Urban Governance, Voice and Poverty in The Developing World
Urban Governance, Voice and Poverty in The Developing World
Urban Governance, Voice and Poverty in The Developing World
Nick Devas
with Philip Amis, Jo Beall, Ursula Grant, Diana Mitlin, Fiona Nunan and Carole Rakodi
London Sterling, VA
First published in the UK and USA in 2004 by Earthscan Publications Ltd Copyright Nick Devas, 2004 All rights reserved ISBN: 1-85383-993-0 paperback 1-85383-992-2 hardback Typesetting by MapSet Ltd, Gateshead, UK Printed and bound in the UK by CPI Bath Cover design by Danny Gillespie Cover photograph Nick Devas (demonstration by street traders at City Hall, Cebu, Philippines) For a full list of publications please contact: Earthscan 812 Camden High Street London, NW1 0JH, UK Tel: +44 (0)20 7387 8558 Fax: +44 (0)20 7387 8998 Email: earthinfo@earthscan.co.uk Web: www.earthscan.co.uk 22883 Quicksilver Drive, Sterling, VA 20166-2012, USA Earthscan publishes in association with WWF-UK and the International Institute for Environment and Development A catalogue record for this book is available from the British Library Library of Congress Cataloging-in-Publication Data Devas, Nick. Urban governance, voice, and poverty in the developing world / Nick Devas with Philip Amis [et al.]. p. cm. Includes bibliographical references and index. ISBN 1-85383-993-0 (pbk.) ISBN 1-85383-992-2 (hardback) 1. Urban poorDeveloping countries. 2. PovertyDeveloping countries. 3. Municipal governmentDeveloping countries. I. Amis, Philip. II. Title. HV4173.U725 2004 339.4'6'091724dc22 2004004915 This book is printed on elemental chlorine-free paper
Contents
List of Photographs List of Figures, Tables and Boxes The Research Team About the Authors Preface Acknowledgements List of Acronyms and Abbreviations 1 Introduction Nick Devas This research and its key message How have urban governance and poverty been considered in previous work? The city case studies Outline of the book Urban Poverty and Governance in an Era of Globalization, Decentralization and Democratization Nick Devas Urban poverty Urban governance Global shifts: globalization, decentralization, democratization Conclusion Economic Growth, Urban Poverty and City Governance Ursula Grant Introduction Do the poor benefit from city economic growth? City economic growth and poverty reduction: Labour markets and economic opportunities City governance and economic growth: What room for manoeuvre? City governance and poverty reduction: What are the possibilities? Surviving in the City: Livelihoods and Linkages of the Urban Poor Jo Beall Introduction Framing urban livelihoods Household livelihood strategies Wider social networks of the urban poor
Urban governance and the social assets of the poor Conclusion 5 Urban Politics: Exclusion or Empowerment? Carole Rakodi Introduction Democratization, decentralization, rights and regime change: The political context Forms of democracy: Urban political structures and arrangements in theory and practice Actors, power relationships and political practices Conclusion Urban Government: Capacity, Resources and Responsiveness Nick Devas New opportunities for responsive city government Constraints on city government Generating and managing the financial resources Responsive city government: Mechanisms of participation and accountability Conclusions: City government and the poor Civil Society Organizations: Do They Make a Difference to Urban Poverty? Diana Mitlin Introduction Civil society and urban development: A background The extent of civil society organizations in the ten cities Civil society and poverty reduction The constraints facing civil society Conclusion Regulating the Informal Sector: Voice and Bad Governance Philip Amis The importance of the informal sector Background to the informal sector The role of city government in enhancing and protecting the assets of the poor in the informal sector Regulation of the informal sector The stories from the cities Emerging themes Concluding remarks
64 67 68 68 69 73 87 91 95 95 96 103 113 120 123 123 125 127 129 133 142 145 146 148 149 152 153 159 162
CONTENTS v
Accessing Land and Services: Exclusion or Entitlement? Fiona Nunan and Nick Devas The challenge Inadequacies in provision Gaining access: Ways and means Access and the nature of urban governance Conclusions: Access to land and services: Exclusion or entitlement?
164 164 165 167 171 184 186 187 187 194 199 200 202 205 219
10 Conclusions: Urban Governance, Voice and Poverty Nick Devas City comparisons Some conclusions on the key research questions Some implications for policy and governance Directions for further research Annex 1: Research Methodology Annex 2: Urban Governance, Partnerships and Poverty Research Working Papers References Index
List of Photographs
Water pump in a Community Mortgage Program area in Cebu City, Philippines (photo: Nick Devas) Chapter 2 Demonstration by street traders at City Hall, Cebu City, Philippines (photo: Nick Devas) Chapter 3 Street market in Cebu City, Philippines (photo: Nick Devas) Chapter 4 Community group, Mombasa, Kenya (photo: Nick Devas) Chapter 5 Waste truck, acknowledging funding support from the Mayor, Barangay Captain and Barangay Council in Cebu City, Philippines (photo: Nick Devas) Chapter 6 Communal water tap, Kumasi, Ghana (photo: Nick Devas) Chapter 7 Informal housing together with formal, low-income housing provided under the Reconstruction and Development Programme, Diepsloot, Johannesburg, South Africa (photo: Jo Beall) Chapter 8 The construction of the New KR Market building in Bangalore, India, has adversely affected urban poor traders and hawkers and disrupted fragile trading relationships (photo: Solomon Benjamin) Chapter 9 Uncollected waste in an informal housing area in Kumasi, Ghana (photo: Fiona Nunan) Chapter 10 Street improvements in Barangay Pasil, Cebu City, Philippines (photo: Nick Devas) Annex 1 Focus group discussion, Mombasa, Kenya (photo: Nick Devas)
Chapter 1
Figures
2.1 Actors and institutions of urban governance 2.2 Actors, relationships and outcomes 25 27
Tables
2.1 5.1 6.1 6.2 8.1 10.1 Urban poverty in case study countries Forms of urban political system Analysis of city revenues Analysis of city expenditures Labour market characteristics of the informal sector Comparative performance of cities 19 92 105 108 147 188
Boxes
1.1 2.1 3.1 3.2 3.3 3.4 4.1 4.2 5.1 5.2 5.3 5.4 6.1 6.2 6.3 6.4 7.1 7.2 The case study cities with approximate populations City level data on urban poverty Mombasa: Unfulfilled economic potential Santiago: Can city economic growth solve poverty? Employment creation programmes: The case of AGETUR in Benin City government business civil society partnerships in Ahmedabad Kumasi: Women traders forge links in support of urban livelihoods Johannesburg: Antisocial networks hurt urban livelihoods and public safety Cebu: Towards democratic decentralization and pro-poor local government Weak local government and informal politics in Bangalore Bangalore: Competing governance circuits The rise and fall of participatory decision-making in Colombo Johannesburg: Transforming the post-apartheid city government Johannesburg: Impact of the citys fiscal crisis Recife: Participatory budgeting Cebu: Sub-city level governance Barangays Civil society in the case cities Issues on the table: Diepsloot, Johannesburg 7 20 40 41 45 50 58 64 75 77 79 85 101 103 116 120 128 136
7.3 7.4 8.1 8.2 8.3 9.1 9.2 9.3 9.4 9.5
Organization of street vendors in Cebu Cebu: Who participates in whose process? Bangalore: Access to land and small business development Maximum Tolerance policy in Cebu Redevelopment in Bangalore: Disrupting patronage politics in the KR Market Cebu: Community Mortgage Program Kumasi: The politics of public toilets Recife: Recognizing the rights of favela dwellers Slum Networking Pilot Project Sanjay Nagar, Ahmedabad Visakhapatnam: Positive impact of upgrading
138 141 151 154 158 174 178 179 181 183
UK Team
International Development Department, School of Public Policy, University of Birmingham: Nick Devas, Philip Amis, Richard Batley, Ursula Grant, Fiona Nunan, Ben Richards, Elizabeth Vidler International Institute for Environment and Development, London: Diana Mitlin, David Satterthwaite, Cecilia Tacoli Department of Social Policy, London School of Economics: Jo Beall, Nazneen Kanji Department of City and Regional Planning, Cardiff University: Carole Rakodi
Additional Study: Agence dExcution des Travaux Urbains (AGETUR) Benin: Blandine Fanou Additional Study: Participatory Budgeting: Brazil: Celina Souza
The research team would like to pay a special tribute to the memory of the late Dr Rose Gatabaki-Kamau, whose untimely death in December 2000 deprived us of a most valued colleague and friend.
Philip Amis is a Senior Lecturer in the International Development Department, University of Birmingham, specializing in urban poverty and urban policy, decentralization and aid management. He has worked as a consultant and researcher primarily in sub-Saharan Africa and south Asia. He has recently been involved in preparing an OECD (Organisation for Economic Co-operation and Development) publication on Harmonising Donor Practices for Effective Aid Delivery. Jo Beall is a Reader in Development Studies at the Development Studies Institute (DESTIN) at the London School of Economics (LSE). She is a specialist on urban social development and urban governance and has researched these issues in south Asia and South Africa. She is co-author of Uniting a Divided City: Governance and Social Exclusion in Johannesburg (Earthscan, 2002) and editor of A City for All: Valuing Difference and Working with Diversity (Zed Books, 1997). Nick Devas is a Senior Lecturer and Director of the International Development Department, University of Birmingham, specializing in issues of decentralized governance, urban management and public finance. He has worked as a consultant and researcher in a number of countries in Africa, Asia and Europe. Together with Carole Rakodi he edited Managing Fast Growing Cities: New Approaches to Urban Planning and Management in the Developing World (Longman, 1993), and is co-author with Ian Blore and Richard Slater of Municipalities and Finance: A Sourcebook for Capacity Building (Earthscan, 2004). Ursula Grant is a researcher at the Overseas Development Institute, London, specializing in chronic poverty, local governance and participatory approaches. She has worked in east and west Africa, and Jamaica. Diana Mitlin is an economist and social development specialist working at the International Institute for Environment and Development (IIED) in London and the Institute for Development Policy and Management, University of Manchester. She has a particular interest in urban poverty and grassroots civil society, participation and effective poverty reduction programmes. Fiona Nunan is the Institutional and Social Development Advisor for the Integrated Lake Management project, Uganda. She worked as a lecturer with the International Development Department, University of Birmingham, from 1994 to 2002, specializing in environmental policy and management, the linkages between environment and poverty, urban environmental issues and peri-urban natural resource management, in Africa and south Asia.
Carole Rakodi, formerly with the Department of City and Regional Planning, Cardiff University, is now Professor of International Urban Development in the International Development Department, University of Birmingham. She is a geographer and urban planner specializing in urban land and housing policy, urban management and governance, and urban poverty and livelihoods. She has worked and carried out research mainly in Africa and also in India. She coedited, with Tony Lloyd-Jones, Urban Livelihoods: A People-centred Approach to Reducing Poverty (Earthscan, 2002) and, with Sam Romaya, Building Sustainable Urban Settlements: Approaches and Case Studies in the Developing World (ITDG, 2002).
Preface
Poverty and governance are both issues high on the agenda of international agencies as well as of many governments in the South. The interconnections between poverty and governance are also receiving increased attention. But, so far, little research has been done on these issues at the sub-national level, and specifically in relation to urban areas. With the rapid growth of cities in the South, it is clear that poverty can no longer be regarded as predominantly a rural phenomenon. Urban areas account for a steadily growing share of the worlds poor people. Thus it is timely to look at how the governance of the cities in the South affects the prospects and livelihoods of the poor of those cities, and how poor people can use their voice to influence the decisions of city government that affect them. This book is based on a three-year research project examining the relationship between urban governance and poverty in ten cities in Asia, Africa and Latin America. The research was carried out between 1998 and 2001 by a team of researchers from the UK working with local researchers in each city. The team members are listed on page ix. The research was funded by the Economic and Social Research programme (ESCOR) of the UK Governments Department for International Development (DFID). Field research involved reviewing existing studies; interviewing key actors from city governments and civil society; holding focus group discussions with groups of the urban poor, community organizations and non-governmental organizations (NGOs); and analysing data on poverty, service provision and finance. In five of the cities (Bangalore, Cebu, Johannesburg, Kumasi and Recife), more detailed case studies were prepared on the ways in which particular groups of the urban poor had sought to achieve improvements in their situation. Annex 1 gives further details about the research process. Annex 2 lists the 31 working papers from this research and indicates how they may be obtained.
Acknowledgements
The authors of this book gratefully acknowledge the enormous contribution of the other members of the team, both in the UK and in the cities concerned. Their contribution was not just in the carrying out of the research in the cities but also in the development of the ideas and approaches used in this research, as well as in the comparative analysis of the findings. Particular appreciation is due to Carole Rakodi for her detailed comments on the draft chapters of this book, and to Annabel Devas for her proof-reading of the drafts and preparation of the text. The funding for this research from DFIDs ESCOR programme is also gratefully acknowledged. Finally, we would like to express our appreciation for all those in the cities concerned who provided the raw material for this book: the residents, officials, elected representatives, activists and others who participated through interviews, group discussions and provision of information. In particular, we salute those who wrestle daily with grinding poverty and who demonstrate how it is possible to survive and even prosper against overwhelming odds. We salute, too, all those who struggle to improve the governance of cities for the benefit of all and particularly for the poor.
Disclaimer
This research has been funded by the Department for International Development (DFID) of the United Kingdom Government through its ESCOR programme. DFID hopes that this research will be of value to policymakers and practitioners. The facts presented and the views expressed are those of the authors and do not necessarily represent the policies of DFID.
AGETUR AMC ANC ASAG BCC BDA CBD CBO CCUP CDC CDF CMC CMP COMUL CPATODA CSO CWA DACF DESTIN DFID DWUP ESCOR GJMC GNP GONGO GRO IDP IIED ILO KCM KMA LDO MCE MLC MMC NGO ODA OECD PB
Agence dExcution des Travaux Urbains (Benin) Ahmedabad Municipal Corporation African National Congress Ahmedabad Study Action Group Bangalore City Corporation Bangalore Development Authority central business district community-based organization City Commission for Urban Poor (Cebu) Community Development Council (Colombo) Community Development Forum (Johannesburg) Colombo Municipal Council Community Mortgage Program (Philippines) Local Commissions for Tenure Regularization and Urbanization (Recife) Cebu Port Authority Trisikad Drivers Association civil society organization Communal Water Association (Cebu) District Assemblies Common Fund (Kumasi) Development Studies Institute Department for International Development Division for the Welfare of the Urban Poor (Cebu) Economic and Social Research programme Greater Johannesburg Metropolitan Council gross national product Government sponsored NGO grassroots organization Integrated Development Plan (Johannesburg) International Institute for Environment and Development International Labour Organization Kumasi Central Market Kumasi Metropolitan Assembly Land Development Objective (Johannesburg) Metropolitan Chief Executive (Kumasi) Metropolitan Local Council (Johannesburg) Mombasa Municipal Council non-governmental organization Overseas Development Agency (now DFID) Organisation for Economic Co-operation and Development participatory budgeting
PO PPP PR PREZEIS PRSP PT RDP ROSCA Rs SANCO SC/ST SEWA SME SNP SPEED SSE UBSP UNCHS UNDP UN-Habitat UNICEF ZEIS
peoples organization purchasing power parity proportional representation Programme for the Regularization of Zones of Special Social Interest Poverty Reduction Strategy Paper Workers Party (Brazil) Reconstruction and Development Programme (South Africa) Rotating Savings and Credit Association rupees South African National Civics Organization Scheduled Castes and Tribes (India) Self-Employed Womens Association (India) small and medium enterprise Slum Networking Project (Ahmedabad) Squatter Prevention Encroachment Elimination Division (Cebu) small-scale enterprise Urban Basic Services Programme United Nations Centre for Human Settlements (now UN-Habitat) United Nations Development Programme United Nations Human Settlements Programme (formerly UNCHS) United Nations Childrens Fund Zones of Special Social Interest
Chapter 1
Introduction
Nick Devas
City governance matters for the poor. It can make the difference between maintaining a fragile foothold in the city and being swept away into ever deepening poverty. City governance is not just about the formal structures of city government, important as these may be. The position of the poor depends on a host of economic and social forces, institutions and relationships: the markets for labour, goods and services; household, kin and social relationships; and the availability of or lack of basic infrastructure, land, services and public safety. The actions of city government can make matters worse for the poor, through inappropriate and repressive policies and interventions, or they can be supportive, for example, by ensuring access to essential infrastructure and services. But much of what affects the life-chances of the poor lies outside the control of city government, determined by the market and private businesses, by agencies of the central state, or by the collective voluntary action of civil society. The ways in which these elements interact, and the power relationships involved, are critical for the urban poor as they seek to establish and improve their position.
It is these interactions and relationships informal as much as formal that we refer to as city governance. Our concern in this volume is how the processes and relationships of city governance affect the poor, and how far those in poverty are able to use what influence they have to achieve a better outcome. International attention is increasingly focused on issues of poverty. The World Banks World Development Report for 2000/01, Attacking Poverty (World Bank, 2001a) is but one example. Poverty Reduction Strategy Papers (PRSPs) have replaced national development plans as the basis for negotiations between governments in the South and donor agencies. But until recently, poverty in the developing world has been seen mainly as a rural problem. Cities have been perceived as the locations of prosperity and manifestations of uneven development, while urban bias in resource allocation has been blamed for slowing development and perpetuating rural poverty (Lipton, 1977). It is true that the majority of the worlds poorest still live in rural areas, but the rapid growth of cities is changing the balance. Already, almost half of the worlds population is urban, and three-quarters of the worlds population growth now occurs in the urban areas of the developing world (UN-Habitat, 2001a, p3). Urban areas account for an increasing share of the worlds poor (Haddad et al, 1999), and one estimate is that by 2035, half of the worlds poor will be living in urban areas.1 Many of those living in the cities of Africa, Asia and Latin America do so in conditions of severe poverty, without access to adequate shelter or basic services. Some 30 million urban dwellers do not have access to safe drinking water (UN-Habitat, 2001a, p14) and 100 million have no permanent home (World Bank, 2000, p48). Statistics of urban poverty are, of course, bedevilled by questions of comparability, such as how poverty should be defined, where poverty lines should be drawn and to what extent these lines should reflect differences in living costs between rural and urban areas. These are issues that will be addressed further in Chapter 2. But it is clear that urban poverty can no longer be regarded as either an insignificant problem or simply a matter of transitional adjustment. Traditionally, economic growth has been seen as the key to reducing poverty, and there are plenty of examples where sustained economic growth has dramatically reduced absolute levels of poverty. But, as is well known, economic growth alone is not sufficient to eradicate poverty. Indeed, economic growth tends to widen the gap between rich and poor, thereby increasing relative poverty, and even in some cases worsening the position of the poorest in absolute terms. How the benefits of economic growth are distributed is critical. Governments, both national and local, play a significant role in determining the distributional aspect of growth, whether for better or for worse. Democratization and the growth of civil society offer some possibilities for the poor to influence what happens. While economic liberalization and globalization may have improved economic growth prospects in some countries, they have generally led to increased inequality, and in many cases conditions for the most vulnerable have worsened. Structural adjustment policies, privatization and deregulation have
INTRODUCTION 3
reduced the scope for government intervention on behalf of the poor, or made that intervention more indirect.2 Yet what governments national, regional and local do or do not do still has a crucial impact on urban economic growth and on poverty, inequality and exclusion. While the room for manoeuvre for city governments may be quite constrained, how they use it can have a significant impact on the poor.
The key message from the study is that the well-being of the urban poor can be improved by access to economic opportunities, supportive social networks, and greater access to assets (notably land), infrastructure and services. Whether and how these critical elements are available to the poor depends to a significant extent on how city governance functions: local political processes, informal as well as formal; the influence of the various civil society organizations (CSOs) representing the poor; and the capacity of city government to respond. Thus, city governance embracing all those involved in making decisions that affect the livelihoods of the poor is central. Voice matters: there are places where the urban poor are getting their voice heard, but the outcomes depend on a number of factors, including the nature of the particular local democratic institutions and processes, the resources available, and the ability of the poor to organize and articulate their demands. These are all aspects which will be explored further in this book.
How have urban governance and poverty been considered in previous work?
The growing international concern with poverty, along with international commitments to reducing global poverty, has spawned numerous studies and reports (eg World Bank, 2001a; UNDP, 2000a; DFID, 1997), including many linked to national poverty reduction strategy programmes (eg World Bank, 2002b). In the past, most attention has been focused on rural poverty, but the connections between rural and urban livelihoods are increasingly being recognized (eg Tacoli, 1998; Chant, 1998; Jerve, 2001; Volbeda, 2002). Those writing about poverty are also paying greater attention to issues of governance (Shepherd, 2000; Johnson and Start, 2001; Osmani, 2001; Craig and Porter, 2003). However, as already noted, the relationship between poverty and governance at the local or urban level is a relatively new area for research. The urban literature to date could be crudely categorized into five broad but overlapping groups. First, there are the wide-ranging treatments of urbanization internationally, dealing with the general problems of urban growth and its impact on urban society, for example, Gilbert and Gugler, 1992; Smith, 1996; Potter and Lloyd-Evans, 1998; Drakakis-Smith, 2000. Within this broad group are writings on urban sociology, including urban social movements (eg Castells, 1983, 2002; Fainstein and Hirst, 1995; Walton, 1998), and the wider political context, including the impact of the current neo-liberal agenda (Burgess et al, 1997). This urbanization and urban sociology literature has on the whole treated both urban poverty and governance in fairly general and theoretical terms. Second, there is the literature on urban poverty, livelihood systems of the poor and social capital, including the output from location-specific research on these issues, for example, Amis and Rakodi, 1995 and other papers in the same issue of Habitat International; Wratten, 1995; Kanji, 1995; Beall, 1995; Brown and Ashman, 1996; Moser, 1996, 1998. We may also include under this broad category earlier works on the urban informal sector and informal settlements, such as Bromley and Gerry, 1973; Perlman, 1976; Moser, 1978; Moser and Peake, 1987, to name but a few. With the focus in this literature being on how those in poverty survive, less attention has been paid to governance issues, although more recent work (eg Jones and Nelson, 1999; Rakodi and LloydJones, 2002) begins to address this aspect, as does some of the literature on social capital (eg Evans, 1996; Harriss and de Renzio, 1997; Putzel, 1997). Third, and closely related, is the literature on civil society and the role of community organizations and non-governmental organizations (NGOs) in relation to urban poverty, for example (among many): Hulme and Edwards, 1996; Douglass and Friedmann, 1998; Mitlin, 2001b. Some of this literature addresses the interactions between civil society and the state at the local level, based on specific city experience, for example, Klaarhamer, 1989; Peattie, 1990; van der Linden, 1997. The fourth group is the literature on urban planning and management, including analyses of housing, land, services and the environment: Sivaramkrishnan and Green, 1986; Stren and White, 1989; van der Hoff and
INTRODUCTION 5
Steinberg, 1992; Devas and Rakodi, 1993; Davey et al, 1996; Rland 1996; UNCHS, 1996; Vanderschueren et al, 1999 (and other papers produced under the UN-Habitat/World Bank Urban Management Programme); Werna, 2000; Freire and Stren, 2001; Hardoy et al, 2001; Plummer, 2002. These have tended to treat issues in a rather more technocratic way, in some cases with a limited conceptualization of either the livelihood systems of the poor or the political processes of city governance. The last group is literature on urban politics and specific political processes at the city level. Much has been written about urban politics and government in western industrialized countries, (eg Wolman and Goldsmith, 1992; Judge et al, 1995), including analyses of pluralism (Judge, 1995), regime theory (Stoker, 1995; Kantor et al, 1997), policy networks (Rhodes, 1997), partnerships (Peters, 1998; Pierre, 1998), and current issues of city government (Hambleton, 1990; Sharpe, 1995). There have also been many detailed studies on government and politics in particular North American cities, including the seminal work by Dahl (1961) and more recent material from Savitch and Thomas (1991) and Imbroscio (1997), as well as some studies on European cities. Political and policy analysis in the developing world has tended to focus on the national level rather than the city level (Grindle, 1980; Grindle and Thomas, 1990; Hyden and Bratton, 1992; Hyden et al, 2000; Bangura, 2000) and more generally on statesociety relations (Chazan et al, 1992; Migdal, 2001). One stream of literature that has been growing is that on decentralization and its impact on the poor: Crook and Manor, 1998; Blair, 2000; Crook and Sverrisson, 2001; and on parallel issues concerning citizen voice and local government accountability (eg Goetz and Gaventa, 2001; Devas and Grant, 2003). But, with the exception of Latin America, writings on urban governance and politics in the developing world have been comparatively rare, certainly until recently. Significant contributions in this general area include: Stren, 1978; Rland 1992; Mabogunje, 1995; Stren with Kjellberg Bell, 1995; McCarney, 1996; Onibokun, 1997; Halfani, 1996a; Porrio, 1997; and Swilling, 1997. However, much of this writing remains at a fairly general level, often in the form of literature reviews with limited conceptual or empirical basis.3 The research on which this book is based seeks to build on all of these strands of writing on urban issues in the developing world, and to provide some empirical basis for a better understanding of how the processes of city governance affect those living in poverty.
large cities between 0.5 million and 4 million. There is another whole research project (or several!) to be done on small cities and towns. But our city studies can be considered as representative in the sense that they each provide rich descriptions of the critical processes, together with explanations that enable a more generalized understanding of the issues to be formed. We selected the cities with a view to providing some interesting contrasts: those cities where economic growth appeared to have reduced absolute poverty (eg Santiago) and those where there had been little economic growth (eg Mombasa); those where there had been at least some significant pro-poor initiatives (eg Cebu, Colombo, Recife) and those where there had been minimal attempts to address poverty (eg Kumasi, Mombasa); those where there had been radical political change (eg Johannesburg) and those where traditional arrangements had continued largely unchanged (eg Bangalore, Ahmedabad); those where decentralization had given greater scope to city government (eg Cebu, Recife) and those where there had been no real decentralization (Mombasa); those where there was evidence of an active civil society (Cebu, Johannesburg, Ahmedabad) and those where civil society appeared weak (Kumasi, Visakhapatnam); those where donor agencies had played a major role in urban improvement (Visakhapatnam) and those where they had not (Santiago, Mombasa).
Initial studies were undertaken in nine cities (see Box 1.1), from which four were selected for more detailed analysis (Bangalore, Cebu, Johannesburg, Kumasi), with an additional city (Recife) being selected for a detailed study of the participatory budgeting system. In addition, a small study was conducted on the arrangement (known as Agence dExcution des Travaux Urbains (AGETUR)) for labour-intensive urban public works in Benin, as an example of a specific poverty-reduction initiative. The following provides snapshots of the situation in our ten cities. Further details on poverty levels and trends are given in Box 2.1 in Chapter 2.4
Ahmedabad, capital of the state of Gujarat, is an ancient city and the seventh largest in India. Its main industry, textiles, at one time accounted for 80 per cent of the citys employment, but the decline of the sector, together with major restructuring, has greatly reduced this. There has been a significant informalization of the economy in recent years. Despite this, the proportion of the population living in poverty fell substantially between the mid-1980s and the mid-1990s (see Box 2.1). However, polarization along social, economic and ethnic lines has increased in recent years, with regular outbreaks of communal violence. The literacy rate is 69 per cent. According to official statistics, 96 per cent of houses have either individual or shared water supplies, and 86 per cent have either individual or shared toilets. However, these figures exclude slum
INTRODUCTION 7
settlements and so significantly overstate the true situation. In addition, the quality of these services is quite variable. The Municipal Corporation, which covers most of the urban area, has a council elected on a ward basis, with some reserved seats for women and scheduled castes and tribes. The mayor, who is elected annually from among the councillors, has a largely ceremonial role. Executive power resides with the Municipal Commissioner, a state-appointed official, and the eight deputy Municipal Commissioners. There is a great range of non-governmental organizations (NGOs) and civil society organizations (CSOs) operating in the city, including the internationally renowned Self-Employed Womens Association (SEWA), and more than 1200 community-based organizations (CBOs).
Bangalore, the state capital of Karnataka, is Indias sixth largest city and
among the fastest growing of the countrys large cities. In recent years, its economy has been centred on high-tech industries and it has become known as Indias Silicon Valley, sharing in the sectors global boom of the early 1990s and the stagnation at the end of that decade. But the citys economy is also based on a mass of small enterprises, many of them unregistered and dependent on fragile networks of relationships. The economic boom has widened the gap between rich and poor, with prestigious developments alongside slum settlements with the former often forcing out the latter. Nevertheless, available estimates suggest that the proportion of households living in poverty fell
significantly during the decade up to the mid-1990s, although not by as much as in Ahmedabad. However, as with any statistics on poverty, these figures are widely contested. A quarter of the citys population live in defined slum areas. More than half the population depend for water on public taps that are often located far away and may deliver contaminated water. Many houses and even whole communities are without latrines. Bangalore Municipal Corporation was established long ago, and its boundaries do not include large areas where some of the citys poorest live. Councillors are elected on a ward basis, with the mayor being elected indirectly. The externally appointed Municipal Commissioner wields considerable power. There has been significant fragmentation of responsibilities to unelected, statecontrolled bodies, such as the Bangalore Urban Development Authority. Whilst there are many NGOs operating in the city, these are often viewed with some suspicion by the poor.
Cebu City is the core part of Metro Cebu (which includes nine other
municipalities) and is the capital of the Central Visayas region of the Philippines. It is a major port and industrial centre. During the late 1980s/early 1990s it experienced an economic boom, but this was followed by a downturn in the late 1990s. Unemployment is now around 15 per cent, and a large proportion of the population are employed in the informal sector. Statistics on poverty vary widely, depending on the source and definitions, classifying anywhere between 19 per cent and 77 per cent of the population as poor (Etemadi, 2001, p19 and Table A3.1). Literacy is high (98 per cent) but less than 60 per cent of the citys population has access to piped water in their house or on their plot, and more than a third of the population do not have access to proper sanitation. Cebu has a directly elected executive mayor and a legislature elected by proportional representation (PR). Below the city government there are 80 elected barangays. Decentralization in the early 1990s gave the city greater responsibilities and resources. NGOs enjoy something of a privileged position in governance at all levels in the Philippines, and many are actively engaged in issues of urban poverty. There are also a great many active CSOs and CBOs in Cebu.
Colombo is the commercial as well as the political capital of Sri Lanka. Its port handles virtually all the countrys trade. Despite the civil war in the northeast of the country, the national economy has performed well, with an average annual growth rate of over 5 per cent a year during the 1990s. This has been mirrored in Colombo which has benefited from economic liberalization and the establishment of export processing zones. Sri Lanka has a reputation for relatively good human development indicators and has a range of welfare programmes for the poor. Nevertheless there is significant poverty: in 1990/91, 18 per cent of Colombos population were below the official poverty line (World Bank, 1995, p7), and this has probably increased since then as a result of ethnic violence. Around 10 per cent of the citys population live in under-served settlements and a quarter live in sub-standard housing. Although almost all households have access to piped water, supplies are intermittent, pressure is low and many people have to share public taps.
INTRODUCTION 9
Colombo Municipal Council (CMC) is elected on a party-list proportional representation system, with a centrally appointed Municipal Commissioner playing a key role. Despite donor assistance with institutional development, CMC is beset by management and financial problems. This has not been helped by political disputes with central and provincial government. Service and development responsibilities are fragmented between a number of agencies, notably the Urban Development Authority. There are some 50 NGOs, many of them engaged in programmes for the urban poor. In terms of CBOs, Community Development Councils (CDCs) played a significant role in enabling residents to participate in decisions affecting them under the Million Houses Programme during the 1980s, but most of these have become dormant in recent years.
Johannesburg is the most important city in the South African economy, based originally on mining but now highly diversified. Under apartheid, the city was divided into 13 racially segregated local authorities. In the early 1990s these were reorganized into four Metropolitan Local Councils (MLCs), elected partly on a ward basis and partly by proportional representation, under the umbrella of the Greater Johannesburg Metropolitan Council (GJMC). This arrangement proved unsatisfactory and the MLCs were abolished in 2000, leaving the city with one, very large, single tier city government for over 3 million people.5 The legacy of the past is one of huge inequality, and an urban structure that seriously disadvantages the poor. Slow economic growth over the past decade, combined with rapid population growth, has led to growing unemployment (between 17 and 30 per cent, depending on definitions) and increased informalization of the economy. Urban poverty may well have increased, partly because the city is absorbing more of the rural poor, although accurate data are not available. Basic service levels are comparatively high, with 98 per cent of households having a tap in the house or on the plot, and 93 per cent having a flush toilet. However, there are marked differences in service levels between areas, with some squatter settlements still having minimal facilities, and the poor encounter severe problems in paying for these relatively high levels of services. Kumasi, Ghanas second city, has a proud history as the centre of the Asante
kingdom. Traditional authorities still exercise a degree of power, notably in relation to the allocation of land. The local economy is quite dynamic, with numerous small-scale manufacturers and workshops, many of which are informal. Structural adjustment and liberalization have had some positive impact on the national economy, but they have widened the gap between rich and poor. Data on urban poverty do not exist for Kumasi, but the national figures for urban areas excluding Accra suggest a modest reduction in poverty during the late 1980s and early 1990s, although the level appears to have increased again since then. While piped water is available in most areas, the pressure is low and the supply intermittent, and most people have to purchase water from private vendors at high prices. Only 30 per cent of households have satisfactory sanitation, with 40 per cent depending on a limited number of public toilets and many defecating in the open.
Kumasi Metropolitan Assembly (KMA) is elected on a ward basis but without political parties (officially, at least). Decentralization in Ghana gave the KMA additional powers and resources but the process is incomplete: KMAs revenue base is weak; voter turnout is low; most services are still effectively controlled by central ministries or agencies; and the Metropolitan Chief Executive (MCE) is appointed by the centre. The MCE at the time of the study behaved in an unpredictable and autocratic manner, dominating the KMA and preventing the establishment of the sub-city levels of government specified in the law. There are NGOs operating in the city but many exist only on paper and none has sought to engage with city government on behalf of the poor. There is also little in the way of sustained grassroots organization (GRO) in poor communities.
Mombasa is Kenyas second city and its major port and trading centre, serving
much of eastern and central Africa. It has an ethnically diverse population with commercial life dominated by Arabs and coastal Kenyans. Failure to invest in modernizing the port and surrounding infrastructure has meant that the ports throughput has increased only slowly. The local economy, like that of Kenya as a whole, has been stagnant. The important tourist industry declined due to perceived insecurity and political unrest in the late 1990s. Formal sector employment has not kept pace with population growth and the economy has become increasingly informal. In 1994, one-quarter of households and one-third of individuals fell below the poverty line (Rakodi et al, 2000, p156), and there is a widespread perception that the position of the poor has worsened since then. Mombasa Municipal Council (MMC) is elected on a ward basis with the mayor being elected by the councillors. It has had a reputation for patronage politics, poor management, weak service delivery and the application of regulations that disadvantage the poor. In recent years, the Council has tried to engage more with civil society organizations and to be more responsive to citizens, but its parlous financial situation means that it has no resources to improve services. Although over 80 per cent of residents are said to have access to water, the service is erratic or non-existent. Only 30 per cent of households have water-borne sanitation, two-thirds use pit latrines, many of which pollute the ground water, and the rest defecate in the bush. There has been a considerable growth in the activities and confidence of NGOs and CBOs representing the poor, although the concrete results of their endeavours have been limited. There has been little external donor support to the city in recent years.
Recife municipality, with a population of 1.3 million, is the core of the Greater
Recife metropolitan area, made up of 14 municipalities with a total population of 3.1 million. It is the fourth largest urban centre in Brazil, and is located in the northeast, the poorest region of the country. Economic stagnation in the early 1990s resulted in increased unemployment: officially, 7 per cent of the population is unemployed, but the real level is more like 30 per cent. There is considerable poverty in 1991, 30 per cent of households had incomes below one single minimum wage (Melo et al, 2001, p23) as well as a high degree of inequality. Illiteracy is estimated at 14 per cent. Although 80 per cent of the
INTRODUCTION 11
population have piped water in their homes, only 48 per cent have access to sewerage. Recife has a directly elected executive mayor and a legislature elected by proportional representation. Democratization and decentralization in Brazil have increased the importance of municipal government as well as its resources. One of the most significant innovations has been participatory budgeting, adopted in Recife, as in many other Brazilian cities, in the mid-1990s, but building on earlier local, participatory initiatives such as the Prefitura nos Bairros and Prezeis programmes started in the 1980s.
