consistent evidence to support the overall conclusions and conceptual reformulations.  27 1.6  Organization of the Dissertation This thesis is organized in response to the theoretical and empirical inquiries  necessary to address the three central research questions introduced above. It begins in Chapter Two with a review of the relevant literature along the three broad theoretical discussions of regional development, the geography of production, and urbanization. Elements of each debate relevant for understanding spatial economic restructuring in the lower Yangzi delta will be highlighted. In the context of crucial insights reviewed in the indigenous Chinese scholarship along the same themes, Chapter Two culminates by proposing three specific working hypotheses. These are subsequently incorporated at the end of the chapter into a hypothetical conceptual framework which seeks to capture the critical elements and the underlying processes and mechanisms which determine spatial economic restructuring in the lower Yangzi delta. The three remaining parts of this study specifically address each of the three research questions. Part II (comprised of Chapter Three) focuses on the first question by illustrating the historical and geographical patterns of structural transformation across Jiangsu Province and Shanghai. The chapter elaborates upon the particular features of these patterns not adequately explained by the conventional wisdom. Chapter Three concludes this part of the thesis by introducing some of the circumstances of spatial economic restructuring in Kunshan. Part III will address the second research question by examining in three chapters (Four to Six) the detailed characteristics of the transformation in Kunshan. Chapter Four will explore the relationships between local government administration, the development of rural industrial enterprises, and the nature of the institutional structures which have emerged to manage local economic activity in Kunshan. Chapter Five reviews the findings from four detailed case studies on transportation, one town, a village, and the nature of the special development zones located in Kunshan. The objectives of this chapter are to assess particular local patterns of change to reveal some  28 of the complex processes and mechanisms which drive local development. Chapter Six will focus specifically on the morphology of rural industrial growth and the spatial proliferation of enterprises in the Kunshan countryside. This chapter will highlight the linkages between specific processes and mechanisms identified in the preceding analysis, and the location of rural industrial enterprises. The four chapters in Parts II and III constitute the empirical heart of the research undertaken for this study. They are followed by Part I V which is comprised of the concluding chapter of this thesis. Chapter Seven will engage the third research question about the theoretical implications of spatial economic restructuring in the lower Yangzi delta. It begins by reviewing the major findings of the research to evaluate the three working hypotheses. The main theoretical implications of this study will then be discussed. Several specific clarifications are incorporated into a revised conceptual framework based on the hypothetical model introduced at the end of Chapter Two. The final section of Chapter Seven concludes the thesis by proposing a planning and management agenda for the lower Yangzi delta that is sensitive to the local level realities of regional change.  29 Notes 1.  See for example: Naughton, B. (1995) Growing out of the plan: Chinese economic reform, 1978-1993. Cambridge: Cambridge University Press.  2.  Perhaps the best example was the "primary stages of socialism" arguments which emerged to justify certain particularly capitalist-like aspects of the reforms. See: Lieberthal, K. (1995) Governing China: From revolution through reform. New York: Norton.  3.  The discussion here is based on Naughton, B. (1995) op. cit., pp. 3-24.  4.  Theroux, P. (1993) Going to see the dragon. Harper's magazine (October), p. 34.  5.  Ibid.  6.  Cited in: Butimer, A . (1995) Review of "Vidal de La Blache: 1845-1918. Un Genie de la Geographie". AAAG 85 (2), p. 406.  7.  Qingming refers to the spring festival of remembrance of ancestors. At the time of the painting, tens of thousands of people from all walks of life converged on the capital and the banks of the Bian River. The original is in the Palace Museum in Beijing.  8.  Cohn, W. (1950) Chinese painting. London: Phaidon, p. 75; Dong, Z. B. (1961) Qingming Shanghe Tu (QingmingFestival by the River). Taibei: National Taiwan University; Giles, H . A . (1918) History of Chinese pictorial art. London: Bernard, p. 168; Liu, Y. L. (1969) Qingming Shanghe Tu zhizonghe yanjiu {Qingming Festival by the River comprehensive research). Taibei: Haitian; Loehr, M . (1980) The great painters of China. New York: Harper Row, pp. 163-167; Whitfield, R. (1965) Chang Tsetuan's "Ch'ing-MingShang-He Tu". Unpublished Ph.D. dissertation: Princeton University.  9.  Liscombe, K. M . (1993) LearningfromMount Hua: A Chinese physician's illustrated travel record and painting theory. Cambridge: Cambridge University Press.  10.  Strassberg, R. E . (1994) Inscribed landscapes: Travel writingfromimperial China. Berkeley: University of California Press, p. 319. See also: Li, C. (1974) The travel diaries of Hsu Hsia-K'o. Hong Kong: Chinese University of Hong Kong.  11.  Loehr, M . (1980) op. cit., p. 165.  12.  Whitfield, R. (1965) op. cit.  13.  Archer, K. (1993) Regions as social organisms: The Lamarckian characteristics of Vidal de la Blache's regional geography. AAAG 83 (3); Gilbert, A . (1988) The new regional geography in English and French-speaking countries. PIHG 12 (2).  14.  McGee, T. G. (1995b) Geography and development: Crisis commitment and renewal. Cahiers de geographie du Quebec 39 (108), p. 527. For reasons which will become clear below, I will not engage the burgeoning discourse on globalization.  15.  Gilbert, A . (1988) op. cit., p. 210.  16.  Pudup, M . B. (1988) Arguments within regional geography. PIHG 12 (3), p. 383.  17.  Ibid.  30 18.  Murphy, A . B. (1991) Regions as social constructs: The gap between theory and practice. PIHG 15 (l),p.22.  19.  After: Gilbert, A . (1988) op. cit., p. 212; and Murphy, A . B. (1991) Ibid.  20.  Murphy, A . B. (1991) Ibid.  21.  I will not engage any further than I already have the burgeoning literature about regions, locality and place, and the new regional geography. For a tiny sample of the most recent contributions, in addition to those already cited, see: Bird, J., et al (Eds.) (1993) Mapping the futures: Local cultures, global change. London: Routledge; Entrikin, J. N. (1994) Place and region. PIHG 18 (2); Massey, D. (1994) Space, place and gender. Cambridge: Polity; Thrift, N. (1993) For a new regional geography 3. PIHG 17 (1).  22.  For a detailed explanation of the hierarchy of administrative units in this area see: Appendix 4.  23.  Chreod Ltd. (1996) The Yangtze delta megalopolis. International conference on towards a sustainable future, Qinghua University, Beijing, April 26; Zhou, Y. X. (1991) The metropolitan interlocking region in China: A preliminary hypothesis. In Ginsberg, N., Koppel, B. & McGee, T. G. (Eds.) The extended metropolis: Settlement transition in Asia. Honolulu: University of Hawaii Press.  24.  Small towns are comprised primarily of townships (xiang) and market towns (jizhen) that are settlements economically and socially integrated with the surrounding rural areas. Also see: Appendix 4. The term "community" will in some cases also refer more generally to the county level unit.  25.  Sigurdson, J. (1977) Rural industrialization in China. Cambridge: Council on East Asian Studies, Harvard University.  26.  The confusion arises because of the way statistics are compiled in China. Spatial designations in the context of industrial statistics, for example, incorporate notions of ownership as well as location. For a good discussion see: Ho, S. P. S. (1994) Rural China in transition: Non-agricultural development in rural Jiangsu, 1978-1990. Oxford: Clarendon, pp. 6-7.  27.  National level statistics regrouped sidelines largely within cultivation and some smaller elements into forestry in 1993.  28.  I have not calculated similar 1994 figures for other regions in China such as the Pearl River (Zhujiang) delta in Guangdong in the south, the Shenyang-Dalian corridor in Liaoning, or the area around Beijing, Tianjin, and Tangshan. However, data presented in Zhou, Y. X. (1991) op. cit., p. 106, and more recently in Chreod Ltd. (1996) op. cit., p. 9, very clearly establish the lower Yangzi delta as China's dominant economic region.  29.  For a definition of gross domestic product, see Appendix 2.  30.  The term Sunan will be used hereafter to refer to the part of the lower Yangzi delta defined earlier within southern Jiangsu and excluding Shanghai.  31.  SHTJNI1995, pp. 21, 26; ZGTJNJ1995, pp. 20-23. Growth rates for rural industrial output for Shanghai could not be accurately determined since the relevant data was not provided in constant values. However, it was possible to estimate the average growth in rural industrial output between 1979-1994 for China at about 20 percent based on growth in the collective sector. See: ZGTJNJ 1995, p. 377.  31  32.  Calculated from: SZTJNJ1995, pp. 15,150.  33.  Ibid., p. 39.  34.  Calculated from: KSTJNJ1989, p. 9; KSTJNJ1991, p. 7; JSSXJJ1993, pp. 59, 83; JSSSN1949-1989, pp. 317-318, 393-394; SZTJNJ 1991, p. 15; SZTJNJ 1992, p. 15.  35.  See: Appendix 4.  36.  This delicate white blossom is often regarded by officials in Kunshan as the premier symbol of local development. Local officials also claimed that horticulturalists in the famous Shanghai Botanical Gardens, barely 50 kilometers to the east and despite repeated efforts, were unable to cultivate a similarly exquisite blossom. See: Appendix 1 (April 8,1992); KSXJ (1990), unpaginated preface.  32 Chapter Two REGIONAL DEVELOPMENT AND INDUSTRIALIZATION: TOWARDS MEGA-URBANIZATION This chapter will review and synthesize elements from the often disparate debates on development, the geography of production, and urbanization relevant for the construction of a new conceptual framework for understanding the spatial economic transformation in China's lower Yangzi delta. The need for targeted deliberations along these three sets of theories of spatial change arises as a result of the particular conceptual and empirical realities of this study. First, to understand the nature of regional transformation in the delta, it is necessary to carefully and systematically consider the role of local forces. Several recent commentaries have highlighted the way in which theories of development must reposition concern for locality, considered variously in terms of culture, history, or the socio-economic and political circumstances of place, within the broader context of regional development. Thus, as development is 1  increasingly seen as a localized phenomenon, this augers for a commensurate shift in the intellectual and analytical focus of inquiry. Second, industrial growth, rural 2  industrialization in particular, and the specific location of industries in the Chinese countryside is the rubric which frames the theoretical discussions and the detailed empirical investigations of this study. A critical review of the theories of the geography of production, therefore, also informs our analysis of the precise spatial economic characteristics of places in the lower Yangzi delta. Third, examining theories of 3  urbanization in the context of China's regional development and industrialization also highlights the role of exogenous forces, including external economies, the dynamics of agglomeration, and the role of large cities, particularly in terms of the transition or otherwise to conventional urban forms. This focused elaboration on specific components from at least three substantive theoretical debates should be interpreted neither as inconsistency nor as conceptual  33 indecisiveness. Instead, invoking a plurality of theoretical perspectives explicitly recognizes, and is a deliberate response to, the complex multiplicity of factors which determine the patterns and processes of spatial economic transformation in the lower Yangzi delta. Thus, this chapter begins with the apparent crisis in development theory, 4  the need to accommodate diversity, and a discussion of linkages and the transactional revolution. This leads into an exploration of a new geography of production which confronts the spectacular irrelevance of much of the current discourse and is followed by a critical examination of conventional notions of urban and rural in the context of the Chinese development debate. Taken together, analysis of these debates highlights components of a middle ground which balances local circumstances with the broad theories of location and the production of industrial space, the local and the global, and theory and practice. This reorientation begins with the need for a profound understanding of the place where change occurs. The chapter concludes by proposing a hypothetical model of mega-urbanization which explains the circumstances of local industrial development as part of the regional transformation occurring across the entire lower Yangzi delta. 2.1  Development Theory in Crisis: Beyond the Impasse Whether or not one acknowledges a crisis in development theory it is clear that  fragmentation of the debate presents at once challenges and opportunities for the construction of new approaches i n development studies. F r o m one side press criticisms that the discourse is a sophisticated whitewash of perspectives which rationalize Western dominance. From another side are realizations that Marxist inspired theories flirt with irrelevance as they appear incapable of dealing with real world complexity.  5  Consequently there has also erupted much debate regarding the relationship between development theory and praxis. Each of these issues will be elaborated upon briefly in 6  the following sections along with proposals for overcoming the "impasse" if, indeed, there is one.  7  34 Critics of Marxist and neo-Marxist theories (dependency, articulation of the modes of production, world systems, etc.) highlight at least four dimensions of this work that are said to have caused its stagnation by the mid-1980s.  8  The first refers to the  teleology rooted in structural Marxism in which capital is defined in terms of laws that produce fixed outcomes. This results in a "disorientation" founded in the notion that underdeveloped countries are necessary under capitalism rather than attempting to explain the underlying processes. The second dimension refers to a generalized economistic determinism in which the cultural, social, and political complex is seen as a consequence of national and global economic systems. This interpretation effectively 9  usurps the study of how historical and geographical factors can, first of all, act independently and secondly, how they might influence (or largely determine in some cases) patterns of development. A third dimension refers to the paradox that Marxistinfluenced development theory has been especially weak in building an understanding of the socialist third w o r l d .  10  This is particularly disturbing in light of the dominant role  of the state and its constituent institutional structures in such countries.  11  The fourth  dimension of the critique refers to the propensity of Marxist perspectives to overlook relevant issues leading to highly polemical confrontations with what Booth calls "mainstream literature".  12  One example that will be discussed in more detail later is the  paucity of theorization regarding processes of industrialization especially in contexts where the state plays a particularly central role. Among the numerous commentaries summarizing the critiques of Marxist inspired development theories are suggestions that new approaches are required to illuminate the complex circuits of interdependency of political, social, and cultural factors in the third world in a rapidly evolving global economy.  13  In a superb review of  these new approaches, Corbridge characterized them under the rubric of a more sensitive post-Marxism.  14  While relying on certain commitments of classical Marxism  (a materialist ontology; causal analysis; and a focus on inequalities that emphasize contradictions in the processes of accumulation) post-Marxism departs from the  35 original by rejecting epistemologically privileged exclusivism, the necessities of economistic determinism, functionalist accounts of agency and power, and dualisms that ignore the diversity of both capitalism and socialism.  15  Post-Marxist development theories accommodate certain uniquely geographical insights into the broad tasks of understanding the role of institutional structures and of time and space which begin to address these concerns. Investigations of socialist development strategies that reveal territorial, economic, and political possibilities and contradictions are an example of the former. Understanding development patterns i n China, for example, requires an examination of state-led or inspired organizations, their constitution and powers, and their capacity to mobilize and transfer resources.  16  Moreover, such studies must seriously consider social, cultural, and institutional parameters, and the geo-politics of territorial and sectoral administration to uncover the basis of any transformation. Post-Marxist theories are also sensitive to the circumstances of capital accumulation over space and time and pay attention to the composition and changing dynamics of an evolving world system. More importantly they recognize the complex interplay of the local, meso- and macro-scales in the joint production of place in what Corbridge calls the "spatiality of the development process".  17  Too often, however, issues at the macro-level have dominated these discussions. Much the same can also be said about the apparent resurgence of the new-right, or the counter-revolution in development economics most clearly represented i n the policy priorities of the World Bank and International Monetary Fund. This neo-liberalist ideology most resembles the modernization paradigm without state intervention.  18  However, recent studies from the World Bank have attempted to reconcile neo-classical and "revisionist" perspectives into a "market friendly" explanation of the "East Asian Miracle".  19  Thus, the central assessment that in some economies, under certain  circumstances particular state interventions seem to have led to higher and more equitable growth than otherwise would have occurred. Furthermore, provided a host of  36 conditions are satisfied, similar policies might sometimes benefit other developing countries. Such languid findings from a study of such depth have been excoriated as an elaborate whitewash of the fact that the East Asian experience challenges standard Bank policy.  20  While apparently acknowledging the beneficial role of state intervention  the Bank emphasizes that East Asian governments have simply stuck to "economic fundamentals" that encourage macro-economic stability. Yet there is a nagging sense that a square block has been forced into a round hole. A framework that links rapid growth to the Holy Trinity of rapid accumulation of human and physical resources, efficient allocation, and high rates of productivity improvement, is utilized to evaluate the efficacy of interventionist public policy. Apparent upon closer examination, however, is the genesis of a critical internal reappraisal of World Bank doctrine. There is some evidence of a desire to accommodate alternative approaches: "... for interventions that attempt to guide resource allocation to succeed, they must address failures in the working of markets."  21  Brave words indeed from a coterie of World Bank economists. Moreover, although the East Asian miracle is fundamentally perceived only in economic terms, even the Bank's President accepts the need for further understanding of culture, politics, history, and place-specific factors in development.  22  Perhaps the Bank's advice will become more  sophisticated in response to the complexities that eclipse the parochialism of economic analysis.  23  Against this post-Marxist/neo-liberal backdrop is a small literature on development in the socialist third w o r l d .  24  Urban and regional development in the  socialist third world has been generally characterized by an ideological commitment to egalitarianism and a slower rate of urban growth accompanied by a reduction in the degree of urban primacy. Forbes and Thrift note that such "polarization reversal" is the intended result of ideological convictions and/or the unintended outcome of more fundamental state interests.  25  While models that seek to explain these phenomena i n  the context of a transition to socialism have been useful for revolutionary and post-  37 liberation periods they are not adequate for understanding the socio-spatial implications of recent reforms in countries like C h i n a .  26  Currently, such partially reformed command economies are allowing market forces to play an increasingly important role as the focus of urban and regional development shifts from self-reliance, spatial equality, and economic egalitarianism toward openness, economic efficiency, and entrepreneurialism. This raises fundamental questions about what will happen to city and countryside, and interregional relationships and disparities. In the context of China there are, of course, a number of issues which add to the underlying complexity of these recent reforms. The Chinese government has, for example, always been deeply concerned about the implications of uncontrolled mass migration into large cities.  27  Moreover, since  industrial space in these large cities is dominated by a largely immobile state sector, this has implications for the location of new non-state industry and the emerging patterns of landuse. The key point is that economic transformation, characterized in China by rapid industrialization without a commensurate transfer of large numbers of people into big cities, is unfolding as part of a set of spatial processes. While this is not unique to socialist developing countries, it is more explicit there because of the declared commitment of Marxist-Leninist regimes to the social use of space that serves mass rather than just elite interests.  28  The primacy of these ideological and political aspects  to the organization and function of space forces deeper consideration of these r e a l m s  29  It is necessary therefore, to go beyond the traditional economic domain of regional planning to examine social and political (institutional) organizations i n space.  30  Such an approach potentially enables us to see how spatial economic patterns  are related to historically grounded variations in national and local socio-political structures as they articulate with a post-cold war global economy. It provides an opportunity to challenge the preoccupation in recent urban and regional development literature with advanced and peripheral capitalist societies and the Eurocentrism which imbues such studies.  31  The intersection of political economy and sensitivity and  38 understanding of place-specific factors in development will begin to address the hegemony and parochialism of Western inspired theory, in spite of the disdain by some for the "heavy breathing over locality and a reconstructed regional geography".  32  This approach will also point the way towards a development praxis which will avoid policies based on dogmatic, unilinear, and universalistic views. The present challenge then is the construction of non-reductionist, non-teleological "post-impasse" development theory which balances abstract conceptualizations, which do not result in relevant development praxis, and empirical research.  33  Here I begin to identify  elements of a "middle ground" which conceptually (and literally as will be demonstrated later) seeks to build a new framework for understanding and explaining the rapid spatial economic transformation in regions like China's lower Yangzi delta. Ultimately, it will provide a theoretical basis for challenging present urban and regional planning policies and for proposing meaningful alternatives. Recent discussions of development practice highlight the links between theoretical research and real experiences and circumstances at the local l e v e l .  34  "[H]igher-level work must grow out of and be based upon participatory research at lower levels ..." with the objective of producing theory that has some meaning in practice and can be "genuinely developmental".  35  The issue of scale becomes important  here because such an approach must ensure that both micro- and macro-levels are explicitly related. The methodological implications are clear. According to Edwards "... the crucial thing is always to build from the 'bottom upwards'... to ensure that people's real experiences and concerns provide the raw material for higher-level analysis and synthesis. Any higher-level research which fails to do this may misinterpret what is actually happening and will therefore fail to inform development policy and practice at higher levels in a responsible way." 36  However, a research focus on the linkages between micro- and macro-levels in which neither global nor local studies are privileged will not in itself point the way through the current crisis in development theory. It is also necessary to dispense with  39 certain commonly assumed dyadic relations: global forces define the parameters of local or regional development; large cities drive rural transformation; small-scale indigenous factors are subordinate to larger scale exogenous forces. Theoretical emphasis on the macro-scale forces at the expense of the particular obfuscates understanding and the explanatory power of development research. While it is perhaps true that site specific studies have overlooked the importance of wider economic and political influences, development theory has, in general, concentrated on the macroforces which are most often conceptualized as predominant in shaping local or regional characteristics. O n the contrary, we must always examine wider forces and trends "through the eyes of those who experience and act in them, for it is their perceptions and actions which give meaning to these forces".  37  Moreover, while acknowledging the role of macro-forces, such an approach can also accommodate aspects of development and change that are i n all the most important ways, locally inspired and propelled. This work "avoids any false opposition or one-way determinism between the global and the local ~ but rather insists on studying both the specificities of particular places and the broader forces which shape and are shaped by particular local circumstances and histories." 38  Corbridge does however, caution against an approach to development research which confines itself in theory and practice to the local without accommodating the facts of globalization  3 9  The relevance of the new critical regional geography discussed in  Chapter One now becomes clear in this theoretical context. Some will argue, however, that generalizing from the particular can never provide grand theory, notwithstanding the proclivities of post-modernism, of which more in the next section.  40  2.2  Confronting the Post-Modern Void: Taking Diversity Seriously It is now necessary to briefly reflect upon the premise that there is a crisis in  development theory. There are some who challenge the notion of a crisis and the impasse framework which informs much of the resulting critique, particularly of Marxist and neo-Marxist theories. Peet, in a stinging review of an edited volume of impasse/post-impasse studies claims that what has really happened in development research is a "growing disagreement mainly between modern and post-modern social theorists".  40  Underlying this debate is the notion that development studies have  suffered in the absence of introspective self-criticism which involves rethinking issues of intervention, knowledge, discursive practice and power.  41  Linked to the widespread disillusionment with the modernist project, this Foucauldian critique challenges the Enlightenment notion of truth, linking it to the power and ambitions of Western inspired developmental discourse and practice  4 2  Furthermore, the post-modern mistrust of so-called "totalizing" ways of theorizing is reflected in the quest for historically and geographically produced singularities, across which meanings are unstable, and the repudiation of modernization as the foundation for truth. The intersection of such perspectives has inevitably resulted in an antidevelopmentalism (post-developmentalism) calling for the rejection of concepts of development altogether  4 3  Elements of a critical deconstructionist post-modernism  which attempt to redefine the terms of engagement are perhaps appropriate for those (individuals and agencies) who enter the arena of development burdened with the conventional wisdom of the Enlightenment in all its forms. However, the postmodernist preoccupation with discursive practice provides only infrequent glimpses of what might lie ahead for development theory under this leitmotif. Ill-defined notions of post-developmentalism that visualize possible paths into some yet unknown domain, simply will not do.  41 While attempting to incorporate certain of the insights of post-modernism without surrendering to its "nihilistic excesses", Corbridge with his post-Marxism and Edwards who links development theory and praxis, go some way towards providing a viable theoretical alternative. Meanwhile, in a different vein, Peet offers a reconstituted Marxism with the flowering of a new critical political economy focusing upon transforming social relations, expressed in the politics of class, gender, interregional alliances, and local social movements, to "make 'science' serve the interests of the oppressed", "empower", or "put the last first".  