COMMENTS ON THE PREVIOUS EDITION ‘I am very impressed with the book’s empirical and theoretical range.’ Mark Bouman, Professor and Department Chair, Chicago State University ‘An excellent book, monumental in scope, breathtaking in depth…a great achievement.’ Christopher Smith, Professor of Geography and Planning, SUNY at Albany ‘This remains the most definitive text and the best introductory one.’ Kevin Ward, Senior Lecturer in Geography, University of Manchester ‘It has huge depth and breadth of coverage, with which no other text can (or probably wants to) compete.’ Tim Coles, Lecturer in Human Geography, University of Exeter ‘This text is more comprehensive than any text I know in the subject area. It is nearly 700 pages long and is a tour de force.’ Linda Stainer, School of Geographical Sciences, University of Derby Urban Geography: A Global Perspective offers the most contemporary, comprehensive, and insightful presentation on urban geography, a compelling readable, spectacularly exciting, and pleasingly sophisticated investigation of an extensive range of vital urban issues at the local and global scales. It is a must-read for any urban student and scholar the world over. James O.Wheeler, The Merle Prunty Jr. Emiritus Professor of Geography, University of Georgia An impressive overview of the urban process, from the origins of cities to the problems of contemporary urban regions in the developed and developing worlds. L.S.Bourne, Professor of Geography, University of Toronto Michael Pacione’s Urban Geography: A Global Perspective provides a comprehensive global view of urban geography that is chock full of information and lavishly illustrated with maps, photographs, and illustrations. At the same time it is theoretically smart, rich in historical detail, and questioning about urban futures in a global world. It is the ideal text for undergraduate courses in urban geography and urban studies. Professor Neil Smith, City University of New York I heartily endorse Pacione’s new text. He has set himself the daunting task of introducing students to the full breadth and richness of urban geography and rises magnificently to the occasion. Professor A.G.Champion, University of Newcastle First edition published 2001, Second edition 2005 by Routledge 2 Park Square, Milton Park, Abingdon, Oxon 0X14 4RN Simultaneously published in the USA and Canada by Routledge 270 Madison Ave, New York, NY 10016 Routledge is an imprint of the Taylor & Francis Group This edition published in the Taylor & Francis e-Library, 2005. “ To purchase your own copy of this or any of Taylor & Francis or Routledge’s collection of thousands of eBooks please go to http://www.ebookstore.tandf.co.uk/.” © 2005 Michael Pacione All rights reserved. No part of this book may be reprinted or reproduced or utilised in any form or by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying and recording, or in any information storage or retrieval system, without permission in writing from the publishers. British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library Library of Congress Cataloging in Publication Data Pacione, Michael. Urban geography: a global perspective/ Michael Pacione.—2nd ed. p. cm. Includes bibliographical references and index. 1. Urban geography. I. Title. GF125.P33 2005 910′.9173′2–dc22 2004015892 ISBN 0-203-02352-8 Master e-book ISBN ISBN 0-415-34305-4 (hbk) ISBN 0-415-34306-2 (pbk) << urban_geography >> second edition Today for the first time in the history of humankind, urban dwellers outnumber rural residents. Urban places, towns and cities are of fundamental importance for the distribution of population within countries; in the organisation of economic production, distribution and exchange; in the structuring of social reproduction and cultural life; and in the allocation of power. The second edition of Urban Geography continues to provide an authoritative and stimulating, global introduction to the study of towns and cities. The text synthesises a wealth of material to provide unrivalled depth and breadth for students of urban geography drawing on a rich blend of theoretical and empirical information with which to advance knowledge of the city. The new edition has been extensively revised to reflect feedback from users and to incorporate the latest research and developments in the field. The text is divided into six main parts that explain and discuss: ■ the field of urban geography and the importance of a global perspective; ■ the historical growth of cities from the earliest times and the urban geography of the major world regions; ■ the dynamics of urban structure and land-use change in Western cities; ■ economy, society and politics in the Western city; ■ the economic, social, political and environmental challenges faced by the third world city; and ■ an overview on the future of cities and cities of the future. Urban Geography is superbly illustrated with more than 150 figures, more than fifty photographs and a colour plate section. There are also over 180 case studies and explanatory boxes, drawing insights from across the globe. The text contains a glossary of key terms and words, chapter summaries, key points, study questions and annotated further reading. Michael Pacione is Professor and Chair of Geography at the University of Strathclyde in Glasgow. URBAN GEOGRAPHY A GLOBAL PERSPECTIVE Second edition MICHAEL PACIONE Routledge Taylor & Francis Group LONDON AND NEW YORK TO CHRISTINE, MICHAEL JOHN AND EMMA VICTORIA Contents List of illustrations ix List of tables xxii List of boxes xxix Preface to the second edition xxxix Acknowledgements xl Copyright acknowledgements xli Introduction vl PART ONE: THE STUDY OF URBAN GEOGRAPHY Chapter 1 Urban geography: from global to local Chapter 2 Concepts and theory in urban geography PART TWO: AN URBANISING WORLD 1 2 22 45 Chapter 3 The origins and growth of cities 46 Chapter 4 The global context of urbanisation and urban change 88 Chapter 5 Regional perspectives on urbanisation and urban change 124 Chapter 6 National urban systems 159 PART THREE: URBAN STRUCTURE AND LAND USE IN THE WESTERN CITY 179 Chapter 7 Land use in the city 185 Chapter 8 Urban planning and policy 223 Chapter 9 New towns 256 Chapter 10 Residential mobility and neighbourhood change 274 Chapter 11 Housing problems and housing policy 297 Chapter 12 Urban retailing 326 Chapter 13 Urban transportation 359 PART FOUR: LIVING IN THE CITY: ECONOMY, SOCIETY AND POLITICS IN THE WESTERN CITY 384 Chapter 14 The economy of cities 385 Chapter 15 Poverty and deprivation in the western city 416 Chapter 16 National and local responses to urban economic change 447 Chapter 17 Collective consumption and social justice in the city 476 Chapter 18 Residential differentiation and communities in the city 499 Chapter 19 Urban liveability 541 Chapter 20 Power, politics and urban governance 571 PART FIVE: URBAN GEOGRAPHY IN THE THIRD WORLD 612 Chapter 21 Third World urbanisation within a global urban system 618 Chapter 22 The internal structure of Third World cities 642 Chapter 23 Rural urban-migration in the third world 667 Chapter 24 Urban economy and employment in the Third World 688 Chapter 25 Housing the Third world poor 708 Chapter 26 Environmental problems in the Third world city 748 Chapter 27 Health in the third World City 769 Chapter 28 Traffic and transport in the Third World city 787 Chapter 29 Poverty, power and politics in the Third World city 801 PART SIX: PROSPECTIVE: THE FUTURE OF THE CITY—CITIES OF THE FUTURE Chapter 30 The future of the city—cities of the future 822 823 Notes 853 Glossary 912 Index 931 Illustrations COLOUR PLATES Plates 1–8 between pages 138 and 139; Plates 9–16 between pages 450 and 451. 1 The ancient city: Thebes 2 The fantasy city: Las Vegas 3 The post-industrial city: Glasgow 4 The modern industrial city: Pusan, South Korea 5 The global city: Tokyo 6 The mobile city: San Diego 7 The informal city: Bombay 8 The congested city: Hong Kong 9 The squatter city: Rio de Janeiro 10 The post-colonial city: Singapore 11 The contested city: Belfast 12 The multicultural city: London 13 The emblematic city: Sydney 14 The unregulated city: Rome 15 The defensive city: New York 16 The emerging global city: Shanghai BLACK AND WHITE PLATES 1.1 Marks and Spencer’s store in Hong Kong 2.1 Waterfront living in Manta, Ecuador…and St Katharine’s Dock, 31 London 3.1 The pyramid at Chichen Itza, Mexico 53 3.2 The Piazza San Pietro, Rome 61 3.3 Terraced housing in the Elswick district of Newcastle upon Tyne 69 3.4 The social polarisation of the postmodern city, Denver CO 82 4.1 The exploding postmodern metropolis of Los Angeles CA 118 5.1 The Banco Popular in Harlem, New York City 128 5.2 The ancient Islamic city and twentieth-century new town, Kano, 155 Nigeria 7.1 High-rise and high-status in San Francisco CA 190 7.2 Late modern architecture in Phoenix AZ 217 8.1 Canary Wharf tower in London 241 8.2 Public housing in the socialist city: Moscow 250 9.1 A private retirement community, Sun City Center FL 271 10.1 Moving home in the USA 9 275 10.2 Gentrification on Butler’s Wharf, London 289 11.1 Public housing in a British city: Glasgow 305 11.2 Council tower blocks and housing association tenements, Glasgow 319 12.1 The Rialto bridge, Venice and Las Vegas NV 327 12.2 A combined retail-recreational experience, Las Vegas NV 333 12.3 The MetroCentre, Gateshead, one of Europe’s largest covered malls 350 13.1 Rush-hour on the Chuo line, Tokyo 366 13.2 The end of the road for an urban freeway, San Francisco CA 367 14.1 Commuters and tourists on Westminster Bridge, London 395 14.2 Deindustrialisation on the River Clyde, Glasgow 406 15.1 The polarised city: London 423 15.2 A burnt-out car by local-authority flats in Possilpark, Glasgow 441 16.1 The redevelopment of Baltimore’s Harborplace 457 16.2 Culture-based urban redevelopment, Gastown, Vancouver 460 17.1 Protest at environmental pollution, New Orleans LA 492 18.1 An exclusive guard-gated community in Nevada 518 18.2 Bangladeshis in Tower Hamlets, London 538 19.1 A protected public space on Manhattan Island, New York City 563 20.1 Low-density residential zoning, Cherry Hills, Denver CO 588 20.2 Social commentary outside the US Mint in Denver CO 604 21.1 Sir Stamford Raffles and a symbol of post-industrial society, Singapore 628 21.2 The Burj al Arab hotel, Dubai 635 22.1 Housing in the indigenous part of Kano, Nigeria 649 22.2 The King Abdullah Mosque prominent in the Islamic city of Amman, Jordan 654 23.1 Kejeta bus station, Kumasi, Ghana, a point of entry for ruralurban migrants 673 23.2 Social polarisation in Beijing: the housing of migrant workers and of middle-class suburban dwellers 684 24.1 A Mexican girl prepares cactus leaves for sale from her doorstep 693 24.2 A worker in a female labour-intensive workshop in Beijing 699 25.1 Large-scale public housing in Singapore 712 25.2 Homeless people shelter in drainpipes, Calcutta 713 25.3 Self-constructed squatters’ housing, São Paulo 721 26.1 Water-collection for domestic use, a daily task in Nairobi 751 26.2 Apartment blocks shrouded in smog, central Beijing 760 26.3 Flooding in Jakarta 764 27.1 Street vendors in Hong Kong 778 28.1 Paratransit in Bangkok 793 28.2 Pedal power: the chief form of personal transport in China 795 FIGURES 1.1 Triggers, processes and outcomes in urban geography 3 2.1 The nature of urban geography 23 3.1 Early urban hearths 51 3.2 The Fertile Crescent and ancient cities of the Middle East 52 3.3 Typical form of a Roman imperial settlement and a generalised example from Roman Britain 55 3.4 The urban structure of medieval Lübeck 60 3.5 Burgess’s model of the industrial city 64 3.6 The processes of residential sorting by class in the nineteenth- 66 century British city 3.7 Layout of back-to-back housing in Manchester 70 4.1 The urban world in 1890 89 4.2 The urban world in 1950 89 4.3 The urban world in 1990 90 4.4 The urban world by 2025 90 4.5 The urban population of world regions, 1970, 1994 and 2025 91 4.6 Average annual rate of urban population growth, 1990–5 95 4.7 Generalised stages of differential urbanisation 104 4.8 The stages of urban development model 106 4.9 Megalopolises of the USA 119 4.10 The megalopolitan trend in Western Europe 120 4.11 Doxiadis’s concept of ecumenopolis 121 5.1 Major metropolitan areas in North America 127 5.2 Migration flows for selected US immigrant magnet metropolitan areas 130 5.3 Major urban centres in Latin America and the Caribbean 135 5.4 Major urban centres in Western Europe 140 5.5 Major urban centres in Eastern and Central Europe 143 5.6 Major urban centres in Asia 147 5.7 Major urban centres in Africa 154 6.1 Deriving the hexagonal pattern of market areas for central places 167 6.2 Hierarchical and spatial arrangement of central places 168 6.3 The changing settlement pattern of East Flevoland 171 6.4 Bylund’s model of settlement diffusion 173 6.5 Vance’s mercantile model of urban development 175 7.1 The fringe belts of Newcastle upon Tyne 187 7.2 Burgess’s concentric-zone model of urban land use, applied to 189 Chicago 7.3 Hoyt’s sector model of urban land use, and its application in Sunderland 193 7.4 Harris and Ullman’s multiple-nuclei model of urban land use 193 7.5 Mann’s model of a typical medium-size British city 197 7.6 Kearsley’s modified Burgess model of urban land use 198 7.7 Vance’s urban realms model 199 7.8 White’s model of the twenty-first-century city 200 7.9 Harvey’s model of the circulation of capital 202 7.10 Principal actors in the production of the built environment 208 8.1 Plan of the model settlement of Bournville 229 8.2 Structure of Howard’s garden city 230 8.3 Residential density profiles in selected European cities 249 9.1 The Abercrombie plan for Greater London, 1944 257 9.2 The British New Towns 258 9.3 Proposed private-sector residential communities in South-East 262 England 9.4 Evolution of new communities in the USA 267 9.5 Locations of major new communities in the USA 268 10.1 Reasons for household relocation 275 10.2 A value expectancy model of migration decision-making 277 10.3 A stress model of the residential-location decision process 278 10.4 Downs’ continuum of neighbourhood change 285 10.5 The concept of the rent gap 291 11.1 Public-housing policy in the USA 308 11.2 Types of accommodation in a post-industrial society 314 12.1 Berry’s classification of urban retail locations 331 12.2 A typology of the contemporary urban retail system 332 12.3 Plan of the Mall of America, Bloomington MN 340 12.4 Retail centres in San Antonio TX 341 12.5 New shopping centre construction starts in the USA 342 12.6 Central-area decline and suburban retail growth in the US Midwest, 1958–63 343 12.7 Stages in the evolution of town-centre shopping schemes in the UK 348 12.8 British shopping-centre development: total floor space opened 349 each year, 1965–93 12.9 The layout of out-of-town regional shopping centres in the 353 UK: the MetroCentre, Gateshead and the Meadowhall Centre, Sheffield 12.10 Major out-of-town shopping centres open or under development in the UK 355 13.1 A typology of commuting flows 363 13.2 Dimensions of the urban transport problem 364 13.3 The relationship between car-ownership and bus-ridership 368 13.4 Models of the relationship between transport policy and urban 376 structure 13.5 The mass rapid transit system and concentrated decentralisation in Singapore 379 13.6 Plan of a transit-oriented community 380 15.1 Anatomy of multiple deprivation 419 15.2 A cumulative causation model of the underclass phenomenon in US cities 423 15.3 Incidence of all offences and property offences in England and 424 Wales, 1950–92 15.4 Male mortality rates by social class and age group in England and Wales, 1976–81 426 15.5 Major sources of poverty among ethnic minorities 427 15.6 Deprived households as a percentage of all households in UK local-authority districts 433 15.7 The geography of deprivation in London 434 15.8 Share of population living in underclass neighbourhoods and extreme poverty neighbourhoods, by region, ethnic distribution and urban location 439 15.9 The geography of deprivation in Glasgow 443 16.1 Major urban policy initiatives in the UK 448 16.2 The geography of urban policy in the UK 451 16.3 Major federal urban programmes in the USA 453 16.4 A general model of economic activity within a developed economy 469 17.1 Determining whether a social process is an element of collective consumption 476 17.2 The hypothetical time-space prism of a suburban housewife 487 17.3 The daily time-space prism of a homeless woman on skid row, 487 Los Angeles CA 18.1 The Gold Coast and the slum in Chicago, 1929 501 18.2 The method of social-area analysis 502 18.3 Social areas in the San Francisco Bay region, 1950 503 18.4 The method of factorial ecology 504 18.5 The components of urban ecological structure 507 18.6 The relationship between life cycle and residential mobility 516 18.7 A model of the assimilation process 524 18.8 Residential spatial outcomes of ethnic segregationcongregation 527 18.9 Residential shifts of Jewish population in Cincinnati OH 528 18.10 Distribution of Wolverhampton’s population born in the New Commonwealth or Pakistan 530 18.11 African-American residential expansion in Atlanta GA, 1950– 531 90 18.12 Spatial process of African-American suburbanisation in the US city 532 18.13 Distribution of immigrant ethnic populations in London 537 19.1 A stress model of urban impact 546 19.2 The elements of a legible city and their application to the urban form of Boston MA 556 19.3 The relationship between crime and environmental characteristics, according to Newman 560 20.1 Forms of urban local government 576 20.2 Spatial structure of local government in the UK 580 20.3 Political geography of the Denver Standard Metropolitan 587 Statistical Area: municipalities of the Denver urban area and land-use zoning in Cherry Hills 20.4 Forms of inter-jurisdictional co-operation: the Metro Denver Network and the Louisville-Jefferson County Compact 594 20.5 Hypothetical example of malapportionment 598 20.6 Gerrymandering in a proposed reapportionment of electoral district boundaries in Philadelphia PA 599 20.7 Arnstein’s ladder of citizen participation 601 21.1 Frank’s model of dependent development, illustrated with reference to the USA and Brazil 623 21.2 Wallerstein’s world systems model 624 22.1 A model of the Latin American city 643 22.2 A model of Latin American urban land use 645 22.3 A general model of the African city 646 22.4 The pre-apartheid segregated city 650 22.5 The apartheid city 651 22.6 Structure of the Islamic city 652 22.7 A model of the colonial-based city in South Asia 656 22.8 A model of the bazaar-based city in South Asia 657 22.9 A model of the South-East Asian city 659 22.10 A model of the Indonesian city 660 22.11 A model of the contemporary Chinese city 664 23.1 672 Changing roles of development-related factors in migration 23.2 The migration decision process 678 24.1 The structure of the Third World urban economy 690 24.2 Continuum of employment relationships in the Third World urban economy 695 24.3 The household economy in the Third World city 705 25.1 Major sources of housing for the Third World urban poor 710 25.2 Migration patterns among low-income neighbourhoods in the Third World city 718 25.3 Turner’s typology of squatters 724 25.4 The Klong Toey land-sharing project in Bangkok 740 26.1 Health hazards of garbage-picking 758 27.1 An integrated approach to community health in Addis Ababa, Ethiopia 781 28.1 The relationship between urban form and dominant mode of transport 790 28.2 A protocol for assessing transport options 798 29.1 Relations of power within a society 802 29.2 The bases of social power 805 30.1 Goals of sustainable development 825 30.2 Major dimensions of urban sustainability 827 30.3 The relationship between urban population density and gasoline consumption 838 30.4 Concentrated and decentralised concentrated forms of urban development 840 30.5 Remodelling the automobile-dependent city 840 30.6 Le Corbusier’s ideal city: la ville radieuse 842 30.7 Soleri’s three-dimensional city: Babel II D 843 30.8 Three urban configurations 844 Tables 1.1 Themes and questions in urban geography 13 2.1 Definitions of urban areas in the USA and UK 25 2.2 The expanding scope of urban geography 31 2.3 Analytical value of different theoretical perspectives in urban geography 39 3.1 The fastest-growing US cities and biggest population losers, 1980–90 79 4.1 Urban population and percentage urban in more developed and less developed regions, 1970, 1994 and 2025 92 4.2 Average annual rate of change of urban population, 1965–70, 1990–5 and 2020–5 94 4.3 Urban population, number of cities and percentage of urban population by city-size class: world, more developed regions and less developed regions 96 4.4 The fifteen largest urban agglomerations, ranked by population 98 size, 1950, 2000 and 2015 4.5 Number of megacities, 1970, 1994, 2000 and 2015 101 4.6 Regional distribution of the world’s population in ‘million cities’ and the location of the world’s largest 100 cities 101 4.7 Stages of development of a daily urban system 107 4.8 Great Britain: population change, 1951–91, by functional region 107 zone 4.9 Types of population change in metropolitan areas of the USA, 1980–90 and 1990–6 108 4.10 City and downtown population changes in the USA, 1990–2000 110 4.11 Changing population and employment distributions in the USA, 111 1950–2000 4.12 Population change in US central cities, 1960–2000 114 4.13 Alternative interpretations of the suburbanisation process 116 5.1 Ranking of US metropolitan areas by population, 2000 126 5.2 Net domestic migration and movers from abroad, 1995–2000, for the largest US metropolitan areas 129 5.3 Latin America: urban populations for 1990 and 2001 and urban 132 change since 1950 5.4 Urban populations and levels of urbanisation for all West European countries with 1 million or more inhabitants 137 5.5 Functional types of cities in Western Europe 139 5.6 Urban population in Eastern Europe in 1992 and 2001, and urban change since the 1930s 142 5.7 Levels of urbanisation of the former Soviet republics 1959, 1989 and 2001 144 5.8 Population and urban change, 1950–2001, in Asian countries with 10 million or more inhabitants 146 5.9 Urban population change in African countries with 1 million or 151 more inhabitants, 1950, 1990 and 2001 6.1 Classification of city systems based on levels of closure and interdependence 160 6.2 Classification of British towns 162 6.3 Classification of US metropolitan areas by type and size 163 6.4 US city types 164 6.5 Characteristics of central places in southern Germany 168 7.1 Assumptions and principles of the concentric-zone model of urban land use 189 8.1 The main goals of planning 225 8.2 Planned settlements in eighteenth- and nineteenth-century Britain 227 8.3 Major reports and legislation leading to the 1947 planning system in the UK 233 8.4 Main phases in British post-war urban policy 237 8.5 Main advantages and disadvantages of selected growthmanagement techniques 245 10.1 Ten life-cycle events related to residential adjustment or relocation 276 10.2 Housing groups in the London borough of Southwark 281 10.3 Summary of neighbourhood life cycles 284 10.4 Factors underlying neighbourhood decline and revitalisation 287 10.5 Demand-side and supply-side factors underlying neighbourhood 293 revitalisation 11.1 Housing tenure and dwelling stock in Great Britain, 1914–2001 298 11.2 UK dwelling completions by tenure, 1946–2001 299 11.3 US public-housing completions, 1939–91 310 15.1 The geography of poverty in the USA 417 15.2 Principal models of urban deprivation 417 15.3 The thirty-four US cities past the ‘point of no return’ 435 15.4 Demographic profile of south central Los Angeles, 1990 437 15.5 A typology of explanations of inner-city decline 438 15.6 Suburbanisation of inner-city problems in the USA, 1980–90 444 16.1 Problems and opportunities of culture-led regeneration 461 17.1 Pressures on the welfare state 478 17.2 Forms of restructuring in public-sector service provision 480 18.1 Factorial ecology of Clydeside: factor structures and loadings 504 18.2 Measurement of neighbourhood cohesion in Glasgow 513 18.3 Forms of social disorganisation in the slum 516 18.4 Ethnic composition of central city and non-central city populations in the USA, 1990 and 2000 533 18.5 Ethnic composition of the British population, 2001 535 19.1 The top ten places to live in the USA, according to Money Magazine, 1997 542 20.1 Types of metropolitan government 573 20.2 Number of local governments in the USA, 1942–2002 580 20.3 Number of municipalities in selected US counties, 1920–70 583 20.4 Metropolitan restructuring processes in the USA 596 20.5 Characteristics of successful grass-roots groups 608 12.1 Hierarchy of planned shopping centres 333 12.2 Changes in shopping-centre land values in the Chicago metropolitan area 345 12.3 The impact of town-centre shopping schemes in Britain 348 12.4 Location and size of enclosed centres of over 500,000 ft2 in Britain 351 13.1 The relationship between transport and urban form in Western cities 359 13.2 Changing mode of journey to work in the USA, 1990–2000 361 13.3 Number of vehicles, Great Britain, 1950–2000 362 13.4 Proportion of households with regular use of a car, Great Britain, 1951–2000 362 13.5 Changes in modal split of work trips in the restricted zone, Singapore 372 13.6 Types of alternative work schedules 374 14.1 The post-war ‘Fordist’ expansionary regime of the late 1940s to 392 early 1970s 14.2 The ‘post-Fordist’ regime of flexible accumulation from the mid-1970s to the present 393 14.3 Structure of world output, 1960–90 393 14.4 World output by sector and major region, 2000 394 14.5 Roster of world cities 402 14.6 Employment change by sector in major British cities, 1981–91 411 14.7 Employment structure of major British cities, 2001 412 21.1 Stages of colonial urbanisation in Asia 626 23.1 Principal reasons for migration from rural North-East Thailand 668 23.2 The demographic and mobility transitions among modernising populations 669 23.3 Average annual rates of rural out-migration in Latin America 674 23.4 Principal policy responses to rural-urban migration in the Third 681 World 24.1 The nature of the two circuits in the Third World urban economy 691 24.2 Types of informally organised market-oriented activities in the Third World urban economy 694 25.1 Indicators of housing quality in high- and low-income countries 708 25.2 Types of rental housing used by lower-income groups in many Third World cities 714 25.3 Examples of ‘owner-occupation’ housing used by low incomegroups in many Third World cities 719 25.4 Proportion of city population in extra-legal housing in various Third World cities 728 25.5 Principal means by which people obtain land for housing in Third World cities 729 25.6 Different attitudes by governments to housing problems in cities and different policy responses 734 25.7 Positive and negative features of illegal land subdivision for low-income housing 743 26.1 Major environmental problems in Third World cities 749 26.2 Household environmental conditions in selected cities 754 27.1 Government-community actions to combat health risks in Third 782 World cities 28.1 Number of passenger cars per thousand inhabitants in selected countries 788 30.1 The multiple goals of sustainable development as applied to cities 828 30.2 The metabolism of Greater London 831 Boxes 1.1 Principal characteristics of globalisation 8 2.1 The assumptions of positivism 33 3.1 Childe’s ten characteristics of an urban civilisation 47 3.2 Preconditions for pre-industrial urban growth 48 3.3 Ur of the Chaldees 52 3.4 A model of the pre-industrial city 56 3.5 Plan of Venta Silurum (Caerwent) 57 3.6 The merchant class and urban growth in medieval Europe 58 3.7 The circular and cumulative model of urban growth 63 3.8 Socio-spatial segregation in nineteenth-century Liverpool 66 3.9 The slum district of Little Ireland in Manchester, 1849 69 3.10 Living conditions in the tenement slums of mid-nineteenthcentury Glasgow 71 3.11 Tenement life in Glasgow 73 3.12 Philadelphia: a seventeenth-century planned town 75 3.13 The township and range system of land division 77 3.14 New York: the urban explosion 78 3.15 Los Angeles: the archetypal postmodern metropolis 84 4.1 The demographic-transition model 92 4.2 Tokyo: global megacity 99 4.3 Overurbanisation 102 4.4 The urbanisation curve 103 4.5 The law of the primate city 105 4.6 Edge City, USA: Tyson’s Corner VA 113 4.7 The changing ethnic composition of US central cities 115 5.1 Urban restructuring and the effect of the Mexico-USA border 136 5.2 The economic and urban development of Shanghai 148 5.3 Australasian cities 149 5.4 Underurbanisation in China 153 5.5 The deteriorating infrastructure of African cities 155 5.6 Informal employment in African cities 156 6.1 Urban systems at the intra-national scale 159 6.2 Assumptions underlying Christaller’s central place theory 165 6.3 The missing middle in the Third World urban hierarchy 176 7.1 Land-ownership and the development of the street pattern in eighteenth-century Glasgow 188 7.2 Burgess’s concentric-zone model of urban land use 191 7.3 The trade-off model of urban land use 195 7.4 The over-accumulation crisis and post-war suburbanisation in the USA 204 7.5 Major agents and motives in the capitalist land-development process 206 7.6 Housing submarkets in Baltimore MD 210 7.7 Externalities 214 7.8 BIDs to privatise public space in New York City 220 8.1 The nature of urban policy 224 8.2 The value of planning 225 8.3 Urban growth management: the Dutch approach 232 8.4 Incentive zoning in Seattle 234 8.5 Land-use conflict in the green belt 235 8.6 Common forms of land-use zoning in the USA 243 8.7 Major principles of smart growth 246 8.8 Marxist-Leninist principles underlying the Soviet socialist city 247 8.9 The Soviet micro-region 248 9.1 Milton Keynes New Town 260 9.2 Where will the new houses go? 263 9.3 The new towns of the Paris region 265 9.4 New urbanism: principles of neighbourhood development 271 9.5 Columbia MD 272 10.1 The concept of housing class 281 10.2 The arbitrage model of neighbourhood racial transition 282 10.3 Housing abandonment in St Louis MO 286 10.4 Geography and gentrification in London: a tale of two postcodes 290 11.1 Rent controls 302 11.2 The Hope VI programme 311 11.3 Sleeping rough in Toronto 315 11.4 Homeless in Hollywood 316 12.1 The Minneapolis skywalk system 335 12.2 The anatomy of the shopping mall 339 12.3 Power centres in the USA 340 12.4 The West Edmonton Mall, Edmonton, Alberta 341 12.5 The nature of commercial blight 344 12.6 The polarisation of retail provision in central Chicago 346 12.7 Retail decentralisation in the UK 354 13.1 Major eras of metropolitan growth and transport development in the USA 360 13.2 Accessibility and mobility 364 13.3 Public transport in Curitiba, Brazil 369 13.4 The political feasibility of high-occupancy vehicle lanes 372 13.5 A transit-oriented community: Tama new town, Japan 378 13.6 Pleasant Hill CA: transit village 379 14.1 The advent of megabyte money 386 14.2 A typology of technopoles 390 14.3 Offshoring: a new international division of labour 395 14.4 Unemployment on the Kirkby estate in Liverpool 398 14.5 World city characteristics 399 14.6 London as a world city 404 14.7 Local economic restructuring in Glasgow 408 14.8 Restructuring the Pennsylvanian economy 410 15.1 Life on a low income 421 15.2 Food poverty in the UK 428 15.3 The Lower East Side People’s Federal Credit Union 430 15.4 The South Shore Community Development Bank 430 15.5 Grameen borrowing circles in Chicago 431 15.6 In the depths of the ‘inner city’ 440 16.1 The London Docklands Development Corporation: area, powers and objectives 451 16.2 Detroit’s empowerment zone plan 454 16.3 Creating a better business climate in Houston TX 455 16.4 A tale of two Baltimores 458 16.5 City marketing, festivals and urban spectacle 463 16.6 The West Midlands Enterprise Board 465 16.7 Urban regeneration in Cleveland OH 466 16.8 Linkage policy in San Francisco CA 468 16.9 The Watts Labor and Community Action Committee 471 16.10 The anatomy of a LETS: the West Glasgow Local Exchange Trading System 471 16.11 The black economy in urban Britain 473 17.1 The importance of collective consumption 477 17.2 Welfare-to-work 480 17.3 Public-choice theory 483 17.4 Human needs and wants 484 17.5 Environmental racism 490 17.6 Differential social access to utility services 491 18.1 You are where you live 505 18.2 Territoriality 508 18.3 Changes in the scale of society 512 18.4 Gated communities 519 18.5 Queer spaces 520 18.6 Ethnicity and race 521 18.7 Measuring segregation 522 18.8 The assimilation of migrant groups 523 18.9 Two models of ethnic segregation 525 18.10 Voluntary apartheid in Detroit 526 18.11 The tipping point 532 19.1 The 1995 Kobe earthquake and its aftermath 547 19.2 Flood-plain development in England 552 19.3 Some principles for designing legible cities 557 19.4 Newman’s design guidelines for defensible space 559 19.5 The concept of user-generated design 562 19.6 Danger areas on a Glasgow housing estate 564 20.1 The California tax revolt 577 20.2 Segregated education 582 20.3 The defensive incorporation of Lathrup Village MI 584 20.4 Legal challenges to exclusionary zoning 585 20.5 The central city-suburban fiscal debate in the Glasgow region 589 20.6 Strategies to resolve political fragmentation 590 20.7 Metropolitanism versus municipalism 594 20.8 The Daley machine in Chicago, 1955–76 598 20.9 The community charge: a short-lived tax 606 20.10 The New Jersey Tenants’ Organization 608 21.1 Defining the Third World 619 21.2 Rostow’s stages of economic growth 621 21.3 Determinants of colonial urban form 625 21.4 Delhi: the evolution of an imperial city 630 21.5 Urban primacy in the Third World 633 21.6 Jabotabek: an extended metropolitan region 636 21.7 Exo-urbanisation in the Pearl River delta region 638 22.1 Structure of the Latin American city: the case of Mexico City 647 22.2 Towards the post-apartheid city 653 22.3 Cairo, Egypt 655 22.4 Mumbai, India 658 22.5 The kampungs of Jakarta, Indonesia 661 22.6 Private cities in Greater Jakarta 662 23.1 Not so bright lights: a migrant’s view of Bombay 675 23.2 China’s urban floating population 682 23.3 Migration controls in Jakarta, Indonesia 682 23.4 Promoting secondary city growth in South Korea 685 24.1 Linkages between the two circuits of the Third World urban economy 692 24.2 Daily activity pattern of a part-time domestic worker in Dhaka, Bangladesh 697 24.3 Waste-picking as a survival strategy in Bangalore, India 697 24.4 Gender, work and human development in Manila, Philippines 701 24.5 Sweatshop labour in a Bangladeshi clothing factory 702 24.6 Child labour in Pakistan 703 25.1 Pavement dwellers in Bombay, India 709 25.2 Factors underlying choice of residential location among lowincome groups in Abidjan, Côte d’Ivoire 716 25.3 The development of San Martin, Buenos Aires, Argentina 722 25.4 The development of an illegal subdivision in Karachi, Pakistan 731 25.5 Forced relocation in Phnom Penh, Cambodia 737 25.6 Principles of the land-sharing strategy in Bangkok, Thailand 739 25.7 Site and services schemes in Pakistan: lessons in failure and success 745 26.1 The inadequacy of water supply in Third World cities 752 26.2 Environmental problems in a low-income community in Karachi, Pakistan 754 26.3 Environmental hazards for children at work in India 756 26.4 The 1999 mudslides in Caracas, Venezuela 757 26.5 Air pollution in the Valley of Death 761 26.6 River pollution in Dar es Salaam, Tanzania 762 26.7 The anatomy of a disaster: the 1998 flood in Dhaka, Bangladesh 764 26.8 The ecological footprint of Jakarta, Indonesia 766 27.1 HIV/AIDS: the modern plague 770 27.2 Aetiology of Chaga’s disease 773 27.3 Health problems in low-income settlements in Third World cities 773 27.4 Health-care provision in the slums of Nairobi, Kenya 775 27.5 Urban agriculture in Chinese cities 779 28.1 Traffic and transport in the Bangkok metropolitan region 788 28.2 The bicycle in Third World cities 794 28.3 Cycle rickshaws and traffic congestion in Dhaka, Bangladesh 796 29.1 Clients and radicals in a Lima shanty town, Peru 808 29.2 The definition of an urban social movement 809 29.3 Direct action by the hungry poor 811 29.4 Establishing autonomous housing co-operatives in Karachi, Pakistan 814 29.5 Participative budgeting in Belo Horizonte, Brazil 816 29.6 The Orangi Pilot Project in Karachi, Pakistan 817 29.7 Structure and goals of the Alliance in Mumbai, India 819 30.1 Main urban dimensions of Agenda 21 of the Rio Earth Summit 826 30.2 Cities and environmental issues 828 30.3 London’s ecological footprint 830 30.4 The urban waste economy in Bangalore, India 833 30.5 Project activities with waste scavengers in Bandung, Indonesia 834 30.6 Car-related urban problems 837 30.7 Kansai, Japan: a creative network city 845 30.8 The informational city region of the twenty-first century 847 30.9 Digitale Stad, Amsterdam, Netherlands 848 Preface to the Second Edition The first edition of Urban Geography: A Global Perspective, published in 2001, elicited a gratifyingly favourable response from an international readership of academic professionals and students, and the book has been adopted as a course text in colleges and universities in Britain, North America and elsewhere. This revised second edition benefited from constructive feedback from this international readership. The principal goal of this second edition of the book remains that of providing instructors and students of the contemporary city with a comprehensive introduction to the expanding field of urban studies. The structure of the first edition is maintained, with minor amendments. Each of the thirty chapters has been revised to incorporate developments in the field since publication of the first edition. All the popular study aids are retained, the glossary has been expanded, and chapter references and notes have been updated to reflect the latest research. Users of the first edition will recognise and benefit from the structural continuity, as well as the additional features, of this second edition. For those who are new to the book—if you are reading this it is likely that you share my interest in the city. Cities are fascinating, dynamic and complex socio-spatial phenomena. Knowledge of cities and of city living is of real importance both for academic understanding and in negotiating our daily lives in the urban world of the twenty-first century. Urban Geography: A Global Perspective is designed to provide an accessible introduction to the study of urban geography in the contemporary world. Michael Pacione Acknowledgements Few books make the passage from conception to publication without the assistance of a variety of outside agencies, and this book is no exception. Preparation of this second edition of Urban Geography: A Global Perspective was facilitated by the professional support provided by Andrew Mould, Anna Somerville, Zoe Kruze and Jo Jacomb at Routledge and Liz O’Donnell for proofreading. Academic colleagues, students and an international panel of reviewers offered valuable comments on the first edition of the book that aided preparation of this revised second edition. In writing the book I enjoyed and benefited from reading the work of other researchers engaged in the study of contemporary urban environments. I am grateful to all the authors and publishers who gave permission to reproduce material in the book. While considerable effort has been made to trace and contact copyright holders prior to publication, the author and publishers apologise for any oversights or omissions and if notified will endeavour to remedy them at the earliest opportunity. A number of friends and colleagues from the international geographic community kindly provided some of the photographs to illustrate different themes in the book. I am grateful to Stuart Aitken (for Plate 4.1), John Cole (Plate 25.2), Bruce Connolly (Plate 23.2a and 23.2b), Roman Cybriwsky (Plates 5, 7.1, 13.1 and 19.1), Larry Ford (Plate 26.3), Brian Godfrey (Plate 9), Rob Imrie (Plate 8.1), Hamish Main (Plate 5.2), Arthur Morris (Plates 2.1a and 24.1), Tony O’Connor (Plate 22.1), Jong-Kie Park (Plate 4), Jonathan Rigg (Plate 7), Philip Rushton (Plate 6), Denis Shaw (Plate 8.2), David Simon (Plate 23.1), Neil Smith (Plate 16.1), and Fulong Wu (Plate 16). All other photographs are taken from my own collection. The maps and diagrams in the book were drawn by Sharon Galleitch, Michael John Pacione and Margaret Dunn. The greatest debt is owed to my family—my wife, Christine, who continues to manage the demands of career and home in a way that is beyond the ability of most males, my son, Michael, whose uncanny ability to solve computing problems is a continuing source of comfort, and my daughter Emma who contributed simply by being the person she is. Collectively all three provided a comfortable and supportive environment in which to work and, when required, essential distractions from the task of writing. Michael Pacione Milton of Campsie 1 August 2004 Copyright Acknowledgements The author and publishers would like to thank the following for permission to reproduce illustrations in this work. FIGURES Alexandrine Press for figure 6 from Built Environment by P.Gaubatz, in ‘Understanding Chinese urban form’, Vol. 24(4), p. 265, 1998 (Figure 22.11). American Geographical Society for the figure of The Latin American city’ from the Geographical Review, by L.Ford, in ‘A new and improved model of Latin American city structure’, Vol. 86(3), pp. 437–40, 1996 (Figure 22.1). American Geographical Society for figure 3 from the Geographical Review, by L.Ford, in ‘A model of Indonesian city structure’, Vol. 83(2), pp. 374–96, 1993 (Figure 22.10). Association of American Geographers for figure 11.2 from the Annals of the Association of American Geographers, by S.Rowe and J.Wolch, in ‘Social networks in home and space’, Vol. 80, p. 191,1990 (Figure 17.3). Association of American Geographers for a figure from Resource Paper for College Geography 75–2, by P.Muller, in The outer city’, p. 19, 1976 (Figure 18.12). Association of Pacific Coast Geographers for figure 8 from the Yearbook of the Association of Pacific Coast Geographers, in ‘Order and disorder: a model of Latin American urban land use’, Vol. 57, p. 28, 1995 (Figure 22.2). Blackwell Publishers Ltd for figure 3.11 from Geografiska Annaler Series B, by E.Bylind, in Theoretical considerations regarding the distribution of settlement in inner North Sweden’, Vol. 42, pp. 225–31, © Swedish Society for Anthropology and Geography, 1960 (Figure 6.4). Blackwell Publishers Ltd and Joint Editors for figure 21.1 from the International Journal of Urban and Regional Research, by D.Harvey, in The urban process under capitalism’, Vol. 2, pp. 101–31, 1978 (Figure 7.9). Blackwell Publishers Ltd for a figure of a ‘Stress model of the recreational decision process’ from Geografiska Annaler Series B, by L.Brown and E.Moore, in The intraurban migration process’, Vol. 52, pp. 1–13, © Swedish Society for Anthropology and Geography, 1970 (Figure 10.3). Blackwell Publishers for figures 2.2, 3.1 and 4.1 by J.Friedmann, in Empowerment, pp. 27, 50 and 67, 1992 (Figures 29.1, 24.3 and 29.2). Centre for Urban Policy Research, New Brunswick, NJ for figure 11.1 by G.Galster and E.Hill in The Metropolis in Black and White: Place, Power and Polarization, CUPR, 1992 (Figure 15.2). Elsevier Science for figure 1.1 from Progress in Planning by D.Bennison and R.Davies, in The impact of town centre shopping schemes in Britain’, Vol. 14(1), pp. 1– 104, 1980 (Figure 12.7). Elsevier Science for figure 2 from Habitat International, by D.Shefer and L.Steinvortz, in ‘Rural-to-urban and urban-to-urban migration patterns in Colombia’, Vol. 17(1), pp. 133–150, 1993 (Figure 23.2). Guilford Press Inc. for figure 12.5, by R.Golledge and R.Stimson in Spatial Behaviours, p. 456, 1997 (Figure 10.2). International Institute for Environment and Development for figure 1 from Environment and Urbanization, by C.Hunt, in ‘Child waste pickers in India’, Vol. 8(2), p. 115, 1996 (Figure 26.1). International Thomson Publishing for the figure ‘British shopping centre development’, by C.Guy, in The Retail Development Process, Routledge, 1994 (Figure 12.8). James & James Publishers for figure 6.1, from J.Hardoy et al., in Environmental Problems in Third World Cities, Earthscan, p. 181, 1992 (Figure 30.1). Liverpool University Press for figure 5, from Third World Planning Review, by S.Angel and S. Boonyabancha, in ‘Land sharing as an alternative to eviction’, Vol. 10(2), pp. 107–27, 1988 (Figure 25.4). MIT Press for figure 3, by K.Lynch, in The Image of the City, p. 19, 1960 (Figure 19.2). Orion Publishing Group for figures 1, 9, 10, 14 and 15, by M.Thomson, in Great Cities and Their Traffic, Penguin, 1977 (Figures 13.2 and 13.4). Oxford University Press for figure 10.1, by J.Gugler, in The Urbanization of the Third World, p. 167, 1988 (Figure 24.2). Pearson Education Ltd for figure 3.3, by E.Morris, in History of Urban Form, Longman, p. 52, 1994 (Figure 3.3). Pearson Education Ltd, for figure 2.5, by E.Morris, in British Town Planning and Urban Design, Longman, p. 29, 1997 (Figure 3.7). Pearson Education Ltd for figure 2.2 by J.Dawson, in Shopping Centre Development, Longman, p. 25, 1983 (Figure 12.4). Pearson Education Ltd for figure 7.1b, by R.Potter and S.Lloyd-Evans, in The City in the Developing World, Longman, p. 142, 1998 (Figure 25.2). Pion Ltd London, and A.R.Veness for figure 4 from Environment and Planning D: Society and Space, by A.R.Veness, in ‘Home and Homelessness in the United States’, Vol. 10, pp. 445–68, 1992 (Figure 11.2). Russell Sage Foundation for figure 7.1, by M.White, in American Neighborhoods and Residential Differentiation, Russell Sage Foundation, p. 237, 1987 (Figure 7.8). Taylor & Francis for the figure of ‘Mann’s model of a typical medium-sized British city’, by P.Mann, in An Approach to Urban Sociology, Routledge, 1965 (Figure 7.5). Taylor & Francis for the figure ‘The concept of the rent gap’, by N.Smith, in The New Urban Frontier, Routledge, 1996 (Figure 10.5). Taylor & Francis for the figure ‘A general model of economic activity within a developed economy’, by P. Ekins and M.Max-Neef, in Real Life Economics, Routledge, 1992 (Figure 16.4). Taylor & Francis for the figure ‘Franks model of dependent development’, by J.Dickensen et al., in A Geography of the Third World, Routledge, 1996 (Figure 21.1) Taylor & Francis for figure 3.2, by L.Brown, in Place, Migration and Development in the Third World, 1991 (Figure 23.1). Taylor & Francis for figure 6.6, by D.Hilling, in Transport and Developing Countries, Routledge, 1996 (Figure 28.2). John Wiley & Sons Ltd for figure 2.1, by F.Boal, in D.Herbert and R.Johnston (eds), Social Areas in Cities, Vol. 1, p. 397, 1980 (Figure 18.8). TABLES Elsevier Science for table 3 from Progress in Planning, by L.Bourne, in ‘Reinventing the suburbs’, Vol. 46(3), pp. 163–84, 1996 (Table 4.13). Elsevier Science for table 3.1 from Progress in Planning, by D.Benison and R.Davies, in ‘The impact of town centre shopping schemes in Britain’, Vol. 14(1), pp. 1–104, 1980 (Table 12.3). Elsevier Science for table 3 from Progress in Planning, by R.Heywood, in ‘The emerging social metropolis’, Vol. 47(3), pp. 159–250, 1997 (Table 20.1). ITBP for the table ‘Public housing completions 1939–91, USA’, by E.Howenstine, in G.Hallett (ed.), The New Housing Shortage, Routledge, 1993 (Table 11.3). ITBP for the table ‘The hierarchy of planned shopping centres’, by K.Jones and J.Simmons, The Retail Environment, Routledge (Table 12.1). James & James Publishers for tables 3.1, 3.2 and 4.1, by J.Hardoy and D.Satterthwaite, in Squatter Citizen, Earthscan, pp. 20, 68 and 103, 1989 (Tables 25.2, 25.3 and 25.6). James & James Publishers for tables 3.2 and 5.1, by J.Hardoy et al., in Environmental Problems in Third World Cities, Earthscan, pp. 77 and 152, 1992 (Tables 26.2 and 27.1). Oxford University Press Inc. for tables 1.3 and 2.1, by R.J.Waste, in Independent Cities: Re-thinking US Urban Policy, © 1998 by Oxford University Press Inc. (Tables 15.3, 15.6). Taylor & Francis for the table ‘The Main Goals of Planning’ by A.Thornley, in Urban Planning Under Thatcherism, Routledge, 1991 (Table 8.1). Taylor & Francis for the table ‘Pressures on the welfare state’, by S.Pinch, in Worlds of Welfare, Routledge, 1997 (Table 17.1). Taylor & Francis for table 2.1 by D.Drakakis-Smith, in The Third World City, Methuen, 1987 (Table 21.1). Taylor & Francis for the tables ‘Principal reasons for migration from rural North-East Thailand’ and ‘Principal policy responses to rural-urban migration in the third world’, by M.Parnwell, in Population Movements and the Third World, 1993 (Tables 23.1 and 23.4). BOXES Pearson Education Ltd for figures 1.7 and 3.45, by A.Morris, in History of Urban Form, Longman, pp. 7 and 84, 1994 (Boxes 3.3 and 3.5). Taylor & Francis for material used in box 14.4, by A.King, in Global Cities, Routledge, 1990 (Box 14.5). Taylor & Francis for material used in box 14.7, by S.Dietrick and R.Beauregard, ‘From front-runner to also-ran’, in P.Cooke (ed), The Rise of the Rustbelt, UCL Press, pp. 52–71, 1995 (Box 14.8). Taylor & Francis for material used in box 16.7, by C.Wood, ‘Renewal and regeneration’, in D.Chapman (ed.), Creating Neighbourhoods and Places, E. and F.N.Spon, pp. 194–221, 1996 (Box 16.7). Taylor & Francis for the material in boxes 21.4 and 24.4, by D.Drakakis-Smith, in Pacific Asia, Routledge, 1993 (Boxes 21.4 and 24.4). Every effort has been made to contact copyright holders for their permissions to reprint material in this book. The publishers would be grateful to hear from any copyright holder who is not here acknowledged and will undertake to rectify any errors or omissions in future editions of this book. Introduction The contemporary world is an urban world. This is apparent in the expansion of urban areas and the extension of urban influences across much of the habitable surface of the planet. Today, for the first time in the history of humankind, urban dwellers outnumber rural residents. Urban places—towns and cities—are of fundamental importance: for the distribution of population within countries; in the organisation of economic production, distribution and exchange; in the structuring of social reproduction and cultural life; and in the allocation and exercise of power. Furthermore, in the course of the present century the number of urban dwellers and level of global urbanisation are likely to increase. Even those living beyond the administrative or functional boundaries of a town or city will have their lifestyle influenced to some degree by a nearby, or even distant, city. We inhabit an urban world in which the spread of urban areas and urban influences is a global phenomenon. The outcomes of these processes are manifested in the diversity of urban environments that characterise the contemporary world. The study of towns and cities is a central element of all social sciences, including geography, which offers a particular perspective on and insight into the urban condition. The scope and content of urban geography are wide-ranging, and include the study of urban places as ‘points in space’ as well as investigation of the internal structure of urban areas. Within the general field of urban geography specialised subareas attract researchers interested in particular aspects of the urban environment (such as population dynamics, the urban economy, politics and governance, urban communities, housing or transport issues). This eclectic coverage, allied to the synthesising power of a geographical perspective, is a major advantage for those seeking to understand the complexity of contemporary urban environments. Students of urban geography draw on a rich blend of theoretical and empirical information to advance their knowledge of the city. The breadth of urban geography and the volume of published research may appear daunting to someone approaching the field for the first time. This book synthesises this wealth of material to provide a comprehensive introduction to the study of urban geography in the contemporary world. It is intended primarily for undergraduate students of geography and for those taking urban-based courses in cognate social sciences. STRUCTURE OF THE BOOK In this introduction I explain the aims and objectives of the book, identify the intended readership and indicate how the material is arranged. I also provide instructors and students with detailed guidance on how to make the best use of the book. Following the Introduction, the book is organised into six main parts. Part One lays the foundations for the study of urban geography. In the opening chapter we explore the importance of, and relationship between, global and local factors in the processes of urbanisation and urban change. I highlight the major outcomes of these processes and identify the main themes and issues of importance in urban geography. Chapter 2 introduces a number of key theoretical and conceptual issues, and provides a brief history of the subject in order to establish a framework for analysis in urban geography. Part Two focuses on the world, regional and national scales. In this part of the book we examine the origins and growth of cities from the earliest times to the present day, establish the global context for the processes of urbanisation and urban growth, identify recent developments in the urban geography of the major world regions and examine national systems of cities and different types of urbanised region in the world. Part Three considers urban structure and land use in Western cities. We examine the key agents and processes underlying patterns of urban change and develop an understanding of the construction and reconstruction of urban areas with particular reference to major urban land uses and issues such as post-war suburbanisation, new community development, residential mobility and neighbourhood change, housing problems and policies, retailing and transportation. Part Four focuses on economy, society and politics in the Western city. Discussion of the urban economy is set within the context of the changing nature of employment in the global economy and the post-Second World War restructuring of metropolitan spaceeconomies. We consider the nature and incidence of poverty and deprivation, and assess national and local responses to urban economic change. The question of social justice in urban service provision is also addressed. We explore the key social processes of congregation and segregation in the city and examine the concept of community and the different bases of residential differentiation. We then employ the notion of urban liveability to consider differential quality of life within cities along a number of major dimensions. Finally, we discuss the role of local government and the distribution and use of power in the city. In Part Five our attention turns to the urban geography of the Third World, a region that exhibits some of the highest rates of urbanisation and urban growth in the world, as well as the greatest incidences of urban social, economic and environmental problems. I establish the global context for Third World urbanisation before analysing the internal structure of Third World cities. The process of rural-urban migration is identified as a major factor in urbanisation and urban growth. Our attention then focuses on experiences of life in Third World cities, with detailed examination of the urban economy, housing issues, environment, health, transport, and poverty, power and politics in the Third World city. Finally, Part Six employs a prospective viewpoint to consider the future of cities with particular reference to the concept of sustainable urban development, and critically examines the nature of cities in the twenty-first century. URBAN GEOGRAPHY IN COLLEGES AND UNIVERSITIES The importance of urban systems and environments in contemporary society is reflected in the availability of urban-based courses at all levels of the educational system. In the further and higher education sectors, most college and university departments of geography, as well as academic departments in cognate social sciences, offer at least an introductory class in urban geography. In many departments the importance and scope of urban geography are reflected in the provision of several urban classes organised either in terms of a systematic division of the field (with, for example, classes on urban social geography, urban economic geography or urban historical geography) and/or according to a regional specialisation (with a major division between the urban geography of the Western world and the study of urban phenomena in the Third World). The book is designed to be used as a basic resource for those engaged in the teaching and study of urban geography. The following section offers some guidance for students and instructors on how to make best use of the book. HOW TO USE THIS BOOK INSTRUCTORS The book provides an introduction to the study of urban geography. Its principal characteristics are: 1. a comprehensive coverage of the major themes and issues of importance in urban geography; 2. a global perspective which examines urban environments in both the developed world and the Third World, and in all the major world regions; 3. an approach to teaching that relates explanation of theories and concepts to revealed processes and patterns, employing examples and case studies from appropriate environments worldwide; 4. support of the text by a number of userfriendly learning strategies, including numerous boxed studies to amplify key concepts and illustrate particular issues; maps, diagrams and tables of statistics; photographs; an introductory preface for each chapter; a guide to further reading; a list of key concepts, set of study questions, sample seminar discussion topics (listed in Table 1.1) and a student project for each chapter; a comprehensive list of references; a glossary of key terms; and a work-file approach to encourage active student participation in the learning process. Glossary terms have been highlighted initially in bold type. For instructors, the material presented in the book can be used in a flexible way to construct and organise several different types of class in urban geography. Three ways in which the book may be used as a basic resource for class instruction are: 1. in its entirety as a means of providing introductory-level students with a general overview of our urban and urbanising world; 2. as a foundation for the study of the urban geography of particular world regions; or 3. as the basis for more specialised classes dealing with specific urban themes. These options can be combined to provide an integrated course across several years of undergraduate urban geography teaching, with a general first-year introduction to urban geography followed by semester-length classes that develop more detailed in-depth analyses of particular issues and themes in the second and subsequent years of study. We can illustrate the potential application of this flexible structure with reference to a selection of the range of urban geography classes on offer in colleges and universities in Britain, Europe, North America and Australasia. The shaded areas in the following table indicate chapters in Urban Geography: A Global Perspective that could be used as a core learning resource for different classes in urban geography. In addition, depending on study objectives and time available, each of the core groups of chapters identified can be supplemented by cross-reference to linked material elsewhere in the book. STUDENTS The book provides a comprehensive introduction to the study of urban geography. It explains the major theories and concepts underlying processes of urban change in the contemporary world and relates them to revealed patterns and outcomes in towns and cities across the globe. It also incorporates a number of features designed to help your understanding and revision of material learned. Thus, the basic resource of the text is supported by a wealth of additional information provided in a variety of forms. These include: 1. chapter previews and introductions that outline the main objectives and issues to be covered; 2. boxed studies that provide supplementary information on key issues and detailed case studies; 3. maps, diagrams and tables that complement the text with supporting graphical and numerical information; 4. photographs that provide a visual illustration of urban environments throughout the world; 5. further reading that identifies key readings to enable you to follow up the themes covered in each chapter; 6. key concepts that act as a shorthand revision guide by checking your understanding of important issues in each chapter; 7. study questions designed to test your knowledge and promote critical consideration of the theories, concepts, themes and issues discussed in each chapter; 8. a project that enables you to undertake independent research and develop a detailed understanding of a particular topic; 9. a glossary that defines the key terms in urban geography; 10. a comprehensive list of major references that acts as a guide for further independent study; and 11. the work-file approach, which involves you as an active partner in the production of knowledge on the urban environment. STUDY TIPS In addition to the particular guidance offered above on how best to use the book my ‘top ten’ tips for the successful study of urban geography are as follows: 1. To obtain maximum advantage from the book you should make full use of the in-built study features. 2. Ensure that you understand the key concepts and terms. Re-read the text to refresh your knowledge of any you are unsure about. The glossary can also act as a ready reference source for a definition. In addition, there will be several dictionaries of geography available in your college library. 3. Use the end-of-chapter study questions to test your knowledge and understanding. It is also always helpful to try to relate general urban processes and patterns with conditions in a town or city with which you are familiar. (For example, can you identify local incidences of social polarisation, gentrification, ghettoisation, deindustrialisation or suburbanisation?) 4. Take time out to read the boxed studies that accompany each chapter in the book. These not only provide detailed information on particular concepts, issues and situations but can help you relate to the broader themes of the chapter. 5. Study the maps, diagrams and tables. Think about what they are depicting, and identify what trends, patterns and processes they display. 6. Use the photographs as a means of entering into the urban scene. Look at the detail of the environment and think about what is being shown (and not shown). How well does it relate to the theme it represents? 7. The glossary can be used as a quick-fire question-and-answer test for yourself by seeing how well your own definitions match those given. 8. Cultivate the habit of making a regular trip to the college or university library to check recent issues of the main academic journals dealing with urban matters. These include Urban Studies, Urban Geography, Annals of the Association of American Geographers, Transactions of the Institute of British Geographers, Environment and Planning A, Environment and Planning D: Society and Space, Cities, Urban Affairs Quarterly, Journal of the American Planning Association, Town Planning Review, International Development Planning Review (formerly Third World Planning Review), Habitat International, Environment and Urbanization. 9. Engage directly in the production of knowledge by creating your own personal workfile of urban materials. These can be obtained from a variety of sources, including: (a) the textbook—make notes, prepare bullet points, annotate diagrams and write short essays on key issues; (b) additional reading—write summaries of important readings; (c) lecture notes and handouts provided by the instructor; (d) class essays, projects or other assignments; (e) your own maps, sketches or photographs of townscapes with which you are familiar and which are representative of particular themes covered in the class; (f) government reports and statistics—for example, published census data and urban planning documents; (g) the media—newspaper and magazine cuttings, television reports and documentaries; (h) the World Wide Web—this provides a vast array of potentially useful information on urban issues, but must be used with critical appreciation of the sources, which can range from international agencies and government departments, through scholarly research articles published in on-line journals, to non-refereed statements by individuals. You can use a search engine (such as Yahoo, Lycos, Alta Vista, Google or Copernic), to locate sites of interest or, if known, type in a specific address. For example: (i) statistics on human settlements http://www.unhabitat.org/ http://www.unstats.un.org/ (ii) direct access to organisations such as the World Bank http://www.worldbank.org/, United Nations http://www.undp.org/ http://www.un.org/ http://www.cyberschoolbus.un.org/, United Nations Centre for Human Settlements (Habitat) http://www.unchs.org/, World Health Organization http://www.who.int/en/, OECD http://www.oecd.org/home, Group of 77 Third World countries http://www.g77.org/, US Census Bureau http://www.census.gov/ (iii) newspapers, magazines and television such as National Geographic http://www.nationalgeographic.com/, One World magazine http://www.oneworld.net/, The Times http://www.the-times.co.uk/, CNN http://www.cnn.com/, BBC http://www.bbc.co.uk/ (iv) for sites on particular cities and themes such as urbanisation statistics http://www.prb.org/, urbanisation in Asia http://www.asiandemographics.com/, Orangi pilot project, Karachi http://www.urikarachi.org/orangi.htm, traffic conditions in California http://www.dat.ca.gov/, London docklands http://www.lddc-history.org.uk/ http://www.london-docklands.co.uk/, urban planning in the USA http://www.planetizen.com/, smart growth http://www.smartgrowth.org/, new urbanism http://www.cnu.org/, gated communities http://www.gated-cities.com/, mega-malls http://www.mallofamerica.com/ http://www.westedmontonmall.com/, light rail transit http://www.lrta.org/, Tsukuba science city http://www.tsukuba.org/, Tokyo teleport http://www.tokyo-teleport.co.jp/, US National Low Income Housing Coalition http://www.nlihc.org/, sustainable cities http://www.unchs.org/programmes/sustainablecities http://www.sustainable-cities.org/, globalisation and world cities http://www.lboro.ac.uk/gawc, photographs of the world’s cities from space http://city.jsc.nasa.gov/cities/. Ideally, your work-file should be in the form of a loose-leaf binder that allows later material to be inserted in appropriate places. This will also help you organise your thoughts, identify gaps in your knowledge and questions for further investigation, highlight the linkages among different themes, and aid revision. 10. Two final points before we commence our study of urban geography. First of all, bear in mind the fact that knowledge of urban geography will be of both academic and practical importance in the urban world of the twenty-first century. Second, and most important, enjoy the learning experience! PART ONE The Study of Urban Geography 1 Urban Geography from global to local Preview: the study of urban geography; global triggers of urban change—economy, technology, demography, politics, society, culture, environment; globalisation; localisation of the global; the meaning of geographic scale; local and historical contingency; processes of urban change; urban outcomes; daily life in the global village; why study urban geography? INTRODUCTION Urban geography seeks to explain the distribution of towns and cities and the sociospatial similarities and contrasts that exist between and within them. If all cities were unique, this would be an impossible task. However, while every town and city has an individual character, urban places also exhibit common features that vary only in degree of incidence or importance within the particular urban fabric. All cities contain areas of residential space, transportation lines, economic activities, service infrastructure, commercial areas and public buildings. In different world regions the historical process of urban evolution may have followed a similar trajectory. Increasingly, similar processes, such as those of suburbanisation, gentrification and socio-spatial segregation, are operating within cities in the developed world, in former communist states and in countries of the Third World to effect a degree of convergence in the nature of urban landscapes. Cities also exhibit common problems to varying degrees, including inadequate housing, economic decline, poverty, ill health, social polarisation, traffic congestion and environmental pollution. In brief, many characteristics and concerns are shared by urban places. These shared characteristics and concerns represent the foundations for the study of urban geography. Most fundamentally, the character of urban environments throughout the world is the outcome of interactions among a host of environmental, economic, technological, social, demographic, cultural and political forces operating at a variety of geographic scales ranging from the global to the local. In this book we approach the study of urban geography from a global perspective. This acknowledges the importance of macro-scale structural factors in urban development, but also recognises the reciprocal relationship between global forces and locally contingent factors in creating and recreating the geography of towns and cities. A global perspective also demonstrates the interdependence of urban places in the Urban geography: from global to local 3 contemporary world, and facilitates comparative urban analysis by revealing common and contrasting features of cities in different cultural regions. The framework we shall employ in our study of urban geography is depicted in Figure 1.1. In reading from left to right, we move from the global to the local scale. This schematic diagram highlights the major ‘trigger factors’ and processes underlying contemporary urban change, as well as the principal urban outcomes of these processes. In this chapter we examine the nature of these factors and processes, and the relationship between global and local forces in the construction of contemporary urban environments. GLOBAL TRIGGER FACTORS THE ECONOMY Economic forces are regarded as the dominant influence on urban change. Since its emergence in Figure 1.1 Triggers, processes and outcomes in urban geography the sixteenth century, the capitalist economy has entered three main phases. The first phase, from the late sixteenth century until the late nineteenth century, was an era of competitive capitalism characterised by free-market competition between locally oriented businesses, and laissez-faire economic (and urban) development largely unconstrained by government regulation. In the course of the nineteenth century the scale of business increased, consumer markets expanded to become national and international, labour markets became more organised as wage-rate norms spread and government intervention in the economy grew in response to the need for regulation of public affairs. By the turn Urban geography 4 of the century these accumulated trends had culminated in the advent of organised capitalism. The dynamism of the economic system—the basis of profitability—was enhanced in the early decades of the twentieth century by the introduction of Fordism. This economic philosophy was founded on the principles of mass production using assembly-line techniques and ‘scientific’ management (known as Taylorism), together with mass consumption fuelled by higher wages and high-pressure marketing techniques. Fordism also involved a generally mutually beneficial working relationship between capital (business) and labour (trade unions), mediated by government when necessary to maintain the health of the national economy. The third and current phase of capitalism developed in the period following the Second World War. It was marked by a shift away from industrial production towards services (particularly financial services) as the basis for profitability. Paradoxically, the explanation for this shift lay in part with the very success of Fordism. As mass markets became saturated and profits from mass production declined, many enterprises turned to serve specialised ‘niche markets’. Instead of standardised production, specialisation required flexible production systems. This current phase of capitalism is referred to as advanced capitalism or disorganised capitalism (to distinguish it from the organised nature of the Fordist era). The transition to advanced capitalism was accompanied by an increasing globalisation of the economy in which transnational corporations (TNCs) operated often beyond the control of national governments or labour unions (see Chapter 14). The evolution of the capitalist economy is of fundamental significance for urban geography, since each new phase of capitalism involved changes in what was produced, how it was produced and where it was produced. This meant that ‘new industrial spaces’ (based, for example, on semiconductor production rather than shipbuilding) and new forms of city (such as a technopole instead of a heavy industrial centre) were required.1 TECHNOLOGY Technological changes, which are integral to economic change, also influence the pattern of urban growth and change. Innovations such as the advent of global telecommunications have had a marked impact on the structure and functioning of the global economy. This is illustrated by the new international division of labour in which production is separated geographically from research and development (R&D) and higher-level management operations, while almost instantaneous contact is maintained between all the units in the manufacturing process, no matter where in the world each is located. The effects of macro-level technological change are encapsulated in the concept of economic long waves or cycles of expansion and contraction in the rate of economic development (see Chapter 14). The first of these cycles of innovation (referred to as Kondratieff cycles) was based on early mechanisation by means of water power and steam engines, while the most recent (and still incomplete) cycle is based on microelectronics, digital telecommunications, robotics and biotechnology. The different technology eras represented by Kondratieff cycles shape not only the economy but also the pace and character of urbanisation and urban change. Modification of the urban environment occurs most vigorously during up-cycles of economic growth. Urban geography: from global to local 5 Technological changes that directly affect urban form also occur at the local level. Prominent examples include the manner in which advances in transportation technology promoted suburbanisation (see Chapter 7) or how the invention of the high-speed elevator facilitated the development of skyscrapers in cities such as New York. DEMOGRAPHY Demographic changes are among the most direct influences on urbanisation and urban change. Movements of people, into and out from cities, shape the size, configuration and social composition of cities. In Third World countries expectations of improved living standards draw millions of migrants into cities (see Chapter 23), while for many urban dwellers in the West the ‘good life’ is realised through suburbanisation or exurbanisation. The condition of the urban environment can also affect the demographic structure of cities by influencing the balance between rates of fertility and mortality. Intra-urban variations in health are pronounced in Third World cities where well above average mortality rates are recorded in overcrowded squatter areas (see Chapter 27), but sociospatial differences in health status are also characteristic of many Western cities. Demographic changes are related to other ‘trigger factors’ such as economic growth or decline, and political change. For example, population growth in a Third World country may induce political attempts to restrict migration, whether within the country to control ‘overurbanisation’ or between countries, as along the Mexico-USA border. POLITICS Cities reflect the political ideology of their society. The urban impact of political change is demonstrated by the case of Eastern Europe and the former Soviet Union (see Chapter 5). During the middle fifty years of the twentieth century the development of new towns and reorganisation of existing cities reflected the imperatives of a command economy and centralised political apparatus. The planned socialist city was intended to promote national economic development and to foster social and spatial equity in collective consumption. Accordingly, high priority was afforded to urban industrial development and the construction of large estates of public housing. With the collapse of communism and relaxation of strict urban land-use controls, capitalist tendencies such as suburbanisation and social differentiation in housing are becoming increasingly evident. An example of how local-regional changes can have a global impact is the way in which the end of the Cold War influenced the cities of the US sunbelt/gunbelt, whose economies had been dependent on defence-related industries.2 Within Western societies, changes in political ideology and subsequent modifications of economic and urban policy have had major impacts on city development. The rise of the New Right governments of Ronald Reagan in the USA and Margaret Thatcher in the UK led to reductions in public expenditure and increased dependence upon the private sector in urban development. This is evident in the rise of agencies such as urban development corporations and enterprise zones, public-private partnership schemes, property-led urban regeneration, and strategies such as the private finance initiative in the UK (see Chapter 16). Politics and economics exist in a reciprocal relationship, the outcomes of which can have a major impact on urban change. On the one hand, the Urban geography 6 formulation of urban policy may be influenced by political forces such as the opposition of middle-class suburban voters to increased taxation to pay for inner-city services. On the other hand, a political decision by central government not to provide a financial incentive package to attract inward investment by a foreign-owned TNC can affect the future economic prosperity of a city and its residents. SOCIETY Macro-scale social changes can have a significant impact on the character of towns and cities. For example, social attitudes towards abortion or use of artificial methods of birth control may influence the demographic composition of a society and its cities. Popular attitudes towards ethnic or lifestyle minorities can determine migration flows between countries and cities, as well as underlying patterns of residential segregation within cities. This is evident in recent trends towards increased suburbanisation of the AfricanAmerican population in the USA, which has been facilitated not only by the economic advancement of individuals and households but by changing social attitudes to the education, employment and residential status of minorities (see Chapter 18). In a similar manner, the attitude of society towards other minority groups, such as single-parent households, the unemployed, disabled people and elderly people, and towards women, conditions their status and location in the city (see Chapter 19). CULTURE One of the most significant of cultural changes in Western society in the post-Second World War era, particularly from the late 1970s onwards, has been the rise of materialism. This is displayed in conspicuous consumption by those who can afford it. At the urban scale this is manifested in the appearance of a ‘cappuccino society’ characterised by stores selling designer clothes, wine bars, pavement cafés, gentrification, yuppies (young upwardly mobile professionals), marbles (married and responsible but loaded executives), and bumper stickers proclaiming ‘Dear Santa, I want it all.’ It is also evident in an increasing gap between rich and poor in cities (see Chapter 15). While some observers claim to have identified signs of an emerging ‘postconsumerism’, materialism remains a dominant cultural influence in urban society. The effect of cultural change on cities is encapsulated in the concept of postmodernity or post-modernism (see Chapter 2). This embraces social difference and celebrates variation in urban environments, whether expressed in architectural or social terms. Youth cultures have flourished in some urban settings (for example, in inner-city ethnic areas energised by hip-hop and rap music), as have alternative lifestyle communities such as the gay districts of West Hollywood in Los Angeles or Pad-dington in Sydney. Postmodern urbanism is also evident in the growth of cultural industries (related to media and the arts) and in the regeneration (and place marketing) of historic urban districts such as Gastown in Vancouver, Covent Garden in London or the Merchant City in Glasgow. Urban geography: from global to local 7 THE ENVIRONMENT The impacts of environmental change on patterns of urbanisation and urban change are seen at a number of geographic scales. At the planetary scale, global warming due to the greenhouse effect (caused in part by the waste outputs of urban civilisation) may require the construction of coastal defences to protect cities such as Bangkok, Jakarta, Venice and London from the danger of inundation. Equally, any significant deflection of the North Atlantic Drift current (the Gulf Stream) which carries warm water from the Caribbean to the coasts of Britain and northern Europe would affect local weather patterns and require a costly response from cities in terms of urban infrastructure such as improved winter heating systems, public transport and road maintenance. At the local scale, ‘natural’ phenomena such as earthquakes and landslides can force the abandonment of settlements (for example, Third World squatter settlements constructed on marginal lands) or require major works of reconstruction, as in Mexico City after the 1985 earthquake. For reasons of clarity we have examined each of these important ‘trigger factors’ independently. In practice, of course, they are interrelated and operate simultaneously along with other local-scale factors to influence urban change. These trigger factors are integral elements in the processes of globalisation. GLOBALISATION Globalisation is a term used to describe a complex of related processes that has served to increase the interconnectedness of social life in the (post) modern world3 (Box 1.1). For Robertson (1992 p. 8), the concept refers ‘both to the compression of the world and the intensification of consciousness of the world as a whole’.4 Globalisation is evident in three forms: 1. economic globalisation: seen in arrangements for the production, exchange, distribution and consumption of goods and services (such as the rise of TNCs, the new international division of labour, increases in foreign direct investment, flexible forms of production and a global financial system).5 2. political globalisation: seen in arrangements for the concentration and application of power (such as the growth of multi-state political-economic groupings, and consideration of local issues within global context).6 3. cultural globalisation: seen in arrangements for the production, exchange and expression of symbols that represent facts, meanings, beliefs, preferences, tastes and values (such as the global distribution of images and information, and the emergent cosmopolitanism of urban life).7 Significantly, there is a reflexive relationship between the global and the local. While global forces lead to change in the city, cities modify and embed globalisation within a local context. Within the global-local nexus, global forces are generally held to be most powerful and their control more spatially extensive. Local forces are seen to be relatively Urban geography 8 weaker and geographically limited in effect. However, it is important not to allow the discourse of globalisation to obscure the fact that urban BOX 1.1 Principal characteristics of globalisation 1. Globalisation is not a new phenomenon. The processes of globalisation have been ongoing throughout human history, but the rate of progress and its effects have accelerated since the ‘early modern’ period of the late sixteenth century, and more especially over the past few decades in tandem with the transition towards postmodernity. 2. Globalisation involves both an intensification of worldwide social relations through time-space compression of the globe, and local transformations involving enhancement of local identity as well as of local consciousness of the world as a whole. 3. In the global-local nexus, global forces are generally held to be most powerful and their control more spatially extensive. Local forces are seen to be relatively weaker and geographically limited in effect, although certain local actions can have global consequences. 4. Globalisation takes place in cities, but the relationship between the city and global processes is dialectic. While global forces lead to changes in the city, cities modify and embed globalisation within local context. 5. Global forces are mediated by locally and historically contingent forces as they penetrate downwards, coming to ground in particular places. 6. A number of ‘trigger forces’ underlie globalisation but the dominant force is generally regarded as economic. 7. Globalisation reduces the influence of nation-states and political boundaries at the same time that states are organising the legal and financial infrastructures that enable capitalism to operate globally. 8. Globalisation operates unevenly, bypassing certain institutions, people and places. This is evident at the global scale in the disparities between booming and declining regions and at the urban scale by social polarisation within cities. 9. The differing interests of actors means that global forces are sometimes embraced, resisted or exploited at lower levels. 10. The mobility of capital diminishes the significance of particular places, although it may also strengthen local identity by engendering a defensive response by local actors. 11. The location of agents that command, control and finance the global economy defines a global city. These are the bases from which transnational corporations (TNCs) launch offensives around the world. The extent to which cities achieve global status is a major determinant of their prosperity. change is not effected by global forces alone. As we shall see in the course of the book, national, regional and local agents are also of significance. For example, national taxation policies, regional trade-union power and local planning regulations all have an impact on urban development and change. Local actions may also have global consequences, as Urban geography: from global to local 9 when urban-based protests against socially regressive economic activities (such as the exploitation of child labour) result in changes in patterns of trade and consumption at the world scale. Globalisation should not be regarded as leading automatically to the disintegration of local life. Individuals can either disembed themselves from a locality by operating within a global milieu or embed themselves by attachment to a particular place. Neither condition is exclusive. The reflexive nature of the global-local relationship is evident in the world of international finance, where the disembedded electronic space of the international financial system actually compels embedded social relations in specific locations (such as the City of London), to facilitate discussion of new financial products and engaging in inter-personal exchanges of information. More generally, in the placebound daily lives of most people, particularly those out-with the mainstream of advanced capitalism, globalisation may promote a search for local identity in a mobilised world.8 Globalisation is a highly uneven set of processes whose impact varies over space, through time, and between social groups. Global forces bypass many people and places. Many towns in the Third World, as well as in rural areas of Western society, produce mainly for local consumption using local techniques. Even within global cities, certain neighbourhoods where poverty and disadvantage prevail are Plate 1.1 An example of economic and cultural globalisation: the British high-street retailer Marks and Spencer’s store in Hong Kong Urban geography 10 peripheral to the working of the global economy. It should be recognised, of course, that the current situation of social and economic disadvantage in such areas may have been triggered by macro-scale forces such as the investment decision of an executive in a TNC based in a city on the other side of the globe. The unevenness of globalisation is apparent at all levels of society. At the world scale it is seen in the disparities between booming and declining regions, and at the urban scale in the social polarisation between affluent and marginalised citizens. The uneven penetration of globalisation is a question not simply of which institutions, industries, people and places are affected but also of how they are affected.9 Local people and places may be overwhelmed and exploited by the forces of globalisation or they may seek to resist, adapt or turn globally induced change into an opportunity. Resistance is a common response in which global forces are mediated at lower spatial scales. During the 1980s, mass protests were held in a number of Third World cities in an effort to resist the imposition of economic austerity measures by the International Monetary Fund (as a response to the burgeoning Third World debt crisis). More recently, the chamber of commerce in the predominantly Mexican Pilsen neighbourhood of Chicago opposed the siting of a Starbucks coffee shop in an effort to preserve local ethnic diversity from the homogenising cultural influence of a global corporation. Frequently, however, people and organisations must acquiesce and adapt to global forces. The proliferation of clothing sweatshops in Los Angeles and New York City during the 1980s was an attempt by local manufacturers to maintain their competitiveness in world markets. Global connections may also be pursued actively by local agents, particularly if they create new opportunities for local economic growth. In world cities such as London and New York foreign investment in property development, especially during the boom years of the 1980s, boosted commercial property markets by raising rental income, increasing property values and generally expanding business opportunities.10 Provincial cities less attractive to finance capital are usually less able to attain similar benefits. GLOCALISATION: THE LOCALISATION OF THE GLOBAL As we have seen, there is a dialectical or reflexive relationship between global and local processes in constructing contemporary urban environments. The term glocalisation has been used to describe the simultaneous operation of processes of: 1. de-localisation or de-territorialisation evident, for example, in the instantaneity of email communication across the globe; 2. re-localisation or re-territorialisation, whereby global influences interact with and are transformed within local context as, for example, in the creation of historic heritage districts in cities. The effects of ‘glocalisation’ are generally apparent. Persky and Weiwel (1994) have demonstrated how in many US cities a significant share of economic activity continues to serve local markets.11 In labour market terms globalisation is of relevance only for a small minority of workers with the skills necessary to compete in international labour markets. Equally, firms must strike a balance between being sufficiently mobile to take advantage of opportunities that may arise elsewhere, and being sufficiently embedded in Urban geography: from global to local 11 a local context in order to develop links with local suppliers and labour forces. In social terms, globalisation has positive and negative consequences for different peoples and places. On one hand, it may promote the growth of vibrant multicultural urban communities; on the other it can lead to socio-spatial concentrations of disadvantage. Politically, glocalisation focuses attention on cities as both strategic sites for global interests seeking to maximise profit and spaces where local grass-roots and civil society assert their right to liveable places. By combining the deterritorialisation tendency of globalisation and the re-territorialisation of localisation the concept of glocalisation underscores the dialectic nature of the relationship between global and local forces in the construction and reconstruction of contemporary urban environments. THE QUESTION OF SPACE AND SCALE The complexity of the global-local nexus forces us to think about the meaning of spatial scale itself. As we shall see in Chapter 2, space is a social construct. The nature of urban (and other) spaces is the product of social relations among actors with different interests. When we refer to the global and the local as being related dialectically, we do not mean to imply that scale or geographic space per se should be given theoretical or political priority in determining the social processes underlying urban change. When we speak of global and local scales we are, in fact, focusing on the processes operating at these (and other intermediate) scales, not on space or scale per se.12 Further, in the real world the significance of different geographic spaces and scales is in a constant state of flux, with certain spaces/scales declining in social importance and others increasing. Changes in the relative importance of geographic spaces/scales are reflected in changes in the distribution of power among social groups. This may result in a strengthened position for some (for example, a spatially mobile TNC) and a weakened position for others (for example, a place-bound minority community). We shall develop the concept of social power later in the book (see Chapters 20 and 29). Here it is sufficient to recognise that the notion of discrete spatial scales, global and local, is a simplification of a reality in which different actors have interests, and operate simultaneously, at multiple spatial scales. A TNC makes decisions on the local-regional scale in relation to its branch plant operations, negotiates tax advantages with national governments, secures capital in the international financial markets and adapts its planning strategy to satisfy institutional shareholders based in a variety of countries. Equally, local actors opposed to the closure of a factory providing local employment may use direct action through a ‘work-in’ at the plant, liaise with regional trade union organisers, lobby national government to exert pressure on the parent company or employ international legal experts to pursue the option of an employee buyout of the plant.13 LOCAL AND HISTORICAL CONTINGENCY As was suggested above, in addition to the operation of macro-scale trigger factors, the processes of urban change are also influenced by a host of locally and historically contingent factors that interact with and mediate more general forces of change. There are Urban geography 12 as many examples of such forces as there are cities in the world. They range from the physical impact of a constricted site on urban development, as in the high-rise architecture of Hong Kong island, to socio-demographic influences such as the concentration of retirees in seaside towns of southern England, or the particular cultural economy of Las Vegas. More generally, global forces come to ground in cities. As Sassen (1996 p. 630) observes, cities are the locations where much of ‘the work of globalisation gets done’.14 Local context, such as the urban economic base, social structure, political organisation, tax regulations, institutions and competing interest groups, exerts a powerful influence on urban change. The central importance of cities within the global economy and society is affirmed also in the ‘hollowing out of the state’ thesis. This contends that globalisation has led to the supranational and local scales becoming more important loci of economic and political power than the nation-state, which, in contrast to its central role during the Fordist era, has assumed the role of enabler rather than that of regulator of economic activity.15 The importance of local action within a globalising world is also a direct consequence of the post-Fordist drive for economic flexibility and product differentiation. The economic advantages that may accrue to cities that can provide attractive production environments (for example, an educated and skilled labour force, good climate and high-quality life space) stimulate innovation and encourage local communities to emphasise the particular advantages of their city as a place to live and work.16 The concept of a dialectic relationship between processes operating at global, local and other intermediate scales provides an essential backcloth to our study of urban geography. PROCESSES OF URBAN CHANGE The interaction of global ‘trigger forces’ and locally contingent factors results in a number of different processes of urban change. These are summarised in Figure 1.1 and will be discussed in detail later (see Chapters 4 and 21). Urbanisation occurs when cities grow at the cost of their surrounding countryside, suburbanisation and exurbanisation when the inner ring or commuter belt grows at the expense of the urban core, counterurbanisation when the population loss of the urban core exceeds the population gain of the ring, resulting in the agglomeration losing population overall, and reurbanisation when either the rate of population loss of the core tapers off or the core starts to regain population while the ring still loses population. These processes of urban change are visible to varying degrees in metropolitan areas of both the developed world and the Third World. The phenomena of peripheral urbanisation and exo-urbanisation are characteristic of cities in the Third World. The concept of peripheral urbanisation reflects the expansion of capitalism into Third World regions and employs a political economy perspective (see Chapter 2) to describe the impact of global capitalism on national urban systems in the Third World. Exourbanisation is promoted by foreign direct investment in Third World countries leading to a pattern of urban growth based on labour-intensive and assembly manufacturing types of export-oriented industrialisation, as in the Pearl River delta region of China. Urban geography: from global to local 13 The varying processes of urban change underline the value of examining global forces within local context in order to understand the complexity of contemporary urban forms. In practice, the reciprocal interaction between global and local forces leads to a plethora of urban outcomes. URBAN OUTCOMES The outcomes of the processes of globalisation and urban change are depicted in Figure 1.1. Three principal outcomes refer to: 1. changes in urban systems at local, regional, national and global scales; 2. the spread of urbanism; 3. changes in the socio-spatial construction of urban places. Each of these principal outcomes subsumes a host of related processes and outcomes. As we progress through Figure 1.1, from left to right, we obtain more detailed insight into the major themes and issues of relevance to the study of urban geography. In essence, Figure 1.1 provides a set of ‘signposts’ to the subject matter of urban geography. All these topics will be discussed later in the course of the book. At this point, however, it is useful to think about the kinds of question we need to ask to develop an understanding of urban geography. Appropriate questions include where, when and why did the first cities arise? What are the major urban land uses? Further questions focused on the urban economic base, on the social composition of residential environments, urban political structure, urban policy and planning, and the concept of sustainable urban development, also shed light on the dynamics of urban change. The list of questions that could offer insight into the form and functions of cities is almost limitless. (You can think of some of your own.) Some key questions that reflect the nature of urban geography, and the content of this book, are shown in Table 1.1. Examination of Figure 1.1 and Table 1.1 provides prospective students with a useful overview of the scope and content of urban geography. A second method of gaining an understanding of the nature of urban geography is to engage in field observation in the tradition of nineteenth-century social commentators such as Charles Booth in London, Friedrich Engels in Manchester and Jacob Riis in New York. The central area of most cities TABLE 1.1 THEMES AND QUESTIONS IN URBAN GEOGRAPHY Theme Question Urban geography: from global to local Why study urban geography? What is the nature of the relationship between global and local forces in the production of urban environments? How has globalisation influenced the form and function of cities? Concepts and theory in urban geography What do we mean by urban? What contribution have the different major theoretical Urban geography 14 perspectives made to understanding in urban geography? The origins and growth of cities When, where and why did the earliest cities arise? What are the social and spatial characteristics of pre-industrial, industrial and post-industrial cities? The global context of urbanisation Are socio-cultural differences between cities declining as a and urban change result of globalisation? How important is your own city in local, regional, national or global terms? Regional perspectives on urbanisation and urban change Why are cities in the Third World growing faster than cities in the developed world? How did the urban settlement pattern of North America develop? Why are many of the greatest centres of urban life in Europe? National urban systems What functions do cities perform? Does any single city or set of cities dominate the national urban system? How extensive is the hinterland of your city? Land use in the city What are the major urban land uses? What factors and agents underlie land-use change in the city? How do different populations use urban space? Urban planning and policy Why does the impact of urban planning and policy vary between countries? How does public planning affect urban development in the capitalist city? What effect has the collapse of communism had on the form and function of the socialist city? New towns What is the difference between an ‘old town’ and a ‘new town’? Should the growing demand for housing be met by building on greenfield land or on brownfield sites in existing cities? Residential mobility and neighbourhood change Why do households move? How do urban housing markets work? What factors lead to housing abandonment, gentrification and neighbourhood regeneration? Theme Questions Housing problems and housing policy What is the relative importance of different housing tenures in the city? Is there an adequate supply of decent and affordable housing? Why are people homeless? What can governments do to improve the quality and availability of housing? Urban retailing What forms of retail facility are available in the city? What impact has suburbanisation had on the structure of retail provision? Urban geography: from global to local 15 What difficulties confront disadvantaged consumers? Urban transportation What are the dominant modes of travel in the metropolitan area? What is the relationship between transport policy and urban structure? Is the car-oriented city sustainable? The economy of cities What is the economic base of the city? How is the urban economy related to the global economy? To what extent has your city been affected by industrialisation, deindustrialisation or tertiarisation in recent decades? What role does the informal sector play within the urban economy? Poverty and deprivation in the Western city What aspects of multiple deprivation are evident in the city? Is there an inner-city problem? Do particular neighbourhoods suffer from the effects of longterm decline? What can be done to alleviate the problems of poverty and deprivation? National and local responses to urban economic change What strategies have been employed to address the problems of urban economic decline? Is there a role for local economic strategies or for the community economy in urban economic regeneration? What has been the impact of property-led regeneration on the socio-spatial structure of the city? Are cultural industries important? Collective consumption and social justice in the city Is there any evidence of social or spatial inequity in access to public services? To what extent have reductions in public expenditure impacted on the provision of welfare services in the city? Theme Questions Residential differentiation and communities in the city Does the city, or particular neighbourhoods within the city, generate a sense of place for residents? Why does residential segregation occur? What is the extent and incidence of social polarisation in your city? How has it changed over recent years? Urban liveability How would you define a liveable city? What are the main hazards of urban living? Is there a relationship between urban design and social behaviour? How do different populations, such as women, elderly people, children, members of minorities and disabled people, cope with the pressures of urban life? Power, politics and urban governance What form of government controls the city? Does the political geography of the metropolitan area facilitate or constrain good government? Do citizens participate in urban government? Urban geography 16 What is the relationship between urban government and central government? Third World urbanisation within a global urban system How are the processes of urbanisation and urban change in the Third World related to those of the developed world? What evidence is there of the influence of colonialism on urban development? What is the likely role of Third World cities within the global urban system of the twenty-first century? The internal structure of Third World cities What are the main structural differences between cities of the Third World and those of the developed world? How has the abolition of apartheid affected the form of cities in South Africa? Is the concept of the Islamic city valid? Rural-urban migration in the Third World What are the relative contributions of migration and natural population change to urban population dynamics? Why do people migrate to cities and what are the consequences for the urban geography of the Third World? What is the response of governments to rural-urban migration? Urban economy and employment What are the major components of the urban economy? in the Third World What is the role of women and children in the urban economy? How important is the household economy in the coping strategies of the urban poor? Theme Questions Housing the Third World urban poor What are the main sources of shelter for the Third World urban poor? When, where and why do squatter settlements arise? What response have Third World governments made to the problem of low-income housing provision? Environmental problems in Third What are the major environmental hazards faced by poor urban World cities residents in the Third World? How do Third World cities manage the wastes produced by their burgeoning populations? What is the ecological impact of the city on the surrounding area? Health in the Third World city What are the chief causes of morbidity and mortality in the Third World city? Are the health problems in Third World cities different from those experienced in cities in the developed world? What can be done to address the health problems of Third World cities? Traffic and transport in the Third What are the main forms of urban transport? World city What role is played by paratransit? What are the major traffic and transport problems? Poverty, power and politics in the Third World city How do the poor respond to their disadvantaged position within the urban power structure? Urban geography: from global to local 17 How important is clientalism? What is the significance of urban social movements in the Third World city? The future of the city—cities of the future What is sustainable urban development? What is the nature of the relationship between energy consumption and urban form? What will be the form of the city of the future? may be explored easily on foot at a pace that enables the ‘participant observer’ to absorb the structure of the city and engage with the ebb and flow of urban life. In Glasgow a stroll of less than a mile from the central railway station through the old city to the cathedral represents a journey back in time, over a millennium of urban history. In cities with extensive public transportation systems the urban explorer can cover large areas. In London, for example, the Underground rail system affords access to a variety of urban milieux: the financial heart of the City of London (Bank station), the political hub (Westminster), an ethnic Bangladeshi community (Shoreditch), Georgian residential squares and terraces (Leicester Square), an environment of the super-rich (Kensington), disadvantaged council estates (Hackney) and areas of waterfront regeneration (Tower station). In many cities, especially the low-density metropolitan areas of North America, private transportation is a prerequisite for exploration beyond the urban core. All towns and cities can be ‘read’ by an observer trained in the basic principles and concepts of urban geography. Here, we conclude our introduction to the subject by undertaking a ‘virtual field trip’ to experience the daily rhythms of life in an advanced capitalist city. You can undertake a similar exercise for your own city. DAILY LIFE IN THE GLOBAL VILLAGE You awake to the sound of an alarm clock bought in the local hardware store but manufactured in Taiwan. You breakfast on fresh orange juice from Florida, tea or coffee from Sri Lanka or Brazil, bread made from wheat grown on the prairies of North America, jam from Bulgaria, butter from New Zealand, bacon from Denmark and freerange eggs from a local organic farm. Depending on your job, and income, your choice of clothing for the day may have been designed and made in a Milanese fashion house or manufactured under sweatshop conditions by child or female labour in the Far East. As you leave for work in the city you nod a greeting to your neighbour who is teleworking from home, and to her partner who works flexitime to accommodate domestic commitments. You travel into the city in a car constructed in South Korea. In your office the ballpoint pen on your desk was manufactured in Germany. Your notepad is made from timber harvested from a renewable forest plantation in Sweden. You perform your job using a personal computer constructed and sold locally but designed in Silicon Valley, California, and loaded with software produced in Seattle. You may spend much of your working day in cyberspace, using e-mail and teleconferencing facilities to establish real-time communications with business contacts around the globe. All your discussions are in English, the emerging global language. Urban geography 18 In your lunch break you visit a McDonald’s, an icon of cultural globalisation in which the same menu and decor are available in almost every large city of the world. With time to spare, you stroll past the security officer at the entrance to a new shopping mall and enter a travel agency where you make a reservation for a summer trip to Bali. In the adjacent music store you pick up a copy of a CD that is currently a chart hit in over twenty countries. You might spend some time sitting on a public bench in the town square reading a newspaper that, in addition to local and national news and weather, contains reports of a bank robbery in Bogota, floods in Dhaka, share prices on the Tokyo stock market, pollution in Lagos and baseball results from the USA. Across the square a small crowd has gathered around a group of street musicians from Peru who are playing traditional folk tunes of the Andes. By the fountain an unemployed man distributes leaflets protesting against the loss of local jobs as a result of downsizing by a foreignowned TNC. Overhead a Russian jet airliner comes in to land at the city’s international airport. At the street intersection enterprising youths rush to clean car windscreens before the traffic lights turn to green. As you rise to return to the office, a party of Japanese tourists ask for directions to the cathedral. You direct them past the redeveloped harbourside where postmodern pavement cafés and gentrified loft apartments located around a yacht marina are replacing the abandoned warehouses and wharves of an industrial past. On the opposite bank of the river the site of a recent garden festival awaits redevelopment under a public-private partnership scheme to build a concert hall, science museum, I-max theatre and a millennium tower. After work you meet some friends in an Irish theme pub, before going on to eat dinner in an Indonesian restaurant. As the city centre empties of its daytime workforce the space is reoccupied by a night-time population of service workers and entertainment seekers. The homeless search for a sleeping space for the night in the doorways of now silent commercial premises, while in the shadows prostitutes and drug dealers ply their trade. The journey out to your exurban home takes you past inner-suburban areas of social housing where groups of unemployed people congregate on street corners and regard passing strangers with a mix of curiosity and suspicion. Many of the young have never had a job and lack the educational qualifications to enter the labour force, while older men formerly employed in heavy engineering are without the flexible skills needed to compete for a place in the post-Fordist economy. Some inner-city neighbourhoods are occupied by ethnic minorities, while others accommodate lifestyle communities, each contributing to the cultural diversity of the metropolis. Farther out, towards the edge of the city, the proliferation of middle-class gated communities underlines the levels of segregation and social polarisation within the city. Once you are on the urban motorway the lights of the city are soon left behind. Out of sight, out of mind—until tomorrow. WHY STUDY URBAN GEOGRAPHY? You should now be in a good position to answer this question. Urban geography provides an understanding of the living environments of a majority of the world’s population. Knowledge of urban geography is of importance for both students and citizens of the contemporary world. Urban places are complex phenomena. Urban geography untangles Urban geography: from global to local 19 this complexity by explaining the distribution of towns and cities, and the socio-spatial similarities and differences that exist between and within urban places. This book adopts a global perspective for the study of urban geography. The perspective acknowledges: 1. the interconnectedness of global urban society as a result of globalisation; 2. the dialectic relationship between global and local processes in the construction and reconstruction of urban environments; 3. the importance of local and regional variations in the nature of urbanism within the overarching concept of a global economy and society. The concepts, themes and issues introduced in this opening chapter illustrate the complexity of urban phenomena and the explanatory power of urban geography. Concepts and theory provide the essential framework for explanation in any academic subject. Accordingly, proper understanding of urban geography must be based on a combination of theoretical insight and empirical analysis. For the author of a textbook, deciding on how and when to present students with the (sometimes difficult) theoretical content is always a tricky question. This book favours a dual approach. Throughout the text, relevant theory and concepts are integrated with empirical evidence to illuminate particular themes and issues under investigation. In addition, however, since students should be familiar with the key philosophical perspectives and debates at an early stage, these are introduced in the following chapter. Together, the two chapters that comprise Part One of the book provide a foundation for our subsequent discussion of the wide range of themes and issues that constitute the subject matter of urban geography. FURTHER READING BOOKS J.Benyon and D.Dunkerley (2000) Globalization: The Reader London: Athlone Press L.Budd and S.Whimster (1992) Global Finance and Urban Living London: Routledge S.Clark and G.Gaile (1998) The Work of Cities Minneapolis, MN: University of Minnesota Press K.Cox (1997) Spaces of Globalization: Reasserting the Power of the Local New York: Guilford Press P.Dicken (1998) Global Shift: Transforming the World Economy New York: Guilford Press J.Eade (1997) Living the Global City: Globalisation as Local Process London: Routledge P.Knox and P.Taylor (1995) World Cities in a World System Cambridge: Cambridge University Press R.Robertson (1992) Globalisation London: Sage S.Sassen (1994) Cities in a Global Economy Thousand Oaks, CA: Pine Forge Press J.Short and Y-H.Kim (1999) Globalization and the City London: Longman JOURNAL ARTICLES E.Berner and R.Korff (1995) Globalization and local resistance: the creation of localities in Manila and Bangkok International Journal of Urban and Regional Research 19(2), 208–21 N.Brenner (1999) Globalisation as reterritorialisation: the re-scaling of urban governance in the European Union Urban Studies 36(3), 431–51 Urban geography 20 K.Cox (1995) Globalization, competition and the politics of local economy development Urban Studies 32(2), 213–24 M.Frost and N.Spence (1993) Global city characteristics and central London’s employment Urban Studies 30(3), 547–58 D.Graham and N.Spence (1995) Contemporary deindustrialisation and tertiarisation in the London economy Urban Studies 32(6), 213–24 C.Hamnett (1994) Social polarisation in global cities: theory and evidence Urban Studies 31(3), 401–24 J.May (1996) Globalization and the politics of place: place and identity in an inner London neighbourhood Transactions of the Institute of British Geographers 21, 194–215 G.Mohan (2000) Dislocating globalisation: power, politics and global change Geography 85(2), 121–33 S.Sassen (1996) Cities and communities in the global economy: rethinking our concepts American Behavioral Scientist 39, 629–39 D.Walker (1996) Another round of globalization in San Francisco Urban Geography 17(1), 60–94 KEY CONCEPTS ■ urbanisation ■ urban growth ■ urbanism ■ global perspective ■ trigger factors ■ globalisation ■ glocalisation ■ global-local nexus ■ spatial scale ■ local contingency ■ hollowing out of the state ■ global village STUDY QUESTIONS 1. Why study urban geography? 2. Examine the influence of global-scale ‘trigger factors’ in urban change. 3. With reference to particular examples, illustrate how locally or historically contingent factors can mediate the urban impact of globalisation. 4. In his play Coriolanus the English dramatist William Shakespeare asked, ‘What is the city but the people?’ What did he mean? Write an account of a city as seen through the eyes of a member of a particular social group, such as a yuppie, homeless person, public official, property speculator or member of an ethnic or lifestyle minority. PROJECT Most towns and cities can be ‘read’ by an observer with a knowledge of urban geography. Using the example of the ‘virtual field trip’ presented in this chapter as a guide prepare an analytical description of the pattern of daily activity in any city with Urban geography: from global to local 21 which you are familiar. In your account you should: 1. attempt both to describe and to interpret the observed patterns of daily life in your city; 2. consider how the different population groups and areas fit into the life of the city; 3. relate your observations of urban life to the key themes and issues in urban geography identified in this chapter. Which of the general themes and issues are of particular relevance to your city? Does your city exhibit specific local characteristics that influence its form and function? 4. illustrate your account, if you wish, with relevant supporting material such as field sketches, photographs, newspaper cuttings, published statistics or government reports. 