Santiago is by far the largest city in Chile, accounting for around half of the countrys gross national product (GNP). The city, like the country, experienced rapid economic growth during the 1990s, averaging around 8 per cent per year. As a result, absolute poverty levels fell dramatically during the 1990s. However, these figures disguise huge inequalities and increasing problems of social exclusion. Unemployment was around 7 per cent in the late 1990s.There is almost 100 per cent provision of piped water, and 94 per cent for sewerage, with charges related to ability to pay (in principle, at least). The city has expanded rapidly over the years, but there is no government for the city as a whole. Instead, the city is divided into 34 comunas, each with its own council elected on a proportional representation basis, and an executive mayor elected from among the councillors. There are big variations between comunas in terms of resources and service levels. Prior to the restoration of democracy in Chile, there was a highly active civil society engaging in issues concerning urban poverty, but in recent years there has been a retreat into private space and civil society has become almost silent. Visakhapatnam in Andra Pradesh is reputed to be the fastest growing urban area in India, its population having grown ten-fold in the 40 years from 1951. It has the largest and fastest growing port in India and a substantial industrial base. However, inadequate infrastructure is constraining further growth. Employment levels and wage rates are comparatively high, although there are uncertainties over the future of some of the traditional industries. No data on poverty were available for the city, but for the state of Andra Pradesh, urban poverty was estimated to be around 30 per cent in 1999 (Amis and Kumar, 2000, p19). Onequarter of Visakhapatnams population live in areas designated as slums. Official figures of 90 per cent of households having access to piped water are disputed; nearly half of those depend on public taps which provide water for only limited periods. A large proportion of households depend on public toilet facilities which are often not maintained properly. In the 1990s, a slum upgrading project funded by the UK Government provided basic infrastructure and services in more than 80 per cent of slum areas, bringing considerable benefits to residents. The Municipal Corporation, which was only established in 1978, is elected on a ward basis. It is perceived by many to be relatively well managed, and has succeeded in maintaining a budget surplus for a number of years. Compared to some other cities in India, political life is somewhat subdued. The business community plays a dominant role, although backward castes have certain
privileged positions. Similarly, there seems to be a low level of civil society activity, with no NGOs having any direct involvement in urban poverty issues and most CBOs being caste-based organizations pursuing their own interests.
INTRODUCTION 13
where city governments have been able to improve their financial position, overcome some of the constraints and adopt a more responsive and participatory approach to the urban poor. In Chapter 7, Diana Mitlin looks at the role of civil society in relation to urban poverty. It is fashionable to see civil society as the counterweight to government in representing the interests of the poor. Yet grassroots organizations (GROs)6 have many limitations, and may be far from representative of the urban poor. NGOs, too, are often viewed with suspicion by those they claim to serve. Nevertheless, there are examples from the case cities where GROs and NGOs have played a role in influencing city policies and practices in favour of the poor. These cases are critically analysed, as well as those where NGOs and CBOs have been less successful. Philip Amis looks at the particular issue of city governments approach to the informal sector in Chapter 8. All too often, city governments adopt a repressive policy towards informal sector trading, and even where they officially claim to favour the poor, interventions often have damaging effects on the fragile and location-specific income opportunities of the poor. Amis reviews examples from the city studies where the voice of the urban poor has had an influence in redirecting the policies and practices of city authorities in relation to the informal sector. He argues that, in the end, democratic accountability is the strongest bulwark against bad governance. Chapter 9 focuses on the issue of access to land, infrastructure and services. Fiona Nunan and Nick Devas examine the inadequacies of basic infrastructure and services in the case cities, and the impact of these on the poor. Access to land is crucial, not just for housing but also for the scope that it provides for economic activities. Access to basic infrastructure and services probably has a greater positive impact on the poor than most welfare programmes. Nunan and Devas review alternative approaches for improving access for the poor to land and basic services, and report on the ways in which city governments in the case cities have or have not sought to address the needs of the poor. In the final chapter, Nick Devas presents a summary comparison between the ten cities in terms of a number of indicators considered in this study. He then reviews the main conclusions from the study, based on the original research questions, and identifies key policy implications for elected representatives, city officials, community organizations, NGOs and donor agencies. He concludes with some suggested priorities for future research.
Notes
1 From Ravaillon, 2001. Measuring urban poverty is fraught with difficulties, as will be discussed in Chapter 2, and forecasting urban poverty depends on a host of assumptions. Nevertheless, it is clear that the number of urban dwellers living in poverty is increasing rapidly, as is the urban share of world poverty. There is evidence that structural adjustment programmes have been accompanied by increased urban poverty (Moser et al, 1993) and have exerted a greater downward pressure on urban incomes than rural incomes, at least in Africa (Becker et al, 1994).
Chapter 2
Urban Poverty and Governance in an Era of Globalization, Decentralization and Democratization
Nick Devas
This chapter unpacks two of the key concepts in this research: urban poverty and urban governance. It looks at how these concepts have been treated in the literature, and considers how the characteristics of urban poverty and urban governance have been affected by the current processes of globalization, decentralization and democratization.
Urban poverty
Defining and measuring urban poverty
A fundamental difficulty for research of this sort is the inadequacy of statistics on poverty at city level. While some cities have data about the numbers of
people falling below a given poverty line at a particular point in time, few have consistent data over time, let alone data on the more sophisticated indicators of the extent and nature of poverty. Nor are the available statistics in a form that is readily comparable across cities. The first issue is how poverty is defined. Conventionally, a distinction is drawn between absolute poverty and relative poverty. The former relates to those who do not have sufficient income to afford a minimum level of nutrition and basic needs, while the latter is concerned with the position of the poor in relation to the rest of society, and so is an indicator of the degree of inequality. The conventional measure of absolute poverty is the headcount index, which indicates the numbers falling below a specified poverty line. This has the merit of simplicity and so is easily understood, but there are always debates about where the line should be set for a particular country or city. Using the international standard poverty line of $1 per day permits some comparison between countries, albeit very crude. Such measures of poverty suffer from a number of serious problems: the question of how, and by whom, the basket of basic needs is defined on which the poverty line is set, and the extent to which this reflects the perceptions of the poor themselves about what constitutes minimum needs; national poverty lines are, to a certain extent, politically determined, to show either a high or low level of poverty (depending on the purpose), and either an improving or worsening situation; a poverty line is inevitably arbitrary, and there may be a large proportion of the population falling just above the line but highly vulnerable to slipping back into poverty; some countries adopt more than one line in order to try to capture this; the international line is particularly arbitrary, even where attempts are made to adjust for differences between countries in relative purchasing power; a single figure for poverty says nothing about the depth of poverty that is, how far the poor fall below the poverty line; one approach to this is to compute a figure for the poverty gap which measures the additional income needed by the poor to rise above the poverty line; headcount and poverty gap indexes measure only income poverty and take no account of other aspects of deprivation, such as inadequate access to basic services, social exclusion and lack of voice that may be just as important to the poor as inadequate income; even as a measure of the economic position of the poor, monetary income is problematic, since not all incomes are in monetary form and the incomes of the poor are generally highly erratic; those who have an adequate income one month may find themselves destitute the next as a result of ill health, loss of a family member or collapse of income earning activity.
In relation to data on urban poverty, two further issues arise: whether an adequate adjustment is made for the considerably greater costs of living for those in urban areas compared to rural areas, for example, for
housing, water, sanitation, waste disposal, transport, etc; Satterthwaite (1997) concludes that international figures massively underestimate the extent of urban poverty;1 how urban is defined, since many of the poorest urban residents live outside the official municipal boundaries, and the distinction between urban and rural is often somewhat arbitrary.
In order to address the inadequacies of income measures of poverty, other indicators of quality of life can be used, such as infant mortality, life expectancy, illiteracy and access to water and sanitation. Such indicators can be a useful counterweight to crude measures of income, but they are rarely available below the national level. Nor are they unproblematic. For example, published statistics on the availability of water tend to overstate the position by recording all those who have access to a tap even though water may flow only intermittently or not at all through those taps. Average figures also tell us little about the position of the poor. Relative poverty, by contrast, is concerned with the distribution of income, and with the share of national income received by the poor usually the poorest 1030 per cent of the population. By definition, there will always be some degree of relative poverty, since there will always be a poorest 10 per cent of the population, whether or not those concerned live in absolute poverty. Nevertheless, peoples perceptions of their own poverty have as much to do with how their situation compares with the general standard of living as it does with achieving a minimum level of nutrition. Thus both measures can help to illuminate the position of the poor. Absolute and relative poverty can move in opposite directions and often do, as economic growth lifts some out of absolute poverty but widens the gap between rich and poor. The conventional ways of presenting relative poverty information are through a Lorenz curve (which plots the share of income received by each segment of the population) and a Gini index of income distribution. A Gini index of zero represents perfect equality, while an index of one represents complete inequality. It is notable that national level Gini indexes tend to be much higher that is, indicating greater inequality in Latin America (typically 0.500.65) and Africa (typically 0.400.60) than in Asia (typically 0.350.45) and OECD (Organisation for Economic Co-operation and Development) countries (typically 0.250.40). For our case study countries, Gini indexes range from 0.59 for South Africa and Brazil to 0.38 for India and 0.34 for Sri Lanka (World Bank, 2001b, Table 2.7). Such data are rarely available at city level, but if they were, it is likely that Gini indexes would be considerably higher in urban areas than for nations as a whole, reflecting the greater range of incomes in urban areas compared to rural areas. The next problem is measuring poverty. This covers several of the issues already referred to, such as how poverty lines are defined and what is included. There are considerable difficulties in conducting surveys to measure poverty, both in relation to the poor, whose incomes may be extremely variable, and in relation to the rich who are generally adept at disguising their incomes. For the sake of comparability over time, methods and measures need to be applied
consistently, but adjustments also have to be made to reflect changed patterns of expenditure needs and purchasing power of currency. Some countries, such as India, have well established methods for measuring poverty, but even so there are many questions about the reliability of the data produced. A particular problem in relation to urban level statistics is that they are often based on sample surveys from a number of urban centres, with the result that there is insufficient data about any one city for a city level analysis. Writers such as Chambers (1995) and Satterthwaite (1995), have criticized conventional approaches to measuring poverty as reductionist, not only for measuring only what is measurable, thereby establishing the measurable as being what is important, but also for ignoring the perceptions of the poor themselves about what constitutes poverty. As a result of focusing on narrow measures of income poverty, professional policy prescriptions tend to emphasize formal income generation, ignoring the livelihood systems of the poor. In recent years, participatory poverty assessments have gained acceptance as a way of obtaining a more informed view about the nature, extent and causes of poverty, taking greater account of the perceptions of the poor themselves. The problems of defining and measuring poverty over a period of time are well illustrated in Cebu. There have been a number of assessments of urban poverty in that city over the past decade. The official statistics show both a relatively low level of urban poverty and a decline in the level of poverty. Meanwhile, other assessments by independent researchers using multiple dimensions of poverty, including participatory assessments by the urban poor, show not only much higher levels of poverty but also that urban poverty is increasing. While differences in the incidence of poverty can be explained in terms of where the poverty line is drawn, differences in the trend are more difficult to explain. Thus, data on poverty always have to be treated with great caution.
Source: 1990s data from World Bank, 2001b, Table 2.6, using urban population data from UNHabitat, 2001a, to calculate the last column; 1970s data from Pernia (1994, p24)
although the absolute numbers living in poverty did not decrease (and in many cases went up) as a result of urban population growth. More recently, the proportion of the urban population living in poverty has probably been increasing, with the absolute numbers in poverty increasing substantially, as a result of structural adjustment and the economic downturn in Asia. Table 2.1 presents basic headcount data for urban poverty in the case study countries (except Chile). This shows between 14 and 30 per cent of the urban population being below the nationally defined poverty line in the 1990s. In the light of the comments made in the previous section about the limitations of urban poverty data, it seems likely that these figures underestimate the extent of urban poverty. Table 2.1 also shows, for the three countries for which comparable data are available from the 1970s, that even where the percentage of the urban population below the poverty line has fallen significantly, the numbers in poverty can increase and increase substantially. This encapsulates the problem for city governments: even where they are successful in promoting economic growth and implementing policies which enable the poor to escape poverty, in-migration of the rural poor continues to add to the numbers of the poor in cities. Indeed, the greater the success of the city in dealing with poverty, the more attractive it becomes for the rural poor. City-level data on urban poverty, as already noted, are in most cases incomplete and out of date. For some cities, like Kumasi, Johannesburg and Visakhapatnam, even the most basic data on levels of urban poverty are unavailable, let alone data on trends or any more sophisticated measures. Box 2.1 summarizes the available data from the city case studies.
Kumasi. Again, no data are available on poverty in Kumasi. National level data for urban poverty (excluding the capital city) show a slight decline in those below the poverty line from 33 per cent in 1988 to 28 per cent in 1992 (Korboe et al, 1999, p187). However, the impact of continuing structural adjustment, together with economic mismanagement, is likely to have worsened urban poverty in the last ten years. Mombasa. The 1994 Welfare Monitoring Survey identified 24 per cent of households and 32 per cent of individuals as falling below the poverty line in Mombasa. Evidence from the Kenyan Participatory Poverty Assessment suggests that poverty has been getting worse (Gatabaki-Kamau et al, 1999, pp2526). Recife. Brazil is another country with a highly unequal distribution of income. For the country as a whole, the proportion of the urban population falling below the poverty line increased from 23 per cent in 1990 to 25 per cent in 1996 (Melo et al, 2001, Table 4). Over the same period, the share of national income received by the poorest 50 per cent of urban residents fell marginally from 12.8 per cent to 12.3 per cent. For Recife itself, 44 per cent of the population were defined as poor in 1991, while 30 per cent of household heads received less than the official minimum salary (ibid, Tables 11 and 12). Santiago. Among our case countries, Chile shows the greatest reduction in urban poverty, from 45 per cent below the poverty line in 1987 to 23 per cent in 1996 (the poverty line being defined as twice the cost of a basic basket of food sufficient to meet minimum nutritional requirements, Rodrguez and Winchester, 1999, pp3334). During that period, economic growth in Chile was sustained at over 6.5 per cent per year. However, the good performance during that decade has to be placed against the massive increase in poverty that took place in the 1980s, and the slowdown in poverty reduction that has happened since the mid-1990s. Overall, income distribution has slightly worsened during the 1990s. There are also large numbers who are just above the poverty threshold who are extremely vulnerable to any increase in unemployment. For Santiago itself, the proportion living below the poverty line fell from 33 per cent in 1990 to 15 per cent in 1998 (Dockemdorff et al, 2000, p171). But citywide statistics mask considerable intra-city variations: the proportion below twice the poverty line varies between municipalities from under 2 per cent to nearly 30 per cent. Visakhapatnam. Specific data are not available for Visakhapatnam, but figures for the state of Andra Pradesh indicate that the proportion of the urban population below the poverty line fell from 39 per cent in 1987 to 33 per cent in 1992 (Amis and Kumar, 2000, p190). In 1999, the State Government of Andra Pradesh indicated that 29.5 per cent of the states urban population should be regarded as being below the urban poverty line (Kumar and Amis, 1999, p19).
access to essential services, social exclusion and powerlessness. The impact of these aspects of deprivation, and the ways in which individuals in poverty experience them, vary according to gender, ethnicity, age and family circumstances. Some forms of deprivation are location-specific (eg access to services); others are particular to certain groups (the homeless, the elderly). Even within households the experience of poverty can vary: there is ample evidence of intra-household differences in poverty, particularly in terms of gender and age (eg Kyomuhendo, 1999).
It is important to recognize the differentiated nature of poverty, deprivation and vulnerability. It cannot be assumed that the poor all have the same needs and interests just because they are poor. Attempts to address poverty in a generalized way are likely to fail to address the particular needs of specific groups of the poor. For example, general pro-poor policies that help some may have no impact on the poorest. The fact that the poor have differing, and often competing, interests makes it difficult to build effective civil society organizations of the poor. Rivalries and conflicting interests undermine the solidarity that is needed to bring about political change. This is illustrated in a number of the city case studies. It is also important to recognize and understand the livelihood systems of the poor and of the various groups of the poor. Poverty interventions by city governments, non-governmental organizations (NGOs) and donors need to be built on this understanding (Jones, 1999). Considerable progress has been made in this area, but not much of this understanding has yet filtered through to those who make decisions in national or city government. Research among the urban poor has demonstrated the complex arrangements and networks which those in poverty use in order to survive. These networks often cross the urbanrural divide. A useful analytical approach here is the assetsvulnerability framework (Moser, 1998; Rakodi, 1999a). This identifies the various forms of capital natural, physical, financial and social which the poor make use of in maintaining their livelihoods, and the ways in which these forms of capital are vulnerable to risks and shocks. The framework enables an analysis to be made of how assets can be increased and vulnerability reduced. We will return to these issues particularly in Chapters 4 and 8.
increased poverty in the short and even medium-term, unless countervailing measures are taken. Poorly designed interventions, even where they are intended to help the poor, may adversely affect some by destabilizing their fragile livelihood systems. Nevertheless, there are plenty of examples of interventions, by national governments, city governments or NGOs, that have enabled many of the poor to access new income opportunities and improve their situation. Another version of poverty pessimism argues that economic and political lites always obstruct reforms that might benefit the poor. It is true that the rich will be quick to defend their interests against attempts to redistribute resources. But there are also enlightened voices among the lite, whether out of genuine concern, self-interest, or the need to demonstrate responsiveness to the poor majority in order to achieve election to office. As Moore (1999) argues, experience demonstrates that economic and political reforms can be achieved by strategic alliances between the poor and sections of the lite. There are examples from our case studies of the urban poor organizing themselves, sometimes with the support of NGOs, making their voice heard, building alliances, and thereby succeeding in making small but perceptible gains. The question explored throughout this research is how the urban poor have been able to take advantage of the opportunities open to them and use whatever avenues of influence they have to assert their claims. There are enough examples from the cities concerned to show that it is not inevitable that the position of the poor always worsens.
Urban governance
The term governance has been widely adopted in the discourse of international development in recent years, but with varying connotations. It is often associated with normative values, as in the term good governance which features so heavily in the discourse of donor agencies (Leftwich, 2000; World Bank, 1997; DFID, 1997). UN-Habitat (2001b), in its Global Campaign for Urban Governance, identifies a number of norms and principles: participation; decentralization; equity; inclusion; accountability; responsiveness to civil society; efficiency of service delivery; sustainability; security.
While there is widespread agreement on these principles (eg United Nations Development Programme (UNDP), 1997a, 2000b; Friedmann, 1998), the aim of the present research has been to identify what is rather than to prescribe what ought to be.
In this volume, we use the term governance in a neutral sense to refer to the range of relationships between civil society and the state (McCarney et al, 1995; Swilling, 1997). According to Halfani (1997, p147), Governance provides the institutional framework within which the civic public realm is managed. The United Nations Development Programme (UNDP) defines governance as:
the exercise of political, economic and administrative authority to manage a nations affairs. It is the complex mechanisms, processes, relationships and institutions through which citizens and groups articulate their interests, exercise their rights and obligations and mediate their differences. (UNDP, 1997b, p9)
It is, therefore, a much broader concept than government, which Stoker defines as the formal institutional structure and location of authoritative decisionmaking in the modern state (ibid, p34). The standard view until relatively recently was that governments had both the authority and the capacity to govern effectively and to implement their policies and plans. The increased use of the term governance is a recognition that this view of government is no longer valid, if it ever was, and that outcomes depend on the interaction between many actors. Power is fragmented, and not a state monopoly, so that plans and policies can only be achieved through a degree of consensus and partnership. According to UN-Habitat,
Todays governance takes place in a more polycentric system of actors in which the state is less dominant than before. The multiplicity of actors complicates policy-making since no single actor is legitimate enough to direct societal change. Consensus is no longer given by virtue of legitimacy granted to the states actions but must be socially constructed. This requires alliances, coalitions and compromises. (UN-Habitat, 2001b, pp6162)
Hyden (1992, p12) identifies four essential properties of governance: trust (between the various groups in society about the nature and purposes of political action, including the ability to cooperate across basic divisions in society); reciprocity (the quality of social interaction among members of a political community, including the formation and operation of associations); accountability (the effectiveness of the processes by which the governed can hold the governors accountable, without which trust and reciprocity cannot be sustained);
authority (effective political leadership which resolves citizens problems and sustains legitimacy in the public realm).
Hydens framework has, however, been criticized by Swilling as being too statecentric, under-representing the impact of non-state modes of governance over the distribution of urban resources (Swilling, 1997, p6). Our analysis of urban governance includes a whole range of actors and institutions, and it is the relationships and interactions between them that determine what happens within the city. The actors and institutions involved include: private sector businesses, both corporate and informal; civil society, including community-based organizations, NGOs, political parties, religious groups, trade unions and trade associations; and the whole range of governmental agencies of national, regional and local government, including traditional authorities where they exist. These are illustrated in Figure 2.1. Urban governance also, crucially, involves individual citizens and households, of all income groups, inasmuch as they have any influence over what happens.3 Within this, city (or municipal) government is but one element, albeit often the largest and most obvious.
Central government
Utility companies
Traditional authorities
If government (including the local agencies of national/state government as well as city government itself) constitutes one large piece of the picture, civil society constitutes another. The term civil society has also been widely adopted in the discourse of international development, but with varying meanings (Mitlin, 2001b; Robinson and White, 1998; Friedmann, 1998). Here we take it to mean associations for common purposes or action, outside the direct control of the state. It includes not just formal organizations, such as NGOs, political parties and trade unions, but a whole range of more informal networks and groups, including community-based organizations, whether formally constituted or not. Civil society is often seen as a counterweight to the state, and an active civil society as essential to the achievement of both good governance and propoor policies. However, civil society encompasses a great variety of groups with widely differing objectives, many of which may have no concern with equity or the poor. Indeed, many may have explicitly anti-poor objectives of protecting privilege and vested interests. Nor are civil society organizations necessarily democratically accountable or characterized by the principles of good governance identified above. These issues are discussed further in Chapter 7. Within this conceptualization of urban governance, informal relationships are at least as important as the formal. Indeed, it is clear that what actually happens in cities is determined by a multiplicity of informal decisions much more than by the formal decision-making processes of city government. But it is also clear that these relationships are not between equals: differences in political power and influence reflect differences in income, wealth and economic power. Dealings between those with differing power and status often take place on the basis of patronclient relationships, in which the poor may be able to obtain certain benefits but from a position of continued dependence. According to Porrio, Urban governance ... deals with the power relationship among different stakeholders in cities (Porrio, 1997, p2). She goes on to identify two key questions: What is the nature of the relationship among stakeholders at certain political and economic conjunctures? What are the forms of negotiation that strengthen or weaken the position of stakeholders in urban governance? (ibid, p2).
In this research we have sought to explore the influence that the various actors, interests and institutions have on decisions and outcomes, and the impact of these decisions and outcomes on the poor. This has involved looking at the relationships between the various actors, interests and institutions involved, the processes of decision-making, and what gives the various actors and interests power and legitimacy, as well as what constrains that power. This is illustrated in Figure 2.2. Decision-making involves not just the formal processes, but the myriad of informal processes by which resources are allocated, access is achieved and development takes place. In particular, we are concerned with processes and mechanisms, both formal and informal, which enable the poor to have influence over, and thereby to benefit from, the decisions and actions of the various institutions of city governance.
Outcomes
Globalization
Globalization is hardly a new phenomenon, but recent developments in technology and communications have brought about far-reaching changes in the patterns of global production and trade, and in the networks between (and within) states. Global competition, and the ease with which capital can flow around the world, have created both opportunities for, and pressures on, cities. Cities no longer relate solely or mainly to their national context but also to international markets and institutions (OECD, 2001).4 They have been obliged to become more flexible and entrepreneurial, in order to attract inward investment, as well as sustaining local businesses. Bangalore has sold itself as the high-tech capital of India, while Cebu promoted itself as The Philippines Best Seller, and the economic phenomenon of the South. Successful responses by cities to globalization have brought benefits to many. The citizens of Cebu prospered considerably during the boom years, and economic growth in Santiago has had a major impact on poverty in the city. But globalization also brings vulnerability, as the inflow of capital which creates employment can just as quickly flow out again.5 Nor are the benefits equally spread. Just as economic growth benefits some more than others, globalization tends to reinforce existing inequalities. In order to make cities attractive to private capital, resources are invested in facilities that benefit those who make investment decisions. Urban space becomes increasingly segregated, often in the name of security, with privately managed shopping malls and gated housing
developments for the rich that exclude the poor (UN-Habitat, 2001a, pp3238). While some of the poor benefit through employment, others do not, and social exclusion increases. Globalization strengthens the power of global capital and multinational companies at the expense of labour. Local products and culture are replaced by standardized offerings of the multinationals a process often referred to as McDonaldization. Under globalization, while capital is free to move, labour is constrained by border restrictions. The main option for labour seeking to improve its economic opportunities is to migrate to the cities. As cities succeed in responding to the global economic opportunities and experience economic growth, more people and particularly poor people move to the city. While this process may help to reduce poverty nationally, it means cities face a neverending task of meeting the needs of the poor. And while individual countries and cities may have achieved benefits from globalization, the overall trend worldwide has been to greater polarization. In 1996, UNDP reported that the share of global income received by the worlds poorest 20 per cent had declined from 2.3 per cent to 1.4 per cent over the previous 30 years, while the share consumed by the richest 20 per cent had increased from 70 per cent to 85 per cent (UNDP, 1996, p2). One further feature of globalization is that it can undermine city governance as cities compete to offer better facilities, lower taxes and easier regulatory regimes, in what has sometimes been described as a race to the bottom. This process erodes local accountability, as decision-makers become more concerned with the interests of investors than the needs of their citizens. The city case studies identify a number of examples where projects designed to attract international capital have had a deleterious effect on the poor.
Decentralization
Decentralization has been a major feature of the development landscape since the mid-1980s. The World Bank observes that 95 per cent of democracies now have elected sub-national governments, and that governments everywhere are devolving powers to sub-national tiers (World Bank, 2000, p107). But decentralization is not new: many colonial regimes created or strengthened subnational units as a means of extending control, often using indirect rule through local traditional rulers (Mawhood, 1993; Mamdani, 1996, Kasfir, 1993). During the post-independence period, many countries reversed such arrangements, which they regarded as colonial impositions, centralizing government responsibilities in order to consolidate power and build their nations. This centralization of government was underpinned by economic models of state regulation and central economic planning. The recent wave of decentralization has been driven by a number of factors. First, the failure in so many countries of the central state and central economic planning to deliver the desired results and to be responsive, particularly to those at the periphery. Second, the political demand from below for greater local autonomy in decision-making. This was particularly strong in Latin America with the re-emergence of democratic governance in the mid-1980s, and in
central and eastern Europe, with the political liberalization that followed the fall of the Berlin Wall in 1989. Third, in a number of countries, such as the Philippines, Ghana, South Africa and Uganda, decentralized governance was seen (from the centre) as essential to unifying the state following the transition from autocratic rule or civil strife. There have been other reasons, too. In China, economic decentralization was seen as a means of promoting economic growth, although political control still remains highly centralized.6 In Indonesia, the threat of secession by resource-rich or ethnically distinct regions precipitated a radical if belated programme of decentralization. Threats of secession were also a stimulus for decentralization in Russia and Ethiopia. In a number of places, particularly in Africa, the decentralization agenda has been driven by international donors, as a means of trying to improve service delivery at the local level. The arguments for decentralization revolve around three main issues. First, administrative or practical considerations: that it is simply not possible for governments to make decisions at the centre about the detailed management of services and development at the local level. While this issue could be addressed through a deconcentrated administration (ie local offices of the central government), there would immediately arise questions about the accountability of decisions of state officials at the local level, bringing demands for political decentralization.7 Second, economic considerations: that if decisions about local services and development are made locally, reflecting local needs and preferences, this should result in a better allocation of resources and hence greater economic efficiency. Matching decision-making jurisdictions to the optimum areas for service delivery (within which costs are internalized) is the core of the fiscal federalism case for decentralization (Oates, 1972). Third, political considerations: that, in a democratic state, local people demand the right to make decisions about those things that affect their locality. There are counter arguments. First, decentralization could result in the fragmentation of the nation state a particular issue for new and fragile states. Second, local decision-making could undermine the achievement of broader, national goals and common objectives. Third, it could result in economic inefficiency, through the loss of economies of scale (a particular issue with many, very small municipalities created in some central and eastern European countries) and inter-jurisdictional rivalries (an issue with the protectionist development strategies pursued by some provinces in China). Fourth, decentralization could result in increasing inequalities, as rich regions are free to prosper while those in poor regions remain disadvantaged. Fifth, excessive fiscal decentralization (that is, the transfer to sub-national governments of resources and decision-making over taxes and borrowing) could undermine the fiscal position of the central state. There is some evidence of this in both Latin America (Prudhomme, 1995; Tanzi, 1995) and China (World Bank, 2000, p113). Sixth, decentralization may not deliver the intended benefits because of limited technical and managerial capacity at the local level, inadequate systems of local accountability and the increased scope for corruption. There are, of course, ways of mitigating these negative effects through properly constructed systems of inter-governmental relations, including
appropriate assignment of responsibilities, regulations regarding local decisionmaking and carefully designed fiscal transfer regimes. Many of the weaknesses of recent decentralization programmes arise because of deficiencies in the system of inter-governmental relations. Evidence about the overall impact of decentralization is difficult to establish, because of problems of measurement, comparability and the direction of causality. On the whole, though, it seems that decentralization has neither delivered the significant economic gains some claimed it would, nor has it created the economic and political problems feared by others.8 For better or worse, decentralization has focused attention on city governance. Specific issues for this study are whether decentralization has increased the resources available for poverty reduction, and whether it has enlarged the political space for the urban poor to make their voice heard and to improve their position. In relation to the first issue, decentralization certainly has increased resources for some cities, notably Recife, Cebu and Kumasi. Whether or not these have been effectively directed to addressing poverty is another matter. We will return to this point in Chapter 6. It would be nave to assume that decentralization automatically ensures that local decision-making reflects local needs and priorities, particularly of the poor. Certainly there is greater knowledge at the local level about local conditions. But elections at any level are a crude mechanism for ascertaining priorities, so that bringing decision-making closer to citizens does not guarantee more responsive and accountable decision-making. Meanwhile, more specific mechanisms of citizen participation are often extremely weak and fail to include the poor (Devas and Grant, 2003). But this is not to say that the poor cannot make use of the enlarged political space at the local level despite the limitations of the mechanisms available. Elite capture is a term widely used (eg World Bank, 2000, p109; Manor, 1999) to suggest that decentralization can adversely affect the poor by surrendering decision-making to a local lite. Manor (1999, p91) has argued that local lites are most unlikely to be more benevolent than those at higher levels. However, there has been little in the way of research on decision-making at the local level or on the behaviour of local lites. Schneider (2002) argues that political decentralization decreases representation of the interests of the poor by reducing the scope for influence at the national level and obliging them to spread their limited organization resources across jurisdictions. By contrast, the rich are able to apply political leverage at the local level without having to organize. From cross-national data, he concludes that there is some evidence that, while administrative decentralization is associated with pro-poor policy (defined as a high proportion of social expenditure in total public expenditure), political decentralization has a negative impact on pro-poor policy. However, von Braun and Grotes (2000) comparison of human development indicators with the extent of political decentralization concludes that decentralization can be beneficial to the poor under the right circumstances, and that it is political rather than administrative decentralization that makes the greatest difference. This is because of the greater scope for the poor both to hold elected officials accountable and to influence public spending decisions when the poor are
concentrated in certain jurisdictions. In the end, however, everything depends on detailed arrangements for decentralized governance, democratic control and accountability, as well as the form of resource transfers. Craig and Porter (quoting Rodrik, 2000) note that crossnational statistical studies suggest that decentralized participatory democracies provide for better economic growth, greater predictability and stability , and deliver superior distributional outcomes (Craig and Porter, 2003, p66). However, Crook and Manors (1998) study of four states in south Asia and west Africa found that, while decentralization had increased participation in all cases, and had increased the performance of government services in some cases, it had no discernable impact on the responsiveness to poor and vulnerable groups. Blairs study of the impact of decentralization in six countries concludes that DLG [democratic local governance] initiatives have encouraged participation, and have increased representation, but they have provided little in the way of empowerment, and even less in making the distribution of benefits more equitable or reducing poverty (Blair, 2000, p25). The present research does not claim that decentralization per se has an impact on urban poverty, one way or another, but instead looks at how city governance now accorded a more prominent position as a result of decentralization affects urban poverty. Nevertheless, we see evidence in a number of the city studies that bringing decision-making closer to citizens opens up greater opportunities for the poor to have an influence and to make at least incremental gains. However, the impact of claims made at the local level may be only localized and piecemeal, often reinforcing clientelistic relationships, rather than being systemic and sustained. Among our case countries, decentralization has probably had greatest impact in Brazil and the Philippines. In Brazil, the transfer of responsibilities and financial resources has enabled municipal government in Recife and elsewhere to develop a range of social programmes and to introduce participatory budgeting. Decentralization in the Philippines in the late 1980s devolved a range of functions and resources to municipal governments, increasing the scope for cities like Cebu to intervene on behalf of the poor. By contrast, in Ghana, decentralization has remained incomplete: although some responsibilities and resources have been transferred, others remain under the control of the centre, with the appointment of the key officials still resting with central government. In South Africa, there has been a radical restructuring of sub-national government, rather than decentralization in a strict sense, since the former white local authorities have long enjoyed a high degree of local autonomy and substantial resources. By contrast, for those living in the former black local authorities (an imposed and contested form of local government), the reforms of the 1990s have opened up political representation and access to resources. In our other case countries, decentralization, in as far as it has happened at all, has had much less impact. There has been a long-term erosion of local government responsibilities and resources in Kenya, although recent reforms have begun to address that. Cities in India have not experienced significant changes as a result of decentralization, despite some protection of their political
status under the 74th Constitutional Amendment. Indeed, in some cases, there appears to have been an erosion of the role and functions of the municipal government in favour of state agencies and semi-autonomous boards.