44  Noble and arguably  necessary objectives indeed, but a perspective whose explanatory power and options for policy are limited by its universalism. Development studies must, rather, confront the post-modern void by striving to be systematic about diversity and locale without falling into glorified empiricism. Small-scale studies can suggest wider trends in regional development which affect social and institutional structures in similar ways while the precise morphology of local growth or decline may vary. The reorientation proposed here then, begins with a profound understanding of the place where change occurs - the arena of development and all its geographies. Among the conceptual gymnastics and often tortuous prose of the neoresurgent-post-impasse-ist-isms are elements that will constitute part of a new theoretical and analytical framework for understanding the spatial economic transformation of the lower Yangzi delta and provide the basis for establishing meaningful policy priorities. It is now possible to trace three overlapping emphases in the geography of development literature. The first is the focus of orthodox studies on macro-level issues as the predominant parameters of national and subnational patterns of change. The second is the unabashed focus of traditional regional geography upon deeply empirical locality studies, and experienced development practitioners who condone such work as the basis for theoretical generalizations. More recently, under the rubric of more sensitive post-Marxist, neo-liberal, critical political economy, or post-modern theories, the focus has shifted to include examination of how macro-scale forces articulate with  42 place-specific, generally small-scale local circumstances. This third phase has enhanced our sensitivity to historically and geographically grounded variation in national and subnational development characteristics. Moreover, many have pointed towards a theoretical framework that links the various analytical levels as a means of highlighting the roles of all relevant actors.  45  This framework  usually takes endogenous place-specific circumstances as agents that condition the local articulation of external or exogenous forces. This is an improvement on the conventional view that local characteristics of development are considered to be internally generated, self-sustaining, and independent.  46  However, endogenous forces  are seen as important only in terms of their "interplay" with exogenous forces and how this leads to regional variation. Or put another way, local characteristics play a role only in channeling the articulation of external forces. While recognizing the importance of local factors, macro-scale forces still serve as the dominant theoretical point of reference. This fact is most explicit in the work of Brown who refers to the "imprint of development" onto the landscape as largely driven by external forces and that the impact of local factors "which might produce self-sustaining, internally generated, autonomous growth is, by comparison, negligible"  47  If, as this study hypothesizes, patterns of regional development in China are largely determined and driven by local forces, then a simple blending of analytical scales which retains its emphasis on global, exogenous, or city-linked factors is inadequate. I am proposing here that in order to explain the transformation occurring in the lower Yangzi delta it is necessary to move further still towards a theoretical framework which accounts for the greater importance of local factors.  48  The shift in  emphasis suggested here is a subtle, but important one. It moves beyond the confines of established epistemologies to embrace an approach that considers the totality of local circumstances of development relative to, but not defined by, exogenous forces. Thus, while "[generalization remains the objective ... it needs to emanate from, rather than be imposed ori, the locale being studied"  49  43 M y intention is not to dismiss any of the existing theoretical frameworks, which are able to account for particular elements of regional change, but to introduce a further perspective which can explain persistent patterns that do not fit the conventional expectations. While relational patterns associated with these conventional expectations may be observed, especially notions of core-periphery (urban-rural) and urban industrial agglomeration, characteristics of the transformation occurring in the lower Yangzi delta suggest, on the other hand, the need for analysis that reveals much more about the underlying processes driving regional development. This calls for a sense of place that transcends conventional measures of relationships by recognizing that their meaning varies in different contexts, urging a research protocol that questions these relationships rather than accepting them at face value.  50  A s part of this  reorientation it is necessary to reexamine the nature of linkages between "rural" and "urban" (core-periphery relations). 2.3  51  Linkages and the Transactional Revolution In spite of the voluminous geographical literature on rural-urban migration,  much less has been published on the wider interactions and linkages between rural and urban in developing countries.  52  Instead most research concentrated on analyses of  rural and urban development as separate issues.  53  However, publication of Michael  Lipton's famous book in 1977 initiated a debate surrounding the assertion of urban-bias which began to focus research on issues of rural-urban linkages.  54  Subsequent work by  Rondinelli, Stohr, and Taylor among others, which makes reference to the position adopted by Lipton or his opponents, has provided a range of theoretical standpoints from which to view rural-urban interaction.  55  While the ensuing debate revolved around rural-urban linkages relatively few of the detailed analyses of town and country relationships shed light on important theoretical and policy questions.  56  Within specific regional historical circumstances for  example, theoretical clarification is needed to highlight the precise form and impact of  44 the processes and mechanisms of rural-urban interactions. Moreover, regional and national policy and planning strategies (which implicitly involve settlement systems) often proceed as if it is quite clear as to the nature of the circumstances under which rural-urban linkages will influence the development process.  57  Even more  fundamental, however, is the need to carefully reconsider the theoretical significance of the implied rural-urban divide. Challenging town and country as distinct spatial and conceptual categories highlights several important issues. By reducing complex socio-economic relations over space to a simple dichotomy, the rural-urban divide is elevated to the level of a major explanatory variable. The resulting squabbles usually revolve around definitions of urban and rural characteristics which attempt to demarcate the two categories. The rationale for examining rural-urban relations does not lie in seeking some "new found general explanandum", but rather in the fact that the rural-urban dimension is an important component of the wider spatial economic transformation.  58  Yet there  remains a nagging sense that this approach begins with the assumption that processes of regional transformation extendfromthe city into the countryside  5 9  This study will, on  the other hand, emphasize a view from the fringe that is less willing to assume that this transformation results primarily from the urban penetration and encapsulation of the countryside into new social and economic circuits largely controlled by cities.  60  While geographers have set out in clear terms the main categories of interactions between rural and urban, focusing on the identification and measurement of linkages and flows is often difficult, and in any case, addresses only part of a very complex picture. It is necessary to adopt an approach which relies less on the distinction between rural and urban, focusing instead upon the conditions which create such interactions and the circumstances within which they exist. The notion of a "transactional environment" is introduced here to capture this broader perspective. "Transactional" refers to the now standard framework of interactions and interrelationships initially proposed by Preston and elaborated upon by Rondinelli and  45 Unwin: migration; flows of goods, services, energy and technological and social information; financial transfers and the transfer of capital in other forms; and other transactional activities.  61  "Environment" refers to overlapping structural,  organizational, and social relationships and interdependencies including: political and administrative imperatives and hierarchies; legal and regulatory frameworks; patterns of jurisdiction, decision-making and power; socio-economic exigencies and opportunities, and all their various transactional networks. Taken together these elements comprise a transactional environment, the analysis of which can provide insights into the mechanisms and processes that underlie regional change. Based on the theoretical discussions thus far, and preliminary glimpses of the spatial economic transformation in China's lower Yangzi delta, it is now possible to more clearly highlight three interrelated themes which delineate the research agenda for this study: deeper analysis of the role of locality; elaboration and analysis of institutional parameters; and careful consideration of the geography of production. 2.4 A New Geography of Production: Making Space for Place Although "locality" can refer to areas clearly urban or rural in the conventional sense, the term is used here to include processes and mechanisms occurring in areas conceptually (functionally and spatially) between the two. For some elements it refers to village or township levels, for others it includes county level or regional-scale (lower Yangzi delta) circumstances. Conventional geographical analyses of regional restructuring and the concomitant patterns of settlement transition soon confront two characteristics of the lower Yangzi delta: the fact that highly productive agricultural areas also contain significant non-agricultural production activities; and "rural hinterlands" that are equally important as providers of goods and services to central places, as well as markets for the goods and services from those central places. Theoretical and policy endeavours which place a high priority on urban dispersal, decentralization, or deconcentration, usually investigated and implemented from urban  46 centres, either overlook or underemphasize articulation with local elements of territorially based development. The objective here is to provide an analytical and theoretical middle-ground which legitimizes locale. One set of arguments in the development debate revolves around how processes of urbanization or rural change liberate, constrain, or otherwise influence each other. What has unfolded along another path is a set of arguments that "transcend" ruralurban relations ~ most notably class and politics.  62  Here I must echo criticisms from  Corbridge who correctly points out that construction of a simple political economy with a reductionist concept of politics that concentrates on issues of class (urban, rural, or otherwise), cannot readily explicate complex individual or institutional political and organizational allegiances.  63  Instead, I wish to shift the focus to an examination of the  structure and function of local and regional institutional frameworks. The objective is not to seek a more sophisticated political model perse. Rather, it is to develop at once more refined and more dynamic ways of thinking about organizational forms and functions which will enable us to discern the patterns of socio-political and economic relationships as they impact upon regional change. This suggests a theoretical framework that emphasizes analysis of process as part of the effort to trace mutually conditioning interactions between state and society in the Chinese space economy. Furthermore, it points toward a methodology by which the finest details of institutional structure, roles, and capacities are repeatedly juxtaposed with larger patterns and trends.  64  Analysis of the functional dynamics of institutional parameters, and the transactional relations and social networks that create and drive circulation, needs to be merged with an examination of territorial practices, including enterprise strategies, to understand local and regional level industrial development. The proliferation of industrial activity in the lower Yangzi delta, insinuating itself as it has into all corners of the rural landscape, can be explained neither by abstract generalizations nor by mere empiricist micro-level descriptions. Issues of industrialization, and industrial location,  47 provide a forum for building an analytical framework for understanding the geography of economic development, and the structures, processes, and mechanisms, shaped in part by their geography, which underlie such development.  65  While geographers have  traditionally focused on the economic dimensions of spatial relations, the virtue of considering issues of industrial location in the context of development theory is that it broadens the emphasis on productive activities to encompass processes by which industrial production is established and modified in regions. However, as Storper and Walker have shown, issues of industrial location have not figured prominently in theories of economic development  66  Reasons for this can  be traced through the critiques of traditional location theory and its roots in neoclassical economics, the radical reorientation in theories of uneven development, and the enormous literature on empirical and locality studies that largely eschewed theory altogether. The original critiques focused on firm-centred, ahistorical theories which assumed tendencies towards economic equilibrium, and treated space as an adjunct to economic relationships  6 7  Radical alternatives involved exploration of  structural forces and processes that sought connections between society and space to explain the large-scale patterns and geographies of uneven development.  68  Lastly,  Smith has highlighted how the specifics of restructuring have induced detailed empirical studies of particular sectors and industries (especially high-tech), and analyses of individual localities "conceived in the belief that much of the earlier theory involved little knowledge of grassroots changes [and attempted] to generate a place-specific empirical matrix of social, economic, cultural, and political change".  69  The inability of specific local experiences to inform general theory spawned the search for a middle level industrial geography that could provide a bridge between the two. The first significant contribution came from Doreen Massey who proposed a new theory of location based on the notion of spatial divisions of labour and which focused on organizational structure and the peculiarities of place  7 0  Massey's recognition of the  uniqueness of place, and how this emerges in relation to the formation of spatial  48 structures of production, accommodates variation in the way in which economic activity incorporates and contributes to geographical inequality. The layering of successive "rounds of investment" over a multitude of spatial structures is combined with the preexisting nature of places each of which entails differing relations within the space economy. This necessarily requires consideration of cultural, political, and ideological "strata" (with all their local specificities) in addition to layers of economic history. A geological metaphor is most often invoked to schematize Massey's analysis.  71  This  popular, if misleading, representation conceals the key insights of her work: that localities do not merely reflect external processes, but in fact absorb, translate, and reproduce national and global impulses toward restructuring in different ways. "The uniqueness of place and the constantly evolving and shifting systems of interdependence are two sides of the same coin."  72  Massey's approach has been criticized by many who claim she failed to bridge theoretical and empirical investigations of spatial economic restructuring  7 3  While  sensitive to the complexity and detail of specific industries and places, this greater concern with locational difference is most often preceded in practice by a jettisoning of theoretical frameworks that would allow greater comprehension of geographical restructuring.  74  I am less disposed to this critique than to the fact that Massey does not  adequately address the way in which local circumstances articulate with the local and regional space economy. It is one thing to recognize the importance and uniqueness of place in the geography of production; however, Massey's approach, deeply rooted in what Gregory calls "the totalizing discourse of a thoroughly modern Marxism", continues to subordinate local circumstances to the impulse of external forces of change.  75  While Massey's work provides the foundation for a new critical regional geography others have instead focused on perspectives of capitalist production and economic theory to inform industrial location and processes of industrial development over space. This view is most clearly represented in the work of Scott, Storper, and  49 Walker, and various combinations thereof. This so-called "new industrial geography" aimed not only to explain specific patterns and processes of industrial change, but also to explore how capitalist production is structured by space in addition to how capital constructs space.  76  What makes it new is theorization of the mediating processes of the  organization of industrial production seen most often as a consequence of the technical and social divisions of labour in industry.  77  Walker builds on these notions by  proposing a unified approach to industrial location that simultaneously considers such issues as enterprise location, economies of agglomeration, systems of cities and towns, and analysis of linkages, and which treats geography as integral to the matter of organization.  78  The essence of these arguments can be distilled down to a set of theoretical tools involving three interconnected dimensions of industrialization and regional development: their underlying institutional parameters; the mechanisms and processes of their evolution; and their geographical foundations and territorial specificity.  79  This  framework accepts that political, economic, and social processes are largely determined by the ways in which they are embedded in place and social relations. Much of this work, however, willingly adopts social theory at the expense of economic theory  8 0  Walker, in his (premature) eulogy for the geography of enterprise for example, emphasizes an approach to "geographical industrialization" derived from analysis of the social relations of production.  81  While justifiably critical of concepts such as cumulative  causation and the new international division of labour that do not fully account for the disequilibrating forces unleashed by capitalist industrialization, Walker has glossed over important insights to be accrued from an analysis of the enterprise dimension. I am not arguing for a separate geography of enterprise. What is needed, however, is recognition of the role enterprises play in the processes which determine how complex production systems are organized over time and space.  82  Enterprise  location and organization in the lower Yangzi delta occurs within a dense web of administrative power and influence tied to particular places and articulated through  50 complex networks of institutional and transactional relationships. Furthermore, as I will contend later, rather than emerging as a result of new or evolving social relations of production over time and space, the proliferation of industrial development in the delta is largely the result of a recapturing of former transactional networks ~ the shapes (morphology) are new, but the patterns have existed in one form or another since at least the late imperial period. Geography contributes usefully to concepts in social theory by considering seriously the way in which space and place intersect with industrialization and social life. This seems quite reasonable if we accept an industrial politics of place "in which real actions in real places produce history" and, as I claimed earlier, the global is a product of the particular.  83  Problems arise with this new industrial geography, however,  when the analysis is applied to overlapping issues of urbanization, agglomeration economies, the city, and regional development. I do, of course, accept the view that the city, in terms of its agglomeration economies, emerges as a productive context in its own right, but Walker takes this further. Capitalist organization is constituted i n and through spatial relations and cannot exist without cities. Thus, since industrialization is fundamentally linked to capitalism, industrialization without cities is "ridiculous".  84  Scott is even more explicit when he claims that industrialization and urbanization are inextricably linked  8 5  The line in the sand is clearly drawn!  These authors, and others discussed below who work specifically on China, suggest that "urban areas are the heart and soul of regional economic development".  86  They all lay claim to the central role of external economies and agglomeration in the formation of industrial complexes within cities and across regions. While this may be appropriate for Pred's early nineteenth century New York, Cronon's early twentieth century Chicago, Scott's late twentieth century Los Angeles, or Massey's restructuring British Midlands, all "incubated in the heat and light of capitalist transition", even Walker concedes such a synthesized analysis cannot capture the "often quite unexpected manner by which actual organizational ensembles evolve".  87  Touche! Allusions to  51 "conjectural circumstances" pervade the voluminous geography of production literature and point up a lack of theoretical certitude about what makes regional development happen.  88  The authors briefly reviewed here argue effectively for a theory of industrial location and regional change that can cope with the expansion, instability, and disparities that characterize the geography of capitalist industrialization. Especially valid is the emphasis upon the imbrication of spatial relations and place in the fabric of the industrialization process in which general forces and locational specifics are mutually considered. The conceptual framework breaks down, however, and is unable to provide the necessary analytical insights, in the explicit assumption that the broader large-scale general forces fundamentally underlie specific local outcomes. Since national and global forces are most clearly articulated within and through large cities then, according to the internal logic of these theories, large cities are the necessarily dominant nodes driving regional transformation. Applied to China, this approach links regional economic restructuring inextricably to large cities by arguing, among other things, that rural industrial development is incapable of taking place independently or of being self-generated.  89  Intersecting with this logic is another dimension of these theoretical investigations concerning the dispersal and vertical disintegration of production embodied in the post-fordist, flexible accumulation literature. In a delightfully pompous rhetorical exchange Lovering criticizes Scott's theory of flexible accumulation as deterministic and economistic because it neglects particular local contingencies of industrial restructuring. Scott replies that Lovering's emphasis on the multidimensional complexity of production organization is indeterminate since it provides no meaningful analytical specifications.  90  Reference to this debate is useful in highlighting the  inability of theories of industrial organization and location to adequately conceptualize the transactional environment. Conventional views of the transactional structure of a production system which focus on transaction costs tend to ignore the local institutional  52 context and the detailed character of enterprises.  91  Furthermore, as Phelps has  demonstrated, the transactional activities of industrial enterprises are as important as their production activities  9 2  In a superb study of subcontracting linkages in  Guangdong's Pearl River delta, C . K . Leung refers specifically to the need for a comprehensive examination of the transactional environment  93  Leung's work is  relevant for this study because of the way he combines analysis of micro-scale decision patterns of enterprises, their institutional and transactional contexts, and place-specific socio-cultural, political, economic, and historical peculiarities to explain the patterns and processes of regional economic development. The conventional political economic approach, explicitly grounded in Marxism, adopts the view that regional change is the result of a set of necessary social relations of production. What makes the regional geography proposed in this study different is that it accepts not only that these relations are constructed over space, but that they are also largely determined by, and embedded within the unique intersection of circumstances which constitute place. This is most clearly demonstrated in the spatial proliferation of industrial activity in the lower Yangzi delta. The now widely accepted view that places have specific local characteristics that influence industrial location, however, does not fully explain the spatial economic transformation observed in the delta. Why, for example, was there not the same intensity of industrial activity around Nanjing as there was near other large cities in the delta (see Chapter Three)? Moreover, it is striking that the lower Yangzi delta can largely retain its prominence in a national context as the nation's premier agricultural producer region, while at the same time successfully industrializing. It would seem there are aspects of the transactional environment not fully captured in the conventional explanations of the dynamics of regional industrial expansion.  53 2.5  Cities, Towns, and Rural Transformation: The Chinese Development Debate As part of the effort to theorize regional change in China it is essential that some  attempt be made to review the perceptions and insights of indigenous scholarship. I will not endeavour here, however, to provide a detailed description of the historical patterns of urban and regional development in China. Nor will I attempt to summarize the ideological and policy priorities that have influenced patterns and processes in the Chinese space economy. Excellent reviews of these issues, and of the evolution of the associated scholarship, both inside and outside China, have already been published.  94  While reference will be made in subsequent chapters to specific historical circumstances that are relevant for the region of this study, the objective here is to highlight certain elements from the Chinese literature that move us towards a theoretical framework for understanding spatial economic transformation in the lower Yangzi delta. These elements will be traced through a review of the large-city versus small-town debate, studies of the relationship between agriculture and industrial growth, research on rural industrialization, and more general theories of urban and regional development. While discussion of these issues is readily accessible in widely circulated national and regional geographical journals and books, much important scholarship is available only as special reports, or papers prepared for meetings or government bureaux, with limited often internal (neibu) or restricted circulation. It is important to note at the outset that Chinese urbanization and regional development theories and practice, at least until the late 1970s, were fundamentally linked to the shifting politics and ideologies of the planned economy. That is not to say there was no debate, but it could only occur within a clearly delineated set of parameters most often characterized by the apparent theme of a pro-rural antiurbanism and geo-strategic imperatives. Economic geographers concentrated on producing detailed regional resource inventories and politically safe territorial planning and management schemes serving centrally determined economic planning priorities.  54 By the early 1980s, there was a significant theoretical revival underway as geographers faced the planning and policy challenges of newly introduced economic reforms. Several writers identified the need for a revitalized regional geography as the foundation for conceptualizing the spatial economic restructuring that had been unleashed in the late 1970s and was in full swing by the mid 1980s.  95  However, this  scholarship did not really take on any theoretical significance until the emergence of the large-city versus small-towns in regional development debate which, for the first time, focused attention on the underlying dynamics of geographical processes of development. Reduced to its simplest form the debate examined whether growth strategies should promote development of rural townships and small cities or allow the growth and development of large cities and metropolitan areas. O n one side of the issue were proponents of small cities, towns, and even villages which they believe had been undersupported, thus failing to develop rationally into a larger hierarchy of settlements.  96  A s with many such issues in China the essence of the approach was  captured in a new slogan: "Strictly control the development of large cities, rationally develop medium-sized cities, and vigorously promote the development of small cities and towns" (yange kongzhi dachengshi, helifazhan zhongdeng chengshi, jijijianshe xiaochengzheri)?  1  In practice, it was believed, small towns could be developed much  less expensively serving as nodes of regional industrial development. By the mid 1980s this notion of a Chinese road to regional development, distinguished by its small town bias, was being increasingly challenged by outspoken supporters of large city and metropolitan growth strategies.  98  While being careful not to dismiss the continuing  ideological suspicion of large urban areas this new perspective chose to highlight the inevitability and advantages of metropolitan growth. Cities would be allowed to grow based on investments that proponents believed would improve economies of scale and increase industrial efficiency.  99  Implicit in this strategy was the increased mobility of  labour and migration of rural people into large and medium sized cities. Coinciding  55 with this perspective was the renewal of satellite-city policies to relieve population pressure and industrial concentration in core urban districts, and the notion of "keypoint cities" (zhongdian chengshi) which aimed to systematize preferential access to resources and to increase autonomy in urban decision m a k i n g .  100  Many geographers reacted to the official emphasis on small towns by invoking a crude stages theory of regional economic development.  101  The key economic  characteristics of the majority of small towns in China in the early 1980s were seen as unambiguously agricultural, placing them at the lowest stage of development. The underlying weakness of the economic base would, therefore, preclude any efforts to promote local industrial development. The line of argument continued by claiming that urbanization and national economic development were inextricably l i n k e d .  