2 Concepts and Theory in Urban Geography Preview: the scope of urban geography; the concept of urban; the urban as an entity; urban as a quality; the importance of a spatial perspective; the value of the urban dimension; a history of urban geography; environmentalism; positivism; behaviouralism; humanism; structuralism; managerialism; postmodernism; moral philosophy; levels of urban analysis INTRODUCTION As we saw in Chapter 1, in approaching a subject for the first time it is useful to begin by obtaining an overview of the main conceptual approaches, themes and issues that comprise the field. In this chapter we describe the scope of urban geography and its links with other branches of geography. We define the concept of urban and explain the value of an urban geographical perspective for an understanding of contemporary towns and cities. We establish the academic context for the study of urban geography by providing a brief history of the subject. In this discussion we relate work in urban geography to the major theoretical developments in the discipline of geography. Finally, we employ the concept of levels of analysis to illustrate the kind of research undertaken by urban geographers from the global to the local scale. THE SCOPE OF URBAN GEOGRAPHY Urban geographers are concerned to identify and explain the distribution of towns and cities and the socio-spatial similarities and contrasts that exist within and between them. There are thus two basic approaches to urban geography: 1. The first refers to the spatial distribution of towns and cities and the linkages between them: the study of systems of cities. 2. The second refers to the internal structure of urban places: the study of the city as a system. In essence, urban geography may be defined as the study of cities as systems within a system of cities.1 Figure 2.1 indicates the scope of urban geography as well as the subdiscipline’s links with other branches of geography. The diagram also indicates the power of urban geography to synthesise many different perspectives so as to advance our understanding of urban phenomena. This eclectic approach to the analysis of urban places extends beyond geography to incorporate research findings and knowledge across Concepts and theory in urban geography 23 traditional disciplinary boundaries. The integrative power of urban geography is a key characteristic of the subdiscipline. A second principal characteristic of geographical analysis of the city is the centrality of a spatial perspective. This distinguishes urban geography from cognate areas of urban study such as urban economics, urban sociology or urban politics. We shall see later that there is a long-standing debate among social scientists over the relative importance of spatial and social forces for the explanation of urban phenomena. However, as we saw in Chapter 1, it is important to be clear about the place of space in urban geography. By acknowledging the importance of spatial location we are not implying that space per se is the key explanatory variable underlying patterns of human activity in the city. The significance of space varies with context. For example, spatial location is of no real significance in the Figure 2.1 The nature of urban geography electronic hyperspace occupied by flows of finance between cities in the global economy but may be of fundamental importance for the spread of infectious diseases in a Third Urban geography 24 World squatter settlement. The spatial perspective of urban geography is of real analytical value because, as Massey (1985)2 observed, the world does not exist on the head of a pin. DEFINING THE URBAN In approaching the concept of urban it is useful to draw a distinction between the question of what is an urban place, and what is urban. This is more than an exercise in semantics. The distinction between the urban as a physical entity and the urban as a quality helps us to understand the complexity of urban life, and illuminates different approaches to the study of cities. THE URBAN AS AN ENTITY Four principal methods are employed to identify urban places: 1. Population size. Since urban places are generally larger than rural places, at some point along the population-size scale it should be possible to decide when a village becomes a town. In practice, this urban population threshold varies over time and space. In Sweden any settlement with more than 200 inhabitants is classed as urban in the national census, whereas in the USA the population minimum for urban status is 2,500; in Switzerland it is 10,000, rising to 30,000 in Japan. Such diversity reflects social context. Given the sparse distribution of settlement in many areas of Sweden, a threshold of 200 may be appropriate, whereas in a densely settled country such as Japan virtually all settlements would exceed such a low urban threshold population. If not made explicit, these differences may complicate international comparison. 2. Economic base. In some countries population size is combined with other diagnostic criteria to define an urban place. In India, for example, a settlement must have more than 75 per cent of the adult male population engaged in non-agricultural work to be classified as urban. 3. Administrative criteria. The majority of towns and cities in the world are defined according to legal or administrative criteria. The definition of urban places by national governments leads to great diversity, which creates difficulties for comparative research that can be overcome only by urban analysts constructing their own definitions and applying them uniformly across the globe. A second problem with administrative definitions is that these may have little correspondence with the actual physical extent of the urban area. A frequent problem is under-bounding, where the built-up area of the city extends beyond the urban administrative boundary. This may lead to major fiscal difficulties for the central city deprived of taxes from commuters resident beyond the legal boundaries of the city. 4. Functional definitions. To address problems such as underbounding (and its converse, overbounding), urban researchers devised ‘functional urban regions’ which reflect the real extent of urban influence. The concept of the extended urban area was first introduced by the US Bureau of the Census in 1910 and later developed into the Standard Metropolitan Statistical Area (SMSA) in 1960 and, since 1983, the Metropolitan Statistical Area (MSA). Consolidated Metropolitan Statistical Areas Concepts and theory in urban geography 25 (CMSAs) are formed by two or more contiguous MSAs.3 As Table 2.1 shows, the US definition includes measures of population size, centrality and economic function. In 2000 a review of standards for defining US metropolitan statistical areas retained the two main principles established in 1960: 1. settlement form (based on the population size of a central core city); 2. functional integration between central and outlying counties (reflected in journeys to work, with this criterion raised from 15 per cent to 25 per cent). Other criteria for inclusion within a metropolitan area (see Table 2.1) were dropped. The 2000 standards identify two main types of core based statistical areas (CBSAs). These are: 1. metropolitan statistical areas, defined around at least one Census Bureau-defined urbanised area of 50,000 or more population; 2. micropolitan statistical areas, defined around at least one urban cluster of at least 10,000 and less than 50,000 population. Adjacent CBS As that have sufficient employment interchange (measured using journeyto-work data) are grouped to form larger ‘combined statistical areas’. Another significant development was the replacement of a ‘central cities’ classification with one of ‘principal cities’ (defined on the basis of a variety of population and employment data). While this will capture many of the previous central cities it will also reflect recent changes in the US urban landscape by identifying newer outlying employment centres as principal cities. Within metropolitan statistical areas the 2000 TABLE 2.1 DEFINITIONS OF URBAN AREAS IN THE USA AND UK The US Standard Metropolitan Statistical Area is: 1. Either one central city with a total population of 50,000 or more or two contiguous cities constituting a single community with a combined population of 50,000 and a minimum population of 15,000 for the smaller of the two. 2. The remainder of the county to which the central city belongs. 3. Adjacent counties, if: ■ 75 per cent or more of the labour force is non-agricultural. ■ At least 15 per cent of workers in the outlying county work in the central county, or 25 per cent of workers in that county live in the central county, i.e. there are significant journey-towork links between areas. ■ At least 50 per cent of residents in a county meet density requirements or non-agricultural employment thresholds. Metropolitan Statistical Areas are also categorised into four levels based on total population (level A: MSAs of 1 million inhabitants or more; level B: 250,000–1 million; level C: 100,000–250,000; level D: less than 100,000). In level A MSAs, Primary Metropolitan Statistical Areas may be identified. Each PMSA consists of Urban geography 26 a large urbanised county or cluster of counties that demonstrates very strong internal economic and social links, in addition to close ties with neighbouring areas. When PMSAs are defined, the MSA of which they are component parts is redesignated a Consolidated Metropolitan Statistical Area. Not all PMSAs, however, have a central city. The UK Standard Metropolitan Labour Area has a core plus ring with a combined population of at least 70,000 with: 1. A core consisting of a local authority administrative area or number of contiguous areas with a density of five jobs or more per acre (13.75 per hectare); or a single administrative area with 20,000 or more workers. 2. A ring consisting of administrative areas contiguous to the core and sending 15 per cent of their economically active population to that core. standards identify two types of counties as a basis for metropolitan divisions. These are: 1. main counties, with 65 per cent or more of employed residents who remain in the county to work, and with a jobs-to-resident-workers ratio of 0.75 or greater; 2. secondary counties, with a high jobs-to-resident-workers ratio (0.75 or greater), but a lower percentage of employed residents working within the county (50–64.9 per cent). Main counties can stand alone as a metropolitan division or can provide the organising basis for a metropolitan division. Secondary counties must combine with another secondary county or with a main county to form the basis of a metropolitan division. The remaining counties of an MSA are assigned to the main and secondary counties with which they have the highest commuting interchange. Metropolitan divisions, if present in an MSA, will account for all of its territory. Comparison of the 2000 metropolitan standards with the 1990 standards revealed that the new definitions accounted for 90 per cent of the US population (compared with 80 per cent for the 1990 definitions). Clearly, the application of the new urban standards will result in a change of status for many US counties. Geographers in the UK have sought to define a similar set of daily urban systems. A first attempt identified Standard Metropolitan Labour Areas comprising a core plus metropolitan ring that together formed the daily urban system (Table 2.1). A development of this system added an outer ring consisting of all local authorities that send more commuters to the core in question than to any other core, the whole being designated a Local Labour Market Area (LLMA). A conceptually similar scheme is that based on Functional Urban Regions, which have been used to compare changing patterns of urbanisation in Western Europe.4 THE URBAN AS A QUALITY In contrast to definitions of the city as a physical entity, the concept of the urban as a quality is related more to the meaning of urban places and the effect of the urban milieu on people’s lifestyles (and vice versa). Clearly, although cities exist as physical objects, it is by no means certain that they are perceived by their inhabitants in the same way that they are objectively structured. It is reasonable, therefore, to think of a city as having both an objective physical structure and a subjective or cognitive structure. Concepts and theory in urban geography 27 We each have our own conception of what a city is and of our local town or city. The same urban space can be seen in different ways by residents, tourists, workers, elderly people, unemployed people, women and children. For the homeless person the city may be a cold, anonymous and inhospitable place; for the elderly a spatially restricted world; for the wealthy a cornucopia of opportunity and well-being. Understanding these subjective interpretations of the urban is important, because meanings inform us not only about the places to which they refer but also about the people who articulate them and the social context in which they live. Urban geographers and others have sought to identify urban meaning through two main approaches. 1. Cognitive mapping. Geographers, planners and environmental psychologists have employed mental maps or cognitive mapping techniques to explore the subjective world of urban places, with a view to both obtaining a better understanding of human behaviour in the urban environment and improving the quality of urban life.5 Whereas traditional means of cognitive mapping provide subjective spatial representations of urban environments, more recently postmodern approaches seek to ‘map’ the meanings of the city for different ‘textual communities’ who share a common understanding of the ‘text’ and organise their lives accordingly as, for example, in the creation of a ‘suburban mentality’.6 2. Urbanism as a way of life. Early efforts to identify urban places in terms of a distinct lifestyle were based on Wirth’s (1938) concept of a rural-urban continuum.7 This argued that as the size, density and heterogeneity of places increased, so did the level of economic and social disorganisation. Wirth, a member of the Chicago school of human ecology, regarded urbanisation as a process leading to the erosion of the moral order of society due to the concomitant decline of community. He saw the urban as a separate spatial realm with its own environmental influences on individuals, and he contrasted the social disorganisation of urban life (in which much social interaction is of a transitory and superficial nature with ‘unknown others’) with the strong extended family links and communities in small settlements and rural areas. More recent perspectives that acknowledge the interpenetration of social realms have rejected the crude dualism of bipolar concepts such as urban-rural or public-private. Accordingly, although cities do exert a particular influence on their inhabitants the concept of a rural-urban continuum has been criticised: 1. for its Western ethnocentrism (which assumed that the rural-urban change process is universally applicable); 2. by studies which reveal the presence of ‘village communities’ in cities (including Wirth’s own work on the ghetto);8 and 3. for failure to locate the process of urbanisation within the political economy of capitalism (not least the impact of wider social, economic and political changes in rural areas, as demonstrated by the presence of ‘urban’ societies in supposedly rural areas).9 In seeking to reinterpret the meaning of ‘urban’, Harvey (1973, 1985) and Castells (1977) dispensed with the notion of a separate urban realm and concluded that while urbanism (as a way of life associated with residence in an urban area) has a distinctive structure and Urban geography 28 character, it exists within a larger framework created by the forces of capitalism.10 This means that ‘urban lifestyles’ can spread beyond the physical limits of the city. The quintessential diversity of urban life is central to postmodern representations of the city. Informed by processes of globalisation, social polarisation, cultural fragmentation and advances in information and communications technology, these focus on the rise of new cultural groupings and urban spaces, such as those defined by lifestyle communities.11 Postmodern readings of the city as ‘text’ employing urban metaphors, such as the city as jungle, bazaar, organism and machine,12 produce a multitude of representations of cities from the perspectives of different populations. In order to understand the truth of the city we need to acknowledge the ‘reality’ of the city as a concrete construction (thing) and as an abstract representation (idea), and examine how each influences the form of the other. THE SIGNIFICANCE OF SPACE AND PLACE Place is part of but different from space. Place is a unique and special location in space notable for the fact that the regular activities of human beings occur there. Moreover, because it is a site of such activities and all that they entail, place may furnish the basis of our sense of identity as human beings, as well as for our sense of community with others. In short, places are special sites in space where people live and work and where, therefore, they are likely to form intimate and enduring connections. As we see in Chapter 18, even in a globalising world a sense of place is of real importance in people’s daily lives. Paradoxically, the advent of cyberspace has re-focused attention on the importance of places in urban life. There is growing recognition among urban scholars that place is a central concept in the analysis of how urban areas are constructed and come to have meaning for their residents. Furthermore, as the constraints of geographical distance become less important, the specific features of particular locales are becoming more important in the locational decisions of businesses and households. The ‘construction’ of place is also a characteristic of the restructuring of many contemporary cities from being centres of production (for example, the steeltowns of yesteryear) to being centres of consumption (for example, Las Vegas of today), in the sense that they provide the context in which goods and services are compared, evaluated, purchased and used. Places such as London’s Covent Garden or Fisherman’s Wharf in San Francisco obtain a distinctive character that not only reinforces the place’s sense of identity but transforms the locality into an ‘item of consumption’, a process often boosted by city marketing strategies (see Chapter 16). In striving to redefine ‘urban’ in other than empirical terms, and to explain the meaning of urban life, both Harvey and Castells rightly rejected notions of ‘spatial fetishism’ (which assigned causal power to space per se in determining human action) and emphasised the need to examine the role of urban places in capitalist society. However, it is important not to dismiss completely the power of space. Space is more than a medium in which social, economic and political processes operate. The dimensions of space—size, density, distance, direction, territory and location—exert powerful influences on urban development and on human interaction. Distance and direction have Concepts and theory in urban geography 29 a direct effect on social networks, journeys to work and physical access to place-bound facilities. The size and density of residential developments have been shown to influence the incidence of deviant behaviour and social pathologies. The concept of territoriality contributes to the formation of discrete sub-areas within cities, often segregated on ethnic or class lines. The partitioning of space by local government boundaries has implications for the social composition and fiscal health of different areas and the quality of life of residents. Competition between local administrations to attract desirable developments, such as shopping centres, and exclude undesirable facilities, such as mental hospitals, also has a fundamental spatial basis. There is a reciprocal relationship between society and space. As Massey (1984 p. 52) observed, ‘just as there are no purely spatial processes, neither are there any non-spatial social processes’.13 A more recent challenge to the significance of geographical space has accompanied the advent of cyberspaces (such as the Internet), which for some authors has undermined the relevance—conceptual and actual—of ‘real’ space. It is argued that as a result of ‘time-space compression’, cyberspace negates the effects of physical separation and produces new ‘spaceless, placeless social spaces’ in which people can interact.14 While cyberspace undoubtedly has spatial implications, the suggestion that it heralds the ‘end of geography’ is excessive, not least because even in advanced societies connections to cyberspace are unevenly distributed.15 This is illustrated clearly by the contrast between elite, and often physically enclosed, communities with extensive links to the global information superhighway and spatially proximate ‘off-line’ urban spaces occupied by the poor, where time and space constraints are profoundly real.16 THE VALUE OF THE URBAN DIMENSION The Wirthian association of particular lifestyles with different settlement sizes is now regarded as unacceptably simplistic. Some have extended this argument to claim that the concept of urban has become redundant on the grounds that, in Western societies at least, we are all ‘urban’ no matter where we live. This has led some to conclude that it is neither fruitful nor appropriate to study the city in its own right.17 Leaving aside the fact that this represents a Western view of contemporary society which is not yet applicable to all parts of the world, the suggestion that the city cannot be a significant unit of study confuses two different questions. These are: 1. whether cities can be objects of analysis; 2. whether explanations of ‘urban phenomena’ can be restricted to the urban level. As Agnew et al. (1984)18 point out, a negative answer to the second question does not require a negative answer to the first. We can reject the idea of ‘urban explanation’ while accepting the urban as a valid object of analysis. Cities are places—or, more accurately, conglomerations of overlapping and interrelated places—and, as we have seen, places matter! Social relations occur in, and help to constitute, places. Individuals, households, communities, companies and public agencies exist and operate in particular places. People’s social, economic and political relations are embedded in the particularity of specific places. The results of global-level structural forces come to ground in particular places. To maintain that cities are merely elements in the capitalist mode of production Urban geography 30 ignores significant variations between and within urban places, as well as the complex suite of other social and cultural factors shown to be involved in the construction and reconstruction of urban environments.19 More recent arguments against the urban as a frame of reference reflect those that dismiss the value of space, and contend that advances in telecommunications technologies will lead to the effective dissolution of the city on the grounds that, once time has become instantaneous, spatial congregation becomes unnecessary.20 Such technologically deterministic views ignore the relational links between new information technologies and urban form which have been apparent ever since the introduction of the telegraph, wireless and telephone in the nineteenth century.21 New technologies contribute to the reconstruction of urban space but do not render it redundant. Visions of the ‘spaceless city’ greatly overestimate the degree to which virtual reality can substitute for place-based face-to-face interaction.22 As Amin and Graham (1997) conclude, the contemporary city, while housing vast arrays of telematic entry points into the burgeoning worlds of electronic space, is a cauldron of emotional and personal worlds and attachments, an engine of reflexivity, trust and reciprocity.23 While many cities and citizens are linked into an electronic ‘non-place urban realm’,24 place-based relational networks that rely on propinquity and physical interaction—the key characteristics of urban places—remain central to the experience of human social, economic and cultural life. The fundamental importance of towns and cities as focal points in contemporary society, in both the developed and the developing realms, emphasises the significance of the urban dimension and underlines the value of the study of urban geography. A BRIEF HISTORY OF URBAN GEOGRAPHY Urban geography is an established branch of geography that attracts researchers and students in significant numbers, and produces a large and expanding volume of published work to aid our understanding of the city. The advance of urban geography to a central position within the discipline has occurred over the past half-century. As Herbert and Johnston (1978 p. 1) noted: whereas in the early 1950s a separate course on urban geography at an English-speaking university was quite exceptional, today the absence of such a course would be equally remarkable; indeed, in many institutions students can opt for a group of courses treating different aspects of the urban environment.25 Urban geography is a dynamic subdiscipline that comprises a combination of past ideas and approaches, current concepts and issues that are still being worked out (Table 2.2). It may be likened to: a city with districts of different ages and vitalities. There are some longestablished districts dating back to a century ago and sometimes in need of repair; and there are areas which were once Concepts and theory in urban geography 31 Plate 2.1 The variety of urban environments is shown by examples of ‘waterfront living’ in (top) a shanty town built on stilts over a coastal marsh in the port-city of Manta, Ecuador, and (bottom) the gentrified St Katharine’s Dock in central London TABLE 2.2 THE EXPANDING SCOPE OF URBAN GEOGRAPHY Systems of cities Cities as systems Urban origins and growth Site and situation of settlements Regional patterns of settlement Urban morphology Townscape analysis Urban ecology Central place theory Settlement classification Social area analysis Factorial ecology Delimitation of the central business district Population movements Migration decisions Suburbanisation Residential mobility Retailing and consumer behaviour Urban imagery Urban geography 32 Urban origins and growth Site and situation of settlements The role of Regional patterns cities in ofthe settlement national political economy Urban morphology problems in structural context Townscape analysis Urban ecology Economic restructuring Poverty andanalysis deprivation Social area The inner-city problem Factorial ecology Delimitation of the central business district Housing markets and gentrification The urban property Residential mobilitymarket Traffic and transport problems Retailing and consumer behaviour The urban physical environment Urban imagery Housing, health and economy in Third Power World and citiespolitics Territorial justice The urban impact Differential accessoftoglobalisation services Social construction of urban space Cultural diversity in cities Social justice Urban liveability Sustainable cities Edge cities Counter-urbanisation Central place theory Settlement classification Rural-urban migration in the Third World Population movements Migration decisions Suburbanisation Urban and regional planning Globalisation of culture and society The global economy The global urban system World cities and global cities Megacities Technopoles Future urban form fashionable but are so no longer, while others are being rehabilitated. Other districts have expanded recently and rapidly; some are well built, others rather gimcrack.26 Since the late 1970s the scope of urban geography has expanded rapidly. For some commentators the increased diversity is a source of potential weakness that may lead ultimately to its disintegration. For others, including the present writer, the breadth of perspective strengthens urban geography’s position as an integrative focus for research on the city. Urban geographers have approached the study of the city from a number of philosophical perspectives. While the significance of each for the practice of urban geography has changed over time, none of the main approaches has been abandoned completely, and research informed by all perspectives continues to be undertaken under the umbrella of urban geography. It is important for students of urban geography to be familiar with the different philosophies of science which underlie the subject. To aid this understanding we shall structure our discussion of the changing nature of urban geography with reference to the main epistemological developments in the discipline. Concepts and theory in urban geography 33 ENVIRONMENTALISM During the first half of the twentieth century the major concerns of urban geography reflected the more general geographical interest in the relationship between people and environment, and in regional description. Early work on urban site and situation, and on the origins and growth of towns, was largely descriptive. One of the first Englishlanguage texts in urban geography comprised a classification and analysis of over 200 towns in terms of their site, situation, relief and climate.27 Despite their simplicity, these investigations provided a foundation for the more conceptually refined practice of urban morphological analysis which continues to illuminate patterns of urban growth and change to the present day.28 At the inter-urban scale, studies of regional patterns of settlement focused attention on the importance of transportation systems.29 Together with the work of land economists,30 this shifted attention from environmental factors towards the economics of location. The economic analysis of cities as points in space was developed most fully in central-place theory.31 At the intra-urban scale, work on urban ecology also moved attention from the environment to human behaviour, and introduced a social dimension to urban geography. Nevertheless, at mid-twentieth century the focus of urban geography was primarily on land use and related issues. The first major ‘paradigm shift’ to affect urban geography reflected a desire to make geographical investigation more scientific. This led to the introduction of the philosophy of positivism. POSITIVISM Positivism is characterised by adherence to the ‘scientific method’ of investigation based on hypothesis testing, statistical inference and theory construction (Box 2.1). Although evident in the work of Christaller (1933) and Losch (1954) on the spatial patterning of BOX 2.1 The assumptions of positivism 1. Events that occur within a society or which involve human decision-making have a determinate cause that is identifiable and verifiable. 2. Decision-making is the result of a set of laws to which individuals conform. 3. There is an objective world comprising individual behaviour and the results of that behaviour which can be observed and recorded in an objective manner on universally agreed criteria. 4. The researcher is a disinterested observer. 5. As in the study of inanimate matter, there is a structure to human society (an organic whole) that changes in determinate ways according to observable laws. 6. Application of the laws and theories of positivist social science can be used to alter societies in determinate ways (social engineering). Source: Adapted from R.Johnston (1983) Philosophy and Human Geography London: Arnold Urban geography 34 settlements,32 positivism blossomed in urban geography in the late 1950s with the development of the spatial analysis school. The redefinition of urban geography as the science of spatial relationships33 was accompanied by a shift in emphasis away from exceptionalism (the study of the unique and particular) towards a nomothetic approach (aimed at a search for abstract or universal laws). This was aided by the emergence of quantitative analytical techniques fuelled by the ‘quantitative revolution’ in geography of the 1950s and 1960s.34 The new approach led to multivariate classifications of settlement types,35 investigations of the rank-size rule for the populations of urban places,36 and analyses of spatial variations in urban population densities.37 Stimulated by translations of Christaller’s work, urban geographers devoted attention to modelling settlement patterns and the flows of goods and people between places.38 Concepts such as distance decay (the attenuation of a pattern or process over distance) were also introduced in the study of urban phenomena such as consumer behaviour, and trip generation and travel patterns.39 The subsequent expansion of computing power and development of geographical information systems has ensured that modelling and simulation remain a vibrant, if minority, area of urban geography.40 The new methods of spatial science were also applied to analysis of the internal structure of cities. The urban land-use models of the Chicago school of human ecology reflected the positivist philosophy in their proponents’ belief that human behaviour was determined by ecological principles or ‘natural laws’ which stated that the most powerful group would obtain the most advantageous position (e.g. the best residential location) in a given space. During the 1970s the development of a range of multivariate statistical techniques extended the social area approach of the ecologists in the form of factorial ecologies designed to reveal the bases of residential differentiation within the city.41 The positivist spatial science approach was also central to the models of urban structure introduced into geography from neo-classical economics. These were founded on the assumption of Homo economicus or the economic rationality of human behaviour. This stated that individual decisions were based on the goal of utility maximisation—that is, the aim of minimising the costs involved (usually in terms of time and money) and maximising the benefits. For two decades, spatial science was the dominant paradigm in urban geography. However, during the 1970s a growing awareness among geographers of alternative philosophies of science led to strong criticism of positivism: 1. The adequacy of an approach which focused on spatial form to the neglect of underlying causal processes was questioned. It was argued that, since spatial form was largely the outcome of the prevailing social forces, the focus of urban research should switch from the study of spatial relations to the study of social relations. 2. Particular criticism was directed at positivism’s mechanistic view of the role of humankind, and its failure to recognise and account for the idiosyncratic and subjective values that motivate much human behaviour. 3. The science of spatial relations affords no insight into the meaning of urban places to their inhabitants. To explore this sense of place requires an approach that focuses on the daily activities and perceptions of urban residents. Concepts and theory in urban geography 35 One response to a growing dissatisfaction with positivist science and the poor predictive ability of many spatial models was a move towards direct observation of human behaviour and decision-making. This led to the development of behaviouralism in urban geography.42 BEHAVIOURALISM The behavioural approach sought to overcome the shortcomings of spatial analysis by highlighting the role of cognitive processes and decision-making in mediating the relationship between the urban environment and people’s spatial behaviour. Urban geographers employed cognitive mapping techniques to examine a host of issues, including migration,43 consumer behaviour,44 residential mobility,45 residential preferences, perceived neighbourhood areas and images of the city.46 The behavioural approach introduced greater realism into urban studies, as the emphasis on empirical investigation of human behaviour countered the abstract nature of ‘spatial’ theory. But behaviouralism did not break away wholly from the positivist tradition. Much of the methodology of positivism was retained, and although focused on exposing the values, goals and motivations of human behaviour, it was still concerned with seeking law-like generalisations. As a consequence, behaviouralism has attracted much of the same criticism that has been levelled at positivism, in particular its failure to recognise and account for the ‘untidiness’, ambiguity and dynamism of everyday life. HUMANISM The humanistic approach views the individual as a purposeful agent of change in the city rather than a passive respondent to external stimuli. Although it is acknowledged that people do not act free of constraints, the humanist philosophy accords central importance to human awareness, agency, consciousness and creativity. The aim of a humanistic approach is to understand human social behaviour using methodologies that explore people’s subjective experience of the world. In practice, this means a change from the positivist principles of statistical inference based on representative random samples of the population to the principle of logical inference based on unique case studies using methods such as ethnography and analysis of literary texts to demonstrate ‘the social construction of urban space’.47 The humanistic perspective has been criticised for placing excessive emphasis on the power of individuals to determine their own behaviour in the city, and affording insufficient attention to the constraints on human decision-making. This critique was advanced most forcefully by proponents of structuralism (see below), who regarded the focus on the individual in society as a distortion of a reality in which people’s behaviour is conditioned by forces over which they have little control. STRUCTURALISM Structuralism is a generic term for a set of principles and procedures designed to expose the underlying causes of revealed patterns of human behaviour. In practice this means that explanations for observed phenomena cannot be found through empirical study of the Urban geography 36 phenomena alone but must be sought by examination of prevailing social, economic and political structures. Structural analysis in urban geography has been based primarily on the work of Marx.48 According to the Marxian or political economy approach, every society is built upon a mode of production—a set of institutional practices by which the society organises its productive activities, provides for its material needs, and reproduces the socio-economic structure. Capitalism is a specific mode of production (others being slavery, feudalism, socialism and communism). Cities are viewed as an integral part of the capitalist mode of production by providing an environment favourable to the fundamental capitalist goal of accumulation. This is the process by which the value of capital is increased through the continual reinvestment of profits from earlier investments. The effect of this expansionary dynamic is most visible in the changing urban land market and, as we shall see later, in processes such as urban redevelopment, gentrification and suburbanisation. The political economy approach entered urban geography in the early 1970s in response to the continuing social problems of urban areas, highlighted in the USA by the civil rights movement. In seeking to uncover the structural forces underlying observed social problems located in the dynamics of the capitalist system, it was argued that: 1. Capitalist society is characterised by conflict between socio-economic groups over the distribution of resources. A key resource is power, most of which is held by an elite who are able to manipulate the majority. 2. Since quantitative spatial analysis describes patterns but fails to reveal underlying causes, any proposals or policies based on this analysis will be supportive of the status quo and unable to lead to progressive social change. Much attention has been directed to the analysis of urban property and housing markets, and studies of residential patterns.49 Notwithstanding the exercise of ‘constrained choice’ by individuals, the political economy approach interpreted urban residential segregation primarily as a result of decisions by those with power in the property market, including building-society managers,50 estate agents51 and local-authority-housing managers.52 Harvey (1976)53 offered an incisive exposition of the relationship between urban residential patterns and the dominant political economy of monopoly capitalism. The dominance assigned to social structure over human agency in the structuralist perspective was rejected by humanistic geographers. Other critics have attacked the emphasis attached to class divisions in society to the neglect of other lines of cleavage such as gender, ethnicity and sexuality, all of which cut across class boundaries and which exert a significant influence on urban lifestyle and the processes of urban restructuring.54 Nevertheless, the political economy approach has had a major impact in urban geography and has provided real insight into the economic and political forces underlying urban change. MANAGERIALISM Doubts over the analytical value of class in modern societies led some writers to abandon Marx’s class-based analysis in favour of Weber’s concept of ‘social closure’—a process by which social groups seek to maximise their benefits by restricting access to resources Concepts and theory in urban geography 37 and opportunities to a limited circle of ‘eligibles’. (The practice of ‘exclusionary zoning’ of land use in the US city provides a good example.) This perspective on power and conflict in society is closely related to the concept of urban managerialism, which focuses attention on the power of urban managers—professionals and bureaucrats—to influence the socio-spatial structure of cities through their control of, for example, access to public housing, or the allocation of mortgage finance.55 Structuralists are dismissive of managerialism’s focus on intermediate-level decision-makers within a social formation. However, at the interface between consumers and allocators of scarce resources, managerialism introduces a humanistic perspective that can help to expose the operation and rationalities of the distributive process in cities. POSTMODERNISM Postmodern theory began to exert an influence on urban geography in the late 1980s and 1990s. The postmodern perspective is characterised by the rejection of grand theory and an emphasis on human difference. This distances postmodernism from both positivism, with its search for general laws and models, and structuralism, with its base in grand theory relating to the capitalist mode of production. The most visible impact of postmodern thinking on the city is in its architecture, where the ‘concrete functionalism’ of the modern era is replaced by a diversity of styles. In terms of the social geography of the city, the most important contribution of a postmodern perspective is how its focus on difference, uniqueness and individuality sensitises us to the needs and situations of all members of a society. This emphasis on the need to study urban phenomena from the multiple viewpoints of diverse individuals and groups has been reflected in studies of gender differences in urban labour markets,56 as well as of the ‘spaces of exclusion’ occupied by minority groups defined by class,57 marital status,58 sexuality,59 race,60 age61 and disability.62 A major criticism directed at the postmodern approach to the city is its apparently unlimited relativism. Because it privileges the views of all individuals, there appears to be no limit to the range of possible interpretations of any situation—there is, in effect, no ‘real world’.63 This has drawn particular criticism from ‘socially concerned’ urban geographers who decry postmodernism’s inability to address the ‘real’ problems of disadvantaged urban residents. MORAL PHILOSOPHY An approach based on moral philosophy or ethics represents an emergent perspective in urban geography. This seeks to examine critically the moral bases of society. Central to the ethical perspective is the concept of normative judgement that focuses on what should be rather than what is.64 This involves critical evaluation of actual situations against normative conditions as defined by ethical principles. In urban geography, researchers are confronted constantly by ethical issues. Questions addressed include the extent to which there is equity in the distribution of welfare services, employment opportunities and decent housing for different social groups in the city; how to interpret the causes of inner-city poverty (including the relative weight attached to personal deficiencies of the population, or structurally induced constraints on behaviour); and the social acceptability Urban geography 38 of existing urban conditions—for example, what is an acceptable level of air pollution or of infant mortality? Although moral perspectives on the city formed an important part of social science in the nineteenth century,65 the foundations of the current ethical perspective in urban geography lie in the humanistic approach, in the consideration afforded to issues of social justice within the political economy perspective of the early 1970s,66 and in more recent critiques of the ethics of market-oriented individualism.67 There is also a degree of commonality with the postmodernist emphasis on the importance of difference (seen, for example, in feminist critiques of male-centred interpretations of what constitutes a liveable urban environment).68 Most fundamentally, however, the ethical perspective rejects the post-modernist denial of the possible existence of generally applicable moral bases for societal behaviour. Further, an ethical perspective contends that in a society not all manifestations of ‘otherness’ should be fostered; some (racial discrimination or child prostitution, for example) should be constrained.69 As Smith (1994 p. 294) observed, while acknowledging the importance of ‘difference’ and ‘otherness’, ‘we should not allow uncritical deference to other people’s views and cultures to deny the possibility that some kinds of behaviour, ways of life and even moral codes are wrong’.70 IN SEARCH OF COMMON GROUND Each of the major philosophical perspectives considered can claim to illuminate some part of the complex dynamics and structure of the city. But no single approach provides a full explanation of urban phenomena. The question of whether an accommodation is possible among the different approaches has been polarised between those who accept a pluralist stance—‘agreeing to differ’ on the grounds that there is no single way to gain knowledge—and those who insist on the need to make a unitary choice of theoretical framework due to the perceived superiority of a particular theory of knowledge. Others have sought to combine approaches in different ways.71 The latter route, which incorporates a search for a middle ground between the generalisation of positivism and the exceptionalism of postmodern theory, is the approach favoured here. The analytical value of employing different theoretical perspectives is illustrated in Table 2.3, with reference to the question of urban residential structure. To fully understand urban phenomena in the contemporary world requires consideration of both the general structural processes related to the ‘mode of production’ and an empirically informed appreciation of the particular social formations that emerge from the interaction of structural forces and local context. The importance of employing a combined multi-layered ‘realist’ perspective72 that encompasses the global and local scales, social structure and human agency, and theory and empirical investigation in seeking to interpret the city informs the organisation and content of this book. LEVELS OF ANALYSIS IN URBAN GEOGRAPHY As we have seen, the character of urban environments is the outcome of the interplay of a host of private and public interests operating at a variety of geographical scales. In order Concepts and theory in urban geography 39 to understand the geography of towns and cities, therefore, it is necessary to look both within and beyond the settlement, and to examine the complex of factors involved in urban change at all levels of the global-local continuum. Although the factors and processes involved in urban development are not confined to any discrete level of the global-local spectrum, the concept of ‘levels of analysis’ offers a useful organising framework which simplifies the complexity of the real world and illustrates some of the issues of concern to urban geography at different spatial scales. We can identify five main levels of analysis. THE NEIGHBOURHOOD The neighbourhood is the area immediately around one’s home; it usually displays some homogeneity in terms of housing type, ethnicity or socio-cultural values. Neighbourhoods may offer a locus for the formation of shared interests and development of community solidarity. Issues of relevance to the urban geographer at this level include the processes of local economic decline or revitalisation, residential segregation, levels of service provision and the use of neighbourhood political organisations as part of the popular struggle to control urban space. THE CITY Cities are centres of economic production and consumption, arenas of social networks and cultural activities, and the seat of government and administration. Urban geographers examine the role of a city in the regional, national and international economy, and how the city’s socio-spatial form is conditioned by its role (for example, as a financial centre or manufacturing base). Study of the distribution of power in the city would focus on the behaviour and biases of formal organisations as well as the informal arrangements by which public and private interests operate to influence government decisions. The differential socio-spatial TABLE 2.3 ANALYTICAL VALUE OF DIFFERENT THEORETICAL PERSPECTIVES IN URBAN GEOGRAPHY: THE EXAMPLE OF URBAN RESIDENTIAL STRUCTURE Theoretical perspective Interpretative insight Environmentalism Although the notion of environmental determinism is now discredited, the influence of environmental factors on residential location can be seen in the problems of building in hazardous zones, and in the effects of architectural design on social behaviour (see examples in Chapter 19) Positivism Uses statistical analysis of objective social, economic and demographic data (e.g. via factorial ecology) to reveal areas in the city that display similar residential characteristics (see examples in Chapter 18) Urban geography 40 Behaviouralism Addresses the key question of why people and households relocate by examining the motives and strategies underlying the intra-urban migration of different social groups (see examples in Chapter 10) Humanism Explains how different individuals and social groups interact with their perceived environments, as in the differential use of public and private spaces within a city or residential neighbourhood (see examples in Chapter 19) Managerialism Illustrates how urban residential structure is affected by the ability of professional and bureaucratic gatekeepers to control access to resources, such as social housing or mortgage finance (see examples in Chapter 11) Structuralism Examines the ways in which political and economic forces and actors (e.g. financial institutions, property speculators and estate agents) influence the residential structure of a city through their activities in urban land and housing markets (see examples in Chapter 7) Postmodernism Explores the place of different social groups in the residential mosaic of the city by focusing on the particular lifestyles and residential experiences of various populations, such as ethnic minorities, affluent groups, gays, the elderly, disabled, and the poor (see examples in Chapter 18) Moral philosophy Critically evaluates the ethical underpinnings of issues such as homelessness or the incidence of slums and squatter settlements (see examples in Chapter 25) distribution of benefits and disbenefits in the city is also an important area of investigation in urban geography. THE REGION The spread of urban influences into surrounding rural areas and, in particular, the spatial expansion of cities have introduced concepts such as urban region, metropolis, metroplex, conurbation and megalopolis into urban geography. Issues appropriate to this level of analysis include the ecological footprint of the city, land-use conflict on the urban fringe, growth management strategies and forms of metropolitan governance. THE NATIONAL SYSTEM OF CITIES Cities are affected by nationally defined goals established in pursuit of objectives that extend beyond urban concerns. Successive New Right governments in the UK (under Margaret Thatcher and John Major) and the USA (under Ronald Reagan and George Bush) followed an economic policy that focused on national economic development largely irrespective of its consequences for the growth or decline of individual urban areas. Cities were encouraged to become more competitive to attract inward investment. National-level policy guidelines, incentives in the shape of competitive grants, and financial and other controls over the actions of local government have a direct influence on urban decision-making and management. In order to comprehend processes and patterns of urban change, geographers need to have an understanding of national policy and the ways in which it affects the inter- and intra-urban geography of the state. Concepts and theory in urban geography 41 THE WORLD SYSTEM OF CITIES The concept of a world system of cities reflects the growing interdependence of nations and cities within the global political economy. In this urban system, ‘world cities’ occupy a distinctive niche owing to their role as political and financial control centres. This status is evident in the concentration of advanced producer services such as education, R&D, banking and insurance, accounting, legal services, advertising and real estate services. Drawing on a ‘world cities’ perspective enables the urban geographer to reframe many urban questions previously defined in the context of the city or regional boundaries. This is illustrated graphically by the way in which investment decisions by managers in a Japanese-owned multinational company with head-quarters in New York can lead to job losses and deprivation at the neighbourhood level in Liverpool or Lagos. In studying the contemporary city, urban geographers must remain aware of the relationship between global and local forces in the production and re-production of urban environments. The need for such a perspective is reinforced by the process of globalisation, which, as we have seen, emphasises the linkages between different ‘levels of analysis’. In particular, the global and the local must be regarded not as analytical opposites but as two sides of the same coin. The process of globalisation underscores the need for urban geographers to employ a multilevel interdisciplinary perspective in the search for urban explanation. FURTHER READING BOOKS K.Cox and R.Golledge (1981) Behavioural Problems in Geography Revisited London: Methuen D.Gregory and J.Urry (1985) Social Relations and Spatial Structures London: Macmillan. D.Harvey (1973) Social Justice and the City London: Arnold D.Harvey (1985) The Urbanisation of Capital Oxford: Blackwell D.Herbert and R.Johnston (1978) Geography and the urban environment, in D.Herbert and R.Johnston (eds) Geography and the Urban Environment volume 1 Chichester: Wiley, 1–33 D.Ley and M.Samuels (1978) Humanistic Geography London: Croom Helm S.A.Marston, B.Towers, M.Cadwallader and A.Kirby (1989) The urban problematic, in G.Gaile and C. Willmott (eds) Geography in America Columbus, OH: Merrill, 651–72 R.Pahl (1965) Urbs in Rure London: Weidenfeld & Nicolson R.Peet (1998) Modern Geographical Thought Oxford: Blackwell JOURNAL ARTICLES A.Amin and S.Graham (1997) The ordinary city Transactions of the Institute of British Geographers 22, 411–29 B.Berry (1964) Cities as systems within systems of cities Papers and Proceedings of the Regional Science Association 13, 147–63 M.Dear and S.Flusty (1998) Postmodern urbanism Annals of the Association of American Geographers 88(1), 50–72 M.Gottdeiner (2000) Lefebvre and the bias of academic urbanism City 4(1), 93–100 F.Schaefer (1953) Exceptionalism in geography: a methodological examination Annals of the Association of American Geographers 43, 226–49 Urban geography 42 D.Walmsley (2000) Community, place and cyberspace Australian Geographer 31(1), 5–19 L.Wirth (1938) Urbanism as a way of life American Journal of Sociology 44, 1–24 KEY CONCEPTS ■ meaning of urban ■ urbanisation ■ urbanism ■ rural-urban continuum ■ space and place ■ cyberspace ■ social processes ■ spatial processes ■ environmentalism ■ positivism ■ behaviouralism ■ humanism ■ structuralism ■ managerialism ■ postmodernism ■ moral philosophy ■ levels of analysis STUDY QUESTIONS 1. What do you understand by the term ‘urban’? 2. Consider the value of an urban geographical perspective for understanding contemporary towns and cities. 3. With the aid of relevant examples, identify the concerns of the urban geographer at different levels of the global-local spectrum. 4. Dr Johnston (1709–84) said, ‘When a man is tired of London he is tired of life; for there is in London all that life can afford’, yet Shelley (1792–1822) thought that ‘Hell is a city much like London’. Make a list of the positive and negative features of urban life. 5. Select any one of the major theoretical perspectives (such as positivism, structuralism or postmodernism) and write a reasoned critique of its value for the study of urban geography. Concepts and theory in urban geography 43 PROJECT Select a major academic journal dealing with urban issues (e.g. Urban Studies or Urban Geography) and undertake a content analysis over a period of time to illustrate the changing focus of the subject. For each of the journal issues sampled you should classify the papers by their main theme. This information can then be tabulated and/or graphed to indicate the percentage coverage of different themes and hence the changing emphasis of urban geography over time. You should expect to find some enduring issues, some that decline in importance, as well as new entries to the field. PART TWO An Urbanising World 3 The Origins and Growth of Cities Preview: pre-industrial cities; theories of urban origin; hydraulic theory; economic theory; military theories; religious theories; early urban hearths; the spread of urban ism; the Dark Ages and urban revival in Western Europe; the medieval town; early modern urbanism; industrial urbanism; the form of the industrial city; residential segregation; the origins of urban USA; the westward progress of urbanism in the USA; post-industrial urbanism; the post-industrial/postmodern city. INTRODUCTION Three major transformations have altered the course of human life. The first was the revolution that led to the development of agriculture around 7000 BC and the growth of Neolithic farming settlements such as Jarmo in Iraq and Jericho in modern Israel. The second was the pre-industrial revolution that brought cities into being. The third was the industrial revolution of the eighteenth and nineteenth centuries that created the urban industrial forerunners of our present cities. This chapter begins by focusing on the preindustrial revolution to examine the origins and development of the earliest urban settlements, before going on to consider the character of industrial and post-industrial cities. PRECONDITIONS FOR URBAN GROWTH In Chapter 2 we saw how problematic the definition of a city can be. With reference to the pre-industrial city, Wheatley (1971 p. xviii)1 defined urbanism as ‘that particular set of functionally integrated institutions which were first devised some 5,000 years ago to mediate the transformation of relatively egalitarian, ascriptive, kin-structured groups into socially stratified politically organised, territorially based societies’. This emphasis on institutional change relates the growth of cities to a major socio-political restructuring of society, which he regards as a key element in the development of civilisation. In similar vein, Childe (1950)2 provides a listing of ten characteristics of an urban civilisation. These may be separated into five primary characteristics referring to fundamental changes in the organisation of society and five secondary features indicative of the presence of the primary factors (Box 3.1). Thus, for example, a community that was The origins and growth of cities 47 capable of building monumental public works probably had not only the craft specialists to undertake the task but also sufficient surplus to support the work. It is important to recognise that these advances in social organisation occurred in particular environmental settings. The combination of environmental and social forces underlying early urban development is encapsulated in Duncan’s (1961)3 ecological perspective, which emphasises how external (environmental) stimuli and internal (social) interrelationships operated together to promote the growth of cities in the pre-industrial era (Box 3.2). With these preliminary observations in mind, we can consider a number of different theories advanced to explain the origins of urban society. In doing so we should note that, as in much social science, it is not a question of whether one hypothesis is correct and the others wrong. Rather, each contributes some insight on the rise of cities. BOX 3.1 Childe’s ten characteristics of an urban civilisation Primary characteristics 1. Size and density of cities. The great enlargement of an organised population meant a much wider level of social integration. 2. Full-time specialisation of labour. Specialisation of production among workers was institutionalised, as were systems of distribution and exchange. 3. Concentration of surplus. There were social means for the collection and management of the surplus production of farmers and artisans. 4. Class-structured society. A privileged ruling class of religious, political and military functionaries organised and directed the society. 5. State organisation. There was a well-structured political organisation with membership based on residence. This replaced political identification based on kinship. Secondary characteristics 6. Monumental public works. There were collective enterprises in the form of temples, palaces, storehouses and irrigation systems. 7. Long-distance trade. Specialisation and exchange were expanded beyond the city in the development of trade. 8. Standardised, monumental artwork. Highly developed art forms gave expression to symbolic identification and aesthetic enjoyment. 9. Writing. The art of writing facilitated the processes of social organisation and management. 10. Arithmetic geometry and astronomy. Exact, predictive science and engineering were initiated. Urban geography 48 BOX 3.2 Pre-conditions for pre-industrial urban growth Population The presence of a population of certain size residing permanently in one place is a fundamental requirement. The environment, level of technology and social organisation all set limits on how large such a population would grow. Particularly important was the extent to which the agricultural base created a food surplus to sustain an urban population. The earliest cities were relatively small in modern terms, with few exceeding 25,000 inhabitants. Environment The key influence of the environment, including topography, climate, social conditions and natural resources on early urban growth is illustrated by the location of the earliest Middle Eastern cities on the Rivers Tigris and Euphrates, which provided a water supply, fish and fertile soils that could be cultivated with simple technology. Technology In addition to the development of agricultural skills, a major challenge for the early urban societies of the Middle East was to develop a technology for river management to exploit the benefits of water and minimise the risk of flooding. Social organisation The growth of population and trade demanded a more complex organisational structure including a political, economic and social infrastructure, a bureaucracy and leadership, accompanied by social stratification. Source: Adapted from O.Duncan (1961) From social system to ecosystem Sociological Inquiry 31, 140–9 THEORIES OF URBAN ORIGINS HYDRAULIC THEORY The importance of irrigation for urban development, especially in the semi-arid climates of the Middle East where the agricultural revolution took place, was identified by Wittfogel (1957),4 who argued that the need for large-scale water management required centralised co-ordination and direction, which in turn required concentrated settlement. The principal characteristics of a ‘hydraulic society’ are that it: 1. permits an intensification of agriculture; 2. involves a particular division of labour; 3. necessitates co-operation on a large scale. The origins and growth of cities 49 As Box 3.2 indicates, these are reflected in Duncan’s (1961) preconditions for urban growth. Agricultural intensification allows a concentration of population, while cooperation leads to a need for managers and bureaucrats. Those who control the water resources (whether a temple elite or secular state) exert power over others (e.g. farmers). The division of labour, centralisation of power and administrative structure all promote concentrated settlement, and hence the emergence of a town. As well as in the Middle East, evidence of a relationship between the adoption of irrigation and rapid population growth, nucleation, monument construction, intense social stratification and expansionist warfare has also been found at Teotihuacán in Mexico, which was the site of a preindustrial city of 70,000–125,000 inhabitants. There is now little doubt that irrigation was a key factor in the growth of pre-industrial cities in the ancient world. The problem lies in disentangling cause and effect. This is particularly difficult if one is seeking to support the belief that urbanisation followed upon the development of irrigation. A more likely scenario is that the institutions of centralised urban government and large-scale irrigation grew side by side. At first, smallscale irrigation schemes would have required a certain amount of administration, which would have expanded the irrigation system. This in turn would have required greater administration and so on, eventually leading to large-scale irrigation works and an urban political organisation with a monopoly of power. ECONOMIC THEORY Several theorists have suggested that the development of complex large-scale trading networks stimulated the growth of urban society5 Certainly, the fact that southern Mesopotamia did not have many raw materials such as metallic ores, timber, building stone or stone for tools made trade essential. This required an administrative organisation to control the procurement, production and distribution of goods. Such an organisation would have been a powerful agent in the community, and its power may well have extended beyond trade into other aspects of society. The need to increase production for trade purposes as well as to feed an expanding population would have led to continued specialisation and intensification, and the growing sedentary population would itself constitute a market for local produce and trade goods. Once again, however, it is unclear whether trade was a cause of city growth or the product of an already existing urban administrative elite. MILITARY THEORIES Some theorists suggest that the origin of cities lay in the need for people to gather together for protection against an external threat, the initial agglomeration leading to subsequent urban expansion. The excavation of a massive defensive wall built on bedrock would appear to indicate the defensive origins of Jericho, but not all early towns have such defences. Wheatley (1971)6 believed that warfare may have contributed to the intensification of urban development in some places by inducing a concentration of population for defensive purposes and by stimulating craft specialisation. Urban geography 50 RELIGIOUS THEORIES Religious theories focus on the importance of a well-developed power structure for the formation and perpetuation of urban places and, in particular, how power was appropriated into the hands of a religious elite who controlled the disposal of surplus produce provided as offerings.7 There is clear evidence of shrines and temples in ancient urban sites and there can be little doubt that religion played a significant part in the process of social transformation that created cities. However, it is unlikely to have been the sole factor. THEORETICAL CONSENSUS? It is doubtful if a single autonomous, causative factor will ever be identified in the nexus of social, economic and political transformations that resulted in the emergence of urban forms of living. A more realistic interpretation is generated if the concept of an ‘urban revolution’ is replaced by the idea of an urban transformation involving a host of factors operating over a long period of time. As Redman (1978 p. 229)8 explained: urbanisation was not a linear arrangement in which one factor caused a change in a second factor, which then caused a change in a third, and so on. Rather, the rise of civilisation should be conceptualised as a series of interacting incremental processes that were triggered by favourable ecological and cultural conditions and that continued to develop through mutually reinforcing interactions. This approach represents a significant departure from monocausal views on the origin of cities. Its value lies in exposing the key roles of social stratification and individual and group decision-making underlying the complex reality of the transformation from nomadic life to settled urban life in the ancient world. For urban geographers, this poses the particular question of where such developments occurred. EARLY URBAN HEARTHS As Figure 3.1 illustrates, there is evidence of early city growth in four areas of the Old World and one area of the New World: 1. Mesopotamia. The first cities are thought to have begun around 3500BC in lower Mesopotamia (Sumer) around the Tigris and Euphrates The origins and growth of cities 51 Figure 3.1 Early urban hearths rivers (Figure 3.2). One of the earliest cities was Ur, which from 2300 BC to 2180BC was the capital city of the Sumerian Empire, which extended north along the Fertile Crescent, possibly as far as the Mediterranean. In 1885BC Ur and the other southern cities were conquered by the Babylonians9 (Box 3.3). 2. Egypt. There is a long-standing debate in archaeology over theories of urban diffusion or independent invention but it is most probable that agricultural and other technologies, possibly including city-building, spread along the Fertile Crescent, then south-west into the Nile valley. By 3500BC a number of the Neolithic farm hamlets along the lower Nile had risen to ‘overgrown village’ status and were clustered into several politically independent units, each containing large co-operative irrigation projects. The transition from settled agricultural communities to cities occurred around 3300 BC when the lower Nile was unified under the first pharaoh, Menes. The early Egyptian cities were not as large or as densely settled as those of Mesopotamia because: ■ The early dynastic practice of changing the site of the capital, normally the largest settlement, with the ascendancy of a new pharaoh limited the growth opportunity of any single city. ■ The security provided by extensive desert on both sides of the Nile meant that once the valley was unified politically, Egyptian cities, unlike those of Mesopotamia, did not require elaborate fortifications and garrisoned troops for protection. Urban geography 52 Figure 3.2 The Fertile Crescent and ancient cities of the Middle East BOX 3.3 Ur of the Chaldees The city of 2000BC was surrounded by a wall 26ft (8m) in height which enclosed an area of 89 acres (36 ha) and a maximum population of 35,000. The walled city was an irregular oval shape, about three-quarters of a mile (1.2km) long by half a mile (0.8km) wide. It stood on a mound formed by the ruins of previous buildings, with the River Euphrates flowing along the western side and a navigable canal to the north and east. The harbours to the north and west provided protected anchorages (A and B in the plan) The temenos or religious precinct occupied much of the north western quarter of The origins and growth of cities 53 the city (C in the plan). The remainder of the city within the walls was densely built up as residential quarters (D in the plan), with two-storey houses of burnt brick below and mud brick above, with plaster and whitewash concealing the change in material. House size and ground plan varied according to the space available and the owner’s means, but generally rooms were arranged around a central paved courtyard that gave light and air to the house. A processional avenue (X–Y in the plan) possibly ran straight through to the temenos. Including those resident outside the walls, a figure of 250,000 has been estimated for the total population of the city-state. Source: A.Morris (1994) History of Urban Form Harlow: Longman Plate 3.1 The pyramid at Chichen Itza in Mexico functioned as a solar calendar for the Mayan civilisation that flourished in Meso-America and the Yucatan between 300 BC and AD 1300 3. The Indus valley. The Harappa civilisation appeared around 2500 BC in the Indus valley in what is now Pakistan. It was distinguished by twin capital cities, a northern one of Harappa in the Punjab and Mohenjo-daro, 350 miles down-river. The planned layout of each city was in marked contrast to the organic growth of Mesopotamian cities such as Ur. Both cities were laid out on a gridiron pattern with wide, straight streets forming 1,200 ft×800 ft (370 m×240 m) rectangular blocks. Socio-spatial segregation was common, with blocks or precincts occupied by a specific group such as potters, weavers, metalworkers and the elite. Each city covered approximately one square mile in area (640 acres/250 ha) and accommodated 20,000 people. Urban geography 54 The Harappa kingdom was ruled from the twin capitals by a single ‘priest-king’ who wielded absolute power. There is some evidence of trade with the Sumerian city-states by 2000 BC but the unchanging material culture and still undeciphered written language suggests that, in contrast to the cities of the Nile valley, the Harappa culture and cities emerged independently. Following a thousand years of stability, the Harappa civilisation was destroyed by invaders around 1500BC. 4. The Yellow River. The valley of the Huangho (Yellow River) was the birthplace of the Shang civilisation that arose around 1800BC. The most significant feature is that individual cities, such as An-Yang, were linked into a network of agricultural villages; a town wall did not separate an urban subculture from a rural one.10 This form of ‘urban region’ is without precedent in the early civilisations of Mesopotamia, the Nile and the Indus. 5. Mesoamerica. The earliest cities in the New World appeared around 200 BC—in southern Mexico (Yucatan), Guatemala, Belize and Honduras. Thus Mesoamerican peoples were entering a stage of development equivalent to the Neolithic of the Old World at a time when Mesopotamian cities had been in existence for 2,000 years. Of the several civilisations that evolved in Mesoamerica, the Mayan, which flourished between AD 300 and AD 1000, was the most culturally advanced. Cities such as Tikal, Vaxactum and Mayapán were centres of small states ruled by a leader drawn from a priest-hood and organised into a loose confederation. Society was highly stratified, with the elite occupying central city land around the palaces and temples, and the lower classes the urban periphery. Although the debate over urban diffusion or independent origin remains unresolved, identification of a number of common features among early urban societies encouraged Sjoberg (1960) to propose a model of the socio-spatial structure of the pre-industrial city11 (Box 3.4). THE SPREAD OF URBANISM By 800 BC cities such as Athens, Sparta and Megara had arisen on the Greek mainland. Greek cities subsequently spread to other parts of the Mediterranean and along the Black Sea coast. The Greek urban diaspora was a direct response to population pressure and the poor agricultural base available to the mainland cities. Individual cities equipped expeditions to establish new cities. A first wave beginning around 750 BC led to settlements on the coast of the Ionian Sea, in Sicily and in southern Italy (e.g. Ephesus, Syracuse and Naples), with a second wave spreading east to reach the Black Sea by 650BC. A significant feature of Greek urban colonisation was that in contrast to the organic layout of mainland cities like Athens, the new independent cities frequently adopted a gridiron plan, irrespective of the topography of a site. The expansion of the Roman Empire, following the defeat of Carthage in the Punic Wars (264–146 BC), carried the practice of city-building, and in particular the gridiron plan, into Western Europe. In Britain, prior to the Roman conquest in 55 BC, systematic town-planning was unknown. Figure 3.3 shows the typical form of a Roman imperial urban settlement. The town perimeter was usually square or rectangular. Within it there were two main cross-streets, the east-west decumanus through the centre of the town, and The origins and growth of cities 55 the north-south cardo, usually bisecting the decumanus towards one end. Secondary streets completed the grid layout to form the insulae or building blocks. The forum (equivalent to the Greek agora) was typically located in one of the angles formed by the intersection of the two Figure 3.3 Typical form of a Roman imperial settlement and a generalised example from Roman Britain Source: A.Morris (1994) History of Urban Form Harlow: Longman Urban geography 56 BOX 3.4 A model of the pre-industrial city Sjoberg’s model attempts to identify some common cross-cultural features of the preindustrial city. Like all models claiming a wide range of application, this model of a generic pre-industrial city has been criticised for oversimplification and overextension. Sjoberg’s model is most appropriate in cities where domination by an elite occurred. He identifies three social groups with social boundaries rigidly defined and often formally codified. The class pyramid was translated into a distinct spatial pattern. The elite occupied the central district around the ceremonial and symbolic institutions, including the religious, educational and political structures. Interspersed within the inner core were the servants of the elite, but the major concentration of lower-class residences was in a zone outside the core. The outcasts were relegated to the periphery, completing the spatial gradation of social status. Within each of the zones further differentiation occurred by occupation or ethnicity; however, there was limited segregation of distinct land-use functions. This generalisation has been questioned by Vance,12 who gives greater emphasis to the pattern of occupational subdistricts and downplays the extent of a zone of lower-class labourers, placing them instead as lodgers scattered within the various subdistricts. Notwithstanding differences of detail, both Sjoberg and Vance depict a walking-scale city with a general social degradation from core to periphery. Despite its simple structure, at a general level the Sjoberg model continues to serve as a useful standard against which individual cities may be assessed. For example, while Langton13 found limited correspondence between the model and the social geography of seventeenth-century Newcastle upon Tyne, Radford14 revealed a close relationship between the model’s description of pre-industrial urban society and the situation in the ante-bellum city of Charleston SC. Source: After G.Sjoberg (1960) The Pre-industrial City New York: Free Press The origins and growth of cities 57 principal streets and normally comprised a colonnaded courtyard with a meeting hall built across one end. The main temple, the theatre and the public baths were also located near the forum. Venta Silurum, a Romano-British town on the western military frontier in Wales, is a good example of Roman town planning15 (Box 3.5). Many of the great modern cities of Europe can trace their urban beginnings to Roman influence, including London, Brussels, Seville, Cologne, Paris, Vienna and Belgrade. With the fall of the Roman Empire in the fourth century AD, Muslim control of the Mediterranean trade routes during the seventh and eighth BOX 3.5 Plan of Venta Silurum (Caerwent) The town dates from AD75 and was founded to replace a tribal hill fort situated on high ground a mile to the north-west. The settlement covered an area of 44 acres (18 ha) and typically was divided in two by the main east-west street that was lined with shops and ran between the two main gates. Two other streets running east-west and four running north-south divided the town into twenty blocks. In a north-south direction there is not one dominant street line but two, each leading to a gate in the defences in a manner typical of a military settlement. centuries and increasing Viking raids from the north in the ninth century, much of Europe entered a Dark Age of economic and cultural stagnation. Under such unsettled conditions long-distance trade of any significance was impossible, towns became isolated and inward-looking, and urban life in Western Europe declined to its nadir by the end of the ninth century. Urban geography 58 URBAN REVIVAL IN WESTERN EUROPE During the tenth century, the reopening of long-distance trade routes and the presence of urban nuclei represented by fortified settlements and ecclesiastical centres stimulated the beginnings of a commercial revival. But the rebirth of urbanism was hindered by the social and economic constraints inherent in the dominant social system of feudalism. Feudalism was a form of social organisation (or mode of production) characterised by the interrelationship of two social groups. Direct producers (peasants working the land) were subject to politico-legal domination by social superiors (feudal lords) who formed a status hierarchy headed by the monarch. Social relations under feudalism are not defined primarily through market interactions (as under capitalism) but are legally defined, and the whole society was organised in terms of a formal and rigid hierarchy of power and dominance. A key feature of the manorial (estate-based) system of economic organisation prevalent in medieval feudalism was the focus on self-sufficiency and local markets. This restricted innovation and commercial expansion. In addition, since the amount of surplus value extracted by feudal lords in rents and taxes was dependent less on market forces than on their expenditure requirements (which increased as they competed for political status through conspicuous consumption), there was insufficient reinvestment in the productive capacity of agriculture and industry.16 The regrowth of towns in Europe coincided with the decline of the feudal system and was related to the increasing economic and political power of an emerging middle class of merBOX 3.6 The merchant class and urban growth in medieval Europe During the centuries of the Dark Ages an economy based on exchange was replaced by an economy of consumption. But a new merchant class was appearing on the edge of feudal society. Probably originating among landless men, escaped serfs, casual harvest labourers, beggars and outlaws, the bold and resourceful among them—the fast talkers, quick with languages, ready to fight or cheat—became chapmen or pedlars, carrying their wares to remote hamlets. They were paid in pennies and farthings and in portable local products, such as beeswax, rabbit fur, goose quills and the sheepskins for making parchment. If they prospered, they could settle in a centre and hire others to tramp the forest paths. The merchants made the towns. They needed walls and wall builders, warehouses and guards, artisans to manufacture their trade goods, cask makers, cart builders, smiths, shipwrights and sailors, soldiers and muleteers. They needed farmers and herdsmen outside the walls to feed them; and bakers, brewers and butchers within. They bought the privilege of self-government, substituting a money economy for one based on land, and thus were likely to oppose the local lord and become supporters of his distant superior, the king. Towns recruited manpower by offering freedom to any serf who would live within their walls for a year and a day. Source: M.Bishop (1971) The Penguin Book of the Middle Ages Harmondsworth: Penguin The origins and growth of cities 59 chants opposed to the economic regulation of their activities by feudal lords (Box 3.6). The urban bourgeoisie of merchants, manufacturers and financiers, often organised into guilds, obtained certain freedoms, notably the right to supervise their own markets. Successful towns obtained charters from the king and, in England, were known as boroughs. In the more urbanised parts of Europe many towns such as Milan, Florence, Cologne and Bruges were able to become self-governing municipalities (communes) capable of independent political action. In northern Italy a fiercely competitive society of rival city-states had emerged by the thirteenth century (providing a foundation for the Italian Renaissance). The economic and political freedoms obtained by the emerging class of burgesses (free merchants and craftsmen) was of central importance in the rise of European urban society and the corresponding decline of medieval feudalism. As the power of the urban bourgeoisie increased, the goal of the economy became expansion and economic profit the function of city growth. THE MEDIEVAL TOWN The power of commerce and trade in medieval urbanism is exemplified in the German city of Lübeck, founded in 1159 by the Duke of Saxony in order to attract merchants from neighbouring lands. Situated at the western edge of the Baltic Sea, Lübeck developed into the nucleus of a German Baltic trading system organised by a confederation of towns known as the Hanseatic League that included Hamburg, Lüneberg, Cologne, Danzig and Riga. Development of regular trade with the Russian city of Novgorod brought furs and forest products, which, along with local salt for preserving fish and meat, were distributed throughout northern Europe in exchange for cloth and other manufactures. The wealth and power acquired by the merchant elite enabled them to govern Lübeck and the other Hanse cities. Within Lübeck the central importance of trade and commerce was reflected in the urban structure (Figure 3.4). The town had a medieval cathedral and monasteries, but it was dominated by the central marketplace. This consisted of the shops of a range of specialist merchants and craftsmen concentrated together in their own areas (e.g. a street of bakers, spice merchants or tailors), with living quarters located above the shop. A mint, the new urban engine of the medieval money economy, was located close to the marketplace. Despite the central importance of commerce, the medieval town did not sever links with the surrounding area but enjoyed a symbiotic relationship with the countryside. In particular, medieval towns, with their poor hygiene and higher mortality, grew only because of the influx of rural immigrants seeking employment and economic opportunity. Many cities, such as Venice and Nuremberg, annexed large areas of their surrounding lands to guarantee food supplies (a rural-urban symbiosis that remains common in modern China). Urban geography 60 Figure 3.4 The urban structure of medieval Lübeck In addition to the commercial function, the main characteristic of the European medieval city that distinguished it as a separate entity was its legal status, which afforded citizens privileges denied to rural dwellers. The dictum that ‘town air makes free’ was a powerful attraction for the rural peasantry and a major force for urban growth. The same privileges did not extend to medieval towns elsewhere. The city of Novgorod, despite its prosperity, was not politically autonomous. Neither were urban freedoms possessed by other important non-European medieval towns such as Baghdad, Delhi or Beijing. Only in the medieval West did cities acquire this special legal status.17 More generally, the new force of capitalism pushed aside the vestiges of feudalism, ushered in the industrial revolution, and paved the way for the emergence of the industrial city. THE PRECONDITIONS FOR INDUSTRIAL URBANISM The complex series of innovations commonly referred to as the industrial revolution emerged in Britain from the mid-eighteenth century onwards, spreading from there to the European mainland and ultimately to other parts of the world during the nineteenth and twentieth centuries.18 The industrial revolution is typically thought of in terms of technology the invention of complex machines and use of inanimate energy sources that The origins and growth of cities 61 greatly increased worker productivity—but important cultural, social and population changes were also involved. It is important to understand these and their effects on the form of the city. For Weber (1958),19 a fundamental change in Western cultural values was a prerequisite for the industrial revolution. Central to this was the concept of profit. In medieval Europe, profit was an alien concept among artisans: the price of a product was fixed by adding to the cost of materials the fair value of one’s labour. In the absence of profit (defined as the excess of the selling price of goods over their cost) there could be no large capital accumulation to invest in the development of industrial society. This change in social outlook was stimulated by the Protestant Reformation, which fostered a new set of values that stressed rationality in interpersonal relations, including trade, hard work and the right to the material rewards of one’s labour. This ‘work ethic’, which originated in northern Europe in the sixteenth century, spread throughout Western society and in Weber’s view represented a prime precondition for the industrial revolution, the development of capitalist society and the emergence of the industrial city, the Plate 3.2 The oval-shaped Renaissance-period Piazza San Pietro in Rome, constructed by Bernini in 1656 to symbolise the allembracing power of the Church, is a potent expression of the link between architecture and ideology foundations of which were laid in the early modern period. Urban geography 62 EARLY MODERN URBANISM In the early modern (post-medieval, pre-industrial) period, from 1500 to 1800, the economic and political powers of the European medieval city were usurped by the expansion of nation-states. Despite this relative decline in urban autonomy, however, the period witnessed increasing levels of urbanisation and the gradual emergence of an urban system in Europe.20 During the early modern period urban development in Europe was characterised by cycles of growth and decline. Three main phases may be identified: 1. The ‘long sixteenth century’, from 1500 to 1650, was a period of general prosperity during which urban growth was widely distributed. 2. From 1650 to 1750 population growth slowed in response to war, plague and famines combined with cyclical downturn in the economy as a result of rising labour costs, falling rent incomes from property and technological stagnation. Urban growth occurred differentially. Cities that were centres of state government (such as Madrid) and port cities engaged in Atlantic trade (such as Amsterdam) experienced growth. By contrast the majority of inland trading centres, ecclesiastical seats and industrial towns suffered at least a relative loss of standing. A net result was a more hierarchical size distribution of urban places. 3. The period of ‘new urbanisation’ between 1750 and 1800 witnessed the growth of smaller cities and the addition of new cities to the urban system. This process was related to the political economy of proto-industrial production that formed as urban and rural-based precursors of factory-based manufacturing industry.21 Widespread urban growth was stimulated by increases in agricultural production and incomes that promoted the growth of regional marketing, administrative and commercial centres. Many small cities with resource-based industries such as metallurgy grew as a consequence of technological innovation, and many rural places that had developed with proto-industrial textile production emerged to become industrial towns as technological change encouraged factory organisation. By the end of the early modern period the twin processes of proto-industrialisation and urbanisation had paved the way for the emergence of industrial urbanism. THE FORM OF THE INDUSTRIAL CITY The nineteenth century witnessed the flowering of industrial capitalism in Western Europe and the rapid growth of the European industrial city. A major driving force was the factory system with its economies of scale, increased productivity and higher levels of output. The need for a large pool of labour as well as ancillary services and markets for their products encouraged factories to cluster together in towns. Successful towns attracted further economic activity that drew in more migrants in search of work in a cumulative process of growth (Box 3.7). Town populations grew apace. By 1800 London was the largest city in the world, with a population of over 900,000. The population of Birmingham increased by 273 per cent between 1801 and 1851 from 71,000 to 265,000, Manchester grew from 75,000 in 1801 to 338,000 in 1851 (a growth of 351 per cent) and Glasgow from 84,000 to 350,000 over the same period (an increase of 317 per cent). Lawton (1972)22 estimated that almost all the 27 million increase in the British The origins and growth of cities 63 population between 1801 and 1911 was absorbed by urban areas. In many of the great nineteenth-century industrial cities, wealth found civic expression in monumental public buildings and cultural institutions. Industrial capitalism also brought a major realignment of social structures with the creation of two main classes: 1. capitalists, who invested in labour with the goal of realising a profit; 2. labour, that sold its skills to the owner of capital in return for a wage. The unequal division of power between the two main classes had a direct effect on the form of the industrial city, which was developed primarily to fulfil the needs of capital. This was particularly evident in the socio-spatial segregation of the classes BOX 3.7 The circular and cumulative model of urban growth The urban economy may be viewed as being made up of two interrelated sectors: 1. the basic sector, comprising activities that produce goods and services sold outside the city to obtain the means needed to purchase imports (e.g. manufactures); 2. the non-basic sector, comprising activities that provide goods and services for the city itself (e.g. retailing). The two sectors are functionally interdependent. Thus, if the basic sector expands, workers in that sector will spend more on services available in the city, creating growth in the non-basic sector. The extent of growth will depend on the respective sizes of the two sectors. In a city with a basic: non-basic ratio of 1:4 an increase of 1000 in basic employment will lead to an increase of 5,000 (1,000+ 4,000) in total employment. This is referred to as the urban economic multiplier and is a key concept in the economic explanation of urban growth In practice given the variety of economic activity in any Urban geography 64 city, the mechanisms of growth are highly complex. These are best seen as a circular and cumulative process whereby growth in one sector triggers expansion through secondary multiplier effects elsewhere in the economy. Consider a situation in which the output of a company manufacturing computer chips expands in response to increased demand. Company profits and employees’ wages will rise and expenditure by shareholders and workers will increase, leading to growth of tertiary activities (such as existing and new shopping and retail activities). The company may decide to move to larger premises, which generates growth in the construction industry. It may also take on additional workers, leading to in-migration, and new housebuilding and urban growth. These changes in the non-basic sector can act as a secondary multiplier effect, lifting the city to a new threshold comprising a larger market that promotes newer forms of manufacturing and service activities. The company research and development team may invent an even-faster computer chip, expanding demand for the company’s product and leading to a further round of growth. The interaction between the basic and non-basic sectors of the urban economy and combination of multiplier effects promotes urban growth. Conversely, the opposite situation may also occur, leading to economic stagnation and urban decline. Figure 3.5 Burgess’s model of the industrial city (Figure 3.5). The burgeoning populations of the nineteenth-century industrial cities placed an enormous strain on urban services and infrastructure. Public sanitation and water supplies were inadequate and often non-existent in the slums. Mid-century London had 200,000 undrained cesspools, and the River Thames was virtually an open sewer. Similar conditions existed in other large industrial cities and it is no surprise that cholera and typhoid were rife, especially, though not exclusively, among the poor. A cholera epidemic in 1832 claimed the lives of 5,000 Londoners. In the same year cholera killed The origins and growth of cities 65 2,800 in Glasgow. The wealthy, with their power to exercise choice, sought to maintain a healthy separation from the masses. RESIDENTIAL SEGREGATION Several processes underlay the pattern of residential segregation in the Victorian city. First, in terms of the physical growth of the city and the evolution of housing areas, much of the house building was undertaken by speculative builders, who were rarely interested in building for the lower end of the market. As a result, the demand for cheap housing remained high, especially following the influx of migrants from the Irish famine in the mid-nineteenth century. This demand was met by multiple occupancy, leading to overcrowded conditions in the central slum areas of most cities. The development of new middle- and high-status housing on the urban periphery by speculative builders allowed vacated inner-city properties to filter down to poorer families. Second, the development of residential segregation was a result of individual locational decisions within the context of a rapidly expanding urban population. Few urban dwellers remained in the same house for long in the nineteenth century, and both migrants and established residents would have been continually re-evaluating the suitability of their residential location. The main economic constraints on residential relocation were those of disposable income, the availability of employment in different areas, and access to accommodation in different sectors of the housing market. Social, demographic and cultural characteristics, as well as knowledge of the city, also influenced residential location. Third, the process of residential differentiation was also influenced by the development of commercial and industrial areas within the city that imposed constraints on the nature of residential development. These constraints were exerted through pressure on land in certain districts, such as the invasion of inner-city housing areas by commercial development; through the provision of employment in particular areas with associated working-class housing close by; and through industrial pollution that prompted those who could do so to move away. Fourth, although land-use planning did not exist in the Victorian city and institutional forces were weaker than they are today, national and local governments affected residential development in several ways. Local by-laws, together with national health and housing legislation, imposed some control over new housing development, while town improvement schemes aimed at better water supply, sewage disposal, the closure of cellar dwellings and the demolition of slum properties affected existing housing. The rate at which new houses were built, and their tenure and form, were also constrained by private-sector factors such as the problem of landownership and the prevailing rate of return on investment. Finally, though based on objective standards, residential differentiation was perpetuated through popular images of different areas. Certain parts of the city could gain a reputation as being, say, of high status, unhealthy or predominantly Irish.23 The operation and outcomes of the processes of residential segregation in the Victorian city are apparent in Liverpool where, as in many nineteenth-century industrial cities, immigration was a major factor underlying urban change (Box 3.8). More generally, Pooley (1979)24 provides a model that summarises the processes of residential Urban geography 66 sorting by social class in the nineteenth-century British city (Figure 3.6). Intra-urban residential mobility is characterised by three main features—distance, direction and frequency—which, in Figure 3.6 The processes of residential sorting by class in the nineteenth-century British city, s.e.g. is socio-economic group Source: C.Pooley (1979) Residential mobility in the British city Transactions of the Institute of British Geographers 4, 258–77 BOX 3.8 Socio-spatial segregation in nineteenth-century Liverpool By 1871 socio-spatial segregation was a feature of the urban landscape of Liverpool. It was based on three principal factors: (1) socio-economic class; (2) ethnicity (primarily Irish/non-Irish); and (3) stage in lifestyle. Though the result is a complex mix of sectors, zones and nuclei, a series of distinct elements can be identified: The origins and growth of cities 67 1. The older central residential districts, approximately within the 1851 built-up area, had developed by 1871 into a number of distinct sectors largely constrained by the location of the docks, principal routes radiating from the centre, and surviving islands of highstatus property. 2. Outside the 1851 built-up area, broadly zonal characteristics are present: population from the central area had moved out into the newer suburbs, within which there were several areas of a transitional nature. Complementary to the mainly outward direction of ‘internal’ population movement, the central zones were largely fed by in-migrants, some of whom were to stay permanently in low-status areas, others of whom soon moved out to the newer suburbs. There was also considerable movement between sectors of low-status housing, often caused by displacement due to urban redevelopment, though non-Irish families generally avoided the Irish areas. 3. The high-status population moved sectorally into the third peripheral zone, while within the suburban ring, lateral movement took socially upward-moving families into the high-status sector to the north. Source: Adapted from R.Lawton and C.Pooley (1976) The Social Geography of Merseyside in the Nineteenth Century: Final Report to the Social Science Research Council London: SSRC combination, produce different patterns of movement within the city. These three main features are, in turn, influenced by the personal characteristics of the household (e.g. socio-economic group, or stage in the life cycle) and by general features of city structure (such as the extent to which distinct social and housing areas have developed). The outcome of these factors is seen in the production of social areas characterised by Urban geography 68 different levels of residential mobility and community stability. The model identifies four types of area, together with their typical locations in the Victorian city: 1. Type A. An area that maintains long-term stability owing to low rates of mobility and the replacement of population loss by in-migrants of a similar character. This type occurred most frequently within high-status areas of the city. 2. Type B. An area that attains short-term stability through frequent short-distant mobility of a circulatory nature, but which over a longer period succumbs to a general centrifugal flow of population. This was the dominant form of mobility in the Victorian city. 3. Type C. An area where there are high rates of longer-distance mobility, as from central and industrial areas under pressure for further commercial use. Rapid community disintegration may take place, and such areas are essentially unstable in character, with a largely transient population often in the process of moving towards the urban periphery. Community disintegration may also occur in high-status areas that are under pressure from a lower-status population. As the high-status migrants leave, the area may become unstable, or alternatively develop as a new, lower-status community 4. Type D. An area of rapid in-migration on the urban periphery receiving large numbers of in-migrants from outside the city as well as from intra-urban population flows. Such areas are characterised by strong community formation and long-term stability. Although the precise form of mobility patterns is dependent on local urban context, the model provides a useful summary of the major causal forces, processes and outcomes of residential mobility in the Victorian city. It is clear that in general the nineteenth-century city formed a dynamic system in which frequent individual residential moves were an integral part of a ‘socio-spatial sorting mechanism’ that influenced the nature of the urban residential mosaic by reinforcing some communities and destroying others.25 HOUSING THE POOR It is important to recognise differences between England, Scotland, North America and continental Europe with regard to housing systems and consequent social patterns in industrial cities. In England, most working-class families rented accommodation from private landlords on weekly terms. Their houses were most likely to be all or part of twoor three-storey terraced houses, whether ‘through houses’ (with a back door and garden), ‘back-to-back’ or built around courtyards, perhaps sharing water supply and toilet facilities. In Scotland and continental Europe the working classes were more likely to live in tenements, managed by a ‘factor’ on behalf of an absentee landlord and let on yearly terms. In North America, although most poor families rented from private landlords, rates of home-ownership were higher, and there was little involvement of philanthropic bodies and less state intervention in housing provision.26 In the industrial towns of nineteenth-century Britain, neither town councils nor employers saw it as their responsibility to build housing for the immigrants. This task was generally left to speculative builders whose modus operandi was to maximise the use of land and The origins and growth of cities 69 minimise expenditure on materials and services. This architectural parsimony reached an apogee in the terraces of back-to-back Plate 3.3 Serried rows of nineteenthcentury red-brick working-class terraced housing run down the hillside to the now disused shipyards on the River Tyne in the Elswick district of Newcastle upon Tyne BOX 3.9 The slum district of Little Ireland in Manchester, 1849 The district of Little Ireland was a typical living environment for 4,000 residents at the bottom of the social order. Engels described how the ‘coal-black, stagnant, stinking river’ Medlock loops through the area. The rows of back-to-back housing held an average of twenty persons per house, with its two rooms, attic and wet cellar. There was a single privy for six houses. Engels (1958 p. 71) presents a graphic description of the area: ‘the cottages are very small, old and dirty, while the streets are uneven, partly unpaved, not properly drained and full of ruts. Heaps of refuse, offal and sickening filth are everywhere interspersed with pools of stagnant liquid. The atmosphere is polluted by the stench and is darkened by the thick smoke of a dozen factory chimneys. A horde of ragged women and children swarm about the streets and they are just as dirty as the pigs which wallow happily on the heaps of garbage and in the pools of filth.’27 Source: Adapted from T.Freeman (1957) Pre-Famine Ireland Manchester: Manchester University Press Urban geography 70 Figure 3.7 Layout of back-to-back housing in Manchester Source: E.Morris (1997) British Town Planning and Urban Design Harlow: Longman housing constructed from the 1830s onwards. These dwellings were built in double rows under a single roof, one side facing the street and the other a parallel back street or alley or a courtyard accessible from the street at intervals through narrow passages or ‘entries’ (Figure 3.7). At the ends of the street or in the corner of a courtyard a standpipe for water supply and a communal earth closet were provided for the use of about twenty houses, or upwards of 150 people. The density of such developments averaged sixty houses or 200 rooms per acre (150 houses or 500 rooms per hectare), and the overcrowding, poor sanitary arrangements, lack of sunlight and absence of through-ventilation in the houses made back-to-backs the despair of medical officers (Box 3.9). As late as 1913 an inquiry into housing conditions in part of Birmingham found that: some 200,000 people were housed in 43,366 dwellings of the back-toback type already long condemned as injurious to health because of lack of ventilation. For the most part they contained only three rooms, and so were overcrowded. In the six worst wards, from 51 per cent to 76 per cent were back-to-backs. Even more serious was the fact that 43,020 houses had no separate water supply, no sinks, no drains, and 58,028 no separate w.c., the closets being com-munal and exposed in courts. This meant that over a quarter of a million people lived in cav-ernous conditions. The real objection to back-toback houses lies not so much in their method of construction as in the degrading and disgusting conditions of their The origins and growth of cities 71 outbuildings, which fre-quently made decency impossible and inevitably tended to undermine the health and morals of the tenants. (Bournville Village Trust 1941 p. 16)28 Even higher residential densities were created by the tenement form of building in working-class areas of nineteenth-century Glasgow (Box 3.10). THE OTHER SIDE OF THE COIN In marked contrast to the slum areas of the nineteenth-century industrial city were the status areas occupied by the elite. Many of Britain’s indus-trial cities were characterised by marked sociospatial differentiation between the upper-class west end and lower-class east end. The genesis of this pattern is illustrated by the westward migration of the upper middle classes in Glasgow which began in the mid-eighteenth century with the movement of the ‘tobacco lords’ away from the old city centre, and culminated in the classbased residential segregation of the Victorian city. The process was driven by two sets of factors. Major push factors were: 1. commercial encroachment into formerly highstatus residential areas adjacent to the central business district; 2. the deteriorating social and physical fabric of the old city centre. The dangers of crime and vice from slum areas were compounded by the threat of cholera, which, though originating in the slums, was no respecter of class. In addition to the push factors, the social elite of nineteenth-century Glasgow had a clearly defined set of criteria against which to judge a potential suburb. Most important was the social composition of the neighbourhood, which had to be exclusively upper class. In Glasgow’s emerging west end this was ensured by: BOX 3.10 Living conditions in the tenement slums of mid-nineteenth-century Glasgow The westward migration of the wealthy from the old town not only introduced sociospatial polarisation into what had been in the eighteenth century a heterogeneous urban structure, but freed land and housing for other uses. Those parts of the old city abandoned by the elite were colonised by a working-class population that was burgeoning in response to the new industries’ demand for labour. Residences previously occupied by a single wealthy family were ‘made down’ to accommodate large numbers of the poor in grossly crowded conditions, characterised most graphically by the wynds and vennels around Glasgow Cross. The end result of decades of making down in such places was addresses such as ‘Bridgegate, No. 29, back land, stair first left, three up, right lobby, door facing’. In 1861 two-thirds of Glasgow’s population of 394,864 lived in houses with only one or two rooms and of these 60 per cent shared a room with at least four others Urban geography 72 Overcrowding was intensified by property speculators building on any available open space to produce ‘backjams’. and ‘backlands’ tenements, which were either added to existing buildings or erected in the erstwhile gardens of the formerly wealthy burgher residences. By the mid-nineteenth century the old city was a congeries of poor-quality housing. One street in a dark ravine of a close housed a husband, wife and two children in a ‘sort of hole in the wall’ measuring six shoe-lengths in breadth, between eight and nine in length from the bed to the fireplace, and of a height that made it difficult to stand upright. Densities of 1,000 persons per acre were commonplace. A notorious example was the rookery in the Drygate, where in a single close 500 people claimed a dwelling. Source: M.Pacione (1995) Glasgow: The Socio-Spatial Development of the City Chichester: Wiley 1. an informal land-use zoning consensus among owners to reserve the area for highquality housing despite the presence of exploitable deposits of coal, iron, brick clay, building sand and freestone; 2. restricted clauses in feu charters, leaseholds and freeholds of houses which laid down what activities were permitted, as well as details of the design of houses, provision of footpaths, type of gas lamps and sewers, and even the nature of the shrubbery. Later feu contracts also often specified a minimum market price for a given house—the whole providing an early example of ‘exclusionary zoning’. The middle classes were also attracted by the picturesque ‘healthful and well aired’ environment of the west end with its rolling drumlin topography, prevailing westerly winds, and location above and to the west of the industrial areas and insalubrious old town. A further attraction was the proximity of the west end to the central business district, where the majority of Glasgow’s leading citizens were engaged.29 As well as spatial or territorial social segregation, in cities such as Glasgow, where ‘high-rise’ tenement building was the norm, vertical segregation was also prevalent. Within the tenement the principal apartments were always on the first floor, away from the noise, smell and dirt of the street but not too far to climb (Box 3.11). The industrial city first developed in Britain, the cradle of the industrial revolution, but soon spread to Europe and North America. THE ORIGINS OF URBAN USA Despite the existence of a Neolithic urban hearth in Mesoamerica, there were no preColumbian settlements in the territory of the future USA that could be considered of urban status. The urban origins of the USA are related unequivocally to the process of colonisation by Spanish, French and especially English settlers. The earliest urban settlements were established during Spanish occupation of the South West in what are now the states of Texas, Arizona, New Mexico and California. This process also extended north from the Caribbean into Florida. Reps (1965)30 has identified St Augustine in Florida, founded in 1565, as the first American city. As in other early Spanish settlements, this combined the functions of (1) military post The origins and growth of cities 73 (presidio), (2) base for trade and farming (pueblo), and (3) religious centre (mission). On the west coast, in California, the Spanish mission settlements formed the pre-urban nuclei for many of the great modern cities from San Diego to San Francisco de Solano.31 Nevertheless, despite the early beginnings, by the mid-nineteenth century there were no urban settlements in the South-West of comparable size to those of the East and rapidly emerging Midwest. English colonisation of North America began in 1585 with several unsuccessful attempts by Crown-appointed adventurers to establish a settlement on Roanoke Island in North Carolina. During the seventeenth century the colonial effort passed to corporate enterprises in the shape of royal chartered joint-stock companies. In 1607 the Virginia Company of London founded Jamestown, which became a centre for the tobacco plantations and which by 1619 had attained a population of around 1,000. Reps (1965)32 has divided these early British settlements into four groups: 1. the earliest tidewater colonies of Virginia and Maryland, including Jamestown and Baltimore; 2. the new towns of New England, the largest of which was Boston; 3. the towns of the middle colonies, dominated by New York and Philadelphia (Box 3.12); 4. the colonial towns of Carolina and Georgia such as Charlestown and Savannah. Also in the seventeenth century the Dutch established Fort Orange (Albany) in 1624 and New Amsterdam (New York) two years later as their main trading posts. Farther north the French were active, founding Quebec in 1608 and Montreal in 1620, giving them effective control of the St Lawrence river valley. By the end of the seventeenth century, however, English colonisation was dominant, although settlement was still largely confined to a 50-mile (80 km) fringe of country inland from navigable waterways. In 1800, of a US population of 4 million, only some 170,000 were settled west of the Allegheny Mountains. The nineteenth century saw a shift in the population centre of gravity as the urban frontier moved westward (Box 3.13). THE WESTWARD PROGRESS OF URBANISM A major obstacle to westward settlement was the difficult transportation links back to the main markets of the eastern seaboard. The importance of developments in transport technology for the expansion of urbanism is illustrated by the impact BOX 3.11 Tenement life in Glasgow The tenements of Glasgow have provided accommodation for millions of citizens over the past 150 years. The quality of this form of housing varied across the city, and was signalled by the character of the front entrance. In working-class tenements the entrance to the close (the passage leading to the stairs) was a simple opening in the wall. In betterclass tenements the close entrance might be decorated to match the windows on the floor Urban geography 74 above, or might be adorned by a porch reached by a flight of steps from street level. The close was the communal part of the tenement, and in working-class communities was a place of social interaction where neighbours met as they entered or departed from the building or took their turn to sweep the close or wash the stairs. At the rear of the tenement was the back court. This enclosed space, or drying area, was the common property of everyone living in the tenement. It sometimes contained a communal wash-house with a cast iron boiler heated by a coal fire, which was available to residents on a rota basis. Strict adherence to the rota was required for a peaceful existence! One could not change one’s wash day to suit circumstances, and the whereabouts of the indispensable item, the washroom key, was a frequent source of friction. The back court also accommodated the ash pits, which replaced the dung heap of earlier generations. Space in the back court was also taken up by the brick w.c. stack, affixed to the back wall of the tenement, and housing a single toilet for communal use on each half-landing. These toilet stacks replaced the dry closet or privy in the back court which served the whole tenement until the Police Act of 1892 made internal sanitation compulsory. As far as the internal structure of the building is concerned, the prime characteristic of the tenement as a building form was its flexibility. An ingenious architect could provide a reasonable variety of different house sizes and plans within the same solid rectangular framework. The four-storey Victorian tenement could contain five large houses of six or more rooms or, on a building plot of the same size, to meet totally different requirements, twenty-five single-apartment houses. The diagram illustrates the layout of a west-end tenement at the higher end of the market (with two flats of four or five main rooms on the upper floor). At the other end of the social scale were one-room and two-room apartments, commonly known as ‘a single end’ and a ‘roorn and kitchen’ respectively. In the diagram a ‘single end’ is grouped between two two-room flats, and this 2:1:2 formation was probably the most common tenement type in Glasgow In the ‘room and kitchen’ arrangement the room or The origins and growth of cities 75 parlour faced the street, as in the better-class tenements. A bed closet was provided on the wall opposite the window, the built-in bed hidden from view behind a door until the practice was forbidden by the Glasgow Buildings Regulation Act of 1900. The parlour would be furnished to the extent of the occupants’ means, and, perhaps surprisingly in view of the cramped size of the flat, was often kept for best and used only ‘for visitors’ or for special occasions such as a wedding or funeral. The family lived and slept in the kitchen at the rear. In the kitchen the open bed recess contained a built-in bed about 2 ft 6 in. (75 cm) above the floor, below which could be stored a smaller bed that could be wheeled out at night for the children. Toilet facilities were provided by a communal w.c. on the half-landing of the stair. It is salutary to note that at the outbreak of the First World War the great majority of Glasgow families lived in one- or two-room flats, and that as late as 1951 this form of accommodation made up half the city’s housing stock. Source: M.Pacione (1995) Glasgow: The Socio-Spatial Development of the City Chichester: Wiley BOX 3.12 Philadelphia: a seventeenth-century planned town In 1681 William Penn (1644–1718) was granted a charter from Charles II establishing him as governor and proprietor of Pennsylvania (in return for cancellation of a £16,000 royal debt owed to Penn’s father). Penn appointed three commissioners to lead the first group of settlers and provided them with a detailed brief for the location and planning of a new town. The gridiron plan served the immediate purpose of quickly dividing up the urban land into equal-size plots to be allocated to settlers by lottery. The layout also reflected Penn’s experience of the Great Plague and the Great Fire of London, and his determination to create a healthy and safe urban environment. Two main avenues 100ft (30m) in width crossed in a central square of 10 acres (4 ha), with minor squares of 8 acres (3 ha) in each quarter of the city. The central square would accommodate a number of public buildings. Penn also instructed that space should be allocated to allow every house to be located in the centre of the plot if the owner desired thereby providing recreation space and Urban geography 76 reducing the fire hazard. After three years Philadelphia had 600 houses, which increased to 2,000 by 1700. Although Penn had envisaged a dispersed pattern of growth between the two rivers, the dominance of the Delaware as a trade route meant that the city expanded outwards from the port area. Not until well into the nineteenth century did the city reach the area of the central square designated in the master plan, a new city hall being completed there in 1890. Subsequently, in the twentieth century, the migration of the central business district (downtown) away from the waterfront led to the gradual decay of the old port area and earliest residential districts, providing opportunities for redevelopment. of the construction of the Erie Canal between the Hudson River and Lake Erie. When the canal reached the obscure village of Buffalo in 1824 the decimation of freight costs over the 363 miles to the Hudson River led to the construction of 3,400 houses in the following year, with the town reaching a population of 18,000 by 1840. Other settlements on the new waterway such as Rochester, Utica, Syracuse and, principally, New York all grew rapidly into prosperous cities (Box 3.14). Borchert (1967)33 has proposed a model of US urban development related to changes in technology. STAGE 1: THE SAILING VESSEL AND HORSE-DRAWN WAGON ERA TO 1830 Throughout the eighteenth and early nineteenth centuries the North American economy was based almost entirely on agriculture and the export of staple products. The main function of the few cities that existed was mercantile, involving the import of manufactured goods from Europe and export of primary produce. The largest cities were on the Atlantic seaboard, with New York, Philadelphia, Boston and Baltimore dominating the trade of the original colonies. The next largest cities were located on the riverine routes into the interior at New Orleans, Quebec and Montreal. Most of the cities beyond the Appalachians were small service centres for surrounding agricultural settlements, the largest in 1830 being Cincinatti with 24,800 inhabitants. STAGE 2: THE AGE OF STEAM AND THE IRON RAIL, 1830–70 The proportion of the US population living in urban areas rose from 8 per cent to 23 per cent over the period 1830–70. In addition to the continued growth of the cities of the eastern seaboard, which had developed as manufacturing bases and key centres for the commercial exploitation of the developing continent, the most dramatic trend was the spread of urbanism in the Midwest associated with the construction of canals and railways. The rail network expanded from 9,000 miles of track in 1850 to 53,000 miles by 1870 and, as well as linking all the east-coast cities, had reached Oakland on the west coast. Inland cities in accessible locations boomed. Chicago on Lake Michigan, with waterway links to the east and the Great Lakes and Erie Canal, grew into a city of 300,000 people by 1870. The only large cities beyond the east coast and the Midwest were New Orleans (population 191,000), with a rich agricultural hinterland, and San The origins and growth of cities 77 Francisco (population 150,000), which developed rapidly following the gold rush of 1849. BOX 3.13 The township and range system of land division The westward expansion of settlement in the USA was based on a national land policy enshrined in the Land Ordinance of 1785. The law required that the territory to be sold should be laid out in regular sections or townships 6 miles square. Each township was divided into thirty-six square sections of one square mile or 640 acres (260ha). Half the land was to be sold as townships, the other half as sections. Section 16 in each township was reserved for the support of schools. The first seven ranges of townships were surveyed by 1786 immediately west of the Ohio river (numbered 1–7 from east to west). An example township in the seventh range is shown divided into its thirty-six sections. Each section could also be divided into quarter-sections. Section boundaries often provided lines for roads, and villages and towns grew up at junctions or were laid out and promoted by property speculators. STAGE 3: THE AGE OF STEAM AND STEEL, 1870–1920 Over the period 1870–1920, the urban population of the USA experienced a fivefold increase to 41 million, and urban development was evident across the continent. On the east coast the New York metropolitan area had quadrupled in population to 5.3 million by 1920, while Philadelphia tripled to reach 1.8 million. In the Midwest, Chicago moved from being the fifth-largest city in North America to being the second largest and experienced a ninefold increase in population to 2.7 million. On the west coast a growing number of cities had populations in excess of 100,000, including Los Angeles. By contrast, in the South, urban development was still at an incipient stage. The continental urbanisation wave was driven by several forces, including: 1. the increasing integration of the North Ameri- Urban geography 78 BOX 3.14 New York: the urban explosion In contrast to Philadelphia, New York began without a master plan, and only after 150 years of organic growth was the first gridiron laid out. When the British captured the city in 1664 the population was around 1,500. Following the opening of the Erie Canal, the city’s population mushroomed from 150,000 in 1820 to 300,000 in 1840, 515,000 in 1851 and 942,000 by 1870. To accommodate expansion, city commissioners were authorised to lay out a gridiron on the undeveloped part of Manhattan Island. As the map shows, the plan of 1811 imposed a uniform grid based on twelve 100ft (30m) wide north-south avenues and 155 east-west streets, 60ft (18m) in width. The southern start line of the commissioners’ plan is clearly seen against the earlier, randomly aligned grids. By 1850 the infilling of the grid had reached Forty-Second Street without any provision for major public open space. In 1858 the culmination of a fourteen-year campaign led to the appointment of Olmsted and Vaux to design Central Park. By 1898 the current five-borough city of New York was constituted with the addition of Brooklyn, Bronx, Queens and Staten Island to Manhattan. The expansion of New York was facilitated also by advances in structural engineering that allowed the Hudson and East Rivers to be bridged and tunnelled under, and that underlay the construction of skyscrapers and high-speed elevators. New York is today a ‘world city’ and one of the command centres of the global economy. The origins and growth of cities 79 can transport system due to the standardisation of rail gauges and increase in transcontinental lines; 2. the influx of large numbers of poor immigrants, which provided a continuous supply of low-cost labour; 3. the introduction of the assembly-line factory system, exemplified by the ‘Fordist’ production system of the Detroit automobile industry, which spread rapidly to all types of manufacturing; 4. the large surpluses being produced by a mechanised agricultural system; 5. the entrepreneurial activity of family-owned corporations. Pittsburgh, for example, became the base of the Carnegie Iron and Steel Corporation, which in 1901 merged with other producers to become US Steel. Consequently, by the early decades of the twentieth century the basic alignment of the North American urban system had been established. STAGE 4: THE AGE OF THE AUTOMOBILE AND AIR TRAVEL, 1920–70 Over the period 1920–70 the proportion of the US population living in urban places of 5,000 or more people increased from 47 per cent to 70 per cent. In geographical terms the major changes in the distribution of the urban population were: TABLE 3.1 THE FASTEST-GROWING US CITIES AND BIGGEST POPULATION LOSERS, 1980–90 The fastest-growing City 1980 The big population losers 1990 % City change 1980 1990 % change 1. Moreno Valley 28,309 CA 118,779 319.6 1. Gary IN 151,968 116,646 −23.2 2. Rancho Cucamonga CA 55,250 101,409 83.5 2. Newark NJ 329,248 275,221 −16.4 3. Piano TX 72,331 128,713 77.9 3. Detroit MI 1,203,369 1,027,974 −14.6 4. Irvine CA 62,134 110,330 77.6 4. Pittsburgh PA 423,960 369,879 −12.8 5. MesaAZ 163,594 288,091 76.1 5. St Louis MD 452,804 396,879 −12.4 6. Oceanside CA 76,698 6. Cleveland OH 573,822 505,616 −11.9 128,398 67.4 Source: adapted from E.Phillips (1996) City Lights New York: Oxford University Press 1. the emergence of super-metropolitan and megalopolitan urban areas; Urban geography 80 2. the spread of metropolitan urban development across most of the continent; 3. the growth of urban areas beyond the old central cities. These spatial trends were facilitated by transport developments, including the widespread use of the automobile and construction of the interstate highway system. The highest rates of urban growth occurred in the sunbelt, with slower growth and in some cases negative growth in the older cities of the North-East (Table 3.1). Between 1940 and 1980 Phoenix grew by 1,107 per cent and Albuquerque by 836 per cent. The phenomenal growth rates of the sunbelt cities were related to the exploitation of petrochemicals and the growth of the microelectronics, computer and aerospace industries. The new industries’ need for a smaller but highly trained and educated workforce freed them from the need to locate close to a large pool of factory labour and laid greater emphasis on other factors, including markets, prevailing wage and tax rates, and a benign climate. Other key factors in the expansion of sunbelt cities included: 1. the growth of federal military expenditure; 2. the migration of large numbers of retirees, often on government pensions, drawn by a warm climate and low cost of living; 3. an improved standard of living and changing lifestyle resulting in more leisure time; 4. the entrepreneurialism and civic boosterism of the political and economic leadership of sunbelt cities, which attracted inward investment.34 In terms of morphology, the cities of the sunbelt exhibit a distinctive character: shaped by the automobile, they have no need for a dominant downtown area but sprawl over the landscape in a low-density form of development encapsulated in the phenomenon of the edge city. STAGE 5: THE AGE OF DECONCENTRATION, 1970 TO THE PRESENT The continuation and extension of these trends to other regions during the latter part of the century allows us to add a fifth stage to Borcherfs (1967) model. This is characterised by a decrease in the size of the larger metropolitan areas, a decline in the density of the urban population, and increasing segregation of people in communities according to socio-economic factors such as class, race, age or language. The counterurbanisation trend is driven by industrial decentralisation, the insecurity and discomfort of metropolitan life as perceived by those able to exercise a choice, a search for amenities and retirement loci, and improved transportation and communications networks which enable people to enjoy an urban lifestyle without being permanently resident in a large metropolitan area. These characteristics are exhibited most clearly in the concept of the post-industrial city. POST-INDUSTRIAL URBANISM Post-industrialism is a social process that has had a major impact on the city. Its principal characteristics are: The origins and growth of cities 81 1. changes in the economy leading to a focus on the service sector rather than on manufacturing; 2. changes in the social structure that afford greater power and status to professional and technological workers; 3. changes in the knowledge base, with greater emphasis on R&D; 4. greater concern for the impact of technological change; 5. the advent of advanced information systems and intellectual technology. According to Bell (1973) the industrial post-industrial watershed was reached in 1956 in the USA when, for the first time in the history of industrial civilisation, the number of white-collar workers exceded the number of blue-collar workers.35 For Bell (1973) the advent of post-industrial societies meant fundamental changes in the nature of economic organisation and social relations. The resultant post-industrial city is characterised by: 1. an employment profile that reflects the twin processes of deindustrialisation and tertiarisation (see Chapter 14) as part of a restructuring of the economic base from a Fordist mode of industrial production to more flexible (post-Fordist) production systems (e.g. high-tech knowledge-based industries); 2. greater integration into the global economic system; 3. restructuring of urban form; 4. emerging problems of increasing income inequality, social and spatial segregation, privatisation of urban space and growth of defensible spaces. Post-industrial urbanism is characterised by fragmentation of urban form and its associated economic and social geographies.36 For some commentators this socio-spatial fragmentation heralds the advent of the postmodern city,37 and is evident in the ‘quartering’ of urban space. THE QUARTERING OF URBAN SPACE It is possible to identify a number of increasingly separate ‘residential cities’. Each has its source in a parallel, (although not always congruent), ‘city of business and work’. The five principal formations of parallel residential and business cities are: (1) the luxury city and the controlling city, (2) the gentrified city and the city of advanced services, (3) the suburban city and the city of direct production, (4) the tenement city and the city of unskilled work, (5) the abandoned city and the residual city. 1. The luxury areas of the city, while located in clearly defined residential districts, are not tied to any quarter. For the wealthy the city is less important as a residential location than as a location of power and profit. If they reside in the city, it is in a world largely insulated from contact with non-members of their class. The controlling city tends to be located in the high-rise centres of advanced services. The controlling city parallels the luxury areas of the residential city in many ways but not necessarily in physical proximity. The luxury city and the controlling city come together in space in the citadel where those at the top of the economic hierarchy live, work, consume and recreate in protected spaces. Urban geography 82 2. The gentrified city serves professionals, managers and ‘yuppies’ who may be doing well for themselves, yet work for others. Their residential areas are chosen for environmental or social amenities or access to white-collar jobs. Gentrified workingclass neighbourhoods, older middle-class areas and new developments with modern and well-furnished apartments all serve their needs. The parallel city of advanced services consists of professional offices with many ancillary services internalised in high-rise office towers and is enmeshed in a technologically advanced communications network. The skyscraper centre is the stereotypical form, as at La Défense in Paris or Canary Wharf in London’s Docklands. 