Democratization
Probably the most significant change to affect city governance has been the emergence or restoration in many countries of democracy at the national and local level. In some cases, this has been the result of protracted political struggles, helping to place a greater value on the democratic institutions created. This is not to suggest, as some have done (Chalker, 1991; World Bank, 1992), that liberal democracy is a precondition for either development or poverty reduction. But democracy does open up the political space in which the poor have the opportunity to make their voice heard. In the first post-apartheid elections in South Africa in 1994, the majority of citizens including most of the poor had, for the first time, a say in the choice of who would make decisions about the things that affected them. The response of the electorate was overwhelming. But things did not change overnight, and many have been disappointed with the slow pace of change. Those with the economic power still hold that power. Government resources are limited, and changing expenditure patterns in favour of the poor has proved very difficult. Meanwhile, some of those elected to represent the poor quickly lost sight of the interests of their constituents. Nevertheless, the political landscape has changed and the poor have a greater chance of being heard than ever they did under the old regime. Democracy has swept through large parts of the world since the mid-1980s: Latin America, central and eastern Europe, many countries in Africa and Asia, including the Philippines, South Korea and more recently Indonesia. Even in India, with its long established democratic tradition, lite control through the dominant party has weakened, and constitutional protection has been introduced for democracy at the local level. What is understood by the term democracy does, however, vary considerably, and there is much debate about what elements are essential. The former communist governments of eastern Europe described themselves as democratic and held regular elections, but few would now accept that these states met the conditions for democracy, including a choice of candidates and freedom of information and the press. However, the western model of liberal democracy presupposes certain values that are contested, such as the value placed on individual rights rather than social hierarchies and kinship relationships which are important in other cultures, and the emphasis on winning and majority rule rather than on consensus and inclusion. Understandings of democracy may clash over issues such as competition between political parties. Political parties provide the opportunity for voter aggregation around common sets of policies, rather than focusing elections on charismatic individuals. But all too often parties fail to develop clear electoral platforms, instead adopting populist rhetoric and reinforcing ethnic or religious divisions (as in Kenya and Zimbabwe, and increasingly in India and Indonesia). They exclude some interests and polarize debate on complex issues in ways that
are often unhelpful. The role of traditional authorities (such as chiefs) and traditional decision-making arrangements is another contentious issue. These often command wide respect from local people, despite not being democratic. They may also, in many cases, allow for different voices to be heard, even within a hierarchical framework. It is clear that democratization does not have any automatic effect on poverty reduction or social inclusion. Despite nearly 20 years of democracy in most of Latin America, incomes and wealth remain far more unequally distributed, and social exclusion remains far deeper, than in countries not noted for democracy, such as China and Indonesia. The emergence or return of multiparty democracy has been blamed in some countries for increased conflict, corruption and declining economic performance. Polidano and Hulme suggest that, while the new populist politics in India has resulted in increased popular participation and improved provision of services in rural areas, it has been accompanied by a marked decline in the quality of governance exemplified by intensified communal politics, corruption, the criminalization of politics and the politicization of the bureaucracy (Polidano and Hulme, 1997, p.6). Even in the so-called developed democracies of the west, it is apparent that the outcome of the democratic process is more often than not determined by the amount of money that candidates or parties can muster, or by the vagaries of the electoral rules. There is a never-ending debate regarding the relative merits of the majoritarian (or first-past-the-post) system and proportional representation, particularly over how far either system represents the interests of disadvantaged groups. Some countries (for example, India, Sri Lanka, the Philippines) seek to address this by guaranteeing representation of women and ethnic minorities through reserved seats. However, there are questions about how far those who occupy such seats succeed in advancing the interests of those they are supposed to represent (Blair, 2000). Elections often fall far short of the requirements for proper democracy, with limited understanding among voters about their rights and their choices; the absence of clear manifestos from candidates and information about the issues involved; and electoral malpractices, such as registration fraud, votebuying, vote rigging and even intimidation. Once elected, representatives may quickly lose interest in those they were elected to serve, concentrating instead on ways, corrupt or otherwise, of recouping the costs of their campaign. Powerful individuals often dominate local politics, maintaining their position through clientelistic control over political agents and local electors. Disillusionment with the performance of those they have elected creates voter apathy and low turnouts at local elections. Nor is it just the shortcomings of the electoral system. The outcomes of the democratic process for the poor also depend on how far those they elect are able to influence the decisions made, and how far those decisions determine what actually happens. All too often, so-called democratic decision-making structures are subject to excessive pressure from powerful interest groups, marginalizing the voices of those who represent the poor. There are important questions about the accountability of the executive to the elected representatives, about the checks and balances between elected (or appointed)
executives and the local legislature, and about the internal organization, functioning and accountability of the paid administration. A particular focus of concern is the budgetary process and the way that this does or does not convert the aspirations of elected representatives (and their constituents) into actual expenditure outcomes. These all emerge as important issues in the case study cities and will be discussed further in Chapters 5 and 6. Whatever the shortcomings of representative democracy, it does potentially allow the poor to have a voice, not least because they represent the majority in any city in the South. In Indian cities, it is notable that electoral turnout is generally much higher in poor wards than in higher-income wards. Whether for good reasons (perceptions of influence and vote-bargaining with candidates) or bad (navety, vote-buying or intimidation) the poor perceive voting as important. Of course, as we have already noted, the poor are not an undifferentiated mass. Divisions by race or ethnicity, gender, income (for example, the poor versus the very poor), economic activity, interest, location and so on, make it difficult for the poor to take advantage of their numbers in the democratic process. Much depends on their ability to organize and lobby. In this, the organizations of civil society NGOs, community-based organizations (CBOs), trade unions, religious groups and political parties can play a significant role. However, civil society is much stronger in some countries than in others. The stronger civil societies have often emerged from periods of political struggle. In the Philippines, NGOs enjoy a privileged position at both national and local government levels. This is also true, to a lesser extent, in South Africa. Of course, not all civil society groups represent the interests of the poor indeed, the most powerful ones often oppose the interests of the poor, and the institutions of civil society tend to reflect and reinforce existing unequal power relationships. There are also, in many countries of the South, numerous socalled NGOs which are little more than briefcase operators that exploit the poor for their own interests. Even where the role of civil society has been crucial to the political struggle, once gains have been made there is a tendency for reduced activism and for NGOs to become incorporated by the state as deliverers of services. This tendency has been particularly marked in Chile, where greater prosperity and the achievement of political rights have been accompanied by a retreat from public socio-political space into private space (Rodrguez and Winchester, 1999, p2).9 There has been a similar, if less marked, tendency in South Africa as many of the civil society activists from the era of struggle have made the transition into government. We will return to the role of civil society in Chapter 7.
Conclusion
The lack of accurate, disaggregated data on the level and trends in poverty at city level is a serious impediment to effective approaches to urban poverty reduction. So too is the lack of understanding by city officials about the differentiated nature of poverty and of the livelihood systems of the poor. These issues will be explored further in the next chapter.
A broadened understanding of urban governance enables us to begin to trace the roles of the private sector, civil society and the various institutions of government at city level and of the interactions between these actors in determining what happens at city level. Globalization, decentralization and democratization have all had significant impacts on city governance, although much also remains unchanged. Social exclusion, the vulnerability of the poor and their lack of voice remain, and in some cases have become more severe. But the situation is not hopeless, and there are examples of where the poor have made gains, albeit often only incremental and still fragile. The remaining chapters of this book will explore these issues further, drawing on the city case studies, to enable us to understand better the relationships between urban governance and poverty.
Notes
1 Satterthwaite (1997, p1) points out that the World Banks 1988 estimate of 300 million poor people living in urban areas implies that less than a quarter of the developing worlds urban population at that date were poor, a figure far lower than that suggested by various studies of cities in Africa, Asia and Latin America. One reason for this underestimate is that poverty lines are generally based on minimum food requirements, whereas urban living requires considerable non-food expenditure. Based on an urban population of the developing world of 1.94 billion (UN-Habitat, 2001a, Table A2). It could be argued that individuals and households have no power on their own and therefore cannot be considered as part of urban governance. Yet it is clear that individual and household decisions about where to locate, where to trade and how to move about has a profound impact on the way cities develop and function. At the very least, their resistance to complying with the official regulations limits the ability of city government to govern effectively. Sassen (1994) has argued that globalization has reinforced the dominant position of certain global cities, concentrating the power of global capital and effectively controlling the new international division of labour. The implication is that there is little that local actors can do about that situation. Whilst clearly some cities do possess much greater natural and historical advantages than others, there is much that is debatable about what constitutes a global city, about the exclusiveness of the concept, and about the room for manoeuvre which city-level actors have (Vidler, 1999). The process can also occur within cities: for example, the shift of business activity from Johannesburgs central business district (CBD) to the suburban centre of Sandton, which was perceived as being safer, is now being superseded by relocation even further out to Midrand. Sub-national governments in China actually have a considerable degree of autonomy, in practice if not in principle, especially those with strong local economies (Shirk, 1993). It is common to distinguish between administrative decentralization or deconcentration, which transfers tasks to field offices of central government without local autonomy, and political decentralization or devolution of powers and resources to elected local governments with a degree of local autonomy. A third
2 3
6 7
Chapter 3
Economic Growth, Urban Poverty and City Governance
Ursula Grant 1
Introduction
Do the poor benefit from urban economic growth and if so, how? Conventional wisdom suggests that everyone or almost everyone should gain from economic growth, including the poor, if only through a process of trickle down. Yet we know that this is often not the case, and that there is no simple relationship between growth and poverty reduction. Nor have all cities been experiencing sustained economic growth: in many cities, the context is one of structural change or even economic decline rather than economic growth. What are the implications for the livelihoods of the urban poor in these circumstances? As Chapter 2 has indicated, the experience of urban poverty is highly differentiated: some of those living in poverty may gain from economic growth;
others may be disadvantaged, particularly through increasing inequality and exclusion. What happens to particular poor individuals or households depends on a whole variety of circumstances, many but not all of which are outside their control. This chapter is concerned with the process of urban economic growth (or more broadly, economic change, since not all cities are growing), and the extent to which the poor benefit from, or are disadvantaged by, that process. It is also concerned with what city governments can do about both the economic condition of their city, and about the impact of economic change on the poor. In the next section of this chapter we consider whether economic growth (or change) has benefited the poor in the case study cities. In order to do this, we examine the economic structures and trajectories of the cities, what may account for economic change in each case, and what impact this has had on the poor. This is followed by a more detailed analysis of two important aspects of the relationship between economic growth and poverty. The first is the operations of the labour market, including the ways in which the informal sector and the local economy provide economic opportunities. The second is the ability of the poor, through their livelihood and coping strategies, to take advantage of the economic opportunities available.2 The last two sections of the chapter review how the institutions of city governance in our case study cities have used the room for manoeuvre which they have to address these issues: first to promote economic growth and second to enable the poor to benefit from that economic growth (or to protect them from the negative impact of economic change or decline).
their productive potential has been central to the growth process. The common denominator of successful growth policies in the countries of east Asia has been that they enhanced the productivity of the poor, rather than relying on welfare transfers to raise incomes (ibid, p45). This chapter attempts to localize and urbanize the debate about economic growth and poverty reduction. In the literature, there are few examples of successful urban growth strategies that have resulted in subsequent poverty reduction. This is partly a reflection of the inadequacy of data at the urban level. This chapter begins to fill that gap while at the same time acknowledging the limitations of the available data. We adopt a multi-dimensional approach to poverty, drawing largely on Mosers assets/vulnerability framework (Moser, 1998). This framework recognizes that, for poor people, labour represents their most important asset. It is through increasing work opportunities and returns to labour that poor households benefit most directly from economic growth. According to the World Bank, the rate of poverty reduction is, on average, proportional to the distribution-corrected growth of private consumption (World Bank, 2000, p274). The evidence to support this assumption from the ten cities examined in this book is mixed. A number of cities (Ahmedabad, Bangalore, Cebu, Santiago) clearly show some reduction in poverty with economic growth. This tends to be connected to the performance of the national economies within which they are situated. The economies of Chile, the Philippines and Sri Lanka have all been growing quite strongly during the 1990s, but Santiago, Cebu and Colombo have all outperformed their national economies. Johannesburg seems to have been stagnating in line with the South African economy. The Indian cities included in this study are all located in growth states and tend to perform above the national average. Ahmedabad, Bangalore and Visakhapatnam all provide modest support to the idea that economic growth increases income levels for the poorest. The other two African case cities, however, are located in countries characterized by long-term economic difficulties. Mombasa shows little evidence of either growth or poverty reduction, and in Kumasi only slight improvements are evident. This partly reflects the limited local management capacity to create an environment for economic growth, but also the dearth of inward investment in sub-Saharan Africa. Mombasa seems to be declining along with the Kenyan economy as a whole, due to a number of factors discussed in Box 3.1. Processing and textile industries have been damaged by economic liberalization, while tourism has been hit by political unrest and insecurity. In Ghana, the disastrous economic policies of the 1970s and 1980s hit the urban poor particularly hard, but by the late 1990s, Kumasi seemed to be experiencing a modest recovery in tandem with the Ghanaian economy. It is not entirely clear what impact this growth has had on the poor, but it seems that at least some of the urban poor in Kumasi have benefited from the overall stabilization of the economy as a result of structural adjustment. Although measurement and data problems pose difficulties in assessing the relationship between urban economic growth and poverty, it is clear that this
relationship is complex. The distribution of benefits from economic growth is mediated by social and political as well as economic factors. While the general increase in incomes may result in a reduced incidence of absolute poverty, inequality limits the extent to which this benefits all poor people. In Santiago, for example, although economic growth has been accompanied by impressive reductions in poverty, it has also resulted in increases in inequality, spatial segregation and stigmatization, as outlined in Box 3.2.
Other cities also suggest mixed outcomes. In Colombo, relatively high rates of economic growth during the early and mid-1990s were accompanied by reductions in unemployment, but little in the way of a reduction in poverty indeed, some of the poverty indicators suggest an increase in the depth and severity of poverty (Fernando et al, 1999, p18). In Johannesburg, the very modest rate of economic growth during the 1990s was accompanied by indications of increasing poverty, particular as a result of unemployment. The performance of the South African economy has been relatively poor over the last 20 years, with per capita incomes rising by an average of only 1.6 per cent per year between 1980 and 1995 (Standing et al, 1996). This is despite a shift from the import substitution policy of the 1960s and 1970s to a more export oriented policy in the 1990s. In Johannesburg, slow employment growth and changes in employment away from primary sectors towards services have been accompanied by rapid urban population growth in the post-apartheid era. The legacies of apartheid include widespread poverty and high levels of inequality. These have been reinforced in recent years by a process of white capital fleeing to suburban areas. Changes in employment patterns include a move away from primary sectors, notably mining, to services. Although manufacturing remains important, manufacturing employment has declined. The fall in demand for labour has resulted in lower wages for unskilled workers and an unemployment rate reaching 20 per cent (or as high as 33 per cent, depending on definitions) by the mid-1990s (Crankshaw, 1997). New entrants to the labour market, predominantly the young and African workers, have suffered particularly. It is now widely acknowledged that the impact of economic growth on poverty has much to do with the level of inequality within a nation. The same is likely to be true at the city level. According to the World Bank:
Improvements in private consumption per capita are generally associated with a reduction in poverty, but where the distribution is unequal, the poor may not share in the improvement. (World Bank, 2000, p234)
Addressing inequality is much more difficult than promoting economic growth. It is not simply a matter of transfers, even if there were the resources available for these, it is also about increasing the participation of the poor. Policy and investment choices are important, particularly investment in human capital. However, a nuanced approach is necessary to ensure that it is the productive assets of the poor that are developed and utilized, and that inequality in access to, and quality of, service provision is addressed. These are issues to which we will return below. Thus, urban economic growth appears to be a necessary but not sufficient condition for urban poverty reduction. With the general trend to decentralization and local democracy, the room for manoeuvre at the local level has increased, at least in some respects. But the ability of city governments to respond to economic opportunities and to encourage pro-poor growth depends upon the wider context of globalization and structural adjustment on the one hand, and non-economic factors, such as the citys social structure, cultural and religious traditions, and the political order, on the other. The attractiveness of
urban areas to investors, and the ability of the urban population to take advantage of the economic opportunities are, in turn, related to the soft infrastructure healthcare, education, and so on (Harris and Fabricius, 1996, p91). Both local politics and the specificity of urban areas are important in all of this and are further explored in later sections of this chapter.
City economic growth and poverty reduction: 3 Labour markets and economic opportunities
What are the mechanisms through which the benefits of urban economic growth are transmitted to the poor, and how do poor individuals and households respond to the opportunities that emerge from economic growth? The most powerful mechanisms through which economic growth influences poverty are the labour market and employment growth, both formal and informal. If economic growth is to improve the living conditions of the poor it must be inclusive, providing a broad range of remunerative employment. Successful formal sector growth, notably in the manufacturing sector, is often regarded as a prerequisite for poverty reduction (World Bank, 1990; Watkins, 1998). It is the changing patterns of manufacturing and service employment, away from the high wage, unionized and male-dominated industries and business sectors and towards low wage, predominantly female light industries and support services that Sassen (1994) claims can account for the dramatic increases in inequality seen in the cities of the north. As already noted, most of the cities examined in this book have experienced some decline in their manufacturing sectors, while at the same time the service sector has expanded. The impact of this on urban poverty is seen in the changing nature and conditions of the labour market that benefit some and not others. In Santiago, this pattern largely reflects major transformations occurring in the Chilean economy, with the city becoming one of the main financial centres of Latin America. Johannesburg has experienced a similar trend but generally in a condition of economic stagnation. Between 1970 and 1991, manufacturing in Johannesburg declined from 24 per cent to 16 per cent of total employment, while mining declined from 5 to 2 per cent. During the same period, employment in services increased from 29 per cent to 33 per cent, with financial services increasing from 8 to 13 per cent (Beall et al, 2000, p108). The explanation is partly due to the exhaustion of the gold mines in the area, but also to a more general process of urban de-industrialization. In Cebu, despite its success in encouraging inward investment and export-led industrialization, employment in the manufacturing industry increased only slightly in absolute terms, and declined as a percentage of total employment from 24.4 per cent in 1988 to 20.4 per cent in 1998.4 Nearly three-quarters of the citys employment is accounted for by the service sector (Etemadi, 1999, Table A2.0).5 Macro-economic policy has an impact on the returns to labour, through its support for either capital or labour intensive investment patterns. It is generally believed that labour intensive growth is critical for poverty reduction as it
provides more opportunities for the poor to make use of their primary asset, labour. Yet few, if any, of the cities studied appeared to be following labour intensive growth strategies as such. Colombo is perhaps the only one that shows some clear design in promoting labour intensive growth through export production. The Sri Lankan government and the Greater Colombo Economic Commission have actively sought to encourage this sector. Nationally, employment in both manufacturing and services increased between 1980 and 1996. One approach is through the direct creation of employment in the form of labour intensive public works programmes. Such an approach has long been used in rural areas of the developing world both to create employment and to provide infrastructure (notably roads). More recently, the approach has been used quite widely in urban areas of francophone Africa (Farvacque-Vitkovic and Godin, 1998). Box 3.3 provides an example of one such programme, in Benin. This programme has been used to create employment, to build capacity within local construction firms and to provide much needed urban infrastructure. However, there are also weaknesses, notably the fact that the approach is top-down and engineering-driven, implemented by a special purpose agency with little connection to the city government. An alternative strategy, through the encouragement of high-tech industries, raises questions about the impact on employment. This is exemplified by Bangalore, where a US$200 million high-tech park was established on the edge of the city, modelled on the American industrial estate and highly subsidized to new business through generous tax concessions and guaranteed power connections. Yet, after several years, many of the units remain vacant and few employment opportunities have been created for the poor. Meanwhile, the effect on land values has been to force out small enterprises and poorer groups in favour of large real estate developers. This failure can be attributed to a lack of understanding by those involved of the way in which much of Bangalores software industry operates that is, through complex networks of microenterprises working in the front rooms of homes in the middle-income districts of the city. Benjamin (2000) asserts that it is these local economies that form the employment base of urban areas in India. Despite the importance and visibility of the high-tech sector it remains relatively insignificant in terms of total employment in the city indeed, the silk and garment industries are more significant. Furthermore, it is in the ancillary industries that the greatest employment growth has been generated in both these sectors. The strength of the labour market as a mechanism to absorb and influence the productivity of the poor has been particularly strong in south and southeast Asia. In the fast growing cities of that region, significant employment increases are evident. In Visakhapatnam, for example, growth has been associated with its large and successful port. The increase in the registered industrial workforce has been combined with a significant improvement in the casual daily wage rate, which doubled between 1993 and 1998. In most Indian cities there are large casual labour markets where individuals turn up seeking work and where contractors go to employ labour. The daily rate for casual labour therefore underpins an entire local labour market, reflecting local economic conditions,
and is a key mechanism through which the benefits of growth may or may not be passing to poor workers. We are concerned not only with how economic growth impacts on the incomes of the poor but also with the ways in which poor people are able to access and utilize the opportunities created by economic growth. Assets such as human and social capital are key to these processes. Ill health, for example, plays a major role in reducing the productivity levels of poor people. The Visakhapatnam study shows how ill health and the associated lack of strength reduce the capacity of workers to take on available work. Rather than being lazy, as is often assumed, these workers simply need to take breaks to recuperate between jobs. Those in poverty usually lack the education and skills to access employment opportunities in the formal sector. Accordingly, human capital deficiencies undermine the ability of workers to overcome poverty as well as hindering broader economic growth. Human capital is an important contributor to welfare changes over time (Grootaert et al, 1995). The longer people are held in poverty, the harder it is to access opportunities. The positive effects of human capital investments at the household or individual level are only felt over the medium- to long-term. Hence, the poor may make choices in favour of immediate needs which contribute negatively to the perpetuation of their low levels of human capital. For instance, rather than attending school, street children in Colombo and elsewhere are engaged in begging and other work. Similarly, the poor in Kumasi reported difficulties in financing their childrens education and withdraw them from school during times of hardship. Even when children do attend school they often arrive late due to morning queues for taps and latrines. The effects of inadequate service delivery can similarly impact directly on the productivity of informal economic activities. For example, women and girls in Kumasi spend a large proportion of their time collecting and storing water for food preparation. They use private taps at high cost or sources such as polluted local streams, risking associated health problems. Similar patterns can be found in many cities around the world. Inadequate service provision directly constrains the capacity of poor people to overcome vulnerability. In poorly serviced areas people use more time and other resources in seeking alternative service provision, often of poorer quality and higher cost. These higher costs often lead to increased indebtedness. Inadequate service provision also reduces their ability to prioritize longer-term asset accumulation, for example through education and health care, which in turn constrains their prospects of rising out of poverty. These are issues to which we will return in Chapter 9.
that city, the bulk of employment generation is constituted by small and tiny enterprises, which cluster to share functional links. For example, in Mysore Road in west Bangalore, an intensive mix of industrial, fabricating and service activities are centred around a host of small enterprises. It is argued that close proximity and high densities allow economic and other linkages to develop between various local groups, thus forming a large and dynamic sector. The diverse and complex ways in which the local economy develops contrast with the corporate sector, located within the master planned areas of rigid land-use controls from which small-scale enterprises (SSEs) and the poor are generally excluded. Bangalore has become known as the silicon valley of India, with the development of software and high-tech industries, but it is the growth of the small businesses that have developed alongside the formal high-tech sector that has provided the majority of the citys employment growth (Benjamin, 2000). In the case of Cebu, while the national economy of the Philippines was experiencing negative growth during the late 1980s, Cebu City experienced strong economic growth, accounting for 10 per cent of the nations exports in 1994. However the Cebu boom benefited only certain sectors and it was, as it still is, the informal sector that absorbed large sections of the urban poor work force (Etemadi, 2000). It is the informal sector that offers the main income opportunities for poor households since barriers to entry are lower. Increased employment of household members within the informal sector is often required during times of hardship, perhaps as a result of a shock such as the illness or death of a family member. Coping strategies include the entry of previously non-earning household members into the informal sector, and wage earners taking up supplementary activities. In Mombasa, women are disproportionately represented in the informal sector, mainly selling foodstuffs, brewing and so on. Children also play an important role in contributing to the informal strategies of households. However, those engaged in informal sector activities face many obstacles: inadequate access to credit and appropriate sites or premises; bureaucratic licensing requirements and regulatory restrictions; and demands by the police for bribes to operate without licences or in non-permitted locations (Rakodi et al, 2000). Similar patterns are found in each of the ten cities. The local or informal economy often involves the use by individuals or households of their homes or land as productive assets, for example by the conversion or addition of rooms to provide space for economic activities. Letting out rooms previously occupied by the owners family is one coping strategy during times of hardship. Relaxation of the enforcement of the planning laws in South Africa has resulted in the construction of backyard shacks in the townships and in the use of housing as productive capital. As a result, the townships are less homogeneous and the use of public and private space has changed. Prohibition of trading in townships has been ended and hawkers, retail outlets, taxis and so on are now an important part of local life. In Bangalore, small-scale weavers such as those belonging to the traditional weaving caste (Devangas) mostly operate from their homes, with minimal cost and easy access to markets for both inputs and finished products.
Successful small business development in Bangalore, both informal and formal, can in part be attributed to a land tenure system that is flexible, diverse, and allows for mixed uses. By operating outside of the formal planning process, with access to relatively cheap land with loose land-use regulations, small local enterprises can start up and develop in messy but effective ways. These locations provide proximity to markets and suppliers, allowing for financial and subcontracting linkages to develop easily. In turn, productive forms of social capital can form and significant employment can be generated. Controls over land and planning are, therefore, important tools that municipal governments can use to promote local economic activity. Conventional master planning can disrupt the productive activities of the poor, whereas linkages between formal and informal businesses can be enhanced through incremental development, mixed land use and diverse tenure. Similarly, planning regulations can affect how households are able to use housing as a productive asset. Indeed, as Moser (1998, p11) has suggested, the removal of tenure insecurity and other obstacles that constrain households from using their homes as productive assets may be the single most important poverty reduction intervention that urban governments could adopt. These issues will be discussed further in Chapter 8.
boomtowns, namely those cities which have seized upon post-industrial technologies to emerge as financial service centres, partially marketized cities and the marginalized city. The last refers to those that have failed to integrate in the global economy and are concentrated particularly in Africa. The African city is characterized by an absence of export-led industrialization, weak infrastructure and state apparatus, and a low skilled labour force, failing to engage effectively in global competition and becoming relatively poorer and more peripheral (Simon, 1992, pp4850). Vidler (1999) examines the agency of the local within the context of this global game. She asserts that a citys relative economic success is shaped in part by its specific starting point and by the resources potentially available to the city government. These starting points include a citys economic position in the national and international economy, its historical legacy, its political and institutional frameworks, and its local culture and social structure. Different levels and combinations of resources are available in each city, some of which are beyond the city governments control while others are potentially subject to their influence, such as infrastructure provision or the development of a skilled labour force. As Vidler argues, a citys starting point does not mean that its growth (or decline) is predetermined. Indeed, one of the keys to city economic growth is an appreciation on the part of city governments of the resources they have at their disposal, coupled with an understanding of how these might be productively exploited or developed (Vidler, 1999, p13). Cities face different challenges, opportunities and local circumstances, and therefore different levels of manoeuvrability or relative autonomy. Attracting corporate investment is often seen as crucial to urban economic growth. For this, city marketing efforts, sometimes in partnership with the private sector, seek to present positive images and gain international exposure. However, such an approach seems questionable where resources are lacking to provide even the most basic infrastructure. Indeed, there are risks that such promotion results in the destruction of livelihood opportunities for the poor, through repression of the informal sector for the sake of the citys image (as in Mombasa and Kumasi), or through pre-emption of well-located land for prestige developments (as in Bangalore). The challenge facing city governments is both to attract inward investment and to develop a skilled labour force while simultaneously protecting the productive activities of the poorest. One approach is to develop partnerships or alliances with other institutions such as private enterprises and community organizations for the purpose of developing and implementing growth strategies. Such growth coalitions or productive alliances can harness opportunities in ways that might not be possible for city governments on their own. The requirements for successful partnership include a reasonably buoyant and influential private sector, active community organizations and a suitably capable and authoritative local government, together with some common economic interest motivating them to join forces. Box 3.4 reviews the case of Ahmedabad, where a fairly progressive local administration has been able to negotiate innovative partnerships with business and community organizations for urban development. Cebu also provides examples of such alliances. Mombasa,
however, demonstrates how ethnic and political tensions can frustrate attempts to build alliances by undermining trust between local government and entrepreneurs, and between enterprises of different scales and types.
and social capital on which the economic activities of the poor depend. There are examples of where city governments have sought to address these issues, such as the provision in Johannesburg of street lighting to lengthen the day for trading and to facilitate socializing without fear. Upgrading of slum areas in Visakhapatnam has had a profound effect on local level economic activity by poor groups. But the case studies show that it is far easier for municipal governments to destroy jobs, livelihoods and social capital by ill thoughtthrough policies and actions, than it is to create or rebuild them. A first step, therefore, is to prevent such bad governance a point to which we will return in Chapter 8.
Notes
1 2 3 This chapter draws on an earlier working paper written with Philip Amis (Amis and Grant, 2000). A fuller analysis of the livelihood strategies of the poor will be provided in Chapter 4. It should be noted that the city studies did not focus on labour markets and employment as such, so the comments in this section are based on general observations from the city studies and elsewhere rather than on detailed analysis of what is occurring in labour markets in the cities concerned. Since 1998 was the time of the east Asian financial crisis, employment in manufacturing is likely to have been lower in that year than it would otherwise have been. However, it is important to note that statistics on employment may vary in terms of how they treat informal sector workers. This can affect the service sector in particular, since this is where the bulk of urban employment is located. Also, the service sector is sometimes used as a residual category. This might help to explain the very high proportion of workers counted in this category in Colombo, Cebu and elsewhere. A short, supplementary case study was undertaken in Benin, examining the role of employment creation programmes in the city of Cotonou. This work was carried out by Blandine Fanou and Ursula Grant in 2000. Benjamins concept of the local economy overlaps with the more widely used concept of the informal economy, but is rather broader, including much of the small-scale industrial sector, and without the pejorative connotations of illegality often associated with the latter concept. By contrast, the corporate economy covers the formally constituted, mostly large and medium sized enterprises. However, in India, the 74th Constitutional Amendment assigned to municipal corporations responsibilities for economic development and poverty reduction.
4 5
6 7
Chapter 4
Surviving in the City: Livelihoods and Linkages of the Urban Poor
Jo Beall
Introduction
Chapter 2 provided a discussion of how urban poverty has been defined and measured and how issues of definition and measurement can affect policies targeting the urban poor. It was pointed out that urban poverty is multidimensional and more dynamic than implied by poverty datum lines and static policy analyses. Indeed, poverty can be characterized by cumulative deprivations, and one dimension of poverty is often the cause of or contributor to another dimension (World Bank, 2000). The dynamic responses to poverty on the part of the urban poor themselves, combined with the complexity of urban institutional life, leads us to suggest that urban social disadvantage needs to be understood relationally, in terms of analytical constructs that explain processes of social exclusion and marginalization, as well as the relative powerlessness and absence of voice among the urban poor in respect of governance.
In this chapter we explore the livelihoods of the urban poor, understood not simply in terms of the strategies by which they survive and make a living in the present, but how they seek to secure their long-term future. In particular we look at their connectedness and the linkages they make towards these ends. The networks of the urban poor include those of mutual support among close family and wider kin as well as neighbourhood and community-level groups. They also include more extensive associations that assist in the access of information and resources, or that link them into the decision-making arenas of urban governance. The question is asked whether the social resources of the urban poor constitute a form of social capital that extends beyond mutuality and enhances urban governance.
example, the inadequacy of public toilets and having to go to the bush was an issue that loomed large for people. Access to land and housing is a critical element in combating urban poverty and achieving livelihood security and is an area that is mediated most particularly by commercial transactions. As a productive asset, housing is a vehicle by which home-based work can be generated. It can be used as collateral for accessing credit and can generate a rental income, which may be important as the only or as a supplementary income for poor urban households (Beall, 2000b). As important, if not more so, is the need for housing as shelter and as a social asset without which it is difficult to fully participate in society. In Cebu, for example, the poor were identified as squatters or renters living in makeshift shanty houses. In both social and economic terms, security of tenure was perceived to be important for livelihood strategies in the present and for greater livelihood security in the longer run. In the urban context, locality and place are important dimensions of the way in which poverty is experienced. Distance from work opportunities or proximity to environmental hazards can play their part in urban social disadvantage and this was found to be the case in a number of our city studies. For example, in Bangalore, location was shown to critically influence the ability of poor people to secure livelihoods. In Johannesburg, the spatial legacy of apartheid saw the poorest citizens forced to reside in the most far-flung residential areas of the metropole. Santiago is also a very segregated city in socio-economic terms (as illustrated in Box 3.2), and although conditions for the poor have improved markedly since 1990, the spatial distribution of poverty has changed very little, while those areas where poverty persists are socially stigmatized. Like health and housing, education holds benefits in terms of short-term coping strategies as well as a households ability to recover and move permanently from vulnerability to self-sufficiency (de la Rocha and Grinspun, 2001). Seen across the generations, securing education for younger family members is an important livelihood strategy, but one found to be evidently under stress among the urban poor. In Cebu this involved failure to access secondary and higher education. In Mombasa, low and declining school attendance was reported, attributed both to inadequate household resources and rising education costs, making it increasingly difficult for children from poor families to attend even primary school. Similarly, in Kumasi, a major issue for people was not being able to afford school fees. Even in higher income cities such as Santiago, issues of inequality were evident, where the level of service provided by the under-resourced municipal schools fell well below subsidized and private schools. Perceptions of poverty were not only associated with material lack but also with issues of respect and dignity. We found people were deeply fearful of social stigma and exclusion. Social exclusion can derive in part from lack of access to labour markets and economic processes, and in part from exclusion from political and decision-making processes. Both forms of exclusion can impact negatively on the livelihoods and linkages of the urban poor. In Kumasi, for example, low-income people were sensitive to being excluded by those in
authority and identified the poorest amongst them as being those who were not respected in society, who had little access to basic needs and who lived in relative isolation. In Mombasa too, among their many problems poor families emphasized feeling powerless to improve their situation. Perceptions of insecurity were found to accompany the experience of social exclusion and these in turn affected the ways in which poor urban dwellers engaged in social and public life. Trends detected in Johannesburg, Santiago and Indian cities such as Bangalore included a decrease in social contact and use of public space, exacerbated by declining levels of public safety. This in turn tended to lead to self-confinement and disengagement with the urban environment and urban processes, with negative consequences for statesociety relations and urban governance. In Kumasi the poor felt themselves unfairly labelled as thieves. While gated communities are most often associated with high-income urban residents protecting themselves from crime and violence, barricading themselves behind high walls and gates, research in Johannesburg found gated communities among the urban poor as well. Here migrant hostel dwellers were just as likely to shut themselves off in bounded compounds, both to protect themselves from violent crime and hostility in the broader environment as well as to pursue illicit livelihood activities in the face of exclusion from job markets and other forms of income generation (Beall et al, 2002). There is no doubt that the urban poor are energetic and resilient in the face of both adversity and opportunity. Individuals, households and communities use their existing endowments and capabilities to survive, to secure livelihood stability, and to increase their security. They secure themselves against shocks and stresses by working, saving and investing, including in social networks and relationships. It is the poorest who are most susceptible to shocks because it is they who have the least by way of endowments. Understanding vulnerability, therefore, requires building a picture of the resource endowments of poor households as well as the income earning activities in which they engage (de la Rocha and Grinspun, 2001). Moser (1998) has pointed out that the resilience necessary to exploit opportunities and to resist and recover from adversity is closely linked to the assets held by the urban poor. For Moser, it is peoples ability to effectively transform these assets into income, food or other basic necessities that can enhance productivity and constitute a resilience strategy. Such transformation can occur either by intensification of existing livelihood strategies or through developing new or diversified strategies (Moser, 1996, p2). Chambers and Conway (1992) conceptualize livelihood strategies as activities that are improvised and sustained by household members through utilization of tangible assets such as stores and resources, and intangible assets such as claims and access. Efforts might include gaining and retaining access to information, resources and opportunities, dealing with risk, negotiating social relationships and managing social networks within families, communities and the city itself. Every households combination of strategies will be different, depending on their tangible and intangible asset base. In what follows, we explore intensification and diversification of urban livelihood strategies at the household level, before going on to examine the importance of linkages beyond the household towards enhancing livelihood security.
sectors, and to certain assets, notably capital and land. Women are found to be over-represented in the non-conventional or informal economy and to pursue such livelihoods largely as workers rather than as entrepreneurs (Beall, 2002a). Again our city studies proved to be in line with these findings. In Mombasa, for example, it was found that women were not only disproportionately represented in the informal economy but were engaged mainly in selling foodstuffs, brewing and selling illicit liquors, while men were dominant in more lucrative activities, such as the hawking of curios, manufacturing, renting out rooms and water vending. Even in Kumasi, where women had more opportunities as traders than in most other places, opportunities for the poor were constrained by acute competition. This applied both to informal livelihood options such as petty trading as well as to urban agriculture where there were competing demands for land by other uses, notably construction. Moreover, womens central role in the economy of that city is largely ignored by the municipal government, as indicated in Box 4.1.