102  The  regional development envisaged would first see a greater concentration of population and industrial activity in large urban centres. New forms of "dispersed urbanization" proposed by the small town lobby, often linked by critics to the slavish duplication of the policies of advanced capitalist countries, would mean skipping a necessary stage of development and were regarded as financially unviable. Moreover, investment i n urban-style infrastructure in the myriad of smaller settlements would constrain the capital accumulation necessary for improving productivity in large and medium sized cities. While this work was highly critical of the official emphasis on smaller communities it fell short of openly advocating expansion of urban populations, and failed to provide alternatives to the problems of surplus rural labour. Evidence from regional economic restructuring and the conceptual hostility of the pro-urban advocates did, however, stimulate deepening theoretical interest in the processes of urbanization and regional industrial development moving beyond the slogans and platitudes extolling the virtues of small towns. By the late 1980s Chinese academics and bureaucrats were openly reappraising the impact of post-1949 policies towards urbanization and economic development. Investment in urban infrastructure had generally been considered as unproductive and came to lag seriously behind  56 industrial development. The advantages of scale economies were quickly counteracted by severe structural inefficiencies which emerged in what one study called premature diseconomies of scale.  103  Some have also highlighted how decades of artificially  separating rural and urban, administratively, socially, and economically led to a disjunction between processes of regional development and urban based industrial structures. This (en)forced dualism defied the "organic connection" between city and countryside and contributed to an industrialization "divorced" from urbanization characterized in particular by a large number of rural townships that were able to compete effectively with large cities in terms of infrastructure.  104  In a similar view, urban geographers in particular have demonstrated how the concentration on production without seriously considering issues of consumption has led to an underdeveloped tertiary sector.  105  Since links between industrial production  and markets were largely determined artificially under the planned economy, the emergence of an urban based tertiary sector, still in its infancy and unable to adequately cope with the demands of recent reforms, left room for small cities and towns to develop their own organizations and structures. In the context of an underdeveloped hierarchy of settlements this fact is expressed, especially in more developed regions, by the proliferation of community based industrial organizations and local economic development zones that constitute part of a locally developed production complex. Weaknesses in the regional industrial structure were also identified that pointed to the artificial separation between industry and primary activities in agriculture and the production of raw materials, and the limited capacity to utilize labour efficiently.  106  Urbanization was said to lag behind growth of industry as the rate of transfer of agricultural labour and population into non-agricultural employment and urban settlements was lower than the rate of growth of non-agricultural output v a l u e .  107  While the big-city versus small-town debate elicited much rhetoric and extreme positions it did point up a further set of theoretical considerations worth highlighting. A great deal of literature examined the relationship between agriculture and industry and  57 there was a new emphasis on the study of industrial location and the geography of enterprise. Much has been written by Chinese scholars that considers the underlying connections between the agricultural sector and industrial development. Some have focused on the declining proportion of gross domestic product from agriculture as a means of tracing empirically the nation's overall economic development.  108  This work  is often linked to the perceived idiosyncrasies of urban industrial development. Others have instead examined the evolving relationship between accumulation in agriculture and regional industrial investment.  109  This relationship is most often articulated as a  set of five phases along a continuum: 1. "Agriculture supports industry" (yinong bu gong); 2. "Agriculture and industry nourish each other" (nonggong hubu); 3. " A shift to industry nourishing agriculture" (gongyi fanbu); 4. "Industry supports agriculture" (yigong bu nong); 5. "Industry builds agriculture" (yigong jian nong). Once again a complex set of interactions is captured in rhyming slogans. While still based on a set of empirically determined indices this work improves on what went before by attempting to conceptualize the role of capital accumulation and structural shifts in the sectoral distribution of labour. It also makes frequent reference to how such issues are related to specific local circumstances of ownership and management of enterprises and the economic development imperatives of community governments.  110  In a parallel body of literature, geographers have examined how agricultural restructuring has affected patterns and processes in the regional space economy focusing particularly on the dramatic expansion of township and village industries. The relationship between agriculture and this sort of rural industrialization has been theorized in the context of at least four interrelated elements. First, gains in productivity brought about by diversification and commercialization of the agricultural economy have raised rural incomes and savings that have been exploited for local industrial investment.  111  Second, this coincided with the release of surplus agricultural  labourers who, motivated by reforms to increase productivity to maximize personal  58 gain, were available for off-farm employment.  112  Third, these elements in turn  stimulated pent-up demand in the countryside for more manufactured goods, not only to supply the increasingly productive agricultural sector, but also increased local demand for consumer products.  113  Fourth, and the most important element  conceptualized in my view, was the way in which agricultural restructuring revitalized the marketing and commercial functions of small towns.  114  Discussion of these  elements is relevant for this study because of the way in which agriculture is fundamentally linked to industrial expansion and the development of villages and small towns. Historically too, the prosperity of agriculture in some regions, and the relatively high population densities it supported, was seen to foster the creation of numerous agricultural sidelines and off-farm activities which absorbed surplus labour and capital. This propensity for engaging in such activities led to a new social division of labour which underlies the prosperity of small towns and processes of rural industrialization.  115  The significance of space and place in such conceptualizations is more obvious and less contrived than in the corpus of literature on advanced capitalist economies. In important ways the Chinese scholarship on these issues tended to exhibit greater sensitivity to the dynamics of local geographical circumstances when theorizing spatial economic transformation. The literature on the relationship between agricultural restructuring and industrial development has, for example, stimulated a reorientation towards more comprehensive studies of the regional space economy including, but not limited to, the transactional environment, institutional organizations, and deep concerns about the deleterious impact on the rural landscape.  116  This reorientation is a response  to the conventional separation of concepts of industrial location and the geography of production from the broader regional development and planning literature. Industrial location scholarship in China has mirrored trends seen in Western theories, but continues to retain a largely quantitative focus.  117  In addition, several  lengthy studies have appeared recently that catalogue in some detail development and reforms of China's industry.  118  In related work, Yang utilizes a modified theory of  59 location to link issues of regional development and spatial economic change, moving beyond the classical factors of location approach of earlier studies.  119  Transformations  and variations in regional industrial structure were being examined within a theoretical framework that formally considered space and place as they determined patterns in the spatial divisions of labour and the emergence of socio-economic organizations.  120  This  latter focus on what I referred to earlier as institutional parameters, is particularly relevant since it is becoming an increasingly important element in theories of industrial location and models of regional economic development in C h i n a .  121  The territorial structure of particular industrial sectors and the classification of industrial regions has also attracted much attention in the Chinese literature. Emphasis is most often placed upon understanding how industrial "allocation" combines with the building of regional urban systems to provide a rational scale, structure, and spatial pattern of development.  Similarly, much work has been done on special industrial  122  parks and what the Chinese call "special economic and technological development zones" (jingji jishu kaifaqu).  123  In most cases, quantitative models of industrial systems  were constructed using variables of locational grouping, optimum size, and indices of inter-firm cooperation.  124  Still, this work highlights a growing recognition of how more  detailed studies of the geography of enterprise can reveal important facets of spatial economic restructuring.  125  Fei, in two important papers, shows how little attention has  been given to the way in which economic and administrative reforms have fundamentally changed enterprise behaviour, their "spatial evolution", and implications for industrial location and regional economic development.  126  Ultimately, he suggests  how understanding "micro-mechanisms" and processes that determine enterprise location and growth provide insights into the evolution of "macro-regional economic systems".  127  Even old-school geographers, while still clinging to conventional notions of  agglomeration and external economies, recognize that China's evolving industrial complexes are emerging "organically" as the product of a multiplicity of small-scale local circumstances.  128  In terms of development praxis others identify the need to  60 carefully consider local "typicality" (leixingxing) - all the characteristics that determine internal mechanisms of change.  129  This perspective is evident in discussions of the processes and mechanisms that underlie rural industrial development and the numerous models that have been constructed, based on particular local circumstances, to explain variations in rural change.  130  Furthermore, as we saw earlier, official policy pronouncements are  embodied in catchy slogans: "leave the land, but not the countryside" (litu bulixiang); and "enter the factory, but not the city" (jinchang bujincheng). Fortunately, conceptual work on the rural transformation is more sophisticated. Most studies refer to certain ideological and historical preconditions for rural industrial development. While there have been many traumatic vacillations in policies affecting the countryside the importance of rural non-agricultural activities, fostering demand for local production, and reducing rural-urban differences has remained relatively unchanged.  131  This is  consistent with historical circumstances, especially in the more advanced agricultural regions, that underlie the emergence of rural industrial enterprises. Along with increased administrative autonomy are issues of ownership and the broadening mandate of community level governments to provide local social infrastructure and opportunities for employment. With the release of a "great inner force" then, rural industrial development was neither unusual nor unexpected in the Chinese context.  132  Theoretical work on rural industrialization has also provided an alternative perspective on the role of urban areas on regional development. Extended periods of highly productive agriculture, and the intensity of the local transactional environment, have been at least as important in influencing the rural transformation as big cities.  133  Thus, despite the forced separation of rural and urban in China's planned economy, the intensity of development in the lower Yangzi delta has remained relatively high. R u r a l industry here is said to have extended the spatial division of labour within the countryside while at the same time linking rural areas to the wider division of l a b o u r . The most important insight, however, is the way in which the development of rural  134  61 industry has served as a catalyst for changes in the structure of local economic organizations and management.  135  Yet this theoretical perspective, which is useful in  emphasizing the importance of local circumstances, is contradicted by numerous studies which directly linked the rural transformation to the proximity of big cities. Such research usually referred to evidence of urban-rural industrial linkages in the context of elegant regression analyses which established the direction and importance of the influence of large urban areas.  136  Other work by G u et al who refers to the evolution of the "urban fringe" in the context of the "Chinese megalopolis" or from Zhou with his "metropolitan interlocking regions", also links issues of regional development to processes of urbanization.  137  Thus, spatial economic patterns, and the mechanisms and evolution of regional development are conceptualized in terms of urban expansion or decentralization, converged urban systems, urban fringes and corridors, or the extended metropolis.  138  While some have linked this urban theoretical bias to a retreat from the universal platitudes of egalitarian spatial planning in a partially reformed command economy context, I am more inclined to relate it to a kind of pragmatic orthodoxy deeply rooted in pre-reform methods for determining policy and planning priorities.  139  This is  consistent with the view of some Chinese scholars who tend to emphasize macroeconomic regulation of regional economic growth with all the implications for control and management of large cities.  140  The general pro-urban/pro-rural conceptual divide in Chinese theories of development has been challenged by scholars and bureaucrats who are seeking a middle-level framework that aims for better economic integration in the settlement hierarchy.  141  When such ideas were first broached in the mid 1980s big cities were  taken as the "nucleus" (hexiri), medium sized cities as the "link" (niudai), and rural small towns as the "cell" (xibao).  A s Kirkby pointed out, however, "the actual mechanisms  142  and divisions of economic and political responsibility necessary to the workings of such a system are left unstated".  143  Conceptually speaking though, there have been some  62 excellent studies over the last few years that have fleshed out notions of a theoretical middle ground and which hint at a path through the apparent impasse. These are introduced and developed in the next section. 2.6  Rural and Urban in China's Regional Development: Seeking a Middle Ground The need for a new theoretical middle ground is underscored by the inability of  current policy and planning strategies to respond effectively to a rapidly evolving space economy. L i n et al refer to the dramatic impact of development in highly productive agricultural regions, while Miao highlights the problems of coordinating urban and rural industries.  144  Others have focused on the "fusion" between town and country and  issues of urbanization in economic development that illustrate how conventional theoretical frameworks cannot account for certain elements of regional change.  145  The  mechanisms and processes which underlie this change have been viewed in terms of a "rural urbanization" (xiangcun chengshihua or sometimes nongcun chengzhenhua) that attempts to accommodate characteristics not commonly identified in the conventional models.  146  Chief among these are the ownership and management structures of rural  enterprises, and the community level motivation and means for establishing such enterprises. These features are also linked to the geographical and historical circumstances of local development. Moreover, rather than emphasizing the role of big cities, this work tends to examine the importance of linkages between the countryside and cities and towns from a rural perspective. Some have even speculated that it is the intensity of the rural transformation that has in fact stimulated many of the urban state-run industrial reforms.  147  A t the core of this perspective is recognition of a regional countryside and  town system exhibiting a "unified", "integrated", or "organic" spatial economic development.  148  Empirical evidence from the lower Yangzi delta discussed in the  following chapters suggests it is neither clearly rural nor urban in character and it is certainly not urban initiated and driven. Several Chinese scholars and bureaucrats have  63 conceptualized the processes of this spatial economic restructuring in the term chengxiang yitihua - literally translated as "city-countryside integration".  149  However,  the deeper meaning implies an organic whole ~ a kind of ecosystem approach (or ruralurban symbiosis) in which there is a transformation of the countryside through the opening up and linking of smaller cities and towns to the surrounding regions. Although chengxiang yitihua is sometimes associated with the specificities of rural-urban economic and or commercial linkages, I invoke its underlying meaning as a platform for building a theoretical middle ground. Interestingly, few Chinese studies have attempted to merge the processes conceptualized with the array of empirical findings that have identified key socio-spatial features of China's rural transformation. One notable exception was an internally published report from the Department of Geography at Nanjing Normal University.  150  The authors developed a schematic to  indicate the relationship between several key features in the development of rural urbanization in the lower Yangzi delta (see Figure 2.1). Ignoring for the moment that the model overlooks certain elements such as the reproduction of capital and institutional parameters, it also fails to elaborate upon the geographical processes which underlie the features illustrated. While the urban-rural distinction is once again emphasized the conceptual and morphological character of rural urbanization remains unclear. The model does, however, give some sense of how the features shown relate to each other and provides a useful preliminary framework for understanding regional development in the delta. While many of the dimensions of this development are set in regional or national urban centres or even abroad, many local phenomena have an impact at these larger scales. Furthermore, understanding the complexity of the interactions between the local and extra-local is crucial to realistic analysis. Only by examining the structure of this transactional environment and its institutional framework from a local perspective can we begin to challenge the inexorability of the transition to conventional urban forms and the (imminent) demise of rural organization. This approach is largely  64  Figure 2.1 RURAL INDUSTRIALIZATION A N D T H E DEVELOPMENT OF RURAL URBANIZATION  Increased Agricultural Production  Potential  <  Rural . Industrialization  Surplus Labour  UJ  al  <  Surplus Production  •3£  i- o .  U «  -_3 <2  E  <u co  •5 E &8  •3 =  ||| CO ET3  E o  e  C O O Q .  E  Industrial Products  agricu ment  -  tn  Traditional Market Town  p  Regit  IS  [Agri. Products Function! Scale T  Spatial Transfer of Population  New Types of Small Cities and Towns  a. <  k  7  Life-Style and Needs CO  jnce  < m E  "o CO oC c o  1 -o  cn C o • X  CD  co  LU  Development of Rural Urbanization  Source: Human Geography Research Office, 1990, p. 59.  °  65 inspired by the work of Koppel who highlights the processes of rural transformation to provide an analytical alternative to the prevailing rural-urban dichotomies.  151  Thus,  interrelationships and linkages in a restructuring space economy are recognized as having an independent reality and are perceived neither as derivations nor reflections, but "... as representative and indicative of independent social [and spatial] facts - taking form, evolving, and varying for reasons attributable to urbanization and rural development; [and which take into account] other social, economic, and political processes ... that are idiosyncratic to the linkages in specific cultural and historical circumstances." 152  M c G e e provides empirical and conceptual depth to this perspective by explicating the emergence of "new regions of extended urban activity" surrounding large cities in A s i a .  153  A s with Koppel this viewpoint is less concerned with the distinction  between rural and urban, focusing instead upon the transactional environment as it affects settlement and the emergence of particular regions of economic activity. Referred to as "extended metropolitan regions" or "mega-urban regions", they usually incorporate large urban cores with well developed transportation links, peri-urban zones surrounding cities which are within a daily commute, and densely populated rural areas of mixed agricultural and non-agricultural activities called desakota regions.  154  These latter regions are characterized by processes that include rapid growth in labour intensive industry, services, and other non-agricultural activities, and an intense highly transactive environment especially in terms of the movement of people, commodities, and information. The resulting patterns of development that distinguish urban from rural are, therefore, more difficult to delineate. The middle ground proposed here consists of several overlapping elements. The first begins by rejecting the conceptual baggage associated with conventional notions of urban and rural. Mounting evidence of the existence of socio-economic patterns and processes which are, conceptually and spatially, between the two forces reconsideration of such notions. The second refers to the need for a middle level geography of  66 production that balances broader theories of location and the production of industrial space with local experiences and circumstances. This can be achieved by considering the activities of industrial enterprises in their institutional and transactive contexts. The third requires closer examination of local level forces in relation to, but not ancillary to, global and national exogenous (usually urban centred) influences. I will unabashedly privilege the local in this part of the analysis to counter the prevailing focus on large scale issues and theoretical frameworks which have usurped much of the intellectual inquiry. The fourth element aims for a middle ground between theory and praxis which can provide a framework for meaningful analysis and a basis for planning and policy strategies for the management of China's mega-urban regions. 2.7  Mega-Urbanization in the Lower Yangzi Delta: Enterprise Location and the Reconstitution of Local Space The key elements of the theoretical deliberations in the previous six sections can  now be summarized in the context of three specific working hypotheses. First, the patterns and underlying processes and mechanisms of regional development in the lower Yangzi delta are fundamentally linked to intensely localized characteristics and circumstances within the wider Chinese space economy. Second, industrialization and the morphology of spatial economic restructuring in the delta are best understood and explained in terms of the complex interactions and interrelationships which constitute the transactional environment. Third, external economies, the dynamics of agglomeration, and the role of large cities and other exogenous forces, while important, are less significant in determining the character of local and regional transformation in the delta than are endogenous forces. These hypotheses are presented separately here to cope with the analytical imperatives of several overlapping concepts and phenomena. Understanding spatial economic transformation in the delta requires transcending the conventional wisdom of urban and regional development and the de jure organizational frameworks which  67 institutionalize such concepts. While cities are commonly viewed as the nexus of growth, regional development in the lower Yangzi delta appears to be more complex than merely in terms of its purported dependence upon urban centred external forces. The critical parameters and the vitality of regional development in the delta were in fact centred within the multitude of localities, making large cities relatively less important. Moreover, while the primacy of agricultural activities (especially in terms of basic food production) has not diminished, new roles in industrial production and other non-agricultural activities have emerged that create new locally specific opportunities for accumulation making rural areas the foci for socio-economic and institutional transformation. Elements of this rural transformation are captured in Figure 2.2 which incorporates the three working hypotheses into a hypothetical model of megaurbanization for the lower Yangzi delta. The model is situated on a meso-scale level of abstraction between concrete local factors affecting the location of enterprises and the production of industrial space, including locality and the transactional environment, and macro-scale economic reforms, external economies and the dynamics of agglomeration. Clearly, however, I have chosen to emphasize small-scale micro-factors and their administrative and institutional parameters, which largely determine the character of the rural transformation. While the diagram is divided into five main components for discussion and explanation it is important to consider the model in its entirety. A t its core, embedded within the transactional environment, are the fundamental elements which link locality and the geography of enterprise. Thus, the key features of diversification and commercialization of highly productive agriculture and the location of industrial enterprises are linked by locally determined administrative and institutional parameters, and via revitalized local marketing and commercial functions and various transactional networks. These are mediated through towns and villages and the transactional activities of enterprises. This set of central processes (organized vertically in the centre of the model) link the historical and socio-cultural circumstances of  68  Figure 2.2 POST-REFORM RURAL TRANSFORMATION AND MEGA-URBANIZATION IN THE LOWER YANGZI DELTA  tn X OC  O  o  Z  o  i tu  SE o  —I  <9 (3  < tn u  < tn ui O u^  3  Ag = Agricultural  69  V  locality to the production activities of contemporary industrial enterprises. In the top part of the diagram the transformation of agriculture, stimulated by the economic reforms introduced i n Chapter One and which are discussed i n more detail in subsequent chapters, has led to the emergence of other local factors in the context of the place in which change occurs. These are manifest in and influenced by space, especially through issues of ownership (as part of the transactional environment) and the control of land, which are fundamentally linked to the location and production of industrial space. This brings us to the lower part of the diagram. While production activities are partly responsible for the reproduction of capital, community level governments who own and manage enterprises largely through township and village collectives, are also able to influence the monetary practices of local banks, economic entities, and individuals for the benefit of local production activities, and local transport, information, and social infrastructure. Procurement and marketing, access to technology, expertise and capital, and production activities, while also linked to external economies partly as a result of "open door" economic reforms, are profoundly mediated by locally determined administrative and institutional parameters.  155  The same is also  true of externally inspired regional infrastructural development. The potential impact of local production activities upon the wider space economy and the emergence of industrial organizations, such as the large number of designated industrial areas and special economic and technological development zones are also taken into account. The transactional environment is mediated through a number of formal and informal administrative and institutional parameters. These are linked to bifurcation of the role of local governments both as community administrators and as owners and managers of non-agricultural enterprises. Although this bifurcation has deep historical and contemporary roots, it is primarily the disengagement of the central government from local administration, that occurred with economic reforms, that has sharply enhanced the dual role of local governments. Within the transactional environment, processes of representation embedded in various administrative and institutional  70 structures allow for the local mobilization of indigenous and external means of production. These locally determined representations manipulate the transactional network, sometimes creating new ones, in order to maximize community-based production opportunities. In the absence of a meaningful legal and regulatory framework localities are free to exploit all means at their disposal to achieve this objective. Local actors, often with apparently conflicting roles, exercise their influence through these intensely localized economic and bureaucratic structures. This helps to explain the intensity and diffuse nature of local transactional networks, within structures and across space, and accounts for the lesser importance of linkages with external economies and the dynamics of agglomeration. Ultimately, however, I am less concerned with the simple unfolding of the social relations of production across space than with the means whereby local actors, through their transactional networks, construct industrial space by utilizing certain administrative and institutional structures. These transactional networks and institutional structures must be carefully analysed in order to understand and explain enterprise location and the reconstitution of local space in the lower Yangzi delta. The hypothetical model of mega-urbanization proposed here, provides a preliminary checklist of the main elements which underlie spatial economic restructuring in the lower Yangzi delta. In the four chapters which follow, detailed analyses of each of these elements will be undertaken to evaluate the three specific working hypotheses and to assess the integrity of the entire model.  