3. The suburban city of the ‘traditional’ family is sought out by better-paid blue- and white-collar workers, the lower middle class. It provides stability, security and a comfortable world of consumption. The home as a symbol of self, exclusion of those of lower status, physical security against intrusion, political conservatism, comfort and escape from the workaday world (thus often incorporating substantial spatial separation from work) are characteristic. The city of direct production parallels, but is not congruent with, the residential suburban city in either time or space. It includes Plate 3.4 The social polarisation of the postmodern city is displayed by the scavenger collecting discarded drinks cans from rubbish bins in the up-market Sixteenth Street mall in Denver CO not only manufacturing but also the production aspects of advanced services, government offices and the back offices of major firms located in clusters in various locations within the metropolitan area that facilitate easy contact with The origins and growth of cities 83 clients. An example in New York City is the industrial valley for the printing industry between midtown Manhattan and the financial district. 4. The tenement city serves as home for lower-paid workers who often have irregular employment, few benefits, limited job security and little chance of advancement. In earlier days, the residents of these ‘slum neighbourhoods’ were often the victims of clearance and urban renewal programmes. Today they may experience abandonment, displacement, service cuts, deterioration of public facilities and political neglect. Struggles against displacement by urban renewal or gentrification have led to militant social movements in many cities. The city of unskilled work includes the informal economy, small-scale manufacturing, warehousing, sweatshops and unskilled consumer services. These are closely intertwined with the cities of direct production and advanced services and thus are located near them, but separately in scattered clusters. The economic city of unskilled work parallels the tenement city. 5. The abandoned city is the place of the very poor, the excluded, the never-employed and permanently unemployed, and the homeless. In older industrialised countries it will have a crumbling infrastructure and deteriorating housing, typical of slums and ghetto areas of the inner city. In less developed countries the excluded often live in peripheral squatter settlements. The abandoned city is paralleled in economic terms by the residual city the city of the less legal portions of the informal economy. In the developing countries the two cities overlap. In industrialised countries the residual city is the place where otherwise undesirable land uses are located. Many of the most polluting and environmentally detrimental components of the urban infrastructure, necessary for its economic survival but not tied directly to any economic activity, are located here. These include sewage treatment plants, incinerators, AIDS residences, homeless shelters, juvenile detention centres and jails. THE POST-INDUSTRIAL/ POSTMODERN CITY Just as industrialism left its imprint on the nineteenth-century city,38 so postindustrialism/post-modernism promoted changes in the form of the late twentieth-century city. Soja (1995)39 has characterised these trends in terms of six geographies of restructuring: 1. The restructuring of the economic base of urbanisation. This involves a fundamental change in the organisation and technology of industrial production, and the attendant social and spatial division of labour. This is marked by a shift from Fordist to postFordist urbanisation, from the tight organisation of mass production and mass consumption around large industrial complexes to more flexible production systems vertically disintegrated but geographically clustered in ‘new industrial spaces’. 2. The formation of a global system of world cities. This has the effect of expanding both the ‘outreach’ of particular world cities, bringing most of the globe within their effective hinterland, and their ‘inreach’, bringing into the global city capital and labour from all major cultural realms. In brief, the local is becoming globalised and the global localised in a process of ‘glocalisation’.40 3. The radical restructuring of urban form. This restructuring has generated a large number of neologisms, including megacity, outer city, edge city, metroplex, Urban geography 84 technoburb, post-suburbia, technopolis, heterpolis and exopolis to indicate a process whereby the city is ‘simultaneously being turned inside out and outside in’.41 The spatial organisation of the post-industrial/postmodern city is significantly different from that of the early modern city, exemplified in the neat concentric and sectoral models of the Chicago school (see Chapter 7), or even the more disjointed late modern city with its dominant central business district, inner city of poor and blue-collar workers, and sprawl of middle-class dormitory suburbs. Among the most obvious features of the post-industrial/postmodern city is the urbanisation of suburbia, but change has also affected the older central cities including significant reductions in density and the gentrification of former working-class neighbourhoods. As we shall see in Chapter 4, some of the large metropolitan regions have also experienced renewed growth after several decades of population decline. On a larger scale is the emergence of megacities, such as Mexico City. 4. The changing social structure of urbanism. Postmodern urbanism is associated with the development of new patterns of social fragmentation, segregation and polarisation, most evident in highly visible lifestyle differences and a growing gap between rich and poor. The socio-economic structure of the postmodern city is increasingly fluid and fragmented in ways that reduce the interpretative value of simple class-based divisions. 5. The rise of the carceral city. The complex geographies of postmodern cities have made them increasingly difficult to govern via traditional local-government structures. This has promoted the appearance of ‘carceral’ areas of walled-in residential estates protected by armed guards, shopping malls made safe by electronic surveillance, ‘smart’ office buildings impenetrable to ‘outsiders’, and neighbourhood-watch schemes organised by concerned homeowners.42 More positively, these developments have refocused attention on ‘the politics of place’, bringing enhanced local political consciousness over who controls and benefits from the restructuring of urban space. 6. A radical change in urban imagery. This refers to our images of the city and how these affect our behaviour and lifestyle in the postmodern city. Hollywood and Disneyland, and their equivalents elsewhere, produced a modernist hyper-reality as entertainment, but in the postmodern city hyper-reality has diffused from such specialised factories into everyday life.43 The popular media and expanding communications network have helped to promote the effects of hyper-reality on how people eat, BOX 3.15 Los Angeles: the archetypal postmodern metropolis Los Angeles has been described as the quintessential postmodern metropolis—not because it is a model for all other cities, but because the processes of postmodernism are displayed with particular clarity in the city. These may be illustrated with reference to six geographies or restructuring processes: 1. The restructuring of the economic base of urbanisation, seen in: ■ the growth of new technopoles outside the old industrial zone in what were once The origins and growth of cities 85 dormitory suburbs or agricultural land, exemplified by the clusters of hightechnology aerospace and electronics firms and industrial parks in Orange County and the San Fernando Valley; ■ the growth of the film and entertainment industry; ■ the presence of low-skill, labour-intensive, fashion-sensitive growth industries such as the apparel, furniture and jewellery industries, mostly around downtown Los Angeles but increasingly dispersed throughout the metropolitan area; ■ the growing finance, insurance and real estate (FIRE) sector providing business services to both domestic and foreign capital. 2. The formation of a global system of world cities, seen in: ■ the city’s growing importance as a world financial and trade centre; ■ the creation of urban multiculturalism, with more than one-third of the 9 million residents of Los Angeles County foreign-born. 3. The radical restructuring of urban form, seen in the complex geography of population and employment growth in the metropolitan area. The city of Moreno Valley, for example, grew rapidly during the 1980s as families moved there in anticipation of jobs. The non-appearance of the promised jobs led to overcrowded schools, poor social services, gridlocked freeways and four-hour daily commuting journeys. 4. The changing social structure of urbanism. Inequality is a deeply embedded facet of Los Angeles life, and the disparity between rich and poor continues to widen, with an increasing percentage of low-wage workers and a steady rise in the poverty rate (from 11 per cent in 1969 to 15 per cent in 1989) and in levels of homelessness. Economic inequalities often coincide with racial and ethnic divisions, with African-Americans, Latinos and Asians represented disproportionately at the bottom of the economic ladder, and located in inner-city areas such as Watts. The fluidity of postmodern urban restructuring is evident in the rapid racial transition in areas such as Huntington Park and Maywood, which have moved from being 80 per cent Anglo in 1965 to 90 per cent Latino. 5. The rise of the careeral city. The carceral (imprisoning) landscape of Los Angeles has been characterised as ‘the built environment of security-obsessed urbanism’, seen in the gated communities and voracious consumption of private security services, Nimby protest movements, increasing suburban separation, exclusionary zoning regulations and the hardening of the city against the undesirable poor, with absent public lavatories, razor wire-protected trash bins, and overhead sprinkler systems that operate randomly through the night to deter pavement (sidewalk) sleepers. 6. A radical change in the urban imagery. Los Angeles is the world’s leading centre for the production and marketing of hyper-reality. Not only have these image products reached around the globe (who is not familiar with the voyages of the Starship Enterprise?), but they also affect the local urban landscape. In the ‘theme-parked’ environment of Los Angeles one can choose to live in a place that caters for a particular lifestyle—for example, for the elderly, swinging singles or gay/lesbian communities—or occupy a ‘dreamscape’ replica of a Greek island, Little Tokyo or old New England. Urban geography 86 work, dress, sleep, vote, enjoy leisure that is, all the activities that underlie the social construction of urban space. Los Angeles in the late twentieth century assumed a position with regard to urban theory comparable to that of Chicago in the early twentieth century (Box 3.15). A number of academic and popular accounts have constructed Los Angeles as an archetype of contemporary and future urbanisation.44 These should not be accepted uncritically, however. In particular, it is important to reflect on how widespread the phenomenon of the post-modern city is. To suggest that the postmodern metropolis of Los Angeles offers a general model for the interpretation of contemporary urbanisation would be incorrect. As with the urban models of the Chicago school, so the greatest contribution of the California school of urban geography lies not in the modelling of urban form but in the exposition of the forces influencing contemporary urban landscapes which have been identified first and most visibly in Los Angeles. FURTHER READING BOOKS C.Chant and D.Goodman (1999) Pre-industrial Cities and Technology London: Routledge M.Dear (2000) The Postmodern Urban Condition Oxford: Blackwell D.Goodman and C.Chant (1999) European Cities and Technology London: Routledge P.Hall (1998) Cities in Civilisation London: Weidenfeld & Nicolson P.Hohenberg and L.Lees (1995) The Making of Urban Europe 1000–1994 Cambridge, MA: Harvard University Press M.Pacione (1995) Glasgow: The Socio-spatial Development of the City Chichester: Wiley J.W.Reps (1965) The Making of Urban America Princeton, NJ: Princeton University Press G.Roberts and I.Steadman (1999) American Cities and Technology London: Routledge G.Sjoberg (1960) The Pre-industrial City: Past and Present New York: Free Press E.Soja (2000) Postmetropolis Oxford: Blackwell S.Watson and K.Gibson (1995) Postmodern Cities and Spaces Oxford: Blackwell JOURNAL ARTICLES J.Borchert (1967) American metropolitan revolution Geographical Review 57, 301 22 V.G.Childe (1950) The urban revolution Town Planning Review 21, 3–17 J.Curry and M.Kenney (1999) The paradigmatic city: postindustrial illusion and the Los Angeles school Antipode 31(1), 1–28 R.Lawton (1972) An age of great cities Town Planning Review 43, 199–224 C.Pooley (1979) Residential mobility in the British city Transactions of the Institute of British Geographers 4, 258–77 A.Scott (1999) Los Angeles and the LA school: a response to Curry and Kenney Antipode 31(1), 29–36 KEY CONCEPTS ■ theories of urban origin The origins and growth of cities 87 ■ hydraulic theory ■ urban hearths ■ pre-industrial city ■ gridiron plan ■ medieval feudalism ■ medieval urbanism ■ early modern urbanism ■ industrial revolution ■ industrial capitalism ■ socio-spatial segregation ■ tenements ■ basic: non-basic employment ■ post-industrial urbanism ■ postmodern city STUDY QUESTIONS 1. Where, when and why did the first cities appear? 2. Examine the contribution of ancient Greece and Rome to the spread of urbanism. 3. The nineteenth century witnessed the flowering of industrial capitalism in Western Europe. Identify the major features of the socio-spatial structure of the European industrial city. 4. Trace the origins and development of urbanism in the USA. 5. Identify the principal characteristics of the post-industrial/postmodern city. 6. With the aid of a relevant example, examine the use of the gridiron plan in urban development. PROJECT Study the origins and development of a city with which you are familiar. Information may be collected from secondary sources such as maps and books held in your local or institutional library and from primary sources via your own fieldwork, which may involve sketching/photography or land-use mapping. The end product of your research should take the form of an annotated map of the city showing the important growth phases from urban origins to the present day. 4 The Global Context of Urbanisation and Urban Change Preview: global urbanisation trends; the changing distribution of world urban population; causes of urban growth; the size of the world’s settlements; megacities and million cities; the relationship between urbanisation and economic growth; the urbanisation cycle; stages of urban development; urbanisation; counterurbanisation; reurbanisation; suburbanization; exurbanisation; types of urbanised regions INTRODUCTION The global urban pattern is changing in three main ways as a result of: 1. urbanisation: an increase in the proportion of the total population that lives in urban areas; 2. urban growth: an increase in the population of towns and cities; 3. urbanism: the extension of the social and behavioural characteristics of urban living across society as a whole. In this chapter we first examine recent patterns of urbanisation and urban growth at the global scale in order to provide a framework for subsequent discussion of regional, national and local patterns and processes of urban change. THE URBANISATION OF THE GLOBE As we have seen in Chapter 3, the current high level of urbanisation at the global level is a relatively recent phenomenon. At the end of the nineteenth century the extent of world urbanisation was limited, with only Britain, North-West Europe and the USA more than 25 per cent urban in 1890.1 With less than 3 per cent of the world’s population living in towns and cities, levels of urbanisation elsewhere were insignificant. In the USA, urban development was confined primarily to the cities of the east coast and emerging Midwest. An indication of the rapid progression of urbanisation across the globe is provided in Figures 4.1– 4.4. The spread of urbanisation in Europe, North America and the Middle East is apparent, as are the rising levels of urbanisation in Africa and Asia, which were almost wholly rural in 1950. Figure 4.5 confirms that for most countries of the world urbanisation is a contemporary and ongoing process. Over the course of the past half- The global context of urbanisation and urban change 89 century, a world in which most people lived in rural areas has been transformed into a predominantly urban world. This trend has influenced not just the physical location of population but also the organisation and conduct of economic and social life of most people on the planet both urban and rural dwellers. Figure 4.5 also reveals the differential incidence of urbanisation in the world. As Table 4.1 indicates, the more developed regions (MDRs) exhibit high levels of urbanisation. More than 75 per cent of the population of Europe, North America, Japan and Australia New Zealand were urban dwellers in 1994, and by 2025 at least eight out of every ten people in these regions are expected to live in urban areas. Accordingly, in the developed countries the pace of urbanisation has slackened and has in some instances gone into reverse, as is indicated in the demographic-transition model (Box 4.1). By contrast, the less developed regions (LDR) are characterised by rapid urbanisation that is expected Figure 4.1 The Urban world in 1890 Figure 4.2 The urban world in 1950 Urban geography 90 to continue for decades (Figure 4.5 and Table 4.1). As Table 4.1 shows, in 1970, 25 per cent of the population of LDRs lived in urban areas. By 1994 almost 40 per cent were urban dwellers. Since the urban population is growing at an average rate of 3.5 per cent per annum while its rural counterpart grows at less than 1 per cent per annum, it is estimated that by 2025 almost 60 per cent of the population of LDRs (i.e. 4 billion people) will live in towns and cities. Figure 4.3 The urban world in 1900 Figure 4.4 The urban world by 2025 The global context of urbanisation and urban change 91 THE CHANGING DISTRIBUTION OF THE WORLD’S URBAN POPULATION The rapid growth of the world’s population has been accompanied in most countries by the multiplication and growth of urban places. The United Nations estimates that between 1950 and 2025 the number of urban dwellers will increase nearly sevenfold, from 738 million to 5.1 billion. But the world urban population is not distributed evenly among regions. As Table 4.1 reveals, in 1970 the more developed regions and less developed regions had a Figure 4.5 The urban population of world regions (%), 1970, 1994 and 2025 Source: United Nations (1995) World Urbanization Prospects: The 1994 Revision New York: United Nations similar number of urban dwellers, 677 million and 676 million respectively. The year 1970 represents a ‘tipping point’ in the global distribution of urban population. Prior to 1970, most urban dwellers lived in the MDR but this dominance has been declining since 1950, when 442 million (60 per cent) of the 737 million urban dwellers worldwide lived in the MDRs. Since 1970 the number of urban dwellers in LDRs has overtaken that of MDRs, and the gap continues to widen. Currently 1.7 billion urban dwellers (60 per cent of the world urban population) live in the LDRs while 868 million live in the MDRs. By 2025 4 billion of the 5 billion urban dwellers are expected to live in the LDRs. The distribution of urban population is also changing within the LDRs and the MDRs. In the former realm, Asia is a major region of urban growth. Whereas in 1970 Asia was home to 503 million urban dwellers (37 per cent of the world total), by 1994 1.2 billion (46 per cent) of the 2.5 billion global urban dwellers were Asian. It is anticipated that 2.7 billion (more than half the world’s urban dwellers) will live in Asia by 2025. This trend is in marked contrast to the situation in Europe. As Table 4.1 shows, with 423 million urban dwellers in 1970 Europe was second only to Asia. Between 1970 and 1994 Europe added Urban geography 92 110 million urban dwellers with a further 65 million expected by 2025. During the period 1994–2025 Asia is expected to add 1.5 billion urban residents—or about 23 new urban Asians for every new European urban resident. Latin America and the Caribbean are also growing rapidly, with their urban population more than doubling from 163 million in 1970 to 349 million in 1994. The urban population of the region is expected to reach 601 million by 2025, slightly more than the number projected for Europe (598 million). Africa exhibits the fastest urban growth rate of any major world region. From 84 million urban residents in 1970 Africa by 1994 had 240 million, and by 2025 the number of urban dwellers is expected to reach 804 million. All these trends are confirmed by the analysis of urban growth rates (Table 4.2 and Figure 4.6). TABLE 4.1 URBAN POPULATION AND PERCENTAGE URBAN IN MORE DEVELOPED AND LESS DEVELOPED REGIONS, 1970, 1994 AND 2025 Region Urban population (million) Urban share (%) 1970 1994 2025 1970 1994 2025 More developed regions 677 868 1,040 67.5 74.7 84.0 Australia-New Zealand 13 18 26 84.4 84.9 89.1 Europe 423 532 598 64.4 73.3 83.2 Japan 74 97 103 71.2 77.5 84.9 Northern America 167 221 313 73.8 76.1 84.8 Less developed regions 676 1,653 4,025 25.1 37.0 57.0 Africa 84 240 804 23.0 33.4 53.8 428 1,062 2,615 21.0 32.4 54.0 163 349 601 57.4 73.7 84.7 1 2 5 18.0 24.0 40.0 Asia a Latin America Oceania b Source: adapted from United Nations (1995) World Urbanization Prospects: The 1994 Revision New York: United Nations Notes: aExcluding Japan. bExcluding Australia-New Zealand BOX 4.1 The demographic-transition model The global context of urbanisation and urban change 93 This model describes the main stages of population change that accompany industrialisation and urbanisation: 1. Birth rates and death rates are both high, leading to fluctuating population growth determined more by mortality than by fertility. This is associated with conditions in pre-industrial society and in some developing areas today (e.g. sub-Saharan Africa), where most people live in rural environments at a subsistence level. 2. In the early stage of development, death rates begin to fall, owing to improvements in medicine and public health. Family and religious values can contribute to maintain a high birth rate, leading to rapid population growth. Many of the population migrate from rural to urban areas, feeding city growth and urbanisation. This phase is typical of many Third World countries today. 3. With the move to cities, large families are less of an economic asset than they had been in rural society, and more of an expense on the family budget. Birth rates fall to a level more in line with the death rate, leading to a slowdown in population growth. Europe and North America have passed through this phase. 4. This stage has been reached in highly developed countries, such as the UK and the USA, where both birth and death rates are low and population growth rates are near zero. THE CAUSES OF URBAN GROWTH The increasing levels of urbanisation and urban growth identified are the result of a combination of natural increase of the urban population and net in-migration to urban areas. The two major processes reinforce each other, although their relative importance varies. Findley (1993),2 for example, found that for twenty-four developing countries between 1975 and 1990 the average contribution of migration to urban growth was 54 per cent. This underestimates the real situation, however, since many migrants who live and work in the city do not move there permanently (see Chapter 23). Finally, it is important to note that although it is often assumed that the largest migration flows are in the countries of the South, more migration takes place in the USA, where one person in five moves each year, albeit between urban areas, than in all the LDRs. As we shall see in the following chapter, the restructuring of the urban system in the USA during the 1980s was more rapid and fundamental than that taking place in most countries of the LDRs. Urban geography 94 SETTLEMENT SIZE The world’s urban population is distributed among settlements of differing sizes along a continuum from small towns with several thousand people to giant cities with populations of tens of millions. Most of the urban population live in settlements with fewer than 500,000 inhabitants.3 Most of these intermediate settlements function as links between town and country where agricultural surpluses are exchanged for manufactured goods and services in accordance with the precepts of central place theory (see Chapter 6). The distribution of urban population in terms of settlement size is shown in Table 4.3. In 1950 only one city (New York) had a population of 10 million or more, accounting for 1.7 per cent of the world urban population. By 1990 twelve cities had attained this size and shared 7.1 per cent of the world urban population. It is anticipated that by 2015 twenty-seven cities will reach the 10 million mark, accommodating 10.9 per cent of the world urban population. In absolute numerical terms this represents a rise from 12 million people living in a single city in 1950 to 450 million in several giant cities by 2015. Between 1990 and 2015 nearly all the population growth in the largest urban agglomerations is expected to occur in the LDRs. At the other end of the population-size continuum, cities with fewer than 500,000 inhabitants were home to 63.7 per cent of the world urban population in 1950. While their share is expected to decrease TABLE 4.2 AVERAGE ANNUAL RATE OF CHANGE OF URBAN POPULATION, 1965– 70, 1990–5 AND 2020–5 (%) Region 1965–70 1990–5 2020–5 World 2.64 2.53 1.93 Less developed regions 3.58 3.51 2.33 Africa 4.64 4.38 3.34 Asiaa 3.28 3.68 2.31 3.97 2.60 1.26 2.26 3.13 3.32 More developed regions 1.74 0.75 0.45 Australia-New Zealand 2.35 1.32 1.05 Europe 1.68 0.84 0.24 Japan 2.20 0.37 −0.06 North America 1.63 1.29 0.99 Latin America Oceania b Source: adapted from United Nations (1995) World Population Prospects: The 1994 Revision New York: United Nations Notes: aExcluding Japan. bExcluding Australia-New Zealand The global context of urbanisation and urban change 95 Figure 4.6 Average annual rate of urban population growth, 1990–5 slowly in the future, more than 50 per cent of world urban population is still expected to live in such cities in 2015 (Table 4.3). MEGACITIES AND MILLION CITIES One of the most striking features of the global urban pattern is the degree to which the urban population lives in giant cities that dominate the global urban and economic systems. Against the background of a general increase in the number of people living in urban places, it is the metropolitan regions that are proliferating and expanding most rapidly. While noting the difficulties of urban definition and cross-national comparison, we can identify a number of significant trends in the geographical distribution of megacities. Table 4.4 lists the fifteen largest urban agglomerations at different points in time, and enables us to map the major changes over the post-Second World War period. Tokyo, with a population of 27.9 million in 2000, has been the largest city in the world since 1970 and is projected to retain that rank (Box 4.2). By contrast, New York is projected to continue to slip down the ranking over the next twenty-five years. Other expected changes include the entry of Lagos, Karachi, Delhi and Dhaka to replace Rio de Janeiro, Osaka, Buenos Aires and Seoul by the year 2015. By then, Lagos is expected to be the third-largest city in the world after Tokyo and Bombay. Comparison of the lists for 1950 and 2000 demonstrates the remarkable shift in the global distribution of largest cities from the MDRs to the LDRs, a trend that will continue for the foreseeable future. Urban geography 96 TABLE 4.3 URBAN POPULATION, NUMBER OF CITIES AND PERCENTAGE OF URBAN POPULATION BY CITY-SIZE CLASS: WORLD, MORE DEVELOPED REGIONS AND LESS DEVELOPED REGIONS Size class World More developed regions Less developed regions 1950 1970 1990 2015 1950 1970 1990 2015 1950 1970 1990 2015 10 million or more No. of 1 agglomerations 3 12 27 1 2 4 4 0 1 8 23 Population 12 44 161 450 12 33 63 71 0 11 98 378 % urban 1.7 3.2 7.1 10.9 2.8 4.8 7.5 7.2 0.0 1.7 6.9 12.0 From 5 million to 10 million No. of 7 agglomerations 18 21 44 5 8 6 8 2 10 15 36 Population 42 130 154 282 32 61 44 56 10 69 110 226 % urban 5.7 9.6 6.8 6.8 7.2 9.0 5.2 5.7 3.5 10.2 7.7 7.2 From 1 million to 5 million No. of 75 agglomerations 144 249 472 43 73 98 120 32 71 151 352 Population 140 265 474 941 84 136 191 240 56 129 283 701 % urban 19.0 19.6 20.8 22.7 19.1 20.1 22.7 24.2 19.0 19.0 20.4 22.2 From 500,000 to 999,999 No. of 105 agglomerations 175 295 422 59 85 104 123 46 90 191 299 Population 73 122 203 293 42 61 72 84 31 61 132 209 % urban 9.9 9.0 8.9 7.1 9.5 9.0 8.5 8.5 10.5 9.0 9.2 6.6 Fewer than 500,000 Population 470 792 1,284 2,178 272 386 472 540 198 406 812 1,638 % urban 63.7 58.5 56.4 57.1 56.1 54.5 67.0 60.0 56.6 52.0 52.6 61.5 Source: adapted from United Nations (1995) World Urbanization Prospects: The 1994 Revision New York: United Nations Also, the largest cities are becoming larger; the average population of the world’s largest cities was over 5 million inhabitants in 1990, compared with 2.1 million in 1950, The global context of urbanisation and urban change 97 and less than 200,000 in 1800. The number of megacities (defined by the United Nations as cities with 8 million or more inhabitants) is increasing rapidly, particularly in LDRs. Whereas in 1950 only New York and London had a population of 8 million or more, by 1970 eleven cities had become megacities (Table 4.5). Three were located in Latin America and the Caribbean (São Paulo, Buenos Aires and Rio de Janeiro), two in North America (New York and Los Angeles), two in Europe (London and Paris) and four in Asia (Tokyo, Shanghai, Osaka and Beijing). By contrast, in 1994 sixteen of the twentytwo megacities were in LDRs, and by 2015 it is expected that twenty-seven of the thirtythree megacities will be located in LDRs. The geographical shift in the focus of megacity growth is repeated in the distribution of ‘million cities’ (Table 4.6). Some of these giant cities have emerged as global or ‘world cities’ (see Chapter 14). URBANISATION AND ECONOMIC GROWTH Regional differences in rates of urbanisation throughout the world provide clear evidence of the relationship between urbanisation and industrialisation. As we saw in Chapter 3, prior to the industrial revolution urban development was restricted by the amount and value of surplus product that could be generated and accumulated in a single place. The low level of economic growth imposed a ceiling on the number of people who could be sustained in urban places. The key role that cities play in dynamic and competitive economies and the relationship between the scale of a national economy and the level of urbanisation is illustrated by the fact that most of the world’s largest cities are in the world’s largest economies.4 In 1990 the world’s twenty-five largest economies had over 70 per cent of the world’s ‘million cities’ and all but one of its twelve agglomerations with 10 million or more inhabitants. Despite such categorical evidence, the relationship between urbanisation and level of economic development is complex, with many factors at work, and while economic growth may result in increased levels of urbanisation, higher levels of urbanisation in turn can stimulate more economic growth. We can illustrate this in the context of the Third World, where a number of factors can mediate the relationship between urbanisation and economic growth. These include the following: 1. The nature of economic activity within each sector of the national economy can affect urbanisation. For example, the type of agricultural enterprise can affect the scale of urban settlement. High-value crops that provide good incomes for farmers and workers within intensive farming systems can support rapid growth of local urban centres to the point where agricultural surpluses support a relatively urbanised population. 2. The nature of land ownership is important in determining whether profits are spent or invested within the country or taken abroad. 3. Cultural preferences for a type of lifestyle can influence the level of urbanisation. 4. Government policies and the activities of state institutions are also important. An increasing share of national income is controlled by the public sector. Since most government employees are urban residents, most public spending flows to towns and Urban geography 98 cities. Some observers have equated this trend with ‘urban bias’ in the investment strategies of Third World governments and have argued that this TABLE 4.4 THE FIFTEEN LARGEST URBAN AGGLOMERATIONS, RANKED BY POPULATION SIZE, 1950, 2000 AND 2015 Rank Agglomeration and country Population (million) 1950 1. New York, USA 12.3 2. London, UK 8.7 3. Tokyo, Japan 6.9 4. Paris, France 5.4 5. Moscow, Russian Federation 5.4 6. Shanghai, China 5.3 7. Essen, Germany 5.3 8. Buenos Aires, Argentina 5.0 9. Chicago, USA 4.9 10. Calcutta, India 4.4 11. Osaka, Japan 4.1 12. Los Angeles, USA 4.0 13. Beijing, China 3.9 14. Milan, Italy 3.6 15. Berlin, Germany 3.3 2000 1. Tokyo, Japan 27.9 2. Bombay, India 18.1 3. São Paulo, Brazil 17.8 4. Shanghai, China 17.2 5. New York, USA 16.6 6. Mexico City, Mexico 16.4 7. Beijing, China 14.2 8. Jakarta, Indonesia 14.1 9. Lagos, Nigeria 13.5 10. Los Angeles, USA 13.1 The global context of urbanisation and urban change 99 11. Calcutta, India 12.7 12. Tianjin, China 12.4 13. Seoul, Republic of Korea 12.3 14. Karachi, Pakistan 12.1 15. Delhi, India 11.7 2015 1. Tokyo, Japan 28.7 2. Bombay, India 27.4 3. Lagos, Nigeria 24.4 4. Shanghai, China 23.4 5. Jakarta, Indonesia 21.2 6. São Paulo, Brazil 20.8 7. Karachi, Pakistan 20.6 8. Beijing, China 19.4 9. Dhaka, Bangladesh 19.0 10. Mexico City, Mexico 18.8 11. New York, USA 17.6 12. Calcutta, India 17.6 13. Delhi, India 17.6 14. Tianjin, China 17.0 15. Metro Manila, Philippines 14.7 Source: adapted from United Nations (1995) World Urbanization Prospects: The 1994 Revision New York: United Nations BOX 4.2 Tokyo: global megacity Tokyo is the largest urban agglomeration in the world and is approaching the threshold of a 30 million population super-city. It has long been the dominant urban centre in Japan. As early as 1700 Tokyo had a population of over 1 million, when the population of London was 650,000. Following the Second World War, the city’s population grew from 6.9 million in 1950 to 16.5 million in 1970, reaching 26.5 million in 1994. The outer limit of the urbanised area now reaches 100 miles (160km) from the urban core. Tokyo meets all the criteria of world-city status. The city dominates the Japanese urban system. It is the major commercial centre, housing the headquarters of all major Japanese companies; a manufacturing and wholesaling centre; the chief financial centre; Urban geography 100 the seat of government; the cultural capital; the centre for media and advertising industries; and has the largest concentration of institutions of higher learning in the world. The Tokyo metropolitan region is also the world’s largest consumer market. The unprecedented growth of the city has led to urban problems, including escalating land prices, shortage of office space, high-cost housing, congestion and excessive commuting journeys. In recent years the government has sought to encourage decentralisation through the construction of technopoles, small cities of around 50,000 built for high-technology industries, research institutes and colleges near existing medium-size cities. It has also discussed the relocation of capital functions to other parts of the Tokyo metropolitan region. This ties in with the long-range plan for the region, which has among its goals the development of a multi-core urban structure in which the functions of the city are dispersed to reduce distances between place of employment and place of residence. has encouraged the migration of people from rural to urban areas, thereby leading to over-urbanisation (Box 4.3). In general, although there is a close relationship between the level of urbanisation and national levels of economic development, the range of potential influences suggests that different factors may be of importance in different national settings. Furthermore, the relative importance of factors that influence levels of urbanisation may change over time as a country becomes more urbanised. THE URBANISATION CYCLE The simplicity of the logistic-growth urbanisation curve (Box 4.4) should not be taken to mean that urbanisation is a unidirectional (low to high) process. In a large number of advanced countries the level of urbanisation actually decreased between 1965 and 1990. Although this can be a result of statistical underbounding of urban areas (see Chapter 2), it is now generally acknowledged to indicate a process of population redistribution which involves either the relatively faster growth of smaller urban places or the absolute decline of the largest cities.7 The shift in the incidence of strongest population growth away from the largest cities in the national urban system has been termed ‘polarisation reversal’8 or, more commonly, ‘counterurbanisation’.9 Geyer and Kontuly (1993)10 have incorporated these concepts into a theory of differential urbanisation which postulates that large, intermediate-size and small cities go through successive periods of fast and slow growth in a cycle of development. The main stages of this urbanisation cycle are summarised in Figure 4.7. THE PRIMATE CITY PHASE During the initial phase of urbanisation, the primate city phase, an increasing proportion of economic activity and population in a country concentrates in a limited number of rapidly growing primate cities (Box 4.5). This phase can be sub-divided into three stages: The global context of urbanisation and urban change 101 1. First is an early primate city stage in which a primate city attains spatial dominance of the urban system, attracting a large proportion of net inter-regional migration. 2. An intermediate primate city stage follows, in which the rapidly growing primate city is still monocentric in form but with suburbanisation a prominent feature. Suburban nodes, the nuclei of a future multinodal primate city, may emerge. As a result of favourable locational TABLE 4.5 NUMBER OF MEGACITIES, 1970, 1994, 2000 AND 2015 Region 1970 1994 World 2000 2015 11 22 25 33 Less developed regions 5 16 19 27 Africa 0 2 2 3 2 10 12 19 Latin America 3 4 5 5 More developed regions 6 6 6 6 Europe 2 2 2 2 Japan 2 2 2 2 Northern America 2 2 2 2 Asia a Source: adapted from United Nations (1995) World Urbanization Prospects: The 1994 Revision New York: United Nations Note: aExcluding Japan TABLE 4.6 REGIONAL DISTRIBUTION OF THE WORLD’S POPULATION IN ‘MILLION CITIES’ AND THE LOCATION OF THE WORLD’S LARGEST 100 CITIES Proportion of the world’s: Urban population 1950 1990 No. of the world’s 100 largest cities in: Population in ‘million cities’ 1950 1990 1800 1950 1990 Africa 4.5 8.8 1.8 7.5 4 3 7 Eastern Africa 0.5 1.7 – 0.8 – – – Middle Africa 0.5 1.0 – 0.8 0 0 1 Northern Africa 1.8 2.8 1.8 3.2 3 2 5 Southern Africa 0.8 0.9 – 0.8 0 1 0 Urban geography Western Africa 102 0.9 2.6 – 2.0 1 0 1 23.7 23.0 30.1 27.8 3 26 27 2.8 4.2 2.2 3.5 2 2 3 14.4 9.2 21.2 13.1 0 18 13 6.5 9.7 6.7 11.1 1 6 11 Asia 32.0 44.5 28.6 45.6 64 33 44 Eastern Asia 15.2 19.7 17.6 22.2 29 18 21 3.7 5.8 3.4 5.6 5 5 8 11.2 14.8 7.0 14.6 24 9 13 1.8 4.1 0.6 3.3 6 1 2 Europe 38.8 22.8 38.0 17.9 29 36 20 Eastern Europe 11.8 9.3 7.7 6.3 2 7 4 Northern Europe 7.7 3.4 9.0 2.1 6 6 2 Southern Europe 6.5 4.0 6.7 3.2 12 8 6 Western Europe 12.8 6.2 14.6 6.2 9 15 8 1.1 0.8 1.6 1.3 0 2 2 Americas Central America and the Caribbean Northern America South America South-East Asia South-Central Asia Western Asia Oceania Source: adapted from United Nations (1996) An Urbanising World: Global Report on Human Settlements Oxford: Oxford University Press BOX 4.3 Overurbanisation The concept of overurbanisation in less developed countries implies that economic growth is unable to keep pace with urban population growth, thereby leading to the major social and economic problems evident in most large cities of the Third World. It is also argued that Overurbanisation is detrimental to national economic growth, since, with limited investment available to most Third World states, the political influence of cities and resultant high level of investment in urban areas will reduce investment in other productive sectors of the economy. The concept of overurbanisation is not accepted by those who reject the implication that the form of urbanisation in the North represents a model for the South and that any departure from this pattern represents ‘overurbanisation’. Despite the simplicity of the logistic curve (see Box 4.4) urbanisation is not a single process followed by all countries in the process of development, and it would be surprising if urban trends in the South mirrored patterns in the North. The much weaker position of most developing countries in world markets, and the fundamental differences between the world in the nineteenth century and late twentieth century must affect the social economic and political factors The global context of urbanisation and urban change 103 that influence levels of urbanisation. The notion of overurbanisation is useful to the extent that it emphasises the dynamic links between different economic sectors and areas of a country. Given that there are 600,000 agricultural villages in India where the infant mortality rate is still 100 per thousand, literacy is less than 40 per cent, malnutrition is widespread, the supply of drinking water is inadequate, electricity is lacking and half the population live in mud huts,5 it is little wonder that rural-urban migration is a feature of many Third World nations. In such situations the problem of ‘over-urbanisation’ might be addressed via rural development policy, which could also benefit the national economy in countries such as India where a significant part of GDP comes from the agricultural sector. Although it is not inappropriate to use the term ‘overurbanisation’ to describe a situation of high levels of primate city urbanisation, rapid in-migration, saturated urban labour markets and overburdened urban services, in seeking explanations we must look beyond these urban problems to national and international economic and political forces. attributes, certain intermediate-size cities develop faster than others. 3. Phase 2 is followed by an advanced primate city stage in which the primate city becomes so large that owing to agglomeration diseconomies, such as congestion costs, a monocentric urban structure becomes unwieldy. By means of intra-regional decentralisation the primate city develops a multicentred or megalopolitan character, and dominates the rest of the urban system economically and spatially. Expansion of the urban system as a whole may lead to one or more intermediate-size cities becoming as large as existing cities. At some point in the development history of most countries the growth rate of the primate city slows and a process of spatial deconcentration commences. THE INTERMEDIATE CITY PHASE The slowing growth rates of the primate city and spatial deconcentration of urban population are often accompanied by growth of intermediate-size centres close to the primate city. This ‘population turnaround’ or ‘polarisation reversal’ has been documented in a number of countries including the USA, the UK, France, Greece and Brazil.12 This phase can be subdivided into two stages: 1. An early intermediate city stage characterised by the uneven growth of a limited set of intermediate-sized cities that are close to but not contiguous with the primate metropolitan region. Although the primate city is still gaining population in absolute terms, it is starting to lose in relative terms to the intermediate cities. Suburban centres within the primate metropolitan region are now growing faster than the central city. BOX 4.4 The urbanisation curve Urbanisation is a process of population concentration whereby towns and cities grow in relative importance through first an increasing proportion of the national population Urban geography 104 living in urban places and, second, the growing concentration of these people in the larger urban settlements. It has been suspected that all nations pass through this process as they evolve from agrarian to industrial societies. For Davies (1969)6 the typical course of urbanisation for a nation is represented by a logistic curve. The first section of the curve is associated with very high rates of urbanisation associated with large shifts of population from rural areas to towns and cities in response to the creation of an urban economy. This is followed by a longer period of consistent moderate urbanisation. As the urban percentage reaches above 60, the curve begins to flatten, approaching a ceiling of around 80 per cent. This is the level at which rural and urban populations appear to reach a functional equilibrium. At any one time individual countries are at different stages in the urbanisation curve. 2. An advanced intermediate city stage in which the suburbanisation process that characterised the development of the primate city during the advanced primate city stage is reproduced in the fastest-growing intermediate-size cities but on a smaller scale. In addition, in contrast to the early intermediate city stage, all centres within the primate metropolitan region begin to lose population in absolute terms, with the central city losing more than the suburban centres. Figure 4.7 Generalised stages of differential urbanisation Source: H.Geyer and T.Kontuly (1993) A theoretical foundation for the concept of differential urbanization International Regional Science Review 17(2), (2), 151–11 The global context of urbanisation and urban change 105 THE SMALL CITY PHASE The small city phase represents a continuation of the previous stage during which deconcentration takes place from the primate and intermediate-size cities towards small urban centres which may eventually grow at a faster rate than either the primate city or the intermediate-size cities. By the end of this phase the urban system has reached a ‘saturation point’ where the rural population cannot be reduced much further (see the logistic curve in Box 4.4) and rural-urban migration ceases to be a major contributory factor in the urbanisation cycle. Since population growth through natural increase may also be very low in advanced societies (see the demographic transition model in Box 4.1), urban growth in general may be slow. However, as the model shows (Figure 4.7), the ‘small city phase’ marks not only the end of the first cycle of urban development but also the beginning of a new one which follows the sequence of major metropolitan, intermediate-size city, and small city growth. Significantly, the set of major metropolitan areas experiencing the highest rates of net inBOX 4.5 The law of the primate city The law of the primate city refers to a situation in which a single city accommodates a disproportionately large number of a country’s population. In some instances the primate city size distribution is the result of outside or foreign influences on the settlement pattern. In many present-day Third World countries, for example, primate cities developed as a result of the intervention of a colonial power. Bangkok is one such example. Jefferson (1939)11 argued that, in the early stages of a country’s urban development, the city that emerges as larger than the rest develops an impetus to self-sustaining growth that enables it, over time, to attract economic and political functions to such an extent that it dominates the national urban system. Capital cities such as Paris or Vienna occupy this niche. In some countries a variety of forces, such as nationalism in Spain and territorial size in the USA, have led to several cities growing to comparable size rather than the emergence of a single primate city. The law of primacy is most relevant to countries that have a relatively simple economy and spatial structure, a small area and population, low incomes, economic dependence upon agriculture and a colonial past. migration during the urbanisation phase of the second cycle may or may not be the same set of large urban areas from the first cycle (as indicated by differential urban growth rates in the rustbelt and sunbelt areas of the USA). Large urban centres able to retain their dominant position in the national and international urban hierarchy, along with a limited group of rapidly growing intermediate-size urban areas from the first cycle, will constitute the new set of major metropolitan centres. Urban geography 106 A ‘STAGES OF URBAN DEVELOPMENT’ MODEL The concept of a cycle of urbanisation has also been employed by Klaassen et al. (1981) and van den Berg et al. (1982) to study the growth patterns within individual urban agglomerations.13 As Figure Figure 4.8 The stages of urban development model Source: A.Champion (2000) Urbanization, suburbanization, counterurbanization and reurbanization, in R.Paddison and W.Lever (eds) Handbook of Urban Studies London: Sage 4.8 shows, four stages of urban development are envisaged: 1. urbanisation: when certain settlements grow at the cost of their surrounding countryside; 2. suburbanisation or exurbanisation: when the urban ring (commuter belt) grows at the cost of the urban core (physically built-up city); 3. disurbanisation or counterurbanisation: when the population loss of the urban core exceeds the population gain of the ring, resulting in the agglomeration losing population overall; 4. reurbanisation: when either the rate of population loss of the core tapers off, or the core starts regaining population with the ring still losing population. As Figure 4.8 indicates, the model is based on changes in the direction and rate of population movement between urban core and urban ring (which together comprise a functionally related daily urban system). The two types of change are absolute shifts when the directions of population change in the two zones differ, and relative shifts when change occurs in the same direction but at different rates. These trends are summarised in Table 4.7. We have already examined the processes and patterns of urbanisation. Here we The global context of urbanisation and urban change 107 TABLE 4.7 STAGES OF DEVELOPMENT OF A DAILY URBAN SYSTEM Stage of development Classification type Population change characteristics Core Ring DUS I II 1. Absolute centralisation ++ − + 2. Relative centralisation ++ + +++ Suburbanisation/ 3. Relative decentralisation + ++ +++ Exurbanisation 4. Absolute decentralisation − ++ + 5. Absolute decentralisation −− + − Counterurbanisation 6. Relative decentralisation −− − −−− 7. Relative centralisation − −− −−− 8. Absolute centralisation + −− − Urbanisation III Disurbanisation/ IV Reurbanisation Total growth (concentration) Total growth (deconcentration) TABLE 4.8 GREAT BRITAIN: POPULATION CHANGE, 1951–91, BY FUNCTIONAL REGION ZONE (PER CENT) Zone type Rate for decade 1951– 61 1961– 71 Deviation from GB rate 1971– 81 1981– 91 1951– 61 1961– 71 1971– 81 1981– 91 Great Britain 4.97 5.25 0.55 2.50 0.00 0.00 0.00 0.00 Core 3.98 0.66 −4.20 −0.09 −0.99 −4.59 −4.75 −2.59 Ring 10.47 17.83 9.11 5.89 5.50 12.58 8.56 3.39 Outer area 1.74 11.25 10.11 8.85 −3.23 6.00 9.56 6.35 Rural area −0.60 5.35 8.84 7.82 −5.57 0.10 8.29 5.32 Source: adapted from A.Champion (2000) Urbanization, suburbanization, counterurbanization and reurbanization, in R.Paddison and W.Lever (eds) Handbook of Urban Studies Beverly Hills, CA: Sage Urban geography 108 focus on population movements associated with the other major dimensions of urban change identified in Table 4.7. REURBANISATION The empirical evidence for reurbanisation is mixed. A study of 241 functional urban regions (FURs) in Europe found that between 1981 and 1991 the proportion of urban cores gaining population reached 47 per cent, compared with only 22 per cent over the period 1975–81.14 However, it was mainly the smaller FURs (particularly those with ancient cathedrals and universities) that exhibited reurbanisation, not the larger, older urban regions. In the UK reurbanisation occurred in only four of thirty-six FURs (Glasgow, Oxford, Cambridge and Canterbury), with only Glasgow confirming model expectations. On the other hand, there is a growing body of case-study evidence that indicates a recovery of large cities from the high levels of population loss experienced in the 1970s era of counterurbanisation. As Table 4.8 shows, the rate of population loss for all 280 of Britain’s urban areas fell from 4.2 per cent in 1971–81 to 0.1 per cent for 1981–91, while the growth rate for the urban rings declined from 9 per cent to less than 6 per cent. In the USA the 1980s witnessed the re-emergence of the larger metropolitan areas as the fastestgrowing elements of the urban landscape15 (Table 4.9). Overall, metropolitan areas with 1 million or more residents grew by 12 per cent in the 1980s compared with 8 per cent in the previous decade. While much of this growth was in the South and West even the North’s large metropolitan areas switched from a population decline of 0.9 per cent between 1970 and 1980 to a 2.7 per cent increase in the 1980s. TABLE 4.9 TYPES OF POPULATION CHANGE IN METROPOLITAN AREAS OF THE USA, 1980–90 AND 1990–96 Type 1980–90 No. of cities 1990–6 % No. of cities % 1 39 17.0 15 6.6 2 54 23.5 53 23.3 3 40 17.4 31 13.7 4 60 26.1 85 37.4 5 37 16.1 43 18.9 Source: adapted from J.Mercer (1999) North American cities: the micro-geography, in F.Boal and S.Royle (eds) North America: A Geographical Mosaic London: Arnold, 191–206 Notes: Type 1 cities: central-city decline added to metropolitan decline. Type 2 cities: centralcity decline and metropolitan growth. Type 3 cities: central-city stagnation (equal to a less than 5% population change over the period) and metropolitan growth. Type 4 cities: strong central-city growth (between 5.1–19.9% over the period) and metropolitan growth. Type 5 cities: booming central-city growth (20.0% or more over the period) and metropolitan growth The global context of urbanisation and urban change 109 The population growth that has occurred in the central areas of US cities was fuelled by two principal migration streams. First, new migrants, primarily from Latin America and Asia, moved into lower-value areas of cities such as New York and Los Angeles as well as into other metropolitan areas on the west coast (San Diego and San Francisco), in the South-West (Houston) and Florida (Miami) that historically had attracted relatively fewer migrants (see Chapter 5). The second stream comprised a flow of ‘baby-boomers’ (those born just after the Second World War and during the affluence of the 1950s and early 1960s), investing in high-status residential areas. During the 1980s the strongest magnets for adult ‘boomers’ were metropolitan areas with expanding high-tech and defence-oriented economies, including coastal cities such as Boston and Seattle and sunbelt locations like Dallas and Atlanta. Australia and Canada also provide evidence of strengthening metropolitan areas and inner-city growth in the 1980s.16 Evidence for US cities over the period 1990–2000 indicates that the occurrence and extent of downtown ‘population rebound’ varies considerably (Table 4.10). While in some cities the downtown contribution to metro growth is small, in others it represented a significant proportion of total population growth, and in others downtown population growth offset citywide population decline. In general, these empirical observations suggest that: 1. There are widespread signs of renewed growth or reduced population decline for larger metropolitan areas, as well as a population recovery for urban cores. 2. There is no evidence of suburban-ring areas losing out to core areas, not even in relative terms, let alone in accordance with the absolute change associated with the later phase of reurbanisation specified in the ‘stages of urban development’ model. Reurbanisation as defined by Klaassen et al. (1981)17 has not yet emerged as a significant feature in the urban systems of advanced economies. There is also considerable disagreement over the extent to which the inner-city revitalisation that took place in the 1980s will be able to continue and lead to a fundamental change in the form of the Western city. The process of decentralisation, on the other hand, is likely to continue as a major feature of post-industrial urbanisation, albeit in a form very different from the dormitory-style suburbanisation of the early post-Second World War period. We can identify two main forms of population decentralisation. The first, counterurbanisation or urban deconcentration, is characterised by net population movement from metropolitan regions into smaller urban regions and rural areas that lie beyond the primary commuter-sheds of the major cities. The second, suburbanisation, reflects a long-established centrifugal movement of population which progressively has involved a broader range of urban functions than just housing taking place over longer distances as personal mobility has grown and urban centres have expanded to embrace their previously rural hinterlands. Urban geography 110 TABLE 4.10 CITY AND DOWNTOWN POPULATION CHANGES IN THE USA, 1990– 2000 Cities where downtown population growth contributed to city population growth Downtown population 1990 2000 City population Increase 1990 2000 Increase Downtown share of city growth (%) Miami FL 15,143 19,927 4,784 358,548 362,470 3,922 122.0 Boston MA 77,253 80,903 3,650 574,283 589,141 14,858 24.6 Atlanta GA 19,763 24,931 5,168 394,017 416,474 22,457 23.0 San Francisco CA 32,906 43,531 10,625 723,959 776,733 52,774 20.1 Chicago IL 56,048 72,843 16,795 2,783,726 2,896,016 112,290 15.0 Seattle WA 12,292 18,983 563,374 47,115 14.2 16,781 7,322,564 8,008,278 685,714 2.5 New York 6,691 153,927 170,708 516,259 Dallas TX 11,858 15,198 3,340 1,006,877 1,188,580 181,703 1.8 Los Angeles CA 34,655 36,630 1,975 3,485,398 3,694,820 209,422 0.9 Houston TX 7,029 7,565 536 1,630,553 1,953,631 323,078 0.2 Cities where downtown population growth reduced city population loss Downtown population 1990 2000 City population Increase 1990 2000 Decrease Downtown offset of city loss (%) 951,270 −76,704 1.0 736,014 651,154 −84,860 1.7 491 261,229 234,403 −26,816 1.8 27,667 1,070 606,900 572,059 −34,841 3.0 15,039 16,359 1,320 628,088 596,674 −31,114 4.1 74,686 78,349 3,663 1,585,577 1,517,550 −68,027 5.1 Cleveland OH 7,261 9,599 2,338 505,616 478,403 −27,213 7.9 New Orleans LA 6,988 8,051 1,063 496,938 484,674 −12,264 8.0 Pittsburgh PA 6,517 10,216 3,699 369,879 334,563 −35,316 9.5 Detroit MI 34,872 35,618 746 1,027,974 Baltimore MD 28,579 30,067 1,470 Norfolk VA 2,390 2,881 Washington DC 26,597 Milwaukee WI Philadelphia PA The global context of urbanisation and urban change Jackson MS 5,253 6,762 1,509 196,637 184,256 111 −12,381 10.9 Source: adapted from E.Birch (2002) Having a longer view on downtown living Journal of the American Planning Association 68(1), 5–21 COUNTERURBANISATION Signs of a population reversal in rural areas were first identified in the USA,18 but similar trends were soon detected in other advanced nations, including Canada,19 Australia,20 Western Europe21 and Britain.22 Counterurbanisation in Britain dates from the early 1960s, when for the first time areas situated well away from metropolitan influence began to grow faster than the main conurbations and their dependent regions. Population growth in rural Britain was particularly strong in the late 1960s and early 1970s but has continued over recent decades, with net out-migration from the main metropolitan areas to the rest of the UK averaging around 90,000 people per year, a rate of 0.5 per cent.23 The reasons for this reversal of long-established trends are so multifaceted that any attempt to apply a single explanation to the widely diverse changes under way in different regions would be unduly simplistic. Synthesising findings from a range of investigations provides a useful inventory of contributory factors. These include: 1. continuing growth of metropolitan centres and their spillover into adjacent nonmetropolitan counties; 2. decentralisation of manufacturing in pursuit of lower land and wage costs; 3. increased employment in service occupations; 4. early retirement coupled with higher retirement incomes not tied to a particular location; TABLE 4.11 CHANGING POPULATION AND EMPLOYMENT DISTRIBUTIONS IN THE USA, 1950–2000 Census year 1950 1960 1970 1980 1990 2000 Metro populationa 57 49 43 40 37 38 Metro employment 70 63 55 50 45 37 Metro population 43 51 57 60 63 62 Metro employment 30 37 45 50 55 63 Central cities as percentage of: Suburbs as percentage of: Source: US Census Bureau statistics Note: aMetro refers to the metropolitan statistical area (MSA) as defined by the US Census at each census date centred on amenity-rich areas outside the daily range of metropolitan commuting; Urban geography 112 5. increased per capita disposable real income; 6. increased pursuit of leisure activities at all ages, 7. increased enrolments in rural colleges and universities in the USA, especially in the late 1960s and early 1970s as a result of the post-war ‘baby boom’; 8. growth of state government in the USA; 9. levelling off of the loss-of-farm population; 10. growth of an anti-materialist perspective among the young; 11. narrowing of the traditional gap between urban and rural lifestyles with the extension of electricity, water and sewage systems, telecommunications and access to modern facilities; 12. more long-distance commuting; 13. growth associated with energy and extractive industries; 14. completion of the interstate highway system in the USA; 15. lower cost of living in rural areas; 16. growth of anti-urbanism as characterised by increased fear of crime and concern with urban disamenities such as congestion and pollution; 17. growth in importance of military establishments in some US counties during the 1960s; 18. residential preference for lower-density rural living; 19. government decentralisation policies in some countries such as Sweden, France and Britain (in the case of the last-named, via the New Towns programme; see Chapter 9). The list of factors is diverse and many are interrelated. The extent to which each contributes to the population turnaround will depend on local conditions. THE SUBURBANISATION WAVE The suburbanisation process began on a significant scale in the 1920s and accelerated after the Second World War, especially in North America. The USA is the world’s first predominantly suburban nation.24 As Table 4.11 shows, by the early 1960s the suburbs held 51 per cent of the US urban population, by 1980 they accounted for half of total metropolitan employment, and by 1990 about two-thirds of the metropolitan population and 55 per cent of metropolitan employment. By the last of these dates, US suburbia contained more than half the entire national population, having expanded from 41 million to 115 million since 1950 (an increase of 180 per cent). In 2000 the suburbs accommodated 140 million Americans (50 per cent of the total population). The suburban wave was driven by the following factors: 1. The rapid growth of urban population and rising disposable incomes enabled people to meet both the cost of new housing and the associated transport costs. 2. Widespread diffusion of the automobile enhanced individual mobility. The number of US automobiles rose from under 1 million in 1910 to 27 million by 1930, the latter amounting to one for every five persons. 3. New suburbs started to resist annexation by central cities through legal incorporation, which enabled them (and their residents) to shield themselves from the problems of the central city (such as low-quality housing, rising taxes, congestion, racial tension The global context of urbanisation and urban change 113 and crime), and to provide the particular living environment they desired and could pay for. 4. There was a huge pent-up demand for housing. 5. There was a need to generate employment after fifteen years of low investment during the Depression of the 1930s followed by the war years. 6. These goals were promoted by public policies that favoured new house-building over rehabilitation and highway construction over mass transit. Consequently, in the USA, the 1950s represented ‘the largest suburban decade ever’.25 In the UK the most ubiquitous form of new rural residential development has been the residential subdivision or housing estate located as an adjunct to an existing rural settlement within commuting distance of an urban workplace. These dormitory settlements, growing almost solely because of out-migration from central cities, have been termed metropolitan villages, defined as settlements where more than one in five of the workforce is employed in towns or cities.26 In the USA the suburbanisation process has created transit- and freeway-dependent dormitory subdivisions, infill developments and automobile-dependent dispersed suburbia. The functions of suburbs range from undifferentiated residential areas to a more recent mix of specialised retail corridors, high-technology industrial clusters, and high-density office and commercial nodes or ‘edge cities’27 (Box 4.6). The impact of suburbanisation on central cities has been profound. Most strikingly, in the 1960s several US cities, notably New York, became technically bankrupt and unable to finance their current expenditure on services. Rising local taxes and deteriorating local services merely served to accelerate the flight of betteroff residents and footloose firms into the suburbs, leaving behind the less dynamic economic sectors and less wealthy people, notably African-Americans and recent immigrants from overseas. Even in the more stable 1980s, when New York City’s population grew by 3.5 per cent, its white non-Hispanic population fell by 11.5 per cent and the proportion of its total residents accounted for by the ‘minority population’ rose to over 60 per cent in 1990. As Table 4.12 indicates, similar patterns of ‘white flight’ were BOX 4.6 Edge City, USA: Tyson’s Corner VA The post-Second World War movement of housing, industry and commerce to the outskirts of urban areas has created perimeter cities that are functionally independent of the urban core. In contrast to the residential or industrial suburbs of the past, these new cities contain along their superhighways all the specialised functions of a great metropolis: commerce, shopping malls, hospitals, universities, cultural centres and parks. This new peripheral urban form is referred to by various names, including technoburb, post-suburbia, cyberbia, stealth city or edge city. Driving time, not space, determines its fluid boundaries. Garreau (1988)28 defined an edge city as a place that has 5 million ft2 (460,000 m2) or more of leasable office space, 600,000 ft2 (56,000 m2) or more of leasable retail space, more jobs than bedrooms, is perceived by the population as one place, and that grew from practically nothing in the early 1960s. Tyson’s Corner, just beyond the Beltway around Washington DC, is the archetypal edge city In the mid 1960s Tyson’s Corner was a rural corner of northern Virginia Urban geography 114 marked only by the intersection of Interstate 66, the Washington Beltway and the access road to Dulles International Airport. Administratively it is still rural, an unincorporated 6,000 acre (2,400 ha) area that contains 30,000 residents and over 75,000 jobs, all under the jurisdiction of Fairfax County but split between three different county supervisory districts and three county planning districts. Tyson’s Corner does not exist as a postal address; residents’ mail must go to either McLean or Vienna. Within this framework in 1990 was the ninth largest concentration of commercial space in the USA, including more than 20 million ft2 (1.9 million m2) of office space, 3,000 hotel rooms and parking for more than 80,000 cars. The area was also the largest east-coast retail concentration outwith Manhattan. Yet it had little of the apparatus of urban governance or civic affairs. Source: J.Garreau (1991) Edge City New York: Random House recorded by other major cities of the USA. While the exodus of white-flight population from New York and Chicago slowed during the 1990s in other major cities, the rate of white out-migration from TABLE 4.12 POPULATION CHANGE IN US CENTRAL CITIES, 1960–2000 City 2000 population (000) Population change (%) White population change (%) Minority share of total population (%) 1960– 1970– 1980– 1990– 1980– 1990– 1990 70 80 90 2000 90 2000 2000 New York 8,008 1.4 −10.4 3.5 −9.4 −11.5 −6.6 60.5 55.3 Chicago IL 2,896 −4.7 −10.7 −7.4 −4.0 −17.2 −3.8 62.7 58.0 Philadelphia PA 1,518 −3.1 −13.5 −6.1 3.2 −12.3 −19.5 48.4 55.0 Detroit MI 951 −8.5 −19.2 −14.6 7.5 −47.7 −47.6 79.7 89.5 Baltimore MD 651 −2.8 −12.5 −6.4 11.5 −15.8 −28.4 61.6 69.4 Cleveland OH 478 −14.3 −23.6 11.9 5.5 −19.0 −20.7 52.5 58.5 Pittsburgh PA 335 −14.1 −18.5 −12.8 9.5 −14.9 −15.2 28.5 32.4 Cincinnati OH 231 −9.8 −15.0 −5.5 9.1 −11.2 −20.3 39.9 47.0 Source: US Census Bureau statistics Notes: Data refer to central cities only. Cities are ranked by 2000 population size. Hispanics are included in the minority population The global context of urbanisation and urban change 115 BOX 4.7 The changing ethnic composition of US central cities A combination of ‘white flight’ and an influx of Hispanic populations have left US whites as a minority in nearly half the 100 largest cities. According to the 2000 census non-Hispanic whites are now a majority in only fifty-two of the largest 100 cities, compared with seventy a decade earlier. Whites now account for 44 per cent of the 58,441,915 people in the largest 100 cities (compared with 52 per cent in 1990). Some 2.3 million whites or 8.5 per cent of the white urban population left the largest cities between 1990 and 2000. Birmingham AL lost 40 per cent of its white population. Other cities where whites have become a minority include Anaheim CA and Riverside CA where immigration from Mexico is particularly strong. Cities experiencing economic difficulties such as Rochester NY saw ‘white flight’ to the suburbs. The Hispanic population of the 100 largest cities grew from 17.2 per cent to 22.5 per cent, an increase of 3.8 million. The Asian population grew by 1 million, from 5.3 per cent to 6.6 per cent. The black population increased by 876,000 but as a proportion of the urban population fell slightly from 24.6 per cent to 24.0 per cent. the central core increased (Box 4.7). In Detroit the loss of almost half the white population of the central city during the 1990s matched the rate of the previous decade, with the result that nearly nine out often residents of central Detroit are from minority ethnic groups (Table 4.12). Although attempts have been made to classify suburbs, they are better viewed as dynamic entities with a diversity that reflects their role in the post-modern city. The diversity of the suburbanisation phenomenon is encapsulated by Bourne (1996)29 in a list of ten differing interpretations (Table 4.13). The first and more traditional interpretation views suburban development as a ‘natural’ process of accommodating growth by extension of the urban margin, and is characterised by the classical ecological models of the city.30 The second perspective sees the suburbs as an escape route from the social and environmental problems of cities either via individual decisions or through centralised planning initiatives.31 The third and fourth views are based on a structural or political economy perspective which interprets suburbanisation as a tool of government macroeconomic policy and a means of generating employment and promoting capital accumulation for the land development, building and financial sectors.32 The fifth characterises suburbanisation as social engineering, as a means of rescuing the poor from themselves and perhaps as an indirect means of inculcating an assumed superior moral order of the past.33 The sixth and seventh explanations are market-driven and derive from micro-economic theory and the capitalist logic Urban geography 116 TABLE 4.13 ALTERNATIVE INTERPRETATIONS OF THE SUBURBANISATION PROCESS 1. Suburbs as natural ecological extensions: suburbanisation as a natural process of organic and evolutionary growth; expansion takes place from the inside outward to the fringe, but is still tied to the urbanised core for jobs and services. 2. Suburbs as a means of escapism: as a means of escape from the health, housing and environmental problems of the industrial inner cities. 3. Suburbs as macro-economic policy tools: suburbs as Keynesian policy instruments of macroeconomic management and regulation, and for generating local employment multipliers. 4. Suburbs as vehicles for capital accumulation: as a means for landowners, the financial sector and the property industry to capture the social surplus, deriving from the profits from the development of newly built suburban environments on the fringe. 5. Suburbs as a means of social engineering: as a means of rescuing the poor and the disadvantaged from themselves, and of re-establishing a traditional and presumed superior moral order of earlier times and communities. 6. Suburbs as the logical outcomes of rational locators, reflecting the rational decisions of firms and households seeking lower-cost locations and more efficient and less regulated landscapes, within a competitive urban environment. 7. Suburbs as maps of consumer preferences and choices, emphasising the dominant role in suburbanisation of the preferences of individual consumers for more space, new housing, social homogeneity and certain public goods. 8. Suburbs as socio-political strategies: Strategies building on manipulation of the political fragmentation of the metropolis, entrenched local autonomy and the demands for social exclusiveness. 9. Suburbs as asylums: as defensive strategies, driven by fear of others, of the inner city and by uncertainty over property values, and stressing security and exclusion. 10. Suburbs as rural nostalgia, reflecting a desire to return to the countryside and rural roots, but without also severing Source: L.Bourne (1996) Reinventing the suburbs: old myths and new realities Progress in Planning 46(3), 163–84 of individual utility maximisation. One views suburban sprawl as the expected outcome of rational individuals, households and firms seeking more efficient and less costly environments; the other emphasises the dominant role of consumer preferences for more space, new housing, privacy and private consumption.34 The eighth sees suburban development as a socio-political strategy of exclusion designed to satisfy demands for local autonomy, social homogeneity and differential consumption of collective goods and services.35 The final two perspectives return to the view of suburbs as a defensive strategy that is driven either by fear of ‘others’ who are different and who may pose a threat to a preferred lifestyle, or by a desire to recapture an assumed simple rural way of life but without losing the advantages of urban living.36 Suburbs are open to all these The global context of urbanisation and urban change 117 interpretations, with the applicability of each ranging from place to place, and over time. Nowhere else is the postmodern message of difference, and the difficulty of generalisation, more relevant than in suburbia. EXURBANISATION The suburbanisation wave reaches its greatest extent in the phenomenon of extended suburbanisation or exurbanisation. Nelson (1992)37 identified four principal factors to explain exurbanisation: 1. continued deconcentration of employment and the rise of exurban industrialisation; 2. the latent anti-urban and rural location preferences of US households; 3. improved technology that makes exurban living possible; 4. an apparent policy bias favouring exurban development over compact development. These developments on the margins of suburbia represent a transition state between urban and rural life akin to the second-home phenomenon. Exurbia tends to be dominated by middle-class residents, many of whom commute long distances to work in the city or in the newer suburbs, but other groups are also present, including retirees and young households seeking social status, more land and new housing at a lower cost than is available in the suburbs. In the USA the exurbs have captured as much as one-quarter of recent national population growth and 60 per cent of recent manufacturing investment.38 For some this heralds a ‘post-suburban’ era characterised by inner suburban population loss and relative income decline, an increase in suburban employment, a reduction in suburban out-commuting, an increase in exurban population and income, and increased farmland conversion to urban use.39 The acid test for any model is how well it corresponds with reality. The first three stages of population change indicated by the model shown in Figure 4.8 accord well with the pattern of urban development in North America and Western Europe. Urbanisation followed by central city decline and suburban growth have been characteristic features of the US city for several decades, and national urban systems in Europe also appear to have followed the model sequence.40 There is, however, less evidence for the final stage of the model. This casts doubt on the hypothesised progression through all stages. Despite many examples of gentrification and ‘urban renaissance’ in cities of the MDRs, the weight of demographic evidence seems to indicate the continuing dominance of centrifugal trends within urban regions rather than a general shift into a reurbanisation stage.41 In contrast, as we have seen, urbanisation remains the dominant process in the LDRs. TYPES OF URBANISED REGIONS The increasing scale of urbanisation, urban growth and development of national urban systems has given rise to a number of different forms of urbanised regions: Urban geography 118 Plate 4.1 The exploding postmodern metropolis of Los Angeles CA with its restricted high-rise central business district surrounded by lowdensity suburbia reaching into the mountains 1. The city-region. This is an area focused on the major employment centre in a region and encompassing the surrounding areas, for which it acts as the primary high-order service centre. The functional relationship between a city and its region was a key feature of central place theory (see Chapter 6). The city-region remains an appropriate description of monocentred urban areas of up to a million people found in the less densely populated parts of even the most highly urbanised countries. Variants employed for statistical purposes include functional urban regions (FURs) and standard metropolitan statistical areas (SMSAs) (see Chapter 2). 2. Conurbation. This is the term coined in 1915 by Geddes to describe a built-up area created by the coalescence of once-separate urban settlements.42 With improvements in transportation and communications the functional influence of the conurbation has spread beyond the limits of the built-up area, so the term is now widely used in the UK and elsewhere to describe multi-nodal functional urban units. The functional relationships within a conurbation differ from those of a city-region; in essence, while there is a degree of dominance by the largest city, the other urban places also have their own functional linkages. 3. The urban field. This is a unit, similar to the conurbation, used in the USA. An urban field is generally regarded as a core urban area and hinterland of population at least 300,000, with an outer limit of two hours’ driving time. Defined in this manner, urban fields range in population size from 500,000 to 20 million and cover one-third of the USA and 90 per cent of the national population. Urban fields are more spatially extensive than European conurbations, since they are based on higher levels of personal mobility. The southern California ‘urban field’ extends 150 miles from north to south and includes Tijuana in Mexico (in the process creating a transnational city in which the largest ‘Mexican’ city is Los Angeles). The concept may become increasingly relevant for understanding the functional reality of urbanised regions outwith the USA as similar levels of mobility are achieved through improvements in The global context of urbanisation and urban change 119 transport and communications. The urban field is one form of polycentric urban region.43 A second is the polynucleated metropolitan region or megalopolis. 4. Megalopolis. This is the term introduced by Gottmann in 1961 to describe the urbanised areas of the north-eastern seaboard of the USA encompassing a population of 40 million oriented around the major cities of Boston, New York, Philadelphia, Baltimore and Washington DC.44 Gottmann subsequently defined a megalopolitan urban system as an urban unit with a minimum population of 25 million. The central importance of transactional activities (in terms of international trade, technology and culture) would indicate a location at a major international ‘breakpoint’ (such as a port city). A megalopolis would typically have a polynuclear form but with sufficient internal physical distinctness for each constituent city to be considered an urban system in its own right. The cohesiveness of the megalopolitan system depends on the existence of high-quality communications and transportation facilities.45 This megalopolitan phenomenon was identified initially in six zones: the archetype model of the North-Eastern USA, the Great Lakes area extending from Chicago to Detroit, the Tokaido area of Japan centred on Tokyo-Yokohama and extending west to include Osaka Kobe, the central belt of England running from London to Merseyside, the North-West European megalopolis focused on Figure 4.9 Megalopolises of the USA Amsterdam Paris Ruhr, and the area around Shanghai. Since then, twenty-six growth areas of the USA have exhibited megalopolitan patterns (Figure 4.9), while similar trends are evident in Brazil (between Rio de Janeiro and São Paulo), in China46 and in Europe47 (Figure 4.10). Urban geography 120 5. Ecumenopolis. This is the term employed by Doxiades in 1968 to describe a projected urbanised world or universal city by the end of the twenty-first century48 (Figure 4.11). Although highly speculative, the ecumenopolis concept does focus attention on the potential consequences of unrestrained urban growth and underlines the importance that is currently being attached to the concept of sustainable urban development (see Chapter 30). In the next chapter we switch our scale of analysis to provide a detailed examination of recent processes and patterns of urban change within the major regions of the world. Figure 4.10 The megalopolitan trend in Western Europe The global context of urbanisation and urban change 121 Figure 4.11 Doxiadis’s concept of ecumenopolis—an urbanised world at the end of the twenty-first century Source: C.Doxiadis (1968) Ekistics London: Hutchinson FURTHER READING BOOKS B.Berry (1976) Urbanization and Counter-urbanization Beverly Hills, CA: Sage A.Champion (2000) Urbanization, suburbanization, counterurbanization and reurbanization, in R. Paddison and W.Lever (eds) Handbook of Urban Studies Beverly Hills, CA: Sage W.Frey and A.Speare (1988) Regional and Metropolitan Growth and Decline in the United States New York: Russell Sage Foundation J.Garreau (1988) Edge City New York: Doubleday J.Gottmann (1961) Megalopolis: The Urbanized Northeastern Seaboard of the United States Cambridge, MA: MIT Press R.Harris and P.Larkham (1999) Changing Suburbs: Foundation, Form and Function London: Spon United Nations (1995) World Urbanization Prospects: The 1994 Revision New York: United Nations JOURNAL ARTICLES E.Birch (2002) Having a longer view on downtown living Journal of the American Planning Association 68(1), 5–21 Urban geography 122 L.Bourne (1996) Reinventing the suburbs: old myths and new realities Progress in Planning 46(3), 163–84 J.S.Davis, A.C.Nelson and K.J.Dueker (1994) The new ‘burbs: the exurbs and their implications for planning policy Journal of the American Planning Association 60, 45–59 J.Elliot (1997) Cycles within the system: metropolitanisation and internal migration in the US 1965–1990 Urban Studies 34(1), 21–41 A.Fielding (1989) Migration and urbanisation in Western Europe since 1950 Geographical Journal 155, 60–9 H.Geyer and T.Kontuly (1993) A theoretical foundation for the concept of differential urbanisation International Regional Science Review 15(12), 157–77 M.Jefferson (1939) The law of the primate city Geographical Review 29, 226–32 W.Lucy and D.Phillips (1997) The post-suburban era comes to Richmond: city decline, suburban transition, and exurban growth Landscape and Urban Planning 36, 259–75 A.Nucci and L.Long (1996) Spatial and demographic dynamics of metropolitan and nonmetropolitan territory in the United States International Journal of Population Geography 1(2), 165–81 M.Pacione (1980) Differential quality of life in a metropolitan village Transactions of the Institute of British Geographers 5(2), 185–206 N.Phelps and N.Parsons (2003) Edge urban geographies: notes from the margins of Europe’s capital cities Urban Studies 40(9), 1725–19 KEY CONCEPTS ■ urbanisation ■ urban growth ■ urbanism ■ megacities ■ million cities ■ urban bias ■ logistic-growth urbanisation curve ■ urbanisation cycle ■ primate city ■ polarisation reversal ■ demographic transition ■ differential urbanisation ■ suburbanisation ■ exurbanisation ■ counterurbanisation ■ reurbanisation ■ metropolitan village ■ edge city ■ overurbanisation ■ city region ■ conurbation ■ urban field ■ megalopolis ■ ecumenopolis The global context of urbanisation and urban change 123 STUDY QUESTIONS 1. Over the course of the second half of the twentieth century, a world in which most people lived in rural areas was transformed into a predominantly urban world. Examine the changing distribution of the world’s urban population over that period. 2. One of the most striking features of the global urban pattern is the degree to which the urban population live in giant cities. Identify and explain the distribution of the world’s megacities and ‘million cities’. 3. With reference to the theory of differential urbanisation, outline the main stages of the urbanisation cycle. 4. The concept of a cycle of urbanisation has also been employed to study growth patterns within individual urban agglomerations. Explain the four main stages in this model. 5. With the aid of particular examples, examine the evidence for urbanisation, counterurbanisation, suburbanisation or reurbanisation in any country with which you are familiar. 6. Identify the principal characteristics and functions of suburbs. How do these correspond to suburban areas in your locality? PROJECT Use census data to plot the population of your city over time. Identify and explain the significant trends. Use disaggregated data to determine whether certain parts of the metropolitan area have lost population, gained population or remained static since 1950. Explain any changes revealed by your analysis. 5 Regional Perspectives on Urbanisation and Urban Change Preview: urbanisation and urban change in world regions; North America; Latin America and the Caribbean; Western Europe; East and Central Europe; Asia and the Pacific; Africa INTRODUCTION This chapter will examine the diverse processes of urbanisation and urban change within the world’s major regions. In our discussion we shall examine the dynamics of recent urban change in six global regions. In each region we shall identify the main social, economic and demographic forces underlying urban change and the effects of these on the pattern of urban development at present and in the immediate future. NORTH AMERICA Between 1950 and 1990 the population of the USA increased by 65 per cent to reach 248.7 million; in Canada it increased by 95 per cent to reach 27.3 million. The proportion of the population living in urban centres grew from 64 per cent to 82 per cent in the USA and from 61 per cent to 77 per cent in Canada. Over the period, major changes occurred in the distribution of population, and in the relative importance of different cities. The dominance of the larger industrial cities and older eastern gateway ports diminished and complex new urban hierarchies emerged. Economic growth was accompanied by changes in methods of production, communication and transportation, modification of traditional occupational structures and of the spatial division of labour, and an increasingly uneven distribution of wealth, all of which influenced the structure and relative importance of regional and urban economies. DEMOGRAPHIC CHANGE The demographic basis of the post-Second World War urban transformation in North America is the outcome of two principal components: 1. a series of socio-demographic shifts and internal migration movements; 2. immigration. Regional perspectives on urbanisation and urban change 125 Both the USA and Canada experienced a ‘marriage and baby boom’ between 1948 and 1963. The resulting population expansion contributed to the rapid growth of the consumer economy, to increased housing demand and new residential construction, and to the evolution of new urban forms through widespread suburbanisation. Parallel to these demographic changes were equally dramatic changes in household and family composition, especially the proliferation and diversification of household types. Average household size declined from 4.0 persons to 2.7 persons between 1950 and 1970, and is even lower in metropolitan areas. In the USA over 20 per cent of all households have only one person and 13 per cent are female-headed single-parent households. Fewer than half of all households are now family households in the sense of having members related by blood or marriage. The impact of these shifts in living arrangements on metropolitan development include a substantially greater consumption of housing space, urban land and public resources per capita and per household than would otherwise be the case.1 As rates of natural increase in population declined after 1963 they also became more uniform across the continent. As a result, internal migration flows assumed greater importance as determinants of urban growth and population redistribution, and as a source of social change within localities. In any given year 18 per cent of North Americans change their place of residence, and after five years over 50 per cent have moved. With such high mobility levels, the potential for redistributing population and economic activity is high and this contributes to a degree of uncertainty regarding future settlement patterns. Another consequence of the declining rate of natural increase in population has been the growing importance of immigration as a factor in population change. During the 1970s and 1980s immigration flows represented 20 per cent of the USA’s and 30 per cent of Canada’s total population growth. By the early 1990s the USA was admitting 1.1 million official immigrants, and Canada 200,000 each year. The origins of immigrants have shifted away from long-established sources in Europe to countries in the South, especially Mexico and Latin America (for the USA) and Asia. In 2000, of the 28.4 million foreign-born people living in the USA (representing 10.4 per cent of the total population), 51 per cent were born in Latin America, 25.5 per cent in Asia, 15.3 per cent in Europe and the remaining 8.1 per cent in other regions of the world. The foreign-born population from Central America (including Mexico) accounted for nearly two-thirds of the total foreign-born population. As we saw in Chapter 4, nearly one in two of the foreign-born live in a central city in a metropolitan area (compared with 27.5 per cent of the native-born population). The focused residential geography of Latino and Asian peoples (for example in cities such as Miami and Los Angeles), along with the longstanding urban concentration of blacks and redistribution of whites from highimmigration metropolitan areas indicates an increasing ‘demographic balkanisation’ of the US population in particular regions and cities. Significantly for the social composition of cities, many of the new immigrant populations are visibly and culturally distinct from earlier European immigrants. REGIONAL AND URBAN CHANGE As we saw in Chapter 3, the system of metropolitan areas in the USA has expanded and become more complex in the post-war period. By 1990 there were 284 metropolitan areas Urban geography 126 with 193 million residents, or 77.5 per cent of the total US population. Half the US population now lives in large metropolises, with consolidated metropolises such as the New York-New Jersey-Connecticut CMSA being of megalopolitan proportions (Table 5.1). Figure 5.1 depicts the largest metropolitan areas in Canada and the USA. The impact of population mobility on urban patterns since the Second World War has exhibited regional variation, with certain regions and urban areas being prime destinations for inflows, whereas others have lost population. Most significantly, since migrants, on average, tend to be younger (excluding retirement migrations) and better educated than non-migrants, these population flows also transfer considerable wealth, market power and labour skills to the receiving places. They also shift the locations of future generations of population growth. During the 1980s the contribution of net migration (combining immigration and internal migration) to metropolitan population growth in the USA varied widely from a low of −8.9 per cent in Pittsburgh to +27.8 per cent in Phoenix (one of the world’s most rapidly growing cities in the decade). For internal migration only, the contributions ranged from −11.1 per cent for Chicago to +25.2 per cent in Tampa. In terms of international immigration, most migrants are attracted by job opportunities and follow earlier migration paths established by friends and kin. In Canada three-quarters of all recent immigrants have settled in the three largest metropolitan areas, namely Toronto, Montreal and Vancouver, while in the USA gateway cities such as New York, Los Angeles, San Francisco and the border cities of Texas and the South-West have attracted most of these flows. This concentration has transformed the social structure and ethno-cultural character of the recipient cities; in some cases, such as Los Angeles and Miami, the ‘new minority’ constitute a majority of the central city population. The characterisation of these changing urban growth patterns since 1950 in terms of polar extremes (in the USA between the sunbelt and frostbelt or between the sun and gunbelt of the southern states and the northern rustbelt) is generally useful but also potentially misleading, since growth has been uneven within both of these broad regions. We can shed more light on these trends by grouping metropolitan areas according to their TABLE 5.1 RANKING OF US METROPOLITAN AREAS BY POPULATION, 2000 Rank Metropolitan area Population Population (000) % change 1980–90 1990–2000 1. New York CMSA 21,200 3.4 9.6 2. Los Angeles CMSA 16,374 26.3 12.7 3. Chicago CMSA 9,158 1.5 11.1 4. Washington-Baltimore CMSA 7,608 16.1 13.1 5. San Francisco-Oakland-San José CMSA 7,039 15.4 12.6 6. Philadelphia CMSA 6,189 4.3 5.0 Regional perspectives on urbanisation and urban change 127 7. Boston CMSA 5,819 −2.0 10.9 8. Detroit CMSA 5,456 6.5 0.0 9. Dal las-Fort Worth CMSA 5,222 32.5 29.4 10. Houston CMSA 4,670 19.6 25.2 11. Atlanta MSA 4,112 33.0 38.9 12. Miami CMSA 3,876 20.7 21.4 13. Seattle MSA 3,555 23.2 19.7 14. Phoenix MSA 3,252 40.0 45.3 15. Cleveland-Akron CMSA 2,946 −2.6 3.0 16. Minneapolis-St Paul MSA 2,969 15.5 16.9 17. San Diego MSA 2,814 34.1 12.6 18. St Louis MSA 2,604 3.2 4.4 19. Tampa-St Petersburg MSA 2,396 28.1 15.9 20. Pittsburgh MSA 2,358 −6.8 −1.5 Source: US Census Bureau statistics Notes: A consolidated metropolitan statistical area (CMSA) includes two or more primary metropolitan statistical areas (i.e. two or more large cities), while a metropolitan statistical area (MSA) is centred on one large city Figure 5.1 Major metropolitan areas in North America Urban geography 128 Plate 5.1 The Banco Popular in New York’s Harlem district testifies to the growing multicultural identity of US cities main economic bases. Doing so identifies a number of factors underlying urban growth: 1. The fastest-growing urban areas in the USA in the 1980s and early 1990s were, typically, smaller metropolitan areas with economies based on retirement and recreational pursuits (e.g. Fort Pierce FL). The population of these communities grew by 46.9 per cent in the 1980s, compared with an average of 11.9 per cent for all metropolitan areas. Most of these places are in Florida, Arizona or Nevada.2 2. A second group, typically larger places, included the finance and service centres acting as ‘control and management’ centres for the national economy. US metropolitan areas in this category (e.g. New York) grew by 14.5 per cent during the 1980s. 3. Third, cities involved primarily in public administration as national, state or provincial capitals, or as military centres showed an average growth rate of 17.5 per cent in the 1980s. 4. Manufacturing-based cities grew by only 1.5 per cent on average while mining and resource communities declined by 2 per cent on average over the decade. 5. Other cities have grown as regional service centres (e.g. Dallas and Atlanta) or as hosts for expanding high-technology industries (e.g. Silicon Valley in San José, Software Valley near Salt Lake City, and Kanata near Ottawa). Analysis of the 2000 census also suggests that metropolitan areas in the USA may be distinguished by the degree to which they attracted or lost international and domestic migrants during the 1990s. The largest metropolitan areas gained the greatest number of migrants from abroad over the period 1995 2000 but lost most domestic migrants. As Regional perspectives on urbanisation and urban change 129 Table 5.2 shows, nine of the ten largest metropolitan areas adhered to this pattern. The New York metropolitan region gained almost a million migrants from abroad but at the same time lost 874,000 residents. The five-county Los Angeles region gained TABLE 5.2 NET DOMESTIC MIGRATION AND MOVERS FROM ABROAD, 1995–2000, FOR LARGEST USA METROPOLITAN AREAS Metropolitan area Total population 2000 Movers from abroadb Net domestic migrationa Ratec No. To central To cities suburbs New York–Northern New Jersey–Long Island, NY– NJ– CT–PA CMSA 21,199,865 −874,028 −44.4 614,057 389,602 Los Angeles–Riverside– Orange County, CA CMSA 16,373,645 −549,951 −36.8 324,013 375,560 Chicago–Gary–Kenosha, IL– IN–WI CMSA 9,157,540 −318,649 −37.6 172,597 150,422 Washington–Baltimore, DC– MD–VA–WV CMSA 7,608,070 −58,849 −8.6 65,837 234,429 San Francisco–Oakland–San José, CA CMSA 7,039,362 −206,670 −32.3 194,220 179,649 Philadelphia–Wilmington– Atlantic City, PA–NJ–DE– MD CMSA 6,188, 463 −83,539 −14.5 58,131 69,790 Boston–Worcester–Lawrence, MA–NH–ME–CTCMSA 5,810,100 −44,973 −8.5 99,790 93,708 Detroit–Ann Arbor–Flint, MI CMSA 5,456,428 −123,009 −24.2 36,179 72,796 Dallas–Fort Worth, TX CMSA 5,221,801 148,644 33.6 151,679 79,815 Houston–Galveston–Brazoria, TX CMSA 4,669,571 −14,377 −3.5 138,826 75,442 Atlanta, GA MSA 4,112,198 233,303 68.4 15,975 146,997 Miami–Fort Lauderdale, FL CMSA 3,876,380 −93,774 −27.4 60,493 239,412 Seattle–Tacoma–Bremerton, WA CMSA 3,554,760 39,945 12.6 47,001 75,765 Phoenix–Mesa, AZ MSA 3,251,876 245,159 93.6 104,609 30,408 Urban geography 130 Minneapolis–St Paul, MN–WI MSA 2,968,806 34,207 12.9 31,145 34,975 Cleveland–Akron, OH CMSA 2,945,831 −65,914 −23.7 13,969 22,288 San Diego, CA MSA 2,813,833 −6,108 −2.4 63,695 45,127 St. Louis, MO–IL MSA 2,603,607 −43,614 −17.9 13,915 21,432 Denver–Boulder–Greeley, CO CMSA 2,581,506 93,586 42.3 44,472 49,498 Tampa–St Petersburg– Clearwater, FL MSA 2,395,997 103,375 49.5 24,728 42,936 Source: Adapted from W.Frey (2003) Metropolitan Magnets for International and Domestic Migrants Washington, DC: Brookings Institution Notes: aA negative value for net migration or the net migration rate is indicative of net outmigration, meaning that more migrants left an area than entered it. Positive numbers reflect net in-migration to an area. b This category includes movers from foreign countries, as well as from Puerto Rico, US Island Areas and US minor outlying islands. c The net migration rate is based on an approximated 1995 population, which is the sum of people who reported living in the area in both 1995 and 2000 and those who reported living in the area in 1995 but now live elsewhere. The net migration rate divides net migration, inmigration minus out-migration, by the approximated 1995 population and multiples the result by 1,000. nearly 700,000 immigrants but lost 550,000 domestic migrants. For some cities, the pattern of large immigration gains and significant domestic migration losses is not new. Over several decades New York, an established immigrant port-of-entry city, has gained migrants from abroad even as it has Figure 5.2 Migration flows for selected US immigrant magnet metropolitan areas Source: W.Frey (2003) Metropolitan Magnets for International and Domestic Migrants Washington DC: Brookings Institution Regional perspectives on urbanisation and urban change 131 lost significant numbers of residents to other parts of the USA. In fact, the domestic migration losses sustained by New York from 1995 to 2000 were less severe than the city experienced in the late 1970s and late 1980s (Figure 5.2). In other cities, including Los Angeles and San Francisco, net domestic out-migration accelerated during the 1990s. This trend established these two west-coast immigrant ports of entry as ‘redistributors’ of population to fast-growing metropolitan areas in the interior USA; a role played by New York and Chicago in earlier decades. Two main explanations have been proposed for strong domestic out-migration from the largest immigration magnet metropolitan areas. One views the process as an emerging national version of the ‘white flight’ phenomenon that characterised local city-to-suburb movements, especially in the 1950s and 1960s. While this remains a factor underlying metropolitan population movements the ethnic composition of out-migration also reflects the over-all ethnic composition of the metropolitan area. Out-migrants from New York and Chicago are more likely to be white than those leaving Los Angeles or San Francisco (Figure 5.2). A second hypothesis relates out-migration to displacement of lower-skilled workers from immigrant magnet areas by new immigrants who compete disproportionately in lower-skilled labour markets. Evidence from cities, such as Los Angeles and San Francisco, suggests that highest rates of domestic out-migration involved adults without a college degree, while conversely in-migrants tended to be college graduates.3 Geographically, at the national scale the metropolitan areas that gained most inmigrants from elsewhere in the USA over the period 1995 2000 are located in the SouthEast and non-California West. Movement of new residents into these areas followed and fuelled the growth of new industries and expanding urban and suburban developments in cities like Phoenix, Atlanta and Las Vegas. These domestic migration magnets also attracted new immigrant populations, in response to demands for low-skilled labour. Within metropolitan areas immigrants have boosted urban core populations in cities such as New York, San Francisco, Washington DC and Boston (Table 5.2), offsetting their losses of domestic migrants to the suburbs and elsewhere, and contributing to a degree of reurbanisation (see Chapter 4). At the same time, core counties in the Midwest and rustbelt continue to sustain some of the greatest domestic out-migration. The city of St Louis lost 105,000 domestic migrants over the period 1995–2000 but received fewer than 12,000 migrants from abroad. Some cities have considered strategies to attract inmigrants to reinvigorate declining urban neighbourhoods. While not all migrants from abroad head primarily for core urban counties, in general population gains in peripheral counties occur almost entirely from domestic migration. The disparity between growing suburban and exurban counties, where domestic migration dominates, and inner metropolitan counties dependent on migration from abroad to offset population decline is characteristic of both domestic migrant magnet metropolises, such as Atlanta and Denver, and immigrant magnet metropolises, such as New York and Washington DC. It is likely that future patterns of urban growth and change in North America will continue to reflect the high levels of population and economic mobility. Extrapolation of trends evident in the 1990s suggests that among the most significant factors underlying the future metropolitan pattern of North America are immigration, an ageing population, Urban geography 132 lifestyle changes, the uncertain growth in income, revisions in government welfare policies, technological innovations, trade liberalisation and increased global competition. LATIN AMERICA AND THE CARIBBEAN In 1990 the population of Latin America and the Caribbean totalled 440 million, having doubled in size since 1960. Over this period the region shifted from being predominantly rural to predominantly urban. With 71.4 per cent living in urban areas in 1990, this represented a level of urbanisation comparable to that of Europe. There is also a heavy concentration of population in large cities, with, in 1990, more people living in ‘million cities’ in the region than in rural areas.4 By 1990 most countries with more than a million inhabitants had more than half their population in urban areas. By 2001 these urbanisation trends had strengthened, with more than three-quarters of the region’s population living in cities (Table 5.3). The rapidity of urbanisation in the region is related to the scale of TABLE 5.3 LATIN AMERICA: URBAN POPULATIONS FOR 1990 AND 2001 AND URBAN CHANGE SINCE 1950 Country Urban population (000) % urban 1990 1950 1990 2001 1950– 90 2001 Change % urban 1990– 2001 Caribbean Cuba 7,801 8,482 49.4 73.6 75.5 24.2 1.9 Dominican Republic 4,293 5,615 23.7 60.4 66.0 36.7 5.6 Haiti 1,855 3,004 12.2 28.6 36.3 16.4 8.7 Jamaica 1,217 1,470 26.8 51.5 56.6 24.7 5.1 Puerto Rico 2,518 2,987 40.6 71.3 75.6 30.7 4.3 854 969 63.9 69.1 74.5 5.2 5.4 Costa Rica 1,429 2,448 33.5 47.1 47.8 13.6 59.5 El Salvador 2,269 3,935 36.5 43.9 61.5 7.4 17.6 Guatemala 3,628 4,688 29.5 39.4 39.9 9.9 0.5 Honduras 1,985 3,351 17.6 40.7 53.7 23.1 13.0 61,335 74,846 42.7 72.6 74.6 29.9 2.0 2,197 2,943 34.9 59.8 56.5 24.9 −3.3 Trinidad & Tobago Central America Mexico Nicaragua Regional perspectives on urbanisation and urban change Panama 133 1,240 1,639 35.8 51.7 56.5 15.9 4.8 28,158 33,119 65.3 86.5 88.3 21.2 1.8 3,665 5,358 37.8 55.8 62.9 18.0 7.1 Brazil 110,789 141,041 36.0 74.6 81.7 38.6 7.1 Chile 10,954 13,254 58.4 83.3 86.1 24.9 2.8 Colombia 22,604 32,319 37.1 70.0 75.5 32.9 5.5 Ecuador 5,625 8,171 28.3 54.8 63.4 26.5 8.6 Paraguay 2,109 3,194 34.6 48.9 56.7 14.3 7.8 15,068 19,084 35.5 69.8 73.1 34.3 3.3 2,751 3,097 78.0 88.9 92.1 10.9 3.3 17,636 21,475 53.2 90.4 87.2 37.2 3.2 314,161 399,269 41.6 71.4 75.8 29.8 4.4 South America Argentina Bolivia Peru Uruguay Venezuela Latin America and the Caribbean (total) Source: adapted from United Nations (2002) World Urbanization Prospects: The 2001 Revision New York: United Nations economic growth, with Brazil, Mexico, Colombia and the Dominican Republic experiencing both rapid economic growth and high levels of urbanisation. In terms of the level of urbanisation we can identify three groups of nations in the region: 1. The most urbanised, with over 80 per cent of their population in urban areas, includes Venezuela and the three Southern Cone nations of Argentina, Chile and Uruguay with a long tradition of urban development based on rapid immigration from Europe in the late nineteenth and early twentieth centuries, and a land-ownership structure which provided little opportunity for immigrants to acquire farmland, thereby ensuring their settlement in the cities. 