In Kumasi, marginalized trader groups were found to experience severe problems with land availability and were allotted no permanent working place. For instance, retailers were constantly being threatened with relocation, being driven away from pavement stalls and having their stalls demolished by the city authorities. In Cebu too, working conditions associated with the poor included not having permanent stalls or vending spaces, having only limited capital or none at all and thus being unable to redeem confiscated goods or to repair vending units when threatened or harassed by the authorities. In Johannesburg, the well-meaning Metropolitan Local Council cleared the streets of vendors, offering them a covered market in Yeoville instead, but depriving the street traders of prime positions and opportunities in terms of passing trade. Casual labour, characterized by very low and irregular wages, constitutes a critical source of income for poor urban households. This was found to be the case in the local economic clusters in Bangalore described in Chapter 8, and casual work was a prominent feature of livelihood strategies in other cities too. Even in Santiago, a city in the study with one of the most developed economies and where economic growth had impacted to a degree on urban poverty, the poor were found to earn very low wages as casual workers in the formal sector. However, in Ahmedabad, by contrast, there was evidence of an upturn in activity in both the formal and informal economies and a corresponding decrease in casual work. Not only was the type of work conducted important but also the location of the income-earning activities undertaken. Home-based work had implications for the living environment, while factory-based work often took its toll in terms of travel expenses. In the African cities in particular, linkages between urban and rural populations were found to be strong and to play a crucial role in family survival, as households diversified their livelihood activities across the ruralurban divide. For example, many men in Mombasa were temporary migrants who might return home either to invest in a business or when work was not available in town. The same applied to the migrant hostel residents of Johannesburg for whom some livelihood activities, such as the sale of crafts and of marijuana, were dependent on supplies from rural areas. Hence, agricultural seasons could affect the availability of produce and labour in urban centres as well as patterns of migration. Intra-family resource flows included not only urbanrural remittances but other forms of reciprocity that indicate important interdependencies between rural and urban environments. Indeed, such flows were found to be particularly important to survival in times of economic hardship among the poor in the African cities studied. These linkages, along with the broader dilemmas of employment creation, have important policy implications, suggesting that urban planners and managers cannot be expected to address the problems of urban poverty outside of the wider policy picture and interventions at higher levels.
often do not meet even the most basic of household needs, increasing the vulnerability of those already marginalized from the social and economic life of the city. Important among the strategies of the urban poor were found to be modulating patterns of consumption in order to adapt to shifts in household income or shocks to the household resource base, such as price rises, loss of subsidies or periods of illhealth on the part of breadwinners. Our studies confirmed that women generally spent a greater proportion of their income on household needs than did men. As such, in family efforts to cope with urban poverty, they were the household members most likely to engage in belt tightening exercises. Even among better-off households, the study of Kumasi Central Market showed that when both spouses were traders and earning similar amounts, women nonetheless allocated over half their income to household expenditure while men allocated far less. In Kumasi, poor people talked of cutting meals from three to just one a day and buying smaller (and less cost effective) bags of rice. In Mombasa it was found that the poor engaged in expenditure saving by walking to work, eating only once a day, gathering fallen items from the ground to sell or to eat, withdrawing children from school, postponing medical treatment and using selfmedication. Consistent with much of the literature on urban poverty (Bradley et al, 1991; Harpham and Tanner, 1995) illhealth emerged as a critical issue in precipitating poverty and one that was strongly linked to problems of indebtedness. In Visakhapatnam linkages were drawn between the impact of household debt and breakdown in family relations, leading to alcoholism and, in extreme cases, suicide among people living in the slum areas. Going into debt was itself seen as part of everyday household financial management in Visakhapatnam. In many cities credit taken from friends and family with no or low interest was regarded as the simplest way to manage household budgets in times of hardship. In the absence of this option and in times of serious financial crisis, moneylenders were favoured. In Kumasi, the use of petty credit from storekeepers was also cited as a mechanism regularly used for buying necessities such as food, while larger forms of borrowing and savings associations were invoked if collateral were needed for less regular expenditure. Asking for assistance from relatives in times of hardship was cited as a key coping strategy for the poor in Mombasa. It was asserted that seeking material and financial assistance from relatives is not regarded as begging although begging at religious buildings for food and money was also regarded as a crucial strategy for many without family support. Supporting family members is clearly reciprocal, implying that everyone is at one time or another indebted to someone in the family. Other important adaptive institutions for the poor can include mechanisms for pooling income and other resources within households and communities (Moser, 1996). Although our city studies did not identify anything as formalized as the community kitchens found in many Latin American cities, there was ample evidence of arrangements for sharing consumption. For example, in Kumasi, the extended family system played a significant role and, among the Asante people in particular, served to soften the impact of unequal social
disadvantage across different members of an extended kin group, through the practice of sharing homes by family members of different class backgrounds. However, while extended families can provide networks of mutuality and support, this is often an unequal process more burdensome on some members than others. For example, a poor person can expect business support from a rich relative or to have utility bills paid. Rich relatives may take on the responsibility of paying school fees for the children of a poor brother, sister, cousin or a distant relative. However, the obligations to provide for and support the poorest members of a family or community can constitute a drain on the resources of the almost poor, significantly increasing their vulnerability. Reciprocal obligations or social indebtedness can be a particular problem for migrant groups, who are prevailed upon by rural relatives and urban counterparts, and at a time when their own position within the city may be fragile and contingent. Under such circumstances some family members are more vulnerable to refusal than others. In Kumasi, while old age is traditionally associated with respect and wisdom, it was reported as often being experienced as loneliness and neglect as household members struggled to balance multiple social obligations against limited livelihood opportunities. The position of older people was reported to have worsened also due to a weakening of traditions and the scattering of family members in and from urban areas. As care of the elderly is dependent on informal family support systems, increasingly older people in Kumasi are left without social protection or are sent back to their villages to escape urban hardships. The vulnerability of older people in cities was not matched by the contribution they made to livelihood strategies within family and household structures. Importantly, they were found to provide care and to adopt parenting roles when younger adult members were engaged in paid work. In South Africa it was found that for many older people, especially women, their reality was one of responsibility for rearing first their own children and then their grandchildren. In the context of HIV/AIDS, greater numbers of older people are finding themselves heads of households comprising many young children and few contributing members. In these and other household structures, in the context of high levels of poverty and unemployment, whole families are dependent on the pensions of older household members.
commercialization of reciprocal relations, for example in relation to gifts and rituals associated with weddings and funerals. Nevertheless, informal institutions and organizations remain important and act as rule-enforcing mechanisms for reinforcing customs, norms, values, religious beliefs and social and solidarity networks in urban communities. These in turn can mediate access to employment, commodity markets, land and housing, services, personal security in the home as well as sources of wider social support. Social networks are important, therefore, as an asset that poor people in cities can exploit to advance themselves or, in times of adversity, use to help dampen the effects of poverty. Ethnic or hometown associations are important in urban areas of Africa, for example. They were identified in Kumasi, Mombasa and Johannesburg as being active and beneficial among migrants, who rely on ethnic affiliation to counter feelings of insecurity and isolation in these cities. Generally, such ethnic networks tend to operate at community level and are not citywide. In Johannesburg, for example, they were found to operate often within or between migrant hostel complexes, with residents of a particular hostel building all originating from one rural area or town and socializing with people in other hostels from the same rural area. These networks constitute a social function but also act as welfare associations, providing financial support to members in need, such as for funerals. Basic needs or insecurities often compel the urban poor to form new, urbanbased informal support networks. In some low-income neighbourhoods of Kumasi, for example, a communal lifestyle enabled pseudo-kinship relationships to emerge within these urban communities, resulting in family-like reciprocal behaviour. For example, a mother needing to go to the market would leave her children in the care of another household. In return, she would feel obliged to undertake the shopping for them or even feed their children when she minded them in return. In Ghana, communal labour has always been an asset the poor have exploited to meet some of their basic needs, and poor communities in Kumasi were found occasionally to mobilize and use communal labour to provide community facilities such as schools, clinics, drains and bridges, public latrines, and to clean their living environments. Among groups of working people, this system was also used. The mechanics at Suame, the traders of the Kumasi Central Market and the carpenters at Anloga, along with several other poor groups in the city, were noted to have specific days on which they cleaned their surroundings before starting work. Such activities are dependent on a degree of mutual interest existing within a community or group and it was asserted that levels of solidarity were less strong in Kumasi than those that might be found in Ghanas rural areas, where stronger communal traditions still prevail. Moreover, mutuality was not a feature of communities across all our city studies. In Santiago, for example, it was found that, unlike during the dictatorship years when they shared disadvantage and a common struggle, people, including poorer citizens, had retreated into family groups with the arrival of peace and relative prosperity. They were both apathetic and suspicious of efforts at social mobilization beyond the household. The experience of Santiago notwithstanding, evidence from Cebu demonstrated that local level social networks are an important way for people to glean
information about employment opportunities as well as access to government resources and services. Pooling resources as a livelihood strategy was also found to extend beyond household sharing to community level systems for savings and for credit. In Bangalore, for example, it was found that customary rotating savings mechanisms, known locally as chit funds, constituted a critical resource for the urban poor and fed into complex informal markets of one kind or another. In Colombo, traditional funeral societies (Maranadara Samiti) and savings groups (sittu) were found to exist alongside sports clubs, trade unions and community development organizations. The same patterns were observable in Johannesburg, with stokvels or informal savings clubs, bulk buying schemes, communal eating arrangements and burial associations existing side-by-side with church-based and religious organizations in a city which historically has boasted an active civil society. Even for Santiago, evidence was provided of broader social networks in operation, such as the allegados committees. This is a Chilean practice and term applied to extended family or friends living in households where they do not pay rent but contribute in non-monetary ways. As shown by the cases of the allegados committees for Latin America and the hometown associations in Africa, informal social networks often become semi-formalized and in turn, come to constitute a more complex web of associational life. Informal support networks and associational forms can lead to more sustained and organized forms of collective action, not least when livelihoods are threatened. In Cebu, for example, trading groups had organized themselves to fight against stall demolitions. Trade-based associations were among the most active community organizations in Kumasi. Membership groups, by definition, seek to promote the interests of their members. As such they do not always assist in addressing community level or citywide concerns and can work to exclude the most destitute or vulnerable in a society. Among the latter, loose organization can sometimes be found, for instance among the street children of Bangalore. However, these fluid associations are often exploitative. They are rarely sustainable and cannot easily be scaled up into more formalized groupings capable of effective collective action (Beall, 1997). Our findings also point to the need not to romanticize social networks and relations of reciprocity. Moser (1998) has argued that under stresses and shocks, social capital can be eroded and de la Rocha and Grinspun (2001) also suggest that social resources seem to diminish under the pressures of poverty itself. Our research confirmed that the poorest people were those in isolated households unable to rely on supportive social networks and that low-income areas were those most likely to be situated in stigmatized localities, with crime, violence and fear becoming self-fulfilling prophesies. As our study of Johannesburg showed and as illustrated in Box 4.2, social networks can be characterized as much by exploitative relationships and antisocial behaviour as by reciprocity, support and cooperation.
associations and membership organizations reinforce social assets amongst the urban poor but that these are often predicated upon what has been dubbed bonding social capital (Foley and Edwards, 1999) among exclusive groups whose activities benefit the group itself but do not have a wider impact on urban poverty reduction. Kinship networks and very localized forms of association constitute and give rise to important community level assets but ones that rarely go beyond the group. For example, it was found in the case of one informal settlement in Johannesburg that the group which had arrived and established themselves there earlier excluded new migrants from access to land, housing and services (see Box 7.2). We found across our city studies that endowments of such forms of bonding social capital constitute important, if fragile, resources for the urban poor. They can provide safety-nets when social deprivation is exacerbated by shocks, stress and other sources of vulnerability. The question as to whether bonding social capital benefits urban development is more vexed. To the extent that it does, it is when poor urban communities engage in self-help initiatives to provide urban goods and services in the absence of adequate provision by the relevant authorities. In other words, activities such as cleaning market places, maintaining public toilets or looking after neighbourhood infrastructure have a substitution effect, allowing local authorities to abdicate responsibility for providing these services for low-income areas or groups. However, the fact that they are invariably performed in the interests of an exclusive group means that the benefit to the city concerned is fairly limited. The question remains, therefore, under what circumstances if any, are localized and in-group activities scaled up for more generalized goals and ends? The ten city case studies provide some evidence of organizations scaling up and scaling out, a process described in some detail in Chapter 7. However, the move from bonding to bridging social capital, for example through the formation of federations of smaller organizations and their engagement with local government, is neither automatic nor easy to forge. Nevertheless, the city reviews did yield evidence of local organizations of the poor forming themselves into loose but wide federations that afforded them greater voice. The best example was the formation of the Cebu City United Vendors Association, described in Box 7.3. Other examples included informal networks of sangas (revolving credit groups) in Bangalore, caste-based organizations in Visakhapatnam, traders associations in Kumasi and committees for local improvement and development in Santiago. The more formally constituted federations of community-based organizations (CBOs) were found to emerge most particularly in response to government provided services, such as the Home Owners and Water Users Associations in Cebu City. Another example was the rents and rates boycotts in the 1980s and early 1990s in Johannesburg, where township residents, on an organized basis, refused to pay what they saw as illegitimate apartheid local authorities for sub-standard services. When neighbourhood or community-based organizations are scaled up to city level, various forms of capture can come into play. Examples of this were provided by the studies of Visakhapatnam, Bangalore and Johannesburg. In these cities it was found that, in their individual capacities, community level
organizations were often concerned with meeting immediate felt needs rather than taking on wider concerns. Even when their ambitions were greater, such organizations generally have no social, legal or political status, can often claim no right of access to decision-making forums, or find that the representatives of informal local structures are relatively powerless in the more formal political arena. Another problem is that organizations active at the local or even city level can become subverted to national-level agendas and ambitions. For example, in Johannesburg the civic organizations affiliated to the South African National Civics Organization (SANCO) became preoccupied with that organizations oppositional agenda in respect of the national government, rather than with local issues of urban development and governance. A number of the city case studies provided examples of community-based organizations being created by the state or other developmental institutions for project-based and sometimes political purposes. For example, in Santiago, large numbers of organizations were formed to support community projects related to state programmes for poor areas. In Bangalore and Colombo, the Urban Basic Services Programme (UBSP) established Neighbourhood Committees and Community Development Councils (CDCs) respectively, in order to ensure consultation and participation in service delivery. Once objectives were achieved, many such organizations became moribund and ceased to function. In Johannesburg user involvement in local decision-making is not only encouraged but is constitutionally required. Here it was found that organizational pluralism was prevented due to one dominant group or community organization emerging and usurping the consultative process. These city-level experiences also demonstrated that the artificial imposition of community-level structures sometimes served to undermine existing organizations and erode spontaneous social participation. This suggests that government cannot easily construct or forge bridging social capital for its own purposes of urban development. To the extent that partnership relationships that build bridging social capital can be fostered and nurtured between organizations of the working poor and local government, this was found to depend on the intervention of representatives of the former. The obvious example to emerge from the city studies was that of the Self-Employed Womens Association (SEWA) in Ahmedabad, which has successfully engaged local authorities and the business community on behalf of its constituency of women pursuing home-based livelihood strategies or earning a living in the informal economy. SEWA was created in 1972, initially as a response to loss of jobs in the citys textile industry. Today it runs credit, healthcare and social insurance programmes as well as engaging in policy advocacy and political lobbying. In the absence of organizations such as SEWA, the resources of the poor are often destined to remain operative at the micro-level alone, poorly harnessed towards advancing livelihood opportunities through organized linkages.
Conclusion
What is clear across all the cities studied is that social networks and the many different forms of associational life described above are crucially important in maintaining and developing urban livelihoods. However, while clearly being a resource to the urban poor, they are very fragile and take their toll on people, some more than others. People do support each other and communities cooperate. Under certain circumstances they organize themselves more widely. However, under conditions of poverty and stress, without some sort of security in place, it is difficult for poor urban households to engage in reciprocal gestures and for low-income communities to sustain self-help initiatives. Even more difficult to sustain is organized and scaled-up public engagement, except under conditions of extreme provocation, such as during the twilight years of apartheid in South Africa or under the Chilean dictatorship, or when sympathetic representatives and interlocutors work with and on behalf of the urban poor. Under these circumstances, it is very easy for city governments to destroy or damage survival and livelihood strategies by their actions, even when well intentioned. Hence, it is vital that urban policy-makers and planners understand the livelihood strategies, networks and other social assets of the poor, in order to anticipate and respond to the possible impact of interventions. This is illustrated most clearly in relation to policies towards the informal sector, as recounted in Chapter 8. When intentions are less noble, it is also easy for city governments to abdicate responsibility for the urban poor altogether, allowing them to substitute their scarce resources and their limited time and energy for the public provision or coordination of service delivery. What does this mean for urban governance? Clearly as much as they benefit poor households and communities themselves, the social assets and social networks of the poor also constitute a resource for city governments. Moreover, when the urban poor do engage in scaled-up public action around community-based or work related issues, they can and do rock the boat. This confirms that there is no substitute for extending the linkages of the urban poor in pursuit of sustainable livelihoods, not simply in self-help networks of mutuality and support but towards engagement in local governance.
Chapter 5
Urban Politics: Exclusion or Empowerment?
Carole Rakodi 1
Introduction
Urban governance refers to the interactive relationships between and within government and civil society actors in cities. It includes the overlapping domains of political and administrative processes of decision-making and is also about how government organizations react to the needs and demands of urban actors, both organized and unorganized. Chapter 6 concentrates on the organizational and administrative arrangements and practices of city governments. This chapter explores political systems and processes at the urban level, including the creation and operation of political institutions, government capacity to make and implement decisions, and the extent to which those decisions recognize and respond to the interests of the poor. Its aim is to analyse and explain how urban political systems and processes operate and the extent to which they give the poor political voice and influence.
URBAN POLITICS 69
For what purpose and in whose interests government actions are taken depends on three main factors: The political context, which has both local and supra-local dimensions. The former refers to the socio-political public space available to urban actors, including the poor, and their ability to take advantage of that space. The outcome depends on the character of that local political space, the capacity of citizens to engage in politics, the distribution of power between them and how they exercise that power. The supra-local dimensions of the political context include the institutions and rules that govern political behaviour, the political culture and the nature of the national political regime. The way that political intermediation mechanisms (the formal structures and procedures of the urban political system) are designed and how they work in practice. The political actors involved their goals and demands, the resources they bring to the political system, the power relationships between them, and the strategies and tactics they adopt, that is their political practices.
The interrelationships between political structures, actors and practices in particular contexts have varied characteristics and differing outcomes with respect to effective and pro-poor governance. As explained in Chapter 1, this research stops short of a rigorous analysis of outcomes, although it has identified illustrative examples of pro-poor results. Instead, it concentrates on how and in what circumstances poor people are able to gain access to decisionmaking processes and influence political agendas, with the result that public agencies (especially local government) address poverty through their practices and resource allocation decisions. The discussion will first explore dimensions of the political context of the cities under study, to identify what local and supra-local contextual characteristics are most likely to produce urban political space in which avenues of influence are open to poor citizens. Second, it will analyse the structures and procedures of alternative urban political systems, including forms of democracy and representation, arrangements for executive control and, briefly, mechanisms for accountability. Third, it will analyse the articulation between political structures and a variety of urban interests, identifying key actors, explaining how they behave, and revealing relationships between them through an analysis of political practices. Particular attention will be given to the concerns of poor people, their capacity to press their demands and the responses of political decision-makers to such demands.
common in many of the countries in which the case study cities are located in the 1970s and 1980s. These gave way in the 1990s to more democratic regimes in which it was intended that citizenship rights would be exercised through the electoral system and that the voting power of poor people would ensure that their concerns were addressed. In addition, decentralization would improve efficiency and local responsiveness. Even in the well-established democratic system of India, there was recognition that democratic practices were being undermined at the local level by the periodic suspension of many elected local governments and the failure to decentralize effectively. This led to attempts to strengthen local democracy by constitutional change at the beginning of the 1990s. Political arrangements are embedded in a wider set of rules and practices. These are based in part on the constitution and associated law, which specifies political and civil rights and ensures that these rights can be protected. Such rules or norms of political behaviour also arise from socially constructed identities, meanings, informal rules and institutions, and political practices. The latter, in turn, influence the framing of constitutional and legal frameworks. All are shaped by history, previous experiences of democratic arrangements and the nature of the (re-)democratization process. Together, they constitute the political culture and incorporate a conceptualization of citizenship as active or passive. Liberal democratic theory (like liberal economic theory) gives primacy to the individual. It assumes that politics aggregates the prior desires or preferences of individuals into collective interests through the mechanisms of elected government, guided by what have come to be a commonly accepted cluster of rules and institutions. These include universal adult suffrage, regular elections, political competition, associational autonomy and the separation of executive, legislative and judicial powers (Held, 1993; Parekh, 1993). According to this theory, government is constituted and leaders selected by individuals exercising their right to choose. Once voted into office, the elected representative is trusted to act as promised, although it is also recognized that he or she has considerable leeway to make detailed decisions, and that additional political practices such as lobbying can supplement periodic elections. All the city governments studied operate within the context of a formal liberal democratic regime at the national level. However, the political culture, the nature of the previous regime and the characteristics of the processes of regime change vary, helping to explain the nature of the post-transition constitutional settlements and the embeddedness of democratic practices. In some of the countries, earlier authoritarian regimes were cloaked in formal democracy, providing some political space for the organization of opposition. For example, in Brazil between 1964 and the end of the 1980s, the military enlisted the aid of the traditional political lite and the executive branch of government to distribute state resources in order to manufacture consent, and to demonstrate this consent through faade democracy. The regimes need to use the traditional political lite in this way enabled the latter to protect its position, jump onto the democratization bandwagon when pressure for regime change grew too strong to resist and protect its own interests after
URBAN POLITICS 71
democratization (Hagiopan, 1994). Countering lite dominance, however, labour unions have remained a potent political force, despite the attempts of the military to reduce their influence. Allied to broader social movements, they were able to exploit the widening of national and local political space to strengthen their political organization (through the Workers Party), influence the new constitution (and thus the form of new political structures) and develop new political practices. Because, in Brazils federal system, local political arenas provided a modicum of political space, local opposition politics and social activism made a significant contribution to the process of regime change, which was rewarded by a commitment to decentralization in the post-transition settlement. Similarly, in the Philippines, democratic formalism allowed political competition within an lite oligarchical regime. Such competition pitted factions of the lite against each other, resulting in political instability, and excluding many groups from political participation (Rland, 1992). Regime change occurred both because of competition between the ruling families and as a result of wider social struggles to end authoritarian rule. As in Brazil, the Catholic Church played an important role in supporting community-based organization (CBO) and building grassroots awareness. Again, as in Brazil, sections of the old oligarchy continue to play significant roles in democratic government at both the national and local levels. However, organized civil society played a key role in the process of bringing about democracy, increasing general levels of political awareness and winning for it formal recognition in representative political structures. In South Africa, because opposition political parties, especially the African National Congress (ANC), were proscribed, the long and often violent struggle against the apartheid regime depended on other forms of social organization, especially trade unions and civics,2 as well as on widespread mobilization and resistance amongst the population at large. The length, intensity and form of the struggle has resulted in a constitution crafted to ensure political representation of the previously excluded majority and an unambiguous political commitment to redressing apartheids legacy of extreme inequality and widespread poverty. However, fiscal constraints have made it difficult to consolidate the new democratic institutions, achieve the necessary administrative reforms and reduce poverty. Moreover, many of the organizations, including the civics and many non-governmental organizations (NGOs), which were critical to the struggle against the previous regime have had difficulty adapting to the new political and financial situation. In Brazil, the Philippines and South Africa, therefore, the struggle to end authoritarian rule depended on organized civil society as well as wider societal opposition. Despite their difficulties, they have a combination of strong democratic constitutions, political commitment to decentralization and provision for civil society inclusion in representative political institutions. As a result, these countries probably have the best prospects for achieving consolidated and stable democracy at both national and local levels. Chile, India and Sri Lanka provide contrasts. In Chile, although an active civil society played a key role in ousting the Pinochet regime, its role under
democracy has not been sustained. Economic growth, following neoliberal economic reforms in the 1990s, and targeted welfare policies have reduced the incidence of poverty. However, the individualist ethos of Chiles model of democracy, its liberalized economy and the focus of its poverty policies on the most vulnerable has led to social fragmentation and the weakening of collective organization. Social exclusion and spatial segregation have undermined the ability of the poor to make effective claims on the political system. In India the prospects for a more effective and responsive democratic local government system to develop were improved by the 74th constitutional amendment. However, the broad-based Congress Party has lost ground to parties with narrower regional, caste or religious bases, reducing the legitimacy of national democratic processes and hindering progress with decentralization. Local politics remains essentially clientelist, despite the presence of fairly well developed civil society organizations (CSOs) and a relatively independent judiciary. In Sri Lanka, national electoral competition has resulted in peaceful transfers of power (except in the north of the country, where a long-standing civil conflict had at the time of the study superseded democratic politics). However, for Colombo, political and bureaucratic rivalry between local, provincial and national levels of government reduce the city governments room to manoeuvre. In addition, political competition at the city and neighbourhood levels manifests itself in electoral politics, which, as in India, are an uneasy mixture of representative democracy and clientelism. The fragility of democracy, limited political space and undeveloped civil society, especially organizations of the poor, are most marked in the countries where, until recently, one-party rule was accompanied by a refusal to tolerate independent organization outside government and the ruling party. Many such countries also suffer from recurrent crises and their governments lack any real commitment to decentralization. These conditions apply especially to Ghana and Kenya. In Ghana, a legacy of authoritarian politics and state-led approaches to development seems to have resulted in a limited appreciation of the need to secure public support. In addition, the traditional authority structures of the Asante kingdom, which is centred on Kumasi, and which are characterized by hierarchy, deference and unquestioning loyalty to established authority, have an important continuing influence on political culture and behaviour. For example, in Ghana the President nominates (subject to approval by the metropolitan assembly) the metropolitan chief executive (MCE). From 1995 to 2001, the MCE who governed Kumasi in an autocratic, idiosyncratic and ineffective manner was a wealthy member of the Asante royal family, as well as a supporter and financier of the ruling party and personal friend of the President. Presidential support, control over resources and the deference due to a member of the traditional lite enabled the MCE to command majority support in the elected Assembly or sideline it when necessary, despite widespread criticism of his management and allegations of corruption. In Kenya, multi-party elections were restored in 1992 in response to domestic and external pressure, but for a decade there were abuses of the electoral rules. Opposition parties were sometimes refused permits to operate and were denied access to the media; and there was violence, harassment and gerrymandering. Moreover, parties continue to have an ethnic base, the
URBAN POLITICS 73
democratic process is seen as an opportunity to capture personal power and President Moi was able to manipulate state resources to stay in power until the December 2002 elections (Wanyande, 2000). Of all the case study countries, Chiles approach to democratization most closely resembles that recommended by liberal democratic theory. It illustrates many of the shortcomings of that model of representative democracy, especially the passive concept of citizenship that entrusts decisions to elected representatives who may in practice pursue their own interests, and the marginalization of minorities that occurs in majoritarian systems (Held, 1996). Where history and culture have fostered conceptions of active citizenship and organizations of poor people, in contrast, the latter show greater capacity to exercise their political rights. In addition, in such an environment, the legislative base for the democratic system is more likely to incorporate ways of ensuring that the interests of disadvantaged groups are recognized and to provide for forms of direct, deliberative or participatory democracy (Miller, 1993; Held, 1996). A further criticism of liberal models of democracy is that they fail to recognize values, bases for identity and behavioural rules that value consensus and impose obligations on kin and ethnic group members to support each other. While this moral public sphere is deeply rooted in countries such as Ghana and Kenya, the new democratic institutions still lack roots and legitimacy. As a result, voting follows old allegiances, fragmenting those (especially the poor) who share common economic interests, exacerbating conflict and rivalry between ethnic groups and undermining controls on rent seeking and corruption (Dia, 1996; Enemuo, 2000). The implications of these differences in political culture and history for local politics in the case study cities will be explored further below.
Forms of democracy: Urban political structures and arrangements in theory and practice
The scope for local decision-making depends on the extent to which powers and responsibilities are decentralized to local government, both in theory and in practice. Earlier research has demonstrated that, to respond to local needs, operate effectively and develop legitimacy, local government needs a sound legislative framework, a range of significant responsibilities, an adequate resource base, and political and administrative capacity (Davey, 1996; Crook and Manor, 1998; Blair, 2000). Without these, it is likely to be locked into a vicious circle in which it has little legitimacy in the eyes of urban residents and businesses but lacks the political and financial resources to improve its performance. As a result, citizens and potential candidates for electoral office are not encouraged to take the political process seriously. These earlier findings are borne out by the experiences of the case study cities. In Ahmedabad, for instance, despite an almost uninterrupted tradition of elected local government since 1950, the general malaise of democratic institutions in India, together with local communal violence arising from caste and religious divisions, had eroded the Ahmedabad Municipal Corporations (AMCs) legitimacy. The appointment of an effective Municipal Commissioner
in the mid-1990s, however, enabled the AMC to realize a financial surplus for the first time. This enabled it to leverage additional capital funds, initiate innovative partnerships and improve service delivery. As a result, confidence in its capacity to deliver increased and its legitimacy in the eyes of citizens improved. In this section, the legislative basis for decentralization will be discussed first. This will seek to demonstrate how different political interests in central and local government interact with the distribution of responsibilities between central and local agencies to influence responsiveness to the poor as well as effectiveness of service delivery. Second, arrangements for electoral participation at the city level, including the location of legislative and executive control, will be analysed. Finally, the potential contribution of other forms of democratic practice will be reviewed.
URBAN POLITICS 75
BOX 5.1 CEBU: TOWARDS DEMOCRATIC DECENTRALIZATION AND PRO-POOR LOCAL GOVERNMENT
In Cebu City, the restoration of democracy in 1986 brought with it a political commitment to decentralization and democratic local government enshrined in the Local Government Code 1991. According to this legislation, municipal government has responsibility for a range of services and can also play an agency role in delivering certain national programmes, such as the Community Mortgage Program (CMP), which provides loans to enable residents of informal settlements to purchase land and install infrastructure. In line with the constitution, local governments receive a substantial share of central government revenue and also have a range of local revenue raising powers (see Chapter 6). Cebu itself benefited from rapid economic growth in the 1980s and early 1990s. With substantial local taxes and significant revenue sharing from central government, Cebu City is reasonably well resourced. In the first mayoral elections of the current era, first Osmea and then his deputy, Garcia, were brought to power, each for two successive terms, on the basis of a high voter turnout and with clear majorities. Half the councillors elected in 1988 and all of those elected in 1995 and 1998 were affiliated to their party. The mayors had a vision of Cebus economic future as a port, industrial and service city potentially attractive to Asian foreign direct investment, and a clear local identity and commitment. In accordance with the Local Government Code, along with a directly elected mayor and members of the legislative assembly, local bodies with wider representation have been established. Most power rests in the hands of the executive mayor. However, the local bodies provide for mandatory representation of NGOs and peoples organizations (registered community-based organizations, or CBOs). They have functions related to development and poverty reduction, housing and upgrading, schools, street children, street vendors etc. In addition, a sub-city level of government increases responsiveness to neighbourhood concerns (see Box 6.4). Policy consultations are common, for example during the preparation of city development plans. Municipal governments in the Philippines are obliged to allocate 20 per cent of their total financial resources to implement their development plans, together with an additional 5 per cent earmarked for a gender component. Since 1986, municipal decision-making in Cebu has contained significant pro-poor elements. In part this can be explained by the citys relative prosperity: according to government figures (although these are challenged by local civil society), the economic boom years reduced poverty. This also enabled the city government to generate resources for large-scale infrastructure investment as well as basic services, despite a downturn associated with the Asian financial crisis of the late 1990s. In addition, organizations and advocates of the poor formed coalitions to lobby for pro-poor policies. They were able, in successive elections, to lobby and secure the commitment of the successful mayoral candidate to a pro-poor political agenda in return for delivering political support. They were also able, to some extent, to hold the Mayor to account following each election. The strength of civil society can largely be attributed to the number of NGOs, some of which cut their teeth on the politics of confrontation during the struggle to end authoritarian rule. Some NGOs provide continuing support for community organization and some participate in local government through the political structures which provide for representation of NGOs and organizations of the poor. The pro-poor orientation of local policy can also be attributed to national policy commitments which themselves result from strong civil society organization and influence. In particular, a Presidential Commission of the Urban Poor was mirrored in the Cebu City Commission of the Urban Poor (subsequently the Division for the Welfare of
the Urban Poor). The DWUP coordinates the implementation of national and local poverty reduction policies with a focus on social housing, services and the upgrading or relocation of informal settlements. Despite the limitations of these programmes, they contribute to a mosaic of pro-poor expenditure and actions. Cebu City, therefore, has a relatively stable and consolidated democratic political system, which has given voice and influence to the poor. However, local politics still has strong oligarchical and clientelist features. Successive mayors have come from the two most important local families. While their actions have provided some benefits to poor groups, their responses to competing interests have often not favoured the poor. Significant resources have been allocated to prestige projects. The concentration of power and resources in the hands of the mayor and the inability of assembly members to exercise effective scrutiny have reinforced a tendency to allocate resources to meeting the needs of the poor on appeal and by favour rather than by right a tendency exacerbated by the non-ward based system of political representation. Peoples organizations (POs) and NGOs can and do successfully articulate the interests of the poor through a variety of formal and informal mechanisms at both neighbourhood and city levels. However, representation in formal decision-making mechanisms is not always meaningful and often the only effective strategy is a direct appeal to the mayor. Thus, although poor residents have obtained improved access to the resources distributed by the political system, their relationships with the mayor are characterized by an uneasy balance between an electoral contract, indirect access mediated by NGOs, and dependent clientelism.
budgeting (PB). Recife, the case study city, followed in the footsteps of Porto Alegre and Belo Horizonte in instituting such a system (see Box 6.3). The remaining cities bear a closer resemblance to Bangalore than Cebu. In Kumasi, central ministries have been reluctant to cede their functions, as intended under the decentralization programme, to municipal government, which lacks financial and administrative capacity. In Colombo, opposition control of the Colombo Municipal Council (CMC) led central government to retain control over key functions, including urban development planning, housing, water supply and staff recruitment. In Kenya, the retention of key services such as water under the control of national agencies and, until recently, the absence of any centrallocal financial transfers, rendered Mombasa Municipal Council (MMC) unable to deliver services effectively. Finally, in Chile, central government keeps firm control over all the most significant services and programmes (including those intended for the poor) and ensures that Santiago does not become an independent source of political power by dividing it into 34 municipios.
URBAN POLITICS 77
installation. Although residents do vote in state elections, those elected represent large and mixed constituencies, are too remote for much personal lobbying and rarely have a pro-poor political platform. Municipal politics, in contrast, is more accessible and councillors are likely to be more responsive. The outcomes of local elections depend largely on the perceived capacity of candidates to obtain resources. Whereas highincome residents and larger enterprises can pay individually (for private services or bribes), low-income and (many middle-income) residents can only obtain secure tenure and services with the support of an elected representative. This can only be obtained if electoral support is given thus voter turnout (at around 80 per cent) is much higher in low- rather than high-income areas. Local leaders, in turn, owe their position to their ability to maintain appropriate political contacts on the one hand and to deliver votes for relevant candidates on the other. All those involved gain in some way from this hierarchy of patronclient relationships. However, political links are not always necessary and are rarely sufficient councillors, community leaders, entrepreneurs and residents also cultivate contacts in the middle and lower levels of the municipal and state bureaucracies (see Box 5.3). Social and ethnic connections or petty bribery enable many to get permissions, exemptions or services they cannot otherwise access. Often, in residential areas, connection men (the locally resident employees of service providing departments) will enable service connections to be made in return for a fee. Thus small enterprises and low-/middle-income residents depend on the politicians and bureaucrats of municipal government and the lower level bureaucrats of the State government. This lower level circuit is a sphere of clientelist inter-dependence, characterized by informal connections referred to by Benjamin and Bhuvaneswari (2001) as politics by stealth. It is far from the ideal of consolidated democratic decentralization, in which elected representatives balance the interests of their constituents with the need for strategic investment and redistribution, and all citizens gain access to land and basic services as of right. Nevertheless, Benjamin and Bhuvaneswari argue that the informal municipal politics and porous bureaucracy of Bangalore serve low- and middle-income residents much better than State level politics, large-scale decision-making and formal planning. Allocation of responsibility for most policy and investment decisions to the state government, which is more remote and unaccountable than the city government, disadvantages the poor.