71 Notes 1.  2.  Douglass, M . (1995a) Viewpoint ~ Bringing culture in: Locality and global capitalism in East Asia. Third world planning review 17 (3), p. viii; McGee, T. G. (1995b) Geography and development: Crisis commitment and renewal. Cahiers de geographie du Quebec 39 (108); Massey, D. (1993b) Questions of locality. Geography 78 (2), p. 147. Ettlinger, N. (1994) The localization of development in comparative perspective. Economic  geography 70 (2), p. 144.  3.  Massey, D. (1993b) op. cit., p. 147.  4.  Little, D. (1989) Understanding peasant China: Case studies Haven: Yale University Press, p. 25; Massey, D. (1993b) Ibid.  5.  6.  in the philosophy of social sc  Peet, R. (1994) Review of "Beyond the impasse: New directions in development theory."  AAAG 84 (2), p. 339. Peet does refer to one study which undermines this claim, p. 340. Edwards, M . (1993) How relevant is development studies? In Schuurman, F. J. (Ed.) Beyond the  impasse: New directions in development theory. London: Zed.  7.  Schuurman, F. J. (Ed.) (1993a) Beyond the impasse: New directions in development theory.  8.  Booth, D. (1985) Marxism and development sociology: Interpreting the impasse. World development 13 (7); Corbridge, S. (1986) Capitalist world development: A critique of radica  Zed.  development geography. London: Macmillan. 9.  Booth, D. (1993) Development research: From impasse to new agenda. In Schuurman, F. J. (Ed.) op. cit.  10.  See especially: Forbes, D. & Thrift, N. (Eds.) (1987) The socialist third world: Urban development and regional planning. Oxford: Blackwell; Post, K. & Wright, P. (1989) Socialism and  underdevelopment. London: Routledge.  World development  11.  Vandergeest, P. & Buttel, F. (1988) Marx, Weber and development sociology.  12.  Booth, D. (1985) op. cit., p. 762. Also cited in Schuurman, F. J. (1993b) Introduction: Development theory in the 1990s. In Schuurman, F. J. (Ed.) op. cit., p. 13.  13.  16(6).  Toye, J. (1993)  Dilemmas of development: Reflections on the counter-revolution in devel  economics. Oxford: Blackwell.  14.  Corbridge, S. (1989) Marxism, post-Marxism, and the geography of development. In Peet, R. & Thrift, N. (Eds.) New models in geography: The political-economy perspective. London: Un  15.  Ibid. pp. 238 - 245.  16.  Corbridge, S. (1991) Third world development. PIHG 15 (3).  17.  Corbridge, S. (1989) op. cit., p. 247.  18.  Schuurman, F. J. (1993b) op. cit., p. 12.  72 19;  20.  World Bank (1993a) pp. 84 - 86.  The East Asian miracle: Economic growth and public policy. Oxford: OU  Marton, A . M . (1995a) Review of "The East Asian miracle: Economic growth and public policy".  Environment and planning A 27 (4).  21.  World Bank (1993a) op. cit., p. 11.  22.  Ibid., p. v - vii.  23.  For more critique of the counter-revolution in development see: Corbridge, S. (1990) Development studies. PIHG 14 (3); Toye, J. (1993) op. cit.  24.  Corbridge, S. (1991) op. cit.; Forbes, D. & Thrift, N. (Eds.) (1987) op. cit.; Sidaway, J. D. & Simon, D. (1990) Spatial policies and uneven development in the Marxist-Leninist states of the third world. In Simon, D. (Ed.) Third world regional development: A reappraisal. London: Paul Chapman.  25.  Forbes, D . & Thrift, N. (Eds.) (1987) op. cit., p. 6. See also Sidaway, J. D. & Simon, D. (1990) op. cit., p. 30.  26.  One example is the model of urbanization advanced by Murray, P. & Szelenyi, I. (1984) The city in  transition to socialism. IJURR 8 (1).  27.  For excellent reviews of the traditional means of controlling and managing population mobility and recent changes see: Chan, K. W. (1994a) Cities with invisible walls: Reinterpreting urbanization post-1949 China. Hong Kong: OUP; Kirkby, R. J. R. (1985) Urbanization in China: Town and country in a developing economy 1949 - 2000AD. London: Routledge. Very recently, however, economic liberalization has led to increasing mobility of the population by dissolving long-standing social relations. This has in turn weakened the state's ability to micro-manage Chinese society.  28.  Sidaway, J. D. & Simon, D. (1990) op. cit., p. 33.  29.  Ibid., p. 32.  30.  Here I conflate "political" and "institutional". Notwithstanding the voluminous literature that debates the nature of "civil society" in China, political will refer to the nature of power and control particularly in terms of the relations across and through territorial and sectoral administrative structures. These structural relationships are largely institutionalized (formally and informally) and, as will be demonstrated later, have profound implications for the nature of economic (and social) relations of production. Unless otherwise stated, however, these terms will not refer to cultural and social institutions.  31.  McGee, T. G. (1991b) Presidential address: Eurocentrism in geography - The case of Asian urbanization. The Canadian geographer 35 (4); Sidaway, J. D. & Simon, D. (1990) op. cit., p. 32; Seater, D . (1989) Peripheral capitalism and the regional problematic. In Peet, R. & Thrift, N. (Eds.) op. cit., p. 267; Watts, M . J. (1989) The agrarian question in Africa: Debating the crisis.  PIHG 13 (1), p. 32. 32.  Watts, M . J. (1989) op. cit., p. 32.  33.  Schuurman, F. J. (1993b) op. cit., pp. 16, 32.  34.  Booth, D . (1993) op. cit.; Edwards, M . (1993) op. cit.  73 35.  A n example of how this might be undertaken is advanced in Edwards, M . (1989) The irrelevance of development studies. Third world quarterly 11 (1).  36.  Edwards, M . (1993) op. cit., p. 83.  37.  Ibid., p. 84.  38.  Booth, D . (1993) op. cit., p. 55.  39.  Corbridge, S. (1993) Marxisms, modernities, and moralities: Development praxis and the claims of  distant strangers. Environment and planning D: Society and space 11 p. 467.  40.  Peet, R. (1994) op. cit., p. 340.  41.  Ibid., p. 341; Escobar, A. (1984 - 85) Discourse and power in development: Michel Foucault and the relevance of his work to the third world. Alternatives 10; Peet, R. & Watts, M . (1993) Introduction: Development theory and environment in an age of market triumphalism. Economic geography 69 (3); Watts, M . (1993) Development 1: Power, knowledge, discursive practice. PIHG 17 (2).  42.  See for example: Peet, R. (1993) Reason, space, modernity: The future of development geography.  Association ofAmerican Geographers annual meeting, Atlanta. In a more recent, though distinctl non-Foucauldian critique, McGee highlights the "considerable disillusionment" within the geography of development debates. He draws particular attention to the assumption that classical and Marxist theories of economic development would bring about the "condition of 'development'". See: McGee, T . G . (1995b) op. cit., pp. 527-528.  43.  Ibid.  44.  Ibid., p. 8; Chambers, R. (1983)  45.  Schuurman, F. J. (Ed.) (1993a) op. cit.  46.  Brown, L. A . (1988) Reflections on third world development: Ground level reality, exogenous forces, and conventional paradigms. Economic geography 64 (3).  47.  Ibid., pp. 267-268.  48.  In the context of this study, unless otherwise stated, "local" will always refer to small scale usually "rural" areas and is meant to be distinguished from what I later call "city driven" factors. This is necessary because of the now standard dualisms frequently assumed in the conventional models that I wish to challenge here.  49.  Brown, L. A. (1988) op. cit., p. 263.  50.  Ibid.  51.  Notions of how urban and rural are defined and how this influences the theoretical frameworks for development are discussed later. As will become clear, the precise definitions are less important than underlying assumptions regarding how they relate to each other.  Rural development: Putting the last first. Harlow: Longman.  74 52.  Notable exceptions include: Baker, J. & Pedersen, P. O. (Eds.) (1992) The rural-urban interface in Africa: Expansion and adaption. Uppsala: The Scandinavian Institute of African Studies; Cronon, W. (1991) Nature's metropolis: Chicago and the great west. New York: Norton; Funnell, D. C. (1988) Urban-rural linkages: Research themes and directions. Geografiska annaler 70 B (2); Potter, R. B. & Unwin, T. (Eds.) (1989) The geography of urban-rural interaction in developing count London: Routledge.  53.  Potter, R. B. & Unwin, T. (Eds.) (1989) op. cit., p. 11.  54.  Lipton, M . (1977) Temple.  55.  Rondinelli, D. A. (1983) Secondary cities in developing countries: Policies for diffusing urba Beverly Hills: Sage; Rondinelli, D. A. (1985) Applied methods of regional analysis: The spatial dimensions of development policy. Boulder: Westview; Stohr, W. B. & Taylor, D. R. F. (Eds.) (1981)  Why poor people stay poor: A study of urban bias in world developmen  Development from above or below? The dialectics of regional planning in developing c Chichester: Wiley.  56.  Limited theoretical and policy insights are revealed in: Baker, J. & Pedersen, P. O. (Eds.) (1992) op. cit.; Potter, R. B. & Unwin, T. (Eds.) (1989) op. cit.  57.  Funnell, D. C. (1988) op. cit., p. 273.  58.  Ibid., p. 268.  59.  For just two examples which discuss the situation in southern Jiangsu from this perspective see: Lee, Y. S. (1991) Rural non-agricultural development in an extended metropolitan region: The case of southern Jiangsu. In Ginsburg, N., Koppel, B. & McGee, T. G. (Eds.) The extended metropolis: Settlement transition in Asia. Honolulu: University of Hawaii Press; Pannell, C. W. & Veeck, G. (1991) China's urbanization in an Asian context: Forces for metropolitanization. In Ginsburg, N., Koppel, B. & McGee, T. G. (Eds.) op. cit.  60.  A la Harvey, D. (1985) The urbanization of capital: Studies in the history and theory of cap urbanization. Oxford: Basil Blackwell; see also Funnell, D. C. (1988) op. cit.  61.  Preston, D. (1975) Rural-urban and inter-settlement interaction: Theory and analytical structure. Area 7 (3); Rondinelli, D. A. (1983) op. cit.; Unwin, T. (1989) Urban-rural interaction in developing countries: A theoretical perspective. In Potter, R. B. & Unwin, T. (Eds.) op. cit.  62.  Koppel, B. (1991) The rural-urban dichotomy reexamined: Beyond the ersatz debate? In Ginsburg, N., Koppel, B. & McGee, T. G. (Eds.) op. cit., p. 47.  63.  Corbridge, S. (1982) Urban bias, rural bias, and industrialization: An appraisal of the works of Michael Lipton and Terry Byres. In Harris, J. (Ed.) Rural development: Theories ofpeasant economy and agrarian change. London: Hutchinson.  64.  This approach is inspired by: Shue, V. (1988) The reach of the state: Sketches of the Chinese bo politic. Stanford: Stanford University Press; also see: Leung, C. K. (1993) Personal contacts, subcontracting linkages, and development in the Hong Kong - Zhujiang delta region. AAAG 83 (2), p. 297.  65.  Storper, M . & Walker, R. (1989) The capitalist imperative: growth. New York: Basil Blackwell, p. 1.  66.  Ibid.  Territory, technology, and industr  75 67.  Gregory, D. (1981) Alfred Weber and location theory. In Stoddart, D. R. (Ed.) Geography, ideology and social concern. Oxford: Basil Blackwell; Harvey, D. (1985) op. cit.; Massey, D. (1973) Towards a critique of industrial location theory. Antipode 5 (1); Sayer, A . (1985) Industry and space: a sympathetic critique of radical research. Environment and planning D: Society and space 3; Smith, (1989) Uneven development and location theory: Towards a synthesis. In Peet, R. & Thrift, N. (Eds.) op. cit.  68.  For an excellent summary see: Smith, N. (1989) op. cit., pp. 145 -152.  69.  Ibid., p. 155.  70.  Massey, D. (1979) In what sense a regional problem? Regional studies 13 (2); Massey, D. (1984) Spatial divisions of labour: Social structures and the geography ofproduction. New York  ;  See also Smith, N. (1985) op. cit.  71.  Massey, D . (1993a) Classics in human geography revisited: 'Spatial divisions of labour', author's  72.  Massey, D . (1984) op. cit., p. 120.  73.  Martin, R. (1993) Classics in human geography revisited: 'Spatial divisions of labour', commentary 1. PIHG 17 (1); Sayer, A . (1985) op. cit.; Smith, N. (1989) op. cit.  74.  Smith, N. (1989) op. cit., pp. 154 -155.  75.  Gregory, D . (1989) Areal differentiation and post-modern human geography. In Gregory D. & Walford, R. (Eds.) Horizons in human geography. London: MacMillan.  76.  Storper, M . (1987) The new industrial geography, 1985 -1986.  77.  Scott, A. J. (1987) Industrial organization and location: Division process. Department of Geography, U C L A . Cited in Ibid., p. 588.  78.  response. PIHG 17 (1).  Urban geography 8 (6), p. 593.  Walker, R. (1988) The geographical organization of production systems.  D: Society and space 6 (3).  of labour, thefirm,and spa  Environment and planning  79.  See especially: Sayer, A. & Walker, R. (1992) The new social economy: Reworking the division labour. Cambridge: Blackwell; Scott, A. J. (1988a) Flexible production systems and regional development. IJURR 12 (2); Storper, M . & Scott, A. J. (Eds.) (1992) Pathways to industrialization and regional development. London: Routledge; Storper, M . & Walker, R. (1989) op. cit.  80.  Rees, J. (1989) Regional development and policy. PIHG 13 (4); Rees, J. (1992) Regional development and policy under turbulence. PIHG 16 (2).  81.  Walker, R. (1989) A requiem for corporate geography: New directions in industrial organization, the production of place and the uneven development (sic). Geografiska annaler 71B (1).  82.  Dicken, P. & Thrift, N. (1992) The organization of production and the production of organization: Why business enterprises matter in the study of geographical industrialization. TIBG 17, p. 288.  83.  Storper, M . (1987) op. cit., p. 592.  84.  Walker, R. (1988) op. cit., p. 385.  85.  Scott, A . J. (1986) Industrialization and urbanization: A geographical agenda. AAAG 76 (1).  o  76 86.  Pudup, M . B. (1992) Industrialization after (de)industrialization: A review essay. Urban geography 13 (2), p. 194; See also Jacobs, J. (1984) Cities and the wealth of nations: Principles of econom New York: Random.  87.  Cronon, W. (1991) op. cit.; Pred, A. R. (1966) The spatial dynamics of U.S. urban industrial grow 1860 -1914. Cambridge: MIT Press; Pudup, M . B. (1992) op. cit., p. 194; Scott, A . J. (1988b) Metropolis: From the division of labour to urban form. Berkeley: University of California Press; Walker, R. (1988) op. cit., p. 408.  88.  Pudup, M . B. (1992) op. cit, p. 195.  89.  Kwok, R. Y. W. (1992) Urbanization under economic reform. In Guldin, G. E . (Ed.) Urbanizing China. Westport: Greenwood, p. 73; Pannell, C. W. (1992) The role of great cities in China. In Guldin, G. E . (Ed.) op. cit., p. 36.  90.  Lovering, J. (1990) Fordism's unknown successor: A comment on Scott's theory of flexible accumulation and the re-emergence of regional economies. IJURR 14 (1); Lovering, J. (1991) Theorizing postfordism: Why contingency matters (a further response to Scott). IJURR 15 (2); Scott, A . J. (1991a) Flexible production systems: Analytical tasks and theoretical horizons - A reply to Lovering. IJURR 15 (1); Scott, A. J. (1991b) A further rejoinder to Lovering. IJURR 15 (2).  91.  O hUallachain, B. (1993) Industry geography. PIHG 17 (4); Sayer, A . (1989) Postfordism in question. IJURR 13 (4); Sayer, A. & Walker, R. (1992) op. cit.  92.  Phelps, N. A . (1992) External economies, agglomeration andflexibleaccumulation. TIBG 17 (1).  93.  Leung, C. K. (1993) op. cit.  94.  The most recent survey of such work is found in Chan, K. W. (1994a) op. cit., pp. 7 -18; Yan, X. P. (1995) Chinese urban geography since the late 1970s. Urban geography 16 (6); See also: Geographical Society of China (Ed.) (1990) Recent development of geographical science in China Beijing: Science Press, especially chapters by: Li, W. Y., pp. 197 - 203; Shen, D. Q. & Cui, G. H., pp. 204 - 213; and Wu, C. J., pp. 1 -17; Kirkby, R. J. R. (1985) op. cit.; Pannell, C. W. (1980) Geography. In Orleans, L. A. (Ed.) Science in contemporary China. Stanford: Stanford University Press; Pannel, C. W. (1990) China's urban geography. PIHG 14 (2); Yeung, Y. M . & Zhou, Y. X. (1991) Human geography in China: Evolution, rejuvenation, and prospect. PIHG 15 (3).  95.  Wu, C. J. (1990) Territorial management and regional development. In Geographical Society of China (Ed.) op. cit.  96. o  Fei, X. T. (1984) Xiao chengzhen da wenti (Small towns, a big issue). Liaowang {Outlook). January; Fei, X. T. (Ed.) (1986) Small towns in China: Functions, problems and prospects. Beijing: New World Press.  97.  Shen, D . Q. & Cui, G. H . (1990) Urban geography and urban planning. In Geographical Society of China (Ed.) op. cit., p. 210.  98.  Feng, Y. F. (1983) Fazhan xiao chengshi shi woguo chengshihua weiyi zhengque de daolu ma? (Is the development of small cities China's only correct road to urbanization?). JJDL 3 (2); Li, B. R. (1983) Woguo chengzhenhua daolu wenti de taolun (A discussion on China's road to urbanization). CSGH 6 (2); Ma, Q. Y. (1983) Woguo chengzhenhua de tedian ji fazhan qushi de chubu fensan (A preliminary analysis of characteristics and development trends of China's urbanization). JJDL 3 (2).  99.  Ibid.  77 100.  Bai, D. M . (1981) Webring cheng nengqi kongzhi shiqu guimo de zuoyong ma? (Are satellite towns a useful means of controlling the population in urban districts?). Jianzhu xuebao  (Architectural journal) (4).  101.  Feng, Y. F. (1983) op. cit.  102.  Ibid., p. 138.  103.  Hu, B. L. (1991) Industrial guidance Bangkok. Unpublished manuscript.  104.  Ibid., pp. 4, 7 - 8.  105.  Ibid., p. 11.  106.  Ibid., p. 12.  107.  Ibid., p. 4.  108.  Ibid., p. 18.  109.  Cai, Q. M., Zhao, P. S. & Jiang, M . K. (1991) Butong fazhan jieduan xiangzhen qiye yu nongye neizai guanxi chutan (Exploration of the internal relationship between rural enterprises and agriculture at different developing periods). JSJJTT (2); Xu, Y. M . & Wu, Q. (1990) Shinianlai xiangzhen qiye lilun yanjiude huigu (Ten year review of research on rural enterprise theory). JSIITT (10).  110.  Cai, Q. M., Zhao, P. S. & Jiang, M . K. (1991) op. cit., pp. 42 - 43; Xu, Y. M . & Wu, Q. (1990) op. cit., pp. 18 -19.  111.  Cai, Q. M., Zhao, P. S. & Jiang, M . K. (1991) op. cit., p. 39; He, B. S. (1991) Jiangsu nongcun feinonghua fazhan yanjiu (Research on rural non-agricultural development in Jiangsu  in the urbanization process. Asian Institute of Technology,  Shanghai People's Publisher, pp. 30 - 66.  112.  He, B. S. (1991) op. cit., pp. 92 -105.  113.  Lin, F. R., Wang, Z. D. & Tang, Q. L. (1992) Guanyu nongye quyu zonghe kaifa wogan lilun wentide tantao (Inquiry into certain theoretical issues of the comprehensive development of rural areas).  JJDL12(3). 114.  Huang, M . D. (1993) Chengzhen tixide jiegou yu yanhua: Lilun fanxiyu shizheng yanjiu (Th and evolution of urban systems: Theoretical analysis and case study). Unpublished Ph.D. East China Normal University, pp. 81 - 82.  115.  Ibid.  116.  Shi, Y. S. (1992) Xiangcun dilixue fazhande huigu yu zhanwang (Development of rural geography: Retrospect and prospect). DLXB 47 (1); Shuai, J. P. (1993) Woguo xiangzhen qiye fada diqude fazhan dui nongyequ weide yinxiang (Influence of China'sflourishingrural enterprise development on agricultural areas). DLYJ12 (3); Yang, C. Y. (1992) Xiangzhen gongye chengzhang yu fazhan zhanlue yanjiu (Research on the growth and development strategy of rural industry). JJDL 12 (1).  117.  Li. W. Y. (1990) Recent development of industrial geography in China. In Geographical Society of China (Ed.) op. cit.; Lu, D. D. (1990) Zhongguo gongye bujude lilun yu shijian (China's distribution: Theory and practice). Beijing: Science Press.  industr  78 118.  Gong, F. X. (1992) Zhongguo qinggongye jingji wenti tansuo {Exploring the problem of China's light industrial economy). Beijing: Economic Science Press; Lu, D. (1992) Zhongguo gongye jingjide gaige yu fazhan {China's industrial economic reform and development). Beijing: Economic Management Press; Zhou, S. L. & Chen, J. G . (1992) Zhongguo gongyede fazhan yu gaige {China's industrial development and reform). Beijing: Economic Management Press.  119.  Yang, W. Y. (1992) Jingji dilixue: Kongjian jingjixue yu quyu kexue (Economic geography: Spatial economics and regional science). DLXB 47 (6).  120.  Lu, D . D. (1991) Woguo gongye buju yanjiu qude quanmian zhongda jinzhan (Seeking overall significant progress in China's industrial location research). JJDL 11 (2); Shi, Q. W. & Wu, W. (1992) Gongye buju: Zai buju lun (Industrial location and relocation theory). JJDL 12 (2); Yang, K. Z. (1993) Zhongguo diqu gongye jiegou bianhua yu quji chengzhang he fengong (Changes in China's regional industrial structure and its effect on regional growth and specialization). DLXB 48 (6).  121.  Li, W. Y. (1990) op. cit., p. 202; Peng, Q. (1993) Chanye jiegoude chengzhang yu jingjiqude kuozhang (Development of industrial structure and the expansion of economic regions). DLYJ12 (1), p. 80; Yang, W. Z. (1991) Chanye jiegou, chanye buju, chanye zhengce yitihua wenti (Industrial structure, industrial location, industrial policy: Problems of integration). JJDL 11 (1).  122.  Li, W. Y. (1990) op. cit., p. 199; Liang, R. C. (1992) Lun gongyequde xingcheng yu fazhan (On the formation and development of industrial regions). DLKX12 (4); Zhou, Y. X. & Yang, Q. (1995) Geographical analysis of the industrial economic return of Chinese cities. Urban geography 16 (6).  123.  Lin, J. B. (1993) Guanyu kaifaqude jige wenti (On the problems of development zones). JJDL 13 (2); Special Zones Office of the State Council (1991) Zhongguo yanhai chengshi jingji jishu kaifaqu {China's coastal municipal economic and technological development zones). Beijing: State Council; Wei, X. Z . (1991) Guanyu gaojishu chanye jiqi yuanqu fazhande yanjiu (Research on the development of high technology industrial parks). JJDL 11 (1).  124.  Li. W. Y. (1986) Developing regional industrial systems in the Chinese People's Republic. In Hamilton, F. E . I. (Ed.) Industrialization in developing and peripheral regions. London: Croom Helm; Liang, R. C. (1989) Gongyequde dengji leixing jiqi jiegou tezhengde tantao (Inquiry into the categories and structural features of industrial areas). DLXB 44 (1); Shen, X. P. (1987) Shilun gongyequ gongye qiye chengzu bujude jingji xiaoguo he zuijia guimade queding (Discussion on the economic effect of the location of industrial enterprises and the determination of optimum size). DLXB 42 (1).  125.  Huang, M . D. (1991) Xiandai qiye zuzhide quwei xuanze jiqi kongjian yingxiang (Organization of location choice of modern enterprises and their spatial influence). JJDL 11 (4); Li, X. J. (1991) Guanyu gongsi dili yanjiude jige wenti (Some problems in the research of firm geography). JJDL 11 (3); Li, X, J. (1993) Chanye lianxi yu nongcun gongyehuade qiye fenxi (Analysis of industrial linkages and rural industrialization enterprises). JJDL 13 (2).  126.  Fei, H . P. (1993a) Qiye yu quyu jingji xietiao fazhan yanjiu (Research on coordinating enterprises and regional economic development). JJDL 15,(3); Fei, H . P (1993b) Duochang qiye kongjian yanhua moshi yanjiu (Study of the models of the spatial evolution of multi-plant enterprises). DLKX 13 {4).  127.  Fei, H . P. (1993b) op. cit., pp. 329 - 330.  128.  Liang, R. C. (1992) op. cit., p. 343.  79 129.  Shen, D . Q. (1988) Changjiang liuyu kaifade zhengtixing, jieduaroring yu leixingxing (The periodicity, typicality and overall development of the Yangzi River valley). In Si, Y. F. (Ed.) Zhongguo  Kexueyuan Nanjing Dili yu Hupo Yanjiusuo jikan {Memoirs of the Chinese Academy of Nanjing Institute of Geography and Limnology, vol. 5). Nanjing: People's Press, pp. 99 -100. 130.  Sun, Y. S. & Lin, Y. Z. (1988) Nongcun chengzhenhuade guochengji qileixing (Processes and types of rural urbanization). JJDL 8 (1). The article refers to the Sunan (southern Jiangsu) model characterized by medium to large size township and village collective enterprises; the Wenzhou (located in coastal Zhejiang) model characterized by small scale individual and household enterprises; and the Subei (northern Jiangsu) model characterized by a lower level of rural development of mixed collective and individual and household enterprises.  131.  Xu, Y. M . & Wu, Q. (1990) op. cit.  132.  Ibid., p. 18.  133.  This perspective is reinforced by discussions with informants and my research collaborators.  134.  Huang, M . D. (1993) op. cit., p. 107.  135.  Ibid.  136.  He, B. S. (1991) op. cit., pp. 42 - 51; Ho, S. P. S. (1994)  Rural China in transition: Non-agricultural  development in rural Jiangsu, 1978 -1990. Oxford: Clarendon, pp. 61 - 63; Ning, Y. M . & Yan, Z. (1993) Woguo zhongxin chengshide bupingheng fazhan ji kongjian kuosande yanjiu (The uneven development and spatial diffusion of China's central cities). DLXB 48 (2), p. 103; Pang, X. M . (1992) A n analysis of the role of cooperations [sic] between urban and rural industries in the development of rural industry in China. The journal of Chinese geography 3 (1); Zeng, Z. G. & Lu, C. (1989) Jiangsusheng xiangcun jingji lebdngde chubu fenxi (Preliminary analysis of rural economic types in • ' Jiangsu Province). DLYJ8 (3); Zhang, F. B. (1988) Jingjiangxian duiwai jingji he shehuide diyu lianxi (External socio-economic regional interaction in Jingjiang County). In Si, Y. F. (Ed.) op. cit.  137.  Gu, C. L. et al (1993) Zhongguo dachengshi bianyuanqu texing yanjiu (Study of the special characteristics of the fringes of China's mega-cities). DLXB 48 (4); Zhou, Y. X. The metropolitan interlocking region in China: A preliminary hypothesis. In Ginsburg, N., Koppel, B. & McGee, T. G. (Eds.) op. cit. See also Wang, S. F. (1991) Chengzhenhua yu quyu jingji fazhan (Urbanization and regional economic development). JJDL 11 (1).  138.  Gu, C. L. & Chen, Z. G. (1994) Zhongguo dadushi kongjian zengzhang xingtai (Metropolitan spatial growth patterns in China). CSGH18 (6); Shen, D. Q. & Cui, G. H . (1990) op. cit., pp. 209 - 210.  139.  Kirkby, R. J. R. (1985) op. cit., p. 252.  140.  Hu, X. W. (1994) Jiaqiang duiquyu he chengshi fazhande guihua yu tiaokong (Strengthening the planning and readjustment of urban and regional development). CSGH 18 (2); Wang, S. F. (1991) op. cit.; Zhao, J. X. & Xu, C. X. (1992) Woguo quyu jingji zengzhang zhengcede lilun tantao (Theoretical inquiry into China's policy of regional economic growth). JJDL 12 (3).  141.  Wei, Y. H . (1993) Zhongguo quyu fazhan yanjiu: Zhuyao yiti he jinqi jinzhan (Research on China's regional development: Major topics and recent progress). JJDL 13 (4); and for an excellent discussion on a unified theory of "dual" structures see Yang, K. Z. (1992) Eryuan quyu jiegou lilunde tantao (Inquiry into theories of dual regional structures). DLXB 47 (6).  142.  Li, M . B. (1983) Woguo chengzhen fazhande zhanwang (China's urban development and prospects).  Chengxiang jianshe (Urban-rural construction) 12 (1), p. 16.  80  143. Kirkby, R. J. R. (1985) op. cit., p. 243. 144. Lin, F. R., Wang, Z. D. & Tang, Q. L. (1992) op. cit.; Miao, C. H. (1994) Quyu chengxiang gongyede xietiao fazhan zhanlue (Strategies for coordinating the development of urban and rural industries).  JJDL 14 (2).  145. Sui, Y. Z. (1992) Chengxiang ronghe xitongde SD dongtai guotu guihua chutan (Study of the SD territorial planning for the city-countryside fusion system). JJDL 12 (4); Wang, S. F. (1991) op. cit. 146. Chen, J. Y. & Hu, B. L. (1991) Chengshihua renkou yu laodongli wenti yanjiu (Urbanization, population and labour force research problems). In Zhang, B. C , Chen, J. Y. & Zhou, Y. X. (Eds.)  Zhongguo chengshihua daolu hongguanyanjiu {Macro-research on China's road to ur Harbin: Heilongjiang People's Press, p. 89.  147. Ibid., p. 92. 148. Ibid., p. 102; Liu, D. H. & Ma, M. (1991) Renwen dilixue yuqi duiguotu guihua lilun yanjiude gongxian ji fazhan wenti (Theoretical research contributions of human geography to the problems of development of territorial planning). JJDL 11 (4), pp. 16 -17. 149. Duan, X. M. (1993) Nongcun gaige he fazhan mianlinde xin wenti (Facing the new problems of rural reform and development) ZGNCJJ (8), p. 19; Zhang, L. J. (1989) Chengxiang yitihua zhilu {The path towards urban-rural symbiosis). Beijing: Rural Reading Materials Press.  150. Human Geography Research Office (1990) Jiangnan xiangcun chengshihua bijiao yu fazhan yanj  {Comparative and development research of rural urbanization in southern Jiangsu). N  Normal University, Department of Geography.  151. Koppel, B. (1991) op. cit. Koppel also shows how conventional analysis of rural and urban is anchored to other dichotomies: agricultural and industrial; socialist and capitalist; public and private; equity oriented and growth oriented; bottom-up and top-down. 152. Ibid., p. 48. 153. McGee, T. G. (1991a) The emergence of "desakota" regions in Asia: Expanding a hypothesis. In Ginsburg, N., Koppel, B. & McGee, T. G. (Eds.) op. cit., p. 3. 154. Ibid. The term is coined from the Bahasa Indonesian words for village {desa) and for town {kota) and is intended to invoke a conceptualization of landscapes neither clearly rural nor urban, but incorporating elements of the two. 155. By "open door" I refer not only to the policy of opening to the outside world, but also to the liberalization of the domestic economy referred to by the Chinese variously as "socialism with Chinese characteristics", or a "socialist market or commodity economy".  81 P A R T II  THE L O W E R YANGZI DELTA: HISTORY AND PRESENT CONDITIONS  This part of the thesis is comprised of Chapter Three which will focus on addressing the first research question about the nature of spatial economic restructuring in the lower Yangzi delta. The analysis presented here is framed by three objectives. The first is an explicit response to the perspective emphasized i n the previous two chapters which insisted that the importance of locality and the circumstances of place be more carefully considered in the processes of development. Hence, the first sections of Chapter Three will explore the critical historical conditions into and from which recent changes in the lower Yangzi delta have been imbricated. This provides the necessary context for the second objective which includes engaging in a detailed examination of the contemporary patterns of spatial economic transformation. Analysis of these patterns across the wider area of Jiangsu and Shanghai will also highlight the way in which the lower Yangzi delta is distinguished from the surrounding region. The third objective of this part of the thesis is to position Kunshan, firstly, within the most dynamic spatial economic restructuring that has occurred throughout the delta, and secondly, in relation to the elaboration in subsequent chapters about particular elements of mega-urbanization.  