2. The second, with between 50 per cent and 80 per cent in urban areas, includes most of the countries that experienced rapid urban and industrial development between 1950 and 1990—Brazil, Mexico, Colombia, Ecuador, and the Dominican Republic—as well as others such as Bolivia, Peru, and Trinidad and Tobago (Table 5.3). 3. The third, with less than 50 per cent of the population in urban areas, comprises the less populous countries such as Paraguay, Haiti and Costa Rica. By 1990 the region had 300 million urban inhabitants and thirty-six ‘million cities’, including three with more than 10 million (Figure 5.3). The region also had two of the world’s five largest cities (São Paulo and Mexico City) and in total eight of the world’s fifty largest cities. São Paulo is the largest city, reflecting its dominant economic role within the region’s largest economy, with Mexico City second largest. While the region’s urban population increased more than twentyfold between 1900 and 1990, most of the Urban geography 134 major urban centres of today were founded by the Spanish and Portuguese, with some, such as Mexico City and Quito, being even older pre-Colombian cities.5 However, in Brazil and Mexico there was a reordering of the urban system in the twentieth century during which many small cities emerged as prominent urban nodes. In Brazil, Brasilia was a political creation in 1958, while Porto Alegre was an unimportant town in 1800. Belo Horizonte was created as a new city at the end of the nineteenth century, while neither Fortaleza nor Curitiba was an important town before they became state capitals in the nineteenth century. In Mexico, too, the restructuring of the urban hierarchy has created new cities away from long-established centres of economic and political power (Box 5.1). This reordering of the urban system is more comparable to that in the USA than to what has happened in other countries in the region. Restructuring may also affect the internal arrangement of cities. In general, all major metropolitan areas experience a decentralisation of population and of production as they grow. As we have seen, this process has gone furthest in the USA, where decentralisation within a core region is commonplace, but it is also evident in some parts of Latin America. In the Mexico City region mega-urbanisation is linked to a process of territorial restructuring that is creating a polycentric extended metropolitan region. The expanded urban system and integrated road network have stimulated flows of people, commodities and capital among the constituent urban centres, and have tended to break down the differences between rural and urban activities.6 SMALL AND INTERMEDIATE URBAN CENTRES Notwithstanding the number of ‘million cities’ in most Latin American countries, a considerable proportion of the urban population live in urban centres other than the large cities. The 1991 census in Argentina showed that 46.5 per cent of the national population lived in urban centres with fewer than 1 million inhabitants, including 18 per cent in urban places with fewer than 100,000. There are great contrasts in the size, growth rate and economic base among the urban centres with fewer than 1 million inhabitants. These include thousands of small market towns and service centres with a few thousand residents, as well as some of the region’s most prosperous and rapidly growing cities over the past twenty years. The latter include: 1. cities that served agricultural areas producing high-value crops for export (e.g. Zamora in Mexico). The pattern of land-ownership is a major influence on the extent to which high-value crop production stimulates local urban development. Whereas a large number of small but prosperous farmers with intensive Regional perspectives on urbanisation and urban change 135 Figure 5.3 Major urban centres in Latin America and the Caribbean production can promote local urban development, large land holdings or plantations divert much of the economic stimulus to more distant cities, including those overseas; 2. cities that attract substantial numbers of tourists. For example, Cuautla in Mexico grew from being a small market town with 18,000 inhabitants in 1940 to a city of 120,000 in 1991 as a result of the growing importance of tourism. Urban geography 136 It is difficult to predict the scale and nature of BOX 5.1 Urban restructuring and the effect of the Mexico–USA border The distribution of population and of urban population in Mexico has been changed significantly by the economic interaction between settlements in its north and the USA. The increasing population concentration in the north of Mexico is strongly associated with the development of the maquila industries there. This increase in population was strongly concentrated close to the border; the population in the thirty-six municipalities that adjoin the USA grew fourteenfold between 1930 and 1990, from 0.28 million inhabitants in 1930 to 2.9 million in 1980 and close to 4 million by 1990. These industries originated in the mid-1960s, when the government of Mexico began a programme to promote industrial development in the border area. This permitted Mexican and foreign-owned factories within the border area to import machinery, materials and components without paying tariffs as long as the goods produced were re-exported. Most of the maquila industries that developed were owned by US companies, and they could also take advantage of the US tariff regulations. In 1967 there were fewer than 100 plants with around 4,000 workers; by the early 1990s there were 2,000, employing more than half a million people. The devaluation of the Mexican currency against the US dollar in the 1980s also boosted industrial and agricultural exports. Since the early 1970s maquila industries have been allowed to set up in the interior and since 1989 to sell products in the domestic market. During the 1980s some cities away from the frontier were attracting major investments—for instance, new motor vehicle export plants were set up in Chihuahua, Hermosillo and Saltillo. The table lists the border towns that to date have attracted most maquila industries and the ‘twin city’ in the United States with which they are connected. Of these, Ciudad Juárez and Tijuana concentrated most maquila industries in 1990. Urban centre Population (000) Compound growth rates (%) Partner city in USA c.1900 1950 1990 1900– 1930– 1950– 1960– 1970– 1980– 30 50 60 70 80 90 Ciudad Juárez 8.2 124.0 807.0 5.4 5.9 7.8 5.3 2.2 4.0 El Paso Tijuana 0.2 60.0 755.0 12.6 10.3 9.8 8.3 2.4 5.8 San Diego Nuevo Laredo 6.5 57.7 4.1 5.0 4.9 5.1 2.8 Reynosa 1.9 34.1 296.0 3.1 10.3 8.1 6.6 3.3 4.3 McAllen Matamoros 8.3 45.8 319.0 0.5 8.1 7.3 4.3 3.0 5.4 Brownsville Mexicali 00 65 7 637 0 77 10 6 44 21 6 4 Calexico Laredo (and to San Antonio, Houston) Regional perspectives on urbanisation and urban change 137 Source: United Nations Centre for Human Settlements (1996) An Urbanising World Oxford: Oxford University Press urban change in Latin America given that it is so dependent on economic performance. For countries that sustain rapid economic growth, urbanisation is likely to continue. Whatever the economic performance, however, the dominance of the region’s largest metropolitan areas is likely to decline as a result of the emergence of new cities with a comparative advantage. These include important centres of tourism, and cities well located to attract new investment in export-oriented manufacturing or to benefit from forward and backward linkages from high-value export agriculture. In many of the higher-income countries in Latin America the factors that stimulated urban decentralisation in North America and Europe, including the development of goodquality inter-regional transport and communications systems, will also exert an influence on urban structure. WESTERN EUROPE Western Europe is highly urbanised, with over 75 per cent of the population resident in urban areas (Table 5.4). Consequently, between 1980 and 1995 the region exhibited one of the smallest increases in the level of urbanisation of all major world regions. As in North America, changes in household size and composition and in age structure have had an important influence on housing markets and TABLE 5.4 URBAN POPULATIONS AND LEVELS OF URBANISATION FOR ALL WEST EUROPEAN COUNTRIES WITH 1 MILLION OR MORE INHABITANTS Country Urban population (000) 1990 % Urban 2001 1990 2001 Northern Europe Denmark 4,357 4,538 84.8 85.1 Estonia 1,131 955 71.8 69.4 Finland 3,063 3,031 61.4 58.5 Ireland 1,993 2,276 56.9 59.3 Latvia 1,902 1,437 71.2 59.8 Lithuania 2,552 2,532 68.8 68.6 Norway 3,068 3,365 69.3 75.0 Sweden 7,112 7,358 83.1 83.3 Urban geography United Kingdom 138 51,136 53,313 89.1 89.5 6,413 6,408 62.6 60.3 38,050 38,565 66.7 67.1 3,303 6,601 33.5 65.8 29,592 31,073 75.4 77.8 Austria 4,267 5,444 55.4 67.4 Belgium 9,606 9,997 96.5 97.4 France 41,218 44,903 72.7 75.5 Germany 67,699 71,948 85.3 87.7 The Netherlands 13,262 14,272 88.7 89.6 4,070 4,826 59.5 67.3 294,889 314,161 76.6 71.7 Southern Europe Greece Italy Portugal Spain Western Europe Switzerland Total (including all countries) Source: adapted from United Nations (2002) World Urbanization Prospects: The 2001 Revision New York: United Nations settlement patterns. Among the most significant economic changes were the rapid growth in the financial-services sector, which boosted the economy of certain key cities, and the continued decline in most cities that had traditionally specialised in heavy industry and port-based activities. DEMOGRAPHIC CHANGES Northern, Southern and Western Europe have long had the slowest population growth rates of any of the world’s regions, and there has been a steady decline in annual average population growth rates over recent decades. The twelve countries that made up the European Union in 1994 saw their annual average growth rate of 0.96 per cent in 1960 5 fall to 0.71 per cent, 0.61 per cent, 0.37 per cent and 0.27 per cent in the next four fiveyear periods. The slight increase to 0.34 per cent for the period 1985–90 was due entirely to net immigration, which rose from 0.5 per thousand in 1985–90 to 1.6 per thousand in 1985–90.7 The very low population growth rates help explain why it is common for cities to experience a net loss of population, since a relatively low level of net out-migration can still exceed the rate of natural increase.8 Important changes have also occurred in relation to household size and composition, in divorce and cohabitation, and in the level and incidence of childbearing. In much of Europe the average household size is now below 3.0 persons, and more women are starting their family later in life. In addition, over a quarter of households contain only one person, at least 10 per cent of families are headed by a lone parent, and around one in three marriages end in divorce. These trends indicate an increase in the number of Regional perspectives on urbanisation and urban change 139 households if not in population, with significant implications for housing markets, migration and settlement structures. The influence of demographic changes on cities and urban systems is seen clearly in the way the ‘baby boom’ of the late 1940s and early 1950s was associated with strong suburbanisation and deconcentration pressures as couples reaching the family-building stage sought affordable homes with gardens and a safe and pleasant environment within commuting distance of work. Similarly, as the population ages, so retirement migration becomes increasingly important. This generally takes the form of population movement away from larger metropolitan centres down the urban hierarchy, taking advantage of the lower house prices and more congenial living environments of smaller towns and more rural areas. However, the preferred location for such migration has changed from seaside and spa towns in the 1950s to countryside areas in the 1960s and 1970s, and to the Mediterranean sunbelt in the 1980s. EUROPE’S CITIES London and Paris are the largest cities in the region, after which the size ranking becomes more ambiguous depending on the boundaries used (Figure 5.4). Comparing cities by their dominant characteristics provides greater insight into current urban processes. Table 5.5 presents eleven types of cities and their dominant characteristics, with examples of cities that represent each functional type. This highlights the major dichotomy between the growing TABLE 5.5 FUNCTIONAL TYPES OF CITIES IN WESTERN EUROPE City type Characteristics Examples Global Accumulation of financial, economic, political and cultural headquarters of global importance London, Paris Growing hightech/services Modern industrial base, national centre of R&D, production-oriented services of international importance Bristol, Reading, Munich Declining industrial Traditional (monostructured) industrial base, obsolete physical infrastructure, structural unemployment Metz, Oberhausen, Mons, Sheffield Port Declining shipbuilding and ship-repair industries, environmental legacies, in the South burdened by additional gateway functions Liverpool, Genoa, Marseille Growing without modern industrialisation Large informal economy and marginal underclass, uncontrolled development and deteriorating environment Palermo, Thessaloniki, Naples Company towns Local economy depending to a high degree on a single corporation Leverkusen, Eindhoven New towns New self-contained cities with overspill population Milton Keynes, Urban geography 140 in the hinterland of large urban agglomerations Evry Monofunctional satellites New urban schemes within large agglomerations with focus on one function only (e.g. technopole, airport) Sophia-Antipolis, Roissy Small towns, rural centres, rurban belts Smaller cities and semi-urban areas in rural regions, All over Europe along coasts or transport corridors with weak economic potential Tourism and culture Local economic base depending on international Salzburg, Venice tourism and cultural events of European importance Border and gateway Hinterland divided by national border; gateways for Aachen, Basel economic migrants and political refugees Source: K.Kunzmann and M.Wegener (1991) The Pattern of Urbanization in Western Europe 1960–1990 Dortmund: Institute of Town Planning Figure 5.4 Major urban centres in Western Europe Regional perspectives on urbanisation and urban change 141 high-tech/services cities, which include many cities that are not among the largest in Europe, and the declining industrial and port cities, many of which still are among the largest cities although most have declining populations. There is little agreement about the likely future pattern of urban growth in Western Europe, not least because of a lack of knowledge of how the factors that affect population distribution will themselves develop. The advent of a post-Fordist economy embracing ‘flexible specialisation’ has induced major structural change in urban economies, while political changes associated with the European Union and removal of the Iron Curtain have the potential for large-scale economic restructuring. A major debate is whether these changes will lead to the growth of a European core region at the expense of the peripheral zones, with economic activity concentrated in a belt stretching from South-East England through Benelux, South-West Germany and Switzerland to Lombardy and North-West Italy. However, not all places within a favoured zone prosper and not all places outside a favoured zone are necessarily at a disadvantage. Research at the ‘locality’ level suggests that the economic fortunes of different places have more to do with their inherent characteristics and the way in which they position themselves with respect to the global economy and international capital than with their particular geographic location.9 This perspective sees a Europe of the future made up of a large number of individual, largely urban-centred regions that compete for jobs, people and capital investment, sometimes in cooperation with their neighbours but usually in competition. EAST AND CENTRAL EUROPE Until the fall of the Iron Curtain in 1989 the countries of Eastern and Central Europe and the former Soviet Union were united by economic and political characteristics, distinct from those of Western Europe, that affected their settlement system and the form of their cities.10 These included the fact that government direction rather than market forces determined the nature and location of most productive investment. In general, priority was given to industry over services, and in many instances industries were located outside the major cities in places that differed from those locations a market economy would have produced. Also, industries were kept in operation long after they would have been deemed unprofitable in the West, so that until the economic and political changes of the late 1980s most industrial centres had avoided the negative impacts of deindustrialisation. The abolition of an urban land market under communism, the limited role permitted for a private housing market and private enterprises, and the largescale public housing estates produced a very different logic to the form and spatial distribution of urban areas (see Chapter 8). Since 1989, however, there has been a transformation in the economies of Eastern Europe towards more market-oriented growth. This is likely to bring major changes to settlement systems through changes in the scale, nature and spatial distribution of economic activities.11 Urban geography 142 INTERNAL MIGRATION AND URBANISATION IN EAST AND CENTRAL EUROPE Internal migration has generally been from poorer to richer regions and from rural to urban areas.12 In the period up to 1989 a significant part of this migration was attributed to official government policies that promoted the growth of large towns and cities to provide the necessary industrial labour force. At the same time, communist planners also sought to develop urban centres lower down the size hierarchy by means of new town development, often in relation to a particular economic function (such as coal-mining), and by imposition of controls on the growth of the largest cities (e.g. Budapest). As Table 5.6 and Figure 5.5 indicate, many of the region’s main population concentrations are around national capitals and major industrial centres. ECONOMIC AND URBAN CHANGE IN THE FORMER SOVIET REPUBLICS Throughout the Soviet era there was a strong relationship between industrial and urban growth, and the central government’s industrial development policies had a major impact on population and urban change. After 1953 the post-Stalin leadership continued the emphasis on heavy industry and the military sector. The priority attached to industrial modernisation and to the introduction of plastics TABLE 5.6 URBAN POPULATION IN EASTERN EUROPE IN 1992 AND 2001, AND URBAN CHANGE SINCE THE 1930s Country Urban population (million) 1992 Albania 2001 1.2 Pre-war 1.4 Bulgaria 5.9 5.3 Czech Republic 7.8 7.7 Slovakia 3.7 3.1 Hungary Poland Romania 6.1 23.1 % population urban 6.4 24.1 1992 35.0 42.9 b 66.0 67.4 75.2 74.5 15.4 21.4 69.2 57.6 c 59.0 64.8 d 60.9 62.5 c 54.4 55.2 33.2 37.3 2001 a 12.3 12.4 21.4 Bosnia/Herzegovina 1.5 1.8 34.2 43.4 Croatia 2.4 2.7 50.8 58.1 Macedonia 1.1 1.2 53.9 59.4 Slovenia 1.0 0.9 48.9 49.1 Regional perspectives on urbanisation and urban change 143 Montenegro 0.3 50.7 Serbia 4.6 46.5 Yugoslavia 4.9 5.5 48.6 52.0 71.1 72.5 56.4 60.0 Eastern Europe Source: adapted from United Nations (2002) World Urbanization Prospects: The 2001 Revision New York: United Nations Notes: a1938. b1934. c1930. d1939 Figure 5.5 Major urban centres in Eastern and Central Europe and chemical industries led to investment in oil and natural gas production, which explains why several of the region’s rapidly growing cities in the period 1959–89 were in the Volga-Urals area. These include cities such as Novocheboksarsk that had not existed in 1959.13 Although the exploitation of oil and gas stimulated urban and industrial developments in peripheral producing regions, such as Siberia and Kazakhstan, the growing network of oil and gas pipelines and electricity grids also permitted the continuing expansion of the major urban industrial centres in the western parts of the country. The concentration of urban development in European Russia was also promoted by the numerous environmental and social advantages over the east and the higher level of infrastructure provision. An additional factor was the increasing importance of trade links with Eastern Europe (through COMECON) and Western Europe, especially after 1970 (Table 5.7). Urban geography 144 TABLE 5.7 LEVELS OF URBANISATION OF THE FORMER SOVIET REPUBLICS 1959, 1989 AND 2001 (%) Republic Level of urbanisation 1959 1989 Change 2001 1959–89 1989–2001 Baltic Republics Estonia 56.5 71.6 69.4 15 −2.2 Latvia 56.1 71.7 59.8 15 −11.9 Lithuania 38.6 68.0 68.6 29 0.6 Belarus 30.8 65.5 69.6 35 4.1 Ukraine 45.7 66.4 68.0 21 1.6 Moldova 22.3 47.0 41.4 25 −5.6 Russian Federation 52.4 73.6 72.9 21 −0.7 Georgia 42.4 55.8 56.5 13 0.7 Azerbaijan 47.8 53.9 51.8 6 −2.1 Armenia 50.0 67.8 67.2 18 −0.6 Kazakhstan 43.7 57.2 55.8 14 −1.4 Uzbekistan 33.6 40.7 36.6 7 −4.1 Kyrgyzstan 33.7 38.3 34.3 5 −4.0 Tajikistan 32.6 32.6 27.7 0 −4.9 Turkmenistan 46.2 45.4 44.9 −1 −0.5 Total (former USSR) 47.9 65.9 55.0 18 −10.9 Republics bordering Europe Caucasus Central Asia Source: adapted from United Nations (2002) World Urbanization Prospects: The 2001 Revision New York: United Nations An increasing proportion of the urban population was concentrated in large cities. By 1989 61 per cent of the region’s urban dwellers lived in cities with 100,000 or more inhabitants (compared with 49 per cent in 1959), with 22 per cent living in ‘million cities’ (compared with 9 per cent in 1959). This illustrates the failure of government attempts to restrict the growth of the largest cities by means of controls on internal migration. Large cities played a key role in the command economy for two reasons. First, they had the highest-quality infrastructure, skilled labour forces, and relatively low wages. Unlike in a market economy, in the Soviet command economy wage rates, taxes, land costs, rent and Regional perspectives on urbanisation and urban change 145 infrastructure costs were not necessarily higher in the major cities, as they would be normally in Western Europe or North America. Second, in a command economy the fact that success depended as much on access to political and administrative decision-makers and government officials as economic performance enhanced the attraction of major administrative centres. The transition to a market economy has been accompanied by less central direction of urban development and reduced levels of urbanisation as a consequence of emigration to the West and the beginnings of a suburbanisation process in some cities.14 Looking ahead, the fundamental political and economic changes heralded by the break-up of the Soviet Union are likely to have far-reaching implications for population and urban change in the region. Urban growth will be promoted by the emergence of capital cities and national urban systems in the new republics. Cities in general may benefit from an expansion of service activity restrained by Soviet economic-development policy. On the negative side, cities associated closely with the heavy industrialisation of the Soviet era (e.g. Donetsk in Ukraine) may experience the deindustrialisation and outmigration that have affected similar cities and regions in North America and Western Europe. Smokestack industries were so widespread within the command economy that deindustrialisation represents a major challenge for the new independent states. Similarly, policies to reduce military expenditure and restructure the military-industrial complex will also impact upon the economy of established industrial regions and lead to a restructuring of the urban system. ASIA AND THE PACIFIC In 1990 Asia contained three-fifths (3,186 million) of the world’s population and an increasing share (32 per cent in 1990) of its urban population. In 2001 Asia had 58 per cent of the world’s population and 46 per cent of its urban population (Table 5.8). It also contains many of the world’s largest and fastest-growing cities, reflecting the presence of most of the nations with the highest economic growth rates since 1980. However, few generalisations are valid for the region, given the variety of countries, which range from the richest in the world to the poorest, and from the largest and most populous to the smallest and least populous (Figure 5.6). With almost two-thirds of the population still living in rural areas, Asia remains a predominantly rural continent. Significantly, however, this proportion of urban to rural population is partly explained by the criteria employed to define urban places. In Asia, if India and China were to alter their definitions of ‘urban centres’ to those commonly used in many European or Latin American nations the level of urbanisation would increase to 50–60 per cent as hundreds of millions of those now classified as rural dwellers became urban. We can identify three main groups of countries in the Asia-Pacific region according to their levels of urbanisation: Urban geography 146 TABLE 5.8 POPULATION AND URBAN CHANGE, 1950–2001, IN ASIAN COUNTRIES WITH 10 MILLION OR MORE INHABITANTS Country Urban % of population in urban Change in level of population, 2001 areas (level of urbanisation) urbanisation (%) (000) 1950 1990 2001 1950–90 1990– 2001 Eastern Asia China 471,927 11.0 26.2 36.7 15.2 11.5 13,571 31.0 59.8 60.5 28.8 0.7 100,469 50.3 77.2 78.9 26.9 1.7 38,830 21.4 73.8 82.5 52.5 8.9 Afghanistan 5,019 5.8 18.2 22.3 12.4 4.1 Bangladesh 35,896 4.2 15.7 25.6 11.4 9.9 India 285,608 17.3 25.5 27.9 8.3 2.4 Iran 46,204 27.0 56.3 64.7 29.3 8.4 2,874 2.3 10.9 12.2 8.6 1.3 Pakistan 48,425 17.5 32.0 33.4 14.5 1.4 Sri Lanka 4,409 14.4 21.4 23.1 7.0 1.7 Indonesia 90,356 12.4 30.6 42.1 18.2 11.5 Malaysia 13,154 20.4 49.8 58.1 29.4 8.3 Myanmar 13,606 16.2 24.8 28.1 8.6 3.3 Philippines 45,812 27.1 48.8 59.4 21.7 10.6 Thailand 12,709 10.5 18.7 20.0 8.2 1.3 Vietnam 19,395 11.6 19.9 24.5 8.2 5.2 Iraq 15,907 35.1 71.8 67.4 36.7 −4.4 Saudi Arabia 18,229 15.9 77.3 86.7 61.4 9.4 8,596 30.6 50.2 51.8 19.6 1.6 Democratic People’s Republic of Korea Japan Republic of Korea South-Central Asia Nepal South-East Asia Western Asia Syria Regional perspectives on urbanisation and urban change 147 Turkey 44,755 21.3 60.9 66.2 39.6 5.3 Yemen 4,778 5.8 28.9 25.0 23.1 −3.9 1,340,529 18.6 31.2 37.4 12.6 6.2 Total Source: adapted from United Nations (2002) World Urbanization Prospects: The 2001 Revision New York: United Nations Figure 5.6 Major urban centres in Asia 1. The most urbanised group includes Australia, New Zealand, Japan, Hong Kong, Singapore and Korea. In these countries agriculture plays a minor role in the economy and more than half of GDP comes from the service sector. 2. The second group comprises Thailand, Indonesia, Malaysia, the Philippines, Fiji and Pakistan, where agriculture contributed less than one-third of GDP and where (apart from Thailand) between 30 per cent and 50 per cent of the population lives in urban areas. 3. The third group contains China and all South Asian countries except Pakistan. These remain predominantly rural, with agriculture having a greater importance in their GDP. Such classifications are purely indicative, however, and are susceptible to change over time as, for example, the speed of economic growth in China propels the country into the second group. In addition, parts of India and China currently exhibit the economic and urban characteristics of group 2 or even group 1 nations (Box 5.2). Urban geography 148 BOX 5.2 The economic and urban development of Shanghai In 1949 Shanghai was the country’s pre-eminent city, with over half of China’s modern industry. But after 1949, throughout the first phase of heavy industrialisation, it was deprived of significant investment funds because of its location, which was vulnerable to attack from the sea. Its population remained remarkably low, given that it had been the largest industrial centre in the world’s most populous nation. In 1982 the city’s urban core districts contained only 6.27 million inhabitants, with the rest of its 11.8 million inhabitants distributed over what was then over 5,000 km2 (1,900 square miles) of mainly farming and small town communities in neighbouring counties. It was not until the early 1980s that it was again given priority—when it was designated one of the fourteen ‘open cities’, and the Shanghai Economic Zone was established in 1983—but it was still compared unfavourably with Guangzhou. In the early 1990s a combination of strong local pressure and a concentration of ex-Shanghai leaders in the higher echelons of national politics led to a major programme of redevelopment, especially in its eastern hinterland, a large area of agricultural and marginal land generally referred to as Pudong (literally ‘east of the Huangpu river’). Here a new high-technology metropolis is being developed, modelled on Singapore. By 1992 Greater Shanghai’s population had risen to 12.87 million, but the urban core districts now contained 7.86 million inhabitants. There was also a large floating population that was not included in this figure—for instance, in 1993 it had a registered temporary population of over 600,000. Migrants into Shanghai generally find employment in the suburban districts and the six counties annexed to the city, and in almost every zhen and smaller town within Shanghai’s orbit there are established communities of migrants. Most are engaged in new, smaller plants and off-farm enterprises. Shanghai’s official boundaries encompass more than 2,300 square miles (6,000 km2), and many of the remaining pockets of untouched farmland are being developed as housing estates, relocated factories and local enterprises, especially in ribbon developments along each highway. The development of rural industrialisation in the city’s attached counties has also introduced a new sense of urbanness into much of the peri-urban core. Source: United Nations Centre for Human Settlements (1996) An Urbanising World Oxford: Oxford University Press METROPOLITAN AREAS AND EXTENDED METROPOLITAN REGIONS As we have seen, a process of population deconcentration tends to affect the core areas of most metropolitan regions, with relatively slow population growth rates or even population decline within the central city and much higher growth rates in the outer areas. In Asia this process is marked in Jakarta, where the population of the metropolitan region is deconcentrating rapidly.15 During the 1980s central Jakarta grew at a rate of 3.1 per Regional perspectives on urbanisation and urban change 149 cent per year, while the growth rates of the urban population in the three neighbouring districts that constitute the Jakarta greater metropolitan area were 11.7 per cent, 20.9 per cent and 19.8 per cent. In the main cities within the more successful economies there is also a major restructuring of the central areas of cities, with development of office buildings, convention centres, hotels and a diverse range of supporting service enterprises as well as modern transport facilities. All these trends tend to involve the removal of people from central cities. In Australian cities such as Sydney, Melbourne, Brisbane, Adelaide and Perth similar trends are evident in a growing centralisation of wealth combined with the selective dispersal of poverty to suburbs vulnerable to structuralchange processes such as deindustrialisation and unemployment (Box 5.3). This form of urban restructuring contrasts with the ‘doughnut’ stereotype that has been employed to characterise US cities, following several decades of white flight from the urban core. Some of the largest relocations of people, to make way for new urban developments, have taken place in the major cities of Asia. BOX 5.3 Australasian cities The most noticeable feature of the physical form of Australasian cities is their low density. The average density of fourteen persons per hectare (5.6 per acre) is similar to that of the USA and compares with average densities of 54 p.p.h. and 157 p.p.h. for European and Asian cities respectively. Most cities in Australia and New Zealand were developed at a time when suburban living in detached housing was practicable. High wages, high employment, cheap land and good public transport encouraged low-density suburbanization in the late nineteenth and early twentieth centuries. As in the USA, the suburbanisation process was accelerated by a rapid rise in car ownership after the Second World War. The development of residential suburbs was aided by government provision of public housing estates (as in the UK) and was accompanied by suburbanisation of manufacturing, retailing and other services (as in the USA). Subsequently, with a few exceptions, such as Perth, urban transport planners have used the low-density urban structure to justify reductions in public-transport services and increased provision for the automobile. In keeping with a neo-liberal response to globalisation, cities in Australia and New Zealand must compete for investment within the international market place. Waterfront redevelopments, as in Wellington’s Lambton Harbour and Darling Harbour in Sydney, reflect similar developments in North American cities (see Chapter 16). Inner-city redevelopment has been accompanied by the growing centralisation of wealth and increasing suburban poverty. Although there are inner-city areas with high proportions of low-income households (mostly single households living in private rented accommodation, as in the UK), the greatest concentrations of low-income households are in the outer suburbs (mostly families with children). This pattern is in contrast to the model of the typical US city. In Australasian cities, as in the UK and the USA, polarisation of income has increased over recent decades and the geography of poverty and affluence is much sharper. There is also an ethnic dimension to socio spatial variations in income This is evident in a Urban geography 150 concentration of Maori in public housing in many of Auckland’s outer suburbs, and of overseas immigrants (most recently from South-East Asia and Latin America) in the middle and outer suburbs of cities such as Sydney, Melbourne and Adelaide. The extent of ethnic segregation, however, is not as extreme as in the ethnic ghettoes of US cities (see Chapter 18). Sources: G.M.Robinson, R.J.Loughran and R.Tranter (2000) Australia and New Zealand: Economy, Society and Environment London: Arnold; S.Hamnett and R.Freestone (2000) The Australian Metropolis London: Spon In many cities, such as Tokyo, Jakarta and Bangkok, population growth has extended outwards beyond the boundary of the metropolitan area, a trend which is often acknowledged by governments in defining new planning regions (for example, the extended metropolitan region around Bangkok stretches some 60 miles (100 km) from the central core). Some Asian urban agglomerations cross existing (or former) national boundaries. Hong Kong, for example, is the centre of the Hong Kong Zhujiang delta region, and a large part of Hong Kong’s manufacturing production has been relocated in southern Guangdong, where some 3 million workers are employed in factories financed and managed by Hong Kong entrepreneurs. At the same time Hong Kong has emerged as a centre of manufacturing-related producer services within the emerging transnational polycentric metropolitan region.16 FACTORS UNDERLYING URBAN CHANGE IN ASIA Urban change in Asia demonstrates how the growth or decline of cities must be understood in terms of the effects of the globalisation of the world economy on the one hand, and economic, social or political changes that are specific to that city or region on the other. For example, the size and rapid growth of Delhi owe more to its role as capital of India than to a position within the global economy. Similarly, Karachi’s population growth over recent decades to become one of the world’s largest cities has been increased by immigration of refugees, including 600,000 from India after Partition, then from Bangladesh during the 1970s, and Afghanistan and Iran in the late 1970s and 1980s. By contrast, urban dynamics in Singapore are shaped more by the city-state’s role within the global economic system than by its acting as a political or administrative centre. Most of the other large cities in Asia come between these two extremes. Tokyo is the world’s largest urban agglomeration because it is both the national capital of the world’s secondlargest economy and one of the three pre-eminent global cities (see Chapter 14). Tokyo’s role within the global economy has also helped to counter a tendency towards decentralisation from the central city. During the 1980s the city attracted many head offices of major national and international companies, while the deregulation and internationalisation of Japan’s financial markets helped to create a concentration of service activities in the city. Urban trends in all Asian nations are also influenced strongly by government actions. A significant policy change during the 1980s was the relaxation of government controls on urban growth in various countries and the downscaling of programmes directing new investment into peripheral regions. The most dramatic change was in China. During the Regional perspectives on urbanisation and urban change 151 Maoist era (1947–77) urbanisation had three main characteristics. First, the level of urbanisation was low due to government migration controls and a system of foodrationing and household regulation (hukou). Second, industrial and urban centres were developed in inland areas at the expense of coastal areas, in pursuit of spatial balance and for national security considerations. Third, since the Chinese economy was isolated from the rest of the world its urbanisation was unaffected by external forces, such as foreign capital. Following the introduction of economic reforms and the ‘open door’ policy in 1978 three main changes occurred in the pattern of Chinese urbanisation. First, the rate of urbanisation accelerated (Table 5.8), fuelled by rapid rural urban migration in the 1980s and 1990s. This was made possible by changes in the household regulation system that during the 1960s and 1970s had controlled migration, and by the growing private food and housing markets and employment opportunities that allowed people to find a livelihood outside the official system. Second, coastal areas, including Shanghai and the Pearl River (Zhujiang) delta, benefited from preferential policies, including fiscal incentives, administrative autonomy and, most important, designation as special economic zones and open development areas. Third, external factors, especially foreign direct investment, played an increasingly important role in shaping urbanisation and urban growth.17 Much of this urban expansion has been concentrated in large cities.18 The post-Maoist urbanisation process transformed a pre-reform situation of underurbanisation, marked by achievement of high industrial growth without a parallel growth of urban population (Box 5.4). AFRICA Most of the nations with the fastest-growing populations are now in Africa (Table 5.9 and Figure 5.7). TABLE 5.9 URBAN POPULATION CHANGE IN AFRICAN COUNTRIES WITH MILLION OR MORE INHABITANTS, 1950, 1990 AND 2001 Country Urban population, 2001 (000) (%) of population in urban areas (level of urbanisation) 1950 1990 2001 Eastern Africa Burundi 603 1.7 6.3 9.3 Eritrea 730 5.9 15.8 19.1 Ethiopia 10,222 4.6 12.3 15.9 Kenya 10,751 5.6 23.6 34.4 4,952 7.8 23.8 30.1 Madagascar Urban geography Malawi 152 1,745 3.5 11.8 15.1 486 28.8 40.5 41.6 6,208 2.4 26.8 33.3 Rwanda 497 1.8 5.6 6.3 Somalia 2,557 12.7 24.2 27.9 Tanzania 11,982 3.8 20.8 33.3 Uganda 3,486 3.1 11.2 14.5 Zambia 4,237 8.9 42.0 39.8 Zimbabwe 4,630 10.6 28.5 36.0 Angola 4,715 7.6 28.3 34.9 Cameroon 7,558 9.8 40.3 49.7 Central African Rep. 1,575 16.0 37.5 41.7 Chad 1,964 3.9 20.5 24.1 Congo, Republic of the 2,056 30.9 53.5 66.1 Gabon 1,038 11.4 45.7 82.3 16,120 19.1 28.1 30.7 Algeria 17,801 22.3 51.7 57.7 Egypt 29,475 31.9 43.9 42.7 Libya 4,757 18.6 82.4 88.0 Morocco 17,082 26.2 46.1 56.1 Sudan 11,790 6.3 22.5 37.1 Tunisia 6,329 31.2 54.9 66.2 Botswana 768 0.3 23.1 49.4 Lesotho 592 0.3 23.1 28.8 Namibia 561 9.4 31.9 31.4 25,260 43.1 49.2 57.7 Benin 2,774 5.3 29.0 43.0 Burkina Faso 1,999 3.8 17.9 16.9 Mauritius Mozambique Middle Africa Zaire (now Dem. Rep. of Congo) North Africa Southern Africa South Africa West Africa Regional perspectives on urbanisation and urban change 153 Côte d’Ivoire 7,197 13.2 40.4 44.0 Ghana 7,177 14.5 34.0 36.4 Guinea 2,312 5.5 25.8 27.9 Liberia 2,414 13.0 42.1 45.5 Mali 3,606 8.5 23.8 30.9 Mauritania 1,624 2.3 46.8 59.1 Niger 2,366 4.9 15.2 21.1 Nigeria 52,539 10.1 35.2 44.9 Senegal 4,653 30.5 39.8 48.2 Sierra Leone 1,714 9.2 32.2 37.3 Togo 1,579 7.2 28.5 33.9 Total 303,481 11.8 31.7 37.7 Source: adapted from United Nations (2002) World Population Prospects: The 2001 Revision New York: United Nations BOX 5.4 Underurbanisation in China China’s path to urbanisation does not mirror that of developed economies in which the process of urbanisation was linked closely with the level of economic development; nor does it reflect the condition of overurbanisation that characterises many developing economies where increases in urban population outpace economic development (see Chapter 4). China’s current urbanisation status has emerged from a situation of underurbanisation prior to 1978. Underurbanisation was a typical phenomenon of socialist economies, including pre-reform China. This arose as a result of several factors, including state emphasis on industrialisation and a related view of urbanisation as primarily a cost incurred in pursuit of industrialisation. This led to efforts to minimise the costs of urbanisation by, for example, the use of capital-intensive production technologies in manufacturing and labour-intensive modes of production in agriculture, and policy restrictions on rural-urban migration. Such strategies were embedded in an ideological commitment to the ideal of income equality, full employment and a more uniform level of economic and social development across regions. Source: L.Zhang and S.Zhao (2003) Reinterpretation of China’s under-urbanization Habitat International In the period since the early 1960s, when most countries gained formal independence, African cities have changed in four main ways: 1. Most have grown in size. There are two main trends. First, while the largest cities have continued to expand, rates of population growth have slackened since the spectacular Urban geography 154 increases of the 1960s and 1970s. Also, whereas the principal component of large city growth was rural urban migration in the earlier post-independence period, natural increase is now the major growth element in many cities. Second, in many countries medium-sized cities are now growing at least as quickly as the largest cities, possibly owing to the deteriorating condition of infrastructure and public services in the major cities. 2. The deterioration of services and infrastructure Figure 5.7 Major urban centres in Africa Note: Zaire is now the Democratic Republic of Congo is the result of the mismatch between economic and urban growth. As national and urban economies stagnate in absolute terms and urban populations continue to grow (at around 4.5 per cent per year for the region as a whole), the resources needed for roads, sewers, water systems, schools, housing and hospitals cannot keep up with demand (Box 5.5). The impact of these adverse living conditions is distributed differentially between the small elite of upper-level managers, foreign diplomats, senior politicians and successful businessmen on the one hand and the growing number of low-income urban dwellers on the other.19 3. The urban labour market has changed since the 1960s. In the post-independence decade the educational system was expanded and African graduates had little difficulty in finding good jobs in either the public sector or the largescale private sector. Subsequent contraction and privatisation of the public sector have reduced these opportunities. Most African cities also have a burgeoning informal economy which has arisen in direct response to the needs of the poor (Box 5.6) (see Chapter 24). Regional perspectives on urbanisation and urban change 155 4. All these changes have had a major effect on city form. Where once the colonial central Plate 5.2 In Kano, Nigeria, the line of the now demolished old city wall separates the high-density ancient Islamic city from the more spacious twentieth-century new town BOX 5.5 The deteriorating infrastructure of African cities As African cities continued to increase in size during the 1980s and 1990s their declining economic situation led to a precipitous deterioration of basic infrastructure and urban services. In many cities most refuse is uncollected and piles of decaying waste are allowed to rot in streets and vacant lots. Schools are becoming so overcrowded that many students have only minimal contact with their teachers. A declining proportion of urban roads are tarmacked and drained, and many turn into quagmires during the rainy season. Basic medical drugs and professional health care are extremely difficult to obtain, except for the rich. Public transport systems are seriously overburdened, and an increasing number of people are obliged to live in informal housing on unserviced plots where clean drinking water must be purchased from water sellers at a prohibitive cost, and where telephones and electrical connections are scarcely available. The lack of investment in urban infrastructure and services also inhibits economic expansion For example in Lagos unreliable infrastructure services impose heavy costs Urban geography 156 on manufacturing enterprises. Virtually every manufacturing firm in the city has its own electric power generator to cope with the unreliable public power supply. These firms invest 10–35 per cent of their capital in power generation alone and incur additional capital and operating expenses to substitute for other unreliable public services. Generally, in Nigeria, and many other low-income countries, manufacturers’ high costs of operation prevent innovation and the adoption of new technology, and make it difficult for them to compete in international markets. Source: United Nations Centre for Human Settlements (1996) An Urbanising World Oxford: Oxford University Press BOX 5.6 Informal employment in African cities Beginning in the 1970s, and gaining strength during the economic downturn of the 1980s and early 1990s, the urban informal sector has become a powerful force for employment creation in virtually all African cities. The main informal activities include carpentry and furniture production, tailoring, trade, vehicle and other repairs, metal goods fabrication, footwear, and miscellaneous services. Estimates made in the mid-1970s suggested that in the typical African country the informal sector employs 60 per cent of the urban labour force. Figures for individual cities included Abidjan 44 per cent, Kumasi 65 per cent, Lagos 50 per cent and Lomé 50 per cent. The proportion of the urban labour force in informal activities has almost certainly risen since then. A trend of the 1980s and 1990s was the supplementation by informal activities of formal-sector jobs on the part of the public-sector employees. Although many informalsector workers are poor, this is not always the case, as such activities as market-trading, the operation of public transport vehicles or renting out of housing in informal neighbourhoods may involve considerable income. Nevertheless, in general there is a marked gap between formal wages and informal earnings. Source: United Nations Centre for Human Settlements (1996) An Urbanising World Oxford: Oxford University Press business district (CBD) was the focus of urban life, now the centre of gravity has shifted as more of the population have moved to the urban periphery where land is cheaper and more accessible, shelter can be constructed economically using locally available materials, and official planning regulations are rarely enforced (see Chapter 25). Urban encroachment on rural areas on the edge of cities leads to conflict. The peri-urban zone is an area of economic and social change characterised by pressure on natural resources, changing employment opportunities and constraints and changing patterns of land use. While rising numbers of urban inhabitants seeking jobs, housing and waste-disposal sites have a clear interest in the expansion of cities into peri-urban areas, the residents of these areas are often some of the poorest in the city region, dependent on natural resources for food, fuel, water and building materials. The horizontal expansion of the African city attenuates infrastructure such as piped water, electricity, sewerage and roads beyond system capacity and adds significantly to the costs of Regional perspectives on urbanisation and urban change 157 education, health and other social services. The combination of peripheral expansion of cities and declining public resources to service them represents the major challenge for the planning and management of African cities in the twentyfirst century. Having examined urban patterns and processes at the global and major world regional scales, in the next chapter we change the level of analysis to examine urban systems at the national scale. FURTHER READING BOOKS A.Bagnasco and P.Le Gates (2000) Cities in Contemporary Europe Cambridge: Cambridge University Press G.Chapman, A.Dutt and R.Bradnock (1999) Urban Growth and Development in Asia Aldershot: Ashgate A.Gilbert (1996) The Mega-city in Latin America New York: United Nations University Press F.Lo and Y.Yeung (1996) Emerging World Cities in Pacific Asia New York: United Nations University Press P.Marcuse and R.van Kempen (2000) Globalizing Cities: A New Spatial Order Oxford: Blackwell K.Pandit and S.Withers (1999) Migration and Restructuring in the United States: A Geographical Perspective New York: Rowman & Littlefield B.Rakodi (1997) The Urban Challenge in Africa New York: United Nations University Press S.Yusuf and W.Wu (1997) The Dynamics of Urban Growth in Three Chinese Cities Oxford: Oxford University Press JOURNAL ARTICLES A.Aguilar (2002) Megaurbanization and industrial location in Mexico’s central region Urban Geography 7, 649–73 P.Coprani (1999) Rapid urbanization in the Gambia Third World Planning Review 21(2), 155–75 T.Firman (1998) The restructuring of Jakarta metropolitan area Cities 15(4), 229–43 T.Firman and G.Dharmapatri (1995) The emergence of extended metropolitan regions in Indonesia Review of Urban and Regional Development Studies 7, 167–88 J.Klink (1999) The future is coming. Economic restructuring in the São Paulo fringe: the case of Diadema Habitat International 23(3), 325–38 J.Musil (1993) Changing urban systems in post-communist societies in Eastern Europe Urban Studies 30(6), 899 906 C.Pannell (2002) China’s continuing urban transition Environment and Planning A 34, 1571–89 S.Osada (2003) The Japanese urban system 1970 1990 Progress in Planning 59, 125 231 Y.Zhu (1998) Formal and informal urbanisation in China Third World Planning Review 20(3), 267–84 KEY CONCEPTS ■ internal migration ■ immigration Urban geography 158 ■ port-of-entry cities ■ retirement migration ■ sunbelt/gunbelt ■ frostbelt/rustbeit ■ baby boom ■ flexible specialisation ■ command economy ■ smokestack industries ■ deindustrialisation ■ extended metropolitan region ■ underurbanisation ■ informal economy STUDY QUESTIONS 1. Identify the main factors underlying the future metropolitan pattern of North America and consider the likely urban outcome of these trends. 2. Discuss the major functional types of cities in Western Europe. 3. Consider the likely changes to the settlement system of Eastern Europe following the collapse of communism. 4. Asia contains many of the world’s largest and fastest-growing cities. Examine the process of urban change in any one city. 5. Consider the impact of independence on the nature of urban development in Africa. PROJECT Using published data (for example, that produced by UN agencies), examine the relationship between key variables such as level of urbanisation and level of industrialisation, or urbanisation and quality of life (measured in a variety of ways). Draw graphs of these relationships and interpret the revealed patterns. Calculation of a correlation coefficient and, if appropriate, insertion of a regression line with confidence limits can help by identifying statistically significant outliers from the general trend. Remember, however, that a statistically significant relationship does not necessarily indicate a causal relationship between any two variables. 6 National Urban Systems Preview: national urban systems; classifications of cities; principles of central-place theory; assessment and applications of central-place theory; diffusion theories of settlement location; a mercantile model of settlement evolution INTRODUCTION At the national level, cities are part of a complex system of interrelated urban places, and are key elements in the economic, social and political organisation of regions and nations. The interdependence among towns and cities makes it important to view a country as a system of urban places rather than as a series of independent settlements. The concept of an urban system refers to a set of towns and cities that are linked together in such a way that any major change in the population, economic vitality, employment or service provision of one will have repercussions for other places1 (Box 6.1). In this chapter we examine urban functions and the differing types of urban places within a national territory, and review the main theories to describe and explain the development of national urban systems. NATIONAL URBAN SYSTEMS AND THE OUTSIDE WORLD Although the principal focus of this chapter is on the national level, we must remember that national urban systems vary in terms of their degree of closure or openness to outside influences. Pred (1977)2 has suggested a fourfold classification of national city systems based on their degree of openness/ closure and level of internal interdependence (Table 6.1): BOX 6.1 Urban systems at the intra-national scale From the eighteenth century onwards the urban structure of Western nations became more complex. Traditional cities became more functionally integrated into regional and national urban systems, while larger cities expanded and developed their economic and social systems. Three levels of urban system may be identified in Western industrial countries: 1 A national system dominated by metropolitan centres and characterised by a step like Urban geography 160 population-size hierarchy, with the number of places at each level increasing in a regular manner with decreasing size of place. 2. Nested within the national system are regional subsystems of cities, displaying a similar but less clearly differentiated arrangement, usually organised about a single metropolitan centre. 3. Contained within regional subsystems are local subsystems or daily urban systems representing the life space of urban residents. 1. Countries where no real systems of cities exist and which are characterised by little economic and social exchange between settlements and with the outside world (i.e. low interdependence TABLE 6.1 CLASSIFICATION OF CITY SYSTEMS BASED ON LEVELS OF CLOSURE AND INTERDEPENDENCE LEVEL OF CLOSURE Low interdependence LEVEL OF INTERDEPENDENCE High interdependence Low closure High closure Little exchange between a nation’s settlements but considerable external influence. No national system of cities. Example: earlier colonial states (with port cities) Little exchange between settlements or with the outside world. No national system of cities. Example: medieval Europe Strong internal (within nation) and external interdependences. A national system of cities with a high level of ‘openness’. Example: the city systems of most present-day European countries Strong internal interdependencies but relatively low external links (though these links will be large in absolute terms). Example: USA and high closure). There are few present-day examples of this category since even Third World countries generally have some form of urban system, a degree of internal exchange and communication and some external links. The best examples are therefore historical. Many countries of medieval Europe were characterised by a few independent urban centres acting as marketplaces for their surrounding areas, but with limited exchange between centres (see Chapter 3). 2. Countries where little exchange occurs between urban centres, but rather than being independent, each centre has strong external ties (i.e. low interdependence and low closure). Here no national system of cities exists but instead individual centres might form part of an international system of exchange. The port cities of eighteenth- and nineteenth-century colonial states exhibited such features. Even today this form of externally focused urban structure persists in some Third World states owing to inertia, the continued export focus of the economy and the locational preferences of multinational organisations. 3. Countries characterised by towns and cities with a high level of interdependence but where the whole system is also subject to strong exogenous influences (i.e. high National urban systems 161 interdependence and low closure). This includes most countries of Western Europe, which are generally highly developed and relatively small in population or territorial terms. High levels of development are linked to specialisation, exchange and interdependence within the national city system; while small size leads to a high level of international trade (in relation to domestic production) with strong transport and information links to other cities throughout the world. 4. Large developed countries such as the USA and international groupings such as the European Union. In these countries or groups of countries the inter-city movement of goods is stimulated by the absence of tariff and quota barriers on trade, whereas the presence of such barriers to external trade results in a relatively high degree of closure. TYPES OF PLACES IN THE NATIONAL URBAN SYSTEM Most cities accommodate a combination of functions, with the relative importance of each varying between places. Researchers have sought to impose some order on this functional diversity by classifying settlements in order to gain insight into their characteristics. City classification schemes range from simple general description3 to those based on multivariate statistical techniques.4 One of the earliest examples of city classification is Harris’s (1943) analysis of 605 US cities.” Having recognised that all large cities are multifunctional, Harris sought to identify ‘critical levels’ of employment in certain occupations that would separate cities into clearly defined functional types. Harris constructed the classification framework by examining the employment profiles of the cities and then intuitively setting minimum levels of employment for each of nine functional types. For example, ‘manufacturing cities’ were those in which manufacturing accounted for at least 74 per cent of total employment in manufacturing, retailing and wholesale (the largest city in this group being Detroit). Harris’s typology provided a benchmark for subsequent city classifications which, as computational techniques became more powerful, included larger numbers of variables. Criticism of the subjective identification of classes by Harris also led to approaches in which classes were derived statistically from the raw data. A prime example is the multivariate statistical classification of British towns by Moser and Scott (1961), which employed a total of fifty-seven measures of population size and structure, population change, households and housing, economic character, social class, voting characteristics, health and education.6 This represented a broader study of urban difference than the earlier, narrower functional analyses. Application of principal components analysis to the data set produced fourteen types of towns (with two, London and Huyton, excluded, since they were ‘too different from other towns to be included in any group’) (Table 6.2). Despite increasing technical sophistication, city classifications were not recognised as making a major contribution to the theoretical development of urban geography. Many considered the classifications to be essentially descriptive (and therefore outmoded), providing illustrative distribution maps rather than explanations of location. Most fundamentally, by the early 1970s critics were able to argue that the development of different classifications had become an end in itself rather than a basis for understanding urban function or for addressing urban problems.7 In response to this critique a number of Urban geography 162 studies employed city classifications as an initial step in the analysis of urban processes and problems. Typical of these is Noyelle and Stanback’s (1983) examination of the differential growth of sunbelt and snowbelt cities.8 In order to test the TABLE 6.2 CLASSIFICATION OF BRITISH TOWNS Mainly resorts, administrative and commercial 1. Mainly seaside towns 2. Mainly spas, professional and administrative centres 3. Mainly commercial centres with some industry Mainly industrial 4. Including most of the traditional railway centres 5. Including many of the large ports as well as two Black Country towns 6. Mainly textile centres in Yorkshire and Lancashire 7. Including the industrial towns of the northeast seaboard and mining towns of Wales 8. Including the more recent manufacturing towns Suburbs and suburban-type 9. Mainly ‘exclusive’ residential suburbs 10. Mainly older mixed residential suburbs 11. Mainly newer mixed residential suburbs 12. Including light-industry suburbs, national-defence centres and towns within the sphere of large conurbations 13. Mainly working-class and industrial suburbs 14. Mainly newer industrial suburbs Source: C.Moser and W.Scott (1961) British Towns: A Statistical Study of their Social and Economic Differences Edinburgh: Oliver & Boyd view that sunbelt cities were accruing population and employment at the expense of snowbelt cities, they first grouped 140 US cities using a location quotient’ which measured the share of a city’s population in a given industry as a percentage of the share of employment in the same industry within the US economy in total. Four major types of city were identified (Table 6.3). The classification was then used as a basis for an analysis of the flow of people and economic activity between cities in different regions. It was found that although the sunbelt was growing at a faster rate than the snowbelt as a whole, study of particular types of city uncovered significant variations on this general trend. For example, while assembly activities had relocated to suburban and sunbelt locations, high-level producer National urban systems 163 services remained in the larger cities, most of which were in the snowbelt. This kind of research shows the potential utility for city classification (in this case for illuminating the TABLE 6.3 CLASSIFICATION OF US METROPOLITAN AREAS BY TYPE AND SIZE Type of centre I Diversified service centres National nodal Regional nodal Sub-regional nodal II Detroit MI Washington DC New Haven CT Production centres Manufacturing Industrial-military Mining-industrial IV New York NY Philadelphia PA Memphis TN Specialised service centres Functional nodal Government-education Education-manufacturing III Example Buffalo NY San Diego CA Tucson AZ Consumer-oriented centres Residential Resort-retirement Nassau FL Tampa FL Source: T.Noyelle and J.Stanback (1983) The Economic Transformation of American Cities Totowa, NJ: Rowman & Allenheld processes of economic restructuring and population redistribution). City-classification research has also proved of value in comparing different types of city in relation to measures of need, hardship and fiscal stress. Logan and Molotch (1987) related the classification of city types to particular consequences (advantages or disadvantages) for individuals and social groups (in terms of, for example, local rents, wages and wealth, taxes and services, and daily life)9 (Table 6.4). Thus, affluent residents of ‘innovation centres’ have the best of all worlds; Hispanic migrant populations will continue to increase in the ‘entrepôt’ cities, leading to further ethnic segregation; ‘module production centres’ and most ‘retirement centres’ are unlikely to provide income redistribution through taxes; growing ‘headquarters’ cities have the highest rates of housing-price inflation; and ‘leisure playgrounds’ are characterised by a two-tier economy comprising many low-paid service workers and relatively few highly paid managers. Clearly, research on urban function has moved a long way from the early descriptive urban taxonomies to a position where city classifications can contribute to an understanding of urban processes, and the formulation and monitoring of urban policy. Urban geography 164 THEORIES OF THE URBAN SYSTEM Settlements exist because certain activities can be carried on more efficiently if they are clustered together rather than dispersed. Inspection of a map showing the pattern of settlement in a region can usually be disaggregated into three component elements that reflect different urban functions: 1. a linear pattern consisting of transport centres performing ‘break of bulk’ or ‘break in transportation’ services (e.g. transfer of cargo from ship to rail), and for which location is related to the network of transport routes; 2. a cluster pattern consisting of places performing specialised activities such as mining, and for which location is related to the localisation of resources; 3. a uniform pattern consisting of places whose prime function is to provide a range of goods and services to the surrounding area, and which require to be accessible to the population that uses them. Most towns, even if based initially upon specialised functions, develop a service function for the surrounding area. Settlements that interact with and provide goods and services to an adjacent hinterland (as well as to their resident population) have been termed central places. By definition, the location of central places is closely connected with the general distribution of population. If the population of an area is evenly spread, then the settlements that serve it will also be evenly distributed. If the distribution of population is uneven, central places will be concentrated in the most accessible locations. Some central places in favourable locations cater for more people and can offer more specialised services; these centres tend to grow progressively larger. Such differential growth produces various grades of central places characterised by different population sizes and zones of influence. The fact that this arrangement of large and small places has often suggested a degree of regularity encouraged efforts to form generalisations about the size and distribution of central places. The most fully developed theory to explain the spatial organisation of settlements derives from Christaller’s (1933) work in southern Germany.10 Following Christaller’s original statement in the 1930s there have been numerous attempts to test his propositions11 and to modify and refine his ideas, TABLE 6.4 US CITY TYPES Type Examples Key functions Headquarters New York City NY; Los Angeles CA Corporate centres: dominance in cultural production, transport and communication networks; corporate control and co-ordinating functions of many large transnational corporations and international banks Innovation centres Silicon Valley towns, including Santa Clara CA; Austin TX; and Research Triangle NC Research and development of aerospace, electronics and instruments; some are (or were) so involved in military contracting that they are ‘war preparation centres’ Module production Alameda CA (military base); Sites for routine economic tasks (e.g. assembly of Hanford WA (nuclear autos, processing of magazine subscriptions or National urban systems 165 places waste); Omaha NE (the ‘800’ phone exchange centre); Detroit and Flint MI (cars) credit card bills), some located near a natural resource (e.g. mining centre) or government function (e.g. Social Security main office in Baltimore MD) Third World entrepôts (warehouses) Border cities such as San Trade and financial centres for importing, Diego CA; Tijuana, Mexico; marketing and distributing imported goods, Miami FL including illegal goods such as drugs and pirated music; major labour centres because of their large numbers of low-paid workers in sweatshop manufacturing and tourist-oriented jobs such as hotel maids Retirement centres Tampa FL; Sun City AZ Home to growing numbers of ageing Americans. Range: affluent towns that maximise services to less affluent cities dependent on pensions, social security and other public programmes to support the local economy Leisure-tourist playgrounds Tahoe City CA; Las Vegas NV; Atlantic City NJ; Disney World FL; Williamsburg VA Range: theme parks, sport resorts, spas to gambling meccas, historical places, and cultural capitals Sources: adapted from J.Logan and H.Molotch (1987) Urban Fortunes Berkeley CA: University of California Press; E.Phillips (1996) City Lights New York: Oxford University Press with a major reformulation proposed by Losch (1943).12 CENTRAL-PLACE TH EORY Christaller’s spatial-equilibrium theory is fundamentally economic in approach and sets out to predict how, through competition for space, an optimal pattern of settlement will emerge. Like all models, central-place theory represents a simplification of reality and is predicated on a number of assumptions (Box 6.2). Economic principles and geometry Christaller’s theory applied to those settlements that are predominantly concerned with serving the needs of the surrounding area. The significance of this service role cannot be measured simply by the population of the place. While population may be BOX 6.2 Assumptions underlying Christaller’s central-place theory 1. There is an unbounded uniform plain in which there is equal ease of transport in all directions Transport costs are proportional to distance and there is only one type of Urban geography 166 transport. 2. Population is evenly distributed over the plain. 3. Central places are located on the plain to provide their hinterlands with goods, services and administrative functions. 4. Consumers minimise the distance to be travelled by visiting the nearest central place that provides the function that they demand. 5. The suppliers of these functions act as economically rational human beings, that is, they attempt to maximise their profits by locating on the plain to obtain the largest possible market. Since people visit the nearest centre, suppliers will locate as far away from one another as possible so as to maximise their market areas. 6. They will do so only to the extent that no one on the plain is farther from a function than he or she is prepared to travel to obtain it. Central places offering many functions are called higher-order centres; others, providing fewer functions, are lower-order centres. 7. Higher-order centres supply certain functions that are not offered by lower-order centres. They also provide all the functions that are provided in lower-order centres. 8. All consumers have the same income and the same demand for goods and services. a measure of absolute importance it is not a measure of a settlement’s centrality. Centrality is the degree to which a place serves its surrounding area, and this can be gauged only in terms of the goods and services offered. Clearly, there are different orders of goods and services: some are costly, bought infrequently, and need a large population to support them (e.g. furniture, jewellery); others are everyday needs and require a small population (e.g. groceries). From this two concepts emerge: 1. The threshold population. The threshold is defined as the minimum population required for a good or service to be provided that is, the minimum demand to make the good or service viable. 2. The range of a good. This is the maximum distance which people will travel to purchase a good or service. At some range from the central plaee, the inconvenience of travel as measured in time, cost and effort will outweigh the value of or need for the good. From these two concepts an upper and a lower limit can be identified for each good or service. The lower limit is determined by the threshold, the upper limit by the range. Ideally each central place would have a circular trade area. It is obvious, however, that if three or more tangent circles are placed in an area, unserved spaces will exist. In order to eliminate any unserved areas the circular market areas must overlap and, since people in these overlap zones will choose to visit their nearest centre in keeping with the assumption of minimum movement, the final market areas must be hexagonal (Figure 6.1). The resulting hexagonal pattern is the most efficient way of packing market areas on to the plain to ensure that every resident is served. Christaller started by identifying typical settlements of different sizes in southern Germany. He then measured their average population, distance apart, and extent of their hexagonal tributary areas. Christaller also stated that the number of central places at each level of the settlement hierarchy follows a fixed ratio (the K value) from the largest National urban systems 167 Landeshauptstadt (regional capital) to the smallest Marktort (hamlet) (Table 6.5). In its simplest terms, therefore, Christaller’s model proposed that settlements with the lowest level of specialisation (Marktort/hamlet) would be equally spaced and surrounded by hexagonally shaped hinterlands. For every six hamlets there would be a larger, more specialised central place (Amtsort/township centre) which would be located equidistant from other township centres. The Amtsort would have a larger market area for specialised services not available in the hamlet. Further up the hierarchy even more specialised settlements would also have their own hinterlands and would be located an equal distance from each other. In the basic model the smallest centres would be spaced 7km apart. The next higher centres would serve three times the area (and therefore three times the population) of the lower-order centres, and would be located √3×7=12km apart. Similarly, the trade area of centres at the next higher level of specialisation would again be three times larger (Table 6.5). This kind of arrangement is called a K-3 hierarchy; in it the number of central places in the settlement hierarchy follows a geometric progression: 1, 3, 9, 27, etc. Thus lowerorder centres, in order to be provided with higher-order goods and services, nest within the tributary areas of higher order places according to a definite rule (the K value). Figure 6.1 Deriving the hexagonal pattern of market areas for central places Urban geography 168 TABLE 6.5 CHARACTERISTICS OF CENTRAL PLACES IN SOUTHERN GERMANY Place No. of Distance No. of places apart complem (km) entary regions Range of region (km2) Area of region (km2) Typical No. of population types of Place Region goods offered Marktort 486 7 729 4.0 44 40 1,000 3,500 Amtsort 62 12 243 6.9 133 90 2,000 11,000 Kreisstadt 54 21 81 12.0 400 180 4,000 35,000 Bezirksstadt 18 36 27 20.7 1,200 330 10,000 100,000 Gaustadt 6 62 9 36.0 3,600 600 30,000 350,000 Provinzhauptstadt 2 108 3 62.1 10,800 1,000 100,000 1,000,000 Landeshauptstadt 1 186 1 108.0 32,400 2,000 500,000 3,500,000 Source: W.Christaller (1933) Die zentralen Orte Suddeutschenland Jena: Fischer, translated by C.W.Baskin (1966) Central Places in Southern Germany Englewood Cliffs, NJ: PrenticeHall A settlement pattern with these features exhibits what Christaller has called the marketing principle. In this, the major factor influencing settlement distribution is the need for central places to be as near as possible to the population they serve. Thus the K3 hierarchy and nesting pattern produce the Figure 6.2 Hierarchical and spatial arrangement of central places National urban systems 169 maximum number of central places in accordance with the notion of movement minimisation (Figure 6.2). By applying the economic principles in conjunction with the geometric properties of the theory plus the simplifying assumption of an isotropic surface, Christaller derived his general model of the location, size and spacing of settlements. An assessment of central-place theory The main criticisms directed at central-place theory are the following: 1. The theory is not applicable to all settlements. Being limited to service centres, it does not include some of the functions, such as manufacturing industry, that create employment and population. 2. The economic determinism of the theory takes no account of random historical factors that can influence the settlement pattern. 3. The theory makes unrealistic assumptions about the information levels and mental acumen required to achieve rational economic decisions, even if profit maximisation were the only goal of human behaviour. 4. The notion of a homogeneous population ignores the variety of individual circumstances. 5. Christaller’s model assumed relatively little governmental influence on business locational decisions, whereas today national and local governments play a major role in influencing business locations by, for example, offering grants to attract electronics firms into Scotland’s ‘Silicon Glen’ or the lobbying by US sunbelt city mayors to attract investment to their cities. 6. Central-place theory is a static formulation that relates to the distribution of service centres under assumed conditions at one point in time. A particular level of mobility is implied by the assumption that consumers look to their nearest central place to satisfy their needs. Levels of personal mobility have increased greatly since the model was proposed. Consumers do not always visit their nearest store, and multipurpose shopping trips often result in low-order centres being by-passed for low-order goods, thus leading to their decline. Even in the classic field-study area of Iowa, economic restructuring and improved transportation infrastructure have undermined the ‘nearest neighbour’ travel patterns of earlier generations. Also, in many advanced countries, today telecommunications and ‘tele-shopping’ have further eroded the ‘frictional effect’ of distance on consumer behaviour. Christaller (1966 p. 84) was not unaware of the temporal limitations of his theory, pointing out that: the stationary state is only fiction whereas motion is reality. Every factor which adds to the importance of the central place regional population, supply and demand of central goods, prices of the goods, transportation conditions, sizes of the central places and competition between central and dispersed production of a good is subject to continuous change.13 Urban geography 170 Unfortunately, he did not translate these qualifications into a dynamic model of the functional and spatial dimensions of the urban system. Accordingly, the relevance of central-place theory in explaining current settlement patterns is limited. However, recognition of the limitations of central-place theory is not the same as rejecting it. Even as an ideal, the theory is useful. Study of where theory and reality diverge can lead to explanation. Kolars and Nystuen (1974 p. 73) suggest that the main contributions of both Christaller and Losch ‘have been as much to stimulate further geographical thought as to give us any absolute explanations of the real world’.14 While there is little evidence of a complete central-place settlement structure emerging in the real world, the theory has stimulated much work in relation to retailing and consumer behaviour, and, as we see below, in the fields of physical and social planning. Applications of central-place theory Central-place ideas have been employed widely in regional-planning schemes in the USA. Canada, Africa, India, Europe and the Middle East.” For example, an Israeli settlement on the Laklish plains to the east of the Gaza Strip was based on a three-level hierarchy of: 1. ‘A’ settlements of various types (including protective border kibbutzim) housing immigrant settlers and serving as agricultural centres containing facilities used daily; 2. ‘B’ settlements (rural community centres), each planned to serve four to six ‘A’ settlements and to supply facilities and buildings used by them once or twice a week; 3. ‘C’ settlements (regional centres), towns roughly at the geographical centre of their region, providing administrative, educational, medical and cultural facilities, and with factories for crop-processing. The most clearly articulated application of central-place principles has occurred in the Dutch polder-lands.16 The increasing importance attached to a planned settlement pattern can be seen by tracing the development of the Wieringermeer, drained in 1930, and the north-east and east Flevoland polders, drained in 1942 and 1957 respectively. In the first polder the location of the villages (service centres) was not successful. The settlement pattern did not conform to any model distribution, with the planners expecting that a spontaneous process of settling would lead to certain clusters at road intersections. As a result, the three regional villages of Slootdorp, Middenmeer and Wieringerwerf were clustered in the middle of the polder, which meant they had overlapping trade areas and that people living well away from the villages were inconvenienced by long journeys. The lower than expected population growth on the polder exacerbated the problems, with small villages incapable of providing a satisfactory level of service. In the second, north-east, polder a settlement pattern was carefully planned in an attempt to avoid the mistakes of the Wieringermeer. Since it was one of the few places in the world where no historical or physical obstacles frustrated the realisation of a theoretical spatial model, Christaller’s hierarchical system was applied with some modifications. In the middle of the area a regional centre, Emmeloord, was founded (with a target population of 10,000), with ten surrounding villages as local service centres each with target populations of 1,000 2,000. Despite this careful planning, however, the settlement pattern quickly demonstrated a number of shortcomings. Because of National urban systems 171 agricultural mechanisation and the reduced demand for labour, the populations of most villages did not reach the target (threshold) figure and this made it difficult to keep the services feasible and the community viable. Paradoxically, Emmeloord grew more rapidly than anticipated. This was due to the increased accessibility to the rural population of the varied services in the regional centre, largely as a result of the general increase in mobility engendered by the spread of the motor car in the 1960s. Experience gained in the first two polders was applied to the settlement of East Flevoland. The initial settlement plan had been similar to that of the north-east polder, with ten ‘A’ centres having a local-service function surrounding a single ‘B’ or district centre, Dronten. A ‘C’ centre, Lelystad, the capital of the polder’s province, was planned at the junction of the four polders but in the western corner of East Flevoland (Figure 6.3). Because of the diminishing importance of farm employment and the increasing affluence, aspirations and mobility of the population, this pattern was reduced to only four villages in 1959 and eventually to two by 1965 (Figure 6.3). The declining significance of agricultural employment over the course of the development of the first three polders also influenced the location and population composition of settlement in the later southern Flevoland and Markerwaard polders, with less emphasis on placing villages to serve the needs of farming, and the introduction of commuters and other nonagricultural workers from Randstad Holland. The fact that in Southern Figure 6.3 The changing settlement pattern of East Flevoland Urban geography 172 Flevoland, an area within the sphere of influence of Randstad Holland (see Box 8.3), no thought was given to a system of service centres set up according to a hierarchically arranged pattern of villages suggests that in such pressured Durban” areas classical central-place theory is of limited relevance. Reconstructing central-place theory The highest central place in Christaller’s model is the regional capital (Lamdeshauptstadt) of 500,000 people and regional population of 3.5 million (Table 6.5). The omission of higher levels in the settlement hierarchy reflects the model’s basis in southern Germany. Since the time of its inception, globalisation has increased the importance of world or global cities (see Chapter 14), meaning that a modern reconceptualisation of Christaller’s model would incorporate at the apex of the hierarchy: 1. global cities, typically with 5 million or more people within their administrative boundaries and up to 20 million in their hinterlands (for example, New York or London); 2. subglobal cities, typically with 1 million to 5 million people and up to 10 million in their hinterlands (for example, national capitals as well as commercial capitals that are not global cities, such as Milan or Barcelona). As a consequence of increased personal mobility, places at the lower levels of Christaller’s hierarchy have declined in significance as central places having lost any service functions (such as a village store) to become mainly residential villages. Only the Bezirkstadt, with a population of 10,000 and service hinterland of 100,000, retains a significant service function (with, for example, a superstore and limited range of national chain stores). Some of the most significant changes have affected settlements at the next two higher levels, typically county market towns found across much of Southern England, Southern Germany and most of France. Many of these have grown because in depopulating regions they have attracted population outflow from surrounding rural areas, and in more prosperous regions have attracted much of the out-migration from major cities. DIFFUSION THEORIES As we have seen, one of the most serious disadvantages of Christaller’s theory is its static nature, which does not enable it to respond easily to changing social and economic conditions. This has led some writers to suggest that attempts to understand ‘natural’ as opposed to planned settlement patterns in terms of spatial equilibrium theory are of limited value. An alternative, which explicitly acknowledges the importance of the time dimension and historical perspective, is to examine the processes by which settlement spreads across a region from the initial point of colonisation. A number of models have been devised. National urban systems 173 Figure 6.4 Bylund’s model of settlement diffusion Source: E.Bylund (1960) Theoretical considerations regarding the distribution of settlement in inner north Sweden Geografiska Annaler Series B 42, 225–31 Bylund (1960) has proposed, within a deterministic framework, six hypothetical models of settlement diffusion based upon his study of early colonisation in central Lapland.17 His four basic models are shown in Figure 6.4; each differs in the number and location of ‘mother settlements’. The process of clone-colonisation (model B) appears to imitate the actual pattern of colonisation most closely, and this concept is developed in two further models to replicate the known settlement history of the area. The principles underlying these models have also been developed by Morrill (1962) in a probabilistic simulation of central-place patterns over time.18 The idea behind the model is that the behaviour of persons, as seen in the founding and growth of settlements and transport lines, occurs gradually over time and may be described as random within certain limiting conditions. The aim of Morriirs approach is to account for the general pattern of settlement, not the exact location of places. The model employed is of the Monte Carlo type (a stochastic growth process model) in which a set of probabilities related to both human and physical conditions governs choice of behaviour.19 The mechanics of the method have been well illustrated by Abler et al. (1972).20 Morrill (1963) examined the spread of settlement in Sweden using this historicalpredictive approach.21 He begins by acknowledging that as the number, size and location of settlements in any region are the result of a long and complex interplay of forces, any Urban geography 174 study which proposes to explain the origins of such patterns must take into account four major factors: 1. the economic and social conditions that permit and/or encourage concentration of economic activities in towns; 2. the spatial or geographical conditions that influence the size and distribution of towns; 3. the fact that such development takes place gradually over time; 4. recognition that there is an element of uncertainty or indeterminacy in all behaviour. Whereas the first two factors are also explicit considerations in classical spatial equilibrium theories, the latter two are uniquely central to diffusion theory. The historical dimension is crucial not least because, as study of the Dutch polderlands showed, changes in social, economic and technological conditions over time can have a radical effect on the efficient functioning of a settlement pattern. It is also clear that locational decisions made on the basis of incomplete information are subject to error, and consequently real settlement patterns are the outcome of many less-than-perfect decisions. Morrill (1963 p. 9) attempted to simulate the development of the settlement pattern in southern Sweden over the period 1860 1960 in order to discover the major locational forces channelling urban development and migration. He was able to conclude that: in sum, the model results can be considered realistic from the point of view of distribution, that is similarity in spatial structure; and from the point of view of process, as a reasonable recognition and treatment of the pertinent forces. Deviations resulted in the main from oversimplified assumptions rather than from a mistaken approach. The construction of a theory to explain the spread of settlement within a territory also formed the basis of Hudson’s (1969) work in Iowa, in which he attempted to integrate diffusion theory with central-place theory.22 Drawing on the work of plant and animal ecologists he identified three phases of settlement diffusion: 1. colonisation, which involves the dispersal of settlement into new territory; 2. spread, in which increasing population density creates settlement clusters and eventually pressure on the physical and social environment; 3. competition, which produces regularity in the settlement pattern in the way suggested by central-place theory. Empirical testing of these hypotheses using settlement data from six areas of Eastern Iowa, at three different times between 1870 and 1960, found that the suggested increase in settlement regularly over time did occur. Vance (1970)23 also employed an historical diffusion perspective to devise a mercantile model of settlement evolution within a colonial setting (Figure 6.5). This envisaged five main stages in the development of settlement systems, in both core (home country) and periphery (colony): 1. Exploration. This involves the search for economic information by the prospective colonising power. National urban systems 175 2. Harvesting of natural resources. This involves the periodic harvesting of staple products, such as fish and timber, with limited permanent settlement. 3. Emergence of farm-based staple production. Increased permanent urban settlement permits export of agricultural commodities to the mother country that supplies the colony with manufactured goods. The seaports in the colony act as ‘points of attachment’ with the home country. Figure 6.5 Vance’s mercantile model of urban development Source: J.Vance (1970) The Merchant’s World Englewood Cliffs, NJ: Prentice-Hall 4. Establishment of interior depot centres. Settlement penetrates inland along favourable long-distance routes that facilitate the movement of staple products from the interior to coastal points of attachment which also begin to develop manufacturing. Towns established at strategic locations along these routes serve as ‘depots of staple Urban geography 176 collection’. This period also witnesses rapid growth of urban industrial centres in the home country to supply both the overseas and the domestic markets. 5. Economic maturity and central-place infilling. The growth of a manufacturing sector leads to economic maturity accompanied by the emergence of a central-place form of settlement pattern in which the depots of staple collection take on a service function and develop as regional centres. In time smaller central places are founded to serve local needs. Overall, the spatial structure of the colonial urban network is exogenic, being determined largely by external forces (Box 6.3). There are close similarities between the patterns of urban development in the USA modelled by Vance (1970) and those identified in West Africa and Brazil by Taffe et al. (1963).30 Notwithstanding the analytical value of these theories, the great variety of settlement forms and distribution would appear to lend support to Bunce’s contention (1982 p. 96) that ‘settlement patterns are the product of the area which they occupy’31 and Grossman’s view (1971 p. 192) that ‘general laws are meaningless outside the specific cultural and technological context’.32 Despite the pedagogic value of the seminal models considered above, it is difficult to avoid the conclusion that a general theory to explain the location, size and BOX 6.3 The missing middle in the Third World urban hierarchy A characteristic of many national urban systems in the Third World is the underdevelopment of the middle tier of the settlement hierarchy. Johnson (1970)24 showed how the nations of Europe had almost ten times as many central places (towns of over 2,500 people) per village as Middle Eastern states. The resulting argument, that a more even pattern of development in the Third World requires promotion of middle-order settlements, has been taken up by subsequent writers such as Rondinelli (1983).25 This proposal is also related to the thesis of urban bias,26 which maintains that major cities of Third World countries have received a disproportionate amount of national resources and that a greater degree of territorial justice needs to be incorporated into national economic and urban development policy. This approach has been employed in Third World socialist states such as China,27 Cuba28 and Tanzania.29 spacing of settlements is unattainable. Urban geographers now acknowledge the need to take into account the political and social forces operating at a variety of spatial scales in interpreting the changing spatial and functional dimensions of national urban systems. FURTHER READING BOOKS K.Beavon (1977) Central Place Theory: A Reinterpretation Harlow: Longman W.Christaller (1966) Central Places in Southern Germany Englewood Cliffs, NJ: Prentice-Hall National urban systems 177 E.Keeler and W.Rogers (1973) A Classification of Large American Urban Areas Santa Monica, CA: Rand Corporation C.Moser and W.Scott (1961) British Towns: A Statistical Study of their Social and Economic Differences Edinburgh: Oliver & Boyd J.Vance (1970) The Merchant’s World Englewood Cliffs, NJ: Prentice-Hall JOURNAL ARTICLES E.Bylund (1960) Theoretical considerations regarding the distribution of settlement in inner north Sweden Geogrqfiska Annaler 42, 225–31 M.van Hulton (1969) Plan and reality in the IJsselmeerpolders Tijdschrift voor Economische en Sociale Geografie 60, 67–76 R.Morrill (1962) Simulation of central place patterns over time Lund Studies in Geography Series B 24, 109–120 KEY CONCEPTS ■ openness/closure ■ location quotient ■ urban taxonomies ■ break of bulk ■ central places ■ centrality ■ range of goods ■ marketing principle ■ frictional effect of distance ■ diffusion theories ■ mercantile model STUDY QUESTIONS 1. Prepare a list of the largest twenty-five or fifty settlements in your country. Using one of the city classification schemes described in the chapter, allocate each settlement to an appropriate category. Describe and explain the results. 2. Explain the economic principles and geometry underlying Christaller’s central-place theory. 3. Since its inception, central-place theory has been subjected to critical analysis. Examine the main criticisms of the theory. 4. With reference to particular examples illustrate the ways in which central-place theory has been applied in the real world. 5. Explain Vance’s mercantile model of settlement evolution. PROJECT Determine the sphere of influence of either a single settlement or a group of settlements by selecting a range of services (e.g. banking, education, different forms of retailing, hospitals newspaper circulation areas) and interviewing the service providers to ascertain Urban geography 178 where people come from to obtain the services. Each service will have its own catchment area. Plot these on a map and examine the pattern shown. Explain any differences in the catchment areas. Is there a clearly defined sphere of influence for the settlement? If you are studying more than a single settlement, do spheres of influence overlap? Does any one settlement dominate the region in terms of service provision? PART THREE Urban Structure and Land Use in the Western City Plate 1 The ancient city: Thebes. The archaeological site at KarnakLuxor on the bank of the River Nile, south of Cairo, reveals the splendour of the ancient city of Thebes that for centuries was the capital of the Egyptian kingdom. The city’s decline began in 672 BC and it was already a ruin in Roman times Urban geography 180 Plate 2 Fantasy city: Las Vegas. The 47 acre, 2,526 room Luxor resort hotel in Las Vegas NV offers guests a mix of ancient Middle Eastern environments, including a ten-storey sphinx with laser-beam eyes, a thirty-storey glass pyramid and the Hanging Gardens of Babylon Plate 3 The post-industrial city: Glasgow. The abandoned wharves of Prince’s dock on the River Clyde in Glasgow were closed in the 1970s and partially filled to provide the site for the 1988 Glasgow Garden Urban structure and land use in the western city 181 Festival as part of the city’s reimaging strategy. The site is under development as a residential-officeentertainment complex Plate 4 The modern industrial city: Pusan, South Korea. A major factor contributing to the demise of heavy engineering and shipbuilding in old industrial cities of the West was competition from newly industrialising countries such as South Korea. The No.1 dock in the Okpo shipyard in Pusan is the world’s largest, in which six different types of vessel can be constructed simultaneously Urban geography 182 Plate 5 The global city: Tokyo. Tokyo is the largest urban agglomeration in the world, with a population nearing 30 million. The city dominates the Japanese urban system, and the Tokyo metropolitan region is the world’s largest consumer market. Problems arising from the unprecedented growth of the city include escalating land prices, a shortage of office space, high-cost housing and congestion, as in the retail district of Shibuya Urban structure and land use in the western city 183 Plate 6 The mobile city: San Diego CA. The Gordian knot of freeways in San Diego is a concrete metaphor for the mobile automobiledependent lifestyle and low-density dispersed form of metropolitan development typical of the USA Plate 7 The informal city: Bombay. A large proportion of the poor residents of Third World cities earn Urban geography 184 a livelihood in the informal sector of the urban economy by carrying out domestic duties for the more affluent. In the municipally owned and managed Dhobi Ghats of Bombay self-employed washermen and women launder the clothes for a regular clientele Plate 8 The congested city: Hong Kong. Central Hong Kong is one of the most densely populated urban environments in the world. The soaring cost of land is reflected in the sprouting high-rise office and apartment blocks. The return of the territory to China in 1997 has not dampened the city’s capitalist dynamic 7 Land Use in the City Preview: urban morphogenesis; town plan analysis; ecological models of the city; Burgess’s concentric-zone model; Hoyt’s sector model; the multiple-nuclei model of Harris and Ullman; Mann’s model of a British town; Kearsley’s modified Burgess model; Vance’s urban-realms model; White’s model of the twenty-first-century city; urban political economy; major actors in the production of the built environment; growth coalitions; the central business district; urban architecture; architecture and urban meaning; the social construction of the urban landscape INTRODUCTION Although most towns and cities have occupied the same location for centuries, the buildings and other physical infrastructure which comprise the built environment are not fixed but affected continuously by dynamic forces of change initiated by public and private interests. This modification of the urban environment occurs at a variety of scales ranging from the residential relocation decisions of individual households to large-scale projects including public road-building programmes and private house-building schemes. In addition, to differing degrees in different countries, the operation of these ‘market forces’ is influenced (enhanced or constrained) by national and local planning. The net effect of these socio-spatial processes is revealed most clearly in the land-use structure of the city. In this chapter we examine the principal models and theories of urban land use. For analytical convenience these are arranged into four broad types based on the principles of: 1. morphogenesis; 2. human ecology; 3. political economy; 4. postmodernism. URBAN MORPHOGENESIS The study of urban morphogenesis or town-plan analysis has a long history in urban geography. Since its high-point in the 1960s the approach has been sidelined, despite the fact that in its more recent formulations it has sought to advance from description and classification of urban forms1 to analysis of the causal forces underlying changes in the pattern of urban land.2 Urban geography 186 Much current research in the morphogenetic tradition stems from the seminal work of Conzen (1960),3 who divided the urban landscape into three main elements of town plan, building forms and land use, and demonstrated how each reacted at a different rate to the forces of change: 1. Land use is most susceptible to change. 2. Since buildings represent capital investments and are adaptable to alternative uses without being physically replaced, change occurs at a slower rate than with land use. 3. The town plan or street layout is most resistant to change. Conzen also introduced the concepts of the fringe belt and burgage cycle to aid analysis of urban change. The existence of a fringe belt and associated fixation line reflects the fact that urban growth is cyclical rather than continuous, with periods of outward extension alternating with periods of standstill (marked by a fixation line) due to a downturn in the building cycle. A succession of fringe belts can be identified around most towns, related to phases of active growth (Figure 7.1). The burgage cycle indicates the way in which land use on a single plot develops over time. These concepts have been developed by White-hand (1991)4 into an approach that seeks to identify the decision-making behaviour underlying land-use change. This is based on the premise that the town plan at any one time is the outcome of the perceptions, principles and policies of individuals (e.g. landowners) or agencies (e.g. local planning departments) which exercise the necessary power. The westward extension of the city of Glasgow in the eighteenth century illustrates both the economic power of landowners and the influence of the burgage-plot pattern of land-holding on urban form (Box 7.1). More recent evidence of the influence of landowners, developers and planners on urban structure is provided by Whitehand’s (1992) study of residential infilling in Amersham in Berkshire, in which he explores the decision-making processes underlying urban change, focusing on negotiations between developers and the local planning authority.5 In similar vein, Moudon (1992) has studied the evolving residential morphology of the North American city.6 These attempts to explore the backgrounds, motivations and actions of the major agents in the creation of townscapes at the local level represent a major advance on the earlier descriptive classifications of town plans. However, the difficulty of undertaking such detailed investigations increases as one looks further into the urban past. ECOLOGICAL MODELS OF THE CITY According to the ecological perspective developed by the Chicago school of human ecology, the significant processes underlying the spatial configuration of the growing American industrial city were analogous to those found in nature. Hence, competition among land uses for space resulted in the invasion of the most desired parts of a city and eventually the succession of existing land uses by a more dominant activity (as in the expansion of the central business district (CBD) into the surrounding transition zone). Under free-market conditions, certain parts of the city would be occupied by the function Land use in the city 187 Figure 7.1 The fringe belts of Newcastle upon Tyne Source: J.Whitehand (1967) Fringe belts: a neglected concept of urban geography Transactions of the Institute of British Geographers 41, 223– 33 that could maximise use of the site, and in due course natural areas would evolve, distinguished by their homogeneous social or ethnic character (such as a slum or ghetto). On the basis of this tendency for ecological processes to sort similar households, Burgess (1925) derived his general concentric zone model of residential differentiation.7 Figure 7.2 shows Burgess’s interpretation of the land-use structure of Chicago (on the left of the diagram) and the general model arising from it (on the right of the diagram). The characteristics of the five zones are described in Box 7.2. Burgess maintained that the city tends to grow outwards in annular fashion from Zone I to Zone V. It is important to recognise that the concentric-zone model was proposed as an ideal type, not as a representation of reality. Based on the study of one city (Chicago) at one point in time (the 1920s) it offers a description of urban development as this would occur if only one factor (radial expansion from the city centre) determined the pattern of urban growth. Burgess was able to point to many examples of invasion and succession underlying the changing occupancy pattern of different zones in Chicago in the early twentieth century as successive waves of immigrants worked their way from their initial quarters in the zone of transition out to more salubrious neighbourhoods. In the model (Figure 7.2) Urban geography 188 BOX 7.1 Land-ownership and the development of the street pattern in eighteenth-century Glasgow Between 1710 and 1780, eight new streets were developed as the town spread west from Glasgow Cross. The influence of the medieval pattern of land ownership based on burgage plots exerted a controlling influence on these developments. As the street plan shows, whereas a single plot or rig provided sufficient space to form a narrow wynd or vennel, the wider and longer streets were formed by the purchase and amalgamation of several plots. Miller Street (constructed by John Miller, a maltster and town bailie, or magistrate) required eight plots. The large profits to be gained by capitalising on the appreciating land values are confirmed by the fact that the cutting of the street obliged Miller to demolish half his newly built mansion. The first of the new streets, Virginia Street, was opened in 1793 through two acres (0.8ha) of cabbage plots. Virginia Street was a furrow long (furlong, about 200m). Although the plot was evolved for the convenience of tillage, its size and shape were well suited to the speculative builder’s goal of maximising the number of properties fronting on to the new streets. These developments on what had been the old burghal tillage lands set in motion a shift in the focal point of the city west from the Cross. It also signalled the emergence of a marked socio-spatial segregation as the upper classes moved away from the crowded conditions of the old town. Source: M.Pacione (1995) Glasgow: The Socio-Spatial Development of the City Chichester: Wiley this is shown by how some of the early immigrant groups (e.g. Germans) have ‘made it’ to the superior accommodation of Zone III, replacing second-generation American families who had moved out to settle the outer residential zone, Zone IV. Burgess was Land use in the city 189 not unaware of the many other factors that influence city growth. (For example, in a less well-known model he postulated a relationship between residential status and altitude in ‘hill cities’.)8 Although Burgess maintained that his model would apply to the thencontemporary American city, he did not expect any one city to be a perfect example of the theory. Figure 7.2 Burgess’s concentric-zone model of urban land use, applied to Chicago Source: R.Park and E.Burgess (eds) (1925) The City Chicago, IL: University of Chicago Press TABLE 7.1 ASSUMPTIONS AND PRINCIPLES OF THE CONCENTRIC-ZONE MODEL OF URBAN LAND USE 1. Cultural and social heterogeneity of the population 2. Commercial-industrial base to the economy of the city 3. Private ownership of property and economic competition for space 4. Expanding area and population of the city 5. Transport is equally easy, rapid and cheap in every direction within the city 6. The city centre is the main centre for employment and near this centre space is limited; competition for this space is high, and therefore it is most valuable. The opposite is 7. true of peripheral areas No districts are more attractive because of differences in terrain 8. No concentrations of heavy industry 9. No historic survival of an earlier land-use pattern in any district Urban geography 190 Subsequent attempts to apply the model have often been less than successful, partly because they failed to recognise its limiting assumptions (Table 7.1). For example, the model is based on the concept of a city with a large population undergoing rapid Plate 7.1 The high-rise central business district of San Francisco CA provides a backdrop to the lowrise high-status neighbourhood of Nob Hill expansion, with much of the population increase assumed to be due to the arrival of ethnically diverse immigrants from overseas. Both assumptions were met in Chicago, which from 1860 to 1910 increased its population almost twentyfold. The Burgess model was also formulated on the basis of a particular set of economic and political circumstances. In particular, the model assumed private ownership of property and the absence of any city-planning constraints on the use of private property. Under these circumstances, property owners were free to develop their land as they wished. It also Land use in the city 191 meant that only the wealthy could afford to live in the better locations away from innercity slums. These conditions were again met by the Chicago of the 1920s. But in most Western societies today, government has intervened in the housing and property market (see Chapter 8). As a result, the slum housing of the model’s zone in transition has been replaced, in many cases, by public redevelopment schemes. Further, around every major British city there are large estates of council housing provided by the government for people who would be unable to compete for such locations in the open market envisaged by Burgess. The general applicability of the model is further reduced by the gentrification of some inner-city slums, and by the continued BOX 7.2 Burgess’s concentric-zone model of urban land use Zone I The first and smallest zone is the central business district (CBD). This is the focus of the commercial, social and cultural life of the city, and the area where land values are highest. Only activities where profits are high enough to meet the rent demanded can locate in the area. The heart of the zone is the downtown shopping area with large department stores and the most exclusive shops. The area also contains the main offices of financial institutions, the headquarters of civic and political organisations, the main theatres and cinemas, and the more expensive hotels. The CBD is the most accessible area in the city. It has the greatest number of people moving into and out of it each day, and the main transport termini are, therefore, located there. Forming the outer ring of the central area is a wholesale business district with warehouses, light industries and, perhaps, a market. The CBD contains the original nucleus of the settlement, but only scattered pockets of residences remain. Zone II Immediately adjacent to the CBD is the zone in transition. Early in the history of the city this formed a suburban fringe that housed many of the merchants and well-to-do citizens. With the growth of the city, however, industries encroached into this zone from the inner zone, and the quality of the residential environment deteriorated. The inner margins of the zone in transition are industrial and its outer ring is composed of declining neighbourhoods. The once fashionable town houses have been converted into flats, furnished rooms and even small industries. The population of the zone is heterogeneous and includes first-generation immigrants as well as older residents. It is also an area frequented by vagrants and criminals, and rates of crime and mental illness are the highest in the city. Those who own property in the zone are interested only in the long-term profit to be made from selling out to businesses expanding from the central area, and in the shortterm profits that accrue from packing in as many tenants as possible. As a result, property is run-down. The zone is characterised by a highly mobile population. Not surprisingly, as people prosper they tend to move out into Zone III, leaving behind the elderly, the isolated and the helpless. Urban geography 192 Zone III This is termed the zone of independent working men’s homes. The population consists of the families of factory and shop workers who have managed to prosper sufficiently to escape the zone in transition but who still need cheap and easy access to their workplaces. The zone is focused on factories, and its population forms the bulk of what may be termed the respectable working class. Unlike in the ‘childless’ zone in transition, all age groups are represented. Zone IV This is an area of better residences, a zone of private housing or good apartment blocks. It is the home of the middle class. At strategic locations, subsidiary shopping centres have developed as mini versions of the downtown shopping area. Zone V Still farther out from the inner city is the commuter belt within thirty to sixty minutes’ journey time of the CBD. This is essentially a suburban dormitory zone characterised by single-family dwellings. Beyond these five main zones Burgess sometimes recognised two additional areas comprising: Zone VI The surrounding agricultural district Zone VII The wider hinterland of the city. association between high social status and inner-city residence in many European cities.9 The value of the concentric-zone model is therefore limited historically and culturally. The model cannot be applied universally, and even within the USA it has become dated. Nevertheless, while the explanatory power of the model is limited in today’s world, some of the constituent land-use zones can still be recognised, and it remains a useful pedagogic device against which to test real-world cities. The earliest constructive criticism of Burgess’s model emerged from an analysis of the internal residential structure of 142 American cities by Hoyt (1939).10 By mapping the average residential rent values for every block in each city, Hoyt concluded that the general spatial arrangement was characterised better by sectors than by concentric zones (Figure 7.3). The resultant model of urban land use starts with the assumption that a mix of land uses will develop around the city centre, then, as the city expands, each will extend outwards in a sector. In this manner the high-rent neighbourhoods of the wealthy follow a definite path along communication lines, on high ground free from flood danger, towards open country, or along lake or river fronts not used by industry. Conversely, lowincome groups with limited housing choice consume the obsolete housing of the wealthy, now converted into apartments, or occupy less desirable zones. The sectors undergo growth and change over time but according to the model, outward change occurs only Land use in the city 193 within sectors. The whole sector may not be geographically or socially similar at any one time, with, for example, better-quality housing moving towards the periphery, leaving decaying housing nearer the centre. A major contrast between the models of Burgess and Hoyt is that whereas residential change is stimulated on the demand side in Burgess’s model, with immigrants competing for inner-city housing, Hoyt stresses supply-side mechanisms, with the construction of new housing for the middle classes on the urban periphery (and subsequent filtering of vacated dwellings) being the catalyst for sociospatial change. Hoyt’s model does not replace the concentric-zone scheme but extends it by adding the concept of direction to that of distance from the city centre. A major weakness of the theory is that it largely ignores land uses other than residential, and it places undue emphasis on the economic characteristics of areas, ignoring other Figure 7.3 Hoyt’s sector model of urban land use, and its application in Sunderland Figure 7.4 Harris and Ullman’s multiple-nuclei model of urban land use Urban geography 194 important factors, such as race and ethnicity, which may underlie urban land-use change. The excessive simplicity of the concentric ring and sector models of the city was addressed by Harris and Ullman (1945), who observed that most large cities do not grow around a single CBD but are formed by the progressive integration of a number of separate nuclei11 (Figure 7.4). The location and growth of these multiple nuclei are determined by a number of controlling factors: 1. Certain activities require specialised facilities and congregate where these are available. Industry, for example, requires transport facilities and is often located close to railway lines, major roads or port facilities. 