In practice, their operation is influenced by the political context and culture, the extent of local autonomy and resources, the configuration of political interests at the city level and the strategies and tactics adopted by different urban actors. However, certain formal political arrangements do seem to provide more opportunities for the poor and marginalized to influence decision-making than others. The most relevant features of those political arrangements include the basis for electoral representation, the location of executive authority, the rules about terms for holding office and provisions for wider societal representation in governance. The basis for electoral representation The evidence from the city case studies and other research (eg Crook and Manor, 1998) is that a ward-based system of representation tends to encourage local representation and accountability. Especially where poor (and lower middle-income) citizens are geographically concentrated in certain residential
URBAN POLITICS 79
INSTITUTIONAL OUTPUTS Dedicated infrastructure/ civic amenities Central locations Larger parcels of land Fiscal benefits
INSTITUTIONS Parastatals/state level bureaucracy Infrastructure development agencies Financial agencies Development authorities Metropolitan planning agencies Dedicated service providers
INSTITUTIONAL OUTPUTS Upgrading infrastructure Secure land tenure while maintaining a diverse tenure regime Basic civic amenities
LOCAL ECONOMY COALITIONS (A spectrum of richer, middle and mostly poor groups: depending on the particular local economy) Local landed lite; larger entrepreneurs; commercial traders; local politicians, lower and middle level bureaucracy, real estate agents Subcontractors; renters, sales agents, fabricators, political workers, lowest level government workers Workers, hawkers
INSTITUTIONS Mostly local government (some lower level bureaucracy of parastatal agencies) Standing committees (especially Works and Finance) Mayors/Deputy Mayors Councillors Senior technical staff Middle and junior bureaucracy Who are the contacts or connections? (into the system) Middle and junior bureaucrats Middle and junior professional government staff Lower, middle political agents, and some senior political heads
Contact Points Staying in the same neighbourhoods Professional associations/ clubs/resorts Relatives Children to similar schools Liaison agents
Source: Solomon Benjamin, Environment and Urbanization Vol 12, No 1, April 2000, p55
neighbourhoods, candidates depend on their votes and in turn need to respond to their constituents concerns. Such an outcome depends, however, on a number of other factors. First, it depends on councillors having influence on executive decisions. Second, it relies on voters having a conceptualization of representation that leads them to expect political promises to be kept and tangible benefits to be forthcoming, rather than political office being regarded primarily as a means of conferring status on the elected representative. It also depends on ward boundaries being appropriately defined. In urban areas which are rapidly growing or where densification is occurring, tardy redefinition of ward boundaries can result in a democratic deficit, in which the population size of wards (or constituencies) is very large, particularly in those wards where the poor reside. For example, in Mombasa, each councillor represents an average of nearly 30,000 residents, compared to about 15,000 in Johannesburg, Colombo and Kumasi. Indeed, national governments may deliberately adopt such a strategy (together with gerrymandering or reserving seats for its own appointees) where opposition parties tend to win urban parliamentary seats and control over powerful local authorities. In Mombasa for example, the provision for seven nominated councillors in addition to 24 elected councillors enabled the national government to secure a clear majority on the Council. In addition, the positive benefits of a ward-based system may not be realized if votes are mobilized on other bases (eg ethnicity or caste), the Council does not have jurisdiction over relevant activities or lacks resources (as in Bangalore), or councillors are prohibited from standing in successive elections (see below). In Kumasi, for example, the domination of a nominated Mayor, together with the Metropolitan Assemblys limited resources and capacity, gives Assembly Members little scope for positive action. In contrast, in some local authorities (eg Ahmedabad) councillors may be given a development budget to spend at ward level. This enables them to increase their responsiveness to the needs of residents (as well as increasing their chances of re-election). Particularly where boundaries for national constituencies and local administrative jurisdictions coincide, rivalry between the candidates may lead to attempts to capture public programmes in order to command the resources needed to reward constituents and supporters. These are contests which mayors or councillors may lose. This occurs, for example, in Santiago, where the boundaries of national constituencies coincide with those of lower level local government. Although ward-based elections may give more political voice to the poor, other things being equal, they also share the main disadvantages of majoritarian democracy, which limits the influence of minorities and potentially exacerbates rivalry between groups. It is to overcome this weakness that systems of proportional representation (PR) have been adopted in many places. However the closed party list system, which is the usual basis for PR, puts power into the hands of the political parties and reduces the responsiveness and accountability of those elected to their constituents. In Johannesburg, in the 1996 elections, 162 of the 270 councillors were directly elected on a ward basis and 108 under a PR system, to try and get the best of both worlds.3 However, designing an appropriate system of PR is by no means straightforward, as illustrated by Colombos experience: the closed party list system introduced in 1979 was
URBAN POLITICS 81
amended in 1987 and 1991 and proposals to abandon it in favour of a wardbased system were under consideration by the time of the 1997 elections. Other reforms that may increase the representativeness of electoral systems include: Mandatory voting (as in Santiago), which may be one of the few ways of counteracting the worldwide pattern of low (sometimes very low) turnout in local government elections, at least in the short term. However, disillusion with Chilean politics in general and the incumbent governments policies in particular have still led to declining turnout (especially in high-income areas) and increasing numbers of spoilt papers (a means of protest adopted by more vulnerable groups including the young and poor people). The reservation of seats for under-represented groups. The 74th Constitutional Amendment in India, for example, provides for a third of the seats on local councils to be reserved for marginalized social groups (women and scheduled castes and tribes (SC/ST)) and for the largely symbolic office of mayor to be rotated between a scheduled caste, female and general candidate each year. However, those elected to reserved seats are not necessarily accountable to their constituents, as class or other interests may take precedence over the interests of the marginalized group concerned.
The responsiveness of a political system to poor citizens depends only partly on the design of formal arrangements for the election of representatives. A further dimension that affects the relative power of councillors is the location of executive power and whether the chief executive is directly or indirectly elected. In addition, the outcomes of electoral politics reflect the tactics adopted by elected representatives. These depend not just on their reliance on the votes of particular groups, but also on their judgement as to whether formal decisionmaking or informal clientelist strategies will produce better results. Both these aspects are discussed further below. The location of executive authority In theory, political systems in which the mayor is directly elected and performs the role of chief executive generate strong leadership and a clear line for accountability. Accountability may be increased further if the mayor shares decision-making power with an executive committee or the full council, which typically approves the budget. However, direct election of a chief executive is also a high-risk system, since an overwhelming share of power is vested in a single individual, who may turn out to be a bad leader. The risks may be reduced and power more widely distributed if the chief executive is indirectly elected and thus shares power more equally with councillors, although the outcome of such systems is sometimes said to be indecisiveness and unclear accountability. For example in Ahmedabad, most power at the municipal corporation level rests with the Standing Committee of the AMC (12 elected from 129 councillors, by convention reflecting party representation on the Municipal Corporation General Board as a whole). However, the existence of other Standing (and ad
hoc) Committees with executive powers diffuses executive responsibility and clouds lines of accountability. Whether an elected chief executive holds effective power also depends on whether s/he, the council or an external body is responsible for appointing senior paid officials. The Latin American tradition, in which a directly elected mayor makes his or her own appointments to senior posts (confidence posts), has both advantages and disadvantages. The advantages are control and strong direction of policy by a like-minded team if the mayor provides good leadership. The disadvantages include discontinuity between administrations, poor promotion prospects for career officials and weak or partisan leadership if the mayor is ineffectual or biased (Nickson, 1995). If the mayor or council as a whole is responsible for such appointments, their ability to direct policy and implementation is increased but so is the scope for using local authority jobs as patronage resources (for example in Mombasa). However, if central government makes senior appointments, including the chief executive, local responsiveness and accountability are reduced. Although a characteristic of authoritarian regimes, this also occurs in supposedly democratic political systems. The most glaring example in the case study cities is Kumasi.4 However, in Indian cities also, the local authoritys agenda can be largely set by state-appointed senior executive officers (commissioners). For example, in Ahmedabad, good leadership was for a time provided by an able, far-sighted and energetic Municipal Commissioner. However, the government practice of transferring commissioners frequently reduces their ability to achieve much. Terms of political office The rules governing terms of political office also make a difference to the efficacy and accountability of local political systems. If successful candidates are limited to a single term, their political tactics may well focus on maximizing their personal returns, especially if the costs of fighting an election are high. Unless their ambitions include a higher-level political career, responsiveness and accountability are likely to be limited. Where successive terms are permitted, candidates have more of an incentive to be responsive and to demonstrate their probity, although the latter may be threatened by obligations to reward their campaign financiers and supporters. A second term may enable representatives to tackle longer-term and more difficult issues, since the priority during a first term, especially if it is short, must be to deliver visible results. An example was the 100 Day Programme of Colombos new Mayor, elected in 1997, which included improvements to solid waste management, and securing NGO and private sector inputs to renovate out-patient dispensaries, roundabouts, traffic lights, roads and public toilets. Mombasas new mayor, elected at the beginning of 1998, adopted a similar strategy of prioritizing immediate visible improvements. However, if cross-party support is not secured for longer-term initiatives, these are vulnerable to electoral reversals. For example, the development of participatory budgeting in Recife was temporarily threatened between 1997 and 2000 when a right-wing mayor held power. In Colombo, the Colombo Municipal Council (CMC)/central government collaboration to
URBAN POLITICS 83
implement the national Million Houses Programme (especially its settlement upgrading component) was seriously undermined by electoral reversals at national and local levels in the mid-1990s (see Box 5.4). On the contrary, the absence of restrictions on the number of successive elections for the same office that may be contested by a candidate may provide opportunities for the abuse of patronage resources, lessen accountability and fail to provide sanctions for poor performance. The same applies to appointments without mandatory restrictions on the duration of an appointment, it may be hard to displace unsatisfactory incumbents. Citizens and many Assembly Members in Kumasi breathed a collective sigh of relief when the expiry of the term of office of the citys Chief Executive enabled his removal and replacement. Responsiveness, accountability and sustainability at the local level may, therefore, be undermined by a poorly designed electoral system, inadequate arrangements for the exercise of executive authority and dysfunctional rules on terms of office. Local electoral arrangements are important in securing effective and legitimate local politics. However, additional provisions for securing wider societal participation in local decision-making can supplement them. Some alternative arrangements are explored below. Provisions for wider societal representation in local electoral politics Some of the ways in which representation that reflects society at large can be secured have been mentioned above partial proportional representation and reserved seats. Further ways of obtaining wider societal inputs in municipal decision-making include the establishment of advisory mechanisms, increasing transparency by making information and decision-making bodies publicly accessible, and increasing accountability by the establishment of external scrutiny bodies. Legislative provision for the mandatory establishment of consultative/ advisory bodies is more common in Latin America than Africa or Asia (except for the Philippines). Thus the Consejo Econmico y Social Communal in Santiago is comprised of representatives of organized civil society (40 per cent representatives of Neighbourhood Boards, the official CBOs, 30 per cent of other CBOs, 30 per cent of business and commercial entities). The forums that play a role in PB in some Brazilian cities and the consultative Municipal Councils in Filipino cities have similar compositions. Non-governmental urban actors may also be included in non-mandatory standing or ad hoc policy-making, administrative or coordinating arrangements. In Cebu, NGOs are included as members of commissions for the welfare of children and women and family affairs; peoples organizations (POs) are members of committees for the registration of POs and committees with functions related to informal settlement upgrading or relocation; and the members of a committee on street vendors include an NGO and vendors association. In Colombo, the Housing and Community Development Committee, the members of which included the Mayor, CMC officers, NGO representatives and six district leaders, was,
between the mid-1980s and the mid-1990s, a regular forum for coordinating the implementation of the Million Houses Programme in the city. Such arrangements do give greater representation to various societal interests. However, their standing in relation to elected bodies or to each other is not always clear, and the lack of a mandatory basis may weaken their influence.
URBAN POLITICS 85
86 URBAN GOVERNANCE, VOICE AND POVERTY IN THE DEVELOPING WORLD PB does increase the capacity of excluded social groups to influence the decision making process regarding the allocation of public resources. In addition, it increases the access of the poor to basic urban services and contributes to making local expenditure reflect the priorities of the poor. While the accountability mechanisms in the process are fragile, it allows for consultation and deliberation on the use of public resources. Although resources allocated through Participatory Budgeting are small, the evidence is that in an extremely unequal society like Brazil, these programmes are one of the few ways to transform public investments from favours into rights. (Melo et al, 2001, p170).
The barangay level of local government in Cebu also appears to benefit poor residents. Elsewhere, such arrangements are less well established and their weaknesses tend to outweigh their strengths. Nevertheless, they have potential for increasing the responsiveness and accountability of government to local communities and poor citizens. For example, the Integrated Development Planning (IDP) process in South African cities is still in its infancy. However, with political commitment at metropolitan and sub-metropolitan levels, there is still scope to support disempowered groups to develop the skills to enable them to play a full role and ensure its representativeness. A prerequisite for exercising voice, increasing responsiveness and securing accountability at both city and local levels is transparency. This depends on the political culture and legislative provision, such as Right to Information legislation or provisions in the local government or planning legislation. It also relies on media coverage of local government issues and openness to resident views. Although there is, in theory, a free press in all the case study cities, in practice it varies in importance, coverage of local government issues and freedom. The local press actively covered a range of city government issues in Ahmedabad and Visakhapatnam and, more unexpectedly, in Mombasa. Elsewhere, its role was more limited, generally concentrating on service delivery problems (eg in Colombo), especially where it practised self-censorship or was vulnerable to legal action or other sanctions (eg in Kumasi). In the latter, local radio had started to adopt a more active role, although the previous Municipal Chief Executive limited media coverage by selecting those journalists permitted to attend the (irregular) Assembly meetings. In Johannesburg community radio plays a significant role, although in some areas, such as Soweto and Alexandra, it had started to come under the control of the local lite. The evidence from the case study cities shows that, for sub-city levels of government and other mechanisms for participation to be effective, representative, accountable to citizens and capable of holding higher levels of local government to account, they need: a supportive political context in which higher levels are committed as a matter of principle to neighbourhood representation and participation, or see effective lower levels as being in their own political interests; a strong civil society, to play an advocacy role on behalf of the poor, act as a watchdog to ensure accountability and organize citizens to claim their rights to land, services and a say in decisions;
URBAN POLITICS 87
considerable ongoing support from NGOs and other organizations, to inform and empower citizens, elected representatives and local leaders to fulfil their roles and responsibilities; transparency, based on legislative safeguards and a free and active local media.
supporters and elected representatives may be deferential the representative is trusted to make decisions on behalf of the electorate. Alternatively, the representative may be regarded as an agent, elected in the expectation that he or she will deliver particularistic benefits to individuals or communities. This may require that the agent places him or herself in a position in which access to resources can be secured, for example by joining the ruling party. Poor people may identify with a party because of its stated policy commitments, but seem to be more likely to behave instrumentally. They trade votes for specific promises of assistance, or support the candidate they believe is most likely to be both responsive and able to assist with the many problems that characterize informal settlements and the lives of poor households in the city, as illustrated for Cebu and Bangalore in Boxes 5.1 and 5.2. In so-called vote bank politics, the poor are far from passive. Sometimes votes are purchased or candidates appeal to sectarian allegiances. However, vote bank politics can work in the interests of the poor. Local leaders in a competitive political situation such as Bangalore often form alliances with more powerful groups and use their ability to deliver votes to win rewards and concessions for residents. Nevertheless, merely voting for a candidate may not result in the promised assistance, because politicians may be unwilling or unable to fulfil their campaign promises. The most common political practices by poor residents focus on developing personalized ties with elected representatives, party leaders or street level officials. Such ties may be expressed in vote bargaining and clientelism (as in elections for Cebus Mayor, or the support base of ward councillors in cities such as Bangalore. They also take the form of petty bribery as when a minor official is tipped (Kumasi) or given tea (Mombasa) or speed money (Ahmedabad). However, where necessary, residents will approach others who may be perceived as more sympathetic or more likely to secure a positive response. For example, in Kumasi, residents will often approach their MP or the local chief, who can, if they are willing, bypass the Kumasi Metropolitan Assembly (KMA) and contact more senior central government officials or politicians on their behalf. In Bangalore, they will, if they can, contact state level bureaucrats or, less easily, politicians. Because the building of political capital requires long-term relationships with local politicians and bureaucrats, the sheer persistence of the poor is critical to their survival and success.
Political alliances
Poor residents or small-scale enterprises (SSEs) may also form alliances and organize to increase their political influence. But there are obstacles to effective organization. In Bangalore, for example, caste-based and religious rivalries hinder the political alliances of the poor. However, local residential environments and economies bring multiple poor groups into close contact and their mutual interdependence forces informal collaboration across these primary divisions. People are highly conscious of their distinctiveness, but ready to sink their differences in common party allegiances and cross-class activities such as religious festivals in order to maximize their political influence. CBOs in
URBAN POLITICS 89
informal settlements generally form around the need to obtain secure tenure and infrastructure. Once these objectives are achieved, they may become dormant, as in Colombo, unless access to decision-making processes or further resources depend on a CBO being present. Indeed CBOs may be initiated by or in response to external bodies, as in the Philippines Community Mortgage Program (CMP) or South Africas Integrated Development Plan (IDP) process. Their roles are discussed in more detail in Chapter 7. SSEs are generally organized on an occupational or sectoral basis for example taxi drivers, auto-repair workers and market traders in Kumasi. Sometimes they are able to successfully protest an undesired policy or regulatory change, or lobby for support from local government. In some political circumstances, a combination of an open and flexible local political culture and well-organized civil society has led to outcomes that are favourable to small and micro-enterprises. For example, in Cebu City, policy towards street vendors has been worked out locally, through threats of demolition, protest, and resistance, leading to concessions and unwritten policies (see Boxes 7.3 and 8.2). Thus the authorities have conceded tolerance in return for compliance with informal rules. The operators in turn have learnt that they have rights and understand the need for engagement with regular processes of policy and rule-making. Individual vendor groups have formed into federations and common interest organizations of informal sector operators have emerged at city level. The result has been that confrontation and favour seeking behind closed doors has gradually been replaced with more open processes of negotiation between informal sector operators and the local authority. More commonly, however, SSEs are weakly organized, especially across sectors. In addition, as explored in more detail in Chapter 8, their gains are limited or temporary, especially if their interests conflict with those of formal business. High-income residents and formal enterprises have greater organizational capacity and when they organize collectively can exert more influence on decision-making than organizations of more numerous but less economically powerful and socially well-connected poor residents. In 1998 in Johannesburg, for example, the Sandton Residents Association resisted an increase in property rates by organizing a boycott and a large number of appeals against property revaluation, worsening the citys financial crisis. If enterprises are sufficiently large and important to the local economy, they can lobby individually and informally to protect their own interests. Sometimes, business organizations may be represented on city level advisory or decision-making bodies. However, often the relationships between businesses and the local political system are ambivalent. Moreover, such arrangements may not represent the enterprises on which the poor depend. For example, it is the government, not the traders, which appoints a traders representative to the KMA.
Trade unions
In cities with large formal sector wage-based economies and a history of trade unionism, trade unions may play an important political role. However, they are also primarily concerned to protect the interests of their own members, who
are unlikely to include the poorest. Trade unions concentrate on workplace rather than neighbourhood issues, for example, the dockworkers in Mombasa. Their activities may contribute to a leftward leaning political culture, as in Brazil, but they may also resist pro-poor changes where they perceive these to be against the interests of their own members. For example, in Indian and South African cities, municipal workers have resisted reorganization of service delivery, especially when it involved corporatization and private sector participation, because of fears for their jobs, even when one purpose of the reforms was to raise capital to extend services to low-income areas. Where traditional industries, such as textiles in Ahmedabad, have declined, or where economic liberalization has made waged employees more insecure, as in Visakhapatnam, unions are unlikely to be important political actors.
NGOs
The presence, purposes, resources and capacity of NGOs varies between the case study cities, but their contribution to supporting poor peoples organization and political engagement or pressing for pro-poor policies is generally limited and contradictory. The majority of NGOs are concerned with welfare and are engaged with national or local government only insofar as they may deliver a service on its behalf. In this role, they may act as intermediaries between government and poor residents, but their relations with the state are either dependent, or, where the government perceives them as competitors for external resources, competitive. Many such NGOs (especially indigenous ones) are small and poorly resourced. Those NGOs engaged in advocacy on behalf of poor citizens or supporting the organizations of the poor are relatively few in number in most of the case study cities. Moreover, NGOs are often distrusted both by poor residents and by governments, as being self-interested, unrepresentative and unaccountable. If they adopt conflictual tactics, these may be resented by government and, especially if they are used repeatedly, lead to resistance rather than positive change. Where NGOs have played a role in gaining benefits for poor residents, as in Cebu, they have worked alongside local government on particular issues or projects (see also Chapter 7).
Religious organizations
Religious organizations are widespread and numerous in low- and middleincome countries and their number and membership are often growing. Sometimes (as in India) political affiliation is, at least in part, on religious grounds. Sometimes also, the concern of specific religious organizations for social justice has been an important direct or indirect political influence. In the case study cities, for example, branches of the Roman Catholic church in both Brazil and the Philippines were instrumental in conscientizing and organizing the poor, while in Kenya the churches joined other civil society actors in pressing for democratization, constitutional reform and the rule of law at the national level. On the whole, though, in these case study cities, religious organizations are not actors in local politics, concentrating instead on the spiritual and sometimes material welfare of their adherents.
URBAN POLITICS 91
Poor residents are sometimes passive in political terms, because of their lack of knowledge of how government operates, disillusionment with their experience of the political process, preoccupation with day-to-day survival or social sanctions against participation, especially by women, in public life. However, more often they devise active tactics and strategies, many of which diverge considerably from the expectations of traditional liberal democratic theory. Sometimes they are successful in securing political representation, influencing decisions and obtaining redress for grievances. However, even in the minority of countries where there is political commitment to addressing the needs of poor people, conflicting interests and inequalities of knowledge, capacity and resources may result in outcomes falling far short of intentions. Even more commonly, the greater economic, social, organizational and informational power of high- and middle-income residents and formal sector businesses reduces poor residents access to political influence and decisionmaking processes and forces them into clientelist strategies. Conflicts of interest among poor residents and between groups of informal sector operators exacerbate the problem.
Conclusion
The characteristics of urban political systems are influenced by the ways in which societal interests interact with formal and informal political structures and processes. The characteristics of the articulation have implications for the nature of government and the way government bureaucracies operate. They also affect the political practices of politicians and citizens, as expressed in the relationships between those with political power and the electorate, as well as in the preoccupations of everyday politics. In addition, they influence the ways in which civil society is organized and engages in political activity. Using ideas about the nature of democratic political systems and the research findings, an attempt is made to develop a typology of urban political systems in Table 5.1. Notionally, this represents a progression, from authoritarian regimes on the left side of the table, to consolidated representative democracy on the right. In practice, of course, not only is such a progression by no means certain, but also most contemporary urban political systems have mixed characteristics. Nevertheless, we can distinguish between them with respect to the basic features of the political system and power structure. Such features tend to be associated with characteristic relationships with the administrative apparatus, citizens and civil society organizations. These in turn have different implications for the ability of poor residents to exercise political voice and the strategies and tactics employed by CSOs, as well as for the responsiveness and accountability of government. The cities studied have, with considerable hesitation, been located in the suggested typology, although this risks being both much too static and also greatly over-simplified. In both autocratic and oligarchical authoritarian regimes, participation in politics means demonstrating support for leaders, who use their control over resources to respond selectively to citizens needs and priorities, using
Power structure
Command, Formal representative democracy + clientelistic under oligarchy; political system populism Lobbying and Pro-patron Balanced clientelism clientelism pressure Subordinated bureaucracy Porous bureaucracy Personal favours by exchange Implementation, crisis management CSOs organize to lobby patrons, press claims and demand reforms to make the democratic system more responsive and accountable. Welfare and development CSOs tolerated. Compliant CSOs may be used to deliver services. Alternative models developed by CSOs may be influential. Group favours by exchange Group favours by lobbying and pressure
Gratitude
Mass acclaim
Welfare CSOs tolerated if they do not challenge the regime and supplement government resources. Advocacy CSOs demanding political reform suppressed.
Welfare and development CSOs tolerated if they do not challenge the regime. Compliant CSOs may be used to deliver services. GONGOs sponsored by the state. Advocacy CSOs demanding political reforms bought off by concessions. Cebu to 1980s Bangalore Ahmedabad Visakhapatnam Santiago
Policy, implementation and partnership Governments respond to CSOs organized to express voice and call government to account. CSOs enter into partnerships with public agencies.
URBAN POLITICS 93
government resources to coerce the population or manufacture consent. In such circumstances, only CSOs that do not challenge the regime are tolerated. Independent or government sponsored NGOs may be used to deliver services on behalf of the state or fill gaps in the services it provides; sometimes in populist regimes with formal democratic systems, development oriented NGOs may be tolerated. However, advocacy organizations demanding political reform are suppressed or co-opted. Nevertheless, welfare NGOs may start to question state service delivery policy and performance, and development NGOs may develop the capacity of citizens to organize and articulate their demands for political and other changes. Many of the cities studied showed these political characteristics in the 1970s or 1980s. As demands for democratization gather pace, the space for political activity is likely to widen, either gradually (with many setbacks) or suddenly. With the introduction of formal democracy, urban governments have to become more responsive. They continue to use compliant CSOs to deliver services and may respond to criticisms of their performance by expanding such use. They may introduce other reforms such as the use of private sector or community organizations to improve service delivery, sometimes drawing on alternative models developed by NGOs. The increased space for political activity makes it possible for civil society to organize, demand reforms to make the democratic system more accountable and develop the capacity of voters to exercise their political rights. However, the limited capacity and legitimacy of government agencies, especially at the local level, the limited efficacy of mechanisms to ensure good performance and accountability and the immaturity of political institutions result in pervasive clientelist relationships and corrupt practices. These may bring some benefits for citizens, including the poor, but at considerable cost, both in terms of time and money to residents and entrepreneurs, and with respect to the legitimacy of the political-administrative system. Although all the cities studied were located in countries which had reestablished or reinforced democratic politics, in all of them systems of formal democracy coexisted with more or less pervasive clientelist politics, providing differing levels of access to the poor. Given favourable political circumstances and increasing levels of political awareness, democratic consolidation may occur. There is arguably potential for this, at least in Recife and Johannesburg and perhaps in some of the other cities studied. If a virtuous circle of increased responsiveness, leading to increased willingness to pay taxes and user charges, better performance, increased accountability and greater political legitimacy occurs, then access to political influence and basic services may be assured as of right. In turn, the capacity of public sector agencies to deliver is likely to increase and, while CSOs continue to lobby for improved performance and other reforms, they are increasingly likely to be willing to act in partnership with public sector agencies. Gradually, statesociety relations come to be based on an (implicit) social contract, and levels of trust, while always provisional, start to increase. Although conflicts of interest persist, and economically powerful actors often win in battles for resources, the deeper embeddedness of democratic values and the more influential political voice of poor citizens increase both the pressure for redistribution and the proportion of
resources directed to reducing poverty. That the conflicts and contradictions do not disappear, even in societies with well-established democratic systems, strong welfare states and effective and well-resourced local government, is illustrated by the experience of many developed countries. Ironically, low and declining turnout in countries such as the UK demonstrates that local politics is not always alive and well even in these circumstances. The road from democratization and decentralization to consolidated local democracy, effective local government, and constructive dialogue between local government and poor citizens is long and rocky. Some of the case study cities have made greater strides than others towards consolidated democratic institutions and mature political processes capable of resolving conflicts of interest and responding to the poor. These differences can be explained by their political history and culture, the characteristics of the struggle for democratization, the nature of the democratic transition, their administrative capacity and the levels of resources available. Even in these systems, many gains made by the poor are achieved by fostering patronclient relationships with elected representatives and officials at all levels of government, rather than as of right. Where patronclient relationships are pervasive, although poor people may be able to gain limited benefits and prevent harm to their livelihoods and living conditions, local politics is essentially extractive and inefficient. It is a politics in which government is seen as a source of goods to be consumed privately by officials and residents alike. However, it has little to offer on how publicly available goods can be expanded so that all citizens can access basic services as of right, or if these are not delivered, how the responsible agencies can be called to account. It is a politics based on favouritism and special pleading not entirely dysfunctional but ultimately limited in its capacity to reduce poverty and inequality.
Notes
1 Richard Batley also reviewed the findings of the city case studies with respect to political aspects of pro-poor governance. His insights are acknowledged with thanks, especially the typology included in the conclusion of this chapter, which has been developed from an initial draft he prepared. However the views put forward here are those of the author. Civics (civic associations) were community-based organizations that were ostensibly apolitical structures that mobilized around quality of life or development issues in the black townships, represented communities in township-based politics and linked them into the mass democratic movement (Beall et al, 2002). The number of councillors was reduced and the relative proportions of those elected through each channel changed (to 50:50) following the reorganization of Johannesburgs local government. Strictly speaking, district and metropolitan chief executives in Ghana are only nominated by the President, subject to the approval of the district/metropolitan assembly, which can reject the nomination. However, the position of the former MCE of Kumasi was exceptionally strong because of his close family relationship to the First Lady, his financial backing for the governing political party and his link to the Asante royal family.
3 4
Chapter 6
Urban Government: Capacity, Resources and Responsiveness
Nick Devas
organizations (NGOs). The heightened international concern with poverty puts pressure on them to consider how to respond to the needs of the poor and excluded. Perhaps most significantly, the process of democratization, accompanied by the increasing strength of civil society, has forced city governments to start becoming more accountable and more responsive to citizen demands. While the performance of city governments in the South generally remains very poor, there are examples in the case study cities of at least the beginnings of more responsive urban government. What accounts for differences between city governments in terms of performance and responsiveness? And what accounts for whether and how they change? Chapter 5 has examined the issues in terms of the urban political processes, both formal and informal. This chapter focuses on the organizational, financial and administrative aspects of city government. But these aspects are closely inter-related. As indicated in Chapter 5, the ways in which city governments respond to the situations they face depends greatly on the national and local political context, including the legal and institutional framework, and the nature of political processes at national and regional level. This includes the extent of political space available to civil society, and the ability of civil society organizations (CSOs) (particularly those of the poor) to make use of that political space to influence the political and administrative processes. This aspect will be considered further in Chapter 7. City governments themselves face a number of specific constraints institutional, organizational and financial on their ability to respond to the needs of citizens, particularly the poor. How far they are able to overcome these impediments depends in turn on both the national political context and on the qualities and abilities of local leadership. In this chapter, we will examine some of the constraints facing the city governments in the ten case study cites, how these constraints affect their ability to respond to the needs of the poor, and how far they have been able to overcome these constraints. The greatest obstacle in many of the cities is finance. We therefore look in more detail at how city governments generate the resources they need, and how this process of revenue generation impacts on the poor. We also look at how far cities spending patterns benefit the poor, and the processes of decision-making about the use of available financial resources. In the last section we look at some of the specific ways in which city governments have sought to become more responsive to the poor, and at the crucial role of civic leadership in this.
Boundaries
The boundaries of several of the cities in this study have not been extended to take account of the citys growth. Many of the urban poor settle on the periphery, beyond the city governments jurisdiction. This is where population growth is most rapid and where infrastructure and service needs are greatest but resources are most limited. In Ahmedabad, for example, the urban periphery is divided between 163 village, town and municipal councils, together with a number of special purpose agencies. The capacity of the adjoining local governments to address the needs of the poor may be very limited indeed. The result is an uncertain patchwork of provision, or no provision at all (Dutta with Batley, 1999, p72). A similar situation applies in Bangalore. In Santiago, the city is divided among 34 municipalities, with the poor concentrated in certain municipalities, and a ten-fold difference in municipalities revenues per capita between the richest and poorest (Dockemdorff et al, 2000, p182). Cebu City covers only the core of the metropolitan region yet contains a substantial proportion of the conurbations industrial and commercial resources. A similar situation applies in Recife. Under apartheid, the Greater Johannesburg area was divided into 13 racially separate jurisdictions with vast differences in autonomy, political legitimacy, and fiscal and management capacity. The interim post-apartheid arrangement of four local councils under the Greater Johannesburg Metropolitan Council (GJMC) was an attempt to introduce a greater degree of equity between areas of the city. The effect was unsatisfactory, with bizarrely shaped jurisdictions and political resistance to transfers of resources between the local councils (Beall et al, 1999). Consequently, from 2001, the lower level was abolished, leaving a unitary city council for some 4 million people. Whilst this has brought about the objective, long cherished by the African National Congress, of one city, one tax base, it has made for a huge institution remote from its electorate.
Responsibilities
Responsibility for many of the services most vital to the poor often does not lie with the city government. In a number of countries (the Philippines and much of Latin America) decentralization has expanded the powers and responsibilities of city governments. However, in several of the case cities, city governments have quite limited responsibilities: waste management, sanitation and local roads, but not primary healthcare, education, electricity, police and security services, and often not even water, land and housing. These services are typically provided by central government, parastatal agencies or even private companies. This severely constrains the ability of city governments to respond to the needs of the urban poor. In south Asian cities, responsibilities are often divided between a great range of special purpose agencies such as development authorities, slum clearance boards, and public utility companies. These tend to operate in competition with each other and with the municipal government, creating problems for long-term operation and maintenance, and breaking the accountability link with local voters.
For example, out of 23 people on the board of the Bangalore Development Authority (BDA), only two are elected members of the Municipal Corporation (Benjamin and Bhuvaneswari, 1999, p52). Benjamin (2000, p54) observes that, compared to the access which the poor have to the municipal corporation through their elected councillors, state agencies and development authorities are more accessible to middle- and upper-income groups (see Boxes 5.2 and 5.3). In Kumasi, the decentralization law placed education and health services under the Metropolitan Assembly, but in practice the resources for these, and therefore effective control, remain firmly with the central ministries (King et al, 2001, p65). Similarly, Mombasa Municipal Council (MMC) nominally has responsibility for primary education, but with central government paying the teachers, head teachers being responsible for running the schools, and maintenance of schools being met largely from fees, the municipal council has little more than an advisory and supervisory role (Gatabaki-Kamau et al, 1999, p112). With decentralization, some municipal governments have been given increased responsibilities for welfare and poverty alleviation, but often without the resources to meet those responsibilities. In Johannesburg, this unfunded mandate has become a major political issue. According to the Ghanaian Local Government Act, Sub-Metro Assemblies are legally responsible for formulating poverty reduction interventions but they have no resources for this. In India, the 74th Constitutional Amendment proposed that local bodies should be responsible for poverty alleviation programmes but this has not been backed up with commensurate financial resources.
to municipal governments from arbitrary interference from states, problems continue. In Colombo, the provincial government has on a number of occasions obstructed the municipal governments attempts to invest in poor communities, perceiving this as a political threat. It has also sought to control the flow of central government funds to the municipal corporation, creating delays and preventing services from being improved. Overlapping responsibilities and rivalry between central and municipal government has also been an issue in Mombasa, where the centres District Administration has the same boundaries as the Municipal Council. There have been numerous examples of central government interference in the local political scene in Kenya, including the curbing of the powers of mayors in opposition controlled local authorities and the reassignment of senior officials. Bureaucratic controls imposed by the centre often do little to improve local performance, resulting instead in delays, costs and rent seeking opportunities for those charged with enforcement. In Kumasi, the tight control exercised by the government appointed Chief Executive has much to do with the fact that the region is a centre for opposition politics. By contrast, one of the factors in Cebus success in the late 1980s and early 1990s was the good relationship between city and provincial governments, although that seems to have broken down in recent years.
Availability of information
As noted in Chapter 2, the data which city governments possess about urban poverty is woefully inadequate and does little to illuminate the differentiated nature of poverty or the livelihood systems of the poor. City governments also lack essential data about environmental conditions and the impact of these on different groups. In some cases, they lack even basic information about the services for which they are responsible: service levels by area, service reliability, who benefits, who pays and how much, and so on. Citywide statistics sometimes exclude informal or irregular settlements, making nonsense of aggregate figures. Without such information, no meaningful poverty strategy can be developed or implemented.
Financial resources
Lack of financial resources is perhaps the most significant factor preventing urban governments from addressing the needs of the poor. Most of the case study cities have potentially substantial resource bases but are unable to exploit them effectively. Local tax instruments are often unsatisfactory low yielding, politically sensitive, difficult to collect, economically damaging and impinging on the poor. Large amounts of tax remain uncollected because of political, administrative or legal obstacles. Mombasa Municipal Council, for example, is effectively insolvent, and is barely able even to pay its staff. Despite its vast revenue base, Johannesburg was forced into financial crisis because of low collection rates on electricity and water charges as well as
property tax (see Box 6.2). Unwillingness to pay in Johannesburg is a legacy from the days of political struggle but has proved difficult to change. As an example of the tensions surrounding the issue, two councillors were murdered during the governments campaign to improve utility charge collections. Enforcement action against defaulters is undermined by intimidation against utility company staff, and the ease with which illegal (re)-connections can be made. Whilst some of the poor may benefit from non-payment, there are serious problems of inequity between those who pay and those who do not. Later in this chapter we look at the attempts cities have made to address their financial constraints.
Some cities receive significant transfers from the central (or state) government. For example, transfers provide two-thirds of Kumasis budget, through a combination of salary support and the District Common Fund (a national tax sharing arrangement). In Recife, national tax sharing represents over half of the citys budget, although the proportion is declining. In Cebu, decentralization increased the proportion of transfers in the citys budget from 30 per cent to 40 per cent, although cuts in the national tax-sharing rate in 1999 reduced this again. In both Recife and Cebu, these transfers allow the cities concerned full discretion in the use of the money. By contrast, in South Africa, redistribution of national resources in favour of poorer local governments has meant that the transfers received by Johannesburg have steadily declined. These now stand at less than 5 per cent of the citys budget, most of which is in the form of grants for specified purposes (mainly health services). In India, state and central transfers represent only 1020 per cent of the budgets in Ahmedabad, Bangalore and Visakhapatnam, and much of this is in the form of specified grants. Mombasa now receives around 12 per cent of its budget from national tax sharing, but until 1999/2000, there was no system of inter-governmental transfers in Kenya, so Mombasa received nothing at all from the centre.
commonplace. In Kumasi, while the official charge at a health clinic is Cedis 500, it is common for people to have to pay ten times that amount if they want to be treated (Devas and Korboe, 2000, p130). In Bangalore, it is routine for the police to require daily payments from street traders for protection. Elected councillors often perceive their position primarily in terms of the rents they can extract, in order to finance their re-election campaigns. In Kenya, land-grabbing by well-connected politicians has been a major issue.4
impact of cities spending patterns on the poor, and the processes of decisionmaking about resource use.