82 Chapter Three HISTORICAL GEOGRAPHY AND CONTEMPORARY PATTERNS OF C H A N G E This chapter begins with a survey of the natural environment and its relationship to the spatial economic structure of the lower Yangzi delta. Particular emphasis is placed upon commercialization and the growth and spatial proliferation of rural nonagricultural activities, and urban development through the late imperial and Republican periods. The second section briefly discusses the pre-reform (1949-1978) political economy which sets the stage for a detailed review of the reforms since 1978 and the resulting transformation of the rural economy. The fourth section examines recent spatial economic patterns in the delta and highlights specific features not adequately explained by the conventional wisdom of urban and regional development. The chapter concludes by introducing particular circumstances of spatial economic restructuring in Kunshan as a quintessential example of the local character of regional change. 3.1  The Lower Yangzi Delta From the Late Imperial Period Central to this study is the assumption that spatial economic patterns, the  structure of human agency, and the natural environment form a highly interdependent system. Focusing specifically on the lower Yangzi delta permits sustained analysis of the interrelationships among the various dimensions of this region as an integrated entity. While such an approach cannot adequately portray the full complexity of all of China, it can provide important insights into the processes of contemporary regional transformation. I will, however, endeavour to include some comparative references to the rest of China in general and to the northern part of Jiangsu Province (Subei) in particular.  83  3.1.1  Natural environment and spatial economic structure The lower Yangzi delta, including Nantong and Yangzhou in the north, Nanjing  and Zhenjiang in the west, the Lake Tai region of Changzhou, Wuxi and Suzhou to the south, and Shanghai to the east is roughly shaped as a shallow basin sloping towards Lake T a i (see Figure 3.1). The lowest part of the basin between Lake T a i and Kunshan is 1.7 to 3 metres above sea level. The surrounding area slopes upwards to the north and east to an average of 6 metres. Between Changzhou and Nanjing the land rises to between 20 and 50 metres with a few points reaching just over 400 metres. The immediate vicinity of the cities of Nanjing and Yangzhou are on the deltaic plain of the Yangzi River averaging between 2 to 10 metres above sea level. Rainfall across the region averages between 950 and 1150 millimetres per year with the greatest proportion falling between May and August. The average annual temperature is 15° to 16 °C with a mean temperature in January of 3°C and in July of 28°C, and an average of 230 frostfree days. Linked to the interaction between the Yangzi River and ocean tides, and 1  the sinking and inundation of the central part of the lower delta, the region has seen repeated expansion and contraction of lakeshore, rivers and drainage, and oceanfront. Competing forces of silting and erosion and flooding and reclamation created the delta's distinctive topography upon which emerged a regional space economy characterized by interaction and linkages among three broadly defined sub-regions.  2  The first is the high-lying outer ridge areas of present day east, central, and north Shanghai Municipality, the three northernmost counties of Suzhou and most of Nantong (see Figure 3.1). This coastal deltaic plain consists of well drained percolating alluvial (or paddy) soils where the dominant crop historically was cotton. Output of both raw cotton and cloth supplied the low-lying areas of the central basin. While cotton remains the local specialty modern agriculture includes winter cropping of rapeseed (canola) and wheat, and rice during the summer. The lowest-lying area of the central basin, a limnetic plain of poorly draining paddy soils, developed an  84  Figure 3.1 PREFECTURES, COUNTIES, & CITIES: JIANGSU & SHANGHAI, 1991  Lianyungang\  Xuzhou,  [Huaiyin>  Yancheng  .  Hongze  Yangzhou  Nanjing ( Nantongi  Sub-Regions Coastal Plain  Lake Tai  Limnetic Plain  Or  Mid-Level Plain Adminstrative Divisions  ^Xi^Suzhoi  \ \ \ > ^ > y \ ^ -  Shanghai  1  Prefectural Boundary  Kunshan  Prefectural Seat (City) County Boundary  150km  Yangzi River Note: 1. For a detailed explanation of the hierarchy of administrative units in Qiina see Appendix 4. 2. See Appendix 4.  Source:  ZGXZHC1992, pp. 27-29.  85 interdependent system of wet and dry agriculture. Wet rice was grown in fields surrounded by prominent embankments to prevent inundation. Reinforcing these embankments were mulberry bushes which supported thriving local sericulture. Earnings from silk largely financed imports of cotton and cloth. Middle-lying areas, including the southwest part of present day Shanghai, Kunshan, and the vicinity of the city of Suzhou, most of Wuxi and Changzhou, and along the banks of the Yangzi River to Nanjing (see Figure 3.1), are comprised of a mid-level limnetic and deltaic plain. A s these areas required no embankments, yet were easily irrigated, they grew rice almost exclusively, supplying the other two areas with surplus grain. Today, rice is a highly productive summer crop along with winter crops of wheat and canola. Clearly, the success of agriculture in the lower Yangzi delta required vigorous efforts to precisely control and manage water. Moreover, the structure and intensity of intraregional transactional activities historically depended upon an intricate network of waterways and canals for transportation. The density of this irrigation and transport network was 6 to 8 times greater than in the north China plain before the post-1949 efforts to restructure the natural environment. Presently, across the 6 counties of 3  Suzhou, for example, 16 percent of the total area is water (not including any of Lake Tai) and in Kunshan the proportion rises to 22.3 percent.  4  This environment was  ecologically more stable with a mere 20 major floods in the 500 years to 1900. Some 5  have also pointed out how this dense water network contributed to a more stable social and economic environment by facilitating relatively higher incomes, and providing a relatively safe haven from uprisings and war, usually from the north, absorbing population and advanced farming techniques and culture in the process.  6  The complexity of the task of managing and maintaining the canal and irrigation system was reflected in the structure of local level administration. Frequently haphazard and inconsistent over the centuries, the issue of water control is worth mentioning here since it provides a historical context for understanding contemporary local administration in the lower Yangzi delta. While the taxation and corvee labour  86 systems also relied upon such administration it is the nature of water control to which local administrative organizations were most clearly linked. Unlike north China, which depended on massive state led intervention to control the Yellow River, state water control in the lower Yangzi delta, like the central administration, essentially stopped at the county level. While the precise role that small communities historically played in water control is the subject of continuing debate, it is clear that what emerged in the delta was a system in which localities were left to fend for themselves until the situation degenerated into widespread inundation. With the divergence of local interests, longterm regional degeneration of water control was virtually unavoidable, only periodically slowed by central government efforts in response to disaster. Yet the fact of strong 7  local interests, and the organizational structures through which they acted, remained a prominent characteristic throughout. Thus, the scale of water control projects in the lower Yangzi delta, usually too small for state concern and too large for individual households, necessarily resulted in a complex set of interrelationships, driven at the community level (township and village), between the central state, peasants, and beginning around 1350, the urbanizing mercantile and landowning elites.  3.1.2 Rural-urban relations and the urban penumbra Sustained high levels of agricultural output and abundant water transport led to a highly commercialized rural economy in the lower Yangzi delta. The basis for this transformation was the introduction of cotton cloth and silk. In the 500 years or so of late imperial China cotton cloth replaced hemp as the material of choice among peasants while silk became popular among the expanding merchant class and other well-off urbanites. These developments also resulted in the commercialization of grain in the delta during this period. Today, extensive commercialization has also occurred in other traditional agricultural activities such as animal husbandry and aquaculture. Commercialization of cotton and sericulture, and marketization of grain contributed to a dramatic growth in the number of small towns in the lower Yangzi  87 delta during the late imperial period. Linked to the marketing of cloth, silk, and rice this region became the most highly urbanized and commercialized i n China by the late nineteenth century. In the mid-fourteenth century, on the eve of the founding of the Ming Dynasty (1368), one part of the delta, including the 6 counties of present day Suzhou, Jiading County and the former Baoshan County now both i n Shanghai, contained only 4 towns. This increased to 67 by the early seventeenth century and to 90 by the mid-nineteenth century (late Qing Dynasty). For Kunshan during the same 8  period the corresponding numbers of towns was 0, 9, and 15 respectively. Today there are 20 administratively designated townships in Kunshan. While southern Jiangsu became the most highly commercialized and urbanized region in China, by the late nineteenth century only 10.6 percent of the population lived in towns of 2000 or more.  9  Meanwhile, the walled city of Suzhou probably became the largest metropolis in China based on the cotton cloth processing and silk weaving industries.  10  While the increasing involvement of peasants in the commercialization of the delta contributed to the proliferation of small towns, a number of related factors acted to keep by far the greatest proportion of the population residing in the countryside. Relatively stable and productive agriculture stimulated population growth which i n turn exerted pressures on food production and per-capita incomes. A n increasing proportion of labour in excess of that needed for cultivation assumed a greater share of the household's productive activities in agricultural sidelines to offset these pressures. Notwithstanding the vicissitudes of increasingly influential market conditions, the household economy in the delta became relatively more stable with a given unit of land thus able to support more people. Moreover, since small towns emerged mainly as centres of commercial activities, with the exception of a certain amount of silk and cotton cloth production, most processing activities were dispersed throughout the countryside.  11  The increasingly dense peasant population was so well-linked to higher level markets that the standard periodic market towns, which supplied a range of daily  88 necessities and around which socio-economic life in most of rural China was organized, did not exist in the lower Yangzi delta.  12  By the end of the nineteenth century a  complex system of markets, commercial services, and production had emerged that enabled peasants to engage in non-agricultural production without having to leave the farm. While small towns in southern Jiangsu served the immediate community and were linked to external markets, they also catered to the rapidly expanding rural consumer market.  13  Expanding commercialization and marketization in the delta, and  the transformation of the peasant economy and growth of small towns which accompanied them, were based on a highly effective transactional network. Yet the efficiency of local level productive activities and per-capita incomes over the same period eventually either stagnated or declined.  14  It is important to emphasize here that  commercialization in the delta, while resulting in dramatic change, did not in the longrun lead to transformative development. The rise of new industries and of large and small cities with the penetration of international capitalism through the late nineteenth century and the Republican period, added a further dimension to change in the lower Yangzi delta. While some claimed that the rural economy remained to a large extent unaffected by the development of urban capitalism in China others have clearly shown how it fundamentally reshaped peasant production.  15  The basic change which left virtually no peasant household  untouched, was the switch from weaving cloth with home-spun yarn to the use of machine-spun yarn produced in urban mills made from locally produced cotton and silk. The fact that this new system of production sustained involutionary commercial growth and contributed to the growth of many towns and the decline of a few others, is less important to this study than the way in which urban areas became linked to the peasant economy. These interrelationships relied upon commercial services, especially the trade and investment activities of merchants and their organizational and transactional networks. Profiting mainly from the circulation of capital in the exchange of goods and not from investment in production, these activities did nevertheless stimulate the  89 proliferation and modernization of agricultural sidelines and household based processing activities.  16  This helps to explain the lower Yangzi delta's rather underdeveloped urban hierarchy through this period described so well in Skinner's seminal study.  17  Along with  larger cities such as Suzhou and Nanjing, the region became crowded with many smaller towns deeply linked to the peasant economy, without a commensurate growth in small and medium sized cities. Regional prosperity and urbanization i n this underdeveloped hierarchy of settlements was seen to be more "general" throughout the delta.  18  Even  the explosive growth of Shanghai after its founding as a Treaty Port i n 1842, did not diminish the economic and cultural prosperity of the delta, although by many measures Shanghai quickly surpassed older cities like Suzhou.  19  The transformation of sericulture and cotton production during the late imperial and Republican periods, characterized by the proliferation of small scale transactional activities, embedded small towns and large cities within an extensive network of reciprocal exchange relationships. This meant that rural communities assumed a greater variety of roles and, as the density of villages and small towns in the lower Yangzi delta increased, their functional integration with the larger cities became increasingly complex. Such integration tied large cities to the growing concentration of rural populations in what Marme describes as the "urban penumbra".  20  With the  transformation and expansion of agricultural sidelines then, major cities emerged as part of "an area of widely diffused urbanness", a region with whose fortunes their own prosperity tended to move i n tandem  2 1  The role of cities in the lower Yangzi delta,  therefore, needs to be placed in the context of a fundamentally endogenous rural transformation of rather long duration.  90 3.2  Pre-Reform Political Economy Widespread "socialist transformation" implemented in China following  "Liberation" in 1949 was characterized by a process of collectivization i n which agriculture became independent of individual small-holdings, private ownership was reduced, and the means of production were socialized. Along with land reform in the early 1950s, which resulted in the redistribution of land and tools of production to poorer households, agricultural sideline production in the lower Yangzi delta continued to develop rapidly. By 1955 there was another large scale movement underway to 22  reorganize the peasant economy into agricultural producers' cooperatives and handicraft cooperatives, the latter being mostly relocated into towns and administratively separated from agriculture. The commensurate emphasis on the production of grain and on the development of large scale urban industrial enterprises led to the stagnation of small scale non-agricultural production in the countryside. In 1958 the process of collectivization culminated with the policies of the Great Leap Forward and the almost overnight establishment of people's communes (renmin gongshe) loosely based on the former townships in terms of population and spatial extent. Constituent brigades (shenchan dadui), roughly equivalent to administrative villages of perhaps 200 to 400 households, and production teams (shenchan xiaodui), which in the delta approximated the old natural village (zhuang) of between 30 to 50 households, became responsible for all aspects of rural life under the unified management of communes. Since specific tasks were determined and assigned to small production groups (shenchan xiaozhu) comprised of 1 to a few households, all but a few of the most rudimentary family chores had become collectively organized. A new policy for rural industrialization in China also emerged at this time which encouraged communes and brigades to establish enterprises to produce industrial goods. While the central objective was to serve agriculture and the mechanization and electrification of farming, industrial production was also supposed to meet the demands  91 of the peasants for daily necessities, and provide materials and local processing to support China's industrial development. A large number of highly inefficient enterprises were subsequently established in the countryside which resulted in the enormous waste of human and material resources. Considerable disruption in agricultural production also occurred as rural areas concentrated on local nonagricultural activities precipitating disastrous food shortages.  23  The government  responded to the crisis in 1960 by implementing a nation-wide program of retrenchment and adjustment of rural industry. Many commune and brigade enterprises were ordered closed and strict requirements were placed on the proportion of rural labour to be allocated to agriculture. In addition, a large number of peasants who had moved to cities to join the urban industrial workforce were returned to the countryside. While informants frequently referred to this period as the "three difficult years" many also highlighted aspects of the Great Leap Forward which contributed to the later prosperity of rural industrial development in southern Jiangsu. While many commune and brigade enterprises were forced to close or, as with the handicraft cooperatives in the mid-1950s, were relocated to urban areas, a sizeable amount of rural industrial infrastructure was created. This, combined with the experience that many farmers gained from rural industrialization during the Great Leap Forward, and the newly acquired skills of those who were forced to return from urban state enterprises, contributed to a build-up of physical and human capital in rural areas. Generally speaking however, except during the first part of the Great Leap Forward, national policies towards rural industry remained highly restrictive through the early to mid1960s. During the decade of the Cultural Revolution (1966-1976), rural industry came to be viewed as a way of directing resources into agriculture by supporting local selfsufficiency in the material inputs needed for irrigation, mechanization, electrification and application of chemical fertilizers.  24  Although some communes and most brigades  were allowed to operate small collective factories, any enterprises that might divert  92 resources from farming or activities not in direct support of agriculture were vigorously discouraged. Furthermore, in the frenzied ideological iconoclasm of the day individual and even commune and brigade enterprises were frequently attacked as the "tail of capitalism". In the context of central government policies to focus on "grain first" this led to considerable pressure to restrict rural non-agricultural development. In spite of the highly centralized and uniform organizational structure of the collective period after 1949 rural communities and local leadership often had a great influence on rural industrialization. Local and regional authorities in southern Jiangsu often resisted and sometimes, with great political risk, successfully circumvented national campaigns to shut down or readjust commune and brigade enterprises and to concentrate on agricultural production. In the early 1960s, for example, while most of China struggled unsuccessfully to feed itself, the crisis in the lower Yangzi delta was less severe. With relatively higher agricultural productivity, more advanced rural industrialization, and among the nation's highest density of rural population, a national retrenchment policy that treated all regions equally was viewed as unnecessarily severe in the delta. Again i n the early 1970s intense debate erupted locally about how to adopt the spirit of a national campaign to concentrate on grain production. In the end, several counties in southern Jiangsu unofficially permitted, and even encouraged, the establishment of commune and brigade enterprises, protecting those not specifically related to the support of agriculture. Relying on the expertise of returned urban workers, the experience gained during the Great Leap Forward, and taking advantage of the general shortages of industrial goods in the countryside precipitated in part by work stoppages i n urban production during the Cultural Revolution, commune and brigade enterprises in places like Kunshan continued to expand and proliferate counter to the prevailing official policy.  25  Thus, on the eve of reforms in the late 1970s the  southern Jiangsu countryside was industrially more advanced than other rural regions in China.  93 Two other aspects of the collective period are worth mentioning here because of how they became important in the more recent transformation in the delta. First, numerous informants referred to various mass campaigns during which urban workers were returned to the countryside, and in the later stages of the Cultural Revolution i n the early 1970s, when many urban youths were "sent up to the mountains and down to the countryside". Many of these individuals, for various reasons elaborated upon in the next chapter, often ended up living in rural townships in places such as Kunshan. In addition to technological and some management skills they also brought social networks and contacts with family, friends, and former colleagues in cities like Shanghai and Suzhou. The second aspect relates to the underlying influence collectivization had upon the development of local level administrative structures in the delta. The state had forced a shift in economic decision making from the household level to production teams and their commune and brigade organizations. In addition to directing virtually every detail of agricultural and industrial production and procurement, communes and brigades were also responsible for negotiating economic relations with the centre. In a parallel development that occurred with the establishment of the National Rural Industry Administration in 1976, communes and brigades, which later provided the organizational foundation for township and village governments, began to set up formal structures to strengthen the administration and management of local industrial enterprises. A s will be demonstrated later, rural industrial production in southern Jiangsu was to remain largely under the control of townships and villages and their constituent administrative organizations. 3.3  Reforms and Transformation in the Rural Economy The death of Mao, and the end of the Cultural Revolution with the arrest of the  Gang of Four in 1976, signalled the beginning of a reassessment of the policies that affected the rural economy and precipitated a rapid expansion in rural industrial activities. In 1976 and 1977 rural gross industrial output in Jiangsu grew by 48 and 54  94 percent respectively.  26  While this growth continued through the interregnum it was not  until the Third Plenary of the Eleventh Central Committee of the Chinese Communist Party ( C C P ) i n December 1978, when Deng Xiaoping began to consolidate his power, that the first of a series of resolutions regarding the rural economy was formally adopted.  27  The resulting reforms implemented in the late 1970s and early 1980s led to  a deregulation of rural non-agricultural activities and reduced central and local government control of agriculture.  3.3.1  Changes in agriculture This reduced direct government control i n agriculture was most evident i n the  establishment of the household or production responsibility system (baogan  daohu)  which had been widely adopted in Jiangsu by 1983. While land was still owned collectively, management decisions regarding production were left to individual households. After meeting certain fixed obligations based on state procurement, surplus output could be retained for consumption or disposed on the market. Much has been made of the return to household organization in farming, particularly in terms of its supposedly superior incentives in producing dramatic increases in crop yields.  28  Across Jiangsu, total output from cultivation and crop yields per unit area in fact stagnated into the late 1980s and early 1990s. Unit area yields of staple grains, for example, actually showed a marginal average annual decline between 1984 and 1993. The same is also true for cotton, while the unit area yields of oil bearing seeds increased at an annual average rate of only 1 percent during the same p e r i o d .  29  Two other parallel developments in the rural economy occurred in the early 1980s which contributed to the transformation in the countryside. First, the production responsibility system brought about crucial changes in the former collective workpoints system of remuneration which had greatly diluted incentives for hard work and the use of spare-time auxiliary labour for farming and off-farm employment. Direct payment of workers, pay based on performance, profit related bonuses, and piece rate wages  95 became the norm and had a dramatic impact on incentives and efficiency in the agricultural and rural non-agricultural sectors. Second, the capacity to make more flexible and efficient use of labour increased with the relaxation of restrictions on the expansion of non-agricultural activities by rural communities associated with the abolition of commune structures and their replacement by township and village administrations by 1984. The simultaneous development of all rural activities was seen as the most effective way of increasing efficiency in agricultural production and for absorbing underemployed labour released from collective agriculture. Thus, as Huang has suggested, the rural change of greatest long-term importance in post-reform Jiangsu was the deinvolution of crop production that came with the diversification of the rural economy through rural industrialization and the expansion of sideline activities, and not the turn to marketized farming as is commonly assumed.  30  Figure 3.2 illustrates changes in the gross value of agricultural output ( G V A O ) and the sectoral distribution of agriculture in Jiangsu for selected years between 1978 and 1993. While the G V A O grew from R M B 141 billion to R M B 348 billion over the period, nearly 70 percent of this increase occurred in sectors outside cultivation  3 1  Aquaculture (primarily fish and shrimp raising), and animal husbandry (mostly poultry and pigs), accounted for 27 and 9 percent respectively of the increase in G V A O . A further 33 percent of the increase came from growth in household sidelines or so called petty commodity production, including handicrafts, and simple food processing such as the production oidoufu and related products (see Plate 3.1). More significantly, the proportion of G V A O from such sidelines increased from under 2 percent in 1978 to over 20 percent in 1993 reflecting a 35 percent average annual increase in the value of output. Meanwhile, the proportions from aquaculture (2.3 to 6.2 percent) and animal husbandry (15.8 to 22.3 percent) showed only modest growth.  32  Taken together the  combined growth in agricultural activities outside crop production resulted in a decline in the proportion of G V A O from cultivation from over 80 percent in 1978 to about 50 percent i n 1993, despite substantial increases in the state procurement prices for major  96  Figure 3.2  TOTAL AGRICULTURAL OUTPUT BY SECTOR: JIANGSU, SELECTED YEARS  -100  lAquaculture 35 H  in  Husbandry  30-  80  OJ  >  I  o U o  25  H Sidelines  -60 f l  00  o  20 —  pa 't-4  40  15H  pa  10-  h-20  Cultivation  1978  1981  1985  Year  Sources: JSTJNJ1992, p. 107; JSTJNJ1994, p. 135.  1988  1991  1993  97 Plate 3.1 HOUSEHOLD SIDELINE PRODUCTION Preparing baiye doufu (100 leaves doufu) Dianshanhu Town, Kunshan  Plate 32 OFFICE FARMERS Peasants in office attire fertilizing rice seedlings Tongxin Village, Kunshan (see Section 3.3.2 below)  98 crops over the same period (see Figure 3.2). Agricultural productivity also exhibited wide spatial variation. Figure 3.3 plots agricultural output value per mu of arable land for Jiangsu and Shanghai for 1991 as an indicator of relative agricultural productivity.  33  With a few notable exceptions it is  quite clear that the region of the lower Yangzi delta including Suzhou, Wuxi, Changzhou, Shanghai, and the immediate vicinity of Nanjing were agriculturally more productive than the rest of Jiangsu. Unit area output values here were 1.8 to 3.5 times that of most of the northern part of Jiangsu. Ganyu County north of the city of Lianyungang in northeast Jiangsu was located in the heart of the province's petrochemical producing region and benefited from easy access to chemical fertilizers, which helps to explain why it falls within the highest category of agricultural productivity. Ganyu, and other counties along the coast such as Sheyang in Yancheng Prefecture and several counties in Nantong Prefecture, are well known for their aquaculture and fishing which also helps to explain their relatively high levels of agricultural productivity. Furthermore, with the exception of Huaiyin City in Subei, all urban districts fall into the highest categories of agricultural production. Most urban districts in Jiangsu and Shanghai include small areas of arable land cultivated with labour intensive cash crops, especially vegetables, for urban consumption which explains their relatively high levels of agricultural productivity. Interestingly, Kunshan falls within only the third highest category of agricultural productivity. Variations in the distribution of surplus labour may partly explain this classification. Since no such data were available, population density was a reasonable proxy for surplus labour in the countryside. Figure 3.4 plots population density across Jiangsu and Shanghai for 1991. Population density in Kunshan, compared to other parts of the delta, was a relatively low 657 people per square kilometer. Therefore, in terms of G V A O per-capita, Kunshan was actually one of the most productive areas in all of Jiangsu  3 4  The introduction of Figure 3.4 provides the opportunity to examine the demographic context of the patterns of spatial economic change discussed in more  99  Figure 3.3 AGRICULTURAL OUTPUT VALUE PER MU OF ARABLE LAND: JIANGSU AND SHANGHAI, 1991  Agricultural Productivity ( R M B / M U )  1  1064 and above 940 to 1063  Kunshan  806 to 939 649 to 805  150km  Below 649 Note: 1.1991 current values, 15 M U = 1 hectare. Sources: Calculated from: HDDTNJ1992, pp. 270-273,365-377; JSSXJJ1991 pp. 441-442.  