2. Similar activities group together to profit from external economies of association, leading to the emergence of specialised legal districts or financial quarters. 3. Some activities repel each other owing to negative externality effects, as seen in the separation of high-income residences from industry. 4. Some activities which could benefit from a central location in or near the CBD, but which cannot afford the high rents demanded, must locate elsewhere. Warehousing or grocery wholesaling are examples of activities that require large structures and would benefit from a central location but are forced to ‘trade off space for accessibility (Box 7.3). The value of the Harris and Ullman model lies in its explicit recognition of the multinodal nature of urban growth. Furthermore, they argue that land uses cannot always be predicted since industrial, cultural and socio-economic values will have different impacts on different cities. While the Burgess zonal pattern and, to a lesser extent, the Hoyt sectoral pattern suggest inevitable predetermining patterns of location, Harris and Ullman suggest that land-use patterns vary depending on local context. Hence the multiple-nuclei model may be closer to reality. In practice, elements of all these models may be identified in many large Western cities. In London, for example, the annular rings of growth reflect Burgess, and a clear distinction can be drawn between an older and poorer inner city and more affluent and modern outer suburbs. Superimposed on this is a pattern of sector development with a zone of local authority and workers’ dwellings from the latter part of the industrial revolution extending from the East End to Dagenham and beyond. To the north and west an affluent residential sector extends from Mayfair to St John’s Wood into Hampstead and on into the ‘stockbroker belt’ of the Chiltern Hills. Finally, multiple nuclei can be found at various scales, the most evident being the financial centre of the City or the concentration of medical services around Harley Street. One of the most severe criticisms of the ‘classical’ models of urban land use referred to their economic bias and consequent neglect of cultural influences on urban land-use patterns. In an early study, Firey (1947) demonstrated that neither the concentric zone nor a sector theory was adequate in explaining land-use patterns in Boston MA, where noneconomic considerations, centred on ‘sentiment and symbolism’, lay behind the spatial juxtaposition of the fashionable residential area of Beacon Hill and an area populated by low-income immigrants and their descendants.12 Firey’s work was significant in illustrating how social values could override economic competition as the basis for sociospatial organisation. Firey recommended a ‘cultural ecology’ approach instead of urban ecology in order to take into account specific cultural and historical factors influencing a city’s land-use patterns. In this he anticipated many of the arguments of postmodernism. Land use in the city 195 Subsequent refinements of the ecological approach have set aside the crude biotic analogy but have retained useful concepts such as natural areas, albeit reformulated as ‘social areas’ or ‘neighbourhood types’ (see Chapter 18). Other work on ecological patterns in cities has sought to reform the traditional models to provide concepts of more direct relevance to contemporary urban society. Four of these merit further consideration. BOX 7.3 The trade-off model of urban land use The mainspring of the concentric-zone model of urban land use is the expansion of the inner zone outwards. This movement is triggered by excessive demand for central city land. The neo-classical economics ‘trade-off’ model employs the concept of bid-rent curves to explain why demand for land, and therefore land-use patterns, vary across the urban area. The basis of the model is the relationship between accessibility and land rent. The more accessible a location the greater the demand for it, which is reflected in the distribution of land values. In the model the city centre is assumed to be the most accessible and therefore most valuable location. Since some land uses place greater importance on accessibility, they are prepared to pay higher rents for central locations. Urban geography 196 As the model shows, under normal conditions a department store will always outbid a housing developer for a central city site because the store’s success is more dependent on its being accessible to as large a population as possible. Land uses that need to be accessible but which cannot afford to pay the rents demanded for sites in the CBD tend to locate in the surrounding zone in transition. In this way the CBD expands outwards and so sets the pattern of city growth in motion. Different land uses evaluate sites in terms of attributes such as size and proximity to desired facilities (such as clients, workplace or open country). Since each is limited in terms of what it can afford to pay for any site, consumers may have to trade off space against accessibility. (For example, a household may occupy a small city-centre apartment to gain accessibility to workplace or leisure facilities, trading these off against a more spacious single-family dwelling in a suburban commuter location.) As a result of such trade-offs, different land uses occupy different locations in the city in a process that produces the urban land-use pattern. MODIFICATIONS OF THE CLASSICAL URBAN MODELS MANN’S MODEL OF A TYPICAL BRITISH TOWN One of the limitations of the classical ecological models was their specific focus on the US city. Mann (1965) combined elements of the Burgess and Hoyt models in his model of a typical medium-size British city.13 The model also incorporated a climatic consideration relevant to the UK by assuming a prevailing wind from the west. As Figure 7.5 shows, in the model: 1. The best residential area (A) is located on the western fringe of the city, upwind of and on the opposite side of town from the industrial sector (D). 2. The areas of the working class and the main council estates (C) are located close to the industrial zone. 3. The lower middle-class housing (B) borders on each side of the best residential area. 4. The model also identifies a CBD, a transition zone, a zone of small terraced houses in sectors C and D, larger housing in sector B, large old houses in sector A, post-1918 residential areas with post-1945 housing added on the periphery, and dormitory settlements at commuting distance from the city. KEARSLEY’S MODIFIED BURGESS MODEL Kearsley’s model was an attempt to extend Mann’s model of urban structure to take into account contemporary dimensions of urbanisation such as the level of governmental involvement in urban development in Britain, slum clearance, suburbanisation, decentralisation of economic activities, gentrification and ghettoisation14 (Figure 7.6). Manipulation of the model’s various elements—such as the extension of inner-city blight, minimisation of local and central government housing, and expansion of recent lowdensity suburbs—offers a North American variant of the basic model. Land use in the city 197 Figure 7.5 Mann’s model of a typical medium-size British city Source: P.Mann (1965) An Approach to Urban Sociology London: Routledge VANCE’S URBAN-REALMS MODEL By extending the principles of the multiple-nuclei model, Vance (1964) proposed the urban-realms model.15 The key element is the emergence of large self-sufficient urban areas each focused on a downtown independent of the traditional downtown and central city. The extent, character and internal structure of each ‘urban realm’ is shaped by five criteria: 1. the terrain, especially topographical and water barriers; 2. the overall size of the metropolis; 3. the amount of economic activity within each realm; 4. the internal accessibility of each realm in relation to its dominant economic core; 5. interaccessibility among suburban realms. Particularly important here are circumferential links and direct airport connections that no longer require them to Urban geography 198 interact with the central realm in order to reach other outlying realms and distant metropolises (Figure 7.7). Though conceived on the basis of work on the San Francisco Bay area, the model has subsequently been applied to describe the general land-use structure of other US cities.16 Figure 7.6 Kearsley’s modified Burgess model of urban land use Source: G.Kearsley (1983) Teaching urban geography: the Burgess model New Zealand Journal of Geography 12, 10–13 WHITE’S MODEL OF THE TWENTY-FIRST-CENTURY CITY Since publication of the three classical models of urban land use many new forces have come to influence urban growth. These reflect societal changes such as Land use in the city 199 deindustrialisation of the urban economy, the emergence of a service economy, the dominance of the automobile, a decrease in family size, suburban residential developments, decentralisation of business and industry, and increased intervention by government in the process of urban growth. White (1987) proposed a revision of the Burgess model that incorporates these trends in order to guide our understanding of the twenty-first-century city17 (Figure 7.8). The model comprises seven elements: 1. Core. The CBD remains the focus of the metropolis. Its functions may have changed over the years but it still houses the major banks and financial institutions, government Figure 7.7 Vance’s urban realms model Source: J.Vance (1964) Geography and Urban Evolution in the San Francisco Bay Area Berkeley, CA: Institute of Local Government Studies, University of California buildings and corporate headquarters as well as the region’s main cultural and entertainment facilities. A few large department stores retain flagship establishments downtown, but most retailing has moved with the affluent population to the suburbs, with many remaining outlets being speciality stores catering for daytime commuters. 2. Zone of stagnation. While Burgess expected investors from the CBD to expand into the zone in transition, White depicts the area as a zone of stagnation. He argues that rather than extending outwards spatially, the CBD expands vertically. Lack of investment in the zone was compounded by the effects of slum clearance, highway construction, and the relocation of warehousing and transport activities to suburban areas. Although Urban geography 200 some older US industrial cities (such as Cleveland OH) have sought to revitalise the zone through conversion of buildings into entertainment, shopping and residential areas, younger cities (such as Dallas TX) have abandoned the zone altogether. 3. Pockets of poverty and minorities. These comprise highly segregated groups living on the fringes of society, including the homeless, Figure 7.8 White’s model of the twenty-first-century city Source: M.White (1987) American Neighborhoods and Residential Differentiation New York: Russell Sage Foundation addicts, dysfunctional families, the underclass, and members of minorities. The surroundings reflect their status, being dominated by deteriorating housing in blighted neighbourhoods. These slum areas are found mostly in the inner city skirting the zone of stagnation, but a few are also located in older suburbs. 4. Elite enclaves. The wealthy have the greatest choice of housing environment and are able to insulate themselves from the problems of the metropolis. Most live in neighbourhoods on the urban periphery in expensive houses on spacious lots. ‘Gilded’ neighbourhoods also remain in the central areas of older large metropolises. 5. The diffused middle class. These areas occupy the largest area of the metropolis and are spatially concentrated between the outer edge of the central city and the metropolitan fringe. This suburban zone is characterised by social diversity: ■ In the interior sections are older settled neighbourhoods which are now in transition as the original settlers have raised their families and are moving to other dwellings. Some of these neighbourhoods, often adjacent to the central city, are attracting the black middle class. Although large numbers of African-Americans have moved to the suburbs in recent decades they remain highly segregated. Land use in the city 201 ■ Farther out there are the archetypal suburban communities comprising married couples with small children living in single-family detached homes built on spacious lots. The suburbanisation of business and industry means that nucleations of other social groups are also present, with working-class families living in more modest neighbourhoods, the elderly in garden apartments and retirement communities, singles in apartment complexes and ethnic minorities in their own enclaves. 6. Industrial anchors and public sector control. Industrial parks, universities, R&D centres, hospitals, business and office centres, corporate headquarters and other large institutional property holders can exert a major influence on patterns of land use and residential development. Institutional actors and other members of a local growth coalition (see below) can pressure city government to modify zoning, lower taxes and construct infrastructure. The location of such activity (e.g. the siting of a large shopping mall) is of considerable importance in shaping the urban structure. 7. Epicentres and corridors. A distinguishing feature of the evolving twenty-first-century metropolis is the emergence of peripheral epicentres located at the convergence of an outer beltway and axial superhighway and providing a range of services to rival those of the CBD. Corridor developments, as along Route 128 near Boston or the Johnson Freeway in Dallas, can also act as a focus for intensive economic activity. The classical models together with more recent modifications provide a powerful insight into the changing structure of the Western city. A major deficiency, however, is that only limited explicit consideration is given to the processes underlying the revealed patterns of land use. This criticism underlay development of the political economy interpretation of urban change. A POLITICAL-ECONOMY PERSPECTIVE Despite some success in describing general patterns of urban land use the traditional ecological models, and in particular their positivist basis in neoclassical economies, were criticised in the early 1970s as: 1. mechanistic, viewing humans as rational decision-makers operating in an abstract environment; 2. ideological, retaining the myth of value-free research while legitimising market capitalism and retention of the socio-economic status quo; 3. devoid of ethical content, since questions of equity and fairness of social conditions and resource allocation were excluded. Energised by the development of a policy-oriented and relevance perspective in human geography, urban geographers sought interpretations of urban change that revealed the structural forces underlying observed land-use patterns. This led researchers such as Harvey (1975) and Castells (1977) to focus explicitly on the place of the city in the capitalist mode of production.18 This neo-Marxist perspective was based on the premise that: Urban geography 202 if the city is considered to start with as a market where labour, power, capital and products are exchanged, it must be equally accepted that the geographical configuration of this market is not the result of chance; it is governed by the laws of capital accumulation. (Lamarche 1976 p. 86)19 The city in advanced capitalist societies is regarded as a particular built form commensurate with the fundamental capitalist goal of accumulation (the process by which capital is reproduced at an ever-increasing scale through continued reinvestment of profits). Thus, as well as concentrating the means of production through agglomeration, cities also develop an infrastructure that facilitates the geographical transfer of profits in search of optimum investment opportunities. Harvey (1985) refers to this process as the ‘circulation of capital’ and sees it as a key factor in urban development.20 As Figure 7.9 shows, Harvey envisaged three circuits of capital: Figure 7.9 Harvey’s model of the circulation of capital Source: D.Harvey (1978) The urban process under capitalism International Journal of Urban and Regional Research 2(1), 101–31 1. The primary circuit refers to the structure of relations in the production process (e.g. the manufacture of goods for sale). Surplus value (profits) created in the production process is either reinvested in the primary circuit with a view to generating further profit or, in the event of overproduction (or underconsumption), may be channelled via the capital market into the secondary or tertiary circuits. 2. The secondary circuit involves investments in fixed capital, such as the built environment (e.g. property development), in the expectation of realising profits either: ■ in the form of rental income from the use value of the building; or ■ from the enhanced future exchange value (sale price) of the building. 3. The tertiary circuit involves investment in science and technology that leads ultimately to increases in productivity, or investment in improving labour capability through Land use in the city 203 education or health expenditure. Much of this investment is made collectively by the state, since individual capitalists are unlikely to take a long-term view of the potential advantages. Significantly, there is a limit to the process of capital transfer from the primary to secondary circuits (since the market can absorb only so many office buildings or leisure centres). When this point is reached, investments become unproductive and the exchange value of capital put into the built environment is reduced or in some instances lost completely (resulting in bankruptcy for some investors and redundancy for the workforce). Since these events occur in specific locations, the outcome is an uneven geography of development, with some areas (e.g. the CBD) benefiting from an inflow of capital investment and other areas (e.g. ‘off-centre’ commercial areas) tending to decline owing to lack of capital investment. However, the devaluation of exchange value does not necessarily destroy the use value of a building or site that can be used as the basis of further development. Since growth (the accumulation of profits) is a constant requirement of the capitalist mode of production, devaluations of fixed capital (e.g. an existing building) represent one of the main ways in which capitalism can speed the accumulation of new capital value (e.g. by redevelopment). This process is illustrated graphically by Feagin (1987), who describes how a property developer in Houston demolished a structurally sound twenty-storey office block to make way for an eighty-two-storey supertower.21 There is, therefore, a major contradiction in the capitalist city between the capitalist dynamic of accumulation (provoking urban growth and change) and the inertia of the built environment (which resists urban change). As Harvey (1981 p. 113) explained: under capitalism there is then a perpetual struggle in which capital builds a physical landscape appropriate to its own condition at a particular moment in time, only to destroy it, usually in the course of a crisis, at a subsequent point in time.22 As we shall see later (Chapter 14), the transfer of capital from the primary to secondary circuit in the form of speculative investment in the built environment underlay the property boom experienced by the large cities of Europe, North America and Australia during the late 1970s and early 1980s in the wake of the oil price shock of 1973. The transfer of capital from the primary to the secondary circuit of Harvey’s model has also been postulated to be a major cause of post-Second World War suburbanisation in the USA.23 Examination of the relationship between the circulation of capital and postwar suburbanisation in the USA enables us to relate theory to the real world and illustrates how the restructuring of capital also involves a restructuring of space. According to this thesis, as a result of the inability of the domestic market to absorb the industrial surpluses that built up as the American war machine returned to peace-time production, other labour- and capital-absorbing activities were promoted by successive governments in the 1950s and 1960s, including suburban capital formation. By engineering a shift of investment into the secondary circuit, the state and specialist financial institutions avoided a crisis of over-accumulation in the primary circuit and simultaneously stimulated new demand for industrial goods in the housing and transportation sectors (Box 7.4). The role of government in promoting post-war low- Urban geography 204 BOX 7.4 The over-accumulation crisis and post-war suburbanisation in the USA The scale of post-Second World War peripheral expansion in the USA was unprecedented. Between 1950 and 1955, as the US population grew by 11.6 million, 79 per cent of the growth was in the suburbs, and the total volume of new suburban residential construction was three times that in the central cities.24 This process was fostered by federal government housing and transport policy. While the 1949 and 1954 Housing Acts laid out a strategy for slum clearance and the rebuilding of inner-city neighbourhoods, in practice federal subsidies flowed overwhelmingly to suburban housing development. In the period 1950–4 the number of low-rent public housing starts fell from 75,000 units to 20,000 units annually, while funding for private suburban housing was liberalised, with increased upper limits for Federal Housing Administration (FHA) home mortgage insurance and relaxed FHA loan terms for single-family dwellings in suburban areas. The electoral significance of the tax deductions associated with owner-occupied dwelling also acquired increasing significance as the level of home-ownership grew. As Badcock (1984 p. 135) observed, ‘this powerful orientation in the US towards home owners at the expense of tenants, new construction in the suburbs in preference to the rehabilitation or selective redevelopment of inner city housing, and the largest builders in the industry, had far-reaching consequences for the metropolitan built environment’.25 The federal highway programme was also of major importance to post-war suburbanisation. Between 1944 and 1961 the federal government allocated its entire transport budget of US$156 billion to roads.26 The major commitment to highway construction was signalled in the 1956 Federal-Aid Highway Act, which provided 90 per cent federal support for the construction of the 65,156km (40,462 mile) Interstate Highways and Defence System. The urban expressway system, constructed as part of the interstate highway programme, served to connect dormitory suburbs with downtown office jobs. Equally important, the 1956 Federal-Aid Highway Act required that central city bypass routes be built into the urban highway network, and by the early 1970s more than eighty of these circumferential beltways had been constructed in the outer areas of US metropolitan areas.27 The enhanced accessibility of suburban highway-oriented locations attracted regional shopping centres, industrial and office parks, and apartment estates to the interchanges along the beltways. Berry and Kasarda (1977) estimated that between 1960 and 1970 the growth of blue-collar employment in the suburbs was 29 per cent compared with a loss of 13 per cent for the central cities; while for white-collar jobs the suburbs recorded an increase of 67 per cent over the period, compared with 7 per cent for the central cities.28 The over-accumulation/under-consumption thesis aids our understanding of the suburbanisation process in the USA but is of doubtful relevance to the situation in Britain, where successive post-war governments have sought to control the outward spread of cities through the operation of a centrally directed planning system (see Chapter 8). In consequence the physical containment of British cities differentiates them from the leapfrog subdivisions and suburban sprawl of US metropolitan centres. Land use in the city 205 density suburbanisation by subsidising the costs of land development has also been explored for metropolitan Toronto, where, between 1955 and 1964, discriminatory public funding of highway and service infrastructure supported expansion of the outer suburbs.29 The political economy approach affords valuable insight into the key processes and agents responsible for the production of the built environment of the capitalist city. This perspective has exposed the roles of and relationship among various factions of capital in influencing urban change. It also highlights the impact that economic and political processes located outside the territory of any particular city have on its internal structure and development. The political economy perspective has been criticised, however, for its apparent reification of the market, and for giving insufficient attention to human geography. This critique echoes the view of Form (1954), who argued that the social organisation of the land market was more relevant than economic models that depicted the city as a free market in which individuals competed impersonally.30 In a study of Lansing MI, Form identified the relationships between the dominant agents in the urban land market, and the ways in which each influenced land-use patterns. This viewpoint developed into the concept of urban managerialism.31 It also informed the humanistic critique of political economy which refocused research attention on the role of human agency in the production of the built environment (see Chapter 2). MAJOR ACTORS IN THE PRODUCTION OF THE BUILT ENVIRONMENT The land-development industry comprises a variety of builders, subcontractors, architects, marketing agents, developers and speculators, together with their legal and financial consultants. During the conversion of rural land into occupied housing a plot may pass through the ownership of at least five different actors: a rural producer, a speculator, a developer, a builder and finally a household. Assisting with land transfers at each stage is a set of facilitators, including real-estate agents and financiers. Finally, at local and central government level, planners and officials oversee the development process to varying degrees according to the prevailing social formation (Figure 7.10). The motives and methods of these participants vary considerably (Box 7.5). While bearing in mind the interrelated nature of the land-development process, we can, in the interests of clarity, examine the operation of several of these individual actors in detail. SPECULATORS Property speculators—either individual entrepreneurs or corporations—purchase land with the hope of profiting from subsequent increases in property values. Speculative activity is a characteristic of capitalist urban development that occurs throughout the urban arena. As we saw in our discussion of urban political economy, speculation in the central city can contribute to the creation of slums prior to revitalisation of a neighbourhood either through private-sector upgrading or gentrification or through publicly financed rehabilitation (see Chapter 10). The effects of speculative activity on the residents of the existing built-up area can be pronounced, often leading to displacement and the destruction of communities. The impact of speculative Urban geography 206 development on land use is seen most starkly around the peripheries of cities. In the USA, the area under urban use is over 10 million acres (4 million ha), but ‘twice as much land is withdrawn from other uses because of the leapfrogging which characterises much suburban growth’.32 Leapfrog development usually occurs as builder-developers try to avoid land that is tied up in complex legal arrangements or is being held by speculators in anticipation of very high profits. Some indication of the level of profits possible is provided by the fact that in Los Angeles the price of residential land increased at a rate of 40 per cent per year in the late 1970s, in part owing to speculative activity. In areas BOX 7.5 Major agents and motives in the capitalist The motives and methods of different participants in the land-development process vary considerably. Participants include the following groups: 1. Rural producers. Landowners, who are primarily concerned with the productive capability of their land, the most obvious group being farmers. 2. Speculators. These may own land that is still in productive use, but their basic interest lies in its appreciating value. (They hope to buy low and sell high.) Their decisions are based on financial considerations (such as depreciation rates, and the comparative viability of alternative investment opportunities). Speculators assemble land for subsequent development but can also withhold land from the market, thereby forcing prices up. 3. Developers. These can be categorised as subdividers and builders. In the USA the former install basic infrastructure and facilities while the latter construct and sell houses on the prepared lots. In Britain the role of subdivider and builder is usually combined. 4. Households. Those who purchase or lease the units are motivated by the functional utility of the house as a place to live (use value) and the land-development process improvement, or at least maintenance, of the financial investment represented by their property (exchange value). 5. Real-estate agents. These purvey information and act as intermediaries between buyer and seller. Their rewards come from commission charged on each land transaction completed, so agents have a vested interest in promoting property transfers. Some estate agents also engage in speculation. 6. Financiers. These provide the funds necessary to the development process. Their decisions are based on a combined desire to obtain the highest possible rate of return on an investment or loan and to minimise or avoid risk. 7. Other facilitators. Other professionals involved include lawyers who represent clients in disputes that may arise during the development process, as well as consultants who advise other agents. 8. Government. All governments influence the process of urban development, although the level of involvement varies. In Britain a centrally directed planning system determines the framework within which local authorities operate. In North America local governments are less constrained by national legislation. Land use in the city 207 around expanding cities in Japan the multiplier between land values for agricultural use and urban use is over 200-fold,33 while around some British towns such as Reading in Berkshire the price of £2,000 per acre for agricultural land can rise to £1 million if planning permission for residential development exists.34 Such inflationary costs are, of course, built into the final purchase price for suburban housing. On the urban fringe, the effects of land speculation constitute one of the most important impacts on land use. For agriculture, speculative land-holding can have both positive and negative effects. Beneficial interaction can occur through the rental back to farmers of farmland that has been purchased by non-farm interests. Provided that the lease conditions are not onerous, from a purely economic perspective it may be attractive for a farmer to rent land rather than to purchase it and encumber the business with a heavy mortgage, thus releasing more of the farmer’s capital for improvements. On the other hand, rising land prices and land speculation make farm enlargement costly, and where tax is levied on land values, as in the USA, sales of surrounding land can push up property taxes. Tax pressures coupled with other urban shadow effects such as pollution, trespass, theft and vandalism may eventually force the suburban agriculturalist to sell out to speculators. The impact of potential urban development also affects land husbandry practices on the fringe. Where urban pressures are strong, farmers may become active speculators, disinvesting in their farms while anticipating a large capital gain from the sale of their land in the near future. Some farmers may ‘farm to quit’ or attempt to ‘mine’ the soil’s fertility, while others may ‘idle’ their farmland. In the absence of effective landmarket regulation, leaving land idle may be a perfectly rational land-use response to the economic incentives created by the urban-fringe property market. Berry and Plaut (1978), for example, estimated that for every acre converted to urban uses in the North-Eastern USA another was idled owing to urban pressures. In Japan, one-third of paddy fields, amounting to 750,000 ha (2,900 square miles) are left idle by speculators in the hope of conversion to urban use.35 Farmers living under less intense urban pressures may be involved in a more passive form of land speculation, watching the appreciation of land values with a view to selling for a large profit on retirement.36 REAL-ESTATE AGENTS Although the principal role of real-estate agents is as middlemen between buyers and sellers of property, some adopt a broader remit and may operate in the assembly of small land parcels for development or as speculators in the urban land market.37 Direct involvement in the land market can be particularly profitable during periods of inflation and can lead to manipulation of the land market. Gutstein (1975) found that in a resurgent inner-city neighbourhood in Vancouver in the early 1970s some properties were ‘sold’ several times in a single year between holding companies with the same owner in order to force up rent levels in the local market.38 Urban geography 208 Figure 7.10 Principal actors in the production of the built environment Similar instances of land speculation have been recorded in areas of racial transition where realestate agents also often play an active role in block-busting—a process whereby members of minority (black) groups obtain entry to residential areas previously reserved exclusively for the majority (white) population39 (see Chapter 10). A real-estate speculator who has bought a property in a block is able to resell to a black family for a substantial profit when demand from the minority group reaches a critical level.40 Frequently this initial sale stimulates existing home-owners to sell up (owing to fears about falling property prices and/or racial bias) and reduces demand from other majoritygroup buyers.41 It can be argued that the blockbuster is providing much-needed accommodation to groups that are discriminated against in the open housing market. However, in some cases the blockbuster is guilty of exploiting the anxieties of residents (possibly forcing them to sell below the market price) and the housing poverty of minority groups (who are willing to pay above market prices) to realise a significant capital gain. Foreman (1971) estimated that in Chicago during the 1950s and 1960s a small minority of real-estate agents triggered the blockbusting of 70,000 white families over a decade.42 Similarly, Orser (1990) has described how the west Baltimore neighbourhood of Edmondson Village experienced real-estate blockbusting and massive ‘white flight’ in response to racial transition between the late 1950s and the mid-1960s. Over a period of less than a decade almost total racial change took place within one of the two census tracts, a new population of 11,000 replacing a previous population of 9,000.43 Land use in the city 209 The social costs of blockbusting in terms of racial disharmony, resentment and harassment are less easy to calculate. Racial and ethnic steering by real-estate agents is a common practice that contributes to patterns of residential segregation44 (see Chapter 18). More generally, real-estate agents can influence the social composition of neighbourhoods by directing people to particular housing areas on the basis of their perception of the market, their partial knowledge of the city, and the normal operating territory of their company.45 FINANCIAL INSTITUTIONS Financial institutions have grown in importance in both Britain and North America with the decline in private rented housing and the growth of home ownership. In common with developers, financial institutions seek to maximise profits and minimise risks. Since these opportunities are differentiated over the urban area, financiers adopt spatially discriminating lending practices, a fact that will have a significant impact on the location of new construction as well as on maintenance and improvements to existing structures. The practice of red-lining areas by mortgage lenders is well documented. Lambert (1976 p. 33) found in a study of Birmingham that ‘the building societies deliberately steered clear of certain of the older areas…although they usually emphasised that there was no written policy which precluded houses in these areas’.46 A contemporary study of Leeds reported that an applicant for a building-society mortgage on a property in Headingley was informed that ‘it was doubtful if a mortgage would be available…owing to the nearness of the blue zone’—that is, an area with a high proportion of students and immigrants (Weir 1976).47 A survey of mortgage finance managers in Bristol found that all the managers had heard of red-lining…and said they did not practise it. One manager stated categorically ‘we do not red-line’. Later in the discussion he pointed to the St Paul’s area on the map and said that ‘there are certain areas in the city in which we won’t lend’. (Bassett and Short 1981 p. 286)48 Discrimination has also been practised on racial grounds. In the USA, especially prior to the civil-rights legislation of 1968, bank lending to African-Americans was based on the twin criteria of residence in an established black neighbourhood and in a ‘good area’.49 More recently, Stegman (1995) reported that African-Americans and Hispanics were 60 per cent more likely to be rejected for mortgage loans than whites even when financial and employment status and neighbourhood characteristics were controlled for.50 Such practices may be stimulated by economics as well as prejudice, since banks, concerned to protect property values in areas where they have invested, may be uncertain about the long-term effect on values of racial transition. This inherent need to minimise risk continues to influence the spatial pattern of lending practices. Harvey (1975) has demonstrated how various financial agencies in Baltimore segmented the city into a series of submarkets on the basis of the individual lending preferences51 (Box 7.6). The practice of red-lining areas perceived as poor risks has also been followed by the insurance industry.52 Urban geography 210 Despite the passage of the US Mortgage Disclosure Act 1975, which requires lending agencies to disclose the geographic pattern of mortgage awards, discriminatory practices may continue largely through covert means such as discouraging wouldbe borrowers with higher interest rates, high down payments, lower loan to value rates and shorter loan maturity terms for property in red-lined areas.53 In Atlanta GA analysis of mortgage-loan decisions in the 1980s and 1990s revealed that even after controlling for differences in borrower characteristics ‘prime conventional home-purchase capital avoids a broad swath of Atlanta’s inner-city BOX 7.6 Housing submarkets in Baltimore MD Harvey (1975) demonstrated how various financial agencies in Baltimore segmented the city into a series of submarkets on the basis of their individual lending preferences. The table shows the specialised niche in the Under $7,000– $10,000– $12,000– Over $7,000 $9,999 $11,999 $14,999 $15,000 Private State S&Ls Federal S&Ls Mortgage banks Savings banks Commercial banks % of city’s transactions in category 39 42 10 7 — 1 16 33 22 24 3 1 13 21 30 29 5 2 7 21 31 23 15 3 7 20 35 12 19 7 21 19 15 20 24 housing-price structure occupied by different lending institutions. Thus, for the lowest-grade housing, 81 per cent of mortgages derived from either state savings and loan (S&L) companies or from private arrangements. The higher perceived risk attached to this submarket made it unattractive to the larger banks and federal savings and loan firms. The spatial expression of these lending practices is shown in the second table. Thus the inner city was heavily dependent on cash transactions and money raised from private sources. In the traditional ethnic districts of east and south Baltimore, small-scale community-based S&Ls) dominated housing finance. Middle-income white areas of north-east and south-east Baltimore were basically served by federal S&Ls. The affluent areas drew on the financial resources of savings banks and commerical banks as well as federal S&Ls, while areas of high turnover and racial change were characterised by mortage-bank finance combined with Federal Housing Administration (FHA) mortage insurance. The black community of west Baltimore, historically discriminated against, also relied heavily on mortage bank finance supported by FHA insurance even though in income terms it was broadly similar to those white areas setrved by federal S&Ls. Land use in the city 211 District Total Sales Transactions by source of funds (%) Sales ho per insured uses 100 (%) sold prop erties Cash Pri Fed State Mort Comm Sav Other FHA VA vate eral S&Ls gage ercial ings bank bank bank S&Ls Inner city 1,199 1.86 65.7 15.0 646 2.33 64.7 15.0 1. East 553 1.51 67.0 15.1 2. West 760 3.34 39.9 5.5 Ethnic 579 3.40 39.7 4.8 1. E. Baltimore 181 3.20 40.3 7.7 2. S. Baltimore 99 2.40 40.4 8.1 Hampden 497 2.32 30.6 12.5 West Baltimore 322 3.16 28.3 7.4 South Baltimore 2,072 5.28 19.1 6.1 High turnover 1. North- 1,071 5.42 20.0 7.2 west 2. North- 693 5.07 20.6 6.4 east 308 5.35 12.7 1.4 3. North Middle 1,077 3.15 20.8 4.4 income 1. South- 212 3.46 17.0 6.6 west 2. North- 865 3.09 21.7 3.8 east 361 3.84 19.4 6.9 Upper income Note: VA, Veterans’ administration 3.0 2.2 4.0 6.1 5.5 12.0 14.3 9.2 43.2 43.7 2.2 2.2 2.3 2.0 2.4 0.5 0.5 0.4 0.8 1.0 0.2 0.1 0.4 0.9 1.2 1.7 1.2 2.2 2.2 2.2 2.9 3.4 2.3 2.6 3.2 7.7 41.4 0.6 – – 2.2 0.6 0.6 5,102 18.2 26.3 4.0 12.1 11.7 22.3 – 1.6 3.0 3.1 – 14.1 2.0 7,059 6.0 25.8 4.2 8,664 22.7 13.4 13.4 1.9 4.0 9.0 22.7 10.6 8,751 13.6 14.9 32.8 1.2 5.7 6.2 38.2 9.5 9,902 9.7 13.8 40.9 1.1 2.9 4.5 46.8 7.4 9,312 14.4 16.5 29.0 1.4 5.6 5.9 34.5 10.2 9,779 25.3 18.1 13.1 29.8 17.0 8.6 0.7 15.9 1.9 8.7 12.7 31.5 15.5 12,330 9.0 17.7 11.1 12,760 29.2 1.0 10.8 11.7 30.2 17.0 12,848 8.5 15.1 1.1 1.4 0.6 0.7 0.7 Ave rage sale price ($) 3,498 3,437 3,568 6,372 6,769 30.0 19.2 7.0 2.0 8.2 8.2 14.7 9.7 12,751 23.5 10.5 8.6 7.2 21.1 2.8 11.9 3.6 27,413 Source: D.Harvey (1975) The political economy of urbanisation in advanced capitalist societies, in G.Gappert and H.Rose (eds) The Social Economy of Cities Beverly Hills, CA: Sage, 119–62 Urban geography 212 neighbourhoods, as well as suburbs perceived at risk for race and class transition’ (Wyly 2002 p. 95).54 Despite the US Home Mortgage Disclosure Act 1995, and identification of expanding home-ownership as a national priority in the American Home-ownership and Economic Opportunity Act of 2000, red-lining of certain neighbourhoods remains a feature of the geography of housing investment in the city. Given the need for financial institutions to operate within a free-market environment, it is unrealistic to expect them not to engage in what they perceive to be ‘economically rational’ discrimination.55 Nevertheless, the practice of redlining ensures that property values will decline, and generally leads to neighbourhood deterioration. It can also engineer a flow of investment away from inner-city areas towards more affluent suburban home-owners. In a study of Boston, for example, it was discovered that levels of reinvestment of savings deposits in inner-city communities were between 3 per cent and 33 per cent, compared with levels of between 108 per cent and 543 per cent in the outer suburbs.56 This flow of capital to the suburbs reflects the wider interests and operations of mortgage finance institutions, many of which are involved in financing and controlling the suburban activities of large construction companies.57 GROWTH COALITIONS While we have so far considered the major agents of urban change independently, in practice growth coalitions operate to foster urban development. These networks, or growth machines,58 exhibit several key characteristics: 1. The critical actors in growth coalitions tend to be those fractions of capital which are most place-bound, such as rentiers, who rely on an intensified use of land or buildings in a particular area for enhanced profits. 2. Growth-machine activists are concerned particularly with the exchange value of a place and tend to oppose government intervention that might regulate growth. 3. Growth networks are often combined private public coalitions that support themselves through local pro-growth bureaucracies. In the fragmented local government system of the USA (see Chapter 20), local growth elites can exert an influential role in electing local officials, ‘watchdogging’ their activities and scrutinising administrative detail. However, since local-government bureaucracies are also sensitive to citizen demands (for reasons of political legitimation), the pro-growth stance can be modified by popular pressure. 4. The power of pro-growth business coalitions over local political strategy is generally less in the UK than in the USA, in part because of a more centralised political and planning system, and greater central funding of local services. Nevertheless, over the period 1979–97 successive New Right governments in the UK encouraged greater private-sector involvement in local economic planning (see Chapter 16) and enhanced the power of business interests in the process of urban development.59 5. Despite a common commitment to growth, the composition of growth networks can vary from place to place and may extend beyond the capitalist class, with, in some cases, membership embracing labour (usually the construction trade unions) or minority groups. Land use in the city 213 6. Within the pro-growth coalition, fractions of capital can pursue strategies that impose negative externality effects on each other, as when the production of a new shopping centre leads to the decay of shopping facilities in older areas. The main ideological battle is between pro-growth and anti-growth factions. The progrowth ideology proclaims that more development results in increased population, greater aggregate sales, more local tax revenues and more local jobs, owing to increased local spending. Benefits therefore flow to those who need employment and those in favour of lower taxes. The anti-growth ideology emphasises the other side of the coin: development brings more people to an area than local institutions can service, thus any downward trend in taxation is overcome in the later stages of growth by the need for greater fiscal expenditure. In addition, development produces pollution, traffic congestion and social pathologies such as higher crime rates. There is also no guarantee that new jobs will go to locals.60 In some cases, growth networks operate in a corrupt manner to exploit rapid development, as for example, when political leaders raise campaign funds by favouring business elites, or public officials receive payments in return for planning and taxation easements.61 The ability of power groups to manage urban growth led Gale and Moore (1975) to formulate the hypothesis of the manipulated city, which contends that urban form is the outcome of conscious manipulation by an alliance of elite interests with social power.62 The extent of the influence of elite coalitions is difficult to assess, however. In contrast to the more regulated process of urban development in Britain and Europe (see Chapter 8), in the ‘free-market’ North American city, growth coalitions of businessmen and municipal politicians have existed since the nineteenth century, supported by an ideology that considers urban growth and commercial growth to be partners. In Canada, in the early 1970s, half the aldermen on the city councils of Toronto, Winnipeg and Vancouver held occupational connections with development interests, and pro-growth parties held control of City Hall in each case.63 In instances of financially weak cities the business/financial community can exercise close to monopoly control, translating their economic power into political power. The financial crisis of New York City in the mid1970s is a classic example of how a financial coalition can benefit at the public expense. By 1977 20 per cent of New York’s budget was committed to interest payments.64 Similarly, in Cleveland in 1978 the oligopoly of business interests which formed the Cleveland Trust Company (which had interlocking directorates with major local steel, coal, utility and banking companies) was able to exact significant economic concessions from the city in return for arranging a loan. Globalisation has promoted a common ‘growth discourse’ among growth coalitions in many US cities based on the belief that entrepreneur-led effort is required for them to compete successfully for capital and jobs within a hypermobile global economy. The power of growth coalitions to influence urban development remains formidable. In Cleveland during the 1980s the growth coalition aggressively promoted gentrifying neighbourhoods along Lake Erie and the construction of a new downtown sports complex as catalysts to improve the local economic investment climate. In the process consideration of the effects of uneven development on disadvantaged populations was muted.65 It is not surprising that Harvey (1971) characterised much of the political activity of the city ‘as a matter of jostling for and bargaining over the use and control of the hidden mechanisms for redistribution’.66 The Urban geography 214 land-use changes that result from this activity involve an uneven distribution of the net costs and benefits of urban development (Box 7.7). BOX 7.7 Externalities Cities would not have appeared if the collective effect of land-use decisions were not beneficial. However, many of the effects of land-use decisions are negative in that they impose disbenefits on other land users. These negative externalities can take a variety of forms. They may, for example, arise from commercial invasion of residential areas or from traffic problems caused by the continued intensification of city-centre land use, or from the process of urban sprawl. As well as being unpriced, externalities are unplanned (in the sense that their extent and impact are not predetermined by a public agency). They are not always unintended, however, since the pollution of atmosphere or an urban river by a factory may be a deliberate attempt to transfer part of the real cost of production from the producer to society at large. Once the externality problem reaches a critical level, it is likely to provoke a public response in the form of planning intervention, the aim of which is to influence the market mechanism so that either prices are adjusted to reflect full social costs or outputs are controlled to a socially optimal level. Frequently local governments have to carry the costs generated by negative spillovers. Another land-use problem related to the question of externalities is the free-rider issue. A free rider is defined as an individual who refrains from any active initiatives but who benefits from the activities of others. For example, once a neighbourhood begins to decline in status, individual property-owners will frequently, and quite rationally in individual cost-benefit terms, choose to do nothing in the face of the deterioration of their property, waiting instead for someone else to undertake improvements from which they, the property owners, will obtain positive externalities. In most cases the normal effect is that private renewal is postponed indefinitely as individual property-owners await the windfall gains, which in practice have little likelihood of appearing unless government rehabilitation programmes are enacted. THE CENTRAL BUSINESS DISTRICT The central business district or downtown is a principal element of all major models of urban land use. Although some CBDs experience fierce competition from business nuclei elsewhere in the city, most cities retain a strong downtown. Even the archetypal polynucleated metropolis of Los Angeles has a CBD. The key characteristic of the city centre or CBD is its accessibility. Accessibility is a major factor in the locational decisions of central-city land users. Activities requiring an accessible location for their economic viability or functional efficiency tend to gravitate to the CBD. As neo-classical economic theory explains, different activities have different levels of demand for accessibility. The differential need for, and willingness to pay for, accessible sites determines the internal land-use pattern of the central city, with land uses placing greatest value on accessibility outbidding others for central city sites (see Box Land use in the city 215 7.3). The demand for central locations is translated into high land values, which in turn produce a high intensity of land use, expressed most visibly in the high-rise physical structure of the downtown created as developers seek to maximise use of costly sites. The CBD/downtown is typically: 1. the main commercial centre of the city; 2. a centre of retailing; 3. an area where manufacturing once concentrated and where light industry may still exist; 4. a locus for service industries, business offices and financial institutions; 5. a zone with only limited residential land uses. Whereas early geographical studies of the CBD focused on the spatial delimitation of the area,67 more recent analyses have concentrated on the changing nature of the zone. Since the 1950s the character and land-use structure of the CBDs of cities in the UK and the USA have been transformed by several major economic and social processes. These include decentralisation of population (see Chapter 4) and of retail activities (see Chapter 12), deindustrialisation (see Chapter 14), increased socio-spatial polarisation and segregation (see Chapter 18), and reductions in the traditional accessibility of the central city associated with increased levels of car ownership (see Chapter 13). Retail suburbanisation has meant that in many cities of the USA the downtown has become less of a retail and commercial district than an office commercial and cultural and entertainment complex. However, many city centres retain significant advantages for specialist shopping activities, being the point of maximum accessibility to the whole metropolitan area in comparison with the sectoral accessibility of suburban regional centres. Government activities and entertainment and cultural facilities that attract significant numbers of people on a regular basis are also concentrated in the city centre. Many city centres have also developed a strong tourist and convention trade function.68 Furthermore, as we have seen in Chapter 4, in some larger cities a degree of reurbanisation is taking place. In several US cities, in addition to rehabilitation of decaying properties for residential use (see Chapter 11), revitalisation of the CBD has been aided by development of a major facility such as a sports stadium or performing-arts centre, that in turn attracts complementary service functions such as restaurants, hotels and retail outlets. The CBD also remains a major focus of office employment, notwithstanding the growth of ‘suburban downtowns’ or edge cities. Companies often locate their top management in the CBD even if they relocate clerical functions to the suburbs. Even in the electronic age of teleconferencing, corporate managers required to respond to rapidly changing market conditions find value in physical proximity to clients, consultants, bankers, government agencies and their competitors’ headquarters. For some large companies (for example in law, banking or management consultancy) a prestigious downtown address (and distinctive building) assumes symbolic as well as practical significance. More generally, the skyscrapers of the CBD are symbols of both corporate power and the high land values embedded in the capitalist organisation of urban space. The attacks on the World Trade Center in New York on 11 September 2001 was a malign reflection of the symbolic power of architecture. The terrorists were attacking not only a landmark building but a symbol of American global capitalism. Urban geography 216 URBAN ARCHITECTURE Architecture and urban design are linked into the dynamics of urban change. As elements in the political economy of urbanisation, architecture and urban design: 1. stimulate consumption by providing products for different market segments, from new office towers to festival shopping malls; 2. promote the circulation of capital through the creation of a steady supply of new fashions in domestic architecture; 3. aid the legitimation of existing economic and social relations by using the ‘aura’ of urban architecture to suggest the stability, permanence and ‘naturalness’ of the current urban environment. Changes in urban form over time, from pre-industrial to post-industrial/postmodern cities, have been accompanied by change in the dominant form of architecture. In the nineteenth-century city, away from the mass of working-class vernacular construction, architects responded to the processes of industrialisation, modernisation and urbanisation with designs for new public buildings, factories, office buildings and mansion houses that rejected these contemporary processes of urban change and embraced classical features. The results are seen in the impressive edifices of Victorian railway stations and town halls that still dominate the centres of many cities. A second major influence on urban architecture was the modernist movement, which has been characterised as a reaction against the extravagance of bourgeois Victorian society. Promoted by the ideas of William Morris in England and by Frank Lloyd Wright in the USA, modernism was boosted by the ideas of the German Bauhaus school (1919 32), which sought to employ industrial production methods, modern materials and functional designs to promote inexpensive architecture available to all citizens. The international impact of modernism on urban design was heightened by the Swiss architect Le Corbusier, who, in response to the challenge of the automobile age, conceptualised the city as a ‘machine for living’. The most visible impacts of his grand designs on contemporary cities were seen in comprehensive redevelopment plans, high-rise building and urban motorways of the 1960s. In contrast to Le Corbusier’s high-density lifestyle, in the USA Wright’s response to the automobile age was his low-density Broadacre City, designed on the basis of two new technologies: the automobile and mass-produced building using high-pressure concrete, plywood and plastics. The use of new technology and materials was especially evident in the downtowns of post-war cities, where glass-fronted façades that reflected their surroundings became popular for corporate buildings in the 1960s and 1970s. This ‘late modern’ period also produced buildings with dramatic geometrical forms and buildings that emphasised technology with exposed pipes, ducts and elevators. Land use in the city 217 Plate 7.2 The glass façade of an office building in Phoenix AZ, reflects the architectural style of the late modern period The demise of modernist architecture was prefigured by criticism of the uniformity and ‘dullness’ of Corbusian tower blocks,69 and of the insufficient attention given to the needs of people for sociable living spaces.70 For some commentators the demolition of the prize-winning Pruitt-Igoe public-housing project in St Louis in 1972 tolled the death knell of modernist architecture. In contrast to the abstract formalism of modernist architecture, postmodern buildings were designed to be decorative and replete with symbolism. Postmodern architecture is eclectic, often combining (‘double-coding’) modernist styles or materials with historical or vernacular motifs as in ‘neo-traditional’ urban designs for upmarket resort communities such as Seaside in Florida, or in kitsch design as seen in ‘authentic’ Tudor inns. In terms of the broader context of urbanisation and urban change, postmodern architecture (as well as postmodern culture and philosophy) emerged along with the development of global capitalism. As Harvey (1989) observed, although this transition and critique of modernism had been under way for some time, it was not until the international economic crisis of 1973 that the relationship between art and society was disturbed sufficiently to allow postmodernism to become the ‘cultural clothing’ of advanced capitalism.71 The property-development industry was quick to adopt Urban geography 218 postmodern design, to promote product differentiation and to both create and supply demand from an increasingly ‘consumerist’ society. Postmodern design has also stimulated the preservation of historic buildings and urban areas, often linked with the growth of cultural industries and festival shopping developments. The dystopian side of postmodernism is seen most clearly in the proliferation of security and surveillance systems and the creation of ‘fortress’ architecture designed to exclude ‘undesirables’ such as rough-sleepers or panhandlers/beggars from parts of the city.72 ARCHITECTURE AND URBAN MEANING Early attempts to read urban meaning from architectural forms and styles regarded urban architecture as symptomatic of the social and cultural values of its age. Thus Mumford (1938) attributed the development of the avenue in the baroque city of the sixteenth century to the growing militarisation of society and consequent need for troop mobility.73 The difficulty of establishing unequivocally the social values underlying architecture led more recent researchers to analyse urban architecture as a product of specific dominant social groups rather than as a symbol of an historical epoch.74 While Mumford’s attempt to read the city as a text is retained, it is related to particular groups, not to society as a whole. However, there is still a large range of possible interpretations of urban meaning to be derived from analysis of architectural form. The text may be read in many ways— for example, from the viewpoint of class conflict seen in the privatisation of public space, or from that of gender relations, as seen in the characterisation of domestic residential space as primarily a female domain. Further, while architecture can reveal the power of certain groups to influence urban structure, architectural analysis alone is insufficient to explain urban meaning. The diversity of urban imagery also underlines the fact that urban meaning is created and can be manipulated. As we have seen, many actors, from estate agents to local authorities, have a vested interest in presenting places in their most favourable light. Urban meaning may also be contested politically by disadvantaged groups who seek to define urban space in ways that best fit their needs. This is seen in the diverse actions of urban social movements (see Chapters 20 and 29). In short, urban space is socially constructed. THE SOCIAL CONSTRUCTION OF THE URBAN LANDSCAPE The diversity of urban landscapes is a key feature of the postmodern perspective on urban development, although the explanation of how the variety of urban environments is produced runs counter to the economic emphasis of the political economy approach. Postmodernism regards all landscapes as symbolic expressions of the values, social behaviour and individual actions of people marked on a particular locality over time. The built environment of the city is seen as the product of a dialectic interaction between society and space.75 Postmodernism attempts to interpret the nature of this relationship. As we have seen, the approach has been likened to reading the city as a text written by a plethora of different authors which has a series of meanings embedded in it. Postmodern Land use in the city 219 analysis aims to read this text from the viewpoint of the different actors involved in its production.76 The method may be illustrated by considering the concept of patterns of consumption, which is one of the most powerful processes in the urban socio-spatial dialectic. The different patterns of consumption enjoyed by different groups in society are reflected in the polarisation of the retailing landscape as a result of the segmentation of the market into different niches catering for different tastes, preferences and lifestyles. Patterns of consumption influence, and are influenced by, the nature of the physical landscape. The different urban landscapes are the result of this dialectic relationship between social practices and the physical environment. The same urban space can therefore have a different meaning for different social groups, a situation that can lead to conflict over the appropriate use of land. This is evident in debates over public and private space77 (Box 7.8), and is clearly illustrated in the concept of the grade-separated city in which two different environments are interlaced within the architecture of the CBD.78 On the outside—both physically and, often, socially—are streets, pavements (sidewalks), plazas and parks, while inside, ‘eligible’ citizens encounter skywalks, tunnels, concourses and atria. In many North American cities these two environments share the same geographic location in the CBD while functioning as separate entities, one under the purview of the local public sector and the other controlled by a loose collective of private-sector interests. The existence of grade-separated space redefines patterns of interaction between activities and between segments of the downtown population. As Byers (1998 p. 200) observed, ‘in a sense, the downtown environment is turned inside out, or, perhaps more appropriately, outside in. Groups that once shared the same city streets are now spending their days in environments that rarely intersect with one another.’79 In the grade-separated city, distinctions in human-activity patterns between day and night, work-week and weekend, inside and outside, all help to reconfigure the way space in different parts of the CBD is used by different sections of the general public. As we noted in Chapter 2, a major difficulty inherent in the postmodern approach is that the diversity of meanings attached to the urban landscape makes generalisation difficult if not impossible. Fundamentally, just as it is necessary to avoid reification of the economy as the sole force for urban change, so it would be equally short-sighted to view urban space as the product of the actions of voluntaristic agents free of ‘structural’ constraints. The urban landscape is the product of both culture and economy, and a proper understanding of urban environments must be based on explicit acknowledgement of this complexity. In this chapter we have examined the major forces underlying urban structure and land use. The one actor in the urban development process not yet considered is, in some settings, the most influential. The state central and local government exerts both a direct (via planning regulations) and an indirect (via, for example, taxation policy) effect on urban form and structure. We consider this factor in the next chapter. Urban geography 220 BOX 7.8 BIDs to privatise public space in New York City The difference between the urban landscapes of rich and poor is illustrated by the growth of business improvement districts (BIDs) in New York City. A BID can be incorporated in any commercial area, and following the decline of city services in the wake of the fiscal crisis of 1975 a growing number of business and property owners have created BIDs to exert control over the quality of public space in their vicinity. The Grand Central Partnership, a fifty-three-block BID established in 1988, employs uniformed street cleaners and security guards, runs a tourist information booth, has closed a street in front of Grand Central Station to create a new outdoor eating space, issues its own bonds and hires lobbyists to press the state legislature for supplemental funding. Although the architects of this new urban space are apparent, the question of who can occupy the space is less clear. The city government agencies have approved the BID plans, but the local community board representing a variety of local business interests has objected to street closures, and has questioned the effectiveness of the BID services for the homeless and the brusqueness of their removal by the BID security guards. In 1995 an advocacy group, the Coalition for the Homeless, sued the BID for hiring out the homeless as security guards at below the minimum wage. The Times Square BID promised to improve the run-down image of the area and promote a new public culture by establishing a ‘community court’ in a disused theatre, thereby enhancing community control over quality of life. When set up in 1994 the court dispensed community-service sentences for minor offences such as shoplifting and prostitution. BIDs can be equated with an attempt to reclaim public space from the sense of menace that drives shoppers and others to the suburbs, but it is also the case that BIDs reconstruct public space in their own image and nurture visible social stratification by catering for a particular lifestyle. FURTHER READING BOOKS E.Burgess (1925) The growth of the city, in R.Park and E. Burgess (eds) The City Chicago, IL: University of Chicago Press, 47–62 D.Harvey (1981) The urban process under capitalism, in M. Dear and A.Scott (eds) Urbanisation and Urban Planning in Capitalist Society London: Methuen, 91–121 H.Hoyt (1939) The Structure and Growth of Residential Neighborhoods in American Cities Washington, DC: Federal Housing Administration A.Jonas and D.Wilson (1999) The Urban Growth Machine: Critical Perspectives, Two Decades Later Albany, NY: State University of New York Press M.White (1987) American Neighborhoods and Residential Differentiation New York: Russell Sage Foundation Land use in the city 221 J.Whitehand (1991) The Making of the Urban Landscape Oxford: Blackwell J.Whitehand and P.Larkham (1992) Urban Landscapes: International Perspectives London: Routledge, 170–206 S.Zukin (1991) Landscapes of Power: From Detroit to Disneyworld Berkeley, CA: University of California Press JOURNAL ARTICLES T.Banerjee (2001) The future of public space Journal of the American Planning Association 67(1), 9–24 J.Byers (1998) The privatization of public space: the emerging grade-separated city in North America Journal of Planning Education and Research 17, 189–205 R.Checkoway (1980) Large builders, federal housing programmes and post-war suburbanisation International Journal of Urban and Regional Research 4(1), 21–45 L.Feagin (1987) The secondary circuit of capital: office construction in Houston, Texas International Journal of Urban and Regional Research 11(2), 172–91 A.Harding (1991) The rise of urban growth coalitions UK-style? Environment and Planning C 9, 295–317 C.Harris and E.Ullman (1945) The nature of cities Annals of the American Academy of Political Science 242, 7–17 T.Hartshorn and P.Muller (1989) Suburban downtowns and the transformation of Atlanta’s business landscapes Urban Geography 10, 375–95 J.Rust (2001) Manufacturing industrial decline: the politics of economic change in Chicago 1955 1998 Journal of Urban Affairs 23(2), 175–90 E.Soja (1980) The socio-spatial dialectic Annals of the Association of American Geographers 70, 207–25 G.Squires and W.Velez (1987) Insurance red-lining and the transformation of an urban metropolis Urban Affairs Quarterly 23(1), 63–83 C.Teixiera (1995) Ethnicity, housing search and the role of the real estate agent: a study of Portuguese and non-Portuguese real estate agents in Toronto Professional Geographer 47(2), 176–83 KEY CONCEPTS ■ urban morphogenesis ■ town-plan analysis ■ fringe belt ■ burgage cycle ■ natural areas ■ concentric-zone model ■ sector model ■ multiple-nuclei model ■ urban ecology ■ urban-realms model ■ political economy ■ circulation of capital ■ use value ■ exchange value ■ human agency Urban geography 222 ■ blockbusting ■ red-lining ■ growth machines ■ manipulated city ■ postmodernism ■ grade-separated city ■ overaccumulation ■ socio-spatial dialectic STUDY QUESTIONS 1. How can the study of town plans help explain the changing pattern of urban land use? 2. Critically evaluate the contribution of Burgess’s concentric-zone model to our understanding of urban structure. 3. Explain the political-economy interpretation of urban land-use change. Can you identify any real-world examples of these processes at work? 4. Select one of the major actors involved in the production of the built environment and, with reference to specific examples, write an essay to illustrate their role in urban change. 5. With reference to the concept of externalities, prepare a list of positive and negative externality effects of urban development and relate these to events in a city of your choice. 6. Explain what is meant by the privatisation of public space and think of some examples of where it has occurred. PROJECT On a map of your local city draw a transect line (e.g. along a major routeway) from the city centre to the metropolitan edge. In the field, follow this line, mapping the major land uses on either side of the route. Back in the laboratory produce a general land-use map of your urban transect. Explain the variations in land use with distance from the city centre. Compare your empirical findings with the theoretical patterns described by the major models of urban land use discussed in the chapter. 8 Urban Planning and Policy Preview: the roots of urban planning; post-war urban planning in the UK; urban policy in the UK; the redevelopment phase; the social-welfare phase; the entrepreneurial phase; the competitive phase; the ‘Third Way’; urban planning in the USA; land-use zoning; growth management; smart growth; planning the socialist city; socialist urban form; planning for sustainable urban development INTRODUCTION The development of most urban areas is influenced, to some degree, by the processes of urban policy and urban planning. In this chapter we examine the nature and operation of urban policy and planning in the UK, Europe and the USA from its origins in the nineteenth century until the present day. Urban planning and urban policy are concerned with the management of urban change. They are state activities that seek to influence the distribution and operation of investment and consumption processes in cities for the ‘common good’. However, it is important to recognise that urban policy is not confined to activity at the urban scale. National and international economic and social policies are as much urban policy, if defined by their urban impacts, as is land-use planning or urban redevelopment. In effect, urban policy is often made under another name. Urban policy and planning are dynamic activities whose formulation and interpretation are a continuing process. Measures introduced cause changes that may resolve some problems but create others, for which further policy and planning are required. Furthermore, only rarely is there a simple, optimum solution to an urban problem. More usually a range of policy and planning options exists from which an informed choice must be made1 (Box 8.1). Planning is carried out within the broad framework of government policy-making and has its general objectives set out in legislation. A primary objective of the UK planning system is ‘to regulate the development and use of land in the public interest’ (Department of the Environment 1999 p. 2).2 Planning can be undertaken for a variety of reasons (Table 8.1). The aims of urban planning may be contradictory and reveal differing attitudes to the roles of the market mechanism and the state. Central to this debate is the question ‘planning for whom?’ While most of the purposes of planning listed in Table 8.1 assume that benefits should accrue to the ‘public as a whole’ or, in relation to redistribution aims, to the poorer and less vocal sections of society, the question of the validity of these goals and the extent to which they are met has produced polarised views on the value of planning (Box 8.2). Urban geography 224 THE ROOTS OF URBAN PLANNING THE UK With the exception of various forms of ‘planted settlements’ (e.g. Greek colonial towns or medieval planned new towns, bas tides), the forces underlying urban growth have operated largely free of any form of public accountability, so that for much of the historical period urban development proceeded in an unregulated organic fashion. Today a powerful system of planning exists in the UK and Europe, and to a lesser extent in North America, which aims BOX 8.1 The nature of urban policy Urban policy is the product of the power relation between the different interest groups that constitute a particular society. Foremost among these agents are government (both local and national) and capital in its various fractions. Capital and government pursue specific goals that may be either complementary or contradictory. For capital the prime directive is profit maximisation. Government, on the other hand, in addition to facilitating the process of accumulation, must also satisfy the goal of legitimation. These political and economic imperatives have a direct influence on the nature of urban policy. Urban policy is also conditioned by external forces operating within the global system, as well as by locally specific factors and agents. The form of urban policy employed depends on the problem to be addressed and, most fundamentally, on the ideological position of the state. Adherents of market capitalism view the production of unevenly developed cities as the inevitable outcome of technological change in an economic system that readily adapts to innovation. The negative socio-spatial effects of this restructuring that impinge on disadvantaged people and places are regarded as unavoidable consequences of a process that is of benefit to society as a whole. For those on the political right, market forces are the most efficient allocators of capital and labour, and state intervention is considered unnecessary. Policies involving social-welfare expenditure and government financial aid to declining cities are regarded as harmful because they anchor low-wage workers to sites of low employment opportunity, discourage labour-force participation and inhibit labour mobility. Welfare state liberals, on the other hand, while accepting the central role of the market, acknowledge that the institutional and cyclical ‘market imperfections’ that have left certain people and places in prolonged economic distress must be rectified by compensatory government policies. to circumscribe urban development and direct it towards socially beneficial goals. For urban planning to exist, there must be a political consensus that the problems affecting cities can best be tackled through government intervention. Urban planning and policy 225 TABLE 8.1 THE MAIN GOALS OF PLANNING 1. To improve the information available to the market for making its locational choices 2. To minimise the adverse ‘neighbourhood effects’ created by a market in land and development 3. To ensure the provision of any ‘public goods’, including infrastructure or actions that create a positive ‘neighbourhood effect’, which the market will not generate because such activity cannot be rewarded through the market 4. To ensure that short-term advantage does not jeopardise long-term community interest 5. To contribute to the co-ordination of resources and development in the interest of overall efficiency of land use 6. To balance competing interests in the use of land to ensure an overall outcome that is in the public interest 7. To create a good environment, for example in terms of landscape, layout or aesthetics of buildings, that would not result from market processes 8. To foster the creation of ‘good’ communities in terms of social composition, scale or mix of development, and a range of services and facilities available 9. To ensure that the views of all groups are included in the decision-making processes regarding land and development 10. To ensure that development and land use are determined by people’s needs, not means 11. To influence locational decisions regarding land use and development in order to contribute to the redistribution of wealth in society Source: A.Thornley (1991) Urban Planning under Thatcherism London: Routledge This, in turn, requires willingness by individuals to relinquish some of the rights to property which they enjoy in a free market and to accept the principle that land use should be centrally controlled for the public good.6 In general, while electorates in the UK and Europe have accepted the implications of a comprehensive system of urban planning, in the USA government intervention in urban development is more restricted, with zoning being the major mechanism for land-use control. Urban planning emerged as a response to the manifest problems of the nineteenthcentury industrial metropolis (see Chapter 3). Two types of reaction were evident. The first, represented in the work of Marx and Engels, was revolutionary and BOX 8.2 The value of planning Urban planning is a ubiquitous activity in the modern world. Nevertheless, the principles and practice of planning have come under attack from both the left and the right of the political spectrum. In contrast to the positive goals identified in Table 8.1, the value of planning has been dismissed by the far left which regards it as a state apparatus Urban geography 226 attuned to the needs of capital and designed to maintain the unequal distribution of power in society. Reade (1987), for example, concludes that the legitimacy of the propertydevelopment industry and its associated financial institutions is maintained by having a ‘planning system’ that provides a pretence of government intervention in the ‘public interest’.3 For some critics on the right, planning is seen as a major cause of inner-city decline and social unrest through its policies of clearance and decentralisation, rigid land-use zoning and imposition of standards. According to Steen (1981), because of planning, resources are fruitlessly channelled into deprived areas and wasted rather than encouraging wealth creation.4 For others the chief problems of planning lie in its interference with the market and in the fact that, contrary to the goals of planning, a dynamic and prosperous urban economy requires inefficiency in its structure and land use in order to permit innovation and experimentation.5 The New Right attack on the redistributive goals of urban planning in the UK was articulated most forcibly by the Conservatives under Margaret Thatcher during the 1980s. advocated the overthrow of the social and political system responsible for creating the polarised social conditions that characterised urban Britain. The conservative alternative involved basic acceptance of the urban industrial system but the use of state intervention to ameliorate its worst excesses. It was the latter argument, articulated in the UK by the factory and sanitary reformers, and reinforced by the success of a number of early housing and new town schemes, that paved the way for the emergence of modern urban planning.7 The belief that designing new communities offered a means of escape from the problems of the nineteenth-century industrial city was central to the ideas of the Utopian socialists. Robert Owen (1813) proposed the creation of ‘agricultural and manufacturing villages of unity and mutual co-operation’ to house between 1,000 and 1,500 persons and cater for all the social, educational and employment needs of the community. Owen’s model industrial complex at New Lanark in Scotland promised superior working and living conditions, cheap subsidised shops based on bulk-buying of all household necessities, and a community school or ‘Institution for the Formation of Character’, which operated as a day school for children and a night school for adults. Owen’s example led to the development of a number of other model towns.8 The social reformer and architect James Silk Buckingham (1849) proposed a Utopian temperance community of 10,000 to be named Victoria.9 The plan envisaged segregation of land uses, with manufacturing trades and noxious land uses near the periphery, and housing and offices in the inner areas. All dwellings were to have flushing toilets and there would be a variety of house sizes to accommodate different households. Public baths were to be provided in each quarter of the town. A green belt of 4,000 ha (10,000 acres) of agricultural land would surround the settlement. All land was to be owned by the development company and buildings occupied for rent. Although Victoria was never built, many of the ideas were taken up by later urban reformers, including Ebenezer Howard in his designs for garden cities (see below). Several smaller Utopian communities were started (Table 8.2). Among the most significant were Saltaire and Bournville. Saltaire was built by Titus Salt to replace his woollen mills in Bradford by a single large factory and a new town for the workforce. He Urban planning and policy 227 selected a greenfield site crossed by a canal and railway, and between 1851 and 1871 completed a model industrial town of 820 dwellings and a population of 4,389. The town was endowed with a variety of community buildings and parks, and although the residential density of thirty-two houses (170 persons) per acre (eighty houses or 420 persons per hectare) appears crowded by today’s standards, it represented a marked improvement on working-class living conditions of the time. The model town of Bournville built by George Cadbury also involved the transfer of factory production from an inner-city to a greenfield site but was the result of a wider concept than Saltaire. From the outset Bournville was intended to be not only a company town but also a general example to society of how it was possible to provide decent living con- TABLE 8.2 PLANNED SETTLEMENTS IN EIGHTEENTH—AND NINETEENTHCENTURY BRITAIN Settlement name Company/family Location Industry Date started Cromford Arkwright Derbyshire Textiles 1772 Styal Greg Cheshire Cotton textiles 1784 New Lanark Arkwright/Dale/Owen Lanarkshire Cotton textiles 1785 Nent Head London Lead Co. N. Yorkshire Lead 1825 Turton Ashworth Bolton Textiles 1825 Barrow Bridge Gardener/Bazley Lancashire Textiles 1830 Street Clark Somerset Footwear 1833 Meltham Mills Brooks Huddersfield Cotton thread 1836 Copley Akroyd Halifax Textiles 1844 Wilshaw Hirst Huddersfield Textiles 1849 Saltaire Salt Bradford Textiles 1851 Bromborough Pool Price Wirral Candles 1853 Akroydon Akroyd Halifax Textiles 1861 West Hill Park Crossley Halifax Carpets 1864 Bournville Cadbury Birmingham Cocoa 1878 Port Sunlight Lever Wirral Soap 1888 Foyers British Aluminium Loch Ness Aluminium 1895 ditions without endangering company profits. With its cottage houses, large gardens, open space and community facilities built around the chocolate factory, Bournville set new standards of working-class living (Figure 8.1). In quantitative terms the nineteenth-century model towns contributed little to the problems of the factory slums in the large cities, but they did stimulate reformers to Urban geography 228 question the morality and economic necessity of such living conditions. The idealism of the Utopian socialists was mirrored at the turn of the century in the ideas of Ebenezer Howard (1898).10 Howard, with London in mind, was strongly critical of living conditions in the large towns. His alternative was to design a garden city based on the following principles: 1. Each garden city would be limited in size to 32,000 population. 2. It would have sufficient jobs to make it self-supporting. 3. It would have a diversity of activities, including social institutions. 4. Its layout would be spacious. 5. It would have a green belt to provide agricultural produce, recreation space and a limit to physical growth. 6. The land would be owned by the municipality and leased to private concerns, thereby reserving any increases in land value to the community as a whole. 7. Growth would occur by colonisation. Howard did not envisage building isolated towns but advocated a cluster arrangement of six interdependent garden cities, connected by a rapid transport route around a central city of 58,000 population, the whole comprising a ‘social city’ of 250,000 people (Figure 8.2). Howard’s ideal of a ‘town-country’ lifestyle led to the founding, in 1899, of the Garden City Association, which built two garden cities, at Letchworth (1901) and Welwyn (1920), and was a major stimulus to the formation in 1914 of the Town Planning Institute. A second, parallel stimulus to urban planning in the UK was the sanitary-reform movement. This was stimulated by concern over the health of the urban population as a succession of epidemics ravaged the densely packed inner areas of the major British cities (see Chapter 3). The deficiency of public facilities, such as a clean water supply and adequate sewerage system, reinforced arguments for government intervention, and under pressure from enlightened politicians such as Shaftesbury, Torrens, Cross and Chadwick legislation was passed to establish basic levels of sanitary provision and building standards. The 1875 Public Health Act consolidated previous measures and introduced a set of codes in respect of the construction of new streets, and the structure of houses and their sanitary facilities, and also gave local authorities the power to close down dwellings that were unfit for human habitation. These measures did much to reduce levels of morbidity and early mortality in the nineteenth-century city. THE USA The roots of urban planning in the USA can be traced to the ideas of the Progressive Intellectuals of the late nineteenth century. In examining the shortcomings of American capitalism this loose-knit group of sociologists, economists and political scientists, including White, Dewey, Cooley and Park, recognised a need for public intervention in, through not control of, the economy.11 Among suggestions for government regulation of business, employment and politics in cities they advocated the appointment of specially trained experts to manage cities. Designers sought to create more humane living environments. Landscape architects such as Urban planning and policy 229 Figure 8.1 Plan of the model settlement of Bournville Source: S.Ward (1994) Planning and Urban Change London: Chapman Urban geography 230 Figure 8.2 Structure of Howard’s garden city Olmstead, Davis and Vaux designed residential areas as ‘cities in a garden’, which led to the building of Riverside IL, Llewellyn Park NJ and Brookline MA as America’s first planned ‘romantic suburbs’. More generally, the City Beautiful movement, which emerged following the Chicago World Columbia Exposition of 1893, argued for the planned unity of the city as a work of art supported by a master plan for land use and by comprehensive zoning ordinances to maintain that plan. By 1913 over forty cities had prepared master plans, and more than 200 were engaged in some form of major civic improvement.12 In 1917 a new professional organisation, the American City Planning Urban planning and policy 231 Institute, was established.13 In practice, however, the forces of privatisation were too strong to be contained by public officials, and ‘urban planners seldom did more than follow residential and commercial developers with transportation and sewerage systems’.14 The strength of private enterprise within the US political economy has been a major force in shaping the form of US planning and the structure of American urban areas. WESTERN EUROPE In parallel with the development of Utopian idealism in Britain, equally ambitious alternative urban forms were being advanced in Europe in the ideas of the Italian Futurist movement, launched in a manifesto by Marinetti in 1909. The concept of a new, comprehensively planned city was a key idea in futuristic urban designs, which included high-rise building, elevated roadways, land-use segregation and the use of massproduction techniques and new materials such as glass and concrete. The Swiss architect Charles Jeanneret (Le Corbusier)15 proposed a city for 3 million people based on four main principles: 1. As a result of increasing size and central area congestion the traditional form of city had become obsolete. 2. Pressure on the central business district could be reduced by spreading the density of development more evenly. 3. Congestion could be alleviated by building at higher density (1,000 persons to the net residential acre, or about 2,500 to the hectare) at local points, with a high proportion (95 per cent) of intervening open space. 4. An efficient urban transportation system incorporating railway lines and segregated elevated motorways would link all parts of the city. Although his plans were not implemented in their entirety, Le Corbusier’s ideas exerted a profound effect on urban planning and the form of cities. The concept of high-rise, highdensity building was translated into practice in most large cities during the 1950s and 1960s, although in many instances less attention was given to the quality of space surrounding the tower blocks. Notwithstanding differing national circumstances, we can identify three broad phases of urban planning in Western Europe: 1. In the immediate aftermath of the Second World War the focus of attention was on reconstruction and satisfying the backlog of housing and basic infrastructure. 2. By the late 1950s, increasing affluence and the growth of centralised planning systems led to comprehensive slum clearance, city-centre redevelopment schemes, and the construction of urban motorways and large-scale public housing projects. 3. From the late 1970s, growing awareness of the social disruption caused by the largescale remodelling of cities led to greater attention to public participation in planning, and the replacement of redevelopment by rehabilitation. A significant ongoing problem for all large European cities is that of growth management. Here the experience of the multicentred metropolitan region of Randstad may offer lessons for polycentric urban regions elsewhere (Box 8.3). Urban geography 232 POST-WAR URBAN PLANNING IN THE UK Although urban planning in the UK was inspired by reformist reactions to the nineteenthcentury city, its cause was advanced in the 1930s by the growing focus on the distribution of population, and associated land-use issues. The increasing concentration of population in an axial belt stretching from London to Merseyside, high unemployment levels outside this zone, and the threat of urban BOX 8.3 Urban growth management: the Dutch approach The Netherlands is a small country with a population density of over 1,030 persons per square mile (400 per square kilometre). The Randstad city-region contains 36 per cent of the national population on 5 per cent of the land area and incorporates the three large cities of Amsterdam, Rotterdam and The Hague as well as a number of smaller urban centres. Pressure on open space is considerable. Dutch planners seek to maintain the green heart of Randstad by directing growth outwards towards the peripheral regions and, within the urban region, by encouraging higher-density development in rehabilitated inner-city areas. Strong planning controls to protect the centre of Randstad are supported by large-scale public ownership of urban land, with 70 per cent of Amsterdam owned by the city authority. The fact that the polycentric structure of Randstad is functional as well as physical, with different cities specialising in broadly different functions (e.g. heavy industry in Rotterdam, finance in Amsterdam and government in The Hague), also reduces journey times and traffic congestion. Continued planned regulation of urban growth will be needed to prevent the coalescence of individual urban areas in Randstad and enable the ‘green-heart metropolis’ to maintain its open structure. sprawl induced governmental concern. A number of reports on these and other issues (Table 8.3) contributed to the passage of the 1947 Town and Country Planning Act, which established the structure of modern urban planning in the UK.16 The basic principle enshrined in the 1947 Actis that of private ownership of land but public accountability in its use so that landowners seeking to undertake development first had to obtain permission from the local planning authority. The local authority was also given the power to acquire land for public works by compulsory purchase on payment of compensation to the landowner. Local planning authorities were required to prepare and submit quinquennial development plans to the Ministry of Town and Country Planning indicating how land in their area was to be used. A second principle embodied in the 1947 Act was that of community gain, rather than individual gain, from land betterment. This meant that when land was developed, the increase in its value that resulted from the granting of planning permission was reserved for the community by the imposition of a 100 per cent land-development tax. Political opposition ensured that this provision was removed in 1952. Today, compulsory exaction of betterment values from land developers has been replaced by a system of negotiated agreements over planning gain, which represents the benefits that a local authority may require of a developer as a condition Urban planning and policy 233 TABLE 8.3 MAJOR REPORTS AND LEGISLATION LEADING TO THE 1947 PLANNING SYSTEM IN THE UK Date Report or new legislation 1940 Report of the Royal Commission on the Distribution of the Industrial Population (Barlow Report, Cmd 6153) 1942 Report of the Expert Committee on Compensation and Betterment (Uthwatt Report, Cmd 6386) 1942 Report of the Committee on Land Utilisation in Rural Areas (Scott Report, Cmd 6378) 1943 Ministry of Town and Country Planning Act 1943 Town and Country Planning (Interim Development) Act 1944 Town and Country Planning Act 1945 Distribution of Industry Act: 1945 Abercrombie’s Greater London Development Plan 1946 New Towns Act 1947 Town and Country Planning Act for planning permission.17 We can note the similarity, in principle, with the mechanism of incentive zoning in the USA (Box 8.4). The primary objectives of the 1947 planning system at the city-region scale were urban containment, protection of the countryside and the creation of self-contained balanced communities (e.g. New Towns). These goals were advanced by local authorities using the development plan and development control process, and by central government through the New Towns programme, supplemented by the Expanding Towns scheme (which enabled cities with problems of overcrowding to arrange overspill schemes with other local authorities (see Chapter 9)). The first major changes to the 1947 planning system were contained in the 1968 Town and Country Planning Act. This sought to introduce greater responsiveness and flexibility to the plan-making process. The single development plan with its specific land-use focus and five-year life expectancy was replaced by a two-tier system of structure plans and local plans. Structure plans are comprehensive strategic statements designed to translate national and regional economic and social policies into the specific areal context of the local authority. Local plans apply the structure-plan strategy to particular areas and issues, and make detailed provision for development control.18 Most commentators would agree that the 1947 planning system has achieved its objectives: post-war suburban growth has been contained to the extent that coalescence of adjacent cities has been prevented and good-quality agricultural land protected. The Urban geography 234 major mechanism of green belts around the large cities has prevented peripheral sprawl, although they have contributed to inflation in land and housing prices in existing BOX 8.4 Incentive zoning in Seattle Incentive zoning offers bonuses to developers in return for the provision of public benefits. The Washington Mutual Tower gained twenty-eight of its fifty-five storeys as a result of amenities offered by the developer. In addition to the twenty-seven storeys allowed as of right, the developer gained thirteen storeys for a $2.5 million housing donation and between one and two and a half storeys for various other public benefits. settlements within the green belt and provoked an ongoing conflict between developers and planners over the availability of land for housing19 (Box 8.5). Also, the green belt has been powerless in the face of increases in transport technology and in the length of acceptable commuting journey, which have seen some residential development leapfrog into settlements beyond the green belt. But, in general, the UK planning system has prevented the kind of scattered urban development in evidence around cities in North America. Urban planning and policy 235 URBAN POLICY IN THE UK Urban planning takes place within the framework of national urban policy, the priorities of which reflect the ideology of the state. We can identify four major phases in British post-war urban policy (Table 8.4): 1. physical redevelopment; 2. social welfare; 3. entrepreneurial; 4. competitive. THE PHYSICAL REDEVELOPMENT PHASE From the end of the Second World War until the late 1960s urban problems were seen largely in physical terms. The policy response to issues of housing quality and supply, transport, and industrial restructuring focused on slum clearance and comprehensive redevelopment strategies, and the planned decentralisation of urban population via regional policy and New Town development. BOX 8.5 Land-use conflict in the green belt Various agents and interests are involved in determining whether a parcel of land is transferred from rural to urban use (see diagram). In the UK, where a nationally co-ordinated planning system attempts to effect an equitable balance between private profit and public interest, conflict between developers and planners over land for new housing is a key factor determining the character of the green belt. The debate over land availability has intensified since the early 1970s, with, in general, builder-developers arguing that the planning system restricts their ability to obtain a basic factor of production and that development controls inflate the price of land and therefore of houses. The purpose of the planning system, on the other hand, is to ensure the orderly release of building sites within an approved policy framework. The commuter village of Torrance in the green belt to the north of Glasgow is a typical location for the ongoing battle between house-builders and planners around Britain’s cities.20 The issues are complex: in seeking permission to build houses on Urban geography 236 the disputed (Tower Farm) site the developers cited a number of arguments, including the need to satisfy demand for housing in attractive locations, and their willingness to provide, via ‘planning gain’, additional infrastructure and community facilities. The planners argued that demand for new housing could be accommodated on brownfield sites outwith the green belt, and that the local authority would have to meet the ongoing running costs of any facilities. Failure to resolve the dispute at a local level resulted in a public inquiry that concluded in favour of the planning authority. The developers exercised their right of appeal to the Secretary of State. Their appeal was dismissed. Thus, in the case of the Tower Farm site, the land remained part of the green belt. However, while the battle of Tower Farm may be over, both developers and planners live to fight another day in a continuing struggle between private profit and public interest in the production of the built environment. Source: M.Pacione (1990) Private profit and public interest in the residential development process Journal of Rural Studies 6(2), 103–16 THE SOCIAL-WELFARE PHASE In the early 1970s empirical research highlighted the incidence of poverty within Britain’s cities as the long post-war boom faltered (see Chapter 15). Emphasis was placed on supplementing existing social programmes to improve the welfare of disadvantaged individuals and communities. Influenced by US initiatives, such as the Head Start project developed as part of the Model Cities programme, a range of area-based experimental schemes were introduced during the 1970s. These ‘Urban Programme’ initiatives, which included educational priority areas and community development projects,21 operated from a ‘culture of poverty’ perspective (under which poverty is held to be self-reproducing) and aimed to give deprived communities the capacity to solve their own problems. The widespread increase in unemployment in the mid-1970s, in the wake of the Arab oil embargo and world recession, made it clear, however, that the scale of urban deprivation could not simply be the result of the inadequacies of the poor. An alternative ‘structural’ explanation of poverty was enshrined in the final report of the Community Development Project. This rejected a social pathological view of deprivation in the conclusion that ‘there might certainly be in those areas a higher proportion of the sick and the elderly for whom a better co- ordination of services would undoubtedly be helpful, but the vast Urban planning and policy 237 TABLE 8.4 MAIN PHASES IN BRITISH POST-WAR URBAN POLICY Physical redevelopment, 1945–69 Area-based social Entrepreneuralism, welfare projects 1979–91 (the inner-city problem), 1969–79 Competitive policy, 1991–7 Area-based projects run by local government gave way in 1978 to ‘partnerships’ involving central government and designated local authorities, other statutory bodies (such as area health authorities), and local voluntary organisations and local industry (but not labour organisations). ‘Programme areas’ also designated. Local government seen as the natural agent of regeneration. Local government dominated by urban managerialism Greater emphasis placed on the role of the private sector in urban policy (privatism). Creation of business elites and growth coalitions. ‘Privatised’ partnerships and business representation on a range of national and local institutions Patterns of interest representation shifted, requiring new patterns of local leadership with private sector, community representatives and voluntary sector organisations, alongside city councillors in boards and companies at arm’s length from the local authority, Consolidation of trends towards urban governance to prevent municipalisation of policy Home Office and the newly created Department of the Environment were key departments during this period. Urban Deprivation Unit created in the Home Office. Some attempt to coordinate policy Centralisation of power. Shift in local government towards urban governance with private sector having greater role. Confrontation approach to local government (rate capping, cutbacks, abolition, and quangos bypassing local authorities). Urban entrepreneuralism Purported ‘new localism’ developed through the establishment of integrated government offices for the regions and the Challenge Fund. Cabinet committee established to oversee Challenge Fund. In reality, Representational regime Construction of corporatist consensus around reconstruction programme. Centrallocal government partnership in councilhousing redevelopment programme. Close links with construction industry (system-build, high-density, highrise) Internal structures of the state Monolithic state characterised by corporate planning, bureaucratic paternalism, functionalism, uniformity and inflexibility. Constructive partnership between central and local Urban geography government. Department of Economic Affairs created (1964) together with a set of official regions and regional planning councils and regional planning boards across a number of areas but thwarted by inter-department rivalry. Attempts to adopt corporate working and the bending of main programmes to address the urban problem. Cooperation between central and local government. 238 promoted involving leaner and flatter managerial structures, generic roles, team working and flexibility ‘remote control’ exercised via the contract culture, Central-local government relations characterised by an ‘authoritarian decentralism’ or ‘centralist localism’. Local partnerships involved in a process of centrally controlled local regulatory undercutting. New public management (to promote, privatisation and; competition) creating the; enabling authority; Neo-liberal philosophy pursued involving deregulation, liberalisation and privatisation. Social needs subordinated to the needs of business. Emphasis given to property-led initiatives and the creation of an entrepreneurial culture, e.g.: ■ Enterprise Zones, 1979 ■ Urban Development Corporations, 1979 ■ Urban Development Grants, 1982 ■ Derelict Land Grant, 1983 ■ City Action Teams, 1985 ■ Estate Action, 1985 ■ Urban Regeneration Grants, 1987 ■ City Grant, 1988 Competition for funds and competitiveness of business and localities the leading priorities for regeneration policy. Initiatives to improve the competitive advantage of localities, e.g.: ■ City Challenge, 1991 ■ Urban Partnership, 1993 ■ City Pride, 1993 ■ Single Regeneration Budget, 1994 ■ Rural Challenge, 1994 ■ Regional Challenge, 1994 ■ Capital Challenge, 1996 ■ Local Challenge, 1996 ■ SectorChallenge, 1996 Patterns of state intervention The long post-war boom period was dominated by policies to achieve national unity through regional balance, containment of urban growth and the reconstruction of urban areas, involving slum clearance and comprehensive redevelopment. Instruments included establishment of the development-plan system, New Towns, industrial development certificates, office development permits and Plan published in 1965 advocating ‘growth through planning’ Area-based experimental socialwelfare projects attempting to respond to economic, social and environmental problems resulting from structural decline of the economy. Innercities policy, e.g.: ■ Urban Programme 1969 ■ General Improvement Areas, 1969 ■ Educational Priority Areas ■ Community Development Projects, 1969 ■ Inner Area Studies ■ Housing Action Areas, 1974 ■ Comprehensive Community Programmes, 1974 ■ Enhanced Urban Programme, 1978 Inner Urban Areas Act Urban planning and policy 239 (1978) acknowledged the economic nature of the urban crisis. Keynesian demandmanagement techniques used to counter the emerging crisis (‘stagflation’) of the 1970s Source: N.Oatley (1998) Cities, Economic Competition and Urban Policy London: Chapman majority were ordinary, working-class men and women who, through forces outside their control, happened to be living in areas where bad housing conditions, redundancies, layoffs and low wages were commonplace’22 (Community Development Project 1977 p. 4). It followed that to tackle the root causes of urban decline would require more than marginal adjustments of existing social policies. This was acknowledged by the 1977 White Paper on policy for the inner cities, which signalled a more broadly based approach to urban problems, combining economic, social and environmental programmes and involving new partnership arrangements between central and local government to provide a more coordinated response. The emphasis on improving the economic environment of cities was promoted in several ways, including a shift of attention from New Towns to urban regeneration and increased powers to enable local authorities to aid and attract industrial developments. The major vehicle for these measures was the expanded Urban Programme.23 In practice, however, a concerted attack on the urban problem was undermined by the prerogatives of national economic policy, which led to reductions in local-authority finance. The Conservative government elected in 1979 continued the concept of partnership but stressed the involvement of the private sector. THE ENTREPRENEURIAL PHASE The advent of the Thatcher government in 1979 ‘witnessed the fracture of three main pillars upon which post-Second World War social democratic politics was constructed— Fordism, welfarism and Keynesianism’24 (Gafficken and Warf 1993 p. 71). The reorientation of urban policy by the New Right Conservative government was part of a wider agenda to restructure Britain economically, socially, spatially and ideologically around a new consensus of free-market individualism and unequivocal rejection of the social-democratic consensus of the post-war Keynesian welfare state. As Martin (1988 p. 221) observed, the thrust of state policy shifted from welfare to enterprise: the aim has been to reverse the post-war drift towards collectivism and creeping corporatism, to redefine the role and extent of state intervention in the economy, to curb the power of organised labour, and to release the natural, self-generative power of competitive market forces in order to revive private capitalism, economic growth and accumulation.25 Urban geography 240 The Keynesian commitment to the macro-economic goal of full employment was replaced by the objective of controlling inflation by means of restrictive monetary measures, and supply-side flexibilisation. From its inception Thatcherism’ was a doctrine for modernising Britain’s economy by exposing its industries, its cities and its people to the rigours of international competition in the belief that this would promote the shift of resources out of inefficient ‘lame duck’ traditional industries and processes into new, more flexible and competitive high-technology sectors, production methods and work practices.26 The principal mechanisms for achieving this transformation centred on tax cuts and deficit spending, deregulation and privatisation, all of which had geographically uneven impacts. At the urban level these three macro-economic strategies were combined most strikingly in the concepts of the enterprise zone (EZ) and the urban development corporation (UDC) (see Chapter 16). As part of the broad political and economic agenda, urban policy was also used to restructure central government-local government relations. Five processes characterised the changes: 1. displacement involving the transfer of powers to non-elected agencies (e.g. UDCs), thereby bypassing the perceived bureaucracy and obstructiveness of local authorities; 2. deregulation involving a reduction in local authorities’ planning controls to encourage property-led regeneration (e.g. in EZs); 3. the encouragement of bilateral partnerships between central government and the private sector; 4. privatisation, incorporating the contracting out of selected local government services, housing tenure diversification and provision for schools to ‘opt out’ of local education authority control; 5. centralisation of powers through a range of quangos27 (quasi non-governmental organisations). THE COMPETITIVE PHASE By the early 1990s it was evident that the approach to urban policy pursued since 1979 had failed to reverse urban decline. The limitations of a property-based approach to regeneration had been exposed by a slump in the demand for property in the recession of 1989–91. As Turok (1992 p. 376) observed, although property development has potentially important economic consequences, it is ‘no panacea for economic regeneration and is deficient as the main focus of urban policy’.28 Property development lacks the scope, powers and resources to provide the holistic approach required to tackle urban decline. Its limited perspective and goals are unable to guarantee a rise in overall economic activity in a locality and ignore important ‘human issues’ such as affordable housing, education and training, social exclusion and investment in basic infrastructure. Urban planning and policy 241 Plate 8.1 The Canary Wharf tower in London symbolises the changing economic structure of the former docklands The government’s response was to reconstruct urban policy around the ‘initial catalyst’ of competition. This era of competitive urban policy was heralded by the City Challenge initiative, which introduced competitive bidding among local authorities for urban regeneration funds.29 Successful schemes are managed by a multi-agency partnership involving the local authority along with the private, voluntary and community sectors. While local authorities appeared to have a greater role under this arrangement, in practice local autonomy is constrained by the underlying entrepreneurial ethos of the Urban geography 242 partnership organisations and by the need for successful bids to conform to central government guidelines. THE ‘THIRD WAY’ We can add a fifth phase to this chronology of British urban policy commencing in 1997 with the election of New Labour. This marked a move away from the neo-liberal era of policy under the Conservatives towards a situation in which greater attention is paid to the social consequences of economic policy. In policy terms the key priorities include strengthening local and regional economies, increasing economic opportunities for deprived areas, rebuilding neighbourhoods and promoting sustainable development. Pursuit of these goals is based on three key principles: 1. A strategic approach is to be taken which involves the integration of national policies and programmes with EU, regional and local initiatives to ensure a comprehensive attack on the multifaceted problems of social disadvantage. 2. Local authorities are to be given a stronger role in urban regeneration. 3. The ‘sweat equity’ of local people will be encouraged to promote community economic development. Specifically, New Labour’s approach to urban policy is based on CORA: 1. Community involvement, with greater public participation; 2. Opportunity to work or to obtain training and education; 3. Responsibility in the obligation of citizens who can work to do so; 4. Accountability of governments to their publics. We discuss the particular policies and outcomes of this Third Way in relation to urban regeneration in Chapter 16. URBAN PLANNING IN THE USA Although urban planning in the USA shares the same reformist roots as in the UK, its evolution and contemporary structure are very different. By contrast with the situation in Britain, there is no national system of planning in the sense of a common framework with a clearly defined set of physical, social and economic objectives. Planning is not obligatory, and together with the fragmented structure of local government in addition to the federal government and fifty states, there are about 8,000 counties, 18,000 municipalities and 17,000 townships each with the power to plan or regulate land use, i.e. an average of 760 per state30 this means that the content of planning is both local and variable from place to place. In principle a range of techniques for controlling urban growth and land use are available, but in practice the major tool employed is land-use zoning. Specific urban problems, such as the provision of low-income housing, are addressed through federal policy. The first comprehensive zoning ordinance was passed in New York in 1916. The judgement of the US Supreme Court in 1926 that zoning did not infringe the Fourteenth Amendment to the Constitution (which protects against property being taken without due Urban planning and policy 243 process of law) led to widespread adoption of the technique. Under this procedure, the effective control of land use was transferred from the state to the municipalities and townships, which were thereafter permitted to limit the types of development on land within their boundaries,31 including control over the height, bulk and area of buildings constructed after the enactment of zoning regulations. The purposes of such controls were to minimise problems of congestion, fire hazard, shading by high buildings; to control population density; to ensure the provision of urban services; and to promote the general welfare of the public. In practice there are many forms of zoning32 (Box 8.6). This variety, and the underlying presumption in favour of development, means that controls over market-induced physical growth and change are much weaker in the USA than in the UK. Critics of zoning maintain that: 1. It is unnecessary, since market forces will produce a fair segregation of land uses. 2. It is open to corruption, particularly in respect of variances (permitted modifications or adjustments to the zoning regulations). 