Ahmedabad 1996/7
$40
3.9%
65%
Total revenue per capita ($) (excluding borrowing) Total revenue per capita (as % of GNP/capita) Local own revenue (as % of total revenue) Total revenue growth in real terms Total per capita revenue growth in real terms Main local taxes (as % of total revenue)
Overall: static Good (45% pa) (Business tax: good) (esp. Octroi) Slight decline Modest increase
Property tax 26% Octroi 58% Business taxes 20% Property tax 24% Amusements 7% 6086% <20% (dep.on definitions) General grant (LATF): 7%; specified grant for roads: 5% High Specified grants 3%
Market fees 12% Services tax 20% Bus.licences 7% Property tax 11% Property tax 5% N/A 67%
Salaries subsidy (20%); tax share (District Common Fund) 45% High Small (mainly refinancing) High None
Mainly conditional Deficit/ grants for conditional education 12% grants plus compensation for abolished taxes: 23% High High None 20% (mainly for roads and infrastructure)
High N/A
Degree of local discretion High in use of overall resources New borrowing 45% (mainly for (as % of budget) one big project)
Note: All figures are approximate only. Source: City case studies
governments can only levy those taxes which are specified in national (or state) legislation, and national regulations often limit the scope for increasing tax rates. Borrowing is one way that cities have used to augment their resources for capital expenditure. With the development of the municipal bond market in places like India, the opportunities for borrowing have increased. However, loans have to repaid, so unless they are used for projects that generate an adequate return, the burden on city finances increases. In Cebu, borrowing for the South Reclamation Project (a project to reclaim a large area of land close to the central business district) nearly doubled the size of the citys budget in 19992000. Although this project was expected to generate substantial returns, the debt burden was considered to be too great, and a large part of the project was transferred to a state agency. Both Ahmedabad and Bangalore have issued bonds for road construction, and for water and sewerage in the case of Ahmedabad, and both have borrowed from HUDCO (the national housing finance institution) for housing and urban infrastructure. While Johannesburg borrowed substantially in the past, recent borrowing has been mainly to refinance previous loans. The weak financial positions of Mombasa and Kumasi have prevented those cities from borrowing at all in recent years.
rarely come anywhere near covering the costs of the services: typically, they cover about 1015 per cent of expenditure on health in the case cities. A much greater problem for the poor is the unofficial charge levied by those controlling access to the service, since these are often much higher than the official charges. Charging for sanitation (such as public toilets) also has an impact on the poor, but can be justified as a way of covering maintenance costs, thereby ensuring that the service continues to be usable. However, this has been highly problematic in Kumasi, where contracts were awarded to members of the Metropolitan Assembly who then failed to use the revenues for maintenance, thus imposing significant burdens on low-income households (see Box 9.2).8 Charges may be justified where only certain people receive the service. In Kumasi, household waste collection is provided only in the high-income areas, and a charge is levied. However, the charge is not commensurate with the cost of the service, so that high-income residents effectively benefit from a large subsidy. Charging for water can also be justified on the grounds that finance is needed to maintain and expand the system. General under-pricing of water is likely to mean that there are no resources to extend water supply to areas where the poor reside, obliging them to continue using much more expensive vended water or polluted sources. In Bangalore, there is a system of free public standpipes, but resources are lacking for maintenance and extension of the network. In Mombasa, while the piped network nominally serves most of the city, in large areas no water flows because there are no resources to repair the network or increase bulk capacity. In Johannesburg there is near universal provision at relatively high standards, but the problem is the absence of a system of targeted subsidies to enable the poorest to pay the relatively high bills. Low-income households, for example pensioners, can face utility bills that may exceed their entire income (Beall et al, 2001). Although the city established an Indigence Register to identify those against whom enforcement action would not be taken, this did nothing to enable those on the register to pay their accumulated debts. In addition, the register soon became outdated, and there was scope for abuse such as meter switching.9 Alone among the case cities, Santiago has a comprehensive system of relating water bills to ability to pay, based on household income data, although how well it works is open to question. A more common approach is to use progressive block tariffs to charge more to large volume consumers, on the assumption that the rich consume more water than the poor. But in Kumasi that system takes no account of how many households use one tap, thereby penalizing those living in communal blocks, whatever their income.
Cebu 1999
Ahmedabad 1996/7
45% * 2.5% **
20%
25% *
30%
2.3% Debt-service ratio (approx) (% of revenues net of loans) Health 14% Water taps 4% Provision for SC/ST 2%
9%
Health 9% Div. for Welfare of Urban Poor Purchase of lots/ slum upgrading Grants to barangays None
Health 14% Primary educ. 20% Water taps 9% Slum improvement Economic opportunities
Nothing significant
Urban infrastucture in poor areas Education Health Social assistance GJMC budget incorporates 4 MLCs (now abolished) 4 Sub-metros 20 Town councils 403 Unit committees but none really functioning None but 6 regions and 15 micro-regions for participatory budgeting
Part of budget None formally handled through but Wards Zones and Wards have some significance
Note: All figures are approximate only. * Including South Reclamation Project. ** Excluding South Reclamation Project. *** Percentage of net budget (including only net revenue from water, electricity) Source: City case studies; GNP data from World Bank 2001a
The most obvious area of expenditure to benefit the poor is that on health, since the benefits of publicly provided health services are in principle available to all, while higher income groups are likely to opt for private services. Expenditure on health typically represents 1015 per cent of city budgets. Similarly, where the municipality provides primary and nursery education for free, the poor may be expected to benefit, so long as access is not dependent on unofficial charges. In Ahmedabad, 20 per cent of expenditure goes on education. The Indian cities subsidize a network of free public standpipes which are clearly of benefit to low-income groups. In Ahmedabad, water provision accounts for about 10 per cent of the budget, in Bangalore about 4 per cent. But many areas are not served often the areas where the poorest live, and maintenance problems and lack of capacity in the system mean that many standpipes operate only intermittently or not at all. In Johannesburg, waste collection is provided virtually throughout the city, so that some of the 4 per cent of the budget spent on that could be considered as benefiting the poor. Some cities spend money on slum upgrading and local level infrastructure and services. This could be expected to benefit the poor directly, although benefits generally accrue more to property owners than to tenants. By contrast, in many systems, subsidies for public housing provide little benefit for the poor, since they rarely occupy such housing. Cebu has a specific department for the Welfare of the Urban Poor, although its budget is minimal. By law, the city is required to set aside 20 per cent of revenues for a Local Development Fund, and some of this is used for projects that benefit the poor, such as artesian wells, fire hydrants, drainage, purchase of plots and so on. But actual expenditures from this fund are much lower than the statutory requirement. In Bangalore, the law requires 18 per cent of the budget to be allocated for Scheduled Castes and Tribes (SC/ST), but in practice actual expenditures are more like 2 per cent (depending on definitions). By contrast, in Ahmedabad, the city is required to assign 10 per cent of revenues to slum areas and in most years this proportion has been exceeded. In Kumasi, 20 per cent of the District Assemblies Common Fund (DACF) is supposed to be allocated to incomegenerating activities for the poor. But this has not generally been complied with, and what little has been spent in this way has been allocated in a most opaque manner (King et al, 2001, p155). By contrast, participatory budgeting in Recife has resulted in some shift, albeit modest, in the pattern of expenditures towards social expenditure and infrastructure in poor neighbourhoods (Melo et al, 2001, p118). Other expenditures are less obviously of benefit to the poor. Most cities spend 2530 per cent on administration, including the mayor and council, and revenue collection, but figures vary depending on what is included. Most cities spend large amounts on public works, much of which is for roads. The benefits to the poor of such expenditure are likely to be marginal at best, through the effects on economic growth, or negative at worst, as the poor are displaced for new road construction or to enable traffic to flow freely. Expenditure on street lighting, drainage and public parks may benefit the poor, but since it is generally commercial and high-income residential areas that are best served, the benefits to the poor are small.
The huge resources used by Cebu for the South Reclamation Project will have little direct benefit for the poor. However, it is possible to justify such an investment if, as is claimed, it will generate substantial financial returns that can then be used for pro-poor expenditures. Other cities such as Bangalore, Ahmedabad and Johannesburg have all undertaken major loan-financed projects, some of which are of questionable benefit to the urban poor. Whether or not large-scale infrastructure improvements aimed at enhancing local economic growth prospects ultimately benefit the poor is open to debate. But there are clearly risks that such mega-projects actively disadvantage the poor, by pre-empting available land, expelling the poor from valuable locations and upsetting fragile trading relationships. This is demonstrated clearly in the case of major road and market projects in Bangalore (Benjamin, 2000). Loan financing for bulk water supplies and sewerage projects can be justified, so long as costs are recovered from consumers, particularly high-income and commercial consumers. But there are risks that heavy debt-service obligations may pre-empt resources which might otherwise be used to address the needs of the poor. One of the over-riding considerations in the restructuring of the Greater Johannesburg Metropolitan Council after the fiscal crisis of 1997 was to reestablish the citys credit-worthiness with lending agencies. In order to do that, cherished anti-poverty initiatives were abandoned or put on hold. What all this shows is that, unless there is a clear policy of redirecting expenditures towards the poor, which is followed through into actual service and infrastructure delivery, the normal, budgeting-as-usual pattern of city government expenditure will have only incidental benefits for the poor and in some cases could actually disadvantage them. It should also be noted that much of the public spending which relates to the urban poor bypasses city government altogether. In most of our case cities, the bulk of funding for health and education comes from central ministries. Even where such functions are formally decentralized, funding and therefore control may remain with central ministries, as in Ghana. In several cases, water supply is the responsibility of state enterprises, and, in Kumasi, this has now been privatized. In South Africa, funding for housing and related infrastructure probably the most significant public spending on urban poverty flows through provincial rather than municipal governments. In Bangalore, there has been a perceptible shift in responsibilities from the municipal government to state boards, remote from the urban poor. The Karnataka Slum Clearance Board, for example, appears to have delivered virtually nothing of benefit for Bangalores poor residents. Meanwhile, the housing and land-banking projects of the Bangalore Development Authority (BDA) have pre-empted land resources for the benefit of higher-income groups and for revenue generating activities. Ironically, however, the poor have at times benefited by illegally occupying land owned by BDA.
is how far citizens especially the poor are able to have any influence on these decisions. What emerges is that even where the poor are able to exert some influence on the formal budget process, outcomes in terms of actual expenditures depend more on informal processes. These informal processes are often under the control of a small executive group. All the cities have formal systems by which the elected members of the city government as a whole approve the annual budget. In cities with directly elected mayors (Cebu, Recife, Santiago), responsibility for preparation of the budget is in the hands of the executive, ie the mayor and his/her cabinet, with the council or legislature having power only to reject or amend the budget. In the executive council system, councillors have in principle at least a greater opportunity to influence what goes into the budget. But since the locus of executive power is more diffuse, paid officials (chief administrator/clerk and treasurer/finance director) tend to play a greater executive role. Whatever the system, there is much negotiation over the final shape of the budget. The ability of individual councillors to influence the outcome in the interests of their constituents (or, indeed, in their own interests) depends on their position, power and negotiating skills. Whatever the formal arrangements, the executive (taken here to mean those at the centre of the processes, whether the mayor and cabinet or senior officials) have considerable scope to influence the outcome, through their control of both the process and the information. Their construction of the draft budget generally allows little scope for ordinary elected representatives to suggest changes, even if the latter had the information and technical skills to do so. More significantly, formally approved budgets often bear little relationship to what is actually implemented. In the Philippines and in much of Latin America, the budget represents only an authorization to spend if funds are available. In Cebu, budget approval is followed by a process of allotments, which is mainly in the hands of the executive. These allotments depend on the resources actually available and are generally much smaller than the allocations in the approved budget. In Recife, while participatory budgeting has shifted budgetary priorities in favour of the poor, it has had less impact on actual expenditures because resources have not been available to implement all the agreed projects. At the same time, the mayor is empowered to spend up to 30 per cent beyond the budget if resources permit potentially a huge area of discretion. In Mombasa, and quite commonly in Africa, local budgets are based on unrealistic projections of revenues. This is done in order to satisfy the demands of councillors while at the same time meeting the legal requirement for a balanced budget. Since the projected revenues do not materialize, budget cuts have to be made during the year. Because of the difficulty of cutting staff or other overhead costs, cuts fall mainly on maintenance, service operations and capital projects including the pet projects of local councillors. The executive effectively operates a shadow budget which determines how resources are actually used and protects those areas deemed to be essential, notably salaries and councillor allowances. There are other devices at the executives disposal. The power to bring forward or postpone expenditure gives the executive considerable discretion. Supplementary budgets, which are usually subject to less rigorous scrutiny than
the main budget, can provide some room for manoeuvre. Extra-budgetary funds, too, offer considerable scope for discretion. In Cebu, the Mayor controls a number of such funds, with little accountability to the elected legislature. The largest of these relates to the revenue share from gambling (PAGCOR), the accounts for which show all manner of uses, from water pumps and dental supplies to training and contributions for celebrations. Other extra-budgetary funds include a Special Education Fund and the unspent balances from previous years budgets. All these offer scope for mayoral patronage. Of course, it is not only the executive that seeks to manipulate decisions during budget implementation. Councillors, civil society organizations and individual citizens continue to lobby for particular projects during the course of the year. Since the budget may be regarded by those responsible as little more than a general guide, or even just a wish list, there is nothing to stop councillors or others from lobbying at any stage. The mechanisms noted above provide the executive with considerable scope to respond to such lobbying. The result is that effective decisions about resource use differ markedly from those approved during the formal budget process.10 All this is not to say that control by the executive is necessarily bad for the poor. The poor, and organizations representing the poor, may be able to take advantage of executive discretion, by virtue of their persistence in pressing their demands. There are also situations where executive discretion may produce a more rational use of resources for the city as a whole than responding to numerous demands from local communities or councillors for small, unrelated projects. For example, investment in bulk water supply capacity may be required before local water taps can deliver any water. But the opportunities the executive has to manipulate financial resources concentrates power, reinforces clientelistic relationships and undermines the democratic link between poor communities and formal decision-making through local councillors.
there may be a large gap between perceptions and reality. This may occur where the formal budget process in which civil society has engaged is overtaken by other, unofficial processes, and where democratically agreed and published budgets are not implemented. As a result, while organizations of the poor may appear to make gains through their engagement with city government, what is eventually delivered may be much less than what had been agreed. Fiscal crises and resource shortfalls are often the reason for non-delivery. These crises may, under certain circumstances, provoke far-reaching reforms to improve the fiscal situation, albeit at a cost to some, as in Ahmedabad and Johannesburg. More often, they reinforce the informal processes of decisionmaking and the position of those with power over resources, as in Kumasi and Mombasa. In the end, much depends on the attitude and capability of the local leadership mayor, senior elected representatives and senior officials, and how this cascades down to junior officials charged with implementing revenue and expenditure policies. This aspect of civic leadership is one to which we will return at the end of the chapter.
context that is capable of responding to that pressure. In a number of cities, a more pro-poor agenda emerged as a result of a national struggle for democracy that galvanized civil society. In South Africa, the commitment of the postapartheid government to poverty reduction was apparent in both the discourse and the range of anti-poverty initiatives being pursued throughout government, national and local. Local governments were required to identify Land Development Objectives (LDOs) and prepare Integrated Development Plans (IDPs), in consultation with local citizens, in order to address the inequalities of the past. But here, as elsewhere, there are tensions between the demands of those in poverty and the need to promote economic growth within a market economy. In Johannesburg, this was exemplified by the competing claims on the one hand for a radical redistribution of resources and, on the other hand, for economic modernization and high-tech growth. Following the citys fiscal crisis in 1997, the adoption of iGoli 2002 as the development strategy for the city was seen by many as the triumph of the neoliberal agenda, but this is to underestimate the extent to which the citys agenda had already shifted in favour of those disadvantaged by the previous system. Cebu, too, faced tension between promoting economic growth and more specific policies to address the needs of the poor. Economic growth from the mid-1980s to the mid-1990s succeeded in raising living standards for the majority, but made little impact on poverty. NGOs played a key role in keeping poverty on the national and local agenda. Local governments throughout the Philippines were made responsible for formulating and implementing the national anti-poverty agenda in their jurisdictions. Cebu City adopted the Urban Basic Services Programme (UBSP) as the centrepiece of its poverty alleviation strategy, and was the first city government in the Philippines to establish a specific department concerned with urban poverty. Commitment to poverty reduction has been a major issue in every recent mayoral election. But as in any city, there are competing centres of power and wealth which resist significant redistribution, and vested interests which promote major projects that disadvantage the poor. In Santiago, rapid economic growth combined with national systems of targeted welfare payments enabled the worst problems of absolute poverty to be addressed. The discourse has now shifted to issues of inequality and exclusion, over which there is much less political agreement. In Brazil, (re-) democratization and decentralization, combined with the increased resources available to the municipal governments, enabled Recife to embark on a number of initiatives to address the needs of the poor, including recognition of the rights of favela dwellers, the integration of squatter settlements into the city and the adoption of participatory budgeting. Other cities have yet to take the pro-poor agenda seriously. In India there is much pro-poor rhetoric, as well as a number of legal obligations to the poor such as the special treatment for Scheduled Castes and Tribes. But these do not seem to have had much impact on the practices of municipal governments. In Visakhapatnam, more or less the only poverty-focused initiative was the externally-driven slum-upgrading project funded by DFID (Department for International Development). In Bangalore, the principle pro-poor programme,
UBSP, was funded mainly from the centre, while a number of municipal initiatives have been decidedly anti-poor. In Mombasa and Kumasi, poverty reduction has barely made it into political rhetoric, never mind into any effective programmes of action.
very conscious that their lack of a specific local mandate limits their ability to influence decisions, and that the large projects are outside the scope of PB. Many regard themselves as scapegoats, forced to take the blame for local governments poor performance or failure to implement agreed projects. Research shows that residents regard the representative political system as corrupt and clientelistic, and that their awareness of the PB process is limited. The system of PB is less well entrenched in Recife than in some other Brazilian cities such as Porto Alegre and Belo Horizonte. This is partly because, despite a history of left-wing activism in Recife, the citys contemporary politics is fragmented, bringing frequent threats to the process. In Belo Horizonte, PB has been incorporated into local law, giving it a degree of protection, and elected councillors are formally involved in the process, unlike in Recife. Despite its limitations, the PB process has made budgetary choices more transparent, especially since the executive is required to publish information both about the budgetary options and about implementation. Above all, community level participants have gained new, more institutionalized opportunities to participate in rulebased rather than clientelistic decision-making.
Source: Melo et al, 2001 and Souza, 2001
It cannot, of course, be assumed that greater citizen participation and an active civil society always benefit the poor. The most articulate voices, with the best connections to city government, are usually those of the better off. Business interests are generally well organized, and partnerships with the city government to promote economic growth may adversely affect the poor. NGOs claiming to serve the poor may instead promote their own, lite interests and create dependency. In Johannesburg, the most powerful CBO is the Sandton Ratepayers Association, representing the residents of the richest district of the city, whose rates boycott had a crippling effect on the citys finances.
Mechanisms of accountability
Responsive city government also requires accountability to citizens for the decisions made. Yet mechanisms of accountability are seriously weak in virtually all the case cities. In most systems, meetings of the full council or legislature are open to the public, but in reality decisions are taken beforehand in closed sessions. There may be forums for consultation with stakeholder groups but these may have more to do with public relations than real accountability. In Colombo, the mayor holds Division days and public day interviews with electors and tax-payers, as well as dialogues with professional groups. In Cebu, the mayor holds meetings with various organized groups, but such meetings tend to reinforce patronclient relationships rather than achieving real accountability. More significant are the independent studies carried out by NGOs on the performance of the municipal government, such as the efforts by the federations of civil society organizations in Cebu to monitor the performance of the mayor against a poverty-reduction agenda. In Recife, one of the most significant achievements of PB has been to bring budgetary choices into the open and to oblige the executive to report on implementation.
Information is typically seen as something to be withheld rather than distributed. At the most basic level, plans, budgets and audited accounts should be public documents which are easily accessible. This is not generally the case in practice. Nor do accounts, where they are available, necessarily give an accurate picture of the citys financial position.11 In Mombasa, accounts have not been produced, let alone audited, for some years. Cebu does publish monthly revenue and expenditure figures in the press, but these do not tell the complete picture (omitting, for example, the extra-budgetary funds). The format of the budgets and accounts may also be extremely difficult for ordinary people, and even elected representatives to understand or engage with. It is often difficult to see from these how resources are actually being used, or to identify the extent of subsidy for a particular activity or client group.
necessarily be more responsive: indeed it may still be dominated by local lites who maintain a feudal control over their locality. But decision-making at such a level is more accessible to the poor, and there may be greater scope for the poor to organize collectively than at the metropolitan level. Nor is it to say that the metropolitan level is not required: there are many matters which need to be decided at that level, for example about planning, transportation and bulk infrastructure. But the metropolitan level needs to be supplemented by something much more local. Yet few of the case study cities have any elected level below the city, and for those that do, the arrangement is mostly unsatisfactory. In Colombo and the three Indian cities (serving between 1 and 6 million people), there is nothing apart from administrative subdivisions below the municipal level, although Ahmedabad has started to decentralize its administration and budget allocations to the zonal level and even to wards (Dutta with Batley, 1999, p80). For a while, the Community Development Councils in Colombo operated somewhat like a lower tier of local government in the parts of the city occupied by the poor. But in the absence of any statutory basis, their role greatly diminished with a change of government. While the metropolitan area of Santiago is divided into 34 municipalities, these are still very large, with an average population of 150,000, and their resource base is very unequal. Similarly, the metropolitan area of Recife is divided into 14 municipalities (of which Recife municipality is the largest), each with varying financial capacities. In Johannesburg, the abolition of the four metropolitan local councils (MLCs) means that there is now a single metropolitan authority for nearly 4 million people. In Ghana, the 1988 local government law established an elaborate structure of local governance below the metropolitan level. In Kumasi, there are officially 4 sub-metropolitan councils, 24 town councils and 1020 unit committees. This would mean one unit committee for every 500 or so residents truly a locallevel institution. Unfortunately, most of this exists on paper only. The sub-metros operate but are deprived of the resources they are entitled to and so do virtually nothing. The town councils have never been inaugurated, and only a few unit committees function at all. This is largely the result of the opposition of the previous metropolitan chief executive and members of the Metropolitan Assembly to any alternative locus of power in the city. Of all the case cities, only Cebu has an effective system of sub-city local governance, called barangays. These are small scale and hence accessible, having a statutory basis (unlike CDCs in Colombo), and receive an equitable distribution of resources (unlike Johannesburgs MLCs). As a result, they do enable local citizens, including the poor, to have an influence over at least some spending in their locality. It could be argued that such an arrangement offers a more secure institutional framework for local-level decision-making about resource use than does Recifes version of participatory budgeting, which ultimately depends on the political commitment of the mayor.
than is sometimes claimed, certainly in terms of negotiating with city government. Finally, and perhaps most importantly, there is the role of civic leadership. From the case study cities, it is clear that civic leadership can make a difference to city governance and to the outcomes for the poor. City governments may be highly constrained, but there is always some room for manoeuvre. How cities make use of that depends to a large extent on the qualities of the civic leadership. In most systems, civic leadership is focused on the mayor, but it also includes other senior elected representatives and senior paid officials. It may also include some outside the formal structures of city government whose voice is influential. Clearly, some civic leaders are more responsive to the poor and have more vision, more integrity, more dynamism, and a greater capacity to make the system deliver than do others. Recent mayors of Colombo, Cebu and Recife, although open to criticism on various counts, have all been able to achieve results that have been of benefit to the poor. Other mayors have been more concerned with maintaining their hold on power and lining their pockets than with the interests of the poor, or have been too ineffectual to achieve anything of lasting value. Moving beyond pro-poor rhetoric into policy and practice depends in large part on the commitment and dynamism of the mayor and other civic leaders. It also depends on the ability of the leadership to build and sustain supporting coalitions around such issues. This is a challenge where political allegiances are fluid and where, as in much of Latin America, there is a tradition of abandoning all previous initiatives whenever a new administration comes into power. In this context, the role of externally appointed leadership can be significant. In Ahmedabad, it was a particularly dynamic municipal commissioner an appointed not an elected official who is credited with turning around the fortunes of that city. By contrast, the appointed metropolitan chief executive in Kumasi dynamic but ill-motivated seems to have done much to discredit that organization in the eyes of its citizens. While individual leadership qualities are important, they must be exercised within an adequate framework of public accountability, to avoid the emergence of personal fiefdoms. An institutional framework is required which allows dynamic, responsible and responsive leadership to emerge, and which obliges leaders to deliver results for the poor, not just as favours but as a matter of routine. While such a framework seems to be lacking in most of the case study cities, there are signs of movement in the right direction in a number of places. In the end, though, what matters for the poor is what happens in practice: how policies and official processes are translated into actual practices and outcomes. This chapter has highlighted the distinction between the rhetoric and the formal systems on the one hand, and the informal processes and decisionmaking that determine what actually happens, on the other. Even apparently benign policies towards informal sector businesses, land development and service provision may be degraded in the process of execution, or have unintended negative consequences for poor people. These are issues to which we will return in later chapters of this book.
Notes
1 In 1999, Mombasa Municipal Council, already overstaffed and financially insolvent, took on 2000 additional junior staff through political patronage. 2 The Kumasi study records how the Chief Executive made key decisions, including dictating the budget, in the car park of the Metropolitan Assembly offices or at his home (King et al, 2001, p43). 3 There are, of course, circumstances in which some of the poor may benefit, at least temporarily, from certain forms of corruption: illegal electricity and water connections and non-enforcement of trading regulations, for example. But the impact of such arrangements is at best uncertain and uneven, and at worst imposes greater costs on the poor and increases their vulnerability. 4 It was the attempt by Mayor Balala to resist one such land-grab that led to him being ousted as Mayor of Mombasa (Rakodi et al, 2000, p161). 5 An exception is Ahmedabad, which seems to have remarkably high revenues. These come mainly from octroi, a local tax on goods entering the city. This tax is noted for its buoyant revenues but also for its negative impact on the economy. 6 This is somewhat surprising since in many other cities in the South, restrictions on transfers limit local discretion considerably. 7 Exempting low value rented properties may, in practice, benefit the owner (through the higher rents that can be charged) more than the (poor) tenant. 8 It is estimated that, for a family of five each using the toilet once a day, the cost would absorb 10 per cent of the basic wage (Devas and Korboe, 2000, p130). 9 Johannesburg, in common with other local governments in South Africa, has now introduced a progressive tariff structure for water which provides the first 6000 litres per month free of charge. It remains to be seen whether this arrangement can be self-financing without external subsidy. 10 This supports Grindles contention that, in developing countries, policy making occurs mainly at the stage of implementation (Grindle, 1980; Grindle and Thomas, 1990). 11 For example, an analysis of budget implementation in Bangalore for 1997/98 suggested that, in order to show a positive balance at the end of year (as required by law), figures had been adjusted even though there was actually no surplus. More seriously, it appeared that capital expenditures had fallen short of capital receipts in other words, the Corporation appeared to have been using loan funds to cover revenue shortfalls. In Mombasa, deficits are routinely disguised by postponing payments, leading to serious problems of indebtedness and inter-agency arrears.
Chapter 7
Civil Society Organizations: Do They Make a Difference to Urban Poverty?
Diana Mitlin
Introduction
Civic action, the coordinated activities of urban dwellers, has been recognized as an important feature in urban development by a number of urban theorists (Castells, 1983; Douglass and Friedmann, 1998). The significance of collective (rather than individual) action was popularized by Putmans (1993) study of Italy and his analysis of the role of social capital within economic development. For urban theorists, the ongoing role of social movements in influencing the governance of cities has long been recognized, as has the significant role of socially motivated agents such as non-governmental organizations (NGOs). More specifically, many have noted the significance of civil society in reducing poverty (see for example Gorman, 1984; Carroll, 1992). Civil society groups have long been considered to be a means to poverty reduction through
lobbying for improved policies, pressurizing to ensure policies and practices reflect laws and regulations, and through offering direct support for those in need (see, for example, Hirschman, 1984; Turner, 1988). At the same time, civil society groups have been recognized as an end in themselves, providing a political voice and supporting a collective identity cultural, ethnic, class or otherwise (see for example, Bebbingtons (2002, p6) description of the impact of ethnic organizations on the composition of local municipal councils in Ecuador). But what does civil society offer to the urban poor and, particularly, to their efforts to reduce poverty and achieve development? This chapter examines what we have learnt about the role of civil society in regard to poverty reduction and, more specifically, the contribution of grassroots organizations (GROs) and NGOs to addressing the development needs of the urban poor in the ten selected cities. It is an opportunity to look at some critical questions facing civil society in the 21st century. First and foremost, what are the ways in which civil society contributes to poverty reduction? Second, does civil society unambiguously contribute to reducing poverty and inequality? Do all groups within the urban poor benefit equally from the actions of civil society groups? Is there exclusion within civil society and, if so, what are the social processes involved: co-option, oligarchy, litism or market? Third, what are the trends within civil society, its social processes and institutional relationships? What can we understand about how its role might develop in the short to medium term? Finally, how are the actions of civil society influencing the shape and nature of urban development? This study offers an opportunity to extend our understanding of civil society at the level of the city to understand how citizens and their social organizations (be they formal or informal) are shaping and changing the nature of urban opportunities and challenges. It may be helpful, at the outset, to define the terms as they will be used here. Civil society is used as an all-embracing term for voluntary associations between the state and individual citizens and their families (Salamon and Anheier, 1992).1 As such, the definition includes non-government organizations, non-profit associations, informal organizations addressing public interest issues, and self-help groups and associations. The discussion here concentrates on two categories of organizations within this broad umbrella. First, grassroots organizations (GROs) are membership organizations that are (or are considered to be) independent of the state.2 As membership organizations, the risks, costs and benefits are shared among the members, and the leadership may be called to account by members. Most are non-profit, although some operate as cooperative commercial enterprises. Many are informal and operate as loose networks of social relations rather than as regulated organizations. Trade associations of informal and self-employed workers and residents associations are the two most important groups of grassroots organizations in the discussion that follows. Second, non-governmental organizations (NGOs) are defined as professional, non-profit intermediary organizations (often without a membership) that are independent of the state and which undertake a range of activities in order to further developmental objectives. In general, they are also charitable institutions, belonging to a larger family of organizations providing welfare and addressing public interest issues for a wide range of beneficiaries.
The significance of civil societys contribution to development practice and policy has been recognized by authors such as Gorman (1984) and Hirschman (1984). The popularity of civil society organizations (CSOs) grew during the 1980s and 1990s with an increasing number of bilateral agencies creating and extending co-financing programmes to NGOs. Multilateral agencies such as the World Bank and the United Nations Development Programme (UNDP) established consultation arrangements in addition to direct funding and programme collaboration. The major emphasis of development agencies has been on the contribution of NGOs and GROs rather than other civil society organizations. Growing support for civil society from official development assistance agencies continued in the 1990s. However, there was also a more detailed consideration of the potential role of NGOs (and self-help organizations) and of their emerging development experiences, both in traditional small-scale projects and in the more ambitious scaled-up roles that were now being required of them. Edwards and Hulme (1992, 1995) and Hulme and Edwards (1997) reviewed some aspects of programme experiences in their edited volumes and highlighted issues of scale, accountability and relations with the state. At a similar time, Carroll (1992) suggested that the commitment of NGOs and GROs to participation and accountability could not be taken for granted. The next section briefly summarizes some of the issues that emerge in the literature considering the role of NGOs and GROs in urban poverty reduction and urban development. This is followed by a description of the extent of the NGO and grassroots sectors found in the case study cities, from which we identify the contribution of these agencies to urban poverty reduction. Broadly speaking, their contribution was disappointing. Finally, we analyse some reasons for this poor performance, with a focus on those internal to the sector, such as participation, accountability, representativeness, responsiveness and engagement with city government.
community leaders to secure financial and political advantage for both parties. The ability of politicians and/or officials to control access to land tenure, infrastructure and services is used to reinforce the dependency of the urban poor on more powerful structures and processes. Women may have little influence on the grassroots organizations that may or may not address their needs.3 The literature on NGOs is comprehensive and has been growing substantially during the last two decades. One of the most widely used distinctions is that of Korten (1987, pp147149), who distinguishes between relief and welfare, small-scale, self-reliant development and sustainable systems development (in which they extend activities from their own organization into relationships with public and private agencies including advocacy). Carroll (1992, p10) makes a five-fold distinction between NGO purposes: charity, relief, development (subdivided into business development, social development and social business), political action and advocacy of special interests (which may be combined with other assistance). Fowler (1997, p32) summarizes a growing concern throughout the 1990s with a table of NGO pretenders; these include NGOs set up to have a commercial advantage, those for illegal purposes, those created by official donors, those created by governments and those created by politicians. Perhaps more significantly, Fowler (1997, p31) emphasizes that NGOs generally have to mix sustained poverty alleviation and social justice while functioning as businesses to provide public services. While a relatively small proportion of this literature focuses on the work of NGOs with the urban poor, there is sufficient to develop an understanding of the range and scale of NGO activities. Housing improvements are encouraged through a number of strategies including loan finance, technical assistance and community mobilization. Although many micro-finance initiatives have been associated with rural development (most notably the Grameen Bank in Bangladesh), others are located in urban areas (Hurley, 1990). Some emphasis has also been placed on NGOs role in policy-making and advocacy. Perhaps critically, and as exemplified by Karaos et al (1995) in the context of the Philippines, policy success can be difficult to achieve and once achieved may not be easily translated into changes in practice. INTRAC recently reported on a study of 141 NGOs working on urban issues that offers an understanding of the balance of these activities in five southern cities. One-third of the NGOs in their sample have a primary focus on income generation with the second and third categories being education and health respectively (Sahley and Pratt, 2003, p39). Most NGOs were working in four to six sectors. While 61 per cent of NGOs noted they were engaged in service delivery, the agencies put more stress on capacity-building activities (such as training and awareness raising). Drawing on this literature, many NGOs would fall within the developmental category, with some addressing issues of relief and charity as well. However, there is also evidence of alternative approaches and some NGOs place considerable emphasis on partnership with organizations of the urban poor (see, for example, the approach of members of the international network Shack/Slum Dwellers International, Environment and Urbanization, 2001). One particular issue is that of relationships within civil society and specifically between grassroots organizations and the NGOs that are seeking to
support them. A number of earlier studies suggest that their relations are neither simple nor uncontentious (Desai, 1995; Mitlin, 1999). Tensions exist between NGOs that are seeking to control the development process and the urban poor who want greater power over decision-making and activities in development programmes and projects. This is just one of the themes that we will consider below.
mothers clubs respectively. However, grander ambitions are also noted. The creation of mothers clubs in low-income neighbourhoods in Recife is linked to a desire in the city administration to influence (and possibly control) community politics.
in Ahmedabad is one of the few examples of larger scale self-help activities. In this case, micro-enterprise support has been extended to include a more comprehensive programme of upgrading in selected low-income settlements (Biswas, 2003). This experience highlights some of the challenges involved for GROs and support agencies in providing self-help. While they can provide moral support and small assistance to the urban poor, more substantive support to provide essential services and/or to increase informal entrepreneurship requires collaboration with state agencies especially local authorities.
in non-transparent ways. Another is vote banks which can be used to secure investments for poor wards. While some vote banks can rightly be criticized for being simple party-based affairs centred on liquor distribution with the poor groups being bussed to party rallies, others can also be used by poor groups to secure genuine gains from candidates in local elections. Vote bank politics can facilitate access to land and services, protect poor groups from demolition by richer ones, resolve local disputes over property boundaries, and help ensure the bureaucracy is responsive to local needs. Councillors (and their competitors from other political parties) have party workers operating as their representatives. These party workers are generally local community leaders who link to street leaders, more formal office bearers of residential associations (the community) and the party political representatives. Street leaders may approach several party workers to test out their responsiveness to their demands, and may approach the councillors directly (Benjamin and Bhuvaneswari, 2001). There is active and rigorous competition as to who can offer what to the organized urban poor. Ethnicity, income and occupation create and fracture groups and alliances. Often such links work vertically as well as horizontally, drawing the poor together with more powerful groups. However, the processes explored in the study of Bangalore are limited in the extent to which they are really pro-poor. In a very different context, mixed benefits also emerge from a donor (United Nations Childrens Fund, UNICEF) project to improve health services in Mombasa through community pharmacies run by village health committees. The most successful example was the Mtongwe Bamako project, which secured real benefits for those living in that low-income settlement, but in other areas the leaders have used the health facilities for their own private benefit. Thus, the organizational capacity of a community and its internal politics influence the extent to which projects benefit all residents and are sustainable.
rather, residents associations competing for inclusion in the state poverty programme, FOSIS, with little apparent interest in collaborating to challenge the shortcomings of the programme. A particular explanation emerges in the case of organizations established by the state. In Colombo and Santiago, it is suggested that such GROs tend to present projects to the state and implement state programmes, rather than taking on a broader and autonomous agenda. In Kumasi, low-income groups have little capacity to influence decisions of the Assembly. The powerlessness of the poor is reflected in their own perception of their difficulties. As one woman in the low-income community of Anloga argued the KMA [Kumasi Metropolitan Assembly] does not care so much for us because we are poor and we are not Ashanti (King et al, 2001). Their lack of political connections and lack of skills in English are further reasons put forward by the urban poor in Kumasi to explain their exclusion from the political process.
organizations became involved in collaborative projects with city officials, local residents developed their own technical capacity, or required assistance in areas in which the NGOs did not have skills. In Cebu City, NGOs enjoy a particularly good working relationship with local government and state agencies; this involves both service delivery and advocacy. As discussed in Box 7.3, there has been considerable collaboration between CSOs in the city with emerging benefits for the poor. The success of NGOs in influencing government practices has been through a more practical engagement in social issues of common concern, rather than at higher-level municipal policy-making. In part, this opportunity for collaborative programmes may be due to a high level of institutionalized participation by civil society in local government within the Philippines. Some of the contradictions in the assessments (including the diverse perspectives on Kumasi) can be resolved through the understanding that NGOs may have successful programmes at the micro-level in specific areas that are not seen as controversial. However, larger-scale success appears to be harder to achieve. It also appears that a two-stage assessment occurs with regards to advocacy and lobbying. When NGOs secure inclusion and are invited to meetings, there is often a sense of success and achievement. However, in the longer term there is scepticism about the value of such inclusion.