100  Figure 3 . 4 POPULATION DENSITY: JIANGSU AND SHANGHAI, 1991  Below 620  Sources: Calculated from:HDDTN/1992, p p . 270-273, 365-377; JSSXJJ1991 p p . 429-430; WXTJN] 1992, pp. 326-328.  101 detail later. Parts of the lower Yangzi delta, including most of Shanghai, Suzhou, Wuxi, Changzhou, and 2 counties in Zhenjiang along the south bank, and 4 counties in Yangzhou and most of Nantong along the north bank, were the most heavily populated in the region with densities ranging from 750 to 1200 people per square kilometer. A l l but one of the urban centres fall into the highest category, with average city densities ranging from as low as 1300 in Huaiyin in Subei to over 10,000 in Shanghai. The one exception was Lianyungang City in the northeast which administratively included some areas of countryside, as did most other city districts in Jiangsu and Shanghai. But with a population of 530,000 the city exhibited an "urban" density of only 639 people per square kilometer which explains its fourth level categorization. The extensive region of relatively high population density along the north bank of the lower Yangzi delta resulted from a combination of historically productive agriculture, which was able to foster and sustain a greater population, and a rather underdeveloped settlement hierarchy. Except for Nantong City (pop. 579,000), there were no other urban settlements of significant size to absorb large numbers of the local population. O n the other hand, the presence of more concentrated urban centres such as Suzhou (pop. 851,000), Wuxi (pop. 937,000), and Changzhou (pop. 677,000) helps to explain the slightly lower densities in parts of the south bank region i n the vicinity of these cities. That Kunshan and a few adjacent counties show relatively low densities is partly explained by the presence of extensive waterways and lakes for which the density data presented in Figure 3.4 did not compensate.  35  With an aggregate population of over 5  million in 1991, Shanghai's suburban counties fall mostly into the highest categories of population density. The high degree of coincidence of the patterns illustrated in Figures 3.3 and 3.4 confirms the traditional relationship between high levels of agricultural productivity and greater population. I will return to these figures again later, but for the time being they provide the initial empirical evidence that circumstances in the lower Yangzi delta distinguish it from the surrounding region.  102 3.3.2  Township and village enterprises: New shapes, old patterns Although the central government provided little direct support to industrial  development in the countryside, either by way of tax concessions, subsidies, or investments, the adoption of reform measures since 1978 has significantly improved the environment for rural non-agricultural activities. The deregulation of rural activities "released a tremendous amount of productive and entrepreneurial energy in rural areas that had previously been constrained by central planning and tight government control."  36  This is clearly reflected in the data presented in Figure 3.5 which show the  growth in the number of rural enterprises and employees, and the rural gross value of industrial output ( G V I O ) in Jiangsu since 1978. The term rural enterprise (xiangzhen qiye or nongcun qiye) unless otherwise indicated, will hereafter refer to all collectively (jiti qiye) and privately (siying qiye) or individually owned (geti qiye) industrial enterprises at or below the township level. The term township and village enterprise will largely denote enterprises owned collectively by township and village level administrations. Unless otherwise indicated most data presented for these designations will also include a small number of cooperative enterprises (hezuo qiye), and new variations i n forms of ownership such as foreign and domestic joint ventures, joint households (lianhu qiye) and associations, and joint ownership by different administrative units. The number of rural enterprises and employees increased from 56,500 and 2.5 million respectively in 1978 to a high of nearly 105,000 and 6.3 million respectively in 1988 (see Figure 3.5, Parts A and B ) .  37  A national campaign initiated in October 1988  to "improve the economic environment and rectify the economic order" in response to overheated industrial growth and high inflation, resulted in a decline in the number of rural enterprises and employment. The squeeze on credit and reduced access to raw materials during the two year recessionary period which followed, forced many rural enterprises to either close or amalgamate, and saw the return of many laid-off workers  103  Figure 3.5  RURAL INDUSTRIAL DEVELOPMENT: JIANGSU, SELECTED YEARS 120  •3 100J  (A) Enterprises  8(H  .s *  60  I  40H 20-  1978  1981  1985  1988  1991  1993  1991  1993  1991  1993  7 6  (B) Employees  5 2 .£5  4  I  3  I  210  1978  1981  1985  1988  360" 340;  (C) Gross Value of Output  120-  M  i s 3  °  ^ o CO ON  100I  806040200-  1978  1981  1985  Year  1988  Sources: Number of enterprises and employees for 1978 from Ho (1994) p.19; other data calculated from: ]SSSN 1949-1989, P-136; JSTJNJ 1992, p. 124; JSTJNJ 1994, pp. 147,159,160.  104 to farming. Many of the informants interviewed in the lower Yangzi delta region referred to this period of retrenchment in a positive way, emphasizing how it led to greater efficiency in the rural industrial sector.  38  Figures for the number of enterprises  and employees resumed a more gradual rise after 1991 increasing to just over 99,000 and 6.2 million respectively by 1993. More dramatic growth, however, occurred in the rural G V I O which increased at an annual average rate of 32 percent between 1978 and 1993 (see Figure 3.5, Part C ) . This compares with output values from all crop production which increased an average of only 3 percent annually over the same period  3 9  Rural G V I O has almost tripled since 1991, increasing to R M B 355.1 billion in  1993. This figure for Jiangsu Province represents a staggering one-fifth of the nation's total value of rural industrial output!  40  Changes in agriculture and the growth of rural non-agricultural activities have facilitated a dramatic restructuring of the entire rural economy i n Jiangsu. Data for selected years presented in Figure 3.6 captures the essence of this rural transformation. Total social product increased from R M B 21.5 billion in 1978 to R M B 394.1 billion in 1993 (see Figure 3.6, Part A )  4 1  More significant was the change in the distribution of  output value in the material products sectors which accompanied this impressive growth. The proportion of total output from agriculture declined from nearly 62 percent in 1978 to just over 8 percent in 1993, while the share from industry grew from just 29 percent to over 90 percent over the same period. The proportion of the total from construction, transportation, and commerce hovered around 10 percent until 1991 after which it declined to 1.6 percent by 1993. Rural industrial growth was responsible for over 90 percent of the increase in total social product.  42  This shift from agricultural to industrial activities in rural Jiangsu is also reflected in the sectoral distribution of rural labour as illustrated in Part B of Figure 3.6. The most obvious change was the decline in the share of the rural labour force employed in agriculture from 86.3 percent in 1978 to 60.8 percent in 1993, nearly all of which came from decreases in the number of peasants engaged in cultivation  4 3  The  105  Figure 3.6 TOTAL SOCIAL PRODUCT AND LABOUR FORCE BY SECTOR: RURAL JIANGSU, SELECTED YEARS A) Total Social Product (Output of the material products sectors)  1978  1981  1985  1988 Year  1991  1988  1991  1993  B) Labour Forced  1981  1985  Year  1993  Notes: 1. Includes construction, transport, and commerce. 2. Includes rural labour with official registration at or below the township level. Sources: JSTJNJ 1992, p. 78; JSTJNJ 1994, pp. 33,129,131.  106 proportion of rural labour working in industry increased from about 9 percent in 1978 to over 20 percent in 1993 while employment in other sectors increased from 4.4 percent to nearly 27 percent. The construction, transportation, and commercial sectors accounted for over 70 percent of the growth in this latter category. Patterns of rural employment also exhibited wide spatial variation. Figure 3.7 plots the available data for non-agricultural employment as a proportion of the total rural labour force for Jiangsu and Shanghai for 1991. In most of the south bank region of the lower Yangzi delta, including all of Shanghai, Suzhou, and Wuxi, Wujin County around Changzhou City, Yangzhong County in Zhenjiang, and Hanjiang, Jiangdu, and Jingjiang Counties in Yangzhou, more than 51 percent of the rural labour force was engaged in non-agricultural activities corresponding to the two darkest categories illustrated. In 6 of these 23 county level units, including 4 in Shanghai, and Wuxi and Jiangyin Counties in Wuxi Prefecture, more than three quarters of the rural labour force was employed i n non-agricultural sectors. In 20 county level units i n Subei, on the other hand, the proportion dips to below 25 percent, including 6 with less than 10 percent. Not surprisingly, all urban areas in the lower Yangzi delta exhibited high proportions of rural labour outside agriculture. However, since city districts include only a tiny number of officially designated local rural residents, the classifications illustrated for urban areas should not be considered significant.  44  Taken from the official Chinese statistical sources, these figures for Jiangsu and Shanghai need to be considered in the context of first-hand field observations and , interview data which reveal that the sectoral divisions of rural labour were not as straightforward as these figures imply. While the absolute number of rural workers included a small proportion who have moved to work i n urban areas (especially i n construction), the official statistics conceal the fact that there was much occupational overlap and dual employment. Many of those employed primarily in agriculture, for example, spent at least part of their working time engaged in non-agricultural activities. More significant, was the fact that rural workers were frequently released from the  107  Figure 3.7 NON-AGRICULTURAL EMPLOYMENT AS A PROPORTION OF THE RURAL LABOUR FORCE: JIANGSU AND SHANGHAI, 1991  1  I' -' • ' -I  |  25 - 38.9 Below 25  Note:  1. Figures for Shanghai calculated from 1990 data. Sources: Calculated from: JSSXJJ1992, pp. 45-48; SJQTZH1990, p. 13.  108 factories and other non-agricultural activities during the particularly busy stages of cultivation (see Plate 3.2). Development of rural enterprises in Jiangsu has also been accompanied by growth in trade and commercial activities. New procurement and marketing channels were established, initially as a result of the increase in agricultural and sideline products sold directly to end users, and later to the increased rural production and consumption of manufactured goods. Rural residents were also gradually permitted to engage i n the transport and sale of certain goods, and in many areas to establish residence in small market towns for commerce and services if they supplied their own rations for grain. Rural retail sales in Jiangsu, as one measure of such activity, increased from R M B 642 million in 1978 to R M B 4.48 billion in 1992 accounting for 56 percent of total provincial retail sales for that year.  45  A further indication of the vigor of rural commerce was the  rise in the volume of agricultural production sold by farmers directly to urban residents, increasing form R M B 18.3 million in 1978 to R M B 672 million in 1992, accounting for 8.4 percent of that year's total volume of retail sales for the province.  46  Official sanction of township and village industries, individual and institutional entrepreneurship, and a commensurate rise in commercial and trading activities in the countryside, have led to a revitalization of the marketing and commercial functions of small towns. While the magnitude and extent of the spatial economic restructuring that has fuelled this revitalization is a relatively recent phenomenon, historical evidence reviewed in the first part of this chapter suggests that these developments have resulted in a recapturing of former transactional networks in addition to the creation of new ones. Thus, while many aspects of the shapes of this restructuring may be new, in terms of ownership, types of industries, and enterprise structure, the patterns of location and the institutional context of non-agricultural activities were consistent with historical patterns of commercialization and the development of small towns in the lower Yangzi delta - the "new shapes, old patterns" i n the title of this section. Each of these issues mentioned above constitute key elements of the rural transformation and mega-  109 urbanization identified in Figure 2.2. Rural residents and community governments responded aggressively to the newopportunities that emerged within a more permissive and flexible reform environment. This release of energy at the grass-roots level, manifested most clearly in the shift of resources from agriculture, especially cultivation, to more profitable non-agricultural activities, was the underlying cause of spatial economic restructuring in rural Jiangsu through the 1980s and 1990s.  47  A s will be discussed in the following section, however,  rural non-agricultural activities are not evenly distributed throughout the region. 3.4  Spatial Economic Patterns in the Lower Yangzi Delta Changes in the lower Yangzi delta reflect an evolution of the regional space  economy whereby industry and other non-agricultural activities have been integrated into areas of the countryside that were also among the nation's most productive agricultural regions. Some of the highest income households in China were found in these areas, benefiting from nearby markets for their produce along with burgeoning urban and rural demand for locally produced goods, and services such as transportation and construction. As extensive commercial and market transactions proliferated at the fringe of large urban conurbations like Shanghai, the distinction between urban and rural economies has become less clear. This section will discuss the emergence of particular patterns of rural industrial activity across Jiangsu and Shanghai which highlight these features. The spatial distribution of 1991 rural per-capita net incomes for Jiangsu and Shanghai are plotted in Figure 3.8. Work by Veeck has demonstrated that growing percapita incomes show a strong positive relationship to locational variations in nonagricultural employment opportunities.  48  Since, industrial employment was the most  significant factor i n income determination, Figure 3.8 shows where industrial activity was relatively important. It was a zone in the southern part of the lower Yangzi delta between Nanjing and Shanghai, especially around the Lake T a i cities of Suzhou, Wuxi,  110  Figure 3.8 RURAL PER-CAPITA NET INCOME: JIANGSU A N D SHANGHAI, 1991  N  . Hongz  Rural Per-Capita Net Income (RMB)  1  1,500 and above  St  1,100 -1,499 885 -1,099 775 - 884  150km  Below 775 Note: 1.1991 current values. Sources: HDDTNJ1992, pp. 340-343,365-377; JSSXJJ1991, pp. 497-498; WXTJN] 1992,335-337.  Ill and Changzhou (often referred to by the Chinese as Suxichang taken from the names of each city), and parts of Zhenjiang. Rural per-capita net incomes here averaged 2 to 3 times that of the northern part of Jiangsu indicating the relative importance of rural industrial activity. A s with data on agricultural productivity, the high level classification of urban areas should not be considered significant. The second and third highest classifications for the 4 counties in coastal Yancheng probably resulted from the presence of comparatively profitable coastal fishing, and extensive aquaculture and salt production, combined with relatively low population densities (cf. Figure 3.4). A s expected rural per-capita net incomes in Shanghai Municipality were relatively high with 3 of the 9 suburban counties with values over R M B 2000 in 1991. Somewhat surprising, however, was the lower rankings of counties adjacent to the city of Nanjing (pop. 2.25 million) which implied that rural industrial activities were much less developed in these areas. Indeed, as data for rural industrial output plotted in Figure 3.9 shows, rural industry was less vigorous here than proximity to Jiangsu's largest industrial centre might suggest. Except for Jiangning County immediately to the south of Nanjing City, values for rural industrial output in the 4 other counties of the prefecture averaged only R M B 349 million in 1991. These values were as low as the least developed counties in Subei and cannot be wholly explained by the fact that the Yangzi River bisects the region. Development of rural industry i n these counties has likely been constrained by factors relating to the "primacy" of Nanjing City (the provincial capital). Investment in rural industry may have also been limited by potential earnings from employment in the service sector, in construction, or from intensive cash cropping catering to the nearby market. These same factors may also help to explain why values for rural industrial output i n a few counties i n Shanghai were slightly lower than the highest level categorization. More importantly, Figure 3.9 confirms the relative intensity of non-agricultural activity in the Suxichang region near Lake Tai in southern Jiangsu. Output values here were on average 25 to 30 times those of the least developed areas in Subei. Wuxi, the richest county in China, produced 100  112  Figure 3.9 RURAL INDUSTRIAL OUTPUT: JIANGSU AND SHANGHAI, 1991  Jiangning  Value of Rural Industrial Output (xl0 RMB) 9  1  3.75 and above 2.75 - 3.74 1.25-2.74 0.65 -1.24  150km  Below 0.65 Note: 1.1991 current values. Sources: Calculated from: HDDTNJ1992,  pp. 270-273,365-377; JSSXJJ1992, pp. 455-456; WXTJN]1992 pp. 329-331.  113 to 150 times the rural industrial output of the poorest counties in Huaiyin in north central Jiangsu. In addition to the pronounced difference between northern Jiangsu and the lower Yangzi delta region, spatial patterns of rural industrial activity also reveal that the distinction between urban and rural economies within the most developed areas of the delta was becoming less clear. Figure 3.10 plots the gross value of industrial output ( G V I O ) across Jiangsu and Shanghai for 1991. Once again, the Lake T a i region of Suxichang exhibits among the highest levels of industrial activity. More interesting, however, is that county level jurisdictions here show levels of industrial activity that in most cases nearly equaled those of the nearby cities of Suzhou, Wuxi, and Changzhou. The G V I O in these areas averaged 80 percent of comparable values in the urban districts. In per-capita terms, the 1992 G V I O in the 6 counties of Suzhou of R M B 20,413 actually exceeded that in Suzhou City ( R M B 19,976). The 1992 per-capita G V I O in the 3 counties of Wuxi of R M B 20,225 was equivalent to 83 percent of that i n Wuxi City ( R M B 24,506).  49  While some counties in Shanghai also show high levels of  gross output, they were in all cases dwarfed by the R M B 127 billion produced by the city's urban industrial complex. Also of interest in this context were the relatively low values of output in counties near the city of Nanjing which in all cases never amounted to more than 9 percent that of the urban area. The 1992 per-capita G V I O in the 5 counties surrounding Nanjing City was only R M B 3733, equivalent to 24 percent of that in the nearby urban districts ( R M B 15,534).  50  While issues of urban primacy and  constraints on local investment in rural industrial activity may be relevant for these areas, it seems clear that mere proximity to large urban conurbations did not result in the sort of rural economic transformation which was occurring i n the Suxichang region of the delta. Although non-agricultural growth has occurred in all parts of Jiangsu and Shanghai, variations in the pace of industrial development emphasize how regional differences have widened. Figure 3.11 plots per-capita increases in industrial output  114  Figure 3.10 GROSS VALUE OF INDUSTRIAL OUTPUT: JIANGSU AND SHANGHAI, 1991  |  | Below 1.35  Note: 1. 1991 current values Sources: Calculated from: HDDTNJ1992, pp. 270-273,365-377; JSSXJJ1992, pp. 453-454; WXTJNJ1992, pp. 329-331.  115 across Jiangsu and Shanghai for the six years between 1985 and 1991. Suxichang again comprised the greatest contiguous extent of county level jurisdictions with the largest increases. Values for per-capita growth here were ten to twenty times larger than most areas of Subei. In fact, the magnitude of the increases in this part of the lower Yangzi delta were on average seven to ten times larger than total per-capita output values for the least developed parts of northern Jiangsu, especially Xuzhou, Lianyungang, Huaiyin, and Yancheng. The only significant outlier in this pattern was Yizheng County to the west of Yangzhou City. Well situated on the north bank of the Yangzi River, Yizheng was an important transshipment centre for petroleum from the northern Jiangsu oil fields.  51  A s a result, the production of synthetic fabrics and chemicals  became well developed in this area and probably account for its large increase. The patterns illustrated i n Figure 3.11 strongly suggest that, compared with other regions, industrial enterprises in Suxichang were either more efficient or tended to engage in more productive activities or both. What was it about the countryside here that has led to significantly higher levels of rural industrial development? Moreover, is it possible to delineate the spatial extent of such a highly productive non-urban region? The model of mega-urbanization proposed in Chapter Two suggests that there are a number of processes and outcomes that could be examined to help answer these questions. While the preceding analysis was able to clearly establish the existence of particular patterns of transformation, the intention was not to explicitly demarcate a mega-urban region as defined at the end of Chapter T w o .  52  However, it is possible to say with some certainty at this point, that at  least Suxichang, most of Shanghai, and parts of Zhenjiang are likely to be included within such a region. More importantly, this analysis has highlighted patterns that, first of all, distinguish this region from other areas, and secondly, appear to be inconsistent with conventional expectations. The two most obvious of these patterns were the presence of industrial development in areas of the countryside that equaled or, in some cases, exceeded levels of industrial development found in many cities, and the  116  Figure 3.11 INDUSTRIAL OUTPUT PER-CAPITA INCREASE, 1985-1991: JIANGSU AND SHANGHAI  Below 785 Note: 1.1980 constant values. Sources: Calculated from: HDDTN/1992, pp. 270-273,365-377; JSSSN1949-1989, pp. 382,384 394-396- 1SSX111992 pp. 105,107; PVXT/N/ 2992, pp. 326-331.  > OHIJNJ  im  117 persistence of areas near large cities like Nanjing and Shanghai that did not seem to have benefited by proximity to vibrant urban economies to the same extent as other regions. With regard to the first observation, it is worthwhile noting that the most intense rural industrial development largely coincided with areas that exhibited among the highest levels of agricultural productivity (compare Figure 3.3 with any or all of Figures 3.8 to 3.11). This is not surprising in the Chinese context where productive agriculture leads to greater population densities and pressures on arable land which stimulated the development of off-farm activities. The resulting rise in rural incomes also increased local demand for manufactured goods and services and linked prosperous agriculture to the growth of rural non-agricultural activities. With respect to the second point, I remain less sanguine than most about the supposedly central role of large cities and other exogenous forces as they affected the rural restructuring observed in the lower Yangzi delta. The evidence presented thus far, in the context of the historical geography of the region, strongly suggests that there were other processes and mechanisms linked to locality, the transactional environment, and the geography of production that were also at work here. Conceptualized under the rubric of megaurbanization proposed in Chapter Two, these processes and mechanisms need to be further analyzed to understand and explain how they drive the region's spatial economic transformation. This undertaking will also provide insights into the means by which the spatial extent of the lower Yangzi delta mega-urban region may be more precisely defined for the purpose of management and planning. The final section of this chapter, and of this part of the thesis, will introduce some of the key characteristics of the transformation that has occurred in Kunshan. The objective is to position Kunshan in the context of the lower Yangzi delta, and to justify the detailed case studies in the following three chapters which address specific elements of the model of mega-urbanization.  1.18 3.5  At the Edge of Shanghai: Kunshan to the Fore Kunshan epitomizes at once the depth of change, the profound problems, and  tantalizing prospects of spatial economic transformation in China's lower Yangzi delta. Yet the most rapid rural industrialization did not begin here until the mid-1980s, several years behind other counties in Suxichang. With an average of more than 1 mu of arable land per-person in Kunshan, the pressures to diversify into non-agricultural activity were less intense since per-capita output value from agriculture was relatively high. Areas such as Jiangyin in Wuxi, on the other hand, where arable land per-person was under 0.7 mu, had already been vigorously promoting non-agricultural development since the early 1980s.  53  Informants in Kunshan frequently claimed, however, that  despite the later start, industrial development was beginning at a higher level and therefore, was advancing more rapidly than in other areas more famous for their development.  54  Whatever the case, it was clear that since the mid-1980s Kunshan, like  other parts of the lower Yangzi delta, has undergone dramatic industrial growth and rural restructuring. This is illustrated in Part A of Figure 3.12 which plots the available data for the increase and sectoral distribution of total social product in Kunshan for selected years. Total social product i n Kunshan grew from R M B 690 million in 1980 to R M B 16.23 billion in 1992.  55  While the share of output from construction, transportation, and  commerce remained unchanged at about 13 percent, the proportion from industry rose from about half i n 1980 to more than four-fifths in 1992. Meanwhile, the proportion from agriculture decreased from 34 percent in 1980 to 5.5 percent in 1992. Average annual growth of the G V I O in Kunshan was over 35 percent for the period accounting for 98.3 percent of the increase in the gross value of agricultural and industrial output (GVAIO).  56  While total social product for China also showed rapid growth, the  changes in sectoral distribution were less dramatic (see Part B of Figure 3.12). Year on year growth in China's G V I O over the period averaged 14 percent, accounting for 90  119 Figure 3.12  TOTAL SOCIAL PRODUCT AND SECTORAL DISTRIBUTION: KUNSHAN AND CHINA, SELECTED YEARS 1  (A) Kunshan 100  18  90  15  80 "70  12  -60 --50  9  -40  §  6  -30  CQ  3  0  1980  1  1985  Agriculture  20  U CQ  '—y ID  1  u  s  u Du  10  1990  1991  Industry  1992  Other  (B) China 6000  100 90  5000  Q  80 70  4000|  60  CQ a o 02  50  3000  40 2000  t 30  S3  CQ  u SP  I  5 o  u  Ou  20  1000  10  0 1980  1985  Agriculture  1990  Industry  1992  1991  Other  Note:  1. Current values for the year shown Source: Calculated from:/SSSN 1949-1989 , pp. 382,394,398; ]ST]NJ 1990, pp. 30-31; SZTJNJ 1989, pp. 15,32-33; SZTJNJ 1991, pp. 38-39; SZTJNJ 1992, pp. 38-39; SZTJNJ 199?, pp. 40-41; ZGTJNJ 1993, p. 50.  120 percent of the total increase in the G V A I O .  5 7  Annual growth in the G V A O in  Kunshan averaged only 5 percent between 1980 and 1992, while agriculture across China grew 6.2 percent annually and still accounted for more than 16 percent of the nation's total social product in 1992.  58  Per-capita gross domestic product i n Kunshan i n  1993 was R M B 10,526, the fourth highest in Jiangsu behind three other areas in Suzhou and Wuxi, and 6 to 10 times larger than the least developed counties in the province.  59  The G V A I O per-capita in Kunshan in 1993 was R M B 35,673, also fourth highest in Jiangsu and 10 to 20 times that of the 15 least developed counties in the province.  60  However, as will be demonstrated below, the rapid growth of industrial activity in Kunshan has neither required nor resulted in a commensurate shift in residential urbanization  6 1  That is not to say that there has been little or no mobility of local  labour and population. In fact, significant proportions of the officially designated peasant population either resided in or commuted on a daily basis to work i n or near the town centres.  62  According to the official statistical classification 20.2 percent of the  population in Kunshan in 1992 was considered non-agricultural (fei nongye renkou).  63  This administrative designation conceals the numbers actually residing in the built-up township centres, but it does provide a useful baseline. Combined with data provided by informants it is possible to estimate an adjusted level of urbanization in Kunshan of approximately 25 to 30 percent  64  While there have been many attempts to generate  longitudinal estimates of urbanization for all of China the same is much more difficult for small jurisdictions such as Kunshan  6 5  Data in the most recent gazetteer does  suggest, however, that Kunshan was probably about 12 to 15 percent urbanized on the eve of reforms in the late 1970s  6 6  While the level of urbanization in Kunshan has  increased from perhaps 12 to 30 percent in the 14 years of reform since 1978, it is important to clarify the nature of this transition. Firstly, approximately 60 percent to 70 percent of the urban population in Kunshan in 1992 resided in 19 small towns outside the largest central urban settlement.  