3. It can lead to premature use of land resources by owners who fear an unfavourable zoning change (down-zoning). 4. It is unequal in its effect, since a piece of property zoned for commercial use provides its owner with windfall profits at the expense of neighbours who must bear the costs of increased traffic noise and congestion. The most vociferous criticism is reserved for the practice of exclusionary zoning. This refers to the adoption by suburban municipalities of legal regulations designed to preserve their jurisdiction against intrusion of less desired land uses. Regulations requiring large lots, excessive floor space, three or more bedrooms, or excluding multiple-unit dwellings, high-density development or mobile homes, all serve to maintain high-cost housing and effectively exclude lower-income population. Supporters of zoning argue that it is a flexible tool and an effective means of allowing local residents to determine part of the character of their neighbourhood. Certainly its wholesale use during much of the twentieth century has helped to determine the land-use structure of metropolitan America. GROWTH MANAGEMENT IN THE USA A wide range of growth management strategies has been employed by cities and states in an attempt to moderate the negative effects of urban sprawl33 BOX 8.6 Common forms of land-use zoning in the USA Zoning is the division of an area into zones within which uses are permitted as set out in the zoning ordinance. If, however, owing to special circumstances, literal enforcement of the ordinance will result in unnecessary hardship for the landowner, the board of adjustment is empowered to issue a variance or relaxation of zoning conditions. The zoning system is, therefore, characterised by a number of forms of zoning, including: Urban geography 244 1. Cluster zoning and planned unit development. This involves the clustering of development on part of a site, leaving the remainder for open space, recreation, amenity or preservation. The planned unit development is an extension of cluster zoning in which developers are given freedom to design developments to meet market demand but within a negotiated set of criteria relating to pollution, traffic congestion, etc. 2. Special district zoning. This is designed to maintain the special land-use character of a place, such as the Special Garment Center District in New York City, designated to deflect market forces and maintain the garment industry against pressure to convert manufacturing space into offices and apartments. 3. Downzoning. Downzoning is the rezoning of an area to a lower density use and is often the result of neighbourhood pressure to avoid the development of intrusive land use. Since downzoning is likely to reduce the value of undeveloped land, it is likely to be objected to by the landowners. 4. Large lot zoning. This has the ostensible purpose of safeguarding public welfare by ensuring that there is good access for emergency-service vehicles, roads are not too congested and there is ample open space. It can also be employed to exclude undesirable residential development and maintain the social exclusivity of a neighbourhood. 5. Incentive zoning. This is basically a means of obtaining private-sector provision of public amenities by offering zoning bonuses in return for private finance of specific infrastructure. It is similar in principle to the UK concept of ‘planning gain’. (Table 8.5). Many of the techniques in use seek to link residential development to infrastructure provision. The town of Ramapo NY, 35 miles from downtown Manhattan, introduced a timed growth plan in the late 1960s to ensure that residential development proceeded in phase with provision of municipal services. To this end, developers were required to obtain a permit for suburban residential development. Where the required municipal services were available the permit was granted but elsewhere development could not proceed until the programmed services had reached the location or were provided by the developer. The town of Petaluma CA, 40 miles north of San Francisco, introduced, in 1972, an annual development quota of 500 dwellings. Applications to build were assessed against criteria that included access to existing services with spare capacity, design quality, open-space provision, the inclusion of low-cost housing and the provision of public services. The city of Napa CA introduced an urban-limit line intended to cap the population at 75,000 by the year 2000. Beyond the boundary essential public services would not be provided (the actual population in 2000 was 72,585). Several Californian cities, including San Diego and San José, have passed laws requiring voter approval for proposed developments. While ‘voter requirements’ have not prevented new development in the long run, they have affected the power balance between ‘pro-growth’ developers and ‘slow growth’ community interest groups, and have provided current residents with some compensation for the negative aspects of growth.34 In contrast to the UK, where a national planning system governs land development, restricting an individual’s right to develop their land is a politically contentious issue in the USA. This difficulty has been overcome by separating the development value of the Urban planning and policy 245 land from its existing use value and permitting the transfer of development value to another site. Under this system of transfer of development rights (TDR) owners of land can sell their development rights to developers in designated receiving areas in which they are permitted to build at an increased density that reflects the value of the transferred rights. While this mechanism has been employed in several areas, such as Montgomery County in Maryland, overall relatively few TDR programmes have been initiated. TABLE 8.5 MAIN AD VANTAGES AND DISADVANTAGES OF SELECTED GROWTH MANAGEMENT TECHNIQUES Technique Purpose Advantages Disadvantages Urban growth boundaries To set a limit on the extension of urban services Encourage more compact and cheaper to service development. Limit sprawl Need for restrictive zoning outside boundaries, and policies on phased growth within boundaries. Can be changed like zoning. If boundaries are too tight can raise cost of land Concurrency or adequate facilities policies To ensure that adequate infrastructure is in place before a development is built. To create phased development Slow the pace of development. Minimise leapfrog patterns. Put more of infrastructure cost on the developer Developers may fear they will lead to a moratorium on infrastructure and delay their projects Growth-rate caps To set a limit on the percentage rate of annual growth Promote phased/fiscally Can raise the cost of land and affordable development housing. Difficult to monitor Transfer of development rights To transfer Private developers pay development potential the costs. Concentrates from protected lands growth to designated growth areas Impact fees One-time payments from a developer to cover the cost of new services for new development Difficult to agree on sending and receiving areas Place more of servicing Raise cost of development. costs on to developer Setting fees is an imprecise process SMART GROWTH The smart thing about the smart-growth movement that emerged in the USA in the mid1990s is that while advocates share many of the goals of previous anti-sprawl efforts their language and methods are more pragmatic and inclusive. Rather than appealing narrowly to environmental sensibilities the focus is on broader quality-of-life issues based on a more comprehensive view of the urban development process.35 Many of the principles of Urban geography 246 smart growth (Box 8.7) overlap with those of new urbanism (see Chapter 9). Smart growth is concerned to protect land from (premature) development and promote development in desired directions. In 1997 the state of Maryland passed the Neighbourhood Conservation and Smart Growth Act in an effort to curb urban sprawl, encourage investment within growth centres, and protect countryside. The Act required counties and incorporated towns and cities to identify growth boundaries. Within these ‘priority funding areas’ state funds are available for infrastructure development; development outside the growth areas must be funded by the counties or the private BOX 8.7 Major principles of smart growth 1. Mix land uses. 2. Take advantage of compact building design. 3. Create a range of housing opportunities and choices. 4. Foster distinctive, attractive communities with a strong sense of place. 5. Preserve open space, farmland, natural beauty and critical environmental areas. 6. Strengthen and direct development towards existing communities. 7. Provide a variety of transport choices. 8. Make development decisions predictable, fair and cost-effective. 9. Encourage community and stakeholder collaboration in development decisions. developers. State office buildings, economic development funds, housing loans and industrial development financing are targeted within growth areas. The smart-growth policy does not prohibit developments outside priority funding areas and major developers that do not require public sector infrastructure investment can proceed unimpeded. But, more generally, state infrastructure dollars can help curb sprawl by influencing the location and amount of urban development. A key challenge for smartgrowth strategies is to achieve acceptability across the diverse groups with an interest in urban development. As Cullingworth and Caves (2003 p. 186) acknowledged, ‘without the necessary cooperation the system will not work’.36 PLANNING THE SOCIALIST CITY If market capitalism represents the governing philosophy for urban growth and change in the USA, the other extreme of the ideological spectrum is represented by urban planning in the socialist city.37 Following the Bolshevik revolution in 1917, debate over the ideal form of the Soviet socialist city was influenced strongly by the doctrines of Marx and Lenin (Box 8.8). The general principles for planning the socialist city were laid out in the 1935 plan for Moscow: 1. Limited city size. Reflecting the ideas of Ebenezer Howard, the optimum city size was generally considered to be in the range of 50,000 60,000 (to allow both the economic provision of items of collective consumption and the forging of a community ethos). Urban planning and policy 247 In the case of Moscow, with a population of 3.5 million, an upper limit of 5 million was envisaged. An internal passport system was introduced in 1932 to control population movements and hence city size. 2. State control of housing. Regulations to control the allocation of housing space were considered essential on grounds of equity and public health. 3. Planned development of residential areas. This was based on the superblock concept in which all daily facilities would be available within walking distance. Groups of superblock neighbourhoods would constitute a micro-region of 8,000–12,000 population which would offer higher-order services. 4. Spatial equality in collective consumption. The distribution of consumer and cultural services was to be based on the general principle of equal accessibility. BOX 8.8 Marxist-Leninist principles underlying the Soviet socialist city The application of Marxist-Leninist principles differentiated the Soviet socialist city from its Western counterpart in the following ways: 1. nationalisation of all resources (including land); 2. planned rather than market-determined land use; 3. the substitution of collectivism for privatism, most apparent in terms of the absence of residential segregation, the dominant role of public transport, and the conscious limitation and dispersal of retail functions; 4.planned industrialisation as the major factor in city growth; 5. the perceived role of the city as the agent of directed social and economic change in backward and frontier regions alike; 6. cradle-to-grave security in return for some restrictions on personal choice of place of residence and freedom to migrate; 7. directed urbanisation and the planned development of cities according to principles of equality and hygiene rather than ability to pay. Source: J.Bater (1980) The Soviet City London: Arnold 5. Limited journey to work. Norms were established to govern the time spent travelling to work (in large cities a forty-minute trip was deemed a maximum), and public transport was to be the dominant mode. 6. Stringent land-use zoning. To reduce journeys to work, housing and employment foci could not be too far apart, but as industry was to be a major urban employer, strict zoning and the use of green buffers were essential to separate residential areas from noxious industry. 7. Rationalised traffic flows. Heavy traffic flows in cities were to be handled by designated streets to minimise pollution externalities and congestion. 8. Extensive green space. Parks and green belts were an integral part of urban design, with public ownership of all land facilitating its conversion to open space. 9. Symbolism and the central city. The city centre was to be the symbolic heart of the city and locus for public demonstrations. This role was emphasised by grand architecture and the decentralisation of administrative and distributive services. Urban geography 248 10. Town planning as an integral part of national planning. Urban planning was subservient to national economic planning. The overriding dominance of industry and its controlling ministries largely frustrated the ideals of Soviet socialist urban planning. By the death of Stalin in 1953, the urban situation in the Soviet Union could be compared with Pittsburgh or Sheffield in the midnineteenth century. As French (1995) observed, ‘heavy industry dominated and polluted, living conditions were appalling, the workers had inadequate rationed access to food and all daily requirements’.38 During the post-Stalin era, priority in urban planning was necessarily given to housing construction, and the fringes of most cities were built up with micro-regions of apartment blocks in which the planned norms of service provision were achieved rarely, or reached only years after the housing was occupied (Box 8.9). In the latter days of the Soviet system, and with the introduction of Gorbachev’s perestroika programme, the mechanisms of urban planning changed, to enable public participation and to incorporate social planning into purely physical planning for cities. Reality replaced the idealism of the founders of Soviet socialism. The goal of a truly socialist city was never achieved.39 SOCIALIST URBAN FORM Urban form evolves in response to complex inter-actions among public and private forces. As we have seen, in socialist cities planned investment and state regulations were the defining instruments that created distinctive patterns of land use. The spatial character of the (formerly) socialist cities typically exhibited a relatively high density of residential settlement in the urban core; an adjacent band of low-density settlement where industry was located, often in large and polluting factories that would have been considered unacceptable urban investments in Western societies; and a swath of increasing residential density at the periphery, commonly in high-rise apartment blocks built to a standard design. Figure 8.3 portrays this archetypical rising BOX 8.9 The Soviet micro-region The micro-region was the basic spatial unit of planning in the Soviet socialist city. Comprising a number of apartment blocks (kvartaly) the micro-region of population 5,000–12,000 was designed to provide basic services for residents. Theoretically, any flat in a micro-region should be within 100–300m (90–270 yards) of the nearest shops with 80 per cent of the population not over 1,000m (0.6 mile) away. As French (1995) describes, in reality the vast majority of micro-regions do not meet the norms of service provision as a result of the pressing need to provide housing. Where shops have been provided, they are not in readily accessible precincts but scattered through the microregion on the ground floors of the apartment blocks, thereby entailing much time and effort in shopping. In particular, any given micro-region is unlikely to have the full range of shops selling daily and weekly necessities. Thus, journeys to neighbouring microregions are often necessary. Urban planning and policy 249 At a higher level in the urban structure, plans group several micro-regions to form a living region’ (zhiloy rayon) of 30,000–50,000 people with higher levels of service provision. In the largest cities, such as Moscow, plans envisaged groupings of ‘living regions’ into a ‘town region’ (gorodskoy rayon) with a major service centre which would obviate the need for journeys to the city centre except on special occasions. The 1971 Moscow Genplan contained eight such ‘town regions’, one of which was the old central district, the overall city focus. In practice the other seven surrounding ‘town regions’ failed to develop their foci; indeed, such regional centres have not emerged effectively anywhere as yet. Source: R.A.French (1995) Plans, Pragmatism and People London: UCL Press density profile or ‘camel back’ for St Petersburg, where residential density drops off sharply at a radius of only 4 km and then rises consistently to 14 km from the centre. In Budapest residential density also declines at around 4 km from the centre but here the absence of a ‘camel back’ reflects the local authorities’ more liberal policies (compared with the Soviet model) regarding private-housing investment and a lower reliance on high-rise public housing estates. Sofia shows more of a mix of high and Figure 8.3 Residential density profiles in selected European cities Urban geography 250 Plate 8.2 High-rise apartment blocks in Moscow are typical of publichousing provision in the socialist city low densities at all distances from the city centre. By contrast cities in the long-standing market economies, such as London, generally exhibit a smooth and declining density gradient, although the steepness of the drop and extension of relatively high residential densities over considerable distances from the city centre vary as a function of regulation, tax and investment policies affecting the land market. Under state socialism urban land and housing markets did not exist. Before the transition to capitalism almost all housing was state-owned and city governments restricted residential mobility as a way to tackle housing shortages. People lived in the same place for long periods and neighbourhood change was slow. Planners located stores and services according to the demographic profiles of neighbourhoods and, once located, urban functions remained in place for decades.40 In Moscow because large-scale housing construction did not begin until the 1960s most working- and middle-class families lived in old downtown buildings in subdivided, overcrowded apartments shared by several extended households (known as ‘communal apartments’). Blue-collar workers from outside Moscow who filled the least desirable jobs in exchange for a Moscow residence permit were given vacant rooms in communal apartments whose former residents had moved to new single-family units in high-rise housing projects on the city’s periphery. In marked contrast to the capitalist city people from different classes, professions, ages and ethnic groups lived in the same neighbourhood, the same building and even the same communal apartment.41 When an old building became unliveable all inhabitants were resettled to new housing and the upgraded building was converted to government offices or commercial space. (Thus, between 1960 and 1988 the total supply of housing in central Moscow dropped by 50 per cent.) In general, the better infrastructure, prestige and Urban planning and policy 251 accessibility offered by a downtown location contributed to the concentration of all types of economic activities there, including manufacturing, services and state offices. The uneven impact of post-socialist privatisation has reinforced Soviet-era privileging of the city centre in Moscow. With 7 per cent of the urban population in 1997 the central district (CBD) of Moscow employed 25 per cent of city jobs and contained 27 per cent of all institutions and enterprises. Such spatially focused growth stimulated demand for new office and commercial space in the central city which has prompted Moscow city government, like many entrepreneurial cities in the USA, to undertake large-scale redevelopment of the downtown in partnership with private capital. In the transition period since 1989 privatisation and the replacement of central planning with the free market, together with post-industrial restructuring, have catapulted cities such as Moscow from industrial socialism to postindustrial capitalism in a short period. In the post-Soviet city capitalist tendencies such as suburbanisation and counterurbanisation are in evidence, and social differentiation in housing and quality of life is increasingly apparent. In Moscow between 1990 and 1994 the level of private ownership of housing increased from 9.3 per cent to 49.1 per cent. This tenure change and the reintroduction of market mechanisms have been accompanied by increased residential mobility that has enabled some individual citizens to improve their living conditions. As well as suburbanisation, gentrification has appeared in central city redevelopment schemes42 while, more generally, there is a trend towards a ‘European’ urban residential structure characterised by an affluent inner core and a poor periphery43 Nevertheless, while individual choice has been enhanced for those able to participate in the housing market, state planning control of zoning regulations and real-estate taxes (coupled with a limited private mortgage system and declining real incomes for the majority) ensure that, in the short to medium term at least, the social geography of Russian cities will continue to be influenced strongly by local government. As in the former Soviet Union, in China the declining role of state enterprises in the economy, the introduction of land and housing markets and the opening up of cities to foreign investment have meant that the state and the centrally planned economy have a much reduced influence on urban development. As we have seen (Chapter 5), as a result of the dual effects of internal reforms and globalisation processes, Chinese cities have experienced significant transformation in their sociospatial structure in the reform era. Just as in the post-Soviet city, ‘Western’ processes of urban sprawl, spatial segregation and social polarisation are being reproduced in Chinese cities. The transition is evident in Beijing, where the creation of a modern CBD is under way, a real-estate market has emerged and the renewal of inner-city slums is in progress. The rapid growth of the urban economy and population has also resulted in peripheral expansion, with the built-up area of the city expanding by almost 50 per cent, from 335 km2 to 488 km2 between 1978 and 1998. The introduction of marked social and spatial polarisation in Chinese cities, as well as in other ‘transitional societies’ such as Vietnam,44 represents a new problem and challenge for urban authorities. Urban geography 252 TOWARDS PLANNING FOR SUSTAINABLE URBAN DEVELOPMENT Increasingly, urban policy and planning are required to adopt a long-term prospective role. This is particularly so when society, in managing urban change, seeks to strike a balance between economic priorities on the one hand and social and environmental priorities on the other. This issue is central to the question of sustainable urban development. Economic development is fundamental to human well-being, but growth which fails to recognise the limits of natural resources and the finite capacity of global ecosystems to absorb waste is a basis for long-term decline in the quality of life. Sustainable development aims to meet ‘the needs of the present without compromising the ability of future generations to meet their own needs’.45 Cities are major agents in depleting the quality of the environment for future generations. They are significant generators of gases, such as carbon dioxide, which cause global warming and of nitrogen oxides and sulphur dioxide, which contribute to acid rain. The greenhouse effect and ozone depletion are consequences of processes of urbanisation and industrialisation which consume raw materials and energy, and produce environmentally damaging waste products. It is important, however, not to allow the rhetoric of sustainable development to obscure the fact that cities will continue to be net consumers of resources and producers of waste products simply because of the relative intensity of social and economic activity in urban places. Neither should idealism be permitted to cloud the fact that most people will not relinquish voluntarily a cherished lifestyle (such as an urban workplace and rural residence). Furthermore, the goal of sustainability is not an integral element of market capitalism and will encounter opposition from entrenched interests. Most fundamentally, it is unrealistic to expect calls for restraints on economic growth to protect the future environment to be heeded generally in a world where millions of impoverished people face a daily struggle to survive. The importance attached to sustainability in the trade-off between environmental considerations and social, economic and cultural aspirations is clearly a function of general levels of well-being in a society. Thus the concept of sustainable urban development embraces more than environmental issues, and cannot be achieved merely by imposing pollution taxes or by promoting technical developments to reduce the energy consumption of cars and production processes. Sustainability must also address the key question of people’s lifestyle. In essence there are two broad approaches to sustainable urban development: 1. an environmental protection approach with a focus on a municipal programme to reduce the consumption of resources and minimise the environmental impact of development; 2. a holistic approach including an ecological component (stressing the importance of environmentally sound policies), economic aspects (development activities and fiscal issues) and social-equity issues (a fair distribution of resources and the distributional impact of policies). From this broader perspective the ideal sustainable community is characterised by: Urban planning and policy 253 1. environmental integrity: clean air, soil and water, a variety of species and habitats maintained through practices that ensure long-term sustainability; recognition that the manner in which natural resources are used and the impact of the individual, corporate and societ al actions on the natural processes directly influence the quality of life; 2. economic vitality: a broadly based competitive economy responsive to changing circumstances and able to attract new investment and provide employment opportunities in the short and long term; 3. social well-being: safety, health, equitable access to housing, community services and recreational activities, with full allowance for cultural and spiritual needs. The realisation of a sustainable and liveable city requires both an integrated planning and decisionmaking framework and a fundamental shift in traditional values and perspectives. There needs to be a change in focus from curative measures such as pollution reduction to measures based on prevention, from consumption to conservation, and from managing the environment to managing the demands on the environment. This will require change at the individual, community, business and urban levels. Some tentative steps along this road may be identified in the European Sustainable Cities project, which frames the search for sustainable urban environments as a challenge ‘to solve both the problems experienced within cities and the problems caused by cities [while] recognising that cities themselves provide many potential solutions’.46 At the local government level, Agenda 21 of the 1991 Rio Earth Summit agreed a number of proposals relat ing to waste management, energy conservation, the integration of land-use and transport planning, and the protection of natural habitats. The summit proposed that by 1996 most local authorities should have carried out a consultative process with local residents to develop a Local Agenda 21, including targets and timetables. In view of the increasing number of people living in urban settlements, strategies designed to substitute sustainable development for unsustainable growth can have a major influence on future urban policy and planning, and on the form and function of cities in the twenty-first century (see Chapter 30). For some analysts, the way towards an improved urban future lies in the creation of new settlements. We examine this theme in the following chapter. FURTHER READING BOOKS G.Andrusz, M.Harloe and I.Szelenyi (1996) Cities after Socialism Oxford: Blackwell J.Cullingworth and R.Caves (2003) Planning in the USA London: Routledge J.Cullingworth and V.Nadin (1997) Town and Country Planning in the United Kingdom London: Routledge T.Daniels (1999) When City and Country Collide: Managing Growth in the Metropolitan Fringe Washington, DC: Island Press M.Elson (1986) Green Belts London: Heinemann O.Gilham (2002) The Limitless City: A Primer on the Urban Sprawl Debate Washington, DC: Island Press R.Jackson (1981) Land Use in America London: Arnold P.Newman and A.Thornley (1996) Urban Planning in Europe London: Routledge Urban geography 254 M.Pacione (2003) Urban policy, in M.Hawkesworth and M.Kogan (eds) Encyclopaedia of Government and Politics London: Routledge, 795–808 J.Reps (1965) The Making of Urban America Princeton, NJ: Princeton University Press S.Ward (2004) Planning and Urban Change London: Sage JOURNAL ARTICLES W.Ackerman (1999) Growth control versus the growth machine in Redlands, California Urban Geography 20(2), 146–67 J.Claydon and B.Smith (1997) Negotiating planning gains through the British development control system Urban Studies 34(12), 2003–22 F.Gaffiken and B.Warf (1993) Urban policy and the postKeynesian state in the United Kingdom and the United States International Journal of Urban and Regional Research 17(1), 67–84 M.Pacione (1990) Private profit and public interest in the residential development process Journal of Rural Studies 6(2), 103–16 R.Peiser (2001) Decomposing urban sprawl Town Planning Review 72(3), 275–98 KEY CONCEPTS ■ Utopian communities ■ garden cities ■ planning gain ■ betterment value ■ green belt ■ Thatcherism ■ privatisation ■ land-use zoning ■ growth management ■ smart growth ■ socialist city ■ transitional societies ■ sustainable urban development STUDY QUESTIONS 1. Planning seeks to influence the process of urban development for ‘the common good’. Consider what is meant by this concept. Make a list of examples of planning actions that have a positive impact on the common good, and a list that you feel have a negative impact on the common good. 2. Identify the origins and development of the current system of urban planning in the UK. 3. Urban planning takes place within the framework of national urban policy. Examine the post-Second World War development of urban policy in the USA or UK and explain its effects on urban planning. 4. Examine the arguments for and against land-use zoning in the USA. Urban planning and policy 255 5. Identify the characteristics of the socialist city and consider how they may change in response to the demise of the command economies in Eastern Europe. 6. What do you understand by the concept of sustainable urban development? PROJECT Conflict over use of land is an integral part of capitalist urban development. Review back issues of your local newspapers (available in your local library) to identify examples of land-use conflicts. Analyse the reasons for the dispute and examine the motives of those involved. How was the dispute resolved and what effect did the decision have on the urban fabric? Who were the winners and losers in the matter? 9 New Towns Preview: the concept of new towns; the UK New Towns; the neighbourhood unit; private-sector new towns in the UK; pressures for urban development; new towns in Europe; new towns in the Third World; new communities in the USA; federal involvement; the profit motive; masterplanned communities INTRODUCTION The attraction of exchanging a decaying and overcrowded urban environment for life in a new, planned community has a long history that ranges from the Greek colonies in the eastern Mediterranean of the fifth century BC to the present day (see Chapter 3). Modern new towns primarily stem from post-Second World War developments in the UK, where, over a period of three decades, twenty-eight New Towns were constructed as part of a government strategy to alleviate the social, economic and physical problems of older urban areas. In this chapter we trace the roots of modern new-town development in the UK and its subsequent manifestation and reformulation in the USA and elsewhere. We then consider the future of the new town concept and the role of public and private agencies in new community building. THE BRITISH NEW TOWNS The British concept of new towns was inspired by reformist reaction to the evils of the nineteenthcentury industrial city and was promoted in the early twentieth century by the garden city ideal. The post-war British New Towns were conceived in response to the Barlow report (1940) and to Abercrombie’s Greater London Plan (1945).1 The recommendations of the Barlow report on the geographical distribution of the industrial population identified a need for satellite towns to help reorganise congested urban areas. The Abercrombie plan applied the general principles of decentralisation, urban containment, redevelopment and regional balance of opportunity to the particular problems of London. Following the Barlow report recommendations, the Greater London Plan proposed a ban on additional industrial development within the plan area, with planned redevelopment in the inner ring, and an enlarged green belt. The most radical proposal was the planned decentralisation of over a million people from the inner ring to the outer ring. This was to be achieved by the expansion of existing towns and the construction of eight New Towns to accommodate 400,000 people, with full local employment and community facilities (Figure 9.1). The legal power to implement these New towns 257 proposals was contained in the New Towns Act of 1946, which was the basis for the construction of twenty-eight New Towns throughout the UK2 (Figure 9.2). The general characteristics of the New Towns were defined by the Reith committee (1946),3 which recommended that they should: 1. be sited around large, densely built-up urban areas to help reduce their populations; 2. not be located less than 40km from London or less than 20km from the regional cities, to allow the development of independent communities; 3. have populations of between 20,000 and 60,000, the upper limit reflecting one of Figure 9.1 The Abercrombie plan for Greater London, 1944 Howard’s principles and the lower limit based on the concept of a threshold population needed to ensure a satisfactory level of service provision as well as a mixed class structure; 4. normally be built on greenfield sites to give planners full scope and avoid disruption of existing communities (although in practice most New Towns were built around an existing settlement nucleus); 5. avoid building on the best agricultural land. The twenty-eight New Towns created under the 1946 Act were designated in two main phases (Figure 9.2). Of the first fourteen (Mark I) New Towns, the eight in the Outer London ring plus East Kilbride outside Glasgow were intended to advance the objectives advocated by Howard, Barlow and Abercrombie, of dispersing population from overcrowded urban areas. Included in this group was Welwyn Garden City, which was redesignated as a New Town. The remaining five New Towns were designed to aid Urban geography 258 regional development, with, for example, Corby built to house workers for a new steel plant, and Glenrothes to accommodate miners in the East Fife coalfield. The determination of the post-war New Town planners to achieve socially balanced communities Figure 9.2 The British New Towns led to the widespread adoption of the neighbour-hood unit concept as a basis for physical planning. In Crawley New Town this attempt at ‘social engineering’ was evident New towns 259 in proposals to reflect the national class profile by integrating 20 per cent middle-class households in each neighbourhood unit as well as in the town as a whole. Such attempts to determine social structure and activity by physical design ignore the ‘disorganised complexity’ of urban life. In the 1950s growing recognition of the limited social relevance of the neighbourhood unit led to its role being reduced to one of ensuring that essential facilities were located in reasonable proximity to housing. The Conservative government brought the first phase of New Town designations to a halt in 1952, largely on the grounds of cost. The problems of urban overcrowding were to be tackled instead by overspill schemes arranged between local authorities with the financial support of central government under the provisions of the Town Development Act 1952. London made arrangements with thirty-two towns, including Basingstoke and Swindon, to build a total of 55,000 dwellings for overspill Londoners. Significantly, since a Town Development Act for Scotland was not passed until 1957, a second Scottish New Town was started at Cumbernauld in 1956. The New Town concept was not abandoned entirely by the government standstill, since several local authorities built their own new towns, as at Killingworth and Cramlington in Northumberland, and Thamesmead, built by London County Council. By 1961 several factors led to a decision to recommence the New Town programme: 1. The first generation of New Towns were showing signs of becoming profitable. 2. The Cumbernauld plan, with its Radburninspired separation of people and traffic, was attracting international acclaim. 3. There was continued criticism of living conditions in many cities, and increasing displacement of population as a result of comprehensive urban redevelopment schemes. 4. There was growing pressure on local councils to release building land to accommodate overspill from the large cities. Most of the Mark II New Towns were designed to accommodate conurbation overspill, with, for example, Skelmersdale and Runcorn sited close to Merseyside, Redditch and Dawley (renamed Telford) for Birmingham, and Washington for Tyneside. The overspill objective was reflected in higher population densities and increased target populations that reached 250,000 in Milton Keynes and 500,000 for Central Lancashire New Town. The desire for flexibility to accommodate subsequent growth was evident in the openended plans for later New Towns. Runcorn was laid out with three spines radiating from a centre separated by green wedges, while Warrington was built on a linear pattern that could be lengthened if necessary. The increasing role of the automobile in daily life also exerted a strong influence on New Town design, with vehicle-pedestrian separation an integral feature of Cumbernauld, Washington and Milton Keynes.4 In Runcorn emphasis was placed on providing a reliable public-transport system operating on a separate roadway. In addition to having an overspill function, the New Towns were also conceived as regional growth points, with, for example, the Central Lancashire New Town intended to act as a focus for the economic regeneration of the old textile area between Manchester and Preston. Urban geography 260 THE FUTURE OF THE BRITISH NEW TOWNS There is little doubt that the UK New Town programme was a success. In the first twenty-five years, fifteen towns were nearing completion and a dozen more started; they provided new housing and a congenial living environment for 1.4 million people, employment for 650,000, as well as a range of public facilities. Critics, however, point to the fact that the New Towns contributed less than 5 per cent of the national housebuilding programme and, in view of the continuing problems of cities, may be regarded as ‘too few and too late’. One option was to build on post-war plans for a larger number of New Towns, and during the 1960s several regional studies advocated expansion of the New Town programme. The South East Study identified sites for New Towns of 150,000 to house over 500,000 people displaced from London.5 These included proposals for New Towns at Newbury (Berkshire), Stansted (Essex), Ashford (Kent) and Bletchley (Buckinghamshire), the last of which was developed as Milton Keynes (Box 9.1). However, growing concern over the cost of New Towns (the development at Milton Keynes had cost £700 million) and over the potential competitive impact of the New Towns on declining inner cities led to the curtailment of the New Town programme in 1976. The first casualty of the decision was the newly designated sixth New Town for Scotland at BOX 9.1 Milton Keynes New Town The South East Study (1964) proposed a major new city at Bletchley in Buckinghamshire, and in 1967 the Milton Keynes Development Corporation was established to create a new town on an area of 21/900 acres (8,860 ha) with an existing population of 40,000, a planned intake of 150,000 and an expected population of 250,000 by the end of the century. Reflecting the economic and social confidence of the time, the idea was to provide a New Town that could cater for the ‘new affluence’ of the 1970s. A key goal in the plan was to provide a wide choice of living environments for future residents. The plan emphasised: 1. the desire to give maximum employment opportunities, with the emphasis on the service sector (education, communications, technical and office employment) rather than manufacturing; 2. the need for a variety of housing types, with a target of 50 per cent private housing; 3. low-density development, with net residential densities of six to ten dwellings per acre (two and a half to four per hectare), to attract the professional and managerial classes; 4. an efficient transport system based primarily on a grid pattern of town roads; 5. differing degrees of design control ranging from maximum control for specific areas (such as the city centre), through intermediate design areas to encourage good design from outside developers, to do-it-yourself areas (in which design controls cover only safety, sanitation and access) conducive to experimental housing. New towns 261 Milton Keynes has attracted criticism for its lowdensity housing layout, over-elaborate road pattern and poor public transport system; but it continues to grow and has attained many of its objectives. Its particular goal of becoming a ‘city of learning’ was enhanced by its selection as the location for the Open University. Stonehouse in Lanarkshire, which was abandoned. The expertise of the project team was redirected into the Glasgow Eastern Areas Renewal project. The Conservative government of Margaret Thatcher confirmed this switch of policy emphasis in favour of the inner cities, and set in train the privatisation of the New Towns. Under the New Towns and Urban Development Corporations Act of 1985 the Commission for New Towns was to be disbanded and the commercial assets of existing New Towns were to be progressively sold off, with private developers assuming an increasing role in the completion of the unfinished New Towns. Certainly the British New Towns represent a policy that was always intended to terminate once the post-war goal of developing a network of new, improved residential and living environments for former inner-city residents had been achieved.6 Whether this marks the end of New Town building in Britain or a temporary standstill, as in 1952, only time will tell. PRIVATE-SECTOR NEW TOWNS IN THE UK Although the overspill of population generated by large-scale post-war inner-city redevelopment has come to an end, the process of suburbanisation, combined with in situ growth in population and employment, has ensured continuing pressure on ‘outer city’ land for residential development.7 The most recent (1997) government projections estimated a need for 4.4 million new houses in England and Wales by 2016. The continuing problem is how to accommodate this growth. An early indication of a shift in government policy in favour of the development of private new towns came at Lower Earley, adjoining the town of Reading in the M4 growth corridor west of London,8 where over 6,000 dwellings were built by a consortium of private builders between 1977 and the early 1990s. The planning gain negotiated amounted to 8 per cent of the selling prices of the houses and was used to finance roads, school sites, leisure facilities and open space. The commercial success of this scheme led to the formation of Consortium Developments Ltd (CDL) in 1983, a group of nine major private house-builders, with the aim of creating new country towns of at least 5,000 dwellings with associated employment and social amenities. By 1989 CDL had lodged specific proposals to build four new towns in Essex (Tillingham Hall), Hampshire (Foxley Wood), Oxfordshire (Stone Bassett) and Cambridgeshire (Westmere) (Figure 9.3). This presented the Conservative government with the Urban geography 262 Figure 9.3 Proposed private-sector residential communities in SouthEast England political dilemma of whether to support the development plans of the private housebuilding industry (in line with the ‘enterprise culture’) or to heed the landscape protection goals of traditional Conservative voters in the rural areas (‘Nimbyism’). The first of the CDL proposals at Foxley Wood was approved provisionally in 1989, only for the decision to be reversed within a few months by a new Secretary of State for the Environment, who also rejected the plans for Stone Bassett and Westmere as the policy pendulum swung in favour of ‘green issues’. In 1992 CDL disbanded, having spent nine years unsuccessfully attempting to create private new towns. This did not, however, herald the end of private-sector interest in new community construction. As we saw in Chapter 8, the battle over development in the countryside around the major British cities is an ongoing conflict.9 In 1993 planning permission was granted for the development of a privately financed township (Hampton) of 5,200 houses for 13,000 people on a 2,000 acre (800 ha) area of abandoned brick and clay pits to the south of Peterborough. When completed, it will comprise four self-contained residential neighbourhoods including low-cost rental, special-needs and sheltered housing; and a 175 acre (70 ha) business park, hotel and leisure complex. Europe’s largest private-housing project of 10,000 homes is being developed by a consortium of fourteen contractors on a 1,000 acre (400 ha) site to the north of Swindon at a total cost of £900 million. The development is in the form of several ‘new villages’; the first of these (Abbey Meads) opened in 1996 with 200 families, with another 150 homes sold and 400 scheduled for completion within two years. The completed development will include commercial and industrial employment, schools, churches, shops and parkland. The involvement of the local council (Thamesdown) is limited to the supervision of housing design, leaving the private sector New towns 263 to provide the housing and infrastructure required to accommodate the projected need for 10,000–15,000 additional houses (over 27,000 people) in the area by 2005. Further indication of the emergence of the private sector as a major driving force for new town construction is seen in the responses of local councils to the government’s projected requirement for new housing. The Gloucestershire County Council structure plan proposed a new town of 6,000 homes on 670 acres (270 ha) of council-owned land in Berkeley Vale, to the west of Stroud, plus a second new town to the north of Gloucester. The greatest pressure on land is in South-East England (Box 9.2). In November 1997 the newly elected Labour government confirmed the previous government’s decision to build 50 per cent of new housing on open countryside, thus fuelling the greenfield brownfield debate and provoking arguments over sustainable urban development. NEW TOWNS IN EUROPE Somewhat paradoxically, in view of the British government’s withdrawal from New Town development, many countries in Europe and elsewhere have followed Britain’s lead, adopting the New Town principle and adapting the concept to their particular needs.10 In the Netherlands, new towns have been employed in the settlement of the polderlands (see Chapter 6). Central government control is also strong in Israel, where the goals of new-town policy have been to colonise inland territory, strengthen border defences and decongest the three major urban areas of Tel Aviv, Jerusalem and Haifa. In France the principal aim of the new-town programme has been to relieve population pressure on Paris, with five new towns constructed outside the capital (Box 9.3). In Sweden four of the eighteen new communities planned to house 250,000 people BOX 9.2 Where will the new houses go? The garden of England may become Britain’s biggest building site, with at least 116,000 homes expected to be built in Kent by 2011, an analysis of county councils’ plans reveals. The second worst affected county is Essex, where granted. 106,000 new houses will be built, followed by Devon, where 99,000 homes are planned. The counties that will escape the bulk of new housebuilding include Durham, where only 22,400 homes are expected, and Cumbria, with 27,500 new homes expected by 2006. In general the survey found that the greatest pressure for homes will be in the South East, South West, East Midlands and Eastern regions, with the North under the least pressure. Clive Aslet, editor of Country Life, which published the survey today, said: ‘These figures are a uniformly dismal tally. Anyone who keeps their eyes open, travelling around the shire counties, will realise the damage that has already been wrought in recent decades. These plans show the blueprint for development in the future. Enough is enough.’ The survey highlights the pressure on green belt land and Areas of Outstanding Natural Beauty. Government figures show that there will be 4.4 million new households by 2016 because of the growing number of single divorced and elderly people The Urban geography 264 councils’ structure plans give developers an idea of where planning permission is likely to be Green-belt land is threatened in Hertfordshire, villages on the Chiltern Downs in Buckinghamshire may be lost in urban sprawl, housing is expected to multiply near Stansted Airport in Essex, and there are fears for green-belt land in Northumberland. New houses for commuters are planned in Durham, new towns are planned for East Sussex, and Devon villages may become large towns, the survey warns. The Painswick Valley in Gloucestershire is under threat from Stroud District Council, which proposes to build 1,500 new houses. The valley is an Area of Outstanding Natural Beauty immortalised by Laurie Lee, and villages such as Dursley, Cam and Painswick are expected to bear the brunt of Gloucestershire County Council’s structure plan. There are plans to develop green-belt land surrounding Slough for 1,000 new homes. Several Berkshire greenfield sites have been earmarked, including Sandelford, near Newbury, where 1,250 dwellings are expected by 2006. Source: The Times, 13 November 1997 from Stockholm were developed by the municipal authority, which has purchased land since 1904 in anticipation of its eventual use for urban development. In Finland the new town of Tapiola was planned and built by a non-profit housing association created by six large private firms and labour unions. In the former Soviet Union many new towns were specialised in function, such as science cities, and most were intended to colonise the country’s vast territory. NEW TOWNS IN THE THIRD WORLD Osborn and Whittick (1977) list over sixty countries in which planned community development has taken place, as distinct from suburban growth.11 A large number of new towns have been built in countries of the developing world. In Brazil the urbanisation of the interior was aided by the development of the new capital city of Brasilia, begun in 1956. The new city has functioned as a regional growth pole, but the persistence of problems linked with poverty, including the development of shanty towns and an informal economy, testify to the difficulty of implementing central planning in a developing country with high levels of income polarisation and poor basic services. In Asia several states have created new towns. Construction of Islamabad as the capital city of Pakistan began in 1961, and by 2000 it had a population in excess of 1 million. The Indian new town of Chandigarh was designed by Le Corbusier and built during the 1950s and early 1960s on a neighbourhood plan of fifty-three self-contained sectors. Intended to accommodate 500,000 people, by 2000 the population of Chandigarh has risen to 642,000. Both Islamabad and Chandigarh were built at considerable cost, and in view of the region’s poverty the widespread construction of new towns cannot be viewed as a general solution to the housing and infrastructural problems facing Asia’s cities. One Asian state that has made extensive use of new towns is Singapore.12 Following independence in 1959, the government New towns 265 BOX 9.3 The new towns of the Paris region The 1965 Schème d’Aménagement et d’Urbanisme (Urban Management Plan) for Paris reversed the previous objective of controlling the growth of the city and sought instead to accommodate the expected growth (from 8.5 million in 1965 to 14 million by 2000) in an orderly fashion. Eight new towns were initially planned to house a total of 3 million people, but this was reduced to five new towns with a total population of 650,000 by 1990. Unlike the UK New Towns, these were not intended to be independent selfcontained settlements but were linked to the agglomeration by rapid transport links. The Parisian new towns were also intended to accommodate future population growth rather than redistribute population from the inner city, as in Britain. In terms of quality of living environment the Parisian new towns represent a major improvement over the highdensity grandes ensembles (social housing estates), and are now entering a phase of maturity. This was reflected in the revised region plan adopted in 1994, which recommends a shift of policy emphasis from the new towns and outer suburbs to the inner areas affected by industrial decline and attendant social problems, a trend reminiscent of that enacted in Britain two decades earlier. Source: D.Noin and R.White (1997) Paris London: Fulton assumed a dominant role in the provision of housing, much of which has been located in sixteen new towns. Singapore provides a good example of how the new-town concept may be adapted to fit local socio-political circumstances. In addition to the goal of overcoming a housing shortage the relocation of population into public housing was linked with a number of other objectives. These included the residential integration of different ethnic groups, physical redevelopment of central city squatter areas, and the release of land for commercial development in line with the creation of an exportoriented modern urban economy. Since 1989 the Singapore government has been engaged in a programme of upgrading earlier new towns, both as a boost to the construction industry and to improve residential environments. In Malayasia the two ‘intelligent new cities’ of Cyberjaya (a high-tech R&D core) and Putrajaya (an electronic federal administrative centre) are intended to boost the country’s transition from a ‘developng’ to a ‘developed’ nation. NEW COMMUNITIES IN THE USA The origins of the new-community concept in the USA can be traced to the first settlements. As we saw in Chapter 3, many early colonial communities were built according to carefully prepared plans. These were followed in the nineteenth century by a number of company towns, including Lowell MA (1822) and Pullman IL (1880), and in the early twentieth century by several privately developed suburban planned communities such as Forest Hill Gardens, New York (1911), and Palos Verdes Estates near Los Angeles (1923) (Figure 9.4). Urban geography 266 The first attempt to create a garden city was made at Radburn NJ by the New Yorkbased Regional Planning Association of America, which functioned between 1923 and 1933. Although Radburn never honoured three of the basic garden-city principles (the green belt, the provision of industry, municipal ownership and control), it remains one of the most significant US experiments in comprehensive urban planning.13 Among the innovative elements of the Radburn design were: 1. the use of the superblock as a building unit instead of the typical narrow rectangular block; 2. the separation of pedestrian and vehicular traffic through the use of overpasses and underpasses; 3. roads constructed for different functions; 4. a continuous park linking the superblocks. The Radburn development failed, owing to the onset of the Depression. Arguably the most significant contribution of the Radburn experiment for new-town development was the conclusion that even under normal conditions, a garden city could not be completed without federal assistance for land assembly, infrastructure financing and provision of low-income housing.14 The first large-scale federal involvement in newcommunity building took place during the 1930s, when the Suburban Resettlement Program sought to create alternatives to the urban slum. The resulting three green-belt towns (Greenbelt MD, Greenhills OH and Greendale WI) were each encircled by a green belt to limit expansion but did not include sites for industries and were not self-contained new towns in Howard’s sense of the term. Rather, they were small, well-planned suburbs that housed only 2,100 families in total when the programme ended in 1938. By 1957 Greenbelt in Maryland had a population of 7,500, at a gross density of two persons per acre (five per hectare).15 The symbolic importance of the green-belt towns was much greater than their achievement in housing provision. Nevertheless, they were remarkable for: 1. the degree and quality of physical planning; 2. the inclusion of suburban housing for low- and moderate-income groups; 3. social innovations, such as the introduction of consumer co-operatives for retail services, New towns 267 Figure 9.4 Evolution of new communities in the USA Source: G.Golany (1978) International Urban Growth Policies New York: Wiley Urban geography 268 Figure 9.5 Locations of major new communities in the USA reminiscent of Robert Owen’s initiative at New Lanark. Congressional sensitivity to popular criticism of the green-belt towns resulted in government withdrawal from the towns in 1952. Two main objections were raised by opponents. First, they felt that the towns were socialist ventures and part of a trend towards ‘socialistic regimentation disguised as cooperative planning’. Second, the towns placed the government in unnecessary competition with private enterprise, which, it was argued, could provide housing more effectively and more cheaply. The sale of the three projects to private parties brought the federal government about half the original cost of the towns and left the local governments without sufficient tax bases or effective control over undeveloped areas.16 Federal interest in new communities was revived during the 1960s by the Johnson administration, which passed the 1968 Housing and Urban Development Act, Title IV of which enabled the Department of Housing and Urban Development ‘to guarantee, and enter into commitments to guarantee the bonds, debentures, notes and other obliga tions issued by new town developers’. A ceiling of $250 million was put on the programme, with a maximum of $50 million for any one developer. The effects of the 1968 Act on new-community development were small, however. The economic situation in 1960–70 was such that private developers had difficulty in locating finance (‘patient money’) even with a federal guarantee, and only one new community (Jonathan MN) was guaranteed under this legislation. The 1970 Housing and Urban Development Act was more successful. This doubled the ceiling for loan guarantees to $500 million, extended the guarantee cover to public as well as private developers, and established a New community development corporation New towns 269 within the Department of Housing and Urban Development to manage the new towns programme and to undertake new community demonstration projects on federal land. Fifteen new communities stemmed from the 1970 Act (Figure 9.5). However, the impact of the energy crisis of 1973 and a depressed housing market created financial difficulties for most of the new towns which the Ford administration, committed to cutbacks in public spending and withdrawal from public housing and urban-renewal programmes, was unwilling to alleviate. With the exception of Woodlands TX, north of Houston (which survived because of revenue from the developer’s substantial natural-gas holdings), all the towns were foreclosed by the federal government, which sold the assets at a total loss of $570 million. More generally, the principle of federal involvement in new-town construction was undermined by a lack of ongoing political support (with the advent of the Nixon administration), and opposition to the concept of new towns from various interest groups, including: 1. mortgage bankers opposed to federal competition in the financial marketplace; 2. small builders opposed to a perceived bias in favour of large-volume builders; 3. big-city mayors who feared a loss of business, population and tax revenue. This led the Johnson administration to promote ‘new towns in town’ at Cedar-Riverside MN and Roosevelt Island NY. Underpinning this lack of commitment to a publicly supported new-town programme were national reservations about what is properly a public responsibility and what should be reserved to the private sector. These reservations were encapsulated by Eichler and Norwitch (1970 p. 308),17 who stated that If new towns are an anti-megalopolic concept …we in America are all for it. But if in order to have new towns we must grant to the state, especially the federal government, the power to control, own, and develop land, and the regulatory devices to channel growth to that land, we resist …. If private ownership of land and freedom of movement are sacrosanct and if new towns are needed, there is only one answer: private industry, with little or no aid or control from government, must build them. Such attitudes signposted the way ahead for newcommunity development in the USA. PRIVATE-SECTOR NEW COMMUNITIES IN THE USA In contrast to the social-welfare motives underlying public-sector new towns, the profit potential has been the principal force behind private-sector involvement in new community development. Two of the earliest large-scale new communities of the postSecond World War era were Park Forest IL, and Levittown NY.18 Undertaken by builders who had been involved in federal housing programmes during the war, these developments capitalised on the pent-up demand for housing, the easy finance provided by the Veterans Administration and Federal Housing Administration, and highway construction, which opened up peripheral areas for residential development. These and other new communities such as North Palm Beach FL were built for the middle class, Urban geography 270 with little recourse to the broader planning principles pioneered in Radburn, the greenbelt towns or British garden cities. During the 1960s house-builders were joined in new-community construction by entrepreneurs from a variety of backgrounds, including: 1. large landowners with properties ripe for development such as the 88,000 acre (35,000 ha) Irvine ranch in California,19 and the 50,000 acre (20,000 ha) O’Neill ranch, which became the site of Mission Viejo; 2. large corporations with cash available for diversification into real-estate projects, including Gulf Oil in Reston and the Humble Oil and Refining Company’s interest in Clear Lake City; 3. banks and insurance companies with, for example, Chase Manhattan Bank, Teachers’ Insurance and Annuity Association, and Connecticut General Life Insurance Company underwriting the development of Columbia.20 In contrast to the UK, where New Town locations were determined by central government, new communities in the USA are sited on the basis of individual developers’ perception of the economic situation in an area. Underlying factors which may influence this decision include climate (with California, Florida and Arizona favoured retirement areas), favourable legislative mechanisms for developers (such as the opportunity to create special districts to help finance public services) and the availability of large areas of undeveloped land (especially if it is under single ownership, to simplify acquisition). As Figure 9.5 shows, most new communities are located in a few states, including California, Florida and New York. These private-sector developments cater for a variety of residential and lifestyle preferences ranging from retirement communities (such as Sun City AR) to ‘postmodern’ neo-traditional planned developments (e.g. Seaside FL) that seek to simulate a particular version of small-town America for residents.21 Many of the new communities are designed in accordance with the principles of the new urbanism22 (Box 9.4) that have much in common with the smart growth strategies discussed earlier (see Chapter 8). Despite efforts to include subsidised housing for lower-income groups in some new communities, such as Columbia MD (Box 9.5), the market principle ensures that in general those who occupy the new master-planned communities are of uppermiddle income, of above-average education, white and, with the exception of retirement communities, young to middle-aged. The USA’s new communities are not, therefore, new towns in the UK sense of being balanced communities with a range of employment opportunities and housing choice; but they are home to an increasing number of Americans, for whom they provide a particular form of Ebenezer Howard’s ‘town country’ lifestyle. New towns 271 Plate 9.1 In private retirement communities, such as Sun City Center in Florida, residents are furnished with a lifestyle and facilities that suit their particular requirements BOX 9.4 New urbanism: principles of neighbourhood development 1. Neighbourhoods should be compact, pedestrian-friendly and mixed-use. 2. Many activities of daily living should occur within walking distance, allowing independence to those who do not drive, especially the elderly and the young. Interconnected networks of streets should be designed to encourage walking, reduce the number and length of automobile trips, and conserve energy. 3. Within neighbourhoods, a broad range of housing types and price levels can bring people of diverse ages, races and incomes into daily interaction, strengthening the personal and civic bonds essential to an authentic community. 4 Appropriate building densities and land uses should be within walking distance of Urban geography 272 transit stops, permitting public transport to become a viable alternative to the automobile. 5. Concentrations of civic, institutional and commercial activity should be embedded in neighbourhoods and districts, not isolated in remote, single-use complexes. BOX 9.5 Columbia MD Begun in 1967 and located twenty miles (32km) from Washington DC, Columbia covers 15,600 acres (6,300 ha) and will eventually accommodate 110,000 people. In 1995 it had 79,000 residents living in nine ‘villages’. Each village comprised several neighbourhoods of 700–1,200 dwellings which supported a neighbourhood centre with an elementary school, day-care centre, and local shopping and community facilities. Four neighbourhoods support a village centre with higher-order facilities. There is also a larger shopping centre for the whole community in the downtown city centre which contains a 70-acre (28-ha) shopping mall, entertainment facilities and office buildings. The road system also adheres to a hierarchical arrangement, from culs-de sac, pedestrian and cycle paths, neighbourhood streets and collector streets to the main US29 which bisects the town. Around 1,800 acres (730ha) were set aside for industrial parks. Onefifth of the site is reserved for parkland, and a continuous open-space system runs through the town. Socially, Columbia has made a conscious effort to be a racially integrated community, with one-third of residents being black. Subsidised housing units are provided and dispersed over five different sites to avoid the creation of a low-income ghetto. Residents belong to a village association that provides one member to the city-wide Columbia Association. This ‘new community’ of Columbia represents a $2.5 billion market investment in a particular lifestyle. FURTHER READING BOOKS T.Bendixson and J.Platt (1992) Milton Keynes: Image and Reality Cambridge: Granta C.Corden (1977) Planned Cities Beverly Hills, CA: Sage G.Golany (1978) International Urban Growth Policies: New Town Contributions Chichester: Wiley N.Griffin (1974) Irvine: The Genesis of a New Community Washington, DC: Urban Land Institute P.Hall and C.Ward (1998) Sociable Cities Chichester: Wiley A.Jacquemin (1999) Urban Development and New Towns in the Third World: Lessons from the New Bombay Experience Aldershot: Ashgate F.Osborn and A.Whittick (1977) New Towns: Their Origins, Achievements and Progress London: International Textbook Company New towns 273 JOURNAL ARTICLES K.Al-Hindi and C.Staddon (1997) The hidden histories and geographies of neo-traditional town planning: the case of Seaside, Florida Environment and Planning D 15, 349–72 C.Ellis (2002) The new urbanism: critiques and rebuttals Journal of Urban Design 7(3), 261–91 K.Parsons (1990) Clarence Stein and the greenbelt towns Journal of the American Planning Association 56(2), 161–83 A.Stockdale and G.Lloyd (1998) Forgotten needs? The demographic and socio-economic impact of freestanding new settlements Housing Studies 13(1), 43–58 B.Webster (2001) Gated communities of tomorrow Town Planning Review 72(2), 149–69 KEY CONCEPTS ■ neighbourhood unit ■ Nimbyism ■ greenfield-brownfield debate ■green-belt towns ■neo-traditional planned developments ■master-planned communities STUDY QUESTIONS 1. Explain the origins and growth of the new-town concept. 2. Examine the new-town development programme in any country with which you are familiar. Consider how successful it has been in relation to its primary objectives. 3. There is an ongoing debate on whether demand for new housing is best satisfied on greenfield or brownfield land. Make a list of the arguments for and against each form of development. 4. A major contrast between Britain and the USA in new-town development is that whereas in Britain the programme has been driven by the public sector, in the USA the private sector has dominated new community construction. Explain why this is so. Consider the advantages and disadvantages of each approach. PROJECT Select a new town or new community with which you are familiar and write a report explaining its rationale, origins, growth and development to the present day. Consider the future prospects for the new town/community. How does it compare with other new towns, and with the classical models of UK new towns? 10 Residential Mobility and neighbourhood Change Preview: residential mobility; why people move; the residential-location decision process; housing search; housing markets; neighbourhood change; housing abandonment; gentrification; displacement INTRODUCTION Housing is the largest user of space in the city, and exerts a profound influence on the structure of metropolitan regions. We have examined already the key private-sector (Chapter 7) and public-sector (Chapter 8) agents involved in the land-development process. Here we focus on the households that acquire, occupy and exchange the dwelling units produced. We begin by examining the nature of residential mobility in the city, then consider how the aggregate of decisions by individual agents affects the process of neighbourhood change. WHY PEOPLE MOVE Intra-urban mobility constitutes the vast majority of moves made by individuals and households within advanced Western nations. Residential relocation can be voluntary or involuntary (Figure 10.1). Although forced relocations, due to property demolition or eviction, can be significant in particular parts of a city, most individuals and households move by choice. Nevertheless, the stimulus for a voluntary move may be externally induced. It is generally agreed that the most important reasons for relocation refer to the characteristics of the housing unit. In Rossi’s (1955) classic study of residential mobility in Philadelphia,1 more than half the movers cited too little or too much living space as a contributory factor, with 44 per cent identifying it as the primary reason behind their desire to move. Adjustment moves accounted for 52 per cent of reasons for moving home in a study by Clark and Onaka (1983),2 with housing characteristics including space, quality and design of the unit, and a desire to shift from renting to owner-occupation of major importance. Neighbourhood characteristics were of lesser importance, as were accessibility considerations, with households generally prepared to trade off a longer journey to work to obtain housing with more amenities at less cost. As we saw in Chapter 4, in larger metropolitan areas the ‘indifference zone’ for travel to work can extend to over two hours in parts of southern California. Similar extensive ‘commuter-sheds’ surround London and other large cities. Residential mobility and neighbourhood change 275 Figure 10.1 Reasons for household relocation Plate 10.1 Moving home. In the USA one person in five moves home every year. In some instances the home also moves Urban geography 276 Induced moves are related to employment and life-cycle factors (Figure 10.1). Traditionally the life-cycle concept refers to the explicit material needs of families as they move through the various stages of the reproductive cycle. A number of major lifecycle events have been related to residential adjustment or relocation3 (Table 10.1). More recently the life-cycle concept has been reformulated as the less-deterministic concept of the life course. This avoids age stereotypes and acknowledges that economic and cultural factors can induce households at the same life-cycle stage to adopt different residential behaviours. Fundamentally, people move in the expectation of achieving a superior living environment. This premise underlies the ‘value expectancy’ model (Figure 10.2), in which migration behaviour is seen as the result of: TABLE 10.1 TEN LIFE-CYCLE EVENTS RELATED TO RESIDENTIAL ADJUSTMENT OR RELOCATION 1. Completion of secondary education 2. Completion of tertiary education 3. Completion of occupational training 4. Marriage 5. Birth of the first child 6. Birth of the last child 7. First child reaches secondary-school age 8. Last child leaves home 9. Retirement 10. Death of a spouse Source: D.Rowland (1982) Living arrangements and the later family life cycle in Australia Australian Journal of Ageing 1, 3–6 1 individual and household characteristics (e.g. life cycle, household density); 2 societal and cultural norms (e.g. community mores); 3 personal traits (e.g. attitude to risk); 4 opportunity structure (e.g. areal differentials in economic opportunity); 5 information (e.g. volume and accuracy). Residential mobility and neighbourhood change 277 Figure 10.2 A value expectancy model of migration decision-making. Source: R.Golledge and R.Stimson (1997) Spatial Behaviors New York: Guilford Press The first four factors combine to produce a set of migration goals, while the addition of the fifth factor influences the expectation of attaining these goals. Interaction between migration goals and expectancy of achievement results in migration intention that can be either to make an in situ adjustment or to relocate. Significantly, migration decisions are also affected by unanticipated events (which may be either constraints or facilitators). These include changes in family structure, the financial cost of moving, or distance. Enforced mobility may also create incongruities between migration intentions and actual behaviour. The model also acknowledges the causal significance of cultural signals or mores in motivating a desire to move or stay. How people ‘read’ these signals (which is a central concern of the postmodern perspective on urban change) may not necessarily be rational in a positivist sense, as for example when people have an affinity for a particular neighbourhood. THE DECISION TO MOVE The residential-location decision process may be conceptualised as the product of stress created by discordance between a household’s needs, expectations and aspirations and its Urban geography 278 actual living environment (Figure 10.3). As we have seen, the sources of stress may be internal (such as a change in family size) or external (e.g. the expiry of a dwelling lease). Furthermore, they may be dwellingspecific (such as a need for an extra bedroom) or location-specific (e.g. an extended journey to work due to a job change). Once it appears, a stressor affects the relationship between a decision-maker’s achievement level (what he or she already has) and aspirations. The degree to which this occurs varies between individuals according to factors such as socio-economic status, stage in the life cycle, experience and personality. Also, each individual will have a threshold aspiration level that adjusts on the basis of these factors, giving rise to different Figure 10.3 A stress model of the residential-location decision process Source: L.Brown and E.Moore (1970) The intra-urban migration process Geografiska Annaler Series B 52(1), 1–13 levels of tolerance of stress. When tolerable stress becomes intolerable strain, the household must decide on one of three courses of action: 1. to remain at the present location and either: ■ improve the environment, or ■ lower expectations; 2. relocate. Residential mobility and neighbourhood change 279 Environmental improvement embraces a range of activities depending on the nature of the stressors involved. A small dwelling can be enlarged with an extension, while dilapidated dwellings can be rewired and redecorated. Neighbourhood stressors such as crime or noise pollution can be tackled by approaching the local authorities or through residents’ action groups. Households constrained to remain in situ may have to lower their aspirations as a means of coming to terms with the situation. They could change their lifestyle or modify their plans (e.g. delay starting a family). More often it is a psychological matter of ‘dissonance reduction’, or learning to become indifferent to what one cannot attain. A large number of households, however, are able to realise their desire to relocate, and this process involves the search for and selection of a new residence (Figure 10.3). THE SEARCH FOR A NEW HOME Whether the decision to move house is voluntary or forced, all relocating households must: 1. specify an ‘aspiration set’ of criteria for evaluating new dwellings and living environments; 2. undertake a search for dwellings that satisfy these criteria; 3. select a specific dwelling unit. New dwellings may be evaluated in terms of site characteristics (attributes of the dwelling itself) and situational characteristics (the physical and social environment of the neighbourhood). The lower limits of the household aspiration set are defined by the characteristics of the dwelling currently occupied, while the upper limits are set by standards to which the household can reasonably aspire. In most instances these standards will be determined by income constraints but, as the value expectancy model suggests, other factors may be involved, including a desire to avoid certain areas that do not conform to a particular lifestyle. On the basis of their aspiration set, individuals initiate a search procedure to locate a suitable new dwelling. This search has a spatial bias.4 This may be illustrated by conceptualising the city as comprising four types of space: 1. Action space is the most extensive and refers to those parts of the city with which the individual is familiar, and includes a subjective evaluation of places. Generally, the longer an individual lives in a city the larger the action space and the greater the differentiation. New areas are assimilated into the action space as travel and information spread. 2. Activity space is the territory within which daily movement takes place and is normally organised around often-used nodes, including home, workplace, friends’ houses and shopping centres. 3. Awareness space indicates the extent of the territory on which a household has information concerning housing opportunities. This area is conditioned by the household’s action and activity spaces. 4. Search space is a subset of awareness space within which possible new residential locations are evaluated. Urban geography 280 The geography of these spaces affects the outcome of the residential relocation decision. In addition to information gained from active experience of the city, households also obtain information on new housing opportunities from secondary sources, including newspapers and, as we have seen in Chapter 7, real-estate agents.5 The relative importance of different information sources varies for different households. Access to information is also affected by search barriers that can: 1. raise the costs of gathering information (e.g. due to lack of transport, or time constraints on women with young children); 2. limit the choice of housing units and locations available (e.g. due to financial constraints, or discrimination in the housing market).6 As Figure 10.3 indicates, the eventual choice of new dwelling is based on the increase in satisfaction (place utility) produced by a move. It is important to recognise, however, that in all cities there are many households where residential location is constrained to the point where behavioural models are of limited significance.7 These subgroups include poor people, elderly people, unemployed people, transient people and special-needs groups such as lone-parent families and former inmates of institutions, as well as homeless people whose living space is the street. Focusing on the constraints involved in residential relocation underlines the fundamental importance of housing-market structure in conditioning residential mobility. HOUSING MARKETS In market societies private housing is a commodity produced and exchanged for profit. Outside the social- or public-housing sector, where housing may be allocated on the basis of some criterion of need or at a below-market price, the residential choices available to urban households depend primarily on their ability to compete in the housing market, which comprises housing providers, facilitators and con sumers (see Chapter 7). While the wealthy can compete in the market for the most attractive and expensive housing in the best areas, those with low incomes or the long-term unemployed are unlikely to be able to gain access to the private housing market except at the lowest levels. Those unable to buy must rent either from a private landlord or in the public social rented sector. The variety of households served by the housing market is demonstrated by the concept of housing class (Box 10.1). This identifies different sections of society with varying access to the housing market, and indicates that within a city there exist a number of distinctive sub-markets catering for different social groups. These submarkets tend to be localised and are, therefore, reflected in the residential mosaic of the city. In Table 10.2, which refers to the London borough of Southwark, six main housing groups are identified according to income, socio-economic status and national origin. Each group tends to seek accommodation of a different type (distinguished chiefly by tenure) within varying geographic limits. Each housing group is supplied with housing opportunities by different combinations of agencies offering different degrees of security of tenure. Residential mobility and neighbourhood change 281 BOX 10.1 The concept of housing class The concept of housing class was introduced by J.Rex (1971)8 in an attempt to understand the position of immigrant groups in the operation of the British housing market. It denotes different groups of people characterised by their access to particular types of housing, usually defined by tenure. Rex identified seven housing classes: (1) outright owners, (2) mortgagees, (3) tenants in purpose-built public housing, (4) tenants in publicly acquired slum housing awaiting demolition, (5) tenants of whole properties belonging to private owners, (6) house owners who must sublet parts of their property to cover the repayments and (7) lodgers who occupy one or more rooms in a dwelling shared with other households. The concept has been criticised subsequently for its assumption of a common system underlying the desirability of certain types of housing and housing tenure, and for being simply an inductive generalisation from a particular case (Birmingham). However, it does illuminate the important role of constraints as well as choice in urban housing markets. TABLE 10.2 HOUSING GROUPS IN THE LONDON BOROUGH OF SOUTHWARK Housing group Socio-economic Housing Location group opportunities in terms of tenure 1. Upper-income Primarily professional managerial and selfemployed 2. Indigenous, middleincome Skilled and semi- Permanent: owner Owner-occupied skilled occupation unfurnished, throughout London, local authority tenants mainly inner London and locally 3. Indigenous, lowerincome Semi-skilled and unskilled 4. Immigrant, middleincome, nonwhite (West Indian, Pakistani, Greek, etc.) Skilled and semi- Permanent: furnished, skilled owner occupation Temporary: furnished 5. Immigrant, Skilled, semimiddleincome, white skilled and Permanent: owner occupation (high-cost unfurnished) Temporary: furnished preceding ownership Degree of flexibility in location/throughout London Permanent: unfurnished, Mainly inner London, local authority, emphasis on Southwark longstanding owner occupation Temporary: furnished London area, often in immigration community, may be temporary UK residents Permanent: unfurnished, Mainly inner London for local authority furnished jobs, may be temporary Urban geography (mainly Irish) 6. Immigrant, lowerincome, white and non-white 282 unskilled Temporary: furnished UK residents Semi-skilled and unskilled Permanent: furnished, local authority Temporary: furnished Mainly inner London, many with their own communities Source: R.Knox (1995) Urban Social Geography: An Introduction London: Longman In British cities generally, a major distinction is that between the housing markets for owner-occupied dwellings and for public social housing. As Harrison (1983 p. 180) observed, ‘British cities are segregated by class… No laws of apartheid are needed to enforce this separation: it occurs as naturally as oil divides itself from water, by the play of unequal incomes and savings in the housing market.’9 In the US city, ethnicity may be substituted for tenure class, with the housing market for African-Americans commonly representing a separate submarket with its own distinct spatial arrangement (Box 10.2). In addition to, and often overlapping with, class- and ethnicity-based housing markets, submarkets also operate within cities to accommodate the needs and preferences of particular groups differentiated on the basis of age,10 religion11 or lifestyle.12 NEIGHBOURHOOD CHANGE The aggregate of individual household residential decisions is reflected in the changing socio-spatial structure of city neighbourhoods. Neighbourhoods are in a constant state of flux. Some analysts have sought to understand neighbourhood dynamics by identifying stages in the process of neighbourhood change13 (Table 10.3). It is important to recognise, however, that the life-cycle analogy inherent in many models can be misleading, since, unlike biological species, neighbourhoods do not follow a predetermined course of growth or decline. At any stage in a neighbourhood’s decline it may reverse direction (owing to inward investment) and begin a period of revitalisation. Similarly, upwardly mobile neighbourhoods may have their progress halted owing, for example, to negative externalities (such as the construction nearby of a noxious facility). BOX 10.2 The arbitrage model of neighbourhood racial transition Residential mobility and neighbourhood change 283 Racially-segregated housing markets are characteristic of US metropolitan areas. This means that when a neighbourhood changes from ‘white’ to ‘black’ occupancy it actually shifts from one housing market to another. This process is explained by the arbitrage model. In the model city, two housing groups occupy segregated zones, spanned by a transition zone accommodating both black and white households. Within each interior zone, levels of poverty increase towards the inner city, which means that the most affluent black neighbourhoods and the poorest white neighbourhoods lie adjacent to the transition zone. Average housing prices also decline from the outer city to the inner city. Neighbourhood racial change occurs when the transition zone moves into the white interior. The transfer of houses in the transition zone in response to the differences in prices that white and minority households are willing to pay for the same unit is termed arbitrage. The arbitrage process can be stimulated by an inflow of poor minority households to the black interior, that drives up prices for housing there. Households in the black interior are then willing to pay more for housing in the transition zone than white households are willing to pay for the same units. White home-owners who move out of the transition zone are more likely to sell to black households, who are prepared to pay higher prices. As was noted in Chapter 7, speculators may hasten change by buying homes from white households and selling them at higher prices to minority households in a process known as ‘blockbusting’. As more housing in the transition zone shifts to minority ownership the area may become less desirable for the remaining whites owing to concern over the exchange value of their property and/or racial prejudice. These households may sell in a depressed market, allowing the transition zone to move further into the white interior, thereby facilitating neighbourhood change. Downs (1981) identified a five-stage neighbourhood change continuum in which neighbourhoods at any stage can be stable, improving or declining14 (Figure 10.4). Factors that may increase a neigh bourhood’s susceptibility to decline or revitalisation are Urban geography 284 shown in Table 10.4. In the remainder of this chapter we will focus on the extremes of housing abandonment and gentrification. TABLE 10.3 SUMMARY OF NEIGHBOURHOOD LIFE CYCLES Stage Level Popu of lation const density ruction Physical changes: dwelling type (predo minant additions) Social changes Family Social Migr stru status ation cture (income) mobility 1. Suburban SingleHigh isation family (low(new density growth), multiple) ‘homog eneity’ 2. In-filling Multi-family Low, Medium (on decree (increa vacant asing sing land) slowly or stable) Ageing families, older children, more mixing 3. Downg rading (stability and decline) Older families, fewer children 4. Thinning NonLow out residential construction -demolitions of existing units 5. Renewal Public housing Conversions Very of existing low dwellings to multi-family High Other changes Low (but Young High increa families, (incre sing) small asing) children, large households Medium (increa sing slowly), population total down Declining (net densities maybe increa sing) Older families, few children, nonfamily house holds Increa Young sing (net) families, mainly children High net Initial in-migr development ation, stage; cluster high develo mobility pment; largeturnover scale projects, usually on virgin land High Low net First transition (stable) in-migr stage: less ation, homog low eneity in mobility age, class, turnover housing first apartments; some replacements Medium Low net Long period of (declin out-migr depreciation ing) ation, and stagnation; high some nonturnover residential succession Decl Higher Selective ining net out- nonresidential migr succession ation, high turnover Declining High net The second in-migr transition ation, stage: may high take either of turnover two forms, depending on Residential mobility and neighbourhood change 285 conditions Luxury Medium Increa Mixed Increa Medium high-rise sing (net) sing apartments Town house Low Decre Few Incre Low conversions asing (net) children asing Source: L.Bourne (1976) Housing supply and housing market behaviour in residential development, in D.T.Herbert and R.J.Johnston (eds) Social Areas in Cities volume 1 Chichester: Wiley, 111–58 Figure 10.4 Downs’ continuum of neighbourhood change Source: A.Downs (1981) Neighborhoods and Urban Development Washington, DC: Brookings Institution HOUSING ABANDONMENT Housing abandonment has been a major problem in inner-city neighbourhoods of most cities in the North-East USA since the 1960s. In New York City between 1965 and 1968 the number of dwelling units abandoned was sufficient to house 300,000 persons. By Urban geography 286 1978 tax arrears on abandoned property amounted to $1.5 billion, substantially contributing to the city’s fiscal crisis.15 The principal reasons behind large-scale housing abandonment were: 1. ‘white flight’ from the city to suburban residential areas; 2. physical deterioration; 3. lack of demand for housing in neighbourhoods marked by social pathologies; 4. declining rent incomes, owing to increasing municipal real-estate taxes and the limited rentpaying capability of low-income tenants; 5. withdrawal of investment due to red-lining by banks and mortgage lenders; 6. the imposition of stringent rent controls, which prompt landlords to reduce spending on repair and maintenance. If a situation is reached in which the housing unit becomes a liability to the owner, bringing in no rental income but generating taxes, the owner may take the economically rational step of halting payment of taxes due and in effect abandoning the building to the city (Box 10.3). The widespread abandonment of inner-city housing in the major American cities had slowed by the mid-1980s as: 1. The worst housing had been withdrawn from the market over the preceding twenty years. BOX 10.3 Housing abandonment in St Louis MO In the early 1970s St Louis MO had one of the highest housing-abandonment rates in the USA, with an average 4 per cent of the city’s housing stock abandoned. In the Montgomery neighbourhood up to 20 per cent of units were abandoned in 1971, and a domino or negative externality effect operated to add more territory to the impacted area. Montgomery was an area of low-quality housing that experienced a transition from being a white to becoming a black neighbourhood during the 1960s. This process culminated in the late 1960s in a sharp increase in the number of abandoned units. In 1970 nearly 80 per cent of housing was rented; crime and vandalism were rife, market values plummeted to zero and property owners ceased making mortgage payments. Slum landlords moved in, paying a token sum for units, then renting them out cheaply, providing little maintenance. The high foreclosure rate eventually slowed as the low value of the property and the lack of potential resale market made the procedure meaningless. Code enforcement programmes were also self-defeating as these too encouraged abandonment. Similarly, rehabilitation was a futile exercise in the face of terminal blight. As soon as units became vacant, fixtures were stripped, frequently by the owner, leaving demolition by the city as the only viable alternative. Residential mobility and neighbourhood change 287 TABLE 10.4 FACTORS UNDERLYING NEIGHBOURHOOD DECLINE AND REVITALISATION Revitalisation factors Decline factors 1. High-income households Low-income households 2. New buildings with good design or old Old buildings of poor design and no buildings with good design or historical interest historical interest 3. Distant from very low-income neighbourhoods Close to very low-income neighbourhoods or to those shifting to low-income occupancy 4. Inner city gaining (or not losing) population Inner city rapidly losing population 5. High owner-occupancy Low owner-occupancy 6. Small rental units with owners living on premises Large rental apartments with absentee owners 7. Close to strong institutions or desirable amenities, such as a university, a lakefront, or downtown Far from strong institutions and desirable amenities 8. Strong, active community organisations No strong community organisations 9. Low vacancy rates in homes and rental apartments High vacancy rates in homes and rental apartments 10. Low turnover and transience among residents High turnover and transience among residents 11. Little vehicle traffic, especially trucks, on residential streets Heavy vehicle traffic, especially trucks, on residential streets 12. Low crime and vandalism High crime and vandalism Source: Adapted from R.Downs (1981) Neighborhoods and Urban Development Washington, DC: Brookings Institution 2. Continued in-migration of minority groups, particularly Hispanics and Asians, maintained a demand for inner-city housing. 3. Federal low-interest loans and rehabilitation programmes, combined with action by local neighbourhood groups, prevented the decline into abandonment in many neighbourhoods. Nevertheless, the problem of abandoned housing remains severe.16 In Philadelphia in 1995 29,000 residential structures lay abandoned, of which 19,000 were so seriously dilapidated that rehabilitation would cost upwards of $110,000 per unit at a total cost of $2 billion.17 In some extreme cases of market failure, investor confidence can be restored only by public-sector involvement to pump-prime the regeneration of distressed neighbourhoods. Urban geography 288 GENTRIFICATION Gentrification is a process of socio-spatial change where the rehabilitation of residential property in a working-class neighbourhood by relatively affluent incomers leads to the displacement of former residents unable to afford the increased costs of housing that accompany regeneration. Gentrification commonly involves residential relocation by people already living in the city, and as such is not a ‘back to the city’ movement by suburbanites. Gentrification should also be distinguished from both the intensification of existing high-status areas and the process of neighbourhood revitalisation involving ‘incumbent upgrading’ where no spatial mobility is involved in an area’s social transformation.18 The classic example of gentrification occurs in a residential area that is initially middle-class. Typically the original residents move out from the central city as they establish families and their incomes rise, to be replaced by households of successively lower income. Eventually the cost of maintenance and reinvestment in the housing exceeds the financial ability of occupants and the area undergoes significant deterioration. The result is further in-migration of low-income households, overcrowding, subdivision of large houses into rental units to provide a rent income acceptable to landlords, and the eventual transition from owner-occupation to rented tenure. Landlords may decide to act as ‘free-riders’ and take no steps to invest in the maintenance or rehabilitation of their property (possibly in response to declining real profits owing to inflation and/or rent controls, and the presence of alternative investment opportunities). If so, the process of deterioration accelerates, possibly leading to abandonment (including ‘torching’ for insurance pur-poses). As Beauregard (1986 p. 48) observes, ‘the housing stock in this area is now inexpensive, and may attract gentrification’.19 A second form of gentrification refers to ‘loft conversions’, or the creation of improved housing units in mixed-use districts, as has occurred in areas of disused industrial and waterfront premises in many older cities.20 In addition to these manifestations of gentrification produced by individual households, large-scale developers and speculators can also purchase multifamily housing units and derelict industrial spaces for conversion into luxury condominiums and cooperative apartments. We can also identify a gentrification process in which local government takes the initiative through a major urban renewal project (see Chapter 11). Many cities began to experience the gentrification of select central and inner-city neighbourhoods during the 1950s, most notably London and New York. Today gentrification is a feature of the urban geography of the majority of larger cities in the advanced capitalist world including London,21 Paris,22 New York,23 Washington DC,24 Vancouver,25 Adelaide,26 Amsterdam27 and Madrid28 (Box 10.4). The phenomenon has also been identified in parts of Eastern Europe29 and the Third World.30 By the 1970s gentrification was widely regarded as an integral residential component of a larger process of urban restructuring linked with wider economic and social trends in capitalist society. Gentrification was seen as a facet of globalisation, the changing economic base from manufacturing industry to producer services, and the boom in office developments in locations such as Canary Wharf and Battery Park City.31 Gentrification is a circular process. Since the mid-1990s in some global cities, notably London and New York, the Residential mobility and neighbourhood change 289 Plate 10.2 Gentrification has transformed abandoned riverside warehouses on Butler’s Wharf, London, into luxury apartments and stimulated the growth of a ‘cappuccino society’ growth of a financial services economy has led to ‘re-gentrification’ of areas that were originally gentrified twenty-five years previously. In certain neighbourhoods, such as Brooklyn Heights in New York and Battersea in London, the ‘rent gap’ between old gentrified and newly gentrified property is as great as the difference between ungentrified and gentrified property was in the 1970s.32 This price difference is driving a new wave of ‘super-gentrification’ in favoured locations where high incomes from the financial services industry underwrite high-standard renovations of previously gentrified properties. THEORIES OF GENTRIFICATION Theories of gentrification have focused on two alternative types of explanation: 1. In consumption-side explanations, neighbourhood change is accounted for primarily in terms of who moves in and out.33 Recent postmodern interpretations have emphasised Urban geography 290 the role of cultural factors in gentrification,34 although it is necessary to avoid more extreme postmodern formulations in which culture supplants rather than complements economics in the process of gentrification.35 As we saw in our discussion of the value expectancy model (Figure 10.2), culture is an important but not predominant factor in residential location. BOX 10.4 Geography and gentrification in London: a tale of two postcodes West Hampstead has for long been the poor relation of Hampstead in North-West London. Anyone living in NW6, rather than NW3, would pay much less for their home and have fewer facilities to enjoy. Now, though, things are changing. The Burford Group is building a 220,000ft2 (20,000m2) development off the Finchley Road, to be called Gourmand, which will be a massive leisure and entertainment complex. It will include a leisure centre, an eightscreen cinema, restaurants, a health and fitness centre, and a Sainsbury’s superstore that should attract people in droves. Prices may still be 25 per cent lower than in Hampstead proper, but James Armstrong, an agent for Foxton’s, reckons that it will not be long before prices start to catch up. ‘As Hampstead gets more expensive and crowded, so people are happy to move ten minutes further away to a quieter street where lots of pubs and restaurants are opening,’ Armstrong said. Lots of people purchasing in the area have rented in Hampstead for a while, but found it too expensive to buy in. However, in West Hampstead a onebedroom flat will cost about £100,000 as against £145,000 in Hampstead. And West Hampstead is so different. The streets are wider, the houses more spread out and there are fewer red-brick mansion blocks. There are also none of the large multimillion-pound detached houses which are a trade mark of Hampstead. The most expensive houses in West Hampstead are to be found in Fawley Road and Crediton Hill, where an Edwardian six-bedroom property will sell for between £600,000 and £800,000. In Narcissus Road there are pretty twostorey threebedroom properties, reminiscent of Fulham terraced houses, selling for £330,000. In Pandora Road you could buy a five-bedroom house for about £320,000. Unlike Hampstead, the area still has a lot of property ripe for modernisation. Although some local developers are hard at work converting and rebuilding, it is surprising that some of the bigger names have not yet surfaced. ‘A lot of people are buying now for rental investment,’ said Armstrong, ‘because they know that in a year’s time they will have made their money back as prices rise. I see West Hampstead as a Netting Hill without the market or carnival,’ West Hampstead’s only drawback is parking. There are no restrictions on any of the streets, so people dump their cars and head for Finchley Road Tube station. Source: Sunday Times, 1 June 1997 2. Production-side explanations place emphasis on the role of the state in encouraging gentrification36 and the importance of financial institutions in selectively providing the capital for rehabilitation.37 According to Smith (1989), before gentrification occurs in a neighbourhood there must exist a rent gap, which is the difference between the Residential mobility and neighbourhood change 291 potential ground rent and the actual ground rent under the present land use38 (Figure 10.5). This arises as a result of cyclical patterns of disinvestment and reinvestment in the built environment. In urban neighbourhoods where physical deterioration and economic devalorisation have reached a critical level, the rent gap becomes sufficiently wide to attract investors seeking to purchase structures cheaply, undertake renovation, then either occupy the unit or resell or rent it to realise a capital gain. Empirical evidence of a causal link between a rent gap and gentrification is inconclusive, however, and there is ongoing debate over the explanatory value of the concept.39 In Hamnetfs (1991) view, although rent-gap theory can help to explain where gentrification may occur, it fails to explain why and when it occurs—in some neighbourhoods and not in others.40 In contrast to production- or supply-side theories, consumption-side explanations focus on the actors involved in the process. THE AGENTS OF GENTRIFICATION All the main agents of change identified in the urban land-development process (Chapter 7) are involved to varying degrees in effecting gentrification, each pursuing a particular set of objectives. Gov- Figure 10.5 The concept of the rent gap Source: N.Smith (1996) The New Urban Frontier London: Routledge ernment involvement is both direct and indirect. Central government policies to promote home ownership, including taxation policy (e.g. relief on mortgage interest repayments) and grants for home improvement, can facilitate gentrification. In North America more direct influence may be exerted by local governments, which stand to benefit from the replacement of low-income groups, with their attendant demands for social welfare programmes, by middle-class consumers whose incomes boost the local economy and whose investments enrich the tax base. Urban geography 292 Among strategies employed by local governments in the USA to promote desirable neighbourhood change are: (1) advertising certain neighbourhoods judged to have gentrification potential; (2) providing tax abatements for rehabilitation (e.g. the J-51 programme in New York City); (3) using community development funds to rehabilitate and improve public services in selected neighbourhoods; (4) employing code enforcement to make owners rehabilitate or sell their property; (5) designating ‘historic’ neighbourhoods; (6) reducing public-service provision in some neighbourhoods to encourage decline prior to facilitating reinvestment; and (7) re-zoning a mixed-use district or failing to enforce existing zoning statutes to facilitate gentrification. For gentrification to occur, such enabling strategies by government have to be accompanied by financial and property agencies becoming interested in the redevelopment potential of a neighbourhood. Landlord developers and real-estate agents have an important part to play in guiding potential gentrifiers to a neighbourhood, buying property and speculating, and displacing residents (e.g. by raising rents). Property interests, in turn, require the cooperation of financial institutions able to lend capital for investment in the built environment. When all these agents come together in a particular spatial context, gentrification can occur. Potential locations are generally characterised by substandard but structurally sound housing ‘with potential’, clustered to allow a contagion effect to occur, often with a unique spatial amenity such as a view, proximity to good transport links with the central business district (CBD), and the presence of local commercial activities (shops, restaurants) attractive to gentrifiers. Clearly, a satisfactory theory of the gentrification phenomenon must involve both supply-side and demand-oriented explanations. A list of the major demand-side and supply-side factors conducive to neighbourhood revitalisation is shown in Table 10.5. WINNERS AN D LOSERS As with all processes of urban change, the benefits and costs of gentrification are distributed unevenly. Among the main beneficiaries are the following: 1. Local government gains increased property tax revenue, and attracts higher-income residents with their associated spending power. On the other hand, middle- and upperincome newcomers may demand costly services as, for example, in Philadelphia’s Queen Village,41 where residents requested more police protection, better garbage collection, an improved school curriculum, cobblestone streets, buried electricity wires and new landscaping. 2. Initial owners of property benefit from rising property values in the neighbourhood which they capitalise through rental income or the sale of the asset. 3. Incoming owner-occupiers also gain. Early arrivals have the advantages of low property prices and higher potential capital gain but run the risk that no gain will accrue, and must usually invest more in property improvement. Residential mobility and neighbourhood change 293 TABLE 10.5 DEMAND-SIDE AND SUPPLYSIDE FACTORS UNDERLYING NEIGHBOURHOOD REVITALISATION City and metropolitan Neighbourhood Factor Factor Operation Strong downtown Creates demand for business district housing close to with growing downtown jobs employment Proximity to amenity such as lakefront, seafront, park or downtown Enhances long-term value Rising real incomes Increasing households’ ability to rehabilitate housing Good public transport Enhances convenience, especially for households with more than one worker Formation of many small, childless households Increases households that need less space, are orientated to urban amenities, and do not need public schools Access to high-quality Enhances attractiveness public or private schools to households with school-age children Operation Demand side Rapid inIncreases demand for migration of non- good-quality housing poor households No nearby public housing with school-age children who would dominate the public schools Enhances attractiveness to households with school-age children and incomes high enough to support renovation No in-migration of poor households Perceived in community as safe Enhances attractiveness as place to live Proximity to revitalised neighbourhoods Creates expectation that revitalisation will work here as well Permits older neighbourhoods to stabilise Supply side Long commuting times to downtown business district Make living near Single family housing downtown more desirable Simpler to rehabilitate than multi-family housing; fewer management problems Strong restrictions Limit suburban housing on suburban and jobs, enhancing city development housing and jobs Housing with interesting architectural features such as high ceilings, fireplaces, carved woodwork Attractive to young households, which are most likely to rehabilitate Rapid increases in Make city housing more attractive prices of Brick housing Easier to rehabilitate than frame housing, Urban geography 294 suburban housing easier to care for, lasts longer Loose housing market Enables poor households displaced by revitalisation to find adequate housing, possibly reducing resistance to revitalization Multi-family housing suitable for condominium ownership Owner-occupied property better maintained and residency more stable than rented property Rents not controlled Encourages property maintenance and investment in new rental units Financial institution Makes ownership and willing to provide rehabilitation easier mortgages and homeownership loans Easy condominium conversion Increases owneroccupancy Commitment by local Reassures private government to investors of long-term upgrade infrastructure value of homes and public services Strong neighbourhood organisation dominated by homeowners Creates pressure on local Housing and other Encourages private government to enforce structures in relatively investment by owners housing codes and improve good condition and lenders public services Later arrivals assume a smaller risk but gain less from rising property values. 4. Developers derive profits from buying run-down units and rehabilitating them for rent or resale to households who can afford higher occupancy costs than the previous residents. 5. Speculators realise a profit by buying properties, holding them without improvement, then selling at a higher price as the market improves. 6. Citizens may gain generally from a higher tax base and lower crime rates in a revitalised area, which must be offset against increased demand for city services. The major costs of neighbourhood revitalisation fall principally on displaced households who would not otherwise have moved but are forced to do so by rising occupancy costs that they cannot afford to pay.42 DISPLACEM ENT The displacement of former residents is a major social consequence of both public clearance and rehabilitation programmes on the one hand, and ‘free-market’ gentrification on the other. The displacement of working-class families reflects their weak position in society in general and the housing market in particular. Paradoxically, whereas they were once concentrated in the inner city because of their limited purchasing power, they are now being displaced from gentrifying inner neighbourhoods for the same reason.43 Displacement due to the super-heating of a local housing market by gentrification can occur in four ways: 1. the eviction of low-income tenants from buildings scheduled for rehabilitation as upperincome residences; Residential mobility and neighbourhood change 295 2. the involuntary departure of long-term families or elderly residents on fixed or limited incomes because of their inability to pay sharply escalating property taxes; 3. the inability of newly married children of existing residents to afford housing within the area they have traditionally regarded as their community; 4. reluctant migration of residents from an area because of the loss of friends or supportive social, religious or economic institutions. Property abandonment and gentrification are related processes in a cycle of neighbourhood change. Gentrification creates social conflicts but it does upgrade the physical environment. Continued neighbourhood decline, on the other hand, eventually leads to the development of a slum, an area characterised by an impoverished social and physical environment. The question of how best to address this, and other issues related to the quality and quantity of urban housing, are addressed in the next chapter. FURTHER READING BOOKS A.Downs (1981) Neighborhoods and Urban Development Washington, DC: Brookings Institution C.Hamnett (1984) Gentrification and residential location theory: a review and assessment, in D.Herbert and R. Johnson (eds) Geography and the Urban Environment volume 6 Chichester: Wiley, 283–330 P.Rossi (1955) Why Families Move New York: Free Press N.Smith (1996) The New Urban Frontier: Gentrification and the Revanchist City London: Routledge N.Smith and P.Williams (1986) Gentrification of the City London: Allen & Unwin JOURNAL ARTICLES W.Clark and J.Onaka (1983) Life cycle and housing adjustment as explanations of residential mobility Urban Studies 20, 47–57 W.Grigsby, M.Baratz, G.Galster and D.MacLennan (1987) The dynamics of neighbourhood change and decline Progress in Planning 28(1), 1–76 L.Lees (2000) A reappraisal of gentrification Progress in Human Geography 24(3), 389–08 J.Rex (1971) The concept of housing class and the sociology of race relations Race 12, 293–301 K.Temkin and W.Rohr (1996) Neighbourhood change and urban policy Journal of Planning Education and Research 15, 159–70 D.Wilson, H.Margulis and J.Ketchum (1994) Spatial aspects of housing abandonment in the 1990s: the Cleveland experience Housing Studies 9(4), 493–510 KEY CONCEPTS ■ life-cycle concept ■ value-expectancy model ■ aspiration set ■ action space ■ activity space Urban geography 296 ■ awareness space ■ search space ■ housing class ■ neighbourhood-change continuum ■ housing abandonment ■ gentrification ■ incumbent upgrading ■ free-rider ■ rent gap ■ displacement STUDY QUESTIONS 1. Examine the major reasons why people move house. 2. Consider the mechanics underlying the search for a new home and identify any constraints on residential relocation experienced by different groups in society. 3. Explain the phenomenon of housing abandonment. 