Conclusion on the contribution of GROs and NGOs in the case study cities
Generally speaking, the discussion in this section suggests we should be pessimistic about the contribution of GROs to addressing the development needs of their members. While many GROs seek to address these development needs, and some have achieved success, they are often immersed in relationships of dependency, clientelism and patronage. Gains have been secured, particularly through the process of elections and the securing of votes, but there remain questions as to the effectiveness of civil society contribution. In regard to NGOs, serious questions are raised about their capacity and sometimes about their commitment to achieve significant poverty reduction. The following section examines some reasons for this low level of success of civil society in addressing urban poverty. Three particular issues will be explored: GROs that are not really inclusive of the poor; limitations of NGOs; and civil societys lack of participation in citywide political structures and debates.
repeated concerns about low levels of participation across the studies. While low levels of participation might not necessarily be problematic, in situations of widespread poverty a lack of participation suggests that residents do not see such organizations as contributing to their needs. In two cities in which grassroots action has been recognized to be successful, Cebu and Bangalore, a related issue arises that, even when communities and groups have managed to organize and unite for specific objectives, they find it hard to maintain that unity once their goals have been secured. Over time, participation slips away. This suggests that even when participation in GROs is recognized to be a successful strategy, other activities are also important. There are also concerns about how representative leaders are of the interests and characteristics of members. Such a lack of representativeness combined with limited accountability may help to explain the significance of the personal agendas of community leaders. In Cebu, Kumasi and Recife the picture emerging from the studies is that there are commonly low levels of participation by residents and a reluctance of some potential members to be involved. Although there are 60 registered associations at Kumasi Central Market, less than 20 are active; The records are poorly kept and some members exist only in name. They rarely attend meetings or pay membership dues (King et al, 2001, p143). It appears to be relatively easy for organizations to be formed but when problems arise, they stagnate and may die. Such a process is illustrated in the case of one organization for drivers of trisikads (cycle taxis) in the port of Cebu. The committee is divided and no activities are happening following an internal conflict, with members disputing the financial transactions of the treasurer. Etemadi (2001, p69) concludes unless and until the internal conflict is resolved the organization would remain divided. Similar problems are noted in the provision of water in the city. There is a possibility for communities without water to have subsidized communal taps. However, take-up is low with less than 10 per cent of urban poor families using water from such taps. Etemadi (2001, p118) suggests that this is because of a lack of active participation by members, undemocratic if not oppressive management style, irregular or no annual election resulting in a monopoly of leadership and lack of financial transparency and accountability. As a consequence, water vendors are widely used despite prices that are three times those of the municipal water company. A number of reasons are put forward to account for low levels of participation. In Cebu, it is suggested that the poor are so preoccupied by their immediate needs that further activities are not possible. Another explanation is offered in Santiago, a city in which significant removals took place during the 1970s and 1980s. The poor were relocated into peripheral areas with considerable disruption to existing social networks and solidarity. In the context of state-sponsored GROs in Bangalore and Colombo, membership is thought to be short-lived due to a lack of ownership on the part of local residents. However, Mtongwe Bamako, the UNICEF project in Mombasa, while limited in some areas, did succeed in others and continued after UNICEFs withdrawal because of the strength of local support. One of the implications of low participation is that even when there are elections within GROs (rather than less explicit processes), leaders may emerge
with a relatively low degree of member involvement. In Recife, for example, delegates for the participatory budgeting process may be elected with as few as 25 votes due to an apparent lack of interest in the elections. As a result, leadership may be unable to represent the needs of members or be uninterested in doing so. In Kumasi, persuading individuals to take on leadership is itself problematic as, unlike the situation in rural areas, there are no traditional mechanisms to allocate such roles. The young people living in the low-income community of Mossi Zongo identified the following desired leadership qualities: educational background; maturity; God-fearing; permanent residence in the community; eloquence; and social standing. This suggests that it is unlikely that one of the poorest members would be selected as a leader. This concern is reinforced by findings from Johannesburg and Visakhapatnam in which low-income and lowstatus residents were the least likely to take part in leadership roles. With respect to gender, there appears to be relatively low participation by women in leadership positions. In Kumasi, for example, leadership structures in the community are male-dominated and there is little real consultation of the ordinary membership, especially the women. This finding is consistent with earlier literature which also reports that a male lite within the settlement often dominates leadership positions (Harrison and McVey, 1997; Hardoy et al, 1991; Pornchokchai, 1992; Thorbek, 1991; van der Linden, 1997). Some of the consequences of a low degree of representativeness are indicated in the studies. In Mombasa, Chiefs and elders are used to exercising power by filtering information and directing communities, and residents are used to them dominating community decision-making (Gatabaki-Kamau et al, 1999, p89). As has been found elsewhere, the emerging picture is of poor accountability with low membership participation, little membership involvement in decision-making and no process to enable a common agenda to be set (Klaarhamer, 1989; Magutu, 1997; Moser and McIlwaine, 1997; Moser and Holland, 1997; Peattie, 1990). The case study of the Diepsloot community in Johannesburg (Box 7.2) illustrates how the poorest and most vulnerable may be disadvantaged when interests compete within GROs. In this case, existing residents opposed squatters joining their community. The most significant GRO in Diepsloot is the Community Development Forum (CDF), an umbrella organization that assumes sole responsibility for liaising with outside actors in the development processes unleashed by the findings of the Land Development Objective (LDO) (Beall et al, 2001, p68). The Forums contribution to local development is widely acknowledged and it has clearly been successful in addressing some development needs within the community. However, its hegemony has been challenged and there has been violent conflict especially in relation to the housing allocation process. Furthermore, while valuable development work has been done by the Forum, three groups do not appear to have their interests well served: women, who are more concerned with the risk of rape and flooding and with the problems of securing a livelihood than with the concerns of the CDF; children, who have no play or recreational facilities and who may be having to work; and illegal immigrants, who are being prevented from squatting in the river bed (adjacent to the site) by CDF members.
Diepsloot. As a consequence of the CDFs agreement to limit the growth of the settlement, it is now actively preventing new shack development, thus closing off an important reception area for poor in-migrants. All concerned with planning and service delivery in Diepsloot regard the CDF as the legitimate voice of the residents and the key to meaningful participatory democracy. Gradually, infrastructure has been installed in the settlement and houses have been consolidated. However, the Forums ability to engage ordinary citizens in influencing the development process has been limited by both the wider context and internal political contests. It has not succeeded in changing the priorities of service-providing agencies where the latter have professional blind spots regarding the priorities of low-income residents. In addition, acceptance of its leadership is not universal, it neglects some concerns of residents in its responses to consultative processes, and it does not represent dissenting voices or new residents. Monopoly of community organization and consultative processes by a single channel is, therefore, neither stable nor desirable.
Similar issues of contested interests arise in the other two neighbourhood studies in Johannesburg. One of the emerging explanations is a leadership which may be democratically elected but which may not be representative. For example, in Meadowlands (another low-income area in Johannesburg), the African National Congress (ANC) is led by the Veterans League which for the most part involves the home owners, even though nearly a quarter of the residents are tenants. Hence community organization, to the extent that it engages with local government, is firmly the terrain of the more established, respectable older working class (Beall et al, 2001, p113).6 Even without the active intervention of the state creating conditions for dependency, GROs face many difficulties. Box 7.3 discusses trade associations of informal vendors in Cebu and describes issues related to the representative nature of leaders, the interests of the poorest groups and the capacities of the organizations themselves. Also considered is the continuing need to negotiate with the relevant authorities to increase the opportunities and reduce the oppression facing the poor. While successes have been achieved by the traders association, generally the more politically successful organizations are those that represent the higher-income traders.
on the part of the local authority. The role of NGOs in advocacy is particularly difficult in Kumasi, in part because the Ghanaian culture is one in which criticism is taken to mean opposition. A more complex picture emerges in the three cities in countries where radical political changes have led to recognition being given to the potential role of civil society. In Santiago, Johannesburg and Cebu, there are concerns about NGOs over-involvement with service delivery roles. This debate suggests that NGOs have taken up opportunities to engage with state programmes. In part, this is understandable, as many of these NGOs have been actively involved in the struggle for democracy, and in each of these countries staff have moved from the NGO to the state sector following democratization. As new state poverty reduction programmes are created, NGOs are encouraged to participate; in some cases, they may have been responsible for initiating the programme in the first place. However, there are indications that the absorption of NGOs into the state is reducing the ability of NGOs to remain a critical commentator on state practice and policy. Turning to governance within the civil society sector itself, there is a concern about relationships between NGOs and GROs. More specifically, there are fears that NGO staff may be reluctant to delegate power and responsibility to local residents. In Colombo, for example, there are evident tensions between NGOs and GROs. One reason for the poor participation of low-income residents in NGO programmes is that such programmes do not respond to the needs of the urban poor: NGOs have their own objectives and are largely self-interested ventures obtaining funds from government and international donors for their own benefit (Russell, 1999, p104). NGOs in Colombo are also accused by community leaders of deliberately failing to develop the capacity of CDCs and other GROs in order to enable themselves to dominate local processes, and weakening these Councils by draining human resources away from the community into their own organizations. A further concern is that they dominate financial and project-level decision-making. In other cities, NGOs are also treated with suspicion by the poor. King et al (2001, p57) reinforce concerns that beneficiaries do not have a sense of ownership of the projects and therefore take no responsibility for their sustainability. It is argued that NGOs have done little to support self-reliant interventions for communities, partly because they are more concerned with their own survival. This situation is also repeated in Bangalore. Benjamin and Bhuvaneswari (2001, p17) suggest that NGOs participate in the formalization (and techno-managerial zeal) of urban development processes with adverse consequences for the poor. For example, one NGO scheme to create hawking zones had a disastrous effect on the income of hawkers (as discussed in Chapter 8). If one were conventional, we would argue that the NGOs have been coopted by the system. A more radical perspective could hypothesize that the lite NGO groups are now one of the main players in shaping the system to be increasingly repressive to poor groups (ibid, 2001, p194). As a result, It is hardly surprising in the above context that local communities are usually suspicious and sceptical of NGOs and official programmes aimed to improve their lot
(ibid, p218). Such worries echo some of the reservations previously expressed by Thurman (1994), Desai (1995) and Gazzoli (1996). There are fewer concerns in Cebu, where NGOs claim high levels of involvement by local beneficiaries in their projects. The relationships are positively assessed by Etemadi (2001, p92): Their direct relationship is founded on trust and support. However, the success has brought challenges and NGOs face a growing pressure to respond to a deeper relationship with GROs. Demands for financial transparency, while indicative of a degree of empowerment of the poor, are considered to be adding to the pressures on NGOs as they seek to manage funds in ways that they consider to be equitable and effective.
Delegates do not debate structural projects Participatory budgeting was restricted to the discussion of containment walls, staircases in sloping areas, paving, improvements in schools and health posts and so on. In Cebu, despite the recognized strengths of civil society, citywide consultation by local authorities may be concerned with a specific area of local government responsibility (for example, hawking and vending) or with citywide planning processes. Local politicians, even the Mayor of Cebu, make clear their preference for dealing with a single representative or group of representatives in such consultative processes. Nevertheless, the situation of civil society in Cebu appears to be somewhat different from that in the other case cities. Box 7.4 describes citywide approaches that seek to move forward the agenda of the poor in that city. Why is Cebu different? There appear to be three related factors: the relatively high degree to which civil society in the Philippines has been formally included in local government process; the fact that Cebu is not a capital or primary city (and hence subject to fewer political interests and less contesting); and finally the relatively strong personal ties between the political and NGO lites.7
Despite the generally favourable account in Box 7.4, not all the problems have been resolved. As noted above, it is often hard to secure high levels of participation and many are simply reliant on authority without questioning decisions or challenging direction. In 1999, only 5 per cent of the budget in Cebu went on capital projects potentially benefiting the poor. The budgeting system encourages the clientelistic political system that pervades Cebu, like the rest of the Philippines (Devas, in Etemadi, 2001). Devas argues that while the political system is delivering resources to the poor, it is also reinforcing dependency: It also puts a premium on the ability of groups (or individuals) to lobby; clearly some are in a better position to do this than others. Each individual project has to be separately negotiated, preventing proper comparison between competing uses for the resources. Despite such reservations, the balance of power and resources appear to be substantively different in Cebu from that in the other cities discussed here.
Conclusion
Is the picture really so pessimistic? Is there any evidence that democratization and decentralization have increased the capacity of civil society to secure benefits for the poor? From the city case studies, it is clear that some civil society groups are seeking more inclusive and pro-poor urban development approaches. At times, as shown above, such approaches have secured some limited success. What is evident is that success cannot be taken for granted. It may be shortlived as more powerful forces seek to use the urban development process to their own advantage. It may also exclude some of the poorest and most vulnerable among settlements of the urban poor. The emerging picture is, for the most part, of isolated communities whose leadership is anxious to use urban development processes and structures to further their own individual interests and is able to address some immediate development needs, but which has little ambition to substantively address issues of inequity and discriminatory development. As Abers (1998) and Evans (1996) argue, the way in which the state interacts with citizens appears to influence the way in which CSOs operate and hence their capacity to represent themselves and their members. A potentially unrepresentative and often unaccountable grassroots leadership appears to reflect, at least in part, an urban development process that is itself exclusionary. The interests that are served by this process seek to create and maintain a grassroots leadership that meets their needs. The conclusion is that, for the most part, the experiences in the ten cities substantiate the existing findings that relationships between grassroots organizations and the state are characterized by clientelism and patronage. Unable to provide comprehensive services and legal land tenure to everyone, city governments control access to such benefits as a way of influencing votes and, in some circumstances, securing other resources, especially land. Some politicians and/or state officials appear to work with community leaders to secure financial and political advantage for both parties. The ability of politicians and/or officials to control access to land tenure, infrastructure and services is
used at best to influence communities in the choices that they make and at worst may reinforce the dependency of the urban poor on more powerful structures and processes. While vote bank politics have helped to address the needs of the poor and have secured real benefits in some areas, other neighbourhoods remain neglected. The discussion has suggested that three factors in particular have restricted the ability of civil society to act in the interests of the poor. First, the leaders of grassroots organizations may not represent the interests of the poorest or most vulnerable groups. Moreover, that leadership may have only limited accountability to their members. Second, NGOs may be failing to support grassroots activities effectively. In some cases, they may actually compete with GROs for power and other resources, or they may dominate decision-making, reducing the effectiveness of project interventions. Third, civil society may simply not operate at the level of the city and may therefore not be able to take a sufficiently broad perspective. Planning and economic development issues may be ignored in favour of short-term advantage. Politicians and officials may seek to isolate community groups so that they can address their own interests. Looking ahead, it is potentially significant to note that, in Cebu, despite the limited NGO influence on policy-making at the highest level, it is the institutionalization of partnership at a project level that has created space for innovation. NGOs and CBOs may maximize their effectiveness through negotiations around social policies and social provision in a number of groupings (commissions, committees and task forces) that are outside of mainstream government policy-making. This argument reinforces some of the conclusions of Boonyabancha (1998) and Khan (1997). These two authors, writing about Thailand and India respectively, suggest that NGOs and CBOs may be most successful when they are able to work alongside local government in operationally focused activities rather than lobbying from outside. Hence, in addition to the three arguments above, the strategies used by civil society are also important in explaining success. Finally, we should also note that the studies may be limited (and unduly pessimistic) because of the specific focus on city governance. In several of the countries (including Chile, the Philippines and South Africa), NGOs and GROs have sought policy changes at the level of national governments rather than the city. In each of these three countries, significantly large, national programmes to assist the poor, improve access to secure tenure, housing and basic services have been introduced.
Notes
1 2 3 The issue of definitions is considered in more detail in Mitlin (1999, pp45). Although, as explored below, this characteristic is considerably more complicated than first appears. Looking specifically at gender issues, Thorbek (1991, pp7980) summarizes the contrasting experiences of women in residents associations in Bangkok and Colombo. In one, women and men participate equally in addressing infrastructure
4 5 6 7
Chapter 8
Regulating the Informal Sector: Voice and Bad Governance
Philip Amis
Street hawkers and informal sector traders are among the strongest visual images of any urban area in the South. Their activities provide the principal source of income of many of those living in poverty in these cities. Their position is often fragile, dependent not only on the uncertainties of passing trade but also vulnerable to the exploitative behaviour of those with power to enforce oppressive regulations. This chapter is concerned with the relationship between city governance and those engaged in informal sector trading. One of the main arguments of this chapter is that it is easier for city governments through bad governance to destroy the livelihoods of the poor than it is to create, sustain or enhance them. The following sections provide brief overviews, first, of the scope and scale of the informal sector in the ten case study cities, second, of the development of the concept of the informal sector and third, of the role of municipal government in relation to this sector, together with some examples from the
case study cities. From this analysis we will develop the argument that the regulation of the informal sector is often determined at the local level and often through informal processes because of the difficulty and inappropriateness of applying the relevant legislation. In this situation, it is the extent to which those in the informal sector have voice that determines whether or not city government is constrained from applying detrimental policies. This notion of voice the influence that the poor can exert is intimately connected to the nature and practices of democracy at the local level, and is an essential constraint on bad governance.
Johannesburg
Cebu
Visakhapatnam
Bangalore
Ahmedabad
Santiago
No figures but informal sector is small. Evidence of outsourcing and increasing informalization
busy pedestrian sites in central business districts, even though these are often areas of conflict. Some parts of the informal sector seek very marginal locations river banks or adjacent to railway tracks often hazardous sites with almost no economic value. Many informal businesses, however, are located in the homes of their operators, resulting in mixed uses in primarily residential areas. The location of the informal sector is one of the key issues in relation to its regulation.
In contrast to this enterprise approach, Hart defined the informal sector in terms of whether individuals were wage earning (formal sector) or selfemployed (informal sector). He makes the important point that individuals can reside and work within both sectors and move between them (Hart, 1973). Other writers have sought to define the informal sector in terms of its relationship to the state and government. Thus Roberts (1995), following the lead of Portes et al (1989), defines the informal sector as those enterprises, which avoid state regulation, such as fiscal obligations and labour standards, in contexts where similar activities are so regulated. In practice, the informal sector is mainly constituted of small-scale enterprises and the self-employed (Roberts, 1995, p116117). Other work has sought to identify the dynamics within the informal sector. The original position of marginality in the Latin American literature was posited by Quijano in terms of the informal sector as an area which was forever deemed to be marginal and where capital accumulation was not possible (Quijano, 1974). This seemed a particularly depressing interpretation, and one that has been challenged by empirical research that has shown a high level of differentiation within the informal sector. Houses research distinguished between an intermediate sector where capital accumulation was possible and one where the basic activity merely enabled survival (House et al, 1993). Work in Kenya suggests that constraints on expansion are often more to do with the extent of the market for the output of the informal sector than a shortage of capital (House et al, 1993). The profitable sections of the informal sector are often controlled by particular groups, frequently reinforced by caste, ethnicity or other social norms that make them difficult to enter. Illegal drug dealing is the clearest case but
informal shelter, transport and some metalworking have similar characteristics. Ultimately the sanction that supports this control is violence. Veena Das work on the riots in New Delhi associated with Mrs Gandhis assassination in 1984 are revealing about the nature of the informal sector: the fencing mechanisms and regulation of entry into most sectors of the informal economy were a product of several factors caste and kinship networks, defined spheres of influence by politicians and local Big Men, and the constant threat of violence to regulate behaviour. The informal sector does not consist only of those who live on the disaster threshold. Enormous differences in wealth may be found in the resettlement colonies (Das, 1996, p188). To conclude, a working definition of the informal sector needs to contain four notions, namely: an idea of the contested relationship with the state regulation; the small size of the enterprise; the idea of a self-employed or survivalist livelihood strategy; and finally the notion of a dynamic and moving frontier between the three first notions. This is the working definition that we shall use for the rest of the chapter.
The role of city government in enhancing and protecting the assets of the poor in the informal sector
In order to consider the role of city government in relation to the informal sector, it is helpful to adopt the livelihoods framework of the assets of the poor (Moser, 1998). The four main assets are labour, human capital, social capital and productive assets (such as housing). The following suggests the ways in which actions by city governments can potentially enhance or protect these assets of the poor. Labour providing a framework which encourages economic growth generally (but the city governments scope for action may be quite limited see Chapter 3); providing and maintaining infrastructure needed for industrial growth (but responsibilities for major infrastructure may lie outside city government); using planning and regulatory powers to encourage rather than restrict the development of informal and small and medium enterprise (SME) activity.
Human capital provision of primary education; provision of health services; provision of residential infrastructure and services (water, sanitation, etc).
However, much of the responsibility, and many of the resources, for these may lie outside the city government.
Social capital setting the scene and providing the environment and funds which enable community organizations to operate, and for inclusive participation by citizens; maintaining law and order (critical for social capital); provision of street lighting, which can enhance social capital by extending the time for social interaction.
Productive assets regulation (eg planning and zoning) can influence the extent to which housing and land can be used as an asset, including for urban agriculture.
The potential contribution of city government in relation to labour relates both to promoting economic and industrial growth generally and to enabling the development of small and medium enterprises. This involves both providing an appropriate and supportive environment and ensuring that essential infrastructure is provided even where that is outside the direct responsibility of city government. Raj (1993) observes that industrial development in urban areas suffers from a lack of infrastructure. This will also be the case with SMEs even if their dependence on water, power and roads may be less than that of large industries. But the most significant impact of municipal government in relation to labour and employment comes through the application of planning controls and other regulatory activities. Inappropriate application of these can substantially destroy employment opportunities.1 Benjamin (1993) shows the negative impact of restrictive regulations and bylaws in the case of New Delhi in India. The removal of informal traders often as the result of city beautification schemes can have a similar impact. The damaging consequences of repressive regulation of the informal sector will be discussed further in the next section. Box 8.1 illustrates the way in which restrictive master planning, in this case in Bangalore, can stifle and undermine local economic activity. This is contrasted with the mixed land-use settings in non-planned areas that lead to diverse economic activity and offer far greater opportunities for the poor to participate. This box also illustrates the critical importance to the informal sector both of land as a productive asset and of location, and the impact of city government (or other agencies of the state) on these. The case of South Africa provides a contrasting example, with the relaxation of a restrictive planning and regulatory system following the end of apartheid in the early 1990s. In the early 1980s there were estimated to be only 300 hawkers functioning in Johannesburgs inner core, whereas the estimate for the mid1990s was 4000, with a total of 15,000 in the Greater Johannesburg area. While this growth is partly explained by an influx of migrants from elsewhere in Africa, the main explanation is the change in official policy from repression to one of greater tolerance (Tomlinson, 1996, pp188189).2 The impact of the changed policy was also noted in the inner city area of Yeoville: There are informal businesses springing up all over ranging from barbers, to car repairers, to vendors of various kinds. There also has been a
proliferation of spaza [informal kiosks on the street], shops and mini-businesses located randomly around Yeoville, selling sweets and foodstuff . (Beall et al, 2001, p122). In relation to social capital as an asset for the informal sector, the critical issue for city government is maintaining law and order and preventing violence and insecurity that can so easily undermine economic activities. Moser and Holland (1995) have shown the devastating effect crime, violence and insecurity have on social capital formation in Jamaica. Similar problems were noted in the Johannesburg case study: The 19901993 violence was directed against gatherings of people engaged in their day-to-day routines, such as funeral vigils, parties, shebeens and stokvels (savings clubs). As a result trains, buses and taxis came to be commonly used for peaceful social activities, until they too
became sites of political attack. (Beall et al, 2001, p101). In the Yeoville area, the negative impact of crime upon commercial activity was also noted; in particular, the withdrawal of banks and major commercial outlets from the area (Beall et al, 2001, p121).
The above aspects of regulation are concerned with the informal sector as an enterprise. In addition, there are aspects of regulation which are more likely to be of concern to municipal government. These include the following: planning and development control; building bylaws; business licensing or permits; local taxation; public health and safety.
As we shall see in the examples presented below, it is unusual for municipal governments to be concerned with all of these aspects of regulation at the same time. Legality should not to be conceptualized along a single dimension (yes or no, black or white). As we shall see, there are many situations where informal
enterprises may be illegal on one dimension (for example health and safety or social security legislation) but may be paying local tax. Given the inherent difficulties in implementing diverse legislation, municipal (and national) governments tend to be selective or perhaps arbitrary in what they implement. An important theme that emerges is that actual implementation of regulation is substantially affected by the behaviour of bureaucrats at the local level. This is consistent with the observation, frequently made, that where policies may have contradictory elements these are usually resolved by lower level officials in the process of implementation (Grindle, 1980). From the viewpoint of the informal sector, regulation can have negative consequences in terms of the ability of entrepreneurs to survive and to earn a living. In the worst cases, it can involve destruction of assets, eviction and/or resettlement. Informal activity is highly dependent upon both location and the nature of the local urban economy. Informal traders are acutely aware that relocation is harmful to them. They know that what is suggested or imposed on them by municipal government, such as relocation from pitches in the city centre to a peripheral location, is likely to spell economic disaster. The question we are concerned with is: what are the circumstances that determine these actions by city government? The city studies suggest that it is the informal sectors ability to protest (voice through democratic processes, whether formally or informally) and/or to divert regulatory activity (eg through politics by stealth) that determines whether municipal governments are able to implement repressive regulatory policies. In other words, the voice of the informal sector acts as a significant restraint on repressive actions by city government.
Zero growth
No new vendors are allowed after the master list is finalized by 9 February 2001 Each vendor family is limited to one stall/structure
Time
6.00 pm to 6.00 am in some areas
Stall/structure
Light and temporary structure Uniform colour Size of table 2 3 or 3 2 or 3 6m, depending upon area Edges of tables should not extend beyond allowable distance No new structures or extension
Cleanliness/sanitation
Keep surroundings clean, arrange structures orderly Dispose of garbage properly Stalls/structures not to be used as sleeping quarters
Fees
Arcabala: Peso 215 (average P5) (Arcabala is a fee, which represents a combination of entrance fee and space rent; the amount is legislated by the City Council; an increase in the fee is done through a public hearing, in which the vendors bargain for the amount.)
Penalties
P500 for violating anti-loitering ordinance; P300350 for redemption of confiscated goods.
The next episode took place after the election of the mayor following the 1997 election. The newly elected mayor, Najib Balala, came to office with an ambitious programme of improving municipal governance, reducing corruption and beautifying the city. The last of these involved denying licences to curio traders and demolishing kiosks in the city centre, on the basis that tourists who visit Mombasa do not like being hassled by street traders. Balala, who was also chairman of the Coast Tourist Association, was more concerned to protect the interests of large tourist operators and formal retailers than the interests of the street traders (Rakodi et al, 2000, pp161162). Kumasi, which has a very high proportion of local employment in the informal sector, has a similar story. The number of informal sector traders had grown, partly as a consequence of structural adjustment and retrenchment, and far exceeded the number which could be accommodated within the citys central market. The Metropolitan Chief Executive (MCE) (at the time of our study) adopted a ruthless policy toward unofficial traders in the central business district and regularly led raids to destroy such sites. As a result, traders, who are amongst the citys poorest, regularly lost both stock and premises. The Chief Executive was effectively protected from any democratic repercussions from such idiosyncratic and repressive behaviour because of his direct appointment by the President. Meanwhile, in 1998, licence fees for traders were increased by a factor of three, provoking a major dispute between the municipal government and the market traders. This issue was exacerbated as the collection of the licence fees was assigned to a traders association widely perceived to be in the hands of the national ruling party rather than representing the interest of traders (Devas and Korboe, 2000, p130). The overall judgement on the Kumasi Metropolitan Assembly is that: KMA does not have any specific pro-poor interventions, it also stifles the attempts by the poor themselves to earn a decent living through trading activities, for instance, through the seizure of items being sold at unauthorised places (King et al, 2001, p104). In Ahmedabad, a voluntary association, SEWA (Self-Employed Womens Association) has campaigned to protect the rights of vendors and those who work in the unprotected sectors. In particular, it has worked with the families of jobless former textile workers. A study in 1995 estimated that there were 72,500 workers engaged in home-based (but also informal) trades. SEWA has fought a legal battle for these local vendors to protect their rights and has helped them organize into a national alliance for vendors. At the local level, SEWA has assisted those involved in rolling out incense sticks (argabatis) to increase their piece rate from Rs 1 to Rs 5 per 1000 sticks. However their earnings are still below the minimum day wage of Rs 3540 prevailing in other states (Dutta with Batley, 1999, pp5051). The Cebu study provides perhaps the most interesting and complex series of episodes in relation to the informal sector. As noted elsewhere, Cebu has a fast growing local economy, but street vending remains an important survival strategy for the urban poor. During the 1980s the situation of vendors was described as dreadful, with demolitions a daily event. At this time it was the Mayors wife who was active in the programme, claiming that she was sweeping and/or cleaning up the area! (Etemadi, 2001, p48). The previous chapter (Box
7.3) has already outlined the process by which street vendors got together to form the Cebu United Vendors Association in order to protect their livelihoods, first through confrontation and then through negotiation with the city government. Despite many setbacks, they were able to bring about a change of policy. According to Etemadi, the plight of the informal street traders and sidewalk vendors has gained recognition. Despite the city ordinance against illegal vending, Mayor Garcia and the Working Committee on the street vendors regard the issue as one of regulation rather than outright eviction and demolition. As pragmatic politicians seeking office or re-election, they cannot ignore the interests and demands of the urban poor who constitute over 60 percent of the citys residents (Etemadi, 2001, p89). This was to result in a new policy: Street vending is [still] illegal by law [and this was supported by the Supreme Court and other local level ordinances.] In practice however, Cebu City has implemented a policy of maximum tolerance towards the street vendor, in terms of unwritten regulations or guidelines that must be followed instead of outright demolition and/or eviction. City officials have come to accept the reality that this subsector of the informal sector provides an alternative livelihood for the unskilled labour force (Etemadi, 2001, p120). Box 8.2 provides some details of these informal rules. In summary, the story of Cebu is one of a degree of openness by the municipal government and an appreciation of where its political support comes from, but also a change in tactics by those lobbying government. The vendors organizations moved from confrontational politics to a strategy characterized by lobbying and engagement (Etemadi, 2001, p142). The story from Bangalore is also complex. It is concerned ultimately with local democracy and how that does or does not provide political and institutional access resulting in pro-poor and inclusive decisions. While Benjamin makes this argument in general terms, it also applies to the informal sector. As already noted, Benjamin rejects the concept of the informal sector due to its inherent ideological assumptions of linear development trajectories (Benjamin and Bhuvaneswari, 2001, p19), focusing instead on concepts of economic clustering. He also argues that it is only possible to understand local economic activity at a finer grain. In particular, areas of mixed land use allow for a diversity of economic activities, which can mutually benefit local traders in complex trading relations. Such areas allow for entry points for newcomers and for poor households. Drawing on earlier work in east Delhi, the concept that captures this is that of the neighbourhood as a factory (Benjamin, 1993). A related issue is the way such areas allow for an incredibly diverse range of local financial mechanisms to operate. This includes chits and other local, indigenous financial arrangements, which support and are intimately tied up with the development of individual businesses and the local economy. There is much in common here with Rutherfords discussion of local indigenous financial arrangements (Rutherford, 2000). In all this, questions of access to land, and the location of that land, are critical. As such, the role of municipal planning is highly significant. Large parts of Bangalore fall under the Bangalore Development Authoritys (BDA) Master
Plan. The rigid zoning of these areas results in homogeneous land-use patterns, making it difficult for the poor to establish claims. By contrast, in the unplanned areas where local economies flourish, there is a more flexible and incremental process at the local level, involving councillors and officials in allocating or legitimizing claims to land in ways that are accessible to the poor and to small-scale enterprises (SSEs). A key factor in explaining the difference between these two approaches is the accountability of those involved. In the latter case, local politicians have to be responsive at least to some extent to their local, ward-based electorate, and must be seen to be delivering some tangible benefits. In the case of the master planned areas, decisions are made at higher levels, remote from those immediately affected, whether at the state level or through an appointed body such as the Development Authority with minimal accountability to citizens. Box 8.3 illustrates the complex relationship between traders and the processes of urban renewal, but also illustrates the way in which money and bribery are necessary to secure an appropriate urban location. The experience of hawkers in the implementation of hawking zones in two different areas of Bangalore further illustrates the argument. Ironically, the threat to hawkers came from an apparently progressive source. The idea of hawking zones was one supported by a lobby group of academics and concerned NGOs across India. In Bangalore, this group advocated that Bangalore City Corporation (BCC) should demarcate hawking zones. In 1999, the courts directed BCC to declare such zones a judgement that was bound by the Indian Supreme Court 1989 ruling on pavement dwellers in Mumbai. This judgement recognised hawking as a fundamental right but allowed for eviction by state agencies provided allotment for alternative places was made (Benjamin and Bhuvaneswari, 2001, p158). What is interesting is how this policy was implemented in different areas of Bangalore. Hawking zones were to be identified jointly by the BCC and the traffic police. Recognized hawkers were allocated space and given licences. Almost without exception the zones identified were in areas with minimal pedestrian traffic and so were not suitable for hawking business. This had a drastic impact on the hawkers with one exception. This relates to hawkers in Central West Bangalore, who were able through their councillor to organize the licensing system on an as is where is basis. The result was that rather than being displaced as in other cases, these hawkers were able to consolidate and regularize their claims (Benjamin and Bhuvaneswari, 2001, p159). The difference was that, in Central West Bangalore, leaders were able to negotiate on behalf of their members, and local politicians were responsive to the demands of their constituents. Benjamin and Bhuvaneswari (2001) refer to such forms of incremental negotiation as politics by stealth, in which poor groups and their alliances operate in low key and non-visible but persistent ways to achieve their objectives. This has emerged as a strategy to subvert a fundamentally inappropriate planning system, to achieve security of land tenure or to protect traders from harassment. The key to politics by stealth is local councillors who find ways to accommodate the demands from their constituents in conjunction with low-
level officials (the porous bureaucracy). Problems are resolved in unofficial and non-transparent ways, outside the formal, legal system. Paradoxically, as we saw in the case of the hawking zones, moves towards a formal, legal system and greater transparency can undermine the informal political arrangements that enable the poor and their local economies to survive. Thus, the current system of local politics, which may seem to many to be corrupt, can sometimes work to the advantage of the poor, whereas simplistic reforms may worsen their position. The relationship of the informal sector to municipal government in Johannesburg has a somewhat different dynamic. One of the key issues emerging there is the tension between the local African population and the new
often illegal immigrants from other regions of Africa. An example of this was the sudden explosion of hawkers and traders on the streets of Yeoville, an inner area of the city. Given the structure of South Africas housing market, these inner city areas are the only places where recent migrants can live and trade. The most controversial issue was the erection in 1999 of the Yeoville market and the relocation of hawkers from the streets into the market. While local residents seem to have approved of this action, the hawkers felt that their business had suffered from the lack of passing trade. There was a dispute about allocation of stalls in the new market, particularly over whether foreigners had been over-represented. What is interesting about the dispute is that the traders and their association, the Rockey Street Traders Association, did not appear unlike their counterparts in Cebu and Bangalore to have good connections into local politics. Instead, they were seen in terms of community politics more generally as a lot of hot air (Beall et al, 2001, p135).