67  Second, from 1978 to 1992 these small towns,  which grew on average from about 2000 to 5000 people, accounted for nearly 90  121 percent of the growth in Kunshan's urban population over the same p e r i o d .  68  The level  of urbanization, its rate of increase, and the structure of urban settlement in Kunshan strongly suggest that rapid industrial growth here has not resulted i n large scale urban agglomeration. This finding becomes even more significant when it is revealed that the 1992 per-capita G V I O in Kunshan ( R M B 20,318) actually exceeded that in Suzhou City ( R M B 19,976), which was at least 90 percent u r b a n !  69  Other indicators are also instructive. Industrial land-use, for example, comprised an average of 40 percent of the area of town seats, while 7 percent of all of Kunshan was considered built-up in 1992  7 0  In addition to at least 2 special  development zones, all 20 towns had designated industrial areas (though not always contiguous), while more than 60 percent of the 466 villages had areas to which they referred as industrial districts.  71  In 1991 there were 3945 registered industrial  enterprises in Kunshan (all but 60 of them non-state run) accounting for more than 80 percent of total social product and employing half of the 369,076 strong workforce.  72  To emphasize the intensity of non-agricultural development in Kunshan, add to this the 5489 commercial organizations which employed 16,480 people, construction units with 18,120 labourers, a transport and telecommunications sector with 16,278 employees, and other tertiary level activities which employed a further 9.3 percent of the workforce.  73  A l l this in an area with among the nation's highest unit area yields of  staple grains (6270 kg/hectare), including rice (7700 kg/hectare) and wheat (3480 kg/hectare), and other agricultural products such as canola (1815 kg/hectare), vegetable greens (16,095 kg/hectare), and melons (14,340 kg/hectare).  74  It is  noteworthy that Kunshan, along with other rural areas of the lower Yangzi delta, have retained their critical importance in a national context as agricultural producer regions while at the same time undergoing rapid industrialization. These preliminary findings establish Kunshan as part of the most dynamic spatial economic restructuring occurring in the region. For this reason, understanding and explaining the processes and mechanisms working at the local level in Kunshan is to  122 appreciate trends of mega-urbanization across the entire lower Yangzi delta. The next part of this thesis will focus on the specific circumstances in Kunshan which drive local non-agricultural development in general, and which determine enterprise location and the emergence of related physical and institutional infrastructure in particular.  123  Notes 1.  She, Z. X. (Ed.) (1991) Taihu liuyu ziran ziyuan dim ji (Lake Tai basin Beijing: Science Press, Plates 9,10; Wang, C. F. (Ed.) (1990) Jiangsu sheng ditu Province). Guangdong: Guangdong Map Publisher, Plates 6-8.  natural resources atl ce (Atlas of Jiangsu  2.  The information presented here is based on an analysis of several historical studies, atlases, and gazetteers: Bell, L. S. (1992) Farming, sericulture, and peasant rationality in Wuxi County in the early twentieth century. In Rawski, T. G. and Li, L. M . (Eds.) Chinese history in economic perspective. Los Angeles: University of California Press; Hu, H . Y. (1947) A geographical sketch of Kiangsu Province. Geographical review 37 (4); Huang, P. C. C. (1990) The peasant family and rur development in the Yangzi delta, 1350-1988. Stanford: Stanford University Press; KSXZ (1990). Shanghai: Shanghai People's Publishers; She, Z. X. (Ed.) (1991) op. cit., Plates 3, 4, 8-10,18, 23; Wang, Y. C. (1992) Secular trends of rice prices in the Yangzi delta, 1638 -1935. In Rawski, T. G. and Li, L. M . (Eds.) op. cit.  3.  Huang, P. C. C. (1985) The peasant economy and social change University Press, Chapter 3.  4.  Calculated from SZTJNJ1992, p. 11.  5.  The north China plain was subject tofloodingalmost every year. Huang, P. C. C. (1990) op. cit.,  in north China. Stanford: Stan  p. 43. 6.  Huang, M . D. (1993) Chengzhen tixide jiegou yu yanhua: Lilun fenxiyu shizhengyanjiu (Th and evolution of urban systems: Theoretical analysis and case study). Unpublished Ph.D. East China Normal University, Shanghai, pp. 80, 81.  7.  Huang, P. C. C. (1990) op. cit., alludes to a long standing debate in China that focusses attention upon the need to coordinate local efforts in the lower Yangzi delta for regional benefit. Major flood events in 1991 and 1995 that I call "administrativefloods"have again highlighted how lack of adequate cooperation between Shanghai, Jiangsu, and Zhejiang exacerbated the innundations. See, for example, Gao, A . M . (1992) Jiangsu sets its growth strategy. China daily, March 26, p. 4. For many years geographers at East China Normal University have been predicting just such disastrous events. In the conclusion I will raise specific issues relating to administrative structure, jurisdiction, and power that must be resolved to overcome the local squabbling which bedevils regional coordination.  8.  Wang, J. F. (1984) Ming Qing Jiangnan shizhen jiegou ji lishi jiazhi chutan (Jiangnan cities and towns during the Ming and Qing Dynasties: Their structure and historical value). Huadong shifan dame  xuebao: Zhexue shehui kexue ban. (Journal of East China Normal University: Philoso sciences edition) (1), p. 75. 9.  Skinner, G. W. (1977) Regional urbanization in nineteenth-century China. In Skinner, G. W. (Ed.) The city in late imperial China. Stanford: Stanford University Press.  10.  Huang, P. C. C. (1990) op. cit., p. 48; Mote, F. W. (1973) A millennium of Chinese urban history: Form, time, and space concepts in Soochow. Rice University Studies 59 (4), p. 43, claims that Suzhou was never China's largest city; Johnson, L. C. (1993) Preface. In Johnson, L. C. (Ed.) Cities of Jiangnan in late imperial China. Albany: SUNY Press, p. X, suggests, on the other hand, that Suzhou was the largest and most prosperous city in the Empire in the late imperial period.  11.  Shih, J. C. (1992) Chinese rural society in transition: Berkeley: Institute of East Asian Studies, pp. 81-82.  A case study of the Lake Tai area, 13  124 12.  Skinner, G. W. (1985) Rural marketing in China: Revival and reappraisal. In Plattner, S. (Ed.)  Markets and marketing. Lanham: University Press of America.  13.  Densely populated rural areas of the lower Yangzi delta were tied directly to higher level markets by an excellent network of navigable canals. Through an agency system and widespread boat ownership, small towns via hundreds of agent boats could supply tens of thousands of rural households; See especially: Fei, X. T. (1939) Peasant life in China. London: Routledge, pp. 101-102; Marme, M . (1993) Heaven on earth: The rise of Suzhou, 1127 -1550. In Johnson, L. C. (Ed.) op. cit., p. 34.  14.  Huang, P. C. C. (1990) op. cit., 11-18 passim, describes two patterns of development in the peasant economy of the delta in late imperial times: "intensification" in which output or output value expands at the same rate as labour input; and "involutionary growth" and "evolutionary commercialization" (based on the famous study from Geertz, C. (1963) Agricultural involution: The process of ecologica change in Indonesia. Berkeley: University of California Press) in which total output expands, but at the cost of diminished marginal returns.  15.  Hou, C. M . (1963) Economic dualism: The case of China, 1840 -1937. Journal of economic history 23 (3); Murphey, R. (1977) The outsiders. Ann Arbor: University of Michigan Press.  16.  Huang, P. C. C. (1990) op. cit., pp. 133-136; Shih, J. C. (1992) op. cit., pp. 120-125.  17.  Skinner, G. W. (1977) op. cit.; Skinner's lower Yangzi urban region also included parts of northern Zhejiang Province including Hangzhou and Ningbo; Also referred to in Pannell, C. W. and Veeck, G . (1989) Zhujiang delta and Sunan: A comparative analysis of regional urban systems and their development. Asian geographer 8 (1 & 2), p. 137.  18.  Mote, F. W. (1973) op. cit., p. 44.  19.  Ibid., p. 47; According to Skinner, G. W. (1977) op. cit., p. 238, in 1843 Shanghai was ranked fourth in the lower Yangzi delta urban system after Suzhou, Hangzhou, and Nanjing.  20.  Marme, M . (1993) op. cit., p. 38.  21.  Rowe, W. T. (1993) Introduction. In Johnson, L. C. (Ed.) op. cit., p. 12.  22.  Across China the output value of such sideline production nearly doubled to RMB 2.2 billion in the 5 years to 1954. See: Byrd, W. A. and Lin, Q. S. (1990) China's rural industry: A n introduction. In Byrd, W. A. and Lin, Q. S. (Eds.) China's rural industry: Structure, development, and reform. Oxford University Press, p. 9; Ho, S. P. S. (1994) Rural China in transition: Non-agricultural  i  f  development in rural Jiangsu, 1978-1990. Oxford: Clarendon, p. 13.  23.  Estimates vary, but most observers agree that perhaps 20-30 million Chinese died during the 19601962 famine.  24.  Known as the "four transformations" (sihua) they were linked to the development of the "five small industries" (wu xiao gongye) — locally run enterprises that utilized lower level technology to produce iron and steel, cement, energy, chemical fertilizers, and agricultural machinery. For more discussion regarding why China promoted small-scale industry to modernize agriculture see: American Rural Small-Scale Industry Delegation (1977) Rural small-scale industry in the People's Republic Berkeley: University of California Press, Chapters 1, 4, 9; Sigurdson, J. (1977) Rural industrialization in China. Cambridge: Council on East Asian Studies, Harvard University, Chapters 1, 6.  25.  Research in other parts of the lower Yangzi delta, especially near Wuxi, confirms this finding. See: Byrd, W. A . and Lin, Q. S. (1990) op. cit., p. 7.  of C  125 26.  JSSSN1949-1989, p. 138; According to Ho, S. P. S. (1994) op. cit., p. 19, commune and brigade enterprise output across China for the same period increased by an annual average of 38 percent.  27.  For a superb review of the political economy of this period see Lieberthal, K. (1995)  28.  Croll, E . (1994) From heaven to earth: Images and experience of development in Roudedge, Chapter 2; Huang, P. C. C. (1990) op. cit., p. 319.  29.  ISTJNJ1992, pp. 107, 111, 112; ISTJNJ1994, pp. 134,137,138.  30.  Huang, P. C. C. (1990) op. cit., p. 17-18.  31.  Cultivation includes all crop production in the standard Chinese statistical categories mcluding staple grains, cotton, silk, oil seeds and so on.  32.  The figure for Forestry (one of the 5 standard statistical categories for agriculture in China) has been excluded since in all cases it never comprised more than 1 percent of the G V A O in Jiangsu.  33.  Thefiveclassifications of values plotted in this Figure, and in subsequent Figures 3.4, 3.7-3.11, were determined as follows: Data for the county level units was ranked and categorized initially into 5 classifications divided at each of the 20th, 40th, 60th, and 80th percentiles. In all but one case (Figure 3.10) the data for each of the 12 sets of urban districts (11 prefectural seats in Jiangsu plus Shanghai City) were excluded for the purpose of determining each of these percentile divisions. In some cases data for the 9 suburban counties in Shanghai Municipality were also excluded when determining classification intervals. For example, data for agricultural productivity in Figure 3.3 was categorized without including the values from Shanghai. This seems reasonable if we consider the somewhat special circumstances which lead to higher output values for the suburban counties in Shanghai which have greater opportunities to engage in higher value production of cash crops for the enormous urban market within this administrative jurisdiction. In other cases, such as the population density figures illustrated in Figure 3.4, values for Shanghai were included when determining the intervals. Finally, classification intervals were sometimes adjusted slightly to reflect the existence of obvious statistical and spatial clustering.  34.  Veeck, G . (1991) Regional variations in employment and income in Jiangsu Province. In Ginsburg, N., Koppel, B. and McGee, T. G. (Eds.) The extended metropolis: Settlement transition Honolulu: University of Hawaii Press, p. 166. Veeck also suggests, however, that environmental conditions and land quality are more effective estimators of agricultural productivity than labour since the large surplus rural work force in many areas keeps per-capita output values low. Thus, I have calculated output value per unit area of arable land to provide a more reliable comparison of agricultural productivity over space.  35.  I was able to locate surface water area figures for only Changzhou and Suzhou. Thus, Kunshan for example, with water covering 22.3 percent of its administrative area, would yield a corrected population densityfigureof 846 people per square kilometer. See: SZTJNJ1992, p. 11.  36.  Ho, S. P. S. (1994) op. cit., p. 27.  37.  The figure here for the number of employees refers to the actual number of workers employed in rural enterprises which may include people with official registration above the township level and/or from other areas.  38  See: Appendix 1 (Several interviews).  39.  Calculated from: JSTJNJ 1992, p. 107; JSTJNJ 1994, p. 134.  China: From revolution through reform. New York: Norton.  Governing  China. Lo  in Asia.  126  40.  Calculated from: ZGTJNJ 1994, p. 363.  41.  Total social product (shehui zong chanzhi) refers to the gross output value of the material products sectors. See: Appendix 2.  42.  Although data are compared in constant values and state procurement prices in the agricultural sector rose sharply during this period, widespread price deregulation in most industrial sectors has probably inflated the relative value of output from industry. Even when this is taken into consideration, however, it is clear that the growth of rural industry was the most significant component of rural restructuring in Jiangsu.  43.  JSTJNJ1988, p. 125; JSTJNJ1991, p. 91; JSTJNJ1992, p. 78,101; JSTJNJ1994, p. 33,129.  44.  For more details about this phenomenon see Appendix 4.  45.  JSTJNJ 1993, p. 236.  46.  Ibid., p. 237.  47.  This point is repeatedly emphasized by Ho. For example see: Ho, S. P. S. (1994) op. cit., pp. 27, 29, 47.  48.  Veeck, G. (1991) op. cit.  49.  Calculated from: HDDTNJ1993, pp. 230-231.  50.  Ibid., p. 230.  51.  Veeck, G. (1991) op. cit., p. 167.  52.  I prefer the term "mega-urban region" or "mega-urbanization" in this context to capture the nature of the processes at work in the lower Yangzi delta. Terms such as "extended metropolitan region", on the other hand, imply that the spatial economic transformation is primarily driven by the growth and expansion of city based processes of metropohtanization that are pushing out into the adjacent countryside. While such processes are still important, clearly I wish to place greater emphasis on the local factors driving rural restructuring in the delta.  53.  See: Appendix 1 (April 8,1992; April 17,1992).  54.  Ibid.  55.  Data for total social product (of the material products sectors) for Kunshan was unavailable in constant prices. However, growth rates for industrial and/or agricultural output discussed below are determined from data compared in real terms.  56.  Calculated from: KSTJNJ 1989, p. 9; KSTJNJ 1991, p. 7; JSSXJJ1993, pp. 59, 83; JSSSN1949-198 pp. 317-318,393-394; SZTJNJ 1991, p. 15; SZTJNJ 1992, p. 15.  57.  Calculated from: ZGTJNJ 1993, pp. 50-51.  58.  Ibid.; See note 51.  59.  JSTJNJ 1994, p. 335.  127 60.  Ibid., p. 337.  61.  This finding is consistent with numerous recent studies which have examined industrialization and urban growth for all of China. For example: Yan, X. P. (1995) Chinese urban geography since the late 1970s. Urban geography 16 (6), p. 474 refers to several Chinese studies which demonstrate how the shift of rural labour into non-agricultural activities has largely occurred in situ. Yan also claims, however, that this occupational transfer without significant migration into urban centres is only a transitional stage. More will be said about this assertion in the conclusion. Wu, H . X. Y. (1994) Rural to urban migration in the People's Republic of China. The China quarterly 139, p. 692 has also clearly demonstrated how China's non-agricultural share of gross domestic product has grown much faster than its urban share of both total population and employment.  62.  Informants in Shipai and Bacheng townships in Kunshan, for example, reported that daytime populations rose from 4000 to 10,000 and 1500 to 8000 persons respectively, for each township seat. See: Appendix 1 (March 27,1992; June 23,1992). This is consistent with survey data reported in a study from Nanjing Normal University which shows daytime populations in 4 township seats across the lower Yangzi delta ranging between 24 percent to 209 percent larger than the resident populations. See: Nanjing Normal University, Department of Geography (1990) Jiangnan nongcun  juluoyu chengshihua yanjiu: Sunanfada diqu nongcun chengzhenhua tujing tantao (Jia agglomeration and urbanization research: Methods of inquiry into southern Jiangsu ru Nanjing: Nanjing Normal University, Department of Geography, p. 17. On any given day then, approximately a quarter to one half the total population of some towns in Kunshan, and elsewhere in the delta, commute to work in the town seat.  63.  HDDTNJ1993, p. 330; Also see Appendix 4. Individuals designated as non-agricultural (with urban registration) resided almost exclusively in urban administrative areas. As such they had access to certain benefits such as subsidized grain and other food commodities, and social welfare. However, since the beginning of reforms, the original intent and influence of these administrative designations has weakened.  64.  See: Appendix 1 (March 27,1992; June 23,1992; June 26,1992; June 30,1992); and KSTJNJ1991 p. 14. For example: While officially classified non-agricultural residents in Shipai, Penglang, and Diandong townships in Kunshan numbered 1641,1870, and 1578 persons respectively in 1991, local informants estimated the actual number of residents of these township seats to be 4000,6000, and 4000 respectively.  65.  Criteria for what constitutes a city or town in China, related definitions of urban, and changing administrative boundaries have fluctuated greatly since 1949 leading to wide variations in the estimates of the proportion of the population considered urban. For example see: Chan, K. W. (1987) Further information about China's urban population statistics: Old and new. The China quarterly 109; Chan, K. W. (1988) Rural-urban migration in China, 1950-1982: Estimates and analysis. Urban geography 9 (1); Chan, K. W. (1994) Urbanization and rural-urban migration in China since 1982: A new baseline. Modern China 20 (3); Chan, K. W. & Xu, X. Q. (1985) Urban population growth and urbanization in China since 1949: Reconstructing a baseline. The China quarterly 104; Goldstein, S. (1990) Urbanization in China, 1982-1987. Population and development review 16 (4); Kirkby, R. J. R. (1985) Urbanization in China: Town and country in a develop economy 1949-2000AD. London: Routledge; Lee, Y. S. (1989) Small towns and China's urbanization level. The China quarterly 120; Ma, L. J. C. & Cui, G. H . (1987) Administrative changes and urban population in China. AAAG 77 (3); Based on the 1990 census the most recent generally accepted estimate for the level of urbanization in China is 26.2 percent. See: The National Population Census Office (1991) Zhongguo disici renkou pucha de zhuyao shuju (Major figure fourth population census of China). Beijing: State Statistical Publishers, pp. 76, 83; Yu, D. P. (1995) Heli duliang diqu chengshihua shuiping de silu (Considerations for the rational measure of regional urbanization levels). Renkou yu jingji (Population and economics) (5), p. 42 provides an adjusted times series which suggests that China was 27.6 percent urban in 1992.  128  66.  KSXZ (1990), pp. 131-141; As with other estimates based on the official statistics for non-agricultural population, this value needs to be placed within the context of shifting definitions and mass campaigns that often dramatically affected urban and rural populations at the local level. The official non-agricultural population in Kunshan comprised between 16 to 18 percent of the total from 1949 to 1960 after which it declined gradually, reaching 10 percent in the late 1960s and where it remained until the late 1970s.  67.  Calculated from the samefiguresused above to estimate the true level of urbanization. See: Appendix 1 (March 27,1992; June 23,1992; June 26,1992; June 30,1992); md KSTJNJ 1991 p. 14.  68.  Calculated from: Ibid.; KSXZ (1990), pp. 131-141. Several recent studies provide evidence for similar findings across China: Li, M . H . (1994) Woguo renkou qianyi de liuxiang (China's migration flow). Renkou yanjiu {Population research) 18 (3), p. 51; Wang F. (1993) Woguo "sanpu" zi "sipu" jian shizhen renkou zhenzhang goucheng fenxi (Structural analysis of China's urban population growth between the third and fourth census). Renkou yanjiu {Population research) 17 (4), pp. 16-17 shows that between 1985 and 1990, more than 55 million people migrated permanently from rural areas to townships and towns within the same county. This was 4.1 times the number who migrated from provincial, prefectural, and county level units to cities and towns outside those jurisdictions; Wu, H . X. Y. (1994) op. cit., p. 697; Zhu, J. M . (1993) Zhongguo dacheng shiqu renkou zhenzhang de feizhongxinhua yanjiu (A study of the non-centralization of population growth in China's large urban areas). Renkou yanjiu {Population research) 17 (6), pp. 29-30. Meanwhile, Xu, X. Q., Ouyang, N. J. & Zhou, C. S. The changing urban system of China: New developments since 1978. Urban geography 16 (6) show that between 1978 and 1990 small and very large cities grew the most rapidly. However, since they have apparently excluded nearly 150 million urban people from their analysis of city growth, it is possible to infer that a substantial proportion of growth in the urban population of China has occurred in small towns (see their Table 3, p. 497).  69.  Calculated from: HDDTNJ1993, pp. 231, 330.  70.  See: Appendix 1 (April 8,1992).  71.  Ibid.; China daily (1991a) Kunshan races on fast track. October 27, p. 4; Kunshan People's Government (1992) Banhao nongcun gongye xiaoqu: Tigao fazhan xiangzhen  qiye {Good m of rural industrial zones will improve the development of rural enterprises). Kunshan: Ku Government.  72.  KSTJNJ 1991, pp. 7, 67,183.  73.  Ibid., pp. 125,183.  74.  Calculated from: Ibid., p. 34.  129 P A R T III  R E S T L E S S L A N D S C A P E S : R E S T R U C T U R I N G T H EK U N S H A N C O U N T R Y S I D E  This part of the thesis, comprised of Chapters Four to Six, will engage the second research question by presenting a detailed analysis of the patterns and underlying processes and mechanisms of spatial economic restructuring in Kunshan. The overall objective is to discuss all of the elements highlighted in the hypothetical model of megaurbanization not already addressed. These discussions are organized along three dimensions which correspond to each of the three chapters which follow. Chapter Four examines the relationships between local government administration, the development of local non-agricultural activities, and the administrative and institutional structures which have emerged to manage local change in Kunshan. Chapter Five discusses the findings from four specific case studies of transportation, one particular town, a village, and the numerous specially designated development zones in Kunshan. Each case study will assess crucial components of the local transformation to reveal and explore some of the underlying processes and mechanisms which drive local development. Chapter Six will focus on the growth and spatial proliferation of industrial activity in the Kunshan countryside. The analysis will highlight linkages between the processes and mechanisms identified in Chapters Four and Five, and the specific location of rural industrial enterprises. Chapter Six will conclude Part III of the thesis by reviewing how key interactions and interrelationships within the transactional environment were initiated, negotiated, and managed through intensely localized administrative and institutional structures to influence spatial economic restructuring in Kunshan.  130  Chapter Four STRUCTURE OF LOCAL GOVERNMENT AND RELATIONSHIP TO ENTERPRISES In Chapter Two I argued that the conceptual framework proposed for understanding and explaining mega-urbanization in the lower Yangzi delta should necessarily privilege local circumstances. I also suggested the need to examine the details of local institutional structures as they related to larger patterns and trends and, more specifically, to consider the activities of industrial enterprises in their institutional and transactive contexts. The processes by which institutional structures are linked to regional economic development in China are complex, usually ill-defined, and frequently quite puzzling. By undertaking an analysis that is sensitive to local cultural and historical circumstances, however, particular insights can reveal how spatial economic change is affected by evolving institutional structures, and conversely, uncover the socio-political consequences of recent reforms and the resulting transformation of the rural economy. This chapter explores the relationships between the changing structure of community administration, the development of rural non-agricultural enterprises, and the emergence of institutional structures which managed local economic activity in Kunshan. The first section begins with a review of the recent changes i n community administration and how they related to the wider hierarchy of authority and jurisdiction in China. The latter part of this section discusses how reforms in public finances stimulated local development imperatives and the emergence of rural industrial enterprises. These issues are related to the spatial proliferation of rural industrial enterprises introduced in the second section. Details of how community governments in Kunshan formalized local institutional structures and the means by which local cadres were able to influence town and village development are discussed in section three. The chapter concludes with a summary that highlights key elements of mega-urbanization, as conceptualized in Chapter Two, to illustrate how particular findings from this analysis of  131 the structure of local government and its relationship to enterprises challenges conventional views of regional development in China.  4.1 Bifurcation of the Functions of Local Government 4.1.1  Community administration O n July 27, 1989 Kunshan was administratively reclassified from a county (xiari) to  a county level city (xianji shi). Having met certain criteria which accorded it 1  economically more developed status, the new designation was intended to promote further industrialization and urban construction. The inclusion of large areas of the Kunshan countryside into "city" status (the former county boundary remained unchanged) was aimed at better integrating rural and urban sectors at the local level. A b l e to deal more directly with the province, Kunshan also experienced less interference from intermediate levels of administration since, with reclassification, it became equal in planning terms to the prefectural level city of Suzhou. This increased autonomy 2  included greater authority to collect and retain tax revenues from enterprises and to locally approve higher values of external investment in Kunshan. More than anything else, local authorities emphasized how "city" status carried greater prestige than the county designation. Since Kunshan became "more famous", outsiders would be more willing to invest here, and as a city, they could set higher standards for themselves.  3  When pressed for further details one C C P official referred to the new television station and a new ticket office for the Shanghai based China Eastern Airlines - the only one outside Shanghai at the time. In addition, more passenger and freight trains could stop in Kunshan which was "good for opening-up to the outside".  4  Kunshan's city designation also changed its internal relationship with its 20 townships, the lowest level of the urban administrative hierarchy. By 1990 all 20 had been elevated from township (xiang) to designated town (jianzhi zhen) status. A s with 5  the county level reclassification, designated towns in Kunshan acquired new authority to raise funds for local development. In addition, they were also supposed to receive 6  132 county (now city) level allocations for town construction. In practice, however, these funds were inadequate, so town governments increasingly turned to other more substantial and stable sources of revenue. Under the official notion of "town leading the development of the countryside", towns were to act as a "bridge" to revenue generating opportunities in areas of the countryside directly under their jurisdiction.  7  Figure 4.1 illustrates the location of town seats, their respective administrative regions, and Kunshan's 466 villages. Village level administrative boundaries are not shown. In addition to their respective government functions, Yushan and Chengbei town seats formed the built-up core of Kunshan, including the municipal (Kunshan) government seat, and were administered as a single urban unit by the Kunshan level bureaucracy.  8  The municipal government also directly administered at least five  specialized state farms in Zhoushi, Zhengyi, and Penglang towns, the Special Economic and Technological Development Zone straddling Yushan and Lujia towns, and the R e d Flag (Hong Qi) Industrial A r e a in Bacheng, jointly administered with Yushan. While the administrative boundaries illustrated in Figure 4.1 were spatially distinct, they also embodied complex overlapping, and often conflicting, patterns of authority, jurisdiction, and power that were much less clear. This was true not only of different levels in the administrative hierarchy, but also of the various agencies at any given level. Understanding the nature of how these patterns were constructed, negotiated, and operationalized in Kunshan reveals much about the processes and characteristics of local development. Before proceeding, however, it is necessary to provide an administrative context over and through which such patterns emerged. Table 4.1 outlines the structure and development of the Kunshan administration between 1978 and 1992. Only the main units of local government in Kunshan are listed. Excluded are the numerous sub-sections and subordinate offices of the local bureaucracy. Also excluded are the chief decision-making and executive bodies sitting atop these bureaux, the various organs of the C C P , and grass-roots organizations such as the Political Consultative Committee, and the local Congress of People's Deputies.  