4. With reference to particular examples, explain the process of gentrification. Who are the principal winners and losers in the process? PROJECT Read the property pages of the local newspapers over several months and make a note of the location, price and types of property being advertised. Are some parts of the city more dynamic in terms of property sales than others? Is there any evidence of gentrification? Can you identify different submarkets in different areas (e.g. for apartments or detached houses)? Do some real-estate agents focus on particular niche markets in terms of spatial coverage, housing type or market price? How do these residential patterns relate to your knowledge of the social geography of the city? 11 Housing Problems and Housing Policy Preview: trends in housing tenure; the growth of owneroccupation; the decline of private renting; public-sector housing; council-house sales; the residual isation of council housing; public housing in the USA; housing afford ability; homelessness; housing strategies; filtering; clearance; rehabilitation; alternative housing strategies INTRODUCTION The availability of shelter is a basic human need. Western governments have adopted different attitudes towards meeting this need. At one extreme, housing is regarded as a consumer good rather than a social entitlement. In the USA the government provides only 1 per cent of the total stock in the form of social housing, primarily as a complement to private urban-renewal programmes, and this effort is directed mainly at the poor, oneparent households, non-working families with dependent children or the low-income elderly. At the other extreme, in Eastern Europe housing was long considered a universal right and an essential part of the ‘social capital’, although in practice the inability of state resources to meet housing demand led to the promotion of alternative forms of provision, including housing co-operatives and owneroccupation. The intermediate position is illustrated by the states of Western Europe, in which housing is considered to be a limited social right, and where the state has intervened in the housing market to ensure a basic level of shelter for the majority of the population. In the absence of intervention, between one-quarter and one-third of the total population of most Western European countries would be unable to meet the full economic cost of the housing it occupies.1 In this chapter I identify and explain the main trends in the distribution of housing across the major tenures of owner-occupation, private rental and public housing. We examine import ant issues including the residualisation of council housing in the UK, housing affordability, and homelessness, and I provide a critical analysis of the main policy initiatives designed to improve the quality and availability of housing in Western society. TRENDS IN HOUSING TENURE IN THE UK The extent of state support for housing provision and its distribution among tenures varies. In the UK central government has played a key role in determining the quality and quantity of housing available for different social groups, with most assistance generally being directed to the owneroccupied sector (via mortgage-tax relief) and the social rented sector (e.g. via rent subsidies). The private rented sector has remained largely unassisted Urban geography 298 and subject to a lengthy period of rent control and regulation. The effects of this policy regime are revealed in the changing tenure structure of Britain’s housing stock. Over the course of the twentieth century a major transformation was effected in the tenure balance of housing in the UK. Prior to the First World War almost 90 per cent of households rented their accommodation from private landlords, 2 per cent rented from local authorities and the remainder were owner-occupiers (Table 11.1). By 1991 the position had been largely reversed, with only 7 per cent of households still renting privately compared with TABLE 11.1 HOUSING TENURE AND DWELLING STOCK IN GREAT BRITAIN, 1914–2001 (MILLION) Year Rented from local authority or Privately New Town rented No. % No. Total number of dwellings Owneroccupied % No. 1914 0.1 1.8 7.5 88.2 0.9 10.0 8.5 1944 1.6 12.4 8.0 62.0 3.3 25.6 12.9 1951 2.5 17.9 7.3 52.5 4.1 29.6 13.9 1961 4.4 26.8 5.0 30.5 7.0 42.7 16.4 1971 5.8 30.6 3.6 18.9 9.6 50.5 19.0 1981 6.6 31.0 2.7 12.6 11.9 56.4 21.2 1991 5.7 24.8 1.7 7.4 15.6 67.7 23.1 2001 3.6 15.4 2.4 10.6 17.1 74.0 24.7 68 per cent owner-occupation and 25 per cent who were council tenants. Most of the transformation occurred during the post-Second World War period. THE GROWTH OF OWNEROCCUPATION During the 1920s owner-occupation was not the preferred housing option of most people. Most households occupied privately rented accommodation, or council houses produced under the 1923 Housing Act (the Chamberlain Act). The latter, because of the high standards of construction and consequently of rent, were reserved primarily for better-off members of the working class. Owneroccupation of housing became attractive for the following reasons: 1. Local authority housing became more restricted to families displaced from slumclearance schemes following the 1930 Housing (Greenwood) Act. 2 Wages increased in real terms. 3 The period of mortgage repayment was lengthened from fifteen years to twenty or twenty-five years. 4 Local authorities started to act as mortgage guarantors. Housing problems and housing policy 299 5 Levels of car-ownership increased. 6 Lack of effective planning controls over suburban development meant that cheap land could be used for speculative house-building. 7 An increasing proportion of privately rented housing was diverted into the owneroccupied sector as landlords found it more profitable to sell property than remain in the regulated rented sector. As Table 11.1 shows, during the post-war period owner-occupation grew to become the largest sector of the UK housing stock. This was encouraged by several factors, including the following. 1. Government policy favoured owner-occupation via: ■ mortgage-tax relief (although the value of this was reduced progressively and removed finally in 2000); ■ reduced stamp duty on house sales (although again this is subject to change); ■ abolition of the tax on imputed net income derived from a dwelling by the owner; ■ government loans and grants to building societies to fund the purchase of pre-1919 privately rented dwellings and to keep interest rates below market rates; ■ the exemption of homes from capital gains tax; ■ the introduction of the option mortgage scheme to provide cheap loans to first-time buyers from low-income groups; ■ the sale of council houses. 2. Specialist financial institutions such as building societies developed, and banks expanded into the mortgage business. 3. The investment climate has generally been favourable to property development. 4. Households have had increasing difficulty TABLE 11.2 UK DWELLING COMPLETIONS BY TENURE, 1946–2001 Year Private Local authority Housing association Year Private Local authority Housing association 1946 31,297 25,013 – 1974 140, 865 103,279 9,920 1947 40,980 97,340 – 1975 150,752 129,883 14,693 1948 32,751 190,368 – 1976 152,181 129,202 15,770 1949 25,790 165,946 – 1977 140,820 119,644 25,127 1950 27,358 163,670 – 1978 149,021 96,196 22,771 1951 22,551 162,854 – 1979 140,481 75,573 17,835 1952 34,320 186,920 – 1980 128,406 76,597 21,097 1953 62,921 229,305 – 1981 115,033 54,867 19,291 1954 90,636 223,731 – 1982 125,416 33,244 13,137 Urban geography 300 1955 113,457 181,331 – 1983 148,067 32,833 16,136 1956 124,161 154,971 – 1984 159,492 31,699 16,613 1957 126,455 154,137 – 1985 156,530 26,115 13,123 1958 128,148 131,614 – 1986 170,565 21,548 12,531 1959 150,708 114,524 – 1987 183,736 18,809 12,571 1960 168,629 116,358 – 1988 199,485 19,002 12,764 1961 177,513 105,529 – 1989 179,593 16,452 13,913 1962 174,800 116,424 – 1990 159,034 15,609 16,779 1963 174,864 112,780 – 1991 151,694 9,645 19,709 1964 218,094 141,132 – 1992 139,962 4,085 25,640 1965 213,799 151,305 – 1993 137,293 1,768 34,492 1966 205,372 161,435 – 1994 147,370 1,982 36,612 1967 200,348 181,467 – 1995 152,841 2,135 38,284 1968 221,993 170,214 – 1996 145,891 868 32,148 1969 181,703 162,910 – 1997 152,841 468 27,597 1970 170,304 157,067 – 1998 145,891 428 23,325 1971 191,612 134,000 – 1999 149,151 165 23,315 1972 196,457 104,553 – 2000 144,250 302 22,910 2001 140,352 462 21,070 1973 186,628 88,148 8,852 finding alternative accommodation, owing to the decline in the privately rented and local authority sectors. The growth of owner-occupation has been particu-larly marked since 1979, following the election of a Conservative government committed to the creation of a ‘property-owning democracy’. The extent of the government’s commitment was indicated by the unequal assistance provided for different tenure types with, in 1990–1, tax relief on mortgages for owner-occupiers amounting to £7,500 million compared with the £900 million support for council housing.2 The increasing numbers and proportion of owner-occupied housing in the UK is revealed in Table 11.2. In addition to the stimulus provided by housing policies, the growth of homeownership in Britain reflected the belief that ownership of a house repre-sents a sound financial investment. This assumes that over the long term house prices will rise at least as fast as and probably faster than general inflation. Accordingly, it is better to pay mortgage interest on a loan of fixed capital value that results in the acquisition of an asset at diminishing real cost than to pay rent, probably at a similar level, but escalat-ing over time and with no asset to show for it. This argument has proved to be generally correct. How-ever, it is also predicated on the continuing ability of the mortgagee to service repayments. In a period of recession, unemployment combined with high debt/ equity Housing problems and housing policy 301 ratios can result in people’s homes being repossessed for non-payment of loan instalments. In the early 1990s in the UK, rising unemployment with the onset of recession (following a government rise in interest rates to control inflation) meant that 1.8 million people who had taken on large mort-gages to purchase houses during the boom of the late 1980s found themselves in a position of negative equity, where the exchange value of their house had fallen below the amount of the outstanding loan. In 1992 350,000 households were in mortgage arrears of more than six months, and 68,000 houses were repossessed.3 By 1996 400,000 mortgages were in arrears and more than 3 million households were unable to move because of being in a position of negative equity whereby the market value of their house was insufficient to cover the repayment of the mortgage on the property. In general, however, the advantages of owner-occupation have ensured that, since 1971, the majority of the UK housing stock is under this form of tenure. THE DECLINE OF PRIVATE RENTING As Table 11.1 reveals, from a dominant position in the housing stock prior to the First World War the privately rented sector has diminished in importance. Although it accounted for half of UK housing in 1951, by 1991 its share had fallen to less than 10 per cent of the total stock, although there was a small recovery by 2001. The principal factors underlying this decline are: 1. Slum clearance. The private rental sector included a high proportion of the oldest and poorest dwellings, many built before 1919. From the 1930s onwards slum-clearance programmes demolished hundreds of thousands of these houses. 2. Policies to deal with overcrowding. The Housing Acts of 1961, 1964 and 1969 increased controls over multiple occupation, and thereby reduced the number of private tenants. 3. Housing rehabilitation. Rehabilitation aided by improvement grants was often followed by tenant displacement and the sale of properties for owner-occupation, particularly in gentrifying neighbourhoods. 4. Poor investment returns. The comparatively poor rate of return from investment in private rented housing was apparent by the late nineteenth century. With the extension of the principle of limited liability, the development of the stock exchange and building societies, the expansion of government and municipal stock and increased overseas investment opportunities, capital flowed out of private rented property into these alternative investment opportunities. During the twentieth century, increased public intervention and, in particular, rent control further reduced the attraction of housing investment, while the increasing cost of repair and maintenance also eroded profits. 5. The ethic of home-ownership. Although there was some building for private rental in the inter-war period, the bulk of private house construction during the building boom of the 1930s, when interest rates were low and land, materials and labour cheap, was for owner-occupation. 6. Subsidies and tax allowances. In contrast to the subsidies provided to council tenants and the tax advantages for owner-occupiers, only since 1973 have private-rented tenants received rent allowances. Landlords of privately rented housing were not Urban geography 302 permitted to set a ‘depreciation allowance’ against their tax liabilities, and this encouraged them to sell for owner-occupation. An investigation into the decline of the private rented sector by the House of Commons Environmental Committee (1982)4 concluded that private rented housing fell short on all the criteria for sound investment—the level of risk, liquidity, expected return on capital, and management responsibility—when compared with alternative opportunities. In London the large commercial landlords were withdrawing from the market, selling off their freeholds to insurance companies and friendly societies, which in turn were selling long leaseholds to owner-occupiers. Although there were 50,000 purpose-built flats in 1,300 blocks in central London in 1966, less than one-third were still let under the Rent Acts in 1982, the majority having been sold off for prices ranging from £30,000 to £200,000.5 Overall, more than 3 million dwellings have been sold by landlords to owneroccupation since the Second World War. Since relatively little new property has been built for private renting, the remaining private rented stock is old and, owing to a succession of rent controls (Box 11.1), in poor condition. In the USA (where there is negligible competition from public-sector housing) and in France, Germany, Denmark and the Netherlands (where there are only mild rent controls, coupled with subsidies) private renting is also in decline and confined to mainly older and less attractive housing. It should be noted, however, that the shrinkage of the private rented sector has been socially selective. In the larger cities of Europe, North America and Australia the demand for centrally located luxury flats remains sufficient to maintain investment in this type of property.6 BOX 11.1 Rent controls Rent controls represent one of the most contentious pieces of housing legislation ever enacted. Introduced to control profiteering by unscrupulous landlords who sought to take advantage of housing shortages following the First World War, rent controls have tended to persist because of government reluctance to risk electoral unpopularity by their removal. The case for rent control was boosted in the UK between 1957 and 1965 when, following the 1957 Rent Act, which decontrolled the rents of 2.5 million dwellings, the practice of letting furnished accommodation to house-hungry and vulnerable immigrants at extortionate rents (‘Rachmanism’) came to public attention. Overcrowding, illegal subletting and insanitary conditions became intense in areas where immigrants concentrated, leading to spasmodic racial conflict and urban riots from the late 1950s onwards. Although it was introduced for a commendable social purpose, a major defect of rent control has been its role in reducing the supply of private rented accommodation available by restricting the ability of landlords to gain an adequate profit. This has led to physical deterioration of housing stock through neglect and, in some cases, to property abandonment. The reintroduction of rent regulation in the 1965 Rent Act contributed to the ongoing decline in the privately rented housing stock. The Housing Acts of 1980 and 1988 attempted to reinvigorate the private rented sector by introducing a form of decontrol for new lets: (1) assured shorthold lettings for a fixed Housing problems and housing policy 303 period at the end of which the landlord had vacant possession, and with rents set by a rent assessment committee if necessary; and (2) assured tenancies, with a high degree of security of tenure and rents agreed between landlord and tenant. Existing tenants of rentcontrolled housing would continue to be protected by the Rent Acts. A negative aspect of this dual provision was that some landlords sought to harass tenants in rent-controlled flats to switch to an assured shorthold or assured tenancy arrangement. Rent control is, therefore, a double-edged sword: while seeking to protect tenants from exploitation, it has also induced a reduction in the stock of available privately rented accommodation. THE DEVELOPMENT OF PUBLIC-SECTOR HOUSING The decline of private rented housing, the traditional source of shelter for those denied access to the owner-occupied sector, led to direct government intervention in the provision of housing. By 1939 1.18 million houses had been constructed by the public sector, amounting to 10.3 per cent of total stock. The growth of municipal housing also stemmed from: 1. the failure of nineteenth-century housing associations to demonstrate that it was possible to provide decent dwellings in sufficient numbers at a rent affordable by the lowpaid; 2. increased working-class militancy over housing conditions, vividly displayed in the Glasgow rent strikes of 1915; 3. the shortage of housing in the post-First World War era; 4. a gradual shift away from the Victorian ideology of self-help and the pervading belief in the efficiency of the market mechanism. The 1919 Housing (Addison) Act provided subsidies to local authorities to make up losses incurred in providing houses over and above those borne by a one-penny rate.7 Houses built under this Act were intended for the ‘general needs’ of the working class, but the high standards inspired by the garden city ideal and high rents placed them beyond reach of ordinary working-class people. In the Mosspark scheme in Glasgow, where the rent for a threebedroom house was £28 per year, 16.2 per cent of the first tenants were ‘professionals’, 30.3 per cent ‘intermediate class’ and 27.5 per cent ‘skilled nonmanual’. Only 18.0 per cent were manual workers. These garden suburbs remain among the most desirable of the municipal housing stock.8 The 1924 Housing (Wheatley) Act gave further impetus to council-house building with over 500,000 houses constructed, but the poorest of the working class remained in overcrowded slum conditions. The 1930 Housing (Greenwood) Act gave local authorities direct responsibility for rehousing communities displaced by slum clearance, for keeping rents within reach of the poor, and for targeting their efforts at slum dwellers previously excluded from council housing. In total, almost 300,000 slum properties were demolished in the 1930s, 1 million council houses were built under the slum clearance programme, and 4 million people relocated. Although the programme tackled some of the worst physical conditions, the minimum standards employed in the rehousing schemes, the lack Urban geography 304 of community facilities and limited local employment created a basis for future social problems. Britain emerged from the Second World War with a housing shortage since during the conflict there had been very little new construction, maintenance had been neglected, over 200,000 dwellings had been destroyed by bombing and more than 250,000 had been made uninhabitable. In addition, because of Britain’s early urbanisation (see Chapter 3), much of the housing was old and substandard; and the necessary slum clearance programme added to the shortage of housing. Local authorities were given the main role in providing new houses for rent, and for most of the post-war period there was general agreement across the political spectrum of the need for a council-house building programme. By the 1970s the crude housing shortage had been resolved, at least in national terms, although regional shortages remained. Despite significant progress towards the goal of ‘a decent home for every family at a price within their means’,9 by the late 1970s a number of criticisms were directed at the UK housing stock, including the following: 1. The social-housing sector was too large and expensive to support; the subsidised rents were not sufficiently well targeted and benefited many who did not need such a level of support. 2. The below-market level of rents contributed to low levels of maintenance undertaken by many local authorities, which stored up problems for the future. 3. Novel building methods employed in the 1960s and 1970s had given rise to structural problems in some estates (with, for example, flat-roofed housing designed originally for a scheme in Algeria employed in the much wetter climate of Glasgow). 4. Problems of damp penetration, inadequate insulation and costly-to-run heating systems provided poor living conditions and had adverse effects on health. 5. The size of the council-housing operation made effective management difficult. 6. The large, homogeneous council estates lacked a mix of social classes and were often constructed without social amenities. 7. As unemployment rose, problems relating to poverty and deprivation began to concentrate in many estates (see Chapter 15). All these factors contributed to a re-examination of national housing policy. The greatest force for change, however, was the ideological conviction of an incoming Conservative government committed to reducing public expenditure. With reference to local authority housing, a major divide opened between the government’s view of council housing as a ‘welfare service’, assisting households unable to afford other types of housing, and the Labour Party’s belief that the public sector should supply housing for ‘general needs’ (i.e. should satisfy the demand from all households who wish to rent rather than buy). The new government’s housing policy set out to: 1. increase owner-occupation; 2. provide a wider choice of landlord in the social rented sector; 3. revive the privately rented sector; 4. reduce the role of local authorities as direct housing providers. Housing problems and housing policy 305 Plate 11.1 The Easterhouse estate on the periphery of Glasgow is typical of the kind of public housing constructed in British cities during the early post-war period A key component of the Conservatives’ housing policy, introduced in the 1980 Housing Act, was the right of council tenants to buy their homes at a discount price that reflected length of tenancy. The case for the sale of council houses rested on the following arguments: 1. Council housing was rising in cost and the growing gap between cost rents and the ‘fair rents’ charged under the Rent Acts required increasing government subsidies. Some proponents of the ‘right to buy’ argued that if long-standing tenants (those of more than thirty years’ standing) were given their houses and others sold at favourable terms, the savings on subsidies and revenue raised could either lead to a reduction in taxation or allow funds to be directed into other areas. 2. Over the long term, it was thought, it would be of more economic benefit for tenants to own than to rent their council house. Kilroy (1982) demonstrated that over thirty-five years the net outgoings (at 1982 prices) of an average tenant who bought his £14,000 house (at a 40 per cent discount) would be £12,000 less than those of one who continued to rent. The purchaser would also have a well constructed, appreciating asset which they could pass on to their children.10 3. Owner-occupation offers a feeling of independence, freedom over the care and condition of the property, and greater locational mobility. Critics of council-house sales countered these arguments on the grounds that: Urban geography 306 1. The ‘rising costs of council-house subsidies’ case is often based on extreme examples such as inner London boroughs, and ignores the fact that local authorities are able to distribute these costs over the whole of their stock by using the rent surpluses from older homes to keep down rents on newer houses built at a time of higher capital costs and interest rates. 2. The argument that selling houses can contribute to a major cut in public expenditure was rejected by those who questioned the financial calculations employed. They pointed out, for example, that proponents of council house sales included Supplementary Benefit (welfare) payments in the cost of council housing, yet these would still need to be paid to low-income householders if they converted to owneroccupation. Omitting this income support cost meant that in 1977 8 the cost of council house subsidy (at £1,100 million) would have been comparable to the tax relief given to ownermortgagers. 3. Normally only higher-income tenants could afford to buy, even with a discount, and even some of these might face financial difficulty if they reached retirement age before they had paid off their mortgage, or if mortgages were based on contributions from the income of several family members. Also, if family members contributing to mortgage payments moved away, they would be unable to qualify for a mortgage of their own unless they forced the sale of the mortgaged council house. 4. It is mainly the better-sized, higher-quality dwellings that are sold off, leaving families on lower incomes with a reduced chance of improving their housing position through a transfer within the local authority system. 5. The sale of older rent surplus-yielding housing in attractive locations would reduce the possibility of cross-subsidisation and thereby necessitate higher rents for disadvantaged tenants in inner-city locations. The re-election of the Conservative government in 1983 was followed by the 1984 Housing and Building Control Act, which introduced higher discounts (up to a maximum of 60 per cent) and a reduction in the eligibility period (from three to two years); and by the 1984 Housing Defects Act, which provided 90 per cent repair grants to remedy defects not apparent at the time of purchase. Between 1980 and 1991 1.46 million sales had been completed and the Treasury had received over £20 billion, making the right to buy (RTB) scheme one of the largest privatisation initiatives of the Thatcher era. The privatisation of council housing was also promoted in the provisions of the 1988 Housing Act, which enabled local authorities to transfer their housing stock to housing associations with the approval of a majority of tenants. By 1994 thirtytwo local authorities, mostly under Conservative control, had transferred the whole of their stock, amounting to 149, 478 dwellings, and in England, by 2000, the ‘loss’ of local authority housing through ‘large-scale voluntary stock transfer’ (LSVT) had reached 450,000 dwellings. The Conservative view of local authorities as ‘enablers’ facilitating action by other bodies, rather than direct providers of social housing, is evident in Table 11.2, which shows the reducing role of council-house building and the growing importance of the housing association sector as an alternative source of rented accommodation. Housing Housing problems and housing policy 307 associations obtain their loan finance for investment in new-house construction and rehabilitation from the Housing Corporation (which oversees their activities) and, since the 1988 Housing Act, from privatesector institutions. By 1992 there were 2,550 registered housing associations in Britain with a total stock of 750,000 houses (3 per cent of total stock). By 1998 housing associations accounted for 1.2 million dwellings or 5 per cent of total housing stock. Housing associations may cater for general housing needs or for the special needs of groups such as elderly or disabled people. They vary in size from national associations owning in excess of 10,000 houses to a majority which own less than l,000.11 THE RESIDUALISATION OF COUNCIL HOUSING As a consequence of RTB and estate transfers, the public-rented sector in the UK is becoming increasingly residualised. This is evident in several ways: 1. Since 1981, when almost 7 million council houses accounted for 31 per cent of the UK stock, there has been both an absolute and a relative decline in the contribution made by public-rented housing. 2. The quality of the council stock has been declining, with an increased incidence of design and structural faults and difficult-to-let dwellings. Whereas in the inter-war and immediate post-war years council housing was a privileged tenure, it is now widely seen as only for those unable to gain entry to owner-occupation. 3. Since the early 1970s the social composition of council tenants has altered, with a relative decline in the population of non-manual/ skilled manual workers and an increase in personal service/unskilled manual workers. This has led to less of a social mix within the working class, and a higher degree of socio-spatial polarisation.12 4. An increasing and disproportionate number of council tenants (such as lone parents, elderly people, sick people, disabled people and unemployed people) are dependent upon income support. A growing proportion of council tenants belong to an ‘underclass’ dependent on the state rather than the labour market for most of its income. It is important not to confuse the symptoms of residualisation with the causes. The residualisation of council housing in the UK is a consequence of broader economic and social processes, including high unemployment resulting from technological innovation and deindustrialisation, the withdrawal of private landlords from the low-rent market, increased demand for owner-occupation among the more affluent, and government policies aimed at reducing public expenditure. There is an increasing possibility that the future role of council housing will be to provide shelter of last resort to those unable to go elsewhere. PUBLIC HOUSING IN THE USA The restricted role envisaged for council housing by the political right in Britain finds an echo in the history of US attempts to enact a programme of public housing.13 Like many other social programmes, public housing in the USA is a product of the New Deal era as enunciated in the Housing Acts of 1937 and 1949. Two basic means of supplying housing Urban geography 308 to low-income families have been used (Figure 11.1). The first employed a public agency to develop, own and manage housing. By the end of 1974 67 per cent of the total publichousing supply (860,000 units) had been developed in this way. Figure 11.1 Public-housing policy in the USA Source: G.Galtner (1995) Reality and Research Washington, DC: Urban Institute Another 230,000 units (turnkey housing) were produced by private interests and sold to a public agency, usually a local housing authority. The second main approach authorised by the 1965 Housing Act was the leasing of privately owned housing, and by the end of 1974, 169,000 units accounted for 14 per cent of the total public stock. A fundamental reason for the failure of the US public-housing initiative was that it did not have widespread political support, being alien to the dominant market philosophy. Even the initial rationale behind the programme was multifaceted: it was seen as a way of simultaneously reducing the high levels of unemployment in the 1930s and assisting the housing industry, as well as a means of eliminating slums and increasing the supply of cheap, decent housing to the poor. Weak federal controls over development costs and housing quality inevitably led to major difficulties. In particular, local housing authorities received no operating subsidy from federal government, their sole source of income being rental payments. As costs rose, minimum and average rents increased steadily. Prior to the 1974 Housing and Urban Development Act, which introduced a degree of Housing problems and housing policy 309 control, rent payments equivalent to half a tenant’s gross income were common. The 1974 Act marked a major switch of emphasis away from conventional public housing towards the use of leased housing. However, signs of the increased privatisation of public housing had emerged during the 1960s with the 1965 Housing Act authorisation of turnkey housing, the sale of public housing to tenants, and the contracting out of management services to private firms. By the mid-1970s construction of public-housing projects had fallen significantly and some of the most deteriorated schemes, such as Pruitt-Igoe in St Louis, had been demolished. The switch of emphasis in social-housing policy from supply-side to demand-side intervention was reflected in Section 8 of the 1974 Housing and Urban Development Act, which introduced a housing allowance programme for lower-income households (Figure 11.1). Rather than making payments direct to needy tenants, the programme consisted of long-term contracts between the Department of Housing and Urban Development and private developers, landlords or public-housing agencies designating specific apartments for Section 8 subsidies. Approved apartments could be existing units or units to be constructed or rehabilitated. Rental payments for tenants were on a par with those for public housing, amounting to between 15 per cent and 25 per cent of their total income. Housing owners received, from Section 8 funds, the difference between the tenant’s contribution and a ‘fair market’ rent based on prevailing private rents in the locality. Section 8 subsidies have been successful in that tenants have a degree of choice of housing unit, and the scheme helped to reduce the overconcentration of low-income tenants in ‘projects’, many of which have been razed. Federal disengagement from public housing was signalled in the belief of the Reagan administration that ‘the genius of the market economy, freed of the distortion forced by government housing policies and regulations that swing erratically from loving to hostile, can provide for housing far better than federal programs’ (President’s Commission on Housing 1982 p. xvii).14 This resulted in an extension of the demand-side approach via Section 8 certificates and the introduction of vouchers worth a certain amount of rental payment that could be ‘spent’ on any available apartment, and the continued decline in public-housing construction (Table 11.3). In 1985, inspired by the example of the UK, the Department of Housing and Urban Development launched a demonstration project to promote the privatisation of public housing. The residualisation of public housing is well advanced in the USA, with the sector catering for a dependent population characterised by low income, unemployment, reliance on public assistance, and a high concentration of the very old, very young and minorities, often occupying a deteriorating physical environment. Continued federal disengagement from low-cost housing provision was indicated by the 1997 Housing Opportunity and Responsibility Act. The goal is to transform publichousing developments into ‘more self-sufficient communities’. This will be achieved by creating lower-density mixed-income communities through a combination of demolition of the worst high-rise apartment blocks, the sale of public housing to tenants and the construction of new low-rise developments to accommodate a wider range of income groups (by easing the current restrictions on public housing to households whose income is less than 80 per cent of the area median income). Displaced public housing tenants will be provided with Section 8 vouchers to obtain accommodation in the private sector. The Urban geography 310 underlying premise, that mixed-income housing will create a sense of community in which working resi TABLE 11.3 US PUBLIC-HOUSING COMPLETIONS, 1939–91 Year Units Year Units 1939 4,960 1966 31, 483 1940 34,308 1967 38,756 1941 61,065 1968 72,638 1942 36,172 1969 78,003 1943 24,296 1970 73,723 1944 3,269 1971 91,593 1945 3,080 1972 58,590 1946 1,925 1973 52,791 1947 466 1974 43,928 1948 1,348 1975 24,514 1949 547 1976 6,862 1950 1,255 1977 6,229 1951 10,246 1978 10,259 1952 58,258 1979 44,019 1953 58,214 1980 15,109 1954 44,293 1981 33,631 1955 20,899 1982 28, 529 1956 11,993 1983 27, 876 1957 10,513 1984 24,092 1958 15,472 1985 19,267 1959 21,939 1986 15,464 1960 16,401 1987 10,415 1961 20,565 1988 9,146 1962 28,682 1989 5,238 1963 27, 327 1990 6,677 1964 24,488 1991 9,174 1965 30,769 Source: E.Howenstine (1993) The new housing shortage: the problem of housing affordability Housing problems and housing policy 311 in the United States, in G.Hallett (ed .) The New Housing Shortage London: Routledge, 8–67 dents will provide role models for the unemployed, is reminiscent of the philosophy behind nineteenthcentury Utopian communities and the early post-Second World War New Towns in the UK15 (see Chapter 9). In 2000 1.3 million units of public housing accounted for 5 per cent of all rental housing and 1 per cent of the total housing stock in the USA. The poor condition of some public-housing areas, identified by the National Commission on Severely Distressed Public Housing in 1992, led in 1999 to enactment of the Hope VI programme that aimed to revitalise distressed public-housing communities (Box 11.2) In Indianapolis revitalisation of the Near Westside neighbourhood involved the replacement of substandard row houses with lower-density duplex units, similar to those in surrounding neighbourhoods. Units were set aside for households who BOX 11.2 The Hope VI programme Home-ownership and Opportunity for People Everywhere was developed as a direct result of the report of the National Commission on Severely Distressed Public Housing. Originally called the Urban Revitalisation Demonstration, Hope VI funds can be used for the following objectives: 1. improving public housing by replacing severely distressed projects, such as high-rise and barrack-style apartments, with town houses or garden-style apartments that blend into surrounding communities; 2. reducing concentrations of poverty by encouraging a mix of income among publichousing residents and encouraging working families to move into housing in revitalised communities; 3. providing support services, such as education and training programmes, child-care services and counselling to assist public-housing residents obtain and retain jobs; 4. establishing and enforcing high standards of personal and community responsibility through explicit lease requirements; 5. forming partnerships, involving public-housing residents, government officials, the private sector, non-profit groups and the community at large in planning and implementing new communities. participated in a family self-sufficiency programme that required them to enter into a formal contract with specified goals and participation requirements, leading towards selfsufficiency and, in many cases, home-ownership. The hope was that through training programmes, access to employment opportunities and a ceiling-rent programme many public housing families would become home-owners. Nationally, by 2003, 45,000 households had been relocated to other public housing, to private market-rate housing, or with Section 8 vouchers, and 8,000 original households had returned to their redeveloped areas. The scale of demolition has raised some concern that, while providing improved Urban geography 312 housing environments, Hope VI has also reduced the amount of affordable housing available to the lowest-income households. HOUSING AFFORDABILITY Housing affordability is measured by the proportion of household income spent on obtaining housing. An expenditure of more than 30 per cent of income in the USA and over 20 per cent in the UK has been taken to indicate an affordability problem.16 Such measures are rather crude, however, since the same proportion of household income has a different impact for those at different ends of the social scale. Neither does it take account of the subjectivity of affordability. Nevertheless, despite definitional problems there is evidence of a growing affordability problem in both rented and owner-occupied housing, particularly in overheated housing markets.17 Since the early 1980s several structural forces in the UK have operated to make housing more expensive in relation to incomes: 1. Increases in unemployment during periods of recession have reduced incomes and effectively debarred households from the owner-occupied sector. For existing mortgagers this has led to problems over mortgage repayment and in some cases substantial negative equity. 2. As a result of global economic restructuring and the changing nature of labour markets in advanced capitalist societies, those with low levels of skill and limited education have the greatest difficulty in entering the job market and securing an income sufficient to meet the market cost of housing. 3. The private rented sector has continued to decline, and rents have increased following the easing of rent control. 4. There is reduced access to social rented housing as a result of government policies such as the ‘right to buy’, and constraints on new publicsector construction. Evidence of the growing unaffordability of housing for some social groups is found in the 400,000 households in mortgage arrears in 1995 and the repossession of 250,000 houses between 1991 and 1994.18 In the USA between 1997 and 1999 while inflation rose by 6.1 per cent house rents increased by 9.9 per cent and house prices by 16 per cent. In 1997 5.4 million very low-income families (those with incomes below 50 per cent of the local MSA median) paid more than half their income for housing or lived in severely inadequate housing, a situation classified by the Department of Housing and Urban Development as ‘worst-case needs’. This represented an increase of 12 per cent in worstcase needs housing since 1991. A growing proportion of these families are working households. In 2002 in no jurisdiction in the USA would a minimum-wage job (at $5.15 per hour) provide enough income for a household to afford the fair market rent (FMR) for a two-bedroom home. Many low-income earners must work two or three jobs to pay their rent. Between 1991 and 1997 the number of housing units affordable to extremely lowincome families (those with incomes below 30 per cent of median MSA income) fell by 5 per cent or 370,000 units. In New York City an extremely low-income household could afford a monthly rent of no more than $471, while the FMR for a two-bedroom unit was $1,031. Housing problems and housing policy 313 Housing affordability is a problem for places as well as for people. Many people cannot find housing near their workplace or find work at a reasonable distance from where they can afford to live. In many areas workers crucial to the local economy, such as teachers and police, cannot afford to live in the communities they serve. The problem is particularly severe in cities with overheated housing markets. These include high-tech hot-spots in places like Boston, Denver and San Francisco where rents increased by more than 20 per cent in the late 1990s, and global cities such as London. The starkest indication of an affordability crisis, however, is represented by the growing incidence of homelessness, which is most visible in the inner areas of large cities, where the sight of people sleeping in doorways and living in cardboard shelters is redolent of conditions in the Third World city (see Chapter 25). HOMELESSNESS Homelessness is an extreme form of social exclusion, caused by a combination of personal and structural factors. Most homeless people leave their home either because parents or friends are no longer willing to accommodate them or because of the breakdown of a relationship. Structural factors underlying homelessness include insufficient construction of affordable housing, gentrification, cutbacks in welfare budgets, stagnating or falling real incomes, and the rise of part-time and insecure employment. Discriminatory practices can also contribute to the problem for some social groups.19 Those most at risk of homelessness include unemployed people, single mothers, disabled people and frail elderly individuals, runaway youths, battered women and children, immigrants and refugees, substance abusers and deinstitutionalised mental patients (see Chapter 17). While many of the homeless are highly visible (e.g. beggars and rough-sleepers), most are not noticeably different from other citizens. The invisible homeless have to cope with the day-to-day strain of living in temporary accommodation, hostels, bed-and-breakfast hotels, or in cramped conditions with friends and relatives. Figure 11.2 illustrates the types of shelter available to the homeless in post-industrial society. Definitions of homelessness vary between countries, making estimates of the number of homeless persons difficult. In the UK the number of those officially homeless rose from 53,100 households in 1978 to 146,000 households (460,000 persons) in 1994. These figures exclude people not accepted by the local authority as homeless or who are not felt to be in ‘priority need’. The latter category excludes most single people and childless couples who approach local authorities for housing assistance. A 1995 report on single homelessness in London enumerated 47,000 persons living in squats, bed-andbreakfast accommodation or sleeping rough, none of whom had been accepted as being in priority need and therefore officially classified as homeless.20 In Britain as a whole, at any time during the 1990s, over 500,000 people lacked a permanent roof over their heads. In Canada more than 26,500 different individuals make use of Toronto’s emergency shelter system each year; and this does not include the Urban geography 314 Figure 11.2 Types of accommodation in a post-industrial society Source: A.R.Veness (1992) Homes and homelessness in the United States Environment and Planning D: Society and Space 10, 445–68 Housing problems and housing policy 315 BOX 11.3 Sleeping rough in Toronto Several weeks ago, I was too late to get a bed in any of the downtown hostels. I made a tour of the all-night coffee shops. Walking around, I noticed that the bus shelters I passed on Queen St were occupied, as were many doorways. Around 3.00 a.m. I ran out of coffee money, so I headed for a quiet park to lie down. Of course, it started to rain. Well, in the course of the next hour, I scrambled alleyways, fire escapes, across warehouse roofs—and without exception, every little hiding hole was filled with some unfortunate person like myself. Finally, I saw some piles of cardboard by some tractortrailers. I pulled the first pile aside. Someone was sleeping underneath. The same thing with the second pile. And the third had two people under it. All this within 50ft (15m) of trendy Queen West. Source: G.Daly (1996) Homeless London: Routledge number who sleep rough (Box 11.3). In the USA the national total of homeless people in 1995 was estimated at between 500,000 and 600,000.21 In New York half the 80,000 homeless live on the street and the remainder in public or private emergency sheltered accommodation. One in five are parents and children, one-third young adults aged between 16 and 21 years, and the remainder single people, of whom 80 per cent are men. As many as 90 per cent of the residents of sheltered accommodation are from ethnic minority groups even though such groups constitute only 40 per cent of the city population.22 In Los Angeles, where the city council has employed barrel-shaped ‘bumproof benches and sprinkler systems to discourage people from sleeping in parks, the homeless established a series of encampments in the heart of the city (Box 11.4). STRATEGIES TO IMPROVE HOUSING QUALITY The process of neighbourhood decline examined in Chapter 10 may culminate in the creation of slum conditions. Apart from relocating people to new living environments outside the city, the three main approaches to the problem of slums are via the filtering mechanism, by clearance and through rehabilitation. FILTERING Filtering requires no direct government involvement. The raising of housing standards is entrusted to the interplay of market forces whereby the provision of improved accommodation for the lowerincome members of the community is effected through construction of new properties for upperand middle-income groups. In principle a chain of sales is initiated by the insertion of new housing at the top end of the market. Properties vacated by those moving to the new housing filter down the social scale and Urban geography 316 enable individual households to filter up the housing scale, leaving the worst housing vacant and ready for demolition. Empirical research has questioned the ability of the filtering mechanism to work towards the elimination of slums. A major weakness is that the filtering process at best works slowly and many of the vacancy chains are broken before the poor benefit to a significant degree. Filtering also assumes that poor people are mobile physically and economically, whereas unemployed and underemployed people are often unable to move out of the slum dwellings because they cannot afford to devote more of their income to rent. Continued migration of lower-income and, especially in the USA, ethnic minority groups into the urban housing system also increases demand for the relatively small number of units vacated and can thus maintain price and rent levels beyond the means of many households. BOX 11.4 Homeless in Hollywood Los Angeles skid row covers a seventy-block area east of the central business district (CBD), and by the mid-1980s contained the largest population of homeless people in the USA. In the heart of a major city of advanced capitalist society around 1,000 street dwellers survived in over a dozen encampments. Throughout the CBD other homeless people squatted on pockets of disused land, including that under the parking ramps of the Bonaventure Hotel, a symbol of conspicuous consumption in the city. The homeless of skid row are engaged in a struggle for lebensraum in the face of attempts by the civic authorities to evict them (e.g. as part of the clean-up of downtown for the 1984 Olympics and in 1987 prior to the Pope’s visit to Los Angeles). The conflict over the use of space focused attention on the social status of the homeless. While selfhelp agencies talked of a sense of community in the camps, the police referred to them as breeding grounds of crime and the narcotics trade. At one end of the spectrum was Love Camp, a cooperatively run encampment on Towne Avenue that pooled resources to rent portable toilets, shared chores such as cooking, cleaning and firewood search, and ran a newsletter outlining to other homeless groups how to establish a ‘homeless block association’. At the other end, across from Jack’s Market, was a pool of ‘failed robber barons’ who extorted money from anyone trying to pass by. The homeless sought to bring their situation into the political spotlight by, for example, establishing a tent city on state land across from City Hall in the run-up to Christmas 1984. They also confront the new, gentrified users of their turf. On Fourth and Main, where office and government buildings encroach on skid row, clusters of homeless people engage with professionals and office workers on a daily basis. Fashion commentary has become a running part of street repartee, as a way of acknowledging, challenging and reversing the stigma of homelessness: ‘Lady, your suit went out of style five years ago. I know because I designed it.’ Such strategies invert the relationship between commentator and subject, albeit temporarily, and challenge the meaning of the urban space they both occupy in different ways. Source: adapted from S.Ruddick (1996) Young and Homeless in Hollywood London: Routledge Housing problems and housing policy 317 A final criticism of filtering is that, unmitigated, the process will lead to the permanent concentration of poorer households in the worst housing, and such distributional consequences are unacceptable in many countries. The concept of filtering has played an important part in the formulation of housing policy in the USA. Until the early 1930s Britain also relied almost entirely on the filtering process for improving the housing conditions of the working classes. As we have seen, not until the 1930 Housing Act (the Greenwood Act) did the government officially acknowledge that the housing problems of the lower-income groups were unlikely to be resolved through private house-building alone. CLEARANCE In 1955 the UK government estimated a total of 850,000 slum dwellings and proposed an annual clearance rate of 75,000. Unofficial estimates that included the continuing obsolescence of Victorian housing stock suggested a replacement rate of 200,000 per year as a more accurate requirement.23 By 1979 1.5 million dwellings had been demolished or closed and 3.79 million people relocated. This was a remarkable quantitative achievement, and there is no doubt that the slum clearance and redevelopment programmes provided a superior housing and residential environment for most families. However, the scale of the programme, the timing of the process, the effect on existing communities and the stress imposed on residents have all been criticised. Slum clearance can be a lengthy process and the delay and uncertainty that often surround a programme can cast a shadow of planning blight over a neighbourhood and exacerbate the disruptive effect. In Britain it generally took two years for a compulsory purchase order (CPO) to be confirmed and another two years for most of the residents to be rehoused. In the period prior to confirmation of the order, repair and maintenance work practically ceased, and by the time residents were eventually rehoused ‘their homes and the state of the surrounding streets are naturally far worse than when the public-health inspector represented the homes as unfit two years or more previously’ (Gee 1974 p. 6).24 Delays caused by national cutbacks in house-building or failure at the local level to tie demolition to the completion of new dwellings intensify the strain on residents. In Liverpool, for example, during the late 1960s and early 1970s the rate of clearance outpaced the capacity to rebuild in demolished areas. The resultant expansion of vacant land was particularly marked where sites were set aside for nonresidential purposes such as public open space, community centres or schools for which finance was in short supply, or for private industrial or commercial development for which demand was low. Often the policy decision to clear an area would be made five to ten years before a CPO was issued, and although most authorities kept detailed clearance programmes confidential, the probability of an area being affected was advertised by the reduction of public-service investment and by the council buying property and leaving it unoccupied. Empty houses attract the attention of anti-social activities, including those of squatters, fly-tippers, vandals and vermin. Such physical conditions have a demoralising effect on the remaining residents of an area. The extended nature of the redevelopment process would also result in disruption of friendships and neighbouring patterns, leaving individuals isolated. Feelings of insecurity Urban geography 318 arising from the physical and social break-up of a community, the loss of friends and relatives, and the disappearance of familiar features in the environment were compounded by the uncertainty over the future caused by a lack of communication between redevelopment authorities and residents. The majority of families relocated from slums moved into council housing, most of which was built on the edge of cities. These overspill estates were remote from the city centre, poorly served by public transport and, initially at least, lacking in amenities. During the 1960s the need to increase the number of houses to accommodate those displaced by slum clearance and comprehensive redevelopment led to the introduction of high-rise building. Multi-storey flats could be erected speedily at high densities on innercity slum clearance sites and at a cost that, until 1968, was underwritten by government subsidy. Many tenants welcomed the modern living conditions in the tower blocks, but over time major disadvantages appeared in the form of lift breakdown, poor maintenance of communal areas, lack of facilities for young children and isolation of the elderly. In addition, the land economies proved to be less and the construction costs greater than anticipated. A reduction in the Exchequer’s multistorey subsidy meant that by the mid1970s most local authorities had retreated from high-rise building.25 The clearance and redevelopment strategy was also criticised for disrupting workingclass communities. Moreover, there was increasing opposition from the former tenants of condemned properties, who often had to pay more for their council house than they had been paying in tenancies protected by rent controls. Also, residents of clearance areas were often dissatisfied with what they saw as a lack of choice in their rehousing by the local authority, with some municipalities imposing a limit on the number of housing offers they were prepared to make in order to maintain the speed of the clearance programme. Owner-occupiers in designated slum clearance areas also protested about having to accept what many saw as a reduction in housing status, as compensation provision was rarely adequate to enable households to purchase alternative housing. The slum clearance and comprehensive redevelopment era in the UK ended in the mid-1970s in the wake of government concern over costs, growing public pressure for greater participation in planning decisions and the realisation that with 1.8 million unfit dwellings, 2.3 million lacking one or more basic amenities (such as a sink, washbasin or inside w.c.) and 3.7 million in a state of disrepair, clearance and rebuilding alone were not sufficient to address the problem of inadequate housing within a reasonable time scale. Housing problems and housing policy 319 Plate 11.2 The thirty-one-storey Red Road tower blocks erected in Glasgow in the 1960s were designed to house 4,000 people at an average density of 180 persons per acre (440 per hectare). The newly built ‘tenements’ in the foreground reflect the more recent response to meeting public-housing needs through the activities of housing associations URBAN RENEWAL IN THE USA The adverse consequences of the urban renewal process have also been revealed in US cities. The principal complaints have centred on the quality of the new accommodation and the increased costs of replacement housing. Underlying the first issue is the fact that unlike those in the UK, US local authorities are not obliged to provide alternative housing for those displaced by renewal. In a survey of clearance programmes in forty-one cities Anderson (1964) found that 60 per cent of displaced households relocated in other slums, and a general conclusion of US clearance activity is that most households move to neighbourhoods similar to those from which they have been moved, usually on the fringe of the clearance area and often scheduled for clearance in the near future.26 As Hartman (1964 p. 278) concluded, for many households relocation means ‘no more than keeping one step ahead of the bulldozer’.27 In many US urban neighbourhoods racial tipping occurred as displaced black families moved en masse into formerly white communities. Urban geography 320 This aggravated racial tensions and conflicts, and contributed to the racial unrest in US cities. The cost of urban renewal for displaced residents is partly explained by the strength of opposition to the concept of public housing, which ensured that when urban renewal was introduced under the 1954 Housing Act it was defined broadly to embrace redevelopment for non-residential uses, for which 10 per cent (later raised to 35 per cent) of project grants could be used. With the administrative latitude allowed, the proportion of commercial urban renewal could be pushed up to two-thirds. In effect, public funds ‘subsidised the purchase of prime land by private entrepreneurs’.28 Understandably, private developers co-operated enthusiastically with the urban renewal programme. By the end of 1964 970 different redevelopment projects had been approved, accounting for a total of 36,400 acres (14,700 ha) of inner-city land. A major problem for low-income housing provision was the inability of local governments to direct the nature of the redevelopment process. In practice, many slum areas were cleared to allow the expansion of libraries, colleges, hospitals and commercial luxury housing projects. The West End project in Boston, for example, cleared 20 acres (8 ha) of workingclass private housing and replaced it with 2,400 new dwellings designed almost entirely for middle- to upperincome occupancy, with only a 150-unit housing project for elderly people directly aided by federal subsidy.29 Downs (1970) estimated that households displaced by urban renewal suffered an average uncompensated loss amounting to 20–30 per cent of one year’s income.30 Increased housing costs were imposed largely without any direct consideration of the ability or desire of households to absorb these costs. The residents of clearance areas are typically among the poorest and least powerful members of society and, in the USA, most clearance zones are located in ethnic-minority neighbourhoods. In such instances, racial discrimination can compound the difficulty of relocatees’ obtaining decent affordable housing. The evidence of racial discrimination in the allocation of housing in the USA is unequivocal.31 The practice has been furthered by the activities of property developers, estate agents and lending institutions as well as by local government segregation ordinances, restrictive covenants on building occupancy and exclusionary zoning laws. Since the 1968 Civil Rights Act, fair housing legislation has attempted to make discrimination on the grounds of race, colour, religion or national origin illegal in connection with the sale, rental or financing of housing and land offered for residential use. However, the enaction and the enforcement of a law are not the same thing, and housing-market discrimination on racial grounds continues. For the victims this means adverse effects on rents, house prices and access to job opportunities, in addition to the more obvious physical segregation of ethnic groups. In general, AfricanAmericans typically pay between 5 per cent and 20 per cent more than white households for the same housing package, and are less likely to be home-owners and therefore to receive the capital gains ensuing from that tenure status. REHABILITATION In the UK the 1969 Housing Act marked a shift away from clearance and redevelopment towards rehabilitation of existing dwellings as the main tool of urban renewal. The 1969 Act extended the system of housing improvement grants and introduced the General Improvement Area (GIA), under which local authorities could undertake environmental Housing problems and housing policy 321 improvements financed by the Exchequer. Between 1969 and 1979 1.9 million improvement grants were taken up. However, several criticisms could be levelled at the programme: 1. Although the lack of basic amenities was most serious in the pre-1919 housing stock, the proportion of grants taken up by these dwellings was less than for inter-war houses. Similarly, owner-occupied housing accounted for 70 per cent of grants in 1973 even though private rented property was most in need of improvement. 2. In social terms the effects of the improvement grant boom were largely regressive, with the better-housed and better-off sections of the community deriving most benefit.32 In London the majority of improvement grants went to landlords and property developers (the proportion varying between 40 per cent and 70 per cent in different boroughs), and in parts of west London the take-up of grants led directly to gentrification and the displacement of working-class residents. Since properties with vacant possession are worth considerably more to a landlord than their value with sitting tenants (the ratio of 2:5 in West London in 1984 translated into a difference of more than £100,000), the incentive to displace tenants and rehabilitate, to increase value even further, led at times to harassment by neglect until tenants had little option but to vacate. Some property developers made huge profits from the grantaided conversion of large multi-occupied houses into small self-contained flats that were sold or let to higher-income groups attracted to selected areas. 3. The impact of GIAs was limited. While in council-housing areas public ownership ensured systematic improvement on a block contract basis, private-sector GIAs (which accounted for two-thirds of those designated between 1969 and 1973) made a disappointingly small contribution, with only 18 per cent of houses in need of improvement actually improved by 1973. 4. In contrast to the gentrifying areas of London, the provisions of the 1969 Act had little impact in urban areas where there was no middle-class demand for rehabilitation. In many northern industrial towns local authority intervention was minimal and few GIAs were declared. Revision of housing-improvement strategy led to the 1974 Housing Act, which introduced Housing Action Areas (HAAs) to tackle small areas (up to 300 houses) of the worst housing conditions within a five- to seven-year time scale, leaving GIAs to deal with better areas. The Act also envisaged a greater role for housing associations to combat the abuses of the improvement grant system by private landlords. The effect of the legislation is apparent in the fall in grants taken up by private owners from 238,000 in 1973 to 66,000 in 1979, while grant approvals to housing associations increased from 4,000 to 19,000 over the same period. By the early 1980s 500 HAAs covering 56,000 dwellings had been declared. However, this amounted to only 3 per cent of the pre-1919 housing stock. The scale of the remaining problem was revealed in the 1986 English House Condition Survey, which found 909,000 unfit dwellings (4.8 per cent of total stock), 463,000 which lacked one or more of the basic amenities, and 2.4 million in poor repair.33 The 1989 Local Government and Housing Act replaced GIAs and HAAs with Urban Renewal Areas. This recognised that poor housing was but one component of disadvantaged urban environments and set out to tackle local economic regeneration as Urban geography 322 well as substandard housing. Urban renewal areas (URAs) are larger than GIAs or HAAs and must meet the following conditions: 1. They must have a minimum of 300 properties. 2. Over 75 per cent of dwellings must be privately owned. 3. At least 75 per cent of dwellings must be considered unfit or must qualify for improvement grants. 4. At least 30 per cent of households must be in receipt of specified state benefits. By 1995 eighty URAs had been designated, comprising 122,000 dwellings in forty-six local authorities. Initially, grants for repairs to housing were mandatory (demand-led) but since 1996, owing to the flood of applications, they are now awarded at the discretion of the local authority. Poor housing is not confined to the older privately owned or rented stock. From a total publicsector stock of 4,654,000 dwellings in 1985, 3,836,000 (82 per cent) required repair or improvement. The Priority Estates Project attempted to achieve a lasting improvement in conditions on ‘difficult to let’ estates by directly involving tenants in estate management. This initiative provided the basis for a more broadly based programme of estate action launched in 1985 with the aim of: 1. bringing empty properties back into use; 2. reducing levels of crime and ‘incivility’; 3. encouraging tenant participation in decisionmaking; 4. diversifying tenure; 5. attracting private investment. By 1991 350 schemes were in operation and £2,000 million had been diverted into the programme. An assessment of six early estate-action schemes identified problems in involving residents in management as well as the more general difficulty of employing physical planning in pursuit of social and economic goals.34 Reflecting the substantial cost of renovating local authority housing stock in the UK (estimated in 2000 at £19 billion over ten years), local authorities have been encouraged to transfer council stocks to ‘registered social landlords’, a collective term for housing associations and other approved not-for-profit housing organisations.35 In 2002 the city of Glasgow, following a ballot of tenants, transferred its entire stock of 90,000 dwellings to a housing association, thereby relieving the city of a heavy housing debt burden and repairs backlog. ALTERNATIVE HOUSING STRATEGIES An international review of alternative housing strategies reveals a wide range of schemes designed to overcome the shortage of decent and affordable housing. These include: 1. equity sharing, in which the occupier owns part of the equity in the house and rents the rest from a local authority, with the option to buy the remaining equity as they can afford it; 2. sale of public land to private developers on condition that it is used to provide low-cost ‘starter homes’; 3. local authority building for sale, or improvement of older houses for sale; Housing problems and housing policy 323 4. sheltered housing, a common form of accommodation for the elderly in Britain. The Anchor Trust houses 24,000 tenants in 21,000 sheltered units, two-thirds of tenants being dependent on state benefits; 5. self-build housing, whereby the eventual occupiers purchase a plot and organise construction. It accounts for 80 per cent of detached housing in Australia, 60 per cent in Germany and 50 per cent of all new building in France;36 6. homesteading, in which the ‘sweat equity’ of people’s self-help efforts reduces the amount of cash the homesteader is required to put into the scheme. In practice, the skilled nature of the work involved often requires the use of contractors. Although initially seen as a means of providing low-income homes, the high costs of rehabilitation and the skills required work against this. A further problem is that the selection of homesteaders depends on their ability to meet the costs rather than on their housing needs, with the result that, unless restrictions can be applied, gentrification may take place, in contradiction of the original spirit of the idea. The notion of housing improvement through a combination of self-help, private investment and limited government intervention is particularly strong in the USA. More than 4,000 communitybased non-profit housing organisations (CBHOs) have grown up to undertake housing projects for low- and moderate-income residents considered too small or too great a risk by financial institutions and other for-profit developers. The capital is supplied by special housing trusts established by state and local governments and by foundations such as the Social Initiatives Support Corporation and the Enterprise Foundation.37 In addition, the Neighborhood Reinvestment Corporation, established in 1978, organises a modest Neighborhood Housing Services (NHS) programme in 137 cities, providing loans and advice to CBHOs undertaking housing rehabilitation. The criteria for selecting a neighbourhood for an NHS programme are: 1. Housing must be basically sound but showing warning signs of lack of maintenance and deterioration. 2. Mortgages and home loans must be difficult to obtain in the area. 3. At least 50 per cent of the dwellings must be owner-occupied. 4. The area must have distinct boundaries and be large enough to stimulate the imagination of potential participants but small enough to allow for early visible success. On average, NHS areas have 4,900 dwellings, 2,300 buildings and 12,500 people. 5. The median income of the residents must average 80 per cent of that of the city. We can note the similarity between these criteria and those used to denote URAs in the UK. The involvement of the private sector is central to the NHS approach. The NHS programme would not be applied to areas that had declined too far to ensure the participation of private investment. The selection of areas with a good chance of success is considered necessary to maintain the credibility of the programme. An attempt to transfer the NHS idea to the UK in the 1980s failed, owing to the designation of areas of very poor housing and the lack of a significant commercial component in redevelopment.38 Non-profit CBHOs have enjoyed success in a number of US cities including, Portland OR, where the Burnside Community Council used foundation grants to gain control of 40 per cent of the downtown single-room occupancy units (SROs) which were rehabilitated and re-let at reduced rents. For CBHOs to have national Urban geography 324 significance, further resources and central leadership are required. One option recommended by the National Housing Task Force (1988) is for the federal government to set up a national corporation to promote CBHOs with the emphasis on encouraging benevolent lending (by industry, corporations, Churches and foundations) to finance lowincome housing.39 As our discussion of housing policy and problems indicates, the goal of providing a decent home in a reasonable living environment at an affordable price for all citizens is an ongoing challenge that can be addressed only in part by ‘housing policy’. The housing problem is but one element in the complex of social and economic difficulties that confront disadvantaged people and places in urban society. FURTHER READING BOOKS M.Anderson (1964) The Federal Bulldozer Cambridge, MA: MIT Press G.Daly (1996) Homeless London: Routledge S.Harriot and L.Matthews (1998) Social Housing: An Introduction London: Longman J.Leavitt and J.Saegert (1989) From Abandonment to Hope: Community Households in Harlem New York: Columbia University Press S.Lowe (2004) Housing Policy Analysis Basingstoke: Palgrave Macmillan G.McCrone and M.Stephens (1995) Housing Policy in Britain and Europe London: UCL Press E.Meehan (1977) A Decent Home and Environment New York: Ballinger A.Ravetz (2001) Council Housing and Culture London: Routledge W.van Vliet (1990) International Handbook of Housing Policies and Practices New York: Greenwood D.Varady, W.Presier and F.Russell (1998) New Directions in Urban Public Housing New Brunswick, NY: CUPR Press JOURNAL ARTICLES D.Cross (1992) Affordable housing in London Progress in Planning 33, 1–92 R.Farley, C.Steeh, M.Kryson, T.Jackson and K.Reeves (1994) Stereotypes and segregation: neighborhoods in the Detroit area American Journal of Sociology 100(3), 750–80 C.Hartman (1964) The housing of relocated families Journal of the American Institute of Planners 30, 266–86 D.MacLennan and A.More (2001) Changing social housing in Great Britain European Journal of Housing Policy 1(1), 105–34 J.McDonnell (1997) The role of race in the likelihood of city participation in the United States public and Section 8 existing housing programmes Housing Studies 12(2), 231–45 J.Smith (1999) Youth homelessness in the UK: a European perspective Habitat International 23(1), 63–77 L.Takahashi (1996) A decade of understanding homelessness in the USA Progress in Human Geography 20(3), 291–310 KEY CONCEPTS Housing problems and housing policy 325 ■ rent control ■ owner-occupation ■ council housing ■ private-rented sector ■ negative equity ■ slum clearance ■ housing rehabilitation ■ filtering ■ housing associations ■ social housing ■ right to buy ■ residualisation ■ difficuIt-to-let estates ■ turnkey housing ■ fair rent ■ housing affordability ■ homelessness ■ planning blight ■ urban renewal ■ discrimination ■ community-based housing associations STUDY QUESTIONS 1. Identify and explain the changing tenure distribution of housing in the UK. 2. Illustrate the role of the public sector in the provision of housing in either the UK or the USA. 3. Examine the arguments for and against the sale of council housing. Consider the effect of this policy on the public sector. 4. Map the main areas of owner-occupied/privately rented and public-sector housing in a city with which you are familiar. Comment on the revealed distributions. 5. Critically examine the major strategies employed by governments to improve the quality of the national housing stock. PROJECT The desirability of public-sector housing varies across a city. Plot the main areas of public housing on a map of your city or a city with which you are familiar. Consult the records of the public-housing authority to identify the differential quality of housing in the public sector. This can be gauged from statistics on the number of transfer requests received from tenants, vacancy levels, levels of rent arrears, and proportion of abandoned housing. Explain why some areas of public-sector housing are more desirable than others. 12 Urban Retailing Preview: shopertainment; spaces and places of consumption; the spatial switching of retail capital; urban retail forms; the changing structure of urban retailing; suburban shopping centres in North America; retail decentralisation and the central business district in the USA; concentration versus decentralisation in Britain; town-cent re shopping schemes; retail decentralisation in Britain; the impact of regional shopping centres; disadvantaged consumers INTRODUCTION Retailing land uses constitute a significant part of the urban environment in all developed economies. In the UK there are almost 350,000 retail outlets plus 200,000 outlets devoted to services (such as hairdressers and dry-cleaners) usually found in shopping centres. The amount of floor space devoted to retail and service activities in Britain is over 800 million ft2 (74 million m2) and the retail sector employs over 2 million workers, with retail sales contributing 25 per cent of Britain’s gross domestic product.1 In the USA, where retailing and wholesaling activity employs 20 per cent of the national workforce, retailing accounts for over $1.5 trillion in sales annually in over 1.5 million establishments,2 while in the Canadian economy retail sales absorb one-third of disposable income.3 The geography of urban retailing has been studied from two main perspectives. From a cultural perspective retailing is viewed as a particular form of consumption, while from an economic perspective attention is focused on the physical construction of retailing environments. In this chapter we employ both cultural and economic perspectives to illuminate the geography of retailing in the contemporary city. In the first part of the discussion we explain the emergence of ‘shopertainment’ as a mode of consumption, and identify different spaces of consumption. The focus then turns to the relationship between retailing and urban form. We examine the effects of spatial switching of retail capital, and identify the different forms of urban retail areas and the changing structure of the urban retailing system. Attention is then devoted to the growth and nature of suburban and outof-town shopping centres and the impact of this decentralisation dynamic on central shopping areas in British and North American cities. We also consider the role and effectiveness of UK public planning interventions designed to protect the traditional position of the town centre in the urban retailing hierarchy. Urban retailing 327 SHOPERTAINMENT For most people shopping is primarily a means to satisfy basic needs. In addition, for many in contemporary urban societies, shopping is, at least in part, a social activity (often referred to as ‘retail therapy’). Retailing along with dining, entertainment, and education and culture, is one of the principal consumption activities in the post-industrial/ postmodern metropolis. For Hannigan (1998 p. 89) the overlap of these consumer activity systems has given rise to ‘three new hybrids which in the lexicon of the retail industry are known as shopertainment, eatertainment and edutainment’.4 The modern concept of linking shopping and entertainment experiences derives from the post-war suburban malls that have increasingly incorporated leisure facilities. The Plate 12.1 Reconstructing reality: the Rialto bridge Venice (upper) and Las Vegas NV (lower) West Edmonton mall is a world of shopping, entertainment and social space ‘where Spanish galleons sail up Main Street past Marks & Spencer to put in at New Orleans’.5 Retail and entertainment also merge in festival market-places where shopping and dining experiences constitute the entertainment in a visual environment that projects an aura of historic preservation. Such novel consumption spaces capture the customers’ imagination Urban geography 328 and potential spending-power by permitting them to engage in simulated forms of nonshopping entertainment. In Seattle the retailer Recreational Equipment contains a 65 fthigh free-standing artificial rock for climbing, a glass-enclosed wet stall for testing rain gear, a vented area for testing camp stoves and an outdoor trail for mountain-biking. In Niketown, New York, customers can take part in a range of athletic pursuits as well as watch giant video screens of sports events. As Crawford (1992 p. 16) observed, ‘shopping has become intensely entertaining and this in turn encourages more shopping’.6 This convergence is described as shopertainment. SPACES AND PLACES OF CONSUMPTION Retail consumption takes place in a variety of settings, including the home, informal markets, stores, shopping streets and malls. THE MALL Shopping malls have been a feature of US retail geography since the world’s first fully covered mall opened in Southdale MN in 1956. Promoted as ‘tomorrow’s main street today’ and as ‘a whole new shopping world in itself the centre provided a prototype for most subsequent mall developments. During the peak period of the ‘mailing of America’, from 1960 to 1980, almost 30,000 malls were constructed. This process culminated in the development of super-regional mega-malls in West Edmonton and Bloomington MN (see below). The proliferation of shopping malls has also focused attention on the people who frequent them. As well as those who come to shop malls have become a ‘hang-out’ for adolescent ‘mall rats’, a ‘third place’ beyond the home and work/school environments where they can congregate, commune and ‘see and be seen’. Many malls open early in the morning to cater for ‘mall walkers’ who like to practise their keep-fit activities in the safety of the centres.7 The mega-malls of North America and Britain’s regional shopping centres market themselves as ‘open to all’ and provide a range of activities from tea dances to teenage school-break activities in order to support this image. As discussed in Chapter 7, however, public use of mall space is regulated by surveillance systems designed to exclude ‘undesirables’. The ability of the mall concept to adapt to changing market conditions since the late 1980s is evident in the development of new landscapes of consumption, such as specialty centres, downtown mega-structures and festival market-places. The specialty centre is an ‘anchorless’ collection of upmarket shops with a particular retail and architectural theme. It is prone to ‘quaintification’ with typical designs in North America including Mediterranean villages (Atrium Court, Newport Beach CA), mining camps (Jack London Village, Oakland CA) and Spanish-American haciendas (the Pruneyard, San José CA). The downtown mega-structure, on the other hand, is a self-contained complex that includes retail functions, hotels, offices, restaurants, entertainment, health centres and luxury apartments (e.g. Water Tower Plaza, Chicago IL). These worlds ensure that the needs of affluent residents, office workers, conference delegates and tourists can be met entirely within a single enclosed space. The festival market-place combines shopping and entertainment with an idealised version of historical urban community and the street Urban retailing 329 market. The market-place is typically decorated with antique signage and props upon which the modern consumer can act out some history. Street entertainers, barrow vendors and costume staff often support the image. As well as examples in North America (e.g. South Street Seaport, New York City; Faneuil Hall Marketplace, Boston MA; Fisherman’s Wharf, San Francisco CA), specialty centres and festival market-places have become familiar spaces of consumption in UK cities (e.g. Covent Garden, London; Glasgow’s Princes Square and the Corn Exchange in Leeds). The development of shopping malls in the form of streets acknowledges the traditional role of the street as a locus of retail consumption. Universal Studios’ City Walk in Los Angeles and the Fremont Street Experience in Las Vegas both seek to create the diversity of the street in the form of a mall. City Walk, part of MCA’s ‘Entertainment City’, aims to capture the ‘real’ feel of a Los Angeles street, with boutiqued façades borrowed from Melrose Avenue, 3-D billboards copied from Sunset Strip and a faux Venice Beach, complete with sand and artificially induced waves. The sanitised version of street life is promoted as a safe alternative to traditional Los Angeles streets, especially for family shopping. THE STREET The street was the original retail space in cities and remains an important social space. The shopping street can be a destination as well as a thoroughfare. Some streets enjoy a reputation as specialised shopping venues, and retailers are willing to pay high rents to have a presence in such prestigious locations. Bond Street in Mayfair has long been one of the pre-eminent locations for London’s luxury shops but in the mid-1990s this dominance was given a renewed emphasis as large numbers of foreign fashion-designer stores were attracted there. Indigenous UK designers with less financial muscle were often relegated to locations in less prominent streets in Covent Garden and South Kensington. The explanation for this trend involves both economic and cultural factors. Physically there is nothing peculiar about Bond Street; it is not a main street, nor is it pedestrianised. It has no particular attractions along its length and is not close to the main tourist attractions in London. Yet, as in the case of Madison Avenue, New York, Bond Street has become a ‘branded street’ for international fashion houses who seek to exploit the cultural cachet and market potential of the street. Less exclusive street-based spaces of consumption are represented by traditional street markets (and their modern equivalent of the car-boot sale8). Such spaces incorporate both the retailing and the socio-cultural components of the consumption experience. This is particularly evident in ethnic-based spaces of consumption.9 HOME Home shopping began in the USA in the late nineteenth century in the form of mail-order catalogues. (In the 1930s the Sears Roebuck catalogue was even merchandising off-theshelf houses.) By the early 1990s, however, the market share of traditional ‘agency’ catalogues had declined to a point where several went out of business. New entrants to this branch of the non-store retail sector (such as Land’s End) have focused on direct marketing aimed at specific lifestyle groups; often those with limited time for traditional Urban geography 330 modes of consumption, such as professionals or dual-income families. Other forms of home shopping, such as television and Internet sales, are also growing in importance but e-commerce still accounts for a relatively small proportion of total retail sales. In 1998 ecommerce sales in the USA were estimated at $4.6 billion or 0.2 per cent of total retail sales in that year. This was expected to rise to $76 billion or 3 per cent of total retail sales by 2003. In the European Union e-commerce sales were anticipated to capture $48 billion or 2 per cent of total retail sales by 2002.10 The convenience of non-store shopping must be set against the loss of sociality offered by the street, store and mall experience. Though likely to gain an increasing market share, non-place shopping is unlikely to replace more conventional landscapes of consumption. THE SPATIAL SWITCHING OF RETAIL CAPITAL As we saw in Chapter 7, ‘creative destruction’ of the built environment by the spatial switching of capital in search of profit is a leitmotiv of the capitalist city. This ‘perpetual struggle in which capital builds a physical landscape appropriate to its own condition at one particular moment in time only to destroy it…at a subsequent point in time’ (Harvey 1981 p. 113)11 is illustrated clearly in the changing retail landscape. As we see below, since the mid-twentieth century the process of recurrent innovation has engineered shifts of retail activities from downtowns to suburbs, devalorising retail environments of the inner city, before returning to revitalise some abandoned sites in order to create new spaces of profit extraction. The same profit dynamic has stimulated central city redevelopments, such as Faneuil Hall/Quincy Market in Boston MA (see below), and the reinvention of retail centres in affluent innercity suburbs as ‘urban street’ shopping environments. Streetscapes have been rediscovered by corporate US retailers in towns such as Bethesda MD that have attracted ‘anti-chain-store chain stores’. These boutiques cater for people who consider themselves too refined and individualistic to shop at the mall or the mass-market big-box stores. Paradoxically, while the boutiques look and feel independently owned they are not. URBAN RETAIL STRUCTURE Urban geographers have a particular interest in the spatial and hierarchical organisation of retailing within the city. An early attempt to impose order on the complexity of different types of retail outlet and forms of shopping area identified three main kinds of retail area12 (Figure 12.1). These are: 1. Nucleated centres that constitute the main shopping areas in the city. A distinction is drawn between the older, unplanned centres of the inner city, including the metropolitan central business district (CBD), and the newer planned suburban centres. 2. Ribbon developments of unplanned retail units that evolve along highway corridors and which are a conspicuous feature of most US cities. These accommodate large space users and affiliated service activities that require good Urban retailing 331 Figure 12.1 Berry’s classification of urban retail locations Source: B.Berry, J.Simmons and R.Tennant (1963) Commercial Structure and Commercial Blight Research Paper 85 Department of Geography, University of Chicago accessibility for passing trade. Typical activities include gasoline stations, automobile dealers, fast-food restaurants and home-supply stores. 3. Specialised function areas comprising retail and service functions that cluster because of mutual attraction or interdependence. These include the theatres, restaurants and bars in an entertainment zone; the automobile row concentration of garages and autodealers which benefits from comparison shopping; and the planned or unplanned medical district. These categories are not mutually exclusive, with many retail functions being found in more than one location, but the classification offers a general description of the major forms of retail location in the city. This early model also provided a basis for subsequent taxonomies, including that proposed by Jones (1991), which differentiates retail areas according to their morphology, location, functional composition, market size and the type of market served13 (Figure 12.2). As Figure 12.2 illustrates, inner-city retailing has been dominated historically by the unplanned shopping area, comprising the CBD, speciality product areas, and retail clusters at major route intersections. Planned inner-city shopping areas are a more recent phenomenon but have become a common feature in urban North America since the mid1970s. In the downtown core two types of planned centres have emerged: the central city fashion mall, which has often been the focus of a major urban renewal project; and the ancillary retail complex, which has become the normal ground-floor or underground use in major office, hotel or condominium developments. Urban geography 332 Figure 12.2 A typology of the contemporary urban retail system Source: T.Bunting and P.Filion (1991) Canadian Cities in Transition Toronto: Oxford University Press Since the 1970s four other forms of planned centre have emerged in the inner city: theme malls, infill shopping centres, retail mall developments at major intersections and superstores or hypermarkets. Theme malls are a recent phenomenon and are normally tourist-oriented, occupy waterfront locations and promote a distinctive specialityproduct theme and atmosphere. The infill centre is a typical suburban shopping centre that has been developed in the inner city when major retail development companies were confronted with a saturated suburban market and new growth opportunities were restricted to neglected inner-city areas. The development of planned centres at major inner-city intersections represents a modernisation of traditional retailing at these shopping nodes. The superstore represents the return of the major supermarket chains to previously abandoned innercity locations. Typically these single retail units occupy a minimum of 50,000 ft2 (4,600 m2), offer discount prices and a wide product assortment, and rely on extensive trade areas. As Figure 12.2 indicates, the classification of planned suburban shopping centres is essentially hierarchical, ranging from the neighbourhood shopping plaza to the regional and super-regional shopping centre. The characteristics of each reflect different numbers of stores, establishment types, total area, selling area, number of parking spaces, customer volumes, trade area size, rental rates, and sales-per-unit area values (Table 12.1). More recent variants of the suburban shopping centre include the megamall/recreation complex, and shopping complexes directed to distinct market segments such as the family or young urban professionals. Urban retailing 333 Plate 12.2 This factory-outlet fashion mall outside Las Vegas NV contains over 100 stores, and is linked to a casino and hotel complex to provide a combined retail-recreational experience TABLE 12.1 HIERARCHY OF PLANNED SHOPPING CENTRES Measure Neighbourhood Community Regional Superregional Market size (000 population) 20–40 40–200 100–500 500+ Average sales ($ million)a 2–5 4–30 10–150 50+ Floor area (gross 50–100 leasable area) (000 ft2) 100–300 300–500 750+ Average sales/ft2a 100 125 200 200 Anchor tenant Supermarket Junior dept store Major dept store Two or more dept stores Type of business Convenience, service Shopping, convenience Shopping goods Shopping goods Urban geography 334 Average no. of stores 10–25 20–40 40–100 100+ Average no. of business types 10–20 15–30 20–40 40+ Site area (acres) 5–10 10–30 30–100 100+ Location Often within a Intersection of Expressway suburban development two arterial roads and arterial Trade area radius Up to 1.5 miles 1–3 miles 3–7 miles Intersection of expressways 10+ miles Source: K.Jones and J.Simmons (1990) The Retail Environment London: Routledge Note: a1982 dollars The retail strip form can be differentiated primarily according to location. In the innercity main streets, strip intersections, neighbourhood shopping streets and corner-store clusters have served essentially the same functions since the early 1900s. In addition, ethnic minority and gentrified community shopping strips have developed specialised retail functions that may serve a metropolitan-wide market, particularly for fashion goods and restaurants. In the suburbs the strip forms of retailing reflect the dominance of the automobile. These include the unplanned 1950s suburban strip shopping malls that were the first stage of retail expansion into the suburbs.14 Although other classifications of urban retail forms are available,15 including those specifically directed to the structure of retailing in UK cities,16 the classification indicated in Figure 12.2 provides a useful general conceptual framework for understanding the complexity of the urban retail environment. It also underlines the dynamic nature of the urban retail structure as new products, store types, market segments and retail locations emerge as a result of the constant interplay among retailers, consumers, developers and, in the UK, planning authorities. THE CHANGING STRUCTURE OF URBAN RETAILING The urban retail pattern is affected by two main forces of change. Changes of the first type are those that have resulted from the effects of market forces and which may be regarded as a ‘natural’ process of evolution. For most of the post-Second World War period this has been characterised by a general process of decentralisation reflected in the relative decline of retail facilities in the inner city and expansion of trade in the suburbs. Second, sometimes reinforcing these trends and sometimes working against them, there are more specific changes effected by planning. These are most evident in the form of inner-city redevelopment schemes and in the layout of new types of outlying shopping centres.17 These changes in urban retail structure have been driven by a range of economic and social forces. The main factors on the demand side include: Urban retailing 335 1. Changes in residential location. The single most important factor underlying change in the urban retail structure in the post-war era has been the widespread suburbanisation of population. Despite growth in some inner-city areas related to gentrification, the overall metropolitan population trend has been one of decentralisation (see Chapter 4). In general it has been the younger, richer and more mobile elements of society that have migrated to the suburbs, creating new, large sources of demand in areas where few shopping facilities existed previously. An older, poorer and less mobile population has been left behind in the inner city, where its lower level of purchasing power has been insufficient to support the surfeit of shopping facilities that remain. These changes have been most pronounced in the US city, where the earlier pattern of retailing has been altered by the growth of new suburban and outlying centres and the decline of central shopping areas. In the UK, by contrast, strict planning controls have sought to restrain the process of retail decentralisation. 2. Changing consumer attitudes and expectations. These may be characterised as a general demand for more convenience and comfort in shopping which newer rather than older centres can satisfy more readily by providing one-stop shopping in environmentally friendly malls for car-borne consumers (Box 12.1). 3. Growth in female employment. The fact that women account for a higher proportion of the workforce than they did a quarter of a century ago has had two major implications for consumer behaviour: first, through the increased purchasing power created by additions to household incomes; and second, through the time constraints imposed on shopping, particularly for women engaged in full-time employment. One result has been an increase in bulk-buying, especially of food, and a reduction in the frequency of shopping trips. 4. Changing levels of purchasing power. For three decades after the Second World War most sections of society experienced an increase in purchasing power. Coupled with the growth in population numbers, the resultant rise in levels of consumption stimulated expansion of retail trade. Subsequent downturns in retail expenditure reflecting the condition of the national economy have had a differential impact on retailing. Larger companies have maintained their economic health by increasing their BOX 12.1 The Minneapolis skywalk system The Minneapolis skywalk system is an ingenious device to provide a climatecontrolled shoppingmall environment in the downtown area. For at least half the year in Minnesota it is unpleasant to be outside: cold, wind and snow fog up your glasses, ruin your shoes and require several pounds of cold-weather gear. Covered shopping malls have a great advantage over unplanned centres, unless a way is found to link together stores and customers. The downtown areas in both St Paul and Minneapolis have developed skywalk systems in order to compete. The skywalk (a tunnel in the sky) links together downtown buildings at the second storey level, so that shoppers, office workers and visitors can move easily around the city, spending an entire day and visiting several different shopping arcades without going outside. In St Paul the system is supported by the city which buys a right of way through intermediate buildings In Minneapolis the Urban geography 336 skywalk is constructed and managed by individual developers. In both cases the result is a closely integrated and successful downtown area where people can move about easily, ignoring the barriers of traffic or parking garages. In addition, the Minneapolis skywalk system is connected directly to a number of large-scale parking garages on the edge of the downtown that link to the interstate highway leading to the wealthy western suburbs. By integrating several major downtown malls Minneapolis has created a retail attraction that no outlying shopping centre can match. The visitor may get lost temporarily, but it all adds to the sense of exploring an extended bazaar. Sources: K.Jones and J.Simmons (1990) The Retail Environment London: Routledge; J.Byers (1998) The privatisation of downtown public space Journal of Planning Education and Research 17,189–205 market share at the expense of traditional outlets by offering lower prices and attracting customers from farther afield. As we shall see later, however, although the benefits of cheaper shopping are available to the majority of consumers, there remain large numbers of lowincome households, particularly the elderly on limited pensions, single parents and unemployed people, who are unable to buy in bulk or to travel to superstores, hypermarkets and regional shopping centres. 5. Increased mobility. The growth of car ownership and car-borne shopping has been another major factor in the trend towards less frequent shopping trips, involving bulkbuying. On the supply side the most significant developments in urban retailing have been: 1. Structural change. Most far-reaching has been the expansion of the multiple retailers’ share of turnover, largely at the expense of independent shopkeepers. By 1989 the five largest retail firms in the UK accounted for 19 per cent of all retail sales and 41 per cent of food retailing. Independent food retailers increasingly have had to make use of voluntary buying groups (e.g. Mace or Spar) or cash-and-carry warehouses in order to withstand competition from multiple traders. Declining profit margins on food retailing have also altered operating strategies in retailing. Self-service outlets have mushroomed and the pursuit of economies of scale has promoted the development of larger units. Taken together, these trends have led to a reduction in the total number of retail establishments. In North America the greatest decline in the number of small independent shops occurred in the 1950s in response to the spread of supermarkets and suburban shopping centres. The total number of establishments in the USA fell from 537,000 in 1948 to 319,000 in 1963, while over the same period the number of supermarkets increased from 2,313 to 14,518. Correspondingly, in Britain between 1961 and 1971 the number of shops fell from 542,301 to 504,781, with a further decline to 354, 131 by 1980. 2. New technology. The use of bar-coding and electronic point of sale (EPOS) terminals linked with in-store computers has a number of advantages for the efficiency of the retail operation, including speeding the sale of goods through a checkout (and checking the work speed of operators), providing an itemised customer receipt and maintaining stock control. Because of the capital costs involved, the benefits of new technology are most likely to accrue to large retailers, further extending their competitive advantage. The potential impact of new technology on urban retailing Urban retailing 337 patterns is considerable, although the rate of adoption of innovations such as teleshopping remains unclear. In considering the cumulative effects of these demand- and supply-side forces some important differences between the experience in North America and that in the UK must be acknowledged. In particular, the pressures for retail decentralisation have been much greater and more effective in North America than elsewhere. Against a background of a relatively free market economy and the absence of a strong body of state and local government planning laws, a large number of new suburban shopping developments have taken place. At the same time, in many US cities there has been a rapid deterioration of shopping facilities in the inner city and considerable erosion of the former status of the central area. In the UK, in contrast, much stricter and nationally co-ordinated planning policies (see Chapter 8) have attempted to curtail the negative effects of decentralisation. Instead of there being large-scale suburban developments, emphasis has been placed on the redevelopment of the central area and other inner-city shopping facilities. Since the mid-1990s government has assessed proposals for new retail developments against a ‘sequential test’ whereby first preference is accorded to towncentre sites where suitable sites or buildings for conversion are available, followed by edge-of-centre sites, and only then by out-of-centre sites in locations that are, or can be made, accessible by a choice of means of transport.18 Further, it is government policy that out-of-centre developments should not be of a scale that would undermine the vitality and viability of town centres that would otherwise serve the community well. As a safeguard, the Secretary of State must be notified of retail proposals in excess of 20,000 m2 gross floor space (reduced from 40,000 m2), and of proposals objected to by another local authority from where significant retail spending power would be drawn by the new development. Retail developments in other countries of Western Europe have followed a course between the extremes of the USA and the UK.19 THE GROWTH OF SUBURBAN SHOPPING CENTRES IN NORTH AMERICA The planned shopping centre has been the dominant element in the process of retail change affecting North American cities. The origins of the planned centre can be traced to Roland Park, built in Baltimore in 1907, while the first out-of-town planned centre was the Country Club Plaza built on the outskirts of Kansas City in 1926. However, the main period of growth was in the post-Second World War era. Five development phases, each reflecting a different development philosophy, may be identified: 1. The 1950s were characterised by a ‘consequent’ development strategy in which the shopping mall was built after the housing in an area (i.e. when the market was established and known). Most plazas were small, and developed independently to serve the convenience of a single residential community. At first the major department stores, with large property investments in the CBD, were reluctant to move into the suburbs. In 1954, however, construction of the Northland Center in suburban Detroit by Hudson’s department store company signalled the start of the competitive attack on downtown USA. Urban geography 338 2. The 1960s saw a shift in strategy towards ‘simultaneous’ development, in which the shopping plaza and housing stock were built at the same time, with the planned centre designed to be the focus of the community. The large regional shopping plaza flourished as downtown department stores began to develop suburban branches. During this period, shopping centre development became an ‘industry’ with established policies and common procedures that led to a ‘pattern book’ approach to the production of a series of homogeneous shopping environments (Box 12.2). 3. The 1970s witnessed the emergence of the ‘catalytic’ shopping centre whereby the large mall was seen as a growth pole that would stimulate future residential construction. Typically, a super-regional shopping centre, built on a greenfield site at the intersection of two major expressways, would precede residential development by three to five years. In some cities, such as Detroit, the commercial revitalisation of central areas provided a second major focus for shopping-centre development during this period. By the end of the 1970s the North American shopping centre industry appeared to have reached a point of market saturation, except in the high-growth cities of the sunbelt. Developers pursued a number of alternative growth strategies: ■ A number of first-generation regional centres built in the early 1960s were rejuvenated by enclosing and remixing the outlets. ■ Through a strategy of infilling, a number of smaller cities were targeted for enclosed regional or community malls on the edge of town, which resulted in a consequent decline in their downtown cores. ■ Developers began to compete with unplanned shopping areas by building strip malls along arterial roads and renting space to low-rent retailers and services. 4. The 1980s saw the emergence of two new retail forms. The first was the power centre, an agglomeration of big-box retailers that often includes category killers (e.g. Home Depot), discounters (e.g. Wal-Mart), warehouse clubs (e.g. Costco) and other valueoriented retailers, mostly operating in large industrial-style buildings. The first power centre (the 280 Metro Centre) opened in 1986 in Colma CA between San Francisco and San Mateo. By 1998 there were 313 power centres in the USA with a combined GLA of 266 million ft2 accounting for over 5 per cent of national shopping-centre sales20 (Box 12.3). The second significant development saw the emergence of the shopping mall as entertainment centre or tourist attraction, the most ambitious examples being the 3.8 million ft2 (350,000 m2) West Edmonton mall developed in Alberta in 1985 (Box 12.4) and the 3.5 million ft2 (325,000 m2) Mall of America built in 1992 at Bloomington MN, 8 miles (13 km) south of the city centre of Minneapolis. The latter development provides over 400 shops, including four major department stores on three shopping levels with a 7-acre (3 ha) amusement park at the heart of the mall (Figure 12.3). Super-regional developments of this scale and diversity can challenge the attractions of any downtown area, and reflect the polycentric structure of post-industrial metropolitan America. Urban retailing 339 BOX 12.2 The anatomy of the shopping mall In addition to the fundamental importance of location, the profitability of the planned shopping mall is for dependent to a large extent on the careful selection of tenants and a layout designed to manipulate customer flows and maximise sales. Most large shopping malls incorporate a number of common design principles: 1. The large anchor (generative) tenants are situated at the ends of an internal mall so that customers will visit both ends and in so doing pass by the smaller stores. 2. Access is strictly controlled to minimise the number of customer exit routes at intermediate locations. 3. The mall ‘street’ is often curved or zigzagged in order to extend its length, increase the number of store fronts and reduce the perceived length of walking distance. 4. Clusters of closely related or competitive activities (e.g. shoe stores) may provide subfoci within the mall. Food outlets often form a ‘gourmet court’ with communal tables and seating facilities. Retailers serving different age or income groups are kept apart; stores catering for teenagers, for example, are segregated from up-market retailers. In the largest malls the fashion stores and massmarket retailers may be located on different floors or in different wings. Typical early mall layouts are shown in the diagrams Toronto and Montreal: Source: R.Davies (1980) Marketing Geography Corbridge: Retail Planning Associates Urban geography 340 Figure 12.3 Plan of the Mall of America, Bloomington MN As early as 1982 there were fourteen metropolitan areas in which a total of twenty-five major suburban retail concentrations exceeded the retail sales levels of the CBD, the trend being particularly marked in Atlanta GA with three and Indianapolis IN with six centres. By 1991 there were 1,800 regional shopping centres of over 400,000 ft2 (37,000 m2) in the USA, almost all of them in suburban locations. Figure 12.4 provides an example for one city. During the 1990s many markets became saturated owing to the extensive shopping centre building activity of the 1960s, 1970s and early 1980s (Figure 12.5). In addition, escalating construction costs made new shopping centres expensive to build. Owing to these difficulties, and because many of the earlier shopping centres were outdated, the shopping centre industry concentrated on renovation of existBOX 12.3 Power centres in the USA A power centre in the USA is a type of supercommunity shopping centre that includes: 1. More than 250,000 ft2 of gross lettable area (GLA). (The average power centre has a GLA of 375,000 ft2 and the largest more than 1 million ft2.) 2. At least one super anchor store with at least 100,000ft2 of GLA. 3. At least four smaller anchors with a GLA of 20,000–25,000 ft2 each. 4. Only a small number of smaller shops with a GLA of less than 110,000 ft2. 5. Generally an open-air centre. 6. A trading area similar to a regional shopping centre. 7. A unified shopping-centre management. Urban retailing 341 Most power centres are located in California, followed by Florida, Illinois, New York and Pennsylvania. Figure 12.4 Retail centres in San Antonio TX Source: J.Dawson (1983) Shopping Centre Development Harlow: Longman BOX 12.4 The West Edmonton Mall, Edmonton, Alberta With 600 stores and 3.5 million ft2 (325,000m2) of retail space, WEM is the largest shopping centre in North America. It has almost twice the retail floor area of downtown Edmonton and is climatecontrolled. (Edmonton is Canada’s northernmost metropolitan area.) In size alone, the mall dominates the city’s retail environment, but the developer (Triple Five Corporation) has provided further attractions. Recreational facilities (1.4 million ft2, 130,000m2) attract tourists from all over western Canada. Visitors come by bus and plane from as far away as Saskatchewan and British Columbia to see Fantasyland (the amusement park), the Deep Sea Adventure (aquarium), the National Hockey League scale ice rink (where the Oilers practise on occasion) and the 5-acre (2 ha) indoor Water Park. Many of them stay in the on-site hotel. Needless to say, the mall does a lot of business: the Edmonton Planning Department estimates that it accounts for one-third of the city’s retail sales and generates 15,000 jobs. The mall cost almost $1 billion and pays $11 million per year in municipal taxes Central Urban geography 342 Edmonton, aside from the downtown planned centres, looks devastated. Other major plazas in the city are feeling the pinch as the mall lures away their tenants, and all over Western Canada merchants complain that local shoppers are taking their money out of town to the mall. Sources: D.Herbert and C.Thomas (1997) Cities in Space, City as Place London: Fulton; K.Jones and J.Simmons (1990) The Retail Environment London: Routledge ing centres rather than on new construction. In the early 1990s three times more centres were under renovation than under construction in the USA. The number of shopping centres grew by 7.5 per cent each year in the 1970s, by 5.2 per cent in the 1980 and by only 2.3 per cent in the period 1990 7. The US shopping centre industry continues to grow, adding over 1 per cent of new retail selling space per year. However, in the 1990s retail capital in the USA entered a new flexible phase, reflected in the reduced production of large malls, and the growth of smaller and more diffuse shopping centres designed to attract a wide range of Figure 12.5 New shopping centre construction starts in the USA consumers (see above). The US retail landscape now comprises a variety of formats, ranging from super-regional centres, mega-malls and power centres to street-based stores in ersatz cultural and historical settings. THE IMPACT OF RETAIL DECENTRALISATION ON THE US CBD The adverse impact of retail suburbanisation on the CBD was characterised by Berry et al. (1963) as ‘commercial blight’21 (Box 12.5). The impact of decentralisation on cities of Urban retailing 343 the US Midwest between 1958 and 1963 is depicted in Figure 12.6. This indicates the more general finding that the greatest competitive impact of retail suburbanisation was felt by the larger metropolitan areas in which the central area lost the bulk of its ‘external’ market and became increasingly dependent on an ‘internal’ market comprising the poor, elderly and non-white residents. More recently a study of Charlotte NC found that between 1972 and 1991 the amount of retail space in the central area declined to less than one-sixth of its former size under the impact of three super-regional shopping centres and other suburban retail development.22 Several factors indicate that the decline of downtown retailing is not necessarily an inexorable process, however, since: Figure 12.6 Central-area decline and suburban retail growth in the US Midwest, 1958–63 Source: R.Davies (1976) Marketing Geography Corbridge: Retail Planning Association 1. Many city centres retain significant advantages for the provision of specialist shopping opportunities. 2. The CBD accommodates concentrations of office employment, government functions, entertainment and cultural facilities which bring significant numbers of people into the area. 3. Many city centres have developed an important tourist and convention trade function. 4. Urban renewal has led to the gentrification of formerly declining districts and the creation of retail demand by high-income apartment dwellers. Similarly, the Urban geography 344 deindustrialisation of the inner city has provided development opportunities on waterfronts for festival shopping districts. 2. BOX 12.5 The nature of commercial blight Berry et al. (1963) suggested that four main kinds of blight can affect the older and weaker elements of the retail system: 1. Economic blight, which involves the closure of large numbers of businesses as a result of a reduction in the amount of trade area or purchasing power support. 2. Physical blight, which refers to the structural deterioration of buildings, due primarily to age but also because of an inability or unwillingness of owners to undertake maintenance. 3. Functional blight, which refers mainly to the obsolescence of small businesses due to the impact of mass merchandising techniques and the growth of automobile shopping. 4. Frictional blight, which encompasses a wide range of negative environmental effects created by problems such as traffic congestion, litter accumulation and vandalism of vacant properties. The combined result of these conditions is the formation of the commercial slum. Source: B.Berry, J.Simmons and R.Tennant (1963) Commercial Structure and Commercial Blight (Research Paper 85) Department of Geography, University of Chicago The declining centres of a number of cities in the USA have experienced at least partial revival. A number of mixed office, hotel, convention centre and shopping complexes, exemplified by the Renais sance Center in Detroit MI, have added a distinctive but separate commercial precinct to traditional central city retail areas. Speciality, theme or festival centres have also been developed in a number of city centres, including Quincy Market in Boston MA and Ghirardelli Square on the San Francisco CA waterfront.23 Although the central shopping areas (CSAs) of many cities are reclaiming some of the trade share lost to suburban centres, this does not signal a return to a former pattern but rather indicates a segregation of the CSA into two distinct parts comprising a growth area of new, specialised shops catering for fashion demands and a deteriorating area made up of older remnants of the retail trade. As we saw in Chapter 7, the recommercialisation of run-down central city areas by private-sector investment can also promote the ‘privatisation of public space’, as evident in Los Angeles CA24 and New Orleans LA.25 As the suburbanisation wave moves outwards, the need for redevelopment is often more pressing for the older, outlying shopping centres of the metropolitan area than for the central area itself. Table 12.2 shows the changing land values in six different shopping centres in the Chicago metropolitan area over a forty-year period. It is instructive to consider the position of the outlying centre at Sixty-Third and Halsted Streets. In the 1950s this was the largest retail centre in Chicago outside the Loop. Located 8 miles (13 km) from the downtown area at a point where the elevated transit Urban retailing 345 line connected with a network of bus and trolley (tram) routes and surrounded by bluecollar and middle-class neighbourhoods, the centre contained several major department stores and hundreds of smaller outlets. In 1952 total retail sales were higher than in any of the new suburban plazas. In the 1960s a period of racial transition commenced in which AfricanAmericans replaced whites, household incomes declined and retail activity reduced sharply. The major department store (Sears) moved out in the mid-1970s. Although 45,000 households still live in the trade area of the centre, the average household income is less than $10,000; 40 per cent of households earn less than $7,500 and 45 per cent of the population are under 25 years of age. Households with cars shop at Evergreen Plaza, a successful mall used by African-Americans 6 miles (10 km) farther out. The Sixty-Third and Halsted centre survives as a cluster of 100 stores, many locally owned, surrounded by abandoned properties and vacant lots. The cycle of decline is in marked contrast to TABLE 12.2 CHANGES IN SHOPPINGCENTRE LAND VALUES IN THE CHICAGO METROPOLITAN AREA Year Traditional centred State Street, W. Side Madison— 63rd and Halsted Washington (central area) Street (outlying centre) Church-Orrington Evanston (outlying centre) 1929 22,000 10,000 4,500 1933 14,000 6,500 4,000 ; 1940 10,000 10,000 3,000 1952 17,000 9,000 3,500 1960 17,500 5,500 3,500 1967 22,000 3,000 3,200 New planned centred Old Orchard Evergreen Plaza Ford City 1929 4,000 5,000 3,000 1933 2,250 2,250 1,000 1940 800 1,500 600 1952 2,000 3,000 3,000 1960 50,000 50,000 26,000 1967 152,000 130, 680 109,000 Source: R.Davies (1976) Marketing Geography Corbridge: Retail Planning Associates Notes: aPeak land values per front foot ($). bLand values per acre ($) Urban geography 346 the growth cycle enjoyed by North Michigan Avenue (Box 12.6). CONCENTRATION VERSUS DECENTRALISATION IN BRITAIN There is a strong contrast between post-war changes in the urban retailing pattern of Britain and those described for North America. A fundamental difference is the limited amount of suburban development that has been permitted in Britain and the relative protection afforded to the central area and other inner-city shopping areas. There are several factors that explain the application of strict planning controls over retail suburbanisation in Britain: 1. In general, the central shopping area has always held a more dominant position over the rest of the retail system than its counterpart in North America. 