9  The  133  Figure 4.1 KUNSHAN: ADMINISTRATIVE DIVISIONS, 1993 Towns 1. SHIPAI 2. B A C H E N G 3. L U Y A N G 4. ZHOUSHI 5. CHENGBEI 6. XINZHEN 7. BINGXI 8. P E N G L A N G 9. ZHENGYI 10. Y U S H A N  11. LUJIA 12. H U A Q I A O 13. N A N G A N G 14. Z H A N G P U 15. QIANDENG 16. SHIPU 17. Z H O U Z H U A N G 18. JINXI 19. DASHI 20. D I A N S H A N H U 1  2  N  Scale: 6km  LEGEND Town seat • Village Kunshan Government €> Specialized State Farm Red Flag Heavy Industry Area S Special Economic & Tech. Development Zone • Built-up Area 1 1 Lakes  Notes: 1. Formerly Chenmu 2. Formerly Diandong Source: Adapted from KSXZ (1990), unpaginated preface.  —  Town Boundary  —  Kunshan Boundary  134 table illustrates, therefore, changes in the broad mid-level administration of the local government bureaucracy. Two fundamental trends are apparent: the enormous expansion in the size and range of responsibilities of the local bureaucracy; and what might be described as the corporatization of the bureaucratic-administrative mid-section of the Kunshan government. Both trends were linked to the disengagement of the central government from local administration and the reduction of state allocations which financed many of its functions. Thus, while central and provincial authorities still determined local obligations through economic and administrative policies and regulations, their financing and implementation at the county level and below were a largely local enterprise. The term "enterprise" is used deliberately here to connote the way in which local government in Kunshan bifurcated into the dual roles of community administration, and owner and manager of several companies and corporate-like economic entities. Table 4.1 lists only a few of these. In 1992 there were at least 100 companies directly or indirectly affiliated with some part of the Kunshan level government bureaucracy.  10  Some of these firms have emerged as a result of the partial commercialization and marketization of government functions within the old state run command economy structures. The most important of these companies became integral components of the various industrial and commercial bureaux and related exchange and distribution organizations under the Planning and Economic Commissions (see Table 4.1). These companies also provided a significant proportion of the financing for their respective administrative organizations, including supplementary bonuses for state employees and the full salaries of other bureaucrats not covered by state allocations.  11  Other companies  were created solely in order to increase extrabudgetary revenues for the benefit of their respective bureaux and/or their clients, engaging in activities largely unrelated to the administrative functions of local government. The Civil Affairs Bureau, for example, operated several enterprises in this category including a chemical factory and plastic works which employed more than 200 workers, 90 percent of whom were handicapped.  12  135  Table 4.1 STRUCTURE AND DEVELOPMENT OF COMMUNITY ADMINISTRATION IN KUNSHAN, 1978-1992 1978  1985  County Government Office Civil Affairs Bureau  County Government Office Civil A f f a i r s Bureau  Culture and Education Bureau  Culture and Education Bureau  Personnel Bureau  Personnel Bureau Labour Bureau Statistics Bureau  Financial Services Bureau Public Security Bureau  Financial Services Bureau Tax Bureau Public Security Bureau Justice Bureau  Grain Bureau  Grain Bureau  Agriculture Bureau  Food Office Agriculture Bureau  Water Conservancy Bureau  Water Conservancy Bureau  Agricultural Machinery Bureau Economic Diversification Management Bureau  1992 Municipal Government Office* Civil Affairs Bureau Civil Affairs Industrial Company Culture Bureau 1  Education Bureau School Industries Supply and Sales Management Section Personnel Bureau Labour Bureau Labour Services Company Statistics Bureau Price Bureau Financial Services Bureau Tax Bureau Public Security Bureau Justice Bureau Grain Bureau Cereals and Edible Oils Supply and Sales Company Food Company Agriculture Bureau Agricultural Products and Materials Company Water Conservancy Bureau Agricultural Machinery Company Economic Diversification Management Bureau Diversification Management Corporation 1  1  Economic Diversification Management Bureau  1  1  Planning Committee Industry Bureau  Planning Committee Industry and Commerce Bureau  No. 2 Industry Bureau Commerce Bureau Finance and Trade Office Supply and Sales Office  Industrial Company  Goods and Materials Bureau  Goods and Materials Bureau  and Commercial Company Goods and Materials Bureau Goods and Materials Management Company Coal and Petroleum Company  Economic Commission  Economic Commission*  Commerce Bureau Supply and Sales Office  Planning Commission Industry and Commerce Administration and Management Bureau Industry Supply and Sales Company No. 2 Industrial Supply and Sales Company Commerce Bureau Commercial Corporation Supply and Sales Cooperation Office Supply and Sales Cooperation  Industrial Wholesale Company Chemical, Pharmaceutical and Construction Materials Industries Bureau  1  Chemical and Pharmaceutical Supply and Sales Company Construction Materials Company Machinery, Electronics and Metallurgical Industries Bureau  1  Metallurgical, Machinery and Electronics Company Textiles Bureau  1  Textiles Company Light Industry Bureau Light Industry Company  CONTD  136  Table 4.1 (Cont'd) 1978  1985  Rural Industry Bureau Economic Cooperation Commission  1992  Rural Industry Bureau «1 Economic Cooperation Commission Economic and Technical Cooperation Office Economic and Technical Cooperation and Development Company 1  1  Economic System Reform Commission Economic Research Centre 1  Foreign Trade Bureau  Foreign Trade Bureau  Basic Construction Bureau  Public Health Bureau  Public Health Bureau Transportation Bureau  Transportation Bureau  Post and Telecommunications Bureau Science Committee Sports Committee Birth Control Office  Post and Telecommunications Bureau Science and Technology Committee Sports Committee Birth Control Office  Nationalities and Religious Affairs Section Foreign Affairs Office Overseas Chinese Affairs Office  Archives Bureau Environmental Protection Office  Electricity Supply Bureau Broadcast and Television Bureau Staff and Workers Education Office Suzhou Television University Salt Industry Company Tobacco Specialty Sales Bureau  People's Bank of China Agricultural Bank of China  Weather Station Insurance Company People's Bank of China Agricultural Bank of China People's Construction Bank of China Industrial and Commercial Bank  Foreign Economic Relations and Trade Commission Foreign Trade Company Special Economic and Technological Development Zone Management Committee' Industrial Trade Corporation Economic Development Corporation Urban and Rural Construction Bureau Land Management Bureau Real Estate Development Administration Public Health Bureau Transportation Bureau Transportation Corporation Post and Telecommunications Bureau Science and Technology Committee Sports Committee Birth Control Committee Standards and Measures Bureau Reception Bureau Investigations Bureau Nationalities and Religious Affairs Section Foreign Affairs Office Overseas Chinese Affairs Office Overseas Chinese Construction Office Senior Citizens Office Archives Bureau Local Gazetteer Office 1  1  Environmental Protection Bureau Environmental Protection Industry Supply and Sales Company Electricity Supply Bureau Broadcast and Television Bureau Broadcast and Television Company Staff and Workers Education Office Suzhou Television University Salt Industry Company Tobacco Specialty Sales Bureau Tobacco Company  1  Cigarettes, Sugar and Liquor Company Weather Station Insurance Company People's Bank of China Agricultural Bank of China People's Construction Bank of China Industrial and Commercial Bank Bank of Communications  Note: 1.  Interviewed  Sources: Appendix 1 (Several interviews); (1990), pp. 254-264, 514-517.  Kunshan dianhua haobu 1991 (Kunshan telephone directory 1991  137 In addition to the numerous small school-based factories which were part of the educational infrastructure and labour curriculum, the Education Bureau also operated a large and very successful enterprise with nearly 1000 workers that manufactured children's bicycles, tricycles, and strollers. Table 4.1 also illustrates the emergence of specialized organizations which were created to manage and promote exchange and distribution for the large number of enterprises which were not part of the planned economy. Agencies such as the Rural Industry Bureau, the Economic Cooperation Commission, and the Economic and Technical Cooperation and Development Office facilitated access to the means of production, technical and management expertise, and markets necessary for the successful development of local industrial enterprises. The nature of these linkages, interactions and interrelationships, and the institutional parameters which embodied them, are discussed in more detail below and in the two chapters which follow. Suffice it to say at this point that the emergence of these organizations, and their affiliated corporate-like entities, reflected dramatic changes in the local space economy. This transformation has, of course, not gone unnoticed by the central government. As a result, two other very important organizations were established in Kunshan as part of a nationwide effort to undertake structural reform: the Economic System Reform Commission; and the Economic Research Centre (see Table 4.1).  13  Both concentrated  their efforts on investigating and implementing strategies for separating government administration (and by default the C C P ) from the economy. While the corporatization of government bureaucracy in Kunshan was linked to the dramatic growth and structural changes in the local economy, the official view was that the overlap of government administrative and economic functions was unhealthy in the long term. In addition to the large number of government run companies, the fact that agencies such as the Bank of Communications and other banks, which were supposed to operate on purely commercial criteria, were considered part of the local administration, suggests that the practical realities of separating government and the economy were profoundly more complex (see  138 the bottom of Table 4.1). At the town and village levels, overlap of civil administration, CCP affairs, and management of the economy was even more apparent. Town and village CCP secretaries often served in government bureaux and frequently held directorships in the town's Economic Commission or related companies. Criteria used to assess the 14  performance of these local cadres and, therefore, their level of remuneration, have become increasingly linked to the success of local economic development and the welfare of local residents.  15  Combined with fiscal pressures to increase the revenue base and to  raise extrabudgetary funds for local economic development and social welfare, it was easy to see why town and village governments in Kunshan vigorously encouraged the expansion of rural non-agricultural activities.  16  4.1.2 Ownership and management of enterprises The rights and responsibilities of the various levels of government were officially most clearly articulated in terms of the fiscal and budgetary system.  17  In practice, the  precise functioning of community level governments depended largely upon their ability to generate extrabudgetary revenues. In Kunshan, collectively owned town and village industrial enterprises became the most important source of such revenues. Development and management of these enterprises significantly enhanced the economic and administrative power and autonomy of local governments. Part A of Table 4.2 presents the public financial revenues and expenditures at the town level in Kunshan from 1989 to 1991. The proportion of total town revenues from extrabudgetary sources rose from 31.3 percent in 1989 to 45.7 percent in 1991. Nearly three quarters of extrabudgetary revenues in 1991 was comprised of funds transferred directly to local governments from town enterprises.  18  Other extrabudgetary sources included locally collected fees for  education and training and special agricultural taxes. Town government spending showed a similar pattern with extrabudgetary expenditures comprising 65.1 percent of the total in 1991. Town enterprises provided funds for 76.4 percent of extrabudgetary  139 expenditures in 1991. This is slightly lower than the 80.5 percent figure for 1989, although the absolute value of such expenditures increased 71 percent over the same period, accounting for 71.3 percent of the increase in total extrabudgetary spending.  19  Part B of Table 4.2 shows the distribution of 1991 extrabudgetary expenditures from town enterprise raised funds. Enterprise and business expenses included management and consulting fees paid to relevant bureaux as well as other costs normally associated with operating enterprises. Social welfare expenditures included spending on culture, education and health, and public services such as community centres, retirement homes, and pensions. Local administration, which included bonuses and salaries for town cadres not covered through state allocations, accounted for the largest proportion of extrabudgetary expenditures. Payments to support agriculture were also significant (see Table 4.2). These official figures concealed other more dubious government spending on such things as banquets, liquor and cigarettes, and travel. While officially disallowed in China, such expenditures were an integral part of government and enterprise operations. Local governments also disguised other financial arrangements within such extrabudgetary ambiguities. Town governments, for example, were not permitted to run budgetary deficits. However, in Kunshan many towns financed local initiatives by securing sizable bank loans indirectly through local enterprises. Moreover, by nidging enterprise expenditures, town and village governments reduced the tax burden on their collectively owned enterprises, thereby leaving more funds available for discretionary spending.  20  Economic power and management authority in Kunshan were manifest through the operation of collectively owned community enterprises concentrated at the town and village levels.  21  Under the supervision of the Kunshan Rural Industry Bureau every town  established an industrial corporation to oversee the operations of all town and village run industrial enterprises. In addition to its supervisory role, the Rural Industry Bureau provided a range of business services to the town industrial corporations and directly to rural enterprises. The Bureau also regulated and approved the development of  140 Table 4.2 TOWN LEVEL PUBLIC FINANCES IN KUNSHAN  Part A: Revenues and Expenditures, 1989-1991 (million R M B )  REVENUES Total  1989 1990 1991  108.55 125.64 139.46  1  Budgetary  Extrabudgetary Town Enterprises  Other  24.78 31.73 47.21  9.20 14.65 16.52  24.46 33.21 41.75  5.92 11.15 12.87  74.57 79.26 75.73  EXPENDITURES 1989 1990 1991  47.45 70.36 83.94  Part B:  17.07 26.00 29.32  Distribution of Extrabudgetary Expenditures From Funds Transferred by Enterprises, 1991 (million R M B )  Total Enterprise expenses and miscellaneous business activities Social welfare Administration Agriculture  Note: 1.  1  41.75  (%) (100)  4.11 12.86 17.27 7.51  (9.8) (30.8) (41.4) (18.0)  The figures represent funds actually collected and spent for that year. Budgetary surpluses were transferred to higher levels of government. The differences between extrabudgetary revenues and expenditures were listed as net current account balances.  Sources: KSTJNJ1989, pp. 219-221; KSTJNJ1991, pp. 160-163  141 enterprises and often acted on their behalf to seek the necessary permission from other parts of the Kunshan administration, such as the Land Management and Environmental Protection Bureaux, for the construction of factories in the countryside. More importantly, the Rural Industry Bureau would facilitate investment and exchange relationships, technological and management cooperation, and the training of workers and managers.  22  Another more specific mandate of the Bureau was to promote the  development of rural industries that were linked directly to the agricultural sector and which would benefit local farmers.  23  The Bureau itself had 39 employees, only 12 of  whom were funded through state allocations. The bulk of the administration, including the 27 non-state employees, was funded through consulting fees collected from rural industrial enterprises and industrial corporations, and the revenues generated by 6 enterprises owned by the Bureau. Town industrial corporations acted primarily as the branch of town governments that actually owned and operated town enterprises. These corporations, which also existed in many villages, functioned like holding companies with a board of directors who determined enterprise activities in consultation with managers and town governments. Town and village governments, for example, selected directors for their industrial corporations who would then appoint factory managers. There was, of course, much overlap of responsibility in these respective positions since local government officials frequently served as directors or deputy heads in the Industrial Corporations, and were sometimes intimately involved in the day to day operations of specific enterprises  2 4  As  long as these enterprises made money, however, managers and the section heads they appointed usually operated with minimal interference from local governments. Similar to the Kunshan Rural Industry Bureau, industrial corporations also fulfilled other roles relating to the provision of certain services. These included market research for villages that intended to establish an enterprise, advice on the design and implementation of accounting systems, and the allocation and training of workers and management personnel. Similar to an industrial association, the corporation provided  142 "member" enterprises access to services and expertise for a fee. Through their supply and sales organizations the industrial corporations also negotiated access to inputs and markets for local enterprises. With the authority, representation, political connections, and entrepreneurial savvy of the town government supporting them, industrial corporations in Kunshan facilitated linkages between local enterprises and factory managers, emerging markets, the agricultural sector and, most importantly, the partially reformed command economy structures. In this context even the names of towns became important since the industrial corporations and many local enterprises bore the same name. Thus, the town of Chenmu in southern Kunshan, which means "old tomb" and is not an auspicious name for doing business, became Jinxi  ~  "bright and beautiful brook" ~ in 1993. That same year, the town of Diandong ("East of Dian" [Lake]) became Dianshanhu ("Dianshan Lake") Town for the same reason (see Figure 4.1).  25  In addition to its economic functions, the town industrial corporation also  provided "guidance" for the implementation of workplace safety regulations, labour standards including salaries, environmental protection rules, and supervised the finances and accounting practices of enterprises. Moreover, the industrial corporation had the power to force enterprise mergers or break-ups if circumstances were deemed appropriate.  26  Taken together, the administrative, regulatory, and economic functions of the town industrial corporations were said to provide the "internal engine of development".  27  Although town and some village governments had similar companies in the construction, commercial, and services sectors, in value terms they were less important than local industrial corporations. More about the precise transactional relationships and interactions will be discussed below, but it is clear at this point that such organizations embodied complex overlapping, apparently conflicting roles and responsibilities. The nature of the way in which these functions were negotiated and balanced reveals much about the underlying processes and mechanisms driving local development. Informants in Kunshan and other parts of the lower Yangzi delta, repeatedly emphasized the  143 importance of local industrial organizations and the way in which town and village governments "stimulated" and "encouraged" investment in local enterprises and the development of production and marketing linkages. The particular institutional parameters to which these comments referred, parallel the historical patterns of local administration outlined in Chapter Three. I refer especially to the emergence of largely autonomous community level bureaucracies, linked primarily to the water management imperatives of late imperial times, which underlie contemporary patterns of administrative jurisdiction and power. One of the more obvious outcomes of these patterns was the severe restrictions placed upon the development of household level and individual industrial enterprises in Kunshan. Under the pre-reform planned agricultural system, Kunshan, like other counties in the Lake Tai region, was under pressure to fulfill high farm production targets and government purchasing quotas. To ensure these obligations were met, local governments banned production teams and individuals from engaging in non-farm activities. They were compensated for this limitation by payments from commune and brigade enterprises through a system of workpoints.  28  However, the abolition of  transfers from industrial enterprises to production teams and the shift to direct payment of workers with implementation of the household responsibility system in agriculture i n the early 1980s, eliminated powerful disincentives for factory labour. The concomitant substitution of inefficient triple cropping with higher yielding better quality grain and cash crops generated higher farm incomes, stimulated local demand for manufactured goods, and released labour from agriculture to work in the newly more productive and rapidly expanding rural industrial enterprises. Thus, by 1984 the ban on production team (by then reorganized into village small groups, joint households and associations, or small cooperatives) run factories and other private or individual enterprises was lifted. Within a year, however, these privately run industries were posing stiff competition for commune and brigade (by then township and village) enterprises for markets, skilled personnel, and raw materials. Although private industrial enterprises were legally tolerated,  144 measures were put in place by the late 1980s in Kunshan which severely restricted access to loans and subsidized inputs, tightly controlled labour allocation and wage levels, and created a formidable and costly administrative and regulatory environment.  29  The  profitability of township and village industries meant that local authorities were reluctant to encourage the development of private enterprises, thus prohibitive obstacles against the development of sizable ones have persisted. Only in sectors such as commerce, transportation, and food services, that did not compete directly with township and village industries, were private enterprises allowed to flourish. These factors served to reinforce the concentration of economic and administrative power, ownership, and management authority at the town and village levels and helps to explain the proliferation of industrial activities into all corners of the Kunshan countryside. 4.2  Spatial Proliferation of Enterprises Since all town and village governments wished to develop their own enterprises as  a source of extrabudgetary revenues this led to the scattering of factories across the countryside. While a number of these enterprises were located in or near the town seats, most were built among the rice paddies, wheat fields, and canola crops (see Plates 4.1 and 4.2). The locational distribution of town and village enterprises was closely linked to the structure of ownership and the respective spatial extent of administrative jurisdiction.  30  The ownership distribution and gross value of output by ownership for  industrial enterprises in Kunshan and China for 1991 are illustrated in Figure 4.2. The 60 state run industrial enterprises in Kunshan were located primarily in Yushan and Chengbei Towns, and in the Red Flag Industrial A r e a (see Figure 4.1). "Other" enterprises denotes Kunshan level collectives, Kunshan run sino-foreign joint ventures, and wholly foreign owned concerns. Taken together these largely urban state run and other enterprises in Kunshan, while they comprised only 4.2 percent of the number of enterprises, were responsible for 27.5 percent of gross industrial output. Across China the largely urbanized state sector enterprises, though few in number, were responsible  145 Plate 4.1 RURAL ENTERPRISES Shipai Town, Kunshan  Plate 4.2 SURGICAL GLOVE FACTORY Yushan Town, Kunshan  146  Figure 4.2 DISTRIBUTION OF OWNERSHIP AND GROSS VALUE OF OUTPUT BY OWNERSHIP OF INDUSTRIAL ENTERPRISES: KUNSHAN AND CHINA, 1991 Part A) Ownership Kunshan: 3945 enterprises  China: 8.0769 million enterprises  State  Part B) Gross Value of Output Kunshan: GVIO: 7.25 billion RMB (1991 current value) China: GVIO: 2824.8 billion RMB (1991 current value)  Sources: Calculated from: KSTJNJ 1991, p.67;  ZGTJNJ1991,  p.403.  147 for 52.9 percent of the nation's industrial output.  1  output was the contribution of the private sector.  32  Excluded from these totals for gross While individual industrial  enterprises made up 55.1 percent of the total number in Kunshan, they produced only 0.8 percent of industrial output, commensurate with their relative size and importance. More than 89 percent of the 2172 individually owned industrial enterprises were dispersed among Kunshan's 466 villages.  33  China's 6.39 million individual industrial  enterprises produced 5.7 percent of gross output with nearly 93 percent of them classified at the village l e v e l .  34  The most important enterprises in Kunshan were the town and  village level industries which together comprised 40.7 percent of the total number and 71.7 percent of the gross value of output (see Figure 4.2). Across China, township and village enterprises represented 17.5 percent of the total number and accounted for 23.1 percent of gross output. The figures for town enterprises in Kunshan included town collectives and a small number of town run sino-foreign and other joint ventures. Village industries included collective and cooperative enterprises. Virtually all of these 1071 village enterprises were established in rural locations within the jurisdiction of village administrations, while the 537 town enterprises were located near the town seats and throughout the towns' administrative a r e a s  35  Linked to the desire of each administrative jurisdiction to maximize local productive opportunities, the sectoral structure of industry across towns and villages in Kunshan has also diversified. Figure 4.3 illustrates the sectoral distribution of the number of enterprises, employees, and output values of the enterprises surveyed i n Kunshan in 1992. While textiles and related activities were the most important sector, accounting for 34 percent of output value, 31.1 percent of employees, and 25.9 percent of the number of enterprises, chemical, metal fabrication, machinery, building materials, and consumer goods industries also proliferated.  36  This sectoral breakdown of industrial  activities closely parallels the distribution of such activities across the entire Suxichang region  3 7  Informants referred to numerous examples of how town and village  governments, anxious to achieve the same success as neighbouring communities, invested  148  Figure 4.3 SECTORAL DISTRIBUTION OF ENTERPRISES, EMPLOYEES, A N D OUTPUT: K U N S H A N 199l'  Textiles & Garments Chemicals  Notes:  2  3  Percentage  1. One shoe and two textiles factories were too newtoreport a 1991 output. 2. Includes weaving and dyeing, and the production of cloth. 3. Includes industrial chemicals, pharmaceuticals, paints, dyes,etc. 4. Includes household appliances, shoes, toys, furniture, etc. 5. Includes a very large honey manufacturer and a large food commodities trading firm, each responsible for 7.3 percent and 6.5 percent respectively of the total output of the enterprises surveyed, in additiontoseveral smaller service companies engaged in transportation, vehicle repair, etc.  Source: Calculated from enterprise survey data (see Appendix 3)  149 in virtually identical activities. In addition to the development of broadly similar 8  industrial structures across the region, this spontaneous and haphazard growth created enormous problems related to the provision of infrastructure, duplication, and the waste of capital and land. With typical Chinese panache, allusions to such conditions were captured in a local slogan: cun cun dianhuo, chu chu maoyan (in every village fires stir, and everywhere is belching smoke) (cf. Plate 4.1).  39  While such industrial development  was "comprehensive" and relatively successful at the local scale, in regional terms (county level and higher) it remained "irrational" and spatially scattered.  40  In conceptual terms, the diverse structure and spatial proliferation of industrial activities in Kunshan provides further evidence that the rural transformation observed here occurred largely as a response to intensely localized development imperatives, rather than serving merely the exigencies of external (metropolitan) demand or other exogenous forces  4 1  The resulting density of residential, industrial, and agricultural land  use in rural Kunshan is evident in Plate 4.3. The view is to the west across Chengbei and Xinzhen Towns as of mid-1992. It illustrates the classic linear patterns and architecture of southern Jiangsu villages, the rich agricultural lands typical of the region, and examples of some township enterprises. As will be shown in the next chapter, most of the fields in the middle-ground have since been converted into village run industrial enterprises. Official concern with the loss of such highly productive land prompted the creation of specially designated agricultural protection zones in several towns in Kunshan (see Plate 4 . 4 )  42  The sign shown in Plate 4.4 outlines the distribution of land quality and  the proclamation from the Shipai Town People's Government that decreed its preservation for agricultural production. That such measures were necessary in one of China's richest farming regions says much about the proliferation of non-agricultural activities and hints at the inherent tension