2. There has also been a richer historical legacy of buildings to be preserved, as well as a greater mix of non-retailing land uses to be controlled. 3. More specifically, during the years of rapid post-Second World War suburban development in North America many of the central areas and other inner shopping centres of British towns remained damaged from wartime bombing, necessitating reconstruction and new retail investment. 4. There was also less immediate pressure for retailing to move out of the central area since, as late as the mid-1950s, shopping habits in Britain had still not been affected greatly by the impact of the automobile or mass-selling techniques. From the 1970s onwards, however, increasing pressure for suburban development was expressed in a growing number of applications for planning permission to build hypermarkets, superstores, discount stores and various sizes of shopping centre on predominantly greenfield sites. The ensuing debate over the desirability of concentration versus decentralisation focused on three main issues: 1. Economic considerations. Most planners argue that the competitive impact of new suburban shopping facilities will lead to the same relative decline in the status of the central area and other inner-city retail areas that has been experienced in many US cities. They point to an obligation to prevent this, not least because of the vast amount of public investment in postwar renewal schemes. Against this, many sections of the business community contend that high rents and congestion costs in the traditional retail locations mean that planners are supporting inefficient trading practices that ultimately lead to unnecessary price increases for consumers. BOX 12.6 The polarisation of retail provision in central Chicago For years the Loop in Chicago has been one of the great retail concentrations in the USA. It is not spatially dispersed like downtown Los Angeles, nor mixed in with other land uses like New York. The Loop is based on an intense concentration of access to a large and prosperous market. Enclosed by the elevated transit track, which is itself bounded by the lakefront land uses and the Chicago River it is also served by Urban retailing 347 underground transit and commuter lines. Offices for the financial district and government are concentrated in the west of the Loop, and continue to expand. The retail activity is located to the east. In the 1950s, six great department stores were located along State Street, on one block after another. In 1948 the Loop generated $6 out of every $1,000 retail sales in the country, one-eighth of all sales in Chicagoland. However, forty years later both Chicago and the Loop have declined in significance: the $6 has dropped to $1, and the 12 per cent to 3 per cent. Of the six great department stores, only Marshall Field and Carson Pirie Scott remain. What happened? To the usual list of suburban growth, freeway construction, shopping plazas and inner-city decline we can add the dramatic reduction of purchasing power in the immediate trade area south and west of downtown, as shown in the text. At the same time, gentrification and expensive condominiums flourish on the north side of the city. Upscale retailers have detached themselves from the mass markets of the Loop and created a flashy new retailing environment on North Michigan Avenue. It is only a mile or so away but the atmosphere is worlds apart: different customers (different access patterns), different stores (Neiman-Marcus, Saks) and dramatic success. Michigan Avenue has become the greatest fashion retail location between Fifth Avenue and Rodeo Drive, anchored by Water Tower Place (Marshall Field) and the Atrium (with Bloomingdale’s). The malls are designed both to exclude unwanted customers and to attract the well-to-do. Source: K.Jones and J.Simmons (1990) The Retail Environment London: Routledge 2. Environmental considerations. Planners are suspicious of any development that might contribute to urban sprawl and are particularly reluctant to allow new suburban shopping facilities to erode the rural character of the green belt. Opponents argue that this presumes that green-belt land is uniformly attractive, whereas a new shopping facility on a derelict site could contribute to visual upgrading of the landscape. 3. Social considerations. Many planners believe that decentralisation will create two different standards of shopping provision, with most of the benefits of suburban developments accruing to a car-oriented middle-class population, leaving the poorer sections of society dependent upon traditional shopping areas which are likely to run down. A general argument against this is that different standards of retail provision exist already within the urban retail system, and that while minority needs must be addressed, the majority should not be deprived of greater choice. TOWN-CENTRE SHOPPING SCHEMES No country in Western Europe has sought to contain the process of decentralisation to the same extent as Britain. As a result of planning policy, much of the post-Second World War pressure for retail growth has been directed into town-centre developments. Between 1950 and 1980 two distinct stages can be identified in the evolution of town-centre shopping schemes (Figure 12.7). During the first Urban geography 348 Figure 12.7 Stages in the evolution of town-centre shopping schemes in the UK Source: D.Bennison and R.Davies (1980) The impact of town centre shopping schemes in Britain Progress in Planning 14(1), 1–104 TABLE 12.3 THE IMPACT OF TOWNCENTRE SHOPPING SCHEMES IN BRITAIN Economic Environmental Social Positive Negative Positive Negative Positive Negative Adds new stock Reduces old stock Modernises outworn areas Changes traditional shopping Allows for efficient shopping May favour car-borne shoppers Accommodates larger, modern stores Discriminates against small independents Reduces land-use conflicts Creates new points of congestion Provides new shopping opportunities May limit choice to stereotypes Increases rates and revenues Increases monopoly powers Scope for new design standards Intrusive effects Safer on older townscapes Creates new employment Changes structure of employment Provides weather protection Creates artificial atmosphere Improves trade on adjacent streets Reduces trades on peripheral streets Leads to Causes blight upgrading of on other streets some streets Enhances status of central area Affects status of surrounding centers Integrates new transport Creates new stress factors from crowds Provides more Attracts comfort and delinquents amenities and vandals Concentrates shopping in one area Causes pressure Potentially on existing greater social infrastructures interaction Breaks up old shopping linkages Becomes dead area at night Source: D.Bennison and R.Davies (1980) The impact of town centre shopping schemes in Britain Progress in Planning 14(1), 1-104 Urban retailing 349 period, from 1950 to 1965, precinct development was fostered by local authorities initially prompted by the need for core-replacement schemes in war-damaged cities such as Coventry, Hull and Plymouth. Towncentre developments were stimulated by: 1. the abolition in 1954 of the wartime system of licences for new buildings; 2. rising personal incomes, which were reflected in a demand for new retail floor-space; 3. growth in vehicular traffic, leading to local congestion in cities; 4. an increase in the amount of private finance capital available, and a shift in investment focus from residential to commercial property. The last of these factors is reflected in the greater involvement of the private sector from the early 1960s (Figure 12.7). Most of the new central-area shopping centres were the result of collaboration between a city planning department and a propertydevelopment company. The planning authority maintained overall control of the scheme in terms of location, its relation to the existing retail pattern and its integration with other parts of the urban renewal programme, while the centre design and shop layout were left to the property developers, who also managed the precinct as an integrated unit in much the same way as in planned suburban developments in North America. The impact of towncentre shopping schemes is summarised in Table 12.3. The desire of planners to protect existing retail outlets is revealed in the fact that between 1965 and 1980, of 387 new shopping centres built, 315 (81 per cent) were located in town centres.26 RETAIL DECENTRALISATION IN BRITAIN The concentration on the redevelopment of existing retail areas did not eliminate pressure from developers for out-of-town shopping centres, although initially this pressure achieved little result. In 1964 a major planning application for a 1 million ft2 (90,000 m2) centre at Haydock, between Manchester and Liverpool, was refused because of the potential impact on existing town centres in the region. During the remainder of the 1960s and 1970s government opposition meant that few proposals for out-of-town regional shopping centres were Figure 12.8 British shopping-centre development: total floor space opened each year, 1965-93 Urban geography 350 Source: C.Guy (1994) The Retail Development Process London: Routledge submitted. In addition, a property-market slump in 1973-4 (Figure 12.8) reduced the amount of capital available for investment in new shopping centres. Only where a suburban shopping development formed the basis of a new district centre in an area of former underprovision or substantial population growth did local planning authorities relax their opposition. The 760,000 ft2 (70,000 m2) Brent Cross regional shopping centre was developed in 1977 in suburban North-West London in an area without major existing facilities, and the first free-standing super-regional centre was built with 1,065,000 ft2 (100,000 m2) of retail floor space as the central shopping area for Milton Keynes New Town in 1979. The 1980s saw a revival of interest in shoppingcentre development, with fifty proposals for out-oftown regional shopping centres submitted between 1982 and 1991. This trend was encouraged by an apparent change in central government’s philosophy on land-use controls. With the election of the Conservative government under Margaret Thatcher in 1979 there was a general presumption in favour of development, ‘enterprise’ and employment creation. Government liberalisation of planning controls was taken furthest within enterprise zones (EZs), and in several of these, major retail developments were permitted, as in the 1.63 million ft2 (150,000 m2) MetroCentre, built on the site of the ash tip of a disused power station within the Gateshead EZ, and the 1.1 million ft2 (102,000 m2) Meadowhall, which replaced a derelict steelworks outside Sheffield (Table 12.4 and Figure 12.9). In practice, government encouragement of competition among retailers did not herald a major Plate 12.3 The 1.63 million square feet MetroCentre was built on an abandoned industrial site within the Urban retailing 351 Gateshead Enterprise Zone, and is one of the largest covered shopping malls in Europe relaxation of planning controls over greenfield developments. Between 1985 and 1992 only four super-regional centres opened although several others had received planning permission. By 2000 a further six of the proposed regional centres had opened (Figure 12.10). All these had received planning permission prior to the change in government policy on retail development in 1996. In addition, in 1997 an application for expansion of the Merry Hill RSC by 400,000 ft2 of retail and leisure space was refused by the Secretary of State. These trends effectively signalled the end of the ‘third wave’ of retail decentralisation in Britain27 (see Box 12.7). The fourth and fifth waves of retail decentralisation in the UK are characterised by the rise of retail warehouse parks and factory outlet centres respectively. By 1997 planning consent had been granted for 1,600 retail parks (with 230 under construction) and 158 factory outlet centres (with eightyone under construction).28 The centres open or likely to open are all within or close to major conurbations and are in areas of industrial or mining dereliction; none of the completed out-of-town regional shopping centres and few of the proposed sites are in the American mould of edge-of-town developments on greenfield sites. THE IMPACT OF REGIONAL SHOPPING CENTRES IN THE UK Assessing the impact of a new regional shopping centre is complicated by the difficulty of isolating this effect from that of general economic processes relating to the national or global economy. Nevertheless, several studies have sought to gauge the impact of a new out-of-town development on surrounding retail centres.29 It is clear that the competitive impact depends greatly on local context. The MetroCentre, located 4 miles (6 km) west of Newcastle city centre, is one of the largest enclosed shopping centres in Europe, with 1.63 million ft2 (150,000 m2) of retail floor space, an additional 107,000 ft2 (9,900 m2) of leisure floor TABLE 12.4 LOCATION AND SIZE OF ENCLOSED CENTRES OF OVER 500,000 ft2 IN BRITAIN Town Size (000 ft2) Centre Year opened Manchester Arndale 1976 1,189 Newcastle Eldon Square 1976 830 Luton Arndale 1972 700 Watford Harlequin 1992 700 Poole Dolphin 1969 686 Nottingham Victoria 1972 622 In-town Urban geography 352 Kingston upon Thames Bentall 1991 600 Cardiff St David’s 1981 581 Maidstone Chequers 1976 542 Milton Keynes Central Milton Keynes 1979 1,065 Redditch Kingfisher 1973 780 Telford Telford Shopping Centre 1973 775 Peterborough Queensgate 1982 650 Runcorn Shopping City 1971 600 Washington The Galleries 1977 543 Basildon Eastgate 1980 517 Gateshead MetroCentre 1986 1,630 Dudley Merry Hill 1989 1,250 Thurrock Lakeside 1990 1,150 Sheffield Meadowhall 1990 1,100 Hendon Brent Cross 1976 760 Leeds White rose 1997 577 Bristol Cribb’s Causeway 1998 577 Manchester Trafford Centre 1998 1,163 Dartford Bluewater 1999 1,461 Glasgow Braehead 1999 538 London White City 2000 557 New towns Out-of-town space and 10,000 car-parking spaces. Survey evidence indicates that 12 per cent of shopping trips have been directed to the MetroCentre from Newcastle city centre. This has led to a compaction process in Newcastle, with a refocusing of the CBD on the refurbished enclosed central mall (Eldon Square) and contraction of retailing in peripheral shopping streets. In general, however, large-scale and prescient redevelopment of the city centre retailing area, together with the insertion of an urban metro system in the 1970s, has enabled Newcastle to retain a thriving central retail function in the face of out-of-town competition. The impact of other regional shopping centres on nearby central areas has been more severe. The central shopping area of Dudley in the West Midlands lost 70 per cent of its market share to the Merry Hill mall, with more than half the retailers present in 1986 gone by 1992; some of them moved direct to the new 1,240,000 ft2 (115,000 m2) regional shopping centre. In 1997 an application from the developers to expand the centre by the Urban retailing 353 construction of an additional 375,000 ft2 (35,000 m2) of gross floor space for retail and leisure uses was refused by the Secretary of State in order to protect the ‘vitality and viability’ of existing centres. By then, however, the Merry Hill regional shopping centre had effectively replaced Dudley town centre as the main retail focus of the area.30 In similar fashion Meadowhall has also exerted a significant impact on the CSA in Sheffield. Since the mid-1990s the policy priority in favour of existing retail centres has encouraged developers of ‘very large stores’ (over 7,000 m2) to redevelop existing stores (by, for example, Figure 12.9 The layout of out-oftown regional shopping centres in the UK: (above) the MetroCentre, Gateshead and (lower) Meadowhall, Sheffield converting storage space to sales space and using just-in-time logistics) or to seek to link their retail development to urban regeneration programmes (see below). DISADVANTAGED CONSUMERS The post-Second World War changes in the structure of urban retailing have brought material advantages to a majority of the population through increased choice, comfort Urban geography 354 and cheapness in shopping. On the other hand, the increasing size and decentralised location of new large stores, disinvestment in smaller supermarkets by the major food retailers who dominate the grocery market, and the decline in the number of small independent shops have contributed to the formation of a class of disadvantaged consumers. This group includes the poor, elderly and mobility-deprived residents of under-served inner-city areas and, in the UK, peripheral public housing estates.31 BOX 12.7 Retail decentralisation in the UK 1. First wave (late 1960s to early 1970s). The development of out-of-centre and even outof-town sites of the first free-standing superstores and hypermarkets. By 1993 the existing 868 superstores transacted almost 50 per cent of grocery-shopping trade, and the top five retail companies (Sainsbury’s, Tesco, Argyll-Safe way, Asda, DeeGateway) controlled 61 per cent of the British grocery market. 2. Second wave (late 1970s to mid-1980s). The arrival of retail warehouses selling durable goods (electrical white goods, carpets, furniture, DIY products) and garden equipment. More recently the range has expanded to include clothing, footwear and toys. By 1992 there were over 2,000 retail warehouses. 3. Third wave (mid-1980s to early 1990s). The move of high-street chain stores to outoftown centres in accessible locations, and providing a range of comparison goods shopping that was previously reserved for the town centre. A significant extension of this trend was the development of out-of-town regional shopping centres such as the Gateshead MetroCentre. 4. Fourth wave (late 1980s to mid-1990s). The development of the retail warehouse park, of which there were 250 by 1992. Initially, small groups of second-generation purpose-built retail warehouses on edge-of-town sites; later developments exhibit a more co-ordinated layout developed in co-operation with local planning authorities. 5. Fifth wave (mid-1990s–). The emergence of factory outlet centres as a further form of retail decentralisation in which groups of leading manufacturers sell products direct to the public. Though these are common throughout the USA, in 1995 there were only four operating in the UK, with planning applications for another twenty. 6. Sixth wave (2000–). The ultimate level of decentralisation, home-shopping has long been available through catalogue sales but is becoming more widely available via specialised cable television channels and by means of teleshopping in a virtual superstore via the Internet. Urban retailing 355 Figure 12.10 Major out-of-town shopping centres open or under development in the UK Some local authorities in the UK have sought to address the problems of disadvantaged consumers by seeking to influence the geography of retailing. At the neighbourhood level, in 1980 the borough council of Islington in north London identified a number of ‘shopping deficiency areas’ within which new retail investment would be encouraged, Urban geography 356 possibly using the ‘planning gain’ mechanism (see Chapter 8). In principle this approach could also be employed at the metropolitan level, particularly in situations where retail developers seeking planning permission for new large stores are willing to work with local authorities to tackle social exclusion. As well as bringing employment benefits to areas of long-term unemployment and a low-skilled workforce, very large store (VLS) developments can also provide deprived communities with an outlet offering low prices and a wide range of merchandise in areas where the existing retail environment is poor.32 More generally, however, although local authorities can use planning controls to prevent specific developments, they are less able to direct investment to areas of need. In view of this it is necessary to consider alternative ways of assisting mobility-deprived households. One possibility is the use of telecommunications and computer facilities to link individual homes or community focal points to a large store or shopping centre some distance away. The first such scheme in the UK, the Gateshead shopping and information service (SIS), was set up in 1980. The scheme aims to improve the: 1. choice of shopping opportunities available to disadvantaged consumers and to provide them with access to stores offering goods at cheaper prices; 2. access of the relatively housebound to a variety of sources of information on local events, welfare, entertainments, transport timetables and other social services. The system has 1,000 regular users.33 Unlike the commercial teleshopping schemes common in the USA, the Gateshead SIS is operated as a joint venture between the local authority and a major retailer. It is aimed at providing a food and convenience-goods shopping service rather than one for non-food luxury items, and is oriented to disadvantaged rather than affluent consumers. Although electronic home shopping (EHS) and the general growth of e-commerce have considerable potential to overcome mobility disadvantage, the cost of providing terminals means that in the short term, most of the residents of inner-city and deprived council estates will continue to rely on conventional retail outlets for their basic shopping needs. The uneven distribution of modern retail facilities within the metropolitan area, and the differential ability of consumers to benefit from such developments, underline the importance of an efficient urban transportation system. This issue will be considered in Chapter 13. FURTHER READING BOOKS B.Berry, J.Simmons and R.Tennant (1963) Commercial Structure and Commercial Blight Research Paper 85 Chicago, IL: Department of Geography, University of Chicago R.Bromley and C.Thomas (1993) Retail Change London: UCL Press R.Davies (1995) Retail Planning Policies in Western Europe London: Routledge R.Davies and J.Reynolds (1988) The Development of Teleshopping and Teleservices Harlow: Longman J.Dawson (1983) Shopping Centre Development Harlow: Longman K.Jones and J.Simmons (1990) The Retail Environment London: Routledge P.McGoldrick and M.Thompson (1992) Regional Shopping Centres Aldershot: Avebury N.Wrigley and M.Lowe (2002) Reading Retail London: Arnold Urban retailing 357 JOURNAL ARTICLES J.Brooks and A.Young (1993) Revitalising the central business district in the face of decline Town Planning Review 64, 251–71 C. Guy (1998) Controlling new retail spaces: the impress of planning policies in Western Europe Urban Studies 25(5/6), 953–79 B.Hahn (2000) Power centers: a new retail format in the USA Journal of Retailing and Consumer Services 7, 223 31 E.Howard (1993) Assessing the impact of shopping centre development: the Meadowhall case Journal of Property Research 10, 97 119 A.Loukaitou-Sideris (1997) Inner-city commercial strips Town Planning Review 68(1), 1–29 K.Robertson (1995) Downtown redevelopment strategies in the United States Journal of the American Planning Association 61, 429 37 KEY CONCEPTS ■ shopertainment ■ spaces of consumption ■ e-commerce ■ nucleated centres ■ ribbon development ■ specialised function areas ■ power centres ■ theme malls ■ planned shopping areas ■ unplanned shopping areas ■ retail strip ■ regional shopping centres ■ out-of-town centres ■ commercial blight ■ retail suburbanisation ■ town-centre developments ■ super-regional shopping centres ■ disadvantaged consumers STUDY QUESTIONS 1. Identify the major forces influencing the structure of urban retailing in the post-war period. 2. Examine the extent to which planning policies in the UK have constrained the process of retail decentralisation. 3. Examine the effects of retail suburbanisation on inner-city retail areas in either the UK or the USA. Can you provide evidence of commercial blight or the development of polarised levels of retail provision in a city with which you are familiar? 4. Identify the difficulties faced by ‘disadvantaged consumers’. Should public authorities intervene in the market process to influence retail developments in favour of such groups and, if so, in what ways? Urban geography 358 PROJECT Map the distribution of shopping areas in a city with which you are familiar. Gather information, from public sources and/or field visits, on the size and retail composition of each centre. Construct a hierarchy of shopping areas. Compare the geographical and hierarchical distribution of the shopping areas with the models discussed in the chapter. 13 Urban Transportation Preview: patterns of travel demand; urban transport problems; strategic responses to the urban transport problem; private versus public transport; mass transit systems; transport system management; road pricing; auto-restraint measures; non-transportation initiatives; transport and urban structure; transport and sustainable urban development INTRODUCTION There is a close relationship between the nature of urban transportation and urban structure (Table 13.1). When most people had to walk to engage in essential daily tasks, cities were necessarily compact. Citizens lived at or close to their workplaces, circumstances that favoured high-density living environments in small, functionally integrated cities that rarely achieved populations of 50,000.1 Only during the industrial revolution did vehicles of relatively high capacity and speed allow greater distances to be travelled and larger quantities of goods to be exchanged. This relaxed restrictions on city size and established the interdependence between transport technology and urban form. During the nineteenth century the development of railways and trams (streetcars) was critical in separating home and workplace, encouraging functional specialisation of land uses in the city, and in promoting the penetration of the surrounding countryside along the more accessible transport TABLE 13.1 THE RELATIONSHIP BETWEEN TRANSPORT AND URBAN FORM IN WESTERN CITIES Stage Urban functions Transport technology Transport system Urban form 1. Preindustrial Defence, marketing, Pedestrian, political-symbolic, draught animal craft industry Route convergence, radial Compact 2. Early industrial Basic industries, secondary, manufacturing Electric tram, Radial streetcar, public improvements, transport incremental additions High-density suburbanisation, stellate form 3. Broadening Motor bus, Lower-density Additional radials, Urban geography Industrial industry, tertiary service expansion public transport, early cars 4. Postindustrial Addition of Towards quaternary activities universal carownership 360 initiation of ‘ring’ roads (incomplete) suburbanisation, industrial decentralisation Integrated radial and Low-density circumferential road suburbanisation, network widespread functional decentralisation Source: D.Herbert and C.Thomas (1997) Cities in Space: City as Place London: Fulton corridors. In London between 1860 and 1914 the Underground system and suburban railway lines were formative influences in urban development that permitted extensive suburbanisation and growth of commuter settlements well beyond the urban core. Similar patterns were evident around major cities in the North-Eastern USA. In the postSecond World War era widespread car-ownership led to a significant increase in personal mobility and a massive expansion in the built-up areas of cities (Box 13.1). In this chapter we shall examine the increasing volume and complexity of urban travel, and the problems posed for urban transportation. We shall analyse the major strategic responses to the urban transport problem, examining a range of both transportand non-transport-based options. Finally, the essential relationship between transport and urban form will be elaborated and the links between transport and sustainable urban development considered. PATTERNS OF TRAVEL DEMAND In the USA between 1945 and 1977 the total number of vehicles rose from 29.5 million to 128.6 million (+336 per cent) and by the latter date one-third of the 22,843,000 households had more than one car. By 2000 55 per cent of US households owned more than one car. In 2000 the motor-vehicle population of the USA had reached 235 million for a total population of 281.4 million. The percentage of the US population licensed to drive has also increased, from 72 per cent of those eligible in 1960 to 88 per cent in 1990, and to over 90 per cent by 2000. Those increases have in turn fuelled a rise in vehiclemiles travelled. In addition, the general BOX 13.1 Major eras of metropolitan growth and transport development in the USA The relationship between transport technology and urban structure may be summarised as a series of stages: 1. Walking-horsecar era (1800–90) during which dependence on walking and horsedrawn vehicles for urban transport ensured retention of a compact, high-density form of city. 2. Electric streetcar era (1890–1920) in which the invention of the electric traction motor led to the growth of streetcar suburbs clustered around stops on radial trolley routes. Urban transportation 361 3. Recreational automobile era (1920–45) in which the enhanced personal mobility bestowed by the automobile transformed the ‘tracked city’ of electric trolleys and trains into a suburban metropolis. 4. Freeway era (1945–) during which the automobile became a necessity of modern urban life, and highway developments promoted urban sprawl and the growth of edge cities. This transition to a polycentric metropolis may be seen to have had five growth phases: ■ Bedroom community (1945–55) dominated by a post-war residential building boom but only modest expansion of suburban commercial activity. ■ Independence stage (1955–65) acceleration of economic growth led by the first wave of industrial and office parks and, after 1960, by rapid diffusion of regional shopping centres. ■ Catalytic growth (1965–80) characterised by further growth of regional shopping malls and associated office, hotel and restaurant facilities at accessible locations in the suburbs. ■ High-rise/high-technology (1980–90). Suburban downtowns developed as functional equals of the central business district, characterised by high-rise office buildings and clustering of high-technology R&D facilities as part of the postindustrial service economy. ■ Mature urban centres (1990–) in which the suburban downtowns evolve into complexes with diversified land uses and acquire enhanced roles as cultural, entertainment and civic centres, with greater local involvement in resolving problems such as traffic management via, for example, traffic-management associations (TMAs) set up by business and local government. The leading urban transport challenge for the twenty-first century focuses on the efficiencies of moving people about the dispersed, polycentric metropolis. TABLE 13.2 CHANGING MODE OF JOURNEY TO WORK IN THE USA, 1990– 2000 1990 No. 2000 % No. % change % Drove alone 84,215,298 73.2 97,102,050 75.7 2.5 In carpools 15,377,634 13.4 15,634,051 12.2 −1.2 6,069,589 5.3 6,067,703 4.7 −0.6 1, 512, 842 1.3 1,532,219 1.2 −0.1 7,894,911 6.9 7,943,205 6.1 −0.8 Public transport Other means Walked or worked at home Source: US Census Urban geography 362 trend away from walking and public transport to auto use, and lower levels of vehicle occupancy (Table 13.2), have meant that increasingly more cars are being used to serve the same number of travellers.2 Similarly, in the UK the number of cars grew from 2 million in 1950 to 23 million in 2000 (Table 13.3), while the proportion of households owning a car rose from 14 per cent in 1951 to 73 per cent in 2000 (Table 13.4). Since the increase in the length of the road network (which totaled 365,000 km in 1994) has been much less than the increase in traffic, the ‘average daily flow’ on all roads increased by nearly half between 1981 and 1993. Much of this TABLE 13.3 NUMBER OF VEHICLES, GREAT BRITAIN, 1950–2000 (000s) Year Cars All vehicles 1950 1,979 3,970 1960 4,900 8,512 1970 9,971 13,548 1980 14,660 19,199 1990 19,742 24,673 2000 23,196 28,898 increase is the result of the dispersed pattern of urban activities and the increased separation of home and work. TABLE 13.4 PERCENTAGE OF HOUSEHOLDS WITH REGULAR USE OF A CAR, GREAT BRITAIN, 1951–2000 Year No. of cars None No. households (million) 1 2 3 1951 86 13 1 — 14.5 1970 48 45 6 1 18.2 1980 41 44 13 2 19.9 1990 33 44 19 4 21.9 2000 27 45 23 5 24.9 These trends are particularly evident in the US metropolis. People now travel longer distances to work, although travel time has remained steady. In 1975 the average travel distance to work was 9 miles and average travel time twenty minutes. In 1990 the average work trip covered 11 miles and took twenty minutes. Clearly, national average figures conceal considerable variation among groups differentiated by age, gender, travel mode Urban transportation 363 and place of residence. Nevertheless, the constant average travel time despite the increase in distance covered reflects improvements in the transportation system and also an ‘improvement in individual speeds obtained by shifts to the single-occupant vehicle from car pooling, mass transit and walking’ (Pisarski 1992 p. 69).3 Accompanying this trend, the spatial pattern of commuting flows has become increasingly complex (Figure 13.1). Reflecting the dispersed structure of the US metropolis, the suburb to central-city trip has not been the dominant work trip since at least 1970.4 In 1994 the suburb-to-suburb commute accounted for 34 per cent of all metropolitan work trips, with the ‘traditional’ suburb-to-central-city commute constituting 17 per cent and the reverse commute (central city to suburb) 6 per cent. In view of the complexity of flows, it is not surprising that the proportion of work trips made on public transport has continued to decline. By 1990 those commuting by private vehicle accounted for 88 per cent of all work trips.5 Non-work travel has grown at a faster rate than work travel, accounting for 59 per cent of all local travel in the USA in 1969 and 74 per cent by 1990.6 This increase may be attributed to a rise in the number of affluent and dual-income households, which promotes more leisure trips, and a decline in household size Figure 13.1 A typology of commuting flows. CC Central city, UF Urban field Source: D.Plane (1981) The geography of urban commuting fields Professional Geographer 33, 182–8 Urban geography 364 and corresponding rise in the number of households. In combination, these trends of more vehicles on the roads, increased mileage travelled, longer distances covered, increased reliance on and use of a private car and the growing number of households reflect an unprecedented level of personal mobility, but have also created many of the problems associated with urban transportation. THE URBAN TRANSPORT PROBLEM The major facets of the urban transport problem are shown in Figure 13.2. These comprise: 1. Traffic movement and congestion. The primary function of urban transport is to provide mobility for people and goods within the city (Box 13.2), but the efficiency with which this is achieved is reduced by congestion. The major Figure 13.2 Dimensions of the urban transport problem Source: M.Thomson (1977) Great Cities and their Traffic Harmondsworth: Penguin BOX 13.2 Accessibility and mobility The concepts of accessibility and mobility are central to understanding urban transportation. Accessibility refers to the number of opportunities or activity sites available within a certain distance or travel time. Mobility refers to the ability to move between different activity sites As the distances between activity sites have become Urban transportation 365 larger, because of lower-density settlement patterns, accessibility has come to depend more and more on mobility, particularly in privately owned vehicles. Although the need for mobility is a consequence of the spatial separation of different types of land uses in the city, enhanced mobility also contributes to increased separation of urban land uses because improved transportation facilities enable people to travel farther in a given amount of time than they could previously. Sources: S.Hanson (1995) The Geography of Urban Transportation New York: Guilford Press; S.Handy and D. Niemeier (1997) Measuring accessibility Environment and Planning A 29, 1175–94 cause of urban traffic congestion is the increasing number and use of vehicles on the roads. More specifically, it stems from the concentration of travel flows at certain times during the day, with the principal reason for the typical double-peak distribution of daily trips being the journey to and from work. Different parts of the city may experience traffic congestion at different times of the day, depending on the mix of traffic, but most large cities experience serious congestion in their central areas during peak hours. In central London average traffic speed fell from 12.9 mph (20.7 km/hr) in 1972 to 10.9 mph (17.6 km/hr) in 1990. This is close to the postulated equilibrium speed of 10 mph (16 km/hr) for peak-hour traffic in the city centre. According to Thomson (1977), motorists will tolerate speeds as low as this before they begin to avoid the area in large enough numbers to create equilibrium at that critical speed.7 2. Crowding on public transport. In nearly every city the use of public transport is concentrated in the morning and evening rush hours. Whatever the volume of demand, there is invariably insufficient capacity to provide comfortable travel conditions at these times. During conditions of peak-hour loading, passengers are often subjected to lengthy queues at stops, crowding at termini, and excessively long periods of hot and claustrophobic travel in overcrowded vehicles. In Tokyo the metro rail system employs ‘pushers’ to ensure that passengers are forced into trains to allow the automatic doors to close. 3. Off-peak inadequacy of public transport. The difference between peak volume and offpeak usage of public transport means that operators face the financial problem of maintaining sufficient vehicles, plant and labour necessary to provide a peak-hour service which is underused for the rest of the time. A usual response has been to reduce off-peak services, leading to inadequate, unreliable and often non-existent provision of public transport at certain times of the day. 4. Difficulties for pedestrians. Paradoxically, although a large number of trips in cities are made on foot, pedestrians are not often included in urban transportation studies. Pedestrians (and pedal cyclists) encounter two kinds of problem. The first is the problem of accessibility to facilities. The replacement of local outlets (e.g. shops, hospitals, etc.) by larger units serving wider catchment areas puts many urban activities beyond reach of the pedestrian. The second refers to the quality of the pedestrian environment, with footbridges and underpasses often inadequately cleaned and policed, and traffic noise and fumes affecting foot travellers most directly. 5. Environmental impact. Transport is a major source of air pollution in cities, with exhaust gases (carbon dioxide, carbon monoxide, nitrogen oxides, hydrocarbons) and Urban geography 366 other pollutants, such as lead and particulates, contributing to a range of health and environmental problems. Other traffic-induced environmental impacts include noise pollution, visual intrusion, the destruction of natural habitats and segregation of communities by transport routes. 6. Accidents. Road traffic accidents constitute a significant social problem, and the majority occur in urban and suburban areas. A minority of the victims are car occupants and a high proportion are pedestrians, cyclists and motor cyclists. Most transportation studies, however, regard road traffic accidents as an unfortunate offshoot of an urban transport system. 7. Parking difficulties. In most city centres, finding a place to park a car is difficult. In many cities parking is often seen, by drivers at least, as the major problem of urban transport. As we shall see later, since it is physically impossible for a large city to provide car parking space for all who wish to enter the centre, restrictions apply. These may lead to illegal parking, which can impede the flow of traffic. Parking difficulties are different from other aspects of the urban transport problem, however, as they are often maintained deliberately by local authorities in an attempt to reduce other problems (Figure 13.2). RESPONSES TO THE URBAN TRANSPORT PROBLEM Three general approaches may be identified: 1. During the 1950s and 1960s the supply-fix approach emphasised the provision of new infrastructure to increase the capacity of the road system to meet demand. This policy was manifested in large-scale road-building and road-improvement programmes. The emphasis Plate 13.1 Rush-hour passengers crowd into a train on the Chuo line in Tokyo Urban transportation 367 was firmly on the supply side, and transport policy was effectively a vehicleoriented policy. Large increases in highway capacity in many cities tended to exacerbate the problems of traffic congestion and environmental degradation in accordance with Down’s law, which states that on urban commuter expressways peak-hour traffic congestion rises to meet maximum capacity.8 In addition, the increased mobility of car-owners stimulated the development of low-density urban areas with unstructured travel demands, resulting in substantial reductions in the role and effectiveness of public transport. 2. During the late 1960s and early 1970s the character and emphasis of urban transport policy shifted. The policy objective became accessibility rather than simply mobility. Greater emphasis was placed on the exploitation of existing facilities, on minimising the environmental impact of the automobile, and on the equity with which all sections of the urban population were served by transport. The shift to people-oriented noncapital-intensive policy options was stimulated by reductions in the growth rate of many large urban areas, lower levels of national economic growth, the rise of a popular movement opposed to the environmental and social costs of major highway projects, and the energy crisis initiated by the Arab oil embargo of 1973. The resultant transportsystem management strategies sought to manipulate demand and make more efficient use of existing highway capacity. 3. The third approach is by means of non-transportation initiatives, which can range from staggering work hours to designing cities to reduce the need for travel by limiting distances between home and other centres of activity. We can examine each of these three approaches. Plate 13.2 Growing concern with the social and environmental costs of highway construction signalled the end of the road for some urban freeways, such as this project in San Francisco CA Urban geography 368 PRIVATE VERSUS PUBLIC TRANSPORT The major policy debate during the era of the supply-fix approach focused on the relative merits of public and private transport. As Figure 13.3 shows, the emphasis on the provision of additional highway capacity to accommodate suburbanisation and rising levels of car ownership induced a decline in public-transport use. Throughout the industrialised world, declining fare revenues and increasing costs have produced a situation in which few metropolitan mass-transit systems operate without subsidies. The continued provision of transport as a public rather than commercial service depends on the attitude of government. The city of Curitiba in Brazil is often cited as a model for urban transport planners elsewhere (Box 13.3). In the UK the 1974–9 Labour government supported urban public transport, and several cities, including London, introduced publicly subsidised fare-cutting experiments to increase the attractive ness of public transport. This approach was contrary to the competitive-market philosophy of the incoming Conservative government in 1979, which abolished subsidies to public transport and introduced deregulation to open up the sector to competition. This led to cost savings but a reduction in services outwith peak hours. Similarly, in Toronto a subsidised rapid-transit system in the central city integrated with suburban rail and bus services led to an increase in journeys to work by public transit, reaching a peak of 22 per cent in 1988. Subsequent withdrawal of subsidies resulted in a sequence of fare increases, declining quality of service and a 15 per cent reduction in passenger traffic by 1993.9 In some countries, including the USA, Germany and France, travel voucher schemes have been employed to encourage workers to use public transit. The basic model is that the employer issues travel vouchers to be used to pay for travel on public transport. As the vouchers are tax-free there is an incentive for employers and employees to use them instead of a (taxed) wage increase of Figure 13.3 The relationship between car-ownership and busridership Urban transportation 369 equivalent value. Transit makes a minor contribution to mobility in American cities, with only 2.5 per cent of all person trips made by transit in 1990 compared with 86 per cent by automobile. In the USA transit use is concentrated in the large cities, with New York accounting for 40 per cent of national transit passenger mileage. Work trips are the mainstay of transit, and in 1990 accounted for 55 per cent of all weekday transit trips.10 In smaller cities, and for most non-work trips, public transport cannot compete with the private car. The benefits of most transit systems for the alleviation of peak-hour congestion on work trips to and from the central city underlie efforts to expand its use. To be effective such strategies must enhance the attractiveness of public transport and restrict the use of the private car. A good-quality transit system must: 1. offer fares low enough for the poor to afford; 2. operate sufficient vehicles to run a frequent service throughout the day; 3. have routes which reflect the ‘desire lines’ of the travelling public, with extensive spatial coverage of the catchment area to ensure that no one is too far from a transit stop; BOX 13.3 Public transport in Curitiba, Brazil The Curitiba master plan is based on an integrated approach to traffic management, transport and land-use planning. On each of the five major transport arteries priority is afforded to public transport, with one two-way lane reserved for express buses. This central lane is flanked on either side by a local access lane for cars, and a highcapacity one-way route for use by both cars and buses. Separating traffic types and establishing exclusive bus lanes has produced a reliable and efficient public-transit system serving 1.3 million passengers daily, and has promoted densification of development along the bus routes. Curitiba’s buses are privately owned by ten companies, managed by a quasi-public company. With this public-private collaboration, publicsector concerns (e.g. safety, accessibility and efficiency) are combined with private-sector goals (e.g. low maintenance and operating costs). The bus companies receive no subsidies; instead all mass-transit income is pooled and companies are paid on a distance travelled basis. Curitiba’s buses carry fifty times more passengers than they did twenty years ago but people spend only 10 per cent of their annual income on transport. As a result, despite the second highest per capita car-ownership rate in Brazil, (with one car for every three people), Curitiba’s gasoline (petrol) use pet- capita is 30 per cent below that of eight comparable Brazilian cities. 4. increase bus speeds relative to those of cars; 5. provide a co-ordinated multimodal one-ticket system with convenient connections between transport modes. Many cities in North America and Europe have introduced rapid mass-transit systems. In Stockholm 38 per cent of residents and 53 per cent of workers living in satellite new towns commute by rail.11 Other European cities have upgraded tram and light rail Urban geography 370 systems,12 introduced electrified suburban railway lines (as in Copenhagen and Glasgow) or expanded underground metro systems (London and Paris). In the UK the Tyneside Metro was completed in 1981 at a cost of £300 million; the 34-mile (55 km), forty-sevenstation light rail system is integrated with bus services and park-and-ride facilities, and operates with a heavy subsidy. Cities such as Manchester and Sheffield have introduced electric trams running along existing roads and railway lines to limit costs, while Croydon’s 28 km tramlink system utilises converted sections of disused railway track. In London’s Docklands a light rail system constructed to link the City with new office developments at Canary Wharf has since been extended east to Beckton to aid further regeneration of the area. In the USA the Bay Area Rapid Transit (BART) system in San Francisco CA carries more than half of all CBD-bound work journeys. Other large cities such as Washington DC (METRO) and Atlanta GA (MARTA) have developed entirely new rapid rail transit (RRT) systems. More recently, light rail transit (LRT) has been seen as a low-cost alternative. Between 1981 and 1994 LRT systems were opened in San Diego CA, Buffalo NY, Portland OR, Sacramento CA, San José CA, Los Angeles CA, Baltimore IN and St Louis MO.13 Significantly, these public-transit systems have not achieved the passenger volumes anticipated.14 Nevertheless, such is the scale of concern over the impact of the automobile on the quality of city life that federal support for mass transit is ongoing, with the 1991 Intermodal Surface Transportation Efficiency Act (ISTEA) permitting the use of federal highway funds for other modes, including transit as an alternative to the singleoccupancy vehicle form of urban travel. The Transportation Equity Act for the TwentyFirst Century (TEA-21), enacted in 1998, is intended to further ISTEA’s comprehensive flexible approach to urban transportation. Fundamentally, if public transport is to compete successfully with the private car it has to provide a comparable service. It has to be affordable, safe, comfortable, reliable, timely and flexible enough to cater for trips that are not necessarily centrally oriented. Integration between different public transport systems must be easy, leading to ‘seamless travel’ with smart cards to cover different modes and minimum waiting times at interchange stations. Such integration requires financial and political commitment. TRANSPORT SYSTEM MANAGEMENT Reaction against continued large-scale provision of highway infrastructure has resulted in literally hundreds of techniques for better use of existing transport facilities. These include channellisation of traffic, especially at intersections (e.g. segregation of left- and right-turning flows, and bus-only lanes), computer-controlled traffic signals, more traffic light-controlled junctions and mini-roundabouts to increase the flow that can be handled by a single intersection, one-way streets to reduce conflicting traffic movements at junctions, limitations on street parking (e.g. clearways) to increase the carrying capacity of traffic routes, reserved lanes for ‘high occupancy’ vehicles, and reversible traffic lanes in which priority changes with time of day. All these management techniques can be easily implemented at relatively low cost and involve minor physical alterations. More sophisticated options based on the concept of intelligent vehicle highway systems (IVHS) involve the use of information technology, such as geographic information systems and global positioning systems, to improve traffic flow by means of on-board navigation. Urban transportation 371 Advanced vehicle control systems (AVCS) under development include both collisionwarning systems and devices that would automatically take over control in order to avoid a crash. The ultimate IVHS is the ‘smart highway’ that would enable all drivers to switch to autopilot, ensuring maximum throughflow without accidents.15 Currently available strategies for transport system management also include the options of road pricing, auto-restraint policies and encouragement of ride-sharing: Road pricing Although in many countries road travellers pay to use inter-city roads, tolls for travel within urban areas have not been adopted widely. In Norway three cities, Oslo, Bergen and Trondheim, have introduced congestion charging,16 as have Melbourne17 and London.18 In Singapore the extent of traffic congestion stimulated the authorities to introduce road pricing as part of a package of measures. Constraints on traffic in the central area include the prohibition of all large vehicles with three or more axles during peak hours, increased parking fees, a revision of parking regulations to discourage commuter traffic but facilitate shortterm business and commercial trips, and the Area Licensing Scheme (ALS).19 The latter, introduced in 1975, designates the centre of the city as a restricted travel zone. Between 7.30 a.m. and 10.15 a.m. only specially licensed vehicles are permitted into the zone. Licences are sold according to a sliding scale that ensures that company-owned vehicles pay the highest rate. Car-pool vehicles (defined as those carrying four or more persons) are exempt. Table 13.5 indicates the effect of the ALS on the modal split choice for work trips. In 1994 the ALS was extended to a full day, resulting in an immediate reduction of 9 per cent in traffic entering and leaving the restricted zone. In 1998 the ALS was replaced by a full system of electronic road pricing with ‘congestion fees’ deducted automatically as vehicles cross sensors installed along the route. The fee charged varies with the level of congestion, making it the world’s first scheme to pass on real-time congestion charges to motorists.20 Other traffic-management policies in Singapore include: 1. a 140 per cent tax (Additional Registration Fee) on all car imports; 2. an annual registration disc costing up to £1,500; 3. restrictions on the number of new cars permitted. From 1990 no new cars can be bought unless the purchaser has a government Certificate of Entitlement to Purchase, which is obtained via a public auction. The government decides each quarter on how many new cars it will allow, based on the number scrapped and road capacity, and potential owners submit a bid, with the number of successful bids equivalent to the number of new cars permitted. The price of the certificate is determined by the value of the lowest successful bid.21 Advocates of road pricing favour it as an example of using market forces to achieve policy ends. Rather than mandating modification of travel directly, road pricing encourages it through monetary disincentives. Objections to road pricing have centred upon equity, political and practical issues. The last of these is the least serious constraint. The practical difficulties of implementing a system of road pricing can be overcome without excessive cost, and electronic toll-collection technology now makes it feasible to levy charges without interrupting traffic flow. In Orange County CA, one- and two- Urban geography 372 occupant cars are charged for using high-occupancy vehicle (HOV) lanes constructed with private funds down the median of an existing highway (Route 91). One of the strongest arguments against road pricing is that it could discriminate against lowerincome car owners. Most significant, however, has been political opposition by the ‘motoring lobby’ and the general recognition by governments that the social benefits of road pricing could be offset by the political costs (Box 13.4). Nevertheless, the problem of urban traffic ensures that road pricing (or congestion pricing) remains on the political agenda. Auto restraint Road pricing is, in effect, one kind of auto restraint policy in that one of its goals is to promote the use of public transport by making car travel more difficult. Other restraint policies include prohibiting on-street car parking and the provision of additional downtown parking areas. The ultimate form of auto restraint is to ban cars from a section TABLE 13.5 CHANGES IN MODAL SPLIT OF WORK TRIPS IN THE RESTRICTED ZONE, SINGAPORE (%) Mode Inbound home-to-work trips Outbound work-to-home trips Before ALS Before ALS After ALS After ALS Bus 33 46 36 48 Car 56 46 53 43 Car driver 32 20 35 23 8 19 5 12 16 7 13 8 Motor cycle 7 7 7 6 Other 4 2 4 3 Car pool Car passenger Source: L.Wang and T.Tan (1981) Singapore, in M.Pacione (ed.) Problems and Planning in Third World Cities London: Croom Helm, 218–49 BOX 13.4 The political feasibility of high-occupancy vehicle lanes High-occupancy vehicle (HOV) lanes may be provided either by conversion of one or more existing general-purpose lanes or by the construction of a new lane. The add-a-lane strategy generates less opposition from motorists. The convert-a-lane approach meets ISTEA’s goals by improving travel conditions for HOVs and decreasing the capacity for single-occupancy vehicles (SOVs). Inevitably, this leads to dissatisfaction among SOV users Opposition to this efficiency measure was illustrated in 1977 by the court decision Urban transportation 373 in the case of the Santa Monica freeway HOV. Faced with chronic congestion on a 20-km section of the ten-lane freeway, California’s Department of Transport (CALTRANS) converted one of five lanes in each direction into an HOV lane. The initial increase in congestion in the four general traffic lanes led to a lawsuit against the project, and twenty-one weeks after its start, a judge halted it and requested additional environmental impact studies. CALTRANS responded by dropping the project, claiming that convert-alane schemes were ‘not politically feasible’. of the city, either wholly or partially, by prohibiting the entry of certain vehicles (e.g. according to licence plates) on certain days. Historic European cities (such as Vienna, Milan and Gothenburg) have limited traffic penetration in the interests of the liveability of the city centre.22 In Gothenburg the central city is subdivided into five sectors, and while traffic is permitted entry within any sector, crosssector flows are prohibited so as to exclude through traffic. Ride-sharing Three-quarters of the 84 per cent of the American labour force that travel to work by car are solo drivers. Nationally the average private vehicle in the USA transports only 1.15 persons on its trip to the workplace. One way of increasing highway capacity is for transport planners to implement programmes to encourage ride-sharing through car- and vanpooling schemes. Research suggests that to maxi mise the possibility of adoption, such schemes should be targeted at particular populations. Many people are not amenable to giving up the freedom of solo driving or are constrained from participating by other daily activity patterns such as collecting children or shopping. It appears that the most effective programmes are those arranged through employers. Evidence from Phoenix suggests that employees favour van pools over car pools because their private cars need not be used, they need not drive in rush-hour conditions, the arrangements are made for them and they have an opportunity to socialise with colleagues.23 Assisting employers to set up van pooling as an employee benefit is thus often more cost-effective than media campaigns aimed at the general public. Although all the strategies considered can contribute to ameliorating the urban transport problem, no single policy is sufficient to promote optimum use of the urban transport system. What is required is a package of self-reinforcing strategies that are integrated into the overall planning process for the metropolitan area. This means considering both transport and non-transport options. NON-TRANSPORTATION INITIATIVES Non-transport strategies seek to capitalise on the relationship between transportation, daily activity patterns and the aggregate land-use structure of the city. Two main approaches focus on patterns of work and on urban structure respectively. Urban geography 374 Transport and work patterns The problem of the commuting peak may be alleviated by promoting alternative work schedules (Table 13.6) which spread the journey-to-work period or reduce the number of workdays per week. Advances in telecommunications technology can also reduce the need for some trips during the business day and in some cases allow employees to work from home or from regional suburban ‘work stations’. The principle of flexible working hours has been applied widely in West Germany, where more than half the labour force is permitted to work nonstandard hours. The potential benefits for urban traffic movement may be considerable. For example, a study of the San Francisco Oakland Bay bridge traffic corridor found that even small increases in TABLE 13.6 TYPES OF ALTERNATIVE WORK SCHEDULES 1. Flexible work hours (‘flexitime’). Employee chooses his or her own schedule, with some constraints. Employee may be free to vary the schedule daily, vary the lunch breaks, or ‘bank’ hours from one day to the next or one pay period to the next. Typically all employees are required to be at work five days per week during designated ‘core’ periods (usually 9.30 a.m. to 3.30 p.m.) 2. Staggered work hours. Employee works a five-day week, but starting and ending times are spread over a wider time period than usual. Individual employee schedules are assigned by management; employees do not choose them 3. Four-day (or compressed) work week. Employee works the same total number of hours as in a typical five-day work week, but reports to work only four times per week. The four days may be the same each week/ or the extra day off may rotate from week to week 4. Job sharing/part-time work. Employee works less than the standard forty-hour work week, either by working fewer than five eight-hour days, or by working less than eight hours per day. Job-sharing means that two or more people share the same office space and work responsibilities Source: D.Plane (1986) Urban transportation, policy alternatives, in S.Hanson (ed.) The Geography of Urban Transportation New York: Guilford Press, 386–414 the number of flexitime workers can have substantial impacts in alleviating peak-period congestion.24 To date, however, urban transportation planning agencies have devoted only limited effort to promoting such strategies.25 The impact of telecommunications on traffic mitigation is slight as yet, with 1.4 per cent of Californian workers telecommuting on a typical workday, and 6.6 million American telecommuters in 1992,26 although Nilles (1991 p. 425) estimated that by 2011 ‘there could be as many as 50 million telecommuters in the US, saving as much as 380 billion passenger-miles of transportation use’.27 The number of telecommuters in Tokyo is expected to grow to between 9 million and 14 million by 2010, reducing pressure on roads and public transport systems with associated savings of up to 25 per cent of annual spending on public transport.28 On the other hand, a danger is that decentralised teleworking or even telecommuting centres could reinforce the trend towards sprawling Urban transportation 375 edge cities, since living on the urban fringe or beyond would not be inconvenient from a commuting viewpoint. Transport and urban structure As we have seen, a symbiotic relationship exists between transport and urban structure. Thomson (1977) identified four general urban transport strategies based on the degree of car-ownership to be accommodated29 (Figure 13.4). Although no city fits any single strategy, these archetypes offer general guidance on possible transport options for different forms of city: 1. Full motorisation. Since the motor car has many virtues as well as drawbacks, one approach is to structure the city to allow full use of the car. Small towns of up to 250,000 population can usually achieve this by upgrading radial routes and constructing inner ring roads around the town centre with adjoining car parks, but in large towns the provision of sufficient roads and parking space becomes impossible unless one abandons the traditional monocentric city. If there is to be a city centre it will be smaller and could not fulfil a traditional multifaceted role, with employment, shopping and entertainment activities dispersed to outlying centres. The primary goal is to maintain high levels of car accessibility throughout the metropolitan area. Los Angeles comes closest to this model. 2. Weak-centre strategy. In cities where the decentralisation force has proved more dominant than those favouring the presence of a strong city centre, the traditional downtown has diminished in importance to become little more than another suburban centre. The archetype of this weak-centre strategy is a city with a radial road network and additional commuter rail service serving a small central area. The combination of ring and radial routes attracts industrial and commercial development to the interchange points and generates growth of strategic suburban centres. Boston provides a good example of the type. 3. Strong-centre strategy. The world’s greatest cities owe much of their power to the concentration of activity in central areas which were established before the automobile began to exert an influence on urban structure. In such cities the transport system must be designed to maintain the strength of the centre. For this reason, the archetype comprises a radial network of road and rail routes, without high-speed orbital roads except close to the city centre itself. The dominant position of the central area will require a highcapacity public transport system, such as a short-stage, highfrequency underground railway, to distribute the large numbers of people entering the zone (Figure 13.4). This ideal is closely reflected in the urban structure of Tokyo. Urban geography 376 Figure 13.4 Models of the relationship between transport policy and urban structure Source: M.Thomson (1977) Great Cities and their Traffic Harmondsworth: Penguin 4. Low-cost strategy. Many cities are unable or unwilling to pay the high cost of these strategies. In cities where a minority own a private car, substantial public expenditure on new roads or rail systems cannot be justified, socially or economically. Traffic problems require a low-cost approach that maximises the use of existing infrastructure through improved management. The archetype consists of a highdensity city with a major centre served by numerous bus and tram corridors in which nonresidential activities are concentrated. Hong Kong provides an example of this urban form. Urban transportation 377 TRANSPORT AND SUSTAINABLE URBAN DEVELOPMENT As we have seen, the automobile has been a major agent underlying urban sprawl. Proponents of sustainable urban development advocate greater use of public transport and a more comprehensive approach to planning which acknowledges the fundamental relationship between transport and urban form. Common to most perspectives on sustainable urban development is the view that a solution to urban transport problems requires greater mixeduse and higher-density development, as well as integration of transport considerations directly into land-use planning with the aim of enabling individuals to sustain their mobility, but to do so with fewer vehicle trips. In the Netherlands the urban transport problem is addressed within the framework of a national plan for sustainable development. The major aims are to ensure that: 1. vehicles are as clean, quiet, safe and economical as possible; 2. the choice of mode of passenger transport must result in the lowest possible energy consumption and the least possible pollution; 3. the locations where people live, work, shop and spend their leisure time will be coordinated to minimise the need for travel. These goals are to be approached via: 1. a series of measures to convert vehicles to clean running by, for example, establishing targets for the reduction of exhaust emissions; 2. a shift from single-occupancy vehicles to public transport for longer journeys or cycling and walking for shorter trips, by means of improved transit systems and restrictions on car use; 3. concentration of residences, work areas and amenities to produce the shortest possible trip distances. Public mass transit plays an important role in a number of cities that may offer lessons for elsewhere. Stockholm provides an example of the coordinated planning of rail transportation and urban development. Half the city’s 750,000 inhabitants live in satellite communities linked to the urban core by a regional rail system. This transit-oriented multicentred built form is the outcome of postSecond World War regional planning that targeted population growth into rail-served new towns in a pattern prompted by Howard’s garden-city principles. The contrast between the pattern of post-war urban growth in Sweden and that in the USA is significant. Like the USA, Sweden is one of the world’s most affluent countries (with GDP per capita of US$17,900 in 1990) and has a high rate of car ownership (420 cars per 1,000 inhabitants), yet suburban growth follows a radically different path as a consequence of government intervention in urban planning at the regional level. Tokyo is a powerful example of a ‘strong centre’ city, and without the contribution of rail mass transit, road congestion would undermine the functioning of the CBD. Over half the 1,250 miles (2,000 km) of suburban railway lines have been constructed by privatesector consortia that link mass transit with new-town development (Box 13.5). Tokyo’s rail-oriented urban expansion is also promoted by various national and local government policies aimed at reducing car use. These include a range of taxes on car-ownership purchase tax, an annual registration tax, a surcharge based on vehicle weight, high Urban geography 378 gasoline taxes, toll roads, and a requirement to provide off-street parking at one’s residence—and incentives to promote transit riding, such as a tax-free commuting allowance of US$500 per month from employers which is fully deductible against corporate income taxes. The existence of a high-capacity public transportation infrastructure in Tokyo has served to ameliorate some of the problems of congestion and overcrowding within the dominant CBD. However, the development of a polycentric metropolitan structure, with both planned and unplanned subcentres and satellite towns, has resulted in dispersed population and employment growth and increased levels of car dependence in the outer suburban areas beyond the dense transit network. Central co-ordination of transport and urban development is seen at its strongest in Singapore, where near-complete control over urban growth and design, and a pro-transit government, has BOX 13.5 A transit-oriented community: Tama new town, Japan Tama new town is one of several publicly sponsored new towns in the Greater Tokyo region; others include Tsukuba Science City, Ryugasaki, Chiba and Kohoku. Tama new town is a joint venture of the Tokyo metropolitan government and the national Housing and Urban Development Corporation. The town currently has 170,000 inhabitants and 35,000 jobs, with a target population and labour force of 360,000 residents and 130,000 workers respectively. A 1991 survey found that 20 per cent of the employed residents had jobs within the town; of the remaining 80 per cent leaving the town for work two-thirds travelled to central Tokyo, with 70 per cent commuting by rail. The new town is designed around twenty-one residential areas (neighbourhoods) each containing a centrally located junior high school and two primary schools. There is a variety of housing, with public housing consisting of mid-rise and high-rise apartments sited near rail stops and targeted at low- and middleincome households. These are surrounded by privately built single-family detached units sold at market rate. Four of Tama new town’s railway stations form a focus for an urban centre comprising a retail plaza, offices, banks and institutional uses. These urban centres are notable for the conspicuous absence of park-and-ride lots. Even though most Tama residents own a car, they bus-and-ride, walk-and-ride or bikeand-ride instead. Source: M.Bernick and R.Cervero (1997) Transit Villages in the Twenty-First Century New York: McGraw-Hill Urban transportation 379 Figure 13.5 The mass rapid transit system and concentrated decentralisation in Singapore BOX 13.6 Pleasant Hill CA: transit village As the Bay Area Rapid Transit (BART) train travels through Contra Costa county, it passes a series of post-Second World War suburbs: Orinda, Moraga, Lafayette, Walnut Creek, Concord. Here BART is enveloped in a sprawling environment of single-family homes and duplexes, one- to two-storey commercial buildings, and lots of hillside open space. The notable exception, however, is what until recently was the next to the last station on the line: Pleasant Hill. At Pleasant Hill are the beginnings of a new type of suburban community: one with a mix of clustered housing, mid-rise offices, small shops, a hotel and a regional entertainment complex, all huddled around the transit station, and with a street configuration and landscaping that encourage people to walk to and from the station. Housing, the primary land use in the area, lies slightly beyond the station core. In all, over 1,600 residential units built in the past five years lie within a three- to five-minute walk of Pleasant Hill station. Office and commercial developments are located even closer to the station. Pac Tel Corporate Plaza, housing the regional administrative offices of the telecommunications giant, is less than 100 yards (30 m) from the station entrance. At the station core much of the 11 acre (4.5ha) BART parking area has been converted into an entertainment and retail complex. The Pleasant Hill station area illustrates an alternative to the low density car oriented Urban geography 380 suburban development that characterises virtually all of Contra Costa. It is a radical departure from business as usual in the suburbs. It is a transit-oriented development, or what local planners and elected officials refer to as a ‘transit village’. Source: M.Bernick and R.Cervero (1997) Transit Villages in the Twenty-First Century New York: McGraw-Hill Figure 13.6 Plan of a transitoriented community produced one of the most efficient transit-land use symbioses in the world.30 The initial Concept Plan (1971) provided for decentralisation of population from overcrowded inner urban areas to a series of high-density new towns served by an efficient transit system comprising the Mass Rapid Transit (MRT) and feeder bus lines. The 1991 revised Concept Plan envisaged future urban growth based on further development of new towns in a hierarchy of centres linked by the MRT in a pattern of ‘concentrated decentralisation’ (Figure 13.5). The concept of transit-oriented development has also emerged in the USA, where a ‘new urbanism’ is critical of the auto-oriented post-Second World War suburbs.31 This advocates a type of suburban community designed in neo-traditional terms with a mix of land uses, moderate residential densities, pedestrian circulation, and small offices and shops clustered around a transit station (Figure 13.6). The first such community was developed at stations on the BART system (Box 13.6), but similar nodes have grown up along other transit routes,32 including the line of the Washington Metrorail system in suburban Virginia and Maryland. Since being linked to the Metrorail network in 1979, the transit village of Ballston 6 miles (10 km) north-east of Tyson’s Corner has grown into one of Arlington County’s ‘new downtowns’ and has been characterised as a ‘postedge city’. Transit-oriented development may also aid regeneration of decayed inner-city Urban transportation 381 areas, as at Fruitvale in Oakland, where a run-down commercial district has been transformed into a transit village by a combination of grassroots community action financed by federal and foundation grants which pump-primed sufficient development to attract private-sector investment.33 However, although transit villages reflect many of the aims of sustainable urban development, they are likely to appeal to only a small proportion of the population, the majority of which will continue to favour the lifestyle associated with the American dream of a dispersed single-family residence on a large lot. FURTHER READING BOOKS M.Bernick and R.Cervero (1997) Transit Villages in the Twenty-First Century New York: McGraw-Hill R.Cervero (1998) The Transit Metropolis Washington, DC: Island Press I.Docherty and J.Shaw (2003) A New Deal for Transport Oxford: Blackwell S.Hanson (1995) The Geography of Urban Transportation New York: Guilford Press M.Thomson (1977) Great Cities and their Traffic Harmondsworth: Penguin R.Tolley and B.Turton (1995) Transport Systems, Policy and Planning London: Longman V.Vuchic (1999) Transportation for Livable Cities New Brunswick, NJ: Rutgers University Press JOURNAL ARTICLES R.Cervero and K.-L.Wu (1998) Sub-centring and commuting Urban Studies 35(7), 1059–76 J.Constantino (1992) The IVHS strategic plan for the United States Transportation Quarterly 46, 481–90 P.Gurstein (1996) Planning for teleworking and home-based employment Journal of Planning Education and Research 15, 212–24 C.Pooley and J.Turnbull (1999) The journey to work: a century of change Area 31(3), 281–92 A. Sorensen (2001) Subcentres and satellite cities: Tokyo’s twentieth-century experience of planned polycentrism International Planning Studies 6(1), 9–32 C.Willoughby (2001) Singapore’s motorization policies 1960–2000 Transport Policy 8, 125–39 KEY CONCEPTS ■ commuting ■ traffic congestion ■ supply-fix approach ■ transport-system management ■ mass transit ■ smart highway ■ road pricing ■ congestion fee ■ auto restraint ■ ride-sharing Urban geography 382 ■ telecommuting ■ full motorisation strategy ■ weak-centre strategy ■ strong-centre strategy ■ transit-oriented development ■ transit village ■ post-edge city STUDY QUESTIONS 1. Identify the principal dimensions of the urban transport problem. Relate these to your own experience of urban travel. 2. Examine the view that the best way to relieve traffic congestion in towns is to build more highways. 3. Make a list of the main advantages and disadvantages of private (car-based) transport and public (bus or rail) mass transit in cities. 4. With reference to particular examples, critically examine attempts by urban authorities to limit the use of private cars in cities. What is your view of the desirability and feasibility of such strategies? 5. Select one of the major models of the relationship between transport and urban structure discussed in the chapter. Explain its rationale and identify a city to which the model may be applied. 6. Think about the possible impacts of telecommuting on urban structure. How do you see these affecting patterns of urban transport and urban form by the years 2010, 2025, 2050? PROJECT Efforts to relieve urban traffic congestion by increasing the ridership of public transport are undermined by commuters’ reluctance to leave their cars. Select one mode of public transport available in your city (e.g. commuter rail, bus or Underground) and undertake a survey of passengers to identify who uses the system and why, and the perceived advantages and disadvantages of this form of travel. PART FOUR Living in the City: Economy, Society and Politics in the Western City 14 The Economy of Cities Preview: the world/global economy; transnational corporations and the state; new industrial spaces; deindustrialisation and tertiarisation; the nature of work and the division of labour; unemployment; informalisation; urban change within the global economy; world cities; transnational urban systems; the post-war restructuring of the UK space-economy INTRODUCTION The focus of this chapter is the urban economy. To understand fully the economy of cities we must locate our discussion within the framework of the wider phenomenon of postSecond World War global economic restructuring. This process is itself related to the more general concept of globalisation (see Chapter 1). In the context of urban economic change, particular importance is attached to the process of economic globalisation. The driving force behind economic globalisation is capitalism. This is evident in: 1. the growth of multinational and transnational corporations (TNCs); 2. the expansion of trade and foreign direct investment (FDI); 3. the emergence of a new international division of labour (NIDL); 4. the enhanced mobility of money capital across international boundaries; 5. the globalisation of markets for consumer goods; 6. the intensification of international economic competition with the rise of newly industrialising countries (NICs). At the urban scale, economic globalisation is manifested on the one hand in the rise of global cities, and in the deindustrialisation and decline of older industrial cities on the other. Understanding the nature of the global economy is a prerequisite for understanding the changing economy of cities. We begin our discussion in this chapter by defining the concepts of a global/world economy and identify the changing structure of the global/world economy in the post-war era. We examine the role of TNCs and the state within the global economy and the rise of new production systems and industrial spaces as part of the transition to advanced capitalism. Deindustrialisation, tertiarisation, the changing nature of work and the division of labour, and problems of unemployment and informalisation are also discussed. The growth of world cities and of a transnational urban system is examined before attention is directed to the postwar restructuring of the UK space-economy, and parallel developments elsewhere. Urban geography 386 THE WORLD/GLOBAL ECONOMY We can differentiate between a global economy and a world economy. The latter is an economy in which capital accumulation in the form of trade between firms based in different nation-states proceeds throughout the world, and has existed in the West since the sixteenth century.1 A global economy is one with the capacity to work as a unit in real time on a planetary scale.2 Only with the advent of new information and communications technologies in the late twentieth century has the world economy become global, primarily in the BOX 14.1 The advent of megabyte money Money was invented to replace barter in the exchange of goods and services. For centuries, money has served as a medium of exchange, a common unit of account and a store of value. In recent decades it has also become a commodity to be bought and sold. This new function of money was created by the US government’s decision in 1971 to take the dollar off the gold standard and allow exchange rates to float in the international markets. With the collapse of the Bretton Woods agreement, which had pegged the value of the US dollar to gold (and other major currencies to the dollar at a fixed exchange rate) the value of national currencies was determined only by investor confidence. In 1995 US$1.2 trillion was exchanged every day in the currency markets, and 95 per cent of all currency transactions in the world are motivated by speculation. Time is critical to profitmaking in the international money markets; it is the speed of the transaction, sometimes preprogrammed into a computer, that generates a profit or a loss, with the same capital often switching back and forth between economies in seconds. The creation of electronic ‘megabyte money’ was made possible by advances in telecommunications which link dealers in real time within the international financial system.3 The advent of money as a commodity has decoupled the ‘real economy’ from the ‘money economy’, the first and smaller of the two being where products are made, trade is conducted, research carried out and services are rendered. The other, far larger money economy is concerned with speculation and the exchange of credits and debts. For the money markets, the long-term economic health of the real economy is incidental to the primary goal of electronic speculators, which is to realise a profit during the working day. This short-termism can have an adverse effect on the operation of the real economy, with companies affected by rising interest rates, increasing commodity prices and declining exports due to a strong domestic currency. In extreme circumstances, companies and their jobs may be undermined regardless of performance, because of sudden changes in the financial environment in which they operate. context of the international financial system (Box 14.1). The main features of the capitalist world/global economy may be summarised as follows: 1. It comprises a single world market within which production is for exchange rather than use, with prices fixed by the self-regulating market. This means that for a period of The economy of cities 387 time, more efficient producers can undercut others and increase their market share to the detriment of other producers and places. 2. Territorial divisions between states in the world economy lead to a competitive state system in which each seeks to insulate itself from the rigours of the world market (e.g. by trade barriers) while attempting to turn the world market to its advantage (e.g. by offering tax incentives to incoming industries). 3. The modern world economy comprises a core (characterised by relatively high incomes, advanced technology and diversified production) and a periphery. The core needs the periphery to provide the surplus to fuel its growth. Core-periphery relationships exist at all geographical scales, including, for example, differences in levels of prosperity between the ‘sunbelt’ towns of Southern England and the older industrial cities of Northern Britain. 4. The world economy has followed a temporally cyclical pattern of growth and recession characterised by Kondratieff cycles. These alternate phases of growth and stagnation show a high correlation with technological change and emphasise the regenerative effects of technological innovation. The first Kondratieff expansion phase (c. 1790– 1815) was associated with the period of the original industrial revolution (with the introduction of mechanised textile production and improved iron production), while the fifth and latest highlights the role of micro-electronics (in, for example, computing, industrial process control and telecommunications) in transforming the economy. 5. Finally, it is important to reiterate that every part of the world has its own particular relationship to the world economy based on locally specific modes of socio-economic organisation comprising a complex of management strategies, skill requirements, firm structures, labour organisation, infrastructural investments, consumption norms and government policies—all of the institutions and institutional arrangements that impinge upon economic production, investment, consumption and employment. THE WORLD ECONOMY IN THE POST-WAR ERA The end of the post-war industrial boom in the early 1970s heralded a crisis for industrial capitalism. The system shock precipitated by the quadrupling of petroleum prices in 1973 as a result of the OPEC cartel has been indicted as a major cause of the economic downturn. However, as Hamilton (1984) explained, a number of other trends originating in the previous decade or before were also responsible for the recession.4 These included the following: 1. A slowdown in economic growth and falling profits in the industrial core countries of the OECD was associated with falling levels of demand for capital goods (particularly in transport, steel and building). 2. Rising levels of inflation, which reduced profits and hampered capital accumulation, led to greater dependence on the banking sector for investment funds. High interest rates, however, restricted technological investment and so hindered competitiveness. Inflation also raised labour costs. The net outcome was the widespread depression of both capital-intensive (e.g. steel, ship-building, vehicles) and labourintensive industries (e.g. textiles, clothing). 3. Increased international monetary instability, which took two forms: Urban geography 388 ■ under- or overvaluation of exchange rates as a result of the transition in the early 1970s from fixed exchange rates to floating exchange rates (Box 14.1). In countries whose currencies were undervalued (e.g. Germany and Japan), industrial production was stimulated by increased demand for exports, whereas in those countries with overvalued currencies (e.g. oil and gas producers such as Norway and the UK) the loss of international competitiveness led to import penetration and a consequent decline in domestic industrial capacity; ■ indebtedness among NICs and some underdeveloped countries as a result of massive borrowing from the ‘petrodollar’ surpluses of the OPEC states. This stimulated debtor countries to increase exports of cheap manufactured goods to the core regions in order to obtain the necessary foreign exchange. This increased the competitive pressure on the labour-intensive sectors of the core economies. 4. The growth of new social values related to social welfare and environmental protection increased industrial costs and contributed to a higher tax burden for both producers and consumers. 5. The introduction of technological innovations in response to escalating energy and labour costs led to reduced demand in some traditional industrial sectors. For example, energysaving designs in transport cut demand for steel, while innovations in microelectronics reduced demand for electro-mechanical products. 6. A resurgence of political volatility (e.g. in the Middle East and South-East Asia) reduced the area of stable business and constrained world trade. 7. The increasing intensity of international competition arose from the post-war liberalisation of trade, the spread of industrialisation to the periphery, the aggressive role of governments in NICs, and the post-1970 stagnation of world markets. The net effect of these trends was to initiate a process of industrial reorganisation and structural change that signalled a transition from industrial capitalism to post-industrial or advanced capitalism (alternatively described as a move from the fourth to the fifth Kondratieff cycle). TRANSNATIONAL CORPORATIONS AND THE STATE IN THE GLOBAL ECONOMY Transnational corporations are major actors in the post-industrial global economy, directing investment and production on a worldwide basis with only limited consideration for national boundaries. By the early 1990s there were 37,000 TNCs with over 170,000 foreign affiliates. This complex of corporations accounted for over US$2 billion of foreign direct investment and totalled over US$5.5 billion in worldwide sales.5 By 1999 there were 60,000 TNCs worldwide, with 500,000 foreign subsidiaries. The 100 largest non-financial TNCs control 15 per cent of global foreign assets and account for 22 per cent of total world sales for all TNCs. The economic power of some individual TNCs surpasses that of many nation-states. Of the eighty largest economic organisations in the world, thirtyfour (43 per cent) are TNCs, all with head offices in the USA, Japan or Europe. TNCs are key players in the global economy. The economy of cities 389 The economic and political power of TNCs is manifested spatially in the creation of export processing zones in Third World countries. Within these enclaves TNCs benefit from low wage costs, government tax concessions and often freedom from national labour legislation. EPZs are labourintensive manufacturing centres that involve the import of raw materials and export of factory products. Over ninety countries had established EPZs by the end of the twentieth century in response to problems of foreign indebtedness and as a cost-effective way to begin industrialisation. While the most common locations have been coastal, as in the case of China’s special economic zones, one of the largest areas of EPZs is along the USA Mexico border (see Chapter 5). A feature of the capitalist global economy is that because the major economic actors operate within an area larger than that which any political entity can control, capitalists— financial institutions and TNCs—have a freedom to manoeuvre that is structurally based. The exercise of this freedom by ‘footloose’ capital can result in tension between them and territorially bounded national governments, and lead to adverse consequences (e.g. redundancies and economic decline) for less competitive places. It is important to realise, however, that while the growth of a global economy has diminished the power of national governments to control their domestic economy, economic globalisation has not eliminated the economic significance of the nation-state. On the contrary, the deregulation and increasing global integration of financial markets (and corresponding loss of national autonomy) were promoted by nations that stood to gain most from a ‘free market’ in global finance.6 The involvement of the state in the global economy is also evident in governments’ attempts to aid ‘their’ firms in the global competition. Nowhere is this partnership between state and economy more strongly developed than in Japan, where the country’s successful internationalisation of its production and financial capital is based on a close partnership between the Japanese state (the Ministry of International Trade and Industry, MITI) and industry. The state (MITI) played a crucial role in reorienting the economy away from heavy industry into electronics, cars and other consumer durables for world markets, and in directing the flow of Japanese capital into the USA, Britain and mainland South-East Asia.7 Globalisation has altered rather than eliminated the economic role of the nation-state. Traditional economic policies managed within the boundaries of regulated national economies are increasingly ineffective, as key factors such as marketing policy, interest rates or technological innovation are highly dependent on global movements. However, while the power of global forces to influence urban growth or decline is profound, it is not omnipotent. Economic and political action undertaken at a national level or by regional groupings of nationstates can modify the effects of the global economic system. Regulatory and tax policies shape the environments that attract or repel investors; decisions about public investment determine whether infrastructure will be rebuilt or allowed to deteriorate; government procurement policy stimulates the private economy; and intergovernmental transfer payments can prevent the collapse of a local economy. Furthermore, the impacts of global processes are manifested in particular local contexts, the nature of which varies between places. Urban geography 390 NEW PRODUCTION SYSTEMS AND NEW INDUSTRIAL SPACES The flexible production systems of a post-Fordist economy and the technological innovation postulated as a basis for the fifth Kondratieff upswing come together in particular places or new industrial spaces variously referred to as technology parks, science cities or technopoles (Box 14.2). The development of technopoles results from the clustering of ‘specific varieties of the usual factors of production: capital, labour, and raw material, brought together by some kind of institutional entrepreneur, and constituted by a particular form of social organisation’10 (Castells 1996 p. 391). The classic example is the development of ‘Silicon BOX 14.2 A typology of technopoles Four main types of ‘technologically innovative milieux’8 can be identified: 1. There are industrial complexes of hightechnology firms in which R&D and manufacturing are linked. Some, such as Silicon Valley, are created entirely from the growth of new high-technology firms. Other complexes, such as Boston’s Route 128, develop through the transformation of old industrial regions. These techno-industrial complexes arise without deliberate planning. 2. Science cities are research complexes with no direct link to manufacturing. The intention is to generate a high level of scientific excellence through the synergy created by agglomeration. Examples include the Siberian city of Akademgorodok and the Japanese science city of Tsukuba. 3. Technology parks aim to induce new industrial growth, in terms of jobs and production, by attracting high-technology manufacturing firms to a privileged space. These emanate from governments (e.g. Hsinchu in Taiwan) or universities (e.g. Cambridge) initiatives. 4. Technopoles may also be instruments of regional development and industrial decentralisation, the major example being the Technopolis programme initiated in 1984 in Japan.9 Valley’ in Santa Clara CA. From being an agricultural area with a population of 300,000 in 1950 the area was transformed into the world’s most intensive complex of hightechnology activity with a population of 1.25 million in 1980.11 The raw material was knowledge produced by major centres of innovation such as Stanford University, Caltech or MIT. Labour came from the output of scientists and engineers from local universities. Capital was prepared to take the risk of investing in pioneering high technology either because of defence-related spending or for the potentially high rewards. The catalyst that energised these production factors was, in the case of Silicon Valley, the launching of the Stanford Industrial Park. Finally, over the years social networks of different kinds have contributed to the development of a local culture of innovation and dynamism that The economy of cities 391 attracts capital and brainpower from around the world and encourages the circulation of ideas and a spirit of business entrepreneurialism. The combined effect has been to make Silicon Valley a self-sustaining innovative milieu of high-technology manufacturing and services. Subsequent attempts have been made to duplicate the success of Silicon Valley by establishing technopoles in other parts of the world, including Munich (Germany), Grenoble (France), Cambridge and the M4 corridor to the west of London (England),12 Bangalore (India),13 Shenzen (China)14 and Malayasia’s multimedia super-corridor.15 The most comprehensive approach is Japan’s technopolis programme masterminded by the Ministry of International Trade and Industry (MITI). This aims to create twenty-six new science cities in the country’s peripheral areas, in order to simultaneously promote new technologies and develop lagging regions.16 At the intra-metropolitan scale new industrial clusters centred on innovative, knowledge-based, technologically intensive activities such as computer graphics and imaging, software design and multimedia industries as well as technologically ‘retooled’ industries such as architecture and graphic design have been identified as key components of an emerging ‘new economy’ of the inner city. Such firms are attracted to the metropolitan core by the creative habitat, potential for ‘knowledge spillovers’ between companies, opportunities for social interactions across work and non-work life, and the cultural and environmental amenities of the locale. Examples of these new production spaces include Telok Ayer in Singapore, Multimedia Gulch in San Francisco, New York’s Silicon Alley and the inner East End of London. While the nature of the new industrial clusters varies with local context general positive effects can include: 1. a contribution to the regeneration of local economies, with growth in employment and high-value output; 2. preservation of heritage buildings through adaptive re-use; 3. increased vitality of inner-city neighbourhoods; 4. positive regional growth effects. On the down side, critics have pointed to negative impacts similar to those emanating from the development of cultural-industry quarters (see Chapter 16). These include: 1. community dislocation due to physical encroachment of new industrial activities; 2. displacement of existing businesses and residents unable to compete with the new industries’ ability to pay enhanced rents; 3. increased social polarisation due to gentrification and lifestyle differences between established residents and incoming workers. DEINDUSTRIALISATION AND TERTIARISATION The growth of a global economy is part of a transition to advanced capitalism. This change has been depicted as a shift from a post-war Fordist paradigm to a post-Fordist regime of flexible accumulation, as shown in Tables 14.1 and 14.2. A major element in this transition has been the expansion of the service sector of the economy. Between 1960 and 1990 the service sector’s share of global output, measured as a percentage of GDP, rose from 50.6 per cent to 62.4 per cent (Table 14.3). By 2000 70 per cent of output in Urban geography 392 ‘high-income countries’ derived from the service sector (Table 14.4). This appears to be a universal phenomenon, although the scale of tertiarisation is relatively slower in developing countries, where the proportionately smaller size of the service sector is explained in part by the existence of a larger informal sector that fulfils many of the functions undertaken by the formal service sector in advanced economies17 (see Chapter 24). In terms of advanced or higher-order producer services the process of urban tertiarisation has led to the emergence of a number of ‘urban service corridors’ within Asian-Pacific city regions, including those centres on Seattle-Portland, Tokyo-Kyoto and Singapore-Kuala Lumpur.18 The related trend of deindustrialisation is a feature of developed economies (Tables 14.3 and 14.4). Whereas developing nations have witnessed manufacturing growth (in terms of both absolute and TABLE 14.1 THE POST-WAR ‘FORDIST’ EXPANSIONARY REGIME OF THE LATE 1940s TO EARLY 1970s Characteristic Key features Industry Monopolistic; increasing concentration of capital; steady growth of output and productivity, especially in new consumer durable-goods sectors; expansion of private and especially public services Employment Full employment: growth of manufacturing jobs up to mid-1960s; progressive expansion of service employment; growth of female work; marked skill divisions of labour Consumption Rise and spread of mass consumption norms for standardised household durables (especially electrical goods) and motor vehicles Production Economies of scale; volume, mechanised (Fordist-type) production processes; functional decentralisation and multinationalisation of production Labour market Collectivistic; segmented by skill; increasingly institutionalised and unionised; spread of collective wage bargaining; employment protection Social structure Organised mainly by occupation, but tendency towards homogenisation. Income distribution slowly convergent Politics Closely aligned with occupation and organised labour; working-class politics important; regionalist State intervention Keynesian-liberal collectivist; regulation of markets; maintenance of demand; expansion of welfare state; corporatist; nationalisation of capital for the state Space-economy Convergent; inherited regional sectoral specialisation (both old and new industries) overlaid by new spatial division of labour based on functional decentralisation and specialisation: regional unemployment disparities relatively stable Source: R.Martin (1988) Industrial capitalism in transition: the contemporary reorganisation of the British space-economy, in D. Massey and J.Allen (eds) Uneven Redevelopment London: Hodder & Stoughton, 202–31 The economy of cities 393 TABLE 14.2 THE ‘POST-FO RDIST’ REGIME OF FLEXIBLE ACCUMULATION FROM THE MID-1970S TO THE PRESENT Characteristic Key features Industry Rationalisation and modernisation of established sectors to restore profitability and improve competitiveness: growth of high-tech and producer service activities and small-firm sector Employment Persistent mass unemployment; general contraction of manufacturing employment, growth of private service-sector jobs; partial defeminisation (in manufacturing); flexibilisation of labour utilisation; large part-time and temporary segment Consumption Increasingly differentiated (customised) consumption patterns for new goods (especially electronics) and household services Production Growing importance of economies of scope; use of post-Fordist flexible automation; small-batch specialisation; organisational fragmentation combined with internationalisation of production Labour market Competitive; deunionisation and derigidification; increasing dualism between core and peripheral workers; less collective/ more localised wage determination Social structure Trichotomous and increasingly hierarchical; income distribution divergent Politics Dealignment from socio-economic class; marked decline of working-class politics; rise of conservative individualism; localist State intervention Keynesianism replaced by free-market Conservatism; monetary and supply-side intervention rather than demand stabilisation; deregulation of markets; constraints on welfare; self-help ideology; privatising the state for capital Space-economy Divergent; decline of industrial areas (pre- and post-war); rise of new high-tech and product services complexes; increasingly polarised spatial division of labour; widening of regional and local unemployment disparities Source: R.Martin (1988) Industrial capitalism in transition: the contemporary reorganisation of the British space-economy, in D. Massey and J.Allen (eds) Uneven Redevelopment London: Hodder & Stoughton, 202–31 TABLE 14.3 STRUCTURE OF WORLD OUTPUT, 1960–90 (% OF GDP) Agriculture Manufacturing Services 1960 1970 1980 1990 1960 1970 1980 1990 1960 1970 1980 1990 World 10.4 6.9 5.6 4.4 28.4 26.1 22.4 21.4 50.6 56.4 57.0 62.4 Industrial countries 6.3 3.9 3.4 2.5 31.0 27.7 23.6 21.5 51.6 58.3 60.2 64.7 USA 4.0 2.8 2.6 1.7 29.0 25.2 21.8 18.5 57.2 62.7 63.8 70.3 Urban geography Japan 394 13.1 6.1 3.7 2.4 35.1 36.0 29.2 28.9 40.9 47.2 54.4 55.8 8.8 4.8 3.6 3.0 34.7 30.5 24.0 21.5 43.8 54.3 59.2 63.8 Developing countries 31.6 22.4 14.7 14.6 15.6 17.8 17.6 20.7 42.8 48.1 44.2 49.8 Latin America and Caribbean 16.5 11.9 9.5 8.5 21.1 22.8 23.5 22.3 50.6 55.3 54.5 57.5 Africa 45.8 33.1 25.1 30.2 5.8 8.6 7.7 10.9 37.5 44.0 39.9 41.6 Asia South and SouthEast 38.0 23.9 17.8 17.4 14.4 15.1 15.4 20.9 37.1 42.5 37.9 45.9 Asia 44.0 35.6 24.7 18.9 13.7 15.6 19.8 22.9 36.2 41.9 43.5 46.6 Europe Source: International Labour Organisation (1995) World Employment 1995 Geneva: ILO TABLE 14.4 WORLD OUTPUT BY SECTOR AND MAJOR REGIO N, 2000 (%) Agriculture Industry Services World 3.9 29.8 66.3 High-income countries 1.9 28.8 69.5 European Union 2.4 29.2 68.4 UK 1.1 28.7 70.3 USA 1.6 24.9 73.5 Japan 1.4 31.8 66.8 Middle-income countries 9.7 35.3 55.1 25.4 30.9 43.7 Low-income countries Source: World Bank World Development Indicators relative output and employment), the ‘industrial nations’ have experienced a significant decline in manufacturing’s share of total output and employment. In large part this trend reflects the replacement of the old international division of labour, based on sector differentiation (with, for example, primary-sector activity in the developing nations and higher value-added stages of manufacturing in developed nations) by a new international division of labour based on the separation of functions (in which the control and command functions are located in a network of global cities in the developed nations while physical production is increasingly dispersed to a host of developing countries where new technology can be allied to lower labour costs). Most recently the NIDL has affected The economy of cities 395 Plate 14.1 Suburban commuters on their way to work flow past tourists on Westminster Bridge, London the tertiary sector of the economy with the advent of ‘offshoring’ whereby companies move whitecollar jobs abroad (Box 14.3). The expansion of the service sector is evident in cities at all levels of a national urban system from those catering for local-regional markets to others serving a global market. In many cities a new economic core of banking and service activities has emerged to replace an older core based on manufacturing. At the heart of this new urban economy are the producer services provided to firms (e.g. legal, financial, advertising, consultancy and accounting services). Paradoxically, despite their use of the most advanced information technologies, BOX 14.3 Offshoring: a new international division of labour The impact of offshoring on manufacturing industry in the UK and USA is well documented. Now the trend is moving up the management ladder as globalisation and technology make it possible to move white-collar work to wherever the costs are lowest. In recent years many callcentre jobs have gone overseas to cities (such as Bangalore, India), with an educated Englishspeaking low-cost work force. Some see this process as an inevitable part of the creative destruction inherent in capitalism. It is argued that offshoring can be of benefit through: 1. reduced costs and larger profits that may be reinvested in more innovative enterprises at home; Urban geography 396 2. new and expanding subcontractors abroad create new markets; 3. urban economies are forced to develop more dynamic sectors. A report from Deloitte Consulting estimated that 2 million jobs in the financialservices sector could migrate to lower-cost countries, such as India, in the next few years, including 730,000 from Europe. Looking further ahead, capital’s continual search for lowest costs will inevitably see the tide of offshoring wash over countries like India as economies, such as that of China, acquire a competitive edge in the global marketplace. Source: Sunday Times 5 October 2003 producer services tend to be concentrated spatially in the downtowns of major cities. For example, over 90 per cent of jobs in finance, insurance and real estate (FIRE) in New York are located in Manhattan, as are 85 per cent of business service jobs.19 This concentration in high-cost central cities is explained by the agglomeration economies provided by such locations, and the interdependence among highly specialised and innovative service providers, and the desire to maintain close linkages with client firms. This sectoral transformation in the economy is reflected in the changing nature of work in advanced capitalist society. THE NATURE OF WORK AND THE DIVISION OF LABOUR The processes of deindustrialisation and tertiarisation affect the mix of job opportunities available to urban residents. Fundamentally, whereas the massproduction techniques of Fordism did not require a well educated workforce, in the post-Fordist era ‘only a welltrained and highly-adaptable labour force can provide the capacity to adjust to structural change and seize new employment opportunities’ (OECD 1993 p. 9).20 The advent of post-Fordist production strategies and the drive by firms to attract a flexible and skilled labour supply is creating a coreperiphery division within the workforce. The core comprises multi-skilled and well-trained employees in highly paid and reasonably secure jobs, whereas the peripheral labour force is most vulnerable to layoff in times of economic downturn.21 In labourmarket terms this indicates a growing gap between what Drucker (1993) has referred to as knowledge workers and the rest.22 The division of labour according to ‘occupational status’ (based on educational standards and occupational qualifications) is the principal form of labour market segmentation. Other important divisions of labour in postindustrial society include those based on the following: 1. Full-time versus temporary work. Although taking temporary work can be a voluntary response to the demands of competing activities (such as domestic work by females), it is also often involuntary, arising because of: ■ employers’ attempts to develop a ‘flexible’ labour force and reduce costs by paying lower wages to temporary workers and avoiding the health, pension and other benefits attached to permanent contracts; The economy of cities 397 ■ the low wage levels in part-time employment, which require people to take on more than one job. In the USA in 1994 7 million people (6 per cent of the workforce) held 15 million jobs. 2. Gender. The ‘traditional’ notion of waged work as a man’s world and the home as the place for women has been overturned by the growth of female employment which has accompanied deindustrialisation and the rise of a servicebased economy. In the UK women accounted for 38 per cent of the labour force in 1972 but 50 per cent in 1996. By the latter date 70 per cent of working-age women were engaged in the formal labour market. Despite the move towards equalisation of male and female economic activity rates, however, most women’s experience of employment remains qualitatively different from that of most men. This is reflected in: ■ a higher involvement in part-time work. By 1995 part-time work accounted for 27 per cent of all officially recorded jobs in the UK, with women occupying 80 per cent of these; ■ the concentration of women in certain industrial sectors such as the service sector; ■ income differentials, which may be explained in part by the female concentration in part-time low-status employment. Even in full-time employment, however, women receive, on average, 75 per cent of the wage of a male employee, although the difference in part reflects the role of overtime working in some sectors. 3. Space. The spatial segmentation of labour markets is evident in the inner-city ‘jobs gap’ whereby poor households with low skills tend to be anchored in declining neighbourhoods, distant from suburban areas of job growth.23 Explanations for this ‘spatial mismatch’ are complex. Personal attributes, such as level of qualifications, coupled with shortages of suitable jobs in the local area are major factors. Other factors include the friction of distance combined with poor public transportation, and racial discrimination whereby ‘race not space’ may contribute to the spatial division of labour (see Chapter 15). UNEMPLOYMENT The greatest division in the labour market is that between the employed and the unemployed. Economic restructuring—deindustrialisation, tertiarisation, increased automation in both manufacturing and services, the demand for skilled labour, the ‘feminisation’ of labour markets, and requirements of ‘flexibility’ in post-industrial working practices—has contributed to a growth in unemployment and long-term unemployment. Since the mid-1970s long-term unemployment has become a feature of most capitalist economies.24 Typically, many of the long-term unemployed were previously in semiskilled or unskilled jobs. During the 1980s youth unemployment emerged as a major problem, with the ranks of the long-term unemployed swollen by never-employed school leavers. In Glasgow, by 1987 54 per cent of unemployed males and 39 per cent of unemployed females had been out of work for more than a year, while 16 per cent of Urban geography 398 unemployed males and 6 per cent of unemployed females had not worked for over five years. At the same time, youth unemployment levels had reached 73 per cent for males and 52 per cent for females under the age of 25 years. These trends continued to characterise the urban economy during the last decade of the century.25 The socio-spatial impact of unemployment is evident in deprived urban environments such as the Kirkby estate in Liverpool, where the local labour market ‘is increasingly incapable of providing jobs in adequate numbers for those seeking them and is riven by a growing polarisation in the pay and conditions of the work that is available’ (Meegan 1989 p. 198)26 (Box 14.4). INFORMALISATION The growth in service-sector employment has not been sufficient to offset the loss of jobs in manufacturing. In addition, the type of labour required in many segments of the service sector (part-time, temporary and female) does not match the abilities of redundant male manufacturing workers. The organisation of remaining manufacturing industry has also changed, with a growth in small-batch production, high product differentiation and rapid changes in output. This has led to greater subcontracting and an increase in part-time and temporary work, piecework and industrial homeBOX 14.4 Unemployment on the Kirkby estate in Liverpool Economic restructuring has resulted in a growing number of unemployed residents, with both longterm and youth unemployment significant. Registered unemployment is three times the national average, and according to a local resident, ‘You never see anyone going to work in the morning.’ Another Kirkby woman, talking about her 20-year-old son, stated, “He’s never had a job and wouldn’t know how to go about getting one.’ In 1987 only 7 per cent of school leavers in the borough found jobs (compared with 28 per cent in 1979). The registered unemployment rate for young people has effectively been kept down by a combination of government training schemes and pupils ‘staying on’ at school. As the opportunities for full-time work have contracted, the local labour market has polarised between large private and public employers and a collection of small firms. The remaining large companies in the area enjoy increased dominance in the labour market and recruit selectively, often without advertising vacancies so as to avoid ‘virtually unmanageable numbers of applications’. Pay and conditions in the small firms are poor, with wages often set below levels of remuneration on training schemes in the knowledge that the prospect of a ‘real job’ will attract workers. The downgrading of the local labour market is reflected in the reduced standard of living for people on the estate. This is illustrated in increasing numbers of families in receipt of housing benefit (81 per cent in 1987) and free school meals (77 per cent). As an unemployed man stated, ‘I’m always skint because I’m unemployed. I get depressed because I can’t provide for my family. We have fights… A lot of people are in this situation. It’s quite common.’ Source: R Meegan (1989) Paradise postponed: the growth and decline of The economy of cities 399 Merseyside’s outer estates, in P.Cooke (ed.) Localities London: Unwin Hyman, 198– 234 working. In these circumstances employers may reduce their costs by circumventing employee benefits, labour unions and even the need to provide a workplace. The emergence of a ‘downgraded manufacturing sector’27 is an example of the informalisation common in Third World cities but which is increasingly evident in developed countries.28 In Chicago the city’s ‘contingent economy’ based on flexible, short-time and temporary employment expanded during the 1990s as secure decent-paying job opportunities declined. In some inner-city neighbourhoods the major local employers are temp agencies (both mainstream and ‘backstreet’) who specialise in the placement of day-labour through a network of hiring halls where prospective workers gather each day, as components in a ‘regime of precarious employment’.29 Other forms of informal economic activity in industrial societies include: 1. the domestic or household economy, which produces goods and services primarily for own consumption but which accounts for the great part of national child-care and dependent-adultcare services; 2. the neighbourhood or mutual aid economy in which goods and services may be exchanged on a non-commercial or non-monetary basis within a local community; 3. the alternative or counter-culture comprising a form of social economy (such as Amish or Old Order Mennonites) based on a rejection of the values of mainstream society; 4. the underground, hidden or black economy, which includes illegal activities, creative bookkeeping for the purpose of tax avoidance, and unrecorded employment. We shall return to consideration of informal economic activity later in our discussion of local responses to urban economic change (Chapter 16). URBAN CHANGE WITHIN THE GLOBAL ECONOMY For cities, the impact of the emergence of a global economic system is encapsulated in the distinction between the city as an autonomous self-governing polity (which existed in medieval Europe prior to the development of an economy based on the trade of marketed commodities) and present circumstances, under which city development is influenced to a significant degree by forces beyond its control. Today, investment decisions by managers in a TNC with headquarters in one of the ‘command cities’ of the global economic system can have a direct effect on the well-being of families living on different sides of the world. In order to confront such forces, cities BOX 14.5 World city characteristics 1. The most inherent feature of the world city is its global control function in the world economy World cities occupy a position between the world economy and the Urban geography 400 territorial nation-state. 2. Since the institutions linking the world economy with the state are transnational corporations and international finance, world cities contain a disproportionate number of the headquarters of the world’s largest corporations and international banks. 3. The presence of corporate headquarters and global financial institutions presupposes the rapid growth of a highly paid international elite, including a transnational producer service class (in law, banking, insur