Emerging themes
In this section we discuss a number of particular themes that emerge from the city case studies, leading on to a more general discussion of the role of voice in the relationship between the informal sector and city government.
attitude and practices towards the informal sector between municipalities within the same country. This was the case in Ghana in the 1990s: while Kumasi was involved in a policy of harassment of informal sector traders in the central business district (CBD), Accra was developing a more tolerant policy. It is therefore relevant to observe that the relationship is always mediated and the outcome is the result of some form of compromise between those involved. The question is: how is this compromise arrived at? The outcome obviously depends on the parties concerned, principally the municipality and those in the informal sector, but also involves a range of elected and paid officials, community leaders and organizations representing informal sector businesses. The result depends on the relative bargaining strength of those involved, their tactics, and the nature and responsiveness of the political processes, informal as well as formal, through which the negotiations are conducted.
banks effectively encouraged the riot. After three days of rioting, the city government had to back down and allow the units to remain (Roy, 2001, pp206207). The conclusion from this is that it is simply not possible to enforce a planning regime that, at a stroke, makes 200,000 individuals unemployed certainly not in a democratic system which allows citizens a voice. The recent work by Amartya Sen on Development as Freedom (2000) is relevant here. In this analysis, Sen makes a strong case for the idea that development ultimately is about securing five distinct freedoms: political freedoms; economic facilities; social opportunities; transparency guarantees; and protective security (Sen, 2000, p10). These five freedoms are inter-related but ultimately advance the capability of the individual in relation to the state. Developing his now familiar argument concerning the importance of democracy in the prevention of famine, he goes on to discuss the role of democracy in other circumstances, thus:
The preventive role of democracy fits well into the demand for what was called protective security in the listing of different types of instrumental freedoms. Democratic governance, including multi-party elections and an open media, makes it very likely that some arrangements for basic protective security will be instituted The positive role of political and civil rights applies to the prevention of economic and social disasters in general. When things are routinely good and smooth, this instrumental role of democracy may not be particularly missed. But it comes into its own when things get fouled up, for one reason or another. And then the political incentives provided by democratic governance acquire great practical significance. (Sen, 2000, p184)
Concluding remarks
From this review of the city case studies in relation to the informal sector, what can be said to be the key factors which determine the relationship between that sector and city government? The first is the nature of illegality. The activities of informal sector enterprises may transgress any of a number of laws and regulations, but the ability of city government to take action and the extent of public support for that action depends on how serious those transgressions are considered to be. Some activities, like street vending and informal settlements, while they may be technically illegal, have a high degree of public acceptability. Even the illegal tapping of water pipes may have wide acceptance, at least within a particular community, especially where this is the only way for the poor to access public water. Meanwhile activities like dealing in hard drugs and robbery are socially unacceptable as well as illegal. There is likely to be much greater political support for repressive policies towards the latter than the former. Second, the scale of the problem will influence the ability of city government to take action. Where, as in the example from Delhi, huge numbers of people are affected by a planning policy, it is likely to prove difficult for the authorities to enforce that policy.
Third, the ability of those affected to organize and to oppose or lobby effectively is critical. Effective, organized opposition greatly increases the costs to city government of taking action. The success of the street vendors in Cebu although far from total was the direct result of organizing to oppose oppressive policies, and then moving to constructive engagement with the city government in developing more appropriate policies. Fourth, there is the matter of the accountability of those making decisions, not just to citizens in general but to the poor in particular, including those in the informal sector. An elected mayor cannot afford to alienate a very large section of the electorate. Local councillors, at least those representing low-income wards, must demonstrate responsiveness to their constituents. On the whole, the closer the elected representative is to the electors, the greater the chances of accountability. Thus, in Cebu, when the city government sought to ban the use of trisikads (cycle taxis), elected leaders at the local barangay level continued to permit them to operate within their areas. By contrast, in Bangalore, the Development Authority, which is not directly accountable to citizens, can impose policies that effectively exclude the poor majority of the population. In this chapter we have sought to discuss issues associated with the regulation of the informal sector. We have argued that the way this regulation operates is largely determined at the very local level, and at the stage of implementation. Whilst, in practice, much of this regulatory activity is repressive and burdensome for those in the informal sector, there may be scope for those involved, especially where they are able to organize, to negotiate less damaging arrangements. Their ability to do so depends, we have argued, on their voice, which in turn depends on the nature of the local-level democratic processes and the accountability of those elected to their constituents. In the end though, it often easier for municipal governments, in their dealings with the informal sector, to undermine or destroy the livelihoods of the poor, through relocation or repressive regulation, than it is to enhance those livelihoods.
Notes
1 2 This is of course a familiar criticism of the planning process in the North. The earliest and still a classic text is Jane Jacobs (1961) The Life and Death of Great American Cities. A senior official from the city government of Durban claimed that the city had created 30,000 jobs without using any capital, simply by changing the policy towards hawkers. As with any job creation estimate, it is difficult to know the accuracy of the figures or the extent to which jobs were created rather than simply relocated. This is, however, very much a local policy: when the Manila-based owners of the shopping mall are expected to visit the Cebu branch, the street vendors are asked to disappear; once the executives have left, the vendors can return and reclaim their sites (Etemadi, 2001, p95). There are parallels with the movement against the Narmada dams in India, in which the Indian Supreme Court eventually ruled by 2 to 1 in favour of the dams (Roy, 2001).
Chapter 9
Accessing Land and Services: Exclusion or Entitlement?
Fiona Nunan and Nick Devas 1
The challenge
Urban poverty is inextricably linked to peoples access to land and basic services. Those living below the income poverty line almost invariably suffer also from insecure and over-crowded housing, inadequate access to water and sanitation, and environmentally hazardous living conditions (Hardoy et al, 1992; UNCHS 1996; UN-Habitat, 2001a; McGranahan and Satterthwaite, 2002). The economic and social costs of the disease, injury and premature death associated with poor quality housing, infrastructure and services are often underestimated in analyses of poverty. These burdens fall most heavily on vulnerable groups such as women, children and the elderly. Even where basic services like water are available, they are often located far away, operate only intermittently, and are costly. Quite apart from the direct effects on physical well-being, the inadequacy
of such basic services increases the costs to the poor and undermines their ability to earn a livelihood. One of the most direct influences city governments have on the scale and nature of poverty is in what they do or do not do in regard to provision of basic services and in supporting housing construction and improvement. City governments influence on land markets is also important since the possibilities for lower-income groups to obtain housing with infrastructure and services usually depends on the opportunities available to them to obtain land. Yet the potential contribution of city and municipal authorities to poverty reduction is often underestimated, as discussions usually focus on inadequate incomes or consumption, and on the role of national government and international agencies in addressing these. Meanwhile, primary responsibility for the provision of basic environmental services so critical for the life-chances of the poor usually rests with city or municipal government, even if delivery of such services may be delegated to private or non-governmental agencies. Ensuring access by poor people to adequate environmental services and shelter is not simply a technical, engineering matter, but is conditioned by a range of economic, social and political factors. It is bound up with local power structures, including the extent to which low-income groups can influence local government policies and resource allocation, and by the relationships between local government, higher levels of government and other institutions of urban governance. The central theme of this chapter is how the nature of urban governance affects the quality and extent of provision of basic environmental services, housing and access to land for the poor. The chapter outlines the inadequacies of provision in the case cities, and reviews how the poor obtain access when the formal system fails to provide. It then examines a number of key governance issues that impinge on provision of environmental services and shelter: the national policy framework; the responsiveness of city government; political accountability; and the roles of the private sector, community organizations, non-governmental organizations (NGOs) and external funding agencies in strengthening, or at times undermining, access by the poor.
Inadequacies in provision
The city studies highlight the inadequacies in infrastructure provision, even in some of the more economically successful cities such as Cebu and Bangalore. In general, the wealthier the city or nation, the greater the proportion of the population adequately served with environmental infrastructure and services. However, the quality and capacity of local governance also influence this (Satterthwaite, 1997). In relation to water, Santiago is the only case study city where nearly all of the population are adequately served. A high proportion of Johannesburgs population is also well served, although there are still considerable numbers lacking adequate provision. There is also a major issue over the ability of the poor to pay for the relatively high standard of services in Johannesburg, a city
characterized by vast inequalities in income. In the other cities, large sections of the population have no access to piped supplies, while for many more, access is difficult, supplies are erratic and the water supplied is contaminated. For instance, in Bangalore, more than half the population depends on public fountains, many of which supply contaminated water because of poor maintenance or broken pipes. In virtually all cities, those inhabitants lacking formal provision often have to pay high prices to informal providers. In most cities, official statistics understate the inadequacies in water supply by classifying all those with access to piped supplies as adequately served, even if the water supply in the pipes is irregular and of poor quality, or access is through public standpipes shared with hundreds of others. In Mombasa, a large section of the population with water pipes has seen no water in their pipes for several years because the supply as a whole is insufficient to meet demand. With regard to sanitation, most of the low-income population in all but Santiago and Johannesburg has inadequate provision. Official statistics again underestimate the scale of the problem because they assume that if a household has access to a latrine, it has adequate sanitation. Large sections of the population whose only access to sanitation is through poorly maintained communal latrines where queues are common are classified as having adequate sanitation. The case studies of Ahmedabad, Bangalore and Cebu highlight how a significant proportion of the population lack sanitation facilities and defecate in the open or into plastic bags (what is termed wrap-and-throw in Cebu). In Kumasi, nearly 40 per cent of the population depend on 240 poorly maintained public toilets (see Box 9.2). In most of the cities, large sections of the population have no regular garbage collection services. The inadequacies are particularly notable in Kumasi, where only high-income areas are served by (highly subsidized) house-to-house collection. Low-income areas are served by skips, which are emptied infrequently, and not all areas are included. In Mombasa, attempts to privatize garbage collection resulted in increased charges that the poor were unwilling or unable to pay, with the result that the private contractor abandoned the service in the areas concerned. Most urban governments have some influence over who obtains land for housing (and under what terms), its quality and location, and the likelihood of it being provided with infrastructure and services. The ways in which they allocate the land they own or control, define land uses and implement zoning and planning controls influence whether and where low-income households can legally acquire land on which to build their homes. In most of the case study cities, the supply of housing, or of land for housing, falls far short of the need, so that most poor people are forced into over-crowded and insecure slum housing, squat on public land or obtain plots on illegally subdivided private land. Only in Santiago is public and private housing construction on a scale that approaches the level of need. Even here the poor quality and adverse location of housing for low-income groups reinforce the inequalities in society. In Johannesburg, the massive, publicly-funded Reconstruction and Development Programme (RDP) housing schemes still fall far below the level of need of the
growing population. A particular issue in these more prosperous cities is that rising land prices and opposition from higher income groups have limited the ability of city governments to provide housing for low-income groups in appropriate locations, such as close to the centre. In most of the other cities, city governments have made little attempt to provide housing or even to ensure access to land for low-income groups. Indeed, land-use controls and building regulations often impede the efforts of the poor to provide their own shelter. The most positive contributions made by city governments have generally been through ex post regularization and provision of services to informal settlements. But this often occurs many years after the initial settlement and is dependent on the ability of the residents to negotiate with the city authorities in turn often dependent on clientelistic political relationships or ethnic connections.
taps and water sellers. The municipal water company will provide a communal tap where the community establishes a Communal Water Association (CWA) with elected officers to manage the tap. However, less than 10 per cent of urban poor families in Cebu make use of water from communal taps. Etemadi (2001, p118) observes that many CWAs are beset by problems, including lack of participation by members, irregular elections, undemocratic management and lack of financial transparency and accountability. Meanwhile, a large proportion of the urban poor in Cebu purchase water from water sellers, paying at least three times the rate from the municipal water company. In one area, residents reportedly pay almost ten times as much because water sellers transport water from a distance of 400500 metres, and in another, the rate was 17 times the charge for a formal water supply. Small-scale water sellers are not monitored or regulated. On the other hand, water vending is a significant source of income for some low-income households. The quality of the water reaching low-income areas can be very low. In Bangalore, 1020 per cent of the citys water demand is met by groundwater via 80,000120,000 bore-wells and open wells. The quality of the water is very poor and often contaminated. Even piped water needs to be boiled or filtered via ultraviolet light an unrealistic requirement for most poor people. The poor in central Bangalore depend upon public fountains. Although these are free of charge they often have broken taps and pipes or damaged platforms. Poor groups frequently dig pits to collect water, but the lack of drainage and sanitation results in contamination. Private tankers supply a vast population in the intermediate and peripheral zones of the city. The supply of water therefore depends on a range of delivery mechanisms, resulting in varying levels of access to, and quality of, water. Political influence plays a key role in determining whether and when infrastructure and services are provided or improved. Maintaining a reliable water supply is viewed as being of utmost importance in low-income communities. A survey of slums in Bangalore found that, in many cases where the hand-pump needed to be repaired, women had collected together money to pay for the repairs. In other cases, the leader of the sanga (local community organization) approached the local councillor to ask for the repairs to be done. Limited availability of water during the day was found in all the South Asian cities. Certain areas of Colombo, notably those occupied by low-income groups, suffer from low water pressure and therefore limited access to water. People are obliged to collect water at night for use during the day. Storage brings problems of hygiene and increased risk of disease. In Visakhapatnam, while 48 per cent of residents have domestic pipe connections, 42 per cent rely on public fountains. As there is only one such fountain for every 150 residents and water is available only twice a day (for one and a half hours in the morning and an hour in the evening), collecting water implies queuing and hence opportunity costs. Although it is possible for a slum area to receive a water supply once the area has been notified by the Urban Development Authority (ie once the slum has received official recognition), notification can take three to four years from first settlement, and usually requires political patronage. In Kumasi, about three-quarters of the city is served with piped water by the Ghana Water Company (a public corporation until 1999, when it was
converted into a limited liability company). The remaining parts of the city, notably the peri-urban areas, are served by private water vendors, boreholes, wells and streams. Many of these boreholes and wells are highly contaminated, while many of the streams used for water collection are also used for defecation. Within the areas served with piped water, around 6 per cent of the citys population depend on public taps, but the water company views these as problematic and is not proposing to increase the number of such taps. Another 27 per cent buy water either from private taps or from neighbours (Rakodi, 1996). Because of the high initial cost of a water connection, many residents resort to making unauthorized connections. The proliferation of such unauthorized connections results in leakage and risks of contamination, rendering the supply less reliable and the whole system less financially viable. In order to obtain a connection or a repair, residents often make contact with the companys technicians, paying them directly. The payment is generally below the official charges and receipts are not issued. This route is faster and more accessible than going through the formal application process. Those who do not have the resources for such payments may make their own illegal and unmetered connections, or else continue to depend on vended water or contaminated sources. In Mombasa, a particular issue is the inadequacy of the bulk supply. Although official statistics (the 1989 census) indicated that 92 per cent of households had access to piped water, very few parts of the city have a continuous supply, and some areas have had no water for several years. In a survey of local residents, water was identified as the top priority in terms of problems affecting their community (AMREF and OVP/MPND, 1997). In response to the inadequacy of the public provision, many residents have had to cut back on water use or have resorted to obtaining water from wells, boreholes or vendors.
Gaining any firmer foothold in terms of land and shelter often requires ethnic or political connections, and may require group negotiation to lower prices and reduce risks of accessing private land. Around the KR Market area of Bangalore, low levels of infrastructure and relatively insecure tenure make plots affordable for the poor. Land is transformed incrementally, giving the poor the chance to remain and to consolidate their position. The variety and flexibility of tenure arrangements outside the master planned areas of the city also enable the poor to establish themselves. Shelter tends to be constructed incrementally, using a variety of materials. Provision of infrastructure and services in these areas is incremental, influenced by politics, particularly through the consolidation of vote banks. Benjamin and Bhuvaneswari (2001, pp6162) describe a common access route in one ward, Azadnagar:
The poorest groups may move into a locality as renters in a squatter settlement in a thatch house. For this, they would initially borrow an amount from the private financiers to pay the required advance. This may be helped by seeking support from one of the local leaders (using ethnic, employment-based or class linkages) to gain a foothold in the political system. This move can help them squat on the land. Once settled in close proximity to employment opportunities, access to various social circuits helps them consolidate their social and economic contacts. With greater stability, they move on to other tenure forms or to more productive locations. Those groups who manage to accumulate surplus seek to invest in the private layouts.
Access to land is not just important in relation to shelter but also in terms of economic activity. Benjamin and Bhuvaneswari note that land is an active ingredient for the economic strategy of poor groups (2001, p4). Land, they suggest, interfaces with the economy in three ways: 1 2 3 access to productive locations: those areas that provide access to multiple employment opportunities are more advantageous for the poor; diversity of tenure regimes that allow for the poorer (among other groups) to consolidate their claims on contested locations; upgrading of infrastructure and services (the quality of land) which both influences the productive potential and reinforces tenure claims (2001, p4).
They also note that land plays a critical role in the expansion and diversification of local economies. This is illustrated in Azadnagar, west Bangalore, where many small enterprises are based. The range of infrastructure and services available in different locations affects the types of economic activities that can take place there. Although the forms of tenure are heavily influenced by institutional and political factors, the diversity of tenure allows a range of income groups to access space, with different combinations of space size, cost and land use (economic, commercial, residential) to suit different needs, including those of the poor. The same building may be used for both productive activities at the front and accommodation at the back, with multiple rented
rooms (vattarams) with a range of rental options. This contrasts with the high costs, uniform layout and inflexible tenure in formal schemes, such as those developed by the Bangalore Development Authority (BDA), which provide few opportunities for the poor. In Kumasi, most low-income households live in rented housing in single or multi-storied compounds. These are often over-crowded with minimal facilities: 30 per cent of those living in indigenous compound housing lack toilets, water and electricity. Overcrowding is exacerbated by the tradition that homeless family members can expect to be housed by other members of their family a crucial means of support for the poorest. There has been an increase in the number of squatter settlements located on the marginal lands around the city, as well as a growing number of people with no roof over their heads, occupying abandoned buildings or paying on a daily basis for a place to spread their mats (Abugre and Holland, 1998). These people are often young, mostly migrants, and are often organized according to village or kinship lines. The allocation of land in Kumasi (except for state land) is the responsibility of the traditional authorities, that is, chiefs and queen mothers. The level of access depends on whether the applicant is a stranger or indigene. For the indigenous poor, access is determined by family landholdings, family membership and the position of the family and the individual in relation to stool (traditional) lands. While the official cost of obtaining land is relatively modest (drink money for the ancestors), the process has become increasingly commercialized. Thus, while the traditional system of land allocation does enable low-income groups (at least those who are native Asantes) to gain access to land, rising demand for land has resulted in prices that increasingly reflect market conditions, thereby excluding the poor. This observation confirms Jenkins (2001, p630) view that in general, in urban areas traditional and customary rights begin to enter the commodification system, and informal markets may not be as effective when exchange values grow more important. While traditional systems of authority and land allocation may provide some scope for the poor to access housing and land, with increasing pressure on urban areas and the subsequent emergence of land markets, such routes are diminishing.
elections; the consequences of formal private sector involvement in service delivery; the role of community organizations and NGOs; and the role of external funding agencies.
In the provision of housing for the poor, national policy has had a critical influence on access in many cities, even though the policies may not always have been ideal. In the Philippines, the responsibility for the provision of housing was devolved to local government through the 1991 Local Government Code. The 1992 Urban Development and Housing Act provided the impetus for the implementation of a social housing programme. In Cebu, this became one of the functions of the Cebu City Commission for the Urban Poor (CCUP), which was set up following the creation of the Presidential Commission for the Urban Poor at the national level. One national programme, the Community Mortgage Program (CMP), has been particularly effective in facilitating access to housing for the poor see Box 9.1. In South Africa, the chronic housing shortage was a visible legacy of the previous regime, so providing housing has been a major issue for the postapartheid government. During the apartheid years, housing for black people was provided either through standardized public housing in townships or through site-and-service schemes, where people were expected to build their own houses on serviced land. From the mid-1970s, there was a trend for people to build shacks in the backyards of government-owned houses, and, from the 1980s, shanty-towns emerged on land adjacent to the formal African townships. Huchzermeyer (2001, p305) notes that current housing policy in South Africa is based on a fundamental understanding that housing is a basic need. This policy is articulated in the 1994 Housing White Paper, with a commitment to deliver 1 million houses in five years. Huchzermeyer reports that the most common form of subsidy used to promote house building has been the projectlinked subsidy, channelled through private developers, which has led to the development of uniform, free-standing, mostly one-roomed houses with individual freehold title in standardized township layouts, located on the urban peripheries (ibid, p306). It has been suggested that these segregated dormitory developments are themselves poverty traps and do not represent a route out of poverty. The target set in the White Paper was not reached and, subsequently, there has been a shift in thinking away from a blanket solution of free-standing houses on individual plots with freehold tenure. Part of the housing problem now is of beneficiaries leaving their new houses to return to squatting, in part due to the cost of travelling to and from employment opportunities. This illustrates how a better understanding of the socio-political and socio-economic environment of the poor is needed to develop more appropriate solutions to the housing crisis. In South Africa, then, housing is seen as critical to poverty reduction and the national housing policy provides those involved with a framework in which to respond to need. Although the policy framework may not be entirely appropriate, without it urban governance agencies would have had less direction and fewer resources than they have to address the acute housing need. National housing policy has also played a crucial role in Chile. Rodrguez and Winchester (1999) suggest that Chiles housing policy since the early 1990s has been successful in significantly increasing house construction. Diverse housing subsidies are available for homeless families, managed by the Housing and Urban Development Ministry. These include Progressive Housing, a
programme for low-income families, Basic Housing, for families in rented accommodation, PET (Special Programme for Workers) for those belonging to labour unions or guild organizations and Urban Renovation, aimed at families who choose to live in zones selected for Urban Renewal programmes in the centre of a city. In these programmes, the role of the municipality is to identify and organize the demand, while implementation is by the private sector (construction companies, housing corporations and housing cooperatives). Although these programmes have been successful in mass-producing housing and in dramatically reducing the number of people living in precarious settlements, there has been much dissatisfaction with the type, quality and location of housing created.
By contrast, in both Ghana and Kenya, there is little if anything by way of a coherent national policy framework or dedicated national resources for the delivery of urban services and shelter. Indeed, the absence until recently of any form of inter-governmental transfer in Kenya has meant that Mombasa Municipal Council (MMC) has had virtually no resources with which to tackle the backlog of urban infrastructure. As already noted in Chapter 6, conflicts between levels of government, in Mombasa as well as in several of the other case cities, has tended to undermine whatever capacity there is within city government to address the needs of the poor.
demolition of informal settlements was necessary. A memorandum of understanding was agreed between FORGE, an NGO housing advocate, together with ANAK (a community-based organization facilitated by FORGE), and the judiciary and police, to ensure that there is adequate consultation and consideration of relocation options prior to demolition. In Colombo, there was a period when the Million Houses Programme allowed low-income groups a much greater role in improving infrastructure and services through community action planning. It proved difficult, however, to sustain this more participatory model in the face of political changes, as explained in Box 5.4. In Recife, the participatory budgeting (PB) process (described in Box 6.3) has ensured that at least part of the citys budget is spent on infrastructure in low-income neighbourhoods in response to the expressed priorities of the residents. There are, however, particular difficulties in providing infrastructure and services to some informal settlements. For example, many have developed on steep slopes or flood plains, or have unclear plot boundaries. There are also more general problems of speculation and inefficient use of land (eg vast plots which remain undeveloped), which greatly increase the costs of infrastructure and make access to services by the poor more difficult. A further issue, given the limited resources available, is how to accommodate the competing demands for bulk infrastructure and local services. The demand from residents of particular neighbourhoods for improved water supplies, for example, may mean that networks are extended and new taps provided despite there being insufficient capacity within the system to enable the water to be delivered. Similar issues arise in relation to extensions to drainage and sanitation, waste collection, electricity and roads: extending local provision may achieve little if there is not the bulk capacity to handle it. The consequence of expansion to serve the poor is often ever greater overloading and failure of the service for everyone. This is exacerbated where under-pricing of services to those who can afford to pay means that there are no resources to fund increases in capacity. A number of the case studies indicate a reluctance on the part of senior technicians (for example, city engineers) to extend services to underserved areas for fear of jeopardizing the supply to those already served. There is, therefore, a dilemma: greater responsiveness to demands from citizens (of whatever income group) for local-level improvements can pre-empt the available resources needed to improve the bulk supply, leaving everyone dissatisfied. This is indeed the position in many of the case study cities.
The Bangalore case provides a contrast between the influence albeit limited which low-income residents have on the Municipal Corporation through their elected councillors (see Box 5.2) and the absence of political accountability of major players such as the Bangalore Development Authority (BDA). The appointed BDA Board includes only two locally elected representatives out of 23 members. As a result, it has little interest in prioritizing the infrastructure needs of poorer groups. Politics plays a role in facilitating access in the Indian cities and in Cebu. In respect of the latter, Etemadi (2001, p138) reports that: In the short run, elections are said to be helpful, because it is only during this period that the communities can make their request for basic services which are fast tracked and granted. They can bargain services for votes, and there are no demolitions during the period. Thus, elections present an opportunity to secure access to services, but this requires vote bargaining, with unequal results for different communities, depending on their degree of organization, the efficacy of their leadership and their political affiliation. Not only do communities vary in their success in obtaining infrastructure, but also elections are too infrequent to deliver adequate access. Furthermore, the process of demand-making, which depends on political patronage, hinders systematic provision. Kumasi and Recife present contrasting examples in terms of political accountability. The Kumasi Metropolitan Assembly (KMA) is headed by an appointed mayor, officially known as the Metropolitan Chief Executive (MCE).2 The mayor at the time of this study ruled with an individualistic and autocratic style, allocating resources and services largely on the basis of political patronage. One example was the totally inadequate solid waste service. Although many areas had been provided with a dumping site, waste was only transferred from these to a final dumping ground very infrequently. Up until 2001, approval for the cleaning of areas of the city required approval of the MCE, who tended to refuse to provide services for those areas represented by Assembly members who were not his supporters (King et al, 2001, p85). Another example of political patronage was the privatization of the citys public latrines, described in Box 9.2. One reason why the mayor in Kumasi could get away with such arbitrary behaviour, so damaging to the poor, was the fact that he was nominated, not elected, and is therefore not politically accountable to the residents of the city. Recife offers a different experience, with an elected mayor bringing in a new programme to formally recognize slum areas and to work in partnership with a range of stakeholders to deliver appropriate tenure, services and infrastructure. This is described in Box 9.3. Although the presence of an elected mayor is certainly not the only factor in improving the services the poor can access, it can be significant. In terms of the title of this chapter, while Kumasi is largely a picture of exclusion, Recife could be said to be moving in the direction of entitlement.
the poor. Most of this is through small-scale, informal and often semi-legal or illegal activities. The formal (or semi-formal) private sector, while a significant provider of rental housing, has a much more limited role in relation to service provision. One area in which this is changing is waste collection, which private contractors in a number of the case cities provide under contract to local government. While in some cases the involvement of the private sector has improved performance, it has often been associated with reduced levels of service and increased costs for the poor. In part this is because, where services like waste collection are contracted out to the private sector, local governments need new skills in tendering, contracting and supervision skills that are often lacking. In Santiago, the contracted-out service seems to have worked quite well, with five companies between them collecting the bulk of solid waste in the 32 comunas in the late
1990s, although the precarious settlements are not covered (Rodrguez and Winchester, 1999, p124). However, in Visakhapatnam, lack of supervision and monitoring of private contractors was cited as an issue, and it is alleged that private contractors have not been collecting as much garbage as they claimed, and are required, to collect. In Ahmedabad, apart from privatizing the cleaning of a couple of important commercial and recreational areas of the city, the main reforms have been to introduce community-based waste collection systems and to subsidize local private societies to clean certain residential areas. With contracted-out services, low-income communities can easily be disadvantaged, due to a lack of official recognition of slum areas, as well as overcrowding which makes access by collection vehicles difficult. Privatization also makes the conflict between poverty reduction and cost recovery particularly apparent. The Mombasa Municipal Council (MMC) privatized most of the garbage collection service in the late 1990s, partly due to the inadequate rate of collection by the corporations own staff (only half of the 400 tonnes generated daily was being collected). The private companies introduced higher charges which many people on low incomes were unwilling or unable to pay.
Contractors responded by excluding the low-income neighbourhoods they were contracted to cover. This resulted in dumping along roads, with the MMC forced to collect the dumped garbage. It seems that privatization took place without adequate legislation to force all the parties concerned to meet their obligations. Previously, MMC had cross-subsidized the service to the poorer areas from the wealthier areas, a practice that private contractors were not willing to continue. Fears are often expressed about the potential effects of increased private sector participation on access to services by the poor. For example, discussions about the possible commercialization or privatization of certain public services in Johannesburg under the iGoli 2002 strategy provoked protests about the potential impact on the poor as well as about the likely loss of jobs. In every case city, the private sector plays a major role in the supply of rental housing. Housing types and tenure arrangements in the private sector are often more flexible and more appropriate to the needs of the poor than those provided by the public sector. The informal private sector is often the main supplier of low cost housing. However, without an appropriate policy framework and financial inducements, the private sector (whether formal or informal) cannot be expected to contribute to the goals of poverty reduction. On the other hand, the construction of rental housing can provide an additional source of income to low-income home owners, although this is unlikely to benefit the poorest since it requires access to both land and capital. Even where there are financial inducements, these may not end up benefiting the poor. In South Africa, there is much criticism of the way the housing subsidies under the Reconstruction and Development Programme (RDP) have been monopolized by private developers providing low quality, standardized housing units. There are similar concerns about the quality of housing built by private contractors under the various subsidy programmes in Santiago.
voluntary organizations. Various CBOs and NGOs are involved in slum upgrading or other community development initiatives, including measures to improve environmental health among low-income groups. These include: ASAG (Ahmedabad Study Action Group, founded in 1971), SEWA (the SelfEmployed Womens Association founded in 1972), St Xaviers Social Service Society (founded in 1976), Vikas (founded in 1978) and SAATH (founded in 1989). The Slum Networking Project (SNP), described in Box 9.4, was developed jointly by the municipal corporation and the private sector, and sought to involve both CBOs and NGOs from the outset. The city studies illustrate a number of examples where NGOs and CBOs have played a constructive role in mediating between local governments and communities to obtain or improve services and housing. In some cases, they support small-scale initiatives that may act as a pilot and from which lessons can be learnt for broader application. While NGOs and CBOs often strengthen the voice of the poor and can advocate on their behalf, they generally cannot provide infrastructure or services on a large scale. They must, therefore, work closely with local governments and the private sector, as well as with
communities and the informal actors involved in service provision. As Chapter 7 has shown, the limitations of such organizations must also be recognized. For example, in Colombo, among a diverse range of NGOs, there are several whose work centres on improving infrastructure and services or supporting CBOs to improve conditions. However, interviews with community activists found that many disliked both local and international NGOs because of their lack of accountability to low-income groups and community organizations. Similar views were expressed in Bangalore. There are community organizations in low-income areas of Bangalore, popularly known as sangas, which focus on securing land tenure, accessing basic services and upgrading the quality and quantity of services. CBOs and NGOs in the richer areas of the city tend to be concerned with solid waste management, the upkeep and maintenance of roads and parks and antiencroachment measures. Given the limited resources of the municipality, there is a concern that these groups exert their claims to public investments at the expense of poorer groups. In the low-income settlement of Diepsloot in Johannesburg, the Community Development Forum (CDF) has been very effective in negotiating for security of tenure and improved services. Box 7.2 describes the role of the CDF, noting the dangers of such an organization becoming the monopoly channel for such negotiations, and the risks of exclusion of other, even poorer, groups. Although in many of the cities there is a multiplicity of organizations involved, the scale of their activities is still far below what is needed. In addition, where NGOs and CBOs are involved in neighbourhood infrastructure, they are often frustrated by the lack of investment in the trunk infrastructure (water mains and the freshwater sources and treatment plants to serve them, trunk sewers and drains) into which each neighbourhoods network can feed. The costs and complexity of such investments are far beyond the capacity of NGOs or CBOs, but without them neighbourhood schemes to improve water, sanitation and drainage can have little benefit. In some of the cities, however, there appears to be little involvement by NGOs in relation to urban infrastructure or services. This may be because there are few or no NGOs working on issues of urban poverty (Visakhapatnam) or because most NGO funds are directed to the rural areas. In some cities, for instance Mombasa and Kumasi, it may have more to do with the municipal authorities unwillingness to work with NGOs and CBOs, or because the NGOs direct their attention to local offices of national agencies rather than to the city government. In many of the cities, there were NGOs at work on environmental issues such as wildlife conservation or the global environment that have little or no relevance to improved environmental health for the citys low-income groups.
example, the World Bank, the Department for International Development (DFID) and the United Nations Development Programme (UNDP) have each funded projects to improve the provision of basic services such as drinking water, access roads, drainage, sanitation and street lights. However, the sustainability of such measures is open to question in terms of replication and maintenance. Although the KMA was involved in these projects, the Assembly has lacked both the political commitment and the resources to support these initiatives adequately. In such programmes, city governments tend to be involved merely in deciding which areas should benefit rather than in the delivery of the project, which is often assigned to a special project agency. Slum improvement programmes have been a key element of donor assistance for poverty alleviation in urban centres (Amis, 2001). The UNICEFsponsored Urban Basic Services Programme (UBSP), for example, involved improving access to basic services through the development of communitybased organizations and initiatives in a number of cities. In Ahmedabad, the UBSP was launched in 1985, through a partnership arrangement between the United Nations Childrens Fund, the state government and central government. Between the mid-1980s and mid-1990s, the UKs aid agency (the Overseas Development Agency (ODA), now DFID) invested considerable resources in slum upgrading projects in India. One of the most successful, in Visakhapatnam, was estimated to have benefited more than 85 per cent of the citys slum population, and is briefly described in Box 9.5.
lack the resources to purchase services or for bribes, and lack connections and influence, even within local social networks. Regularization of informal settlements is a significant means of delivering services and land for housing to the poor. But for people in poverty the relationships between land tenure, housing services and income opportunities are complex and often fragile. Therefore, interventions need to be based on a proper understanding of how the poor gain access to and make use of the land and services they have. The ability of regularization and upgrading to address the priorities of the poor depends on the nature of the processes involved, the extent of participation by poor people and their representatives, and whether there are safeguards to prevent the poor from being displaced. In the end, what is needed is a shift in the approach of the institutions of city governance from one of exclusion of the poor to one of entitlement. In some of the cities studied, exclusion from land, shelter and basic services remains the dominant experience of poor people. In several of the other cities, there are positive signs of attempts to extend access and inclusion. To shift to a position where the poor receive services and land for shelter as of right, rather than as a favour, requires a rare combination of an appropriate national policy environment, adequate resources, effective mechanisms of political accountability, and an active civil society articulating the needs of the poor. There were a few examples within the case study cities where such a combination of circumstances existed, but not many.
Notes
1 2 The authors gratefully acknowledge the contribution of David Satterthwaite to developing the research and many of the ideas on which this chapter is based. See note 4 of Chapter 5.
Chapter 10
Conclusions: Urban Governance, Voice and Poverty
Nick Devas
In this book we have looked at how city governance in the widest sense impacts on poor people, and the ways in which poor people can have voice and influence over the institutions of city governance. We have examined these issues on the basis of studies of ten cities in Asia, Africa and Latin America. It is necessary to be highly circumspect in drawing general conclusions from a limited number of specific cases, especially since the experience of each city depends on a host of particular conditions: economic, social, political, historical, cultural and geographic. Nevertheless, there are some common themes that emerge, and some important lessons. Given the limited comparative research so far in this area, we believe that these conclusions can be of value to policy-makers (both elected and appointed), planners, community organizations and non-governmental organizations (NGOs), as well as international agencies. In this chapter, we will summarize the emerging themes and review the conclusions from the previous chapters in terms of the original research questions. We will then present some
CONCLUSIONS 187
implications for policy-makers, and conclude with an indication of possible directions for future work. First, though, we make a comparison between the case cities in terms of the issues we have examined in this study.
City comparisons
Table 10.1 presents a very crude ranking of the performance of the ten cities in terms of some key indicators reviewed in this book: economic growth; poverty reduction and inequality; political institutions and processes; participatory mechanisms; the role of civil society; institutional constraints on city government, including finance; access by the poor to land, infrastructure and services; the behaviour of city government towards the informal sector; and specific poverty reduction initiatives. The table seeks to indicate whether the performance in each city is, on balance, positive or negative in relation to the urban poor. Such comparisons are open to a host of criticisms. They are, inevitably, gross over-simplifications of complex issues, especially since each element in the table is made up of a number of sub-elements, each of which may have both positive or negative features. The rating uses a very limited scale and so can only be the crudest approximation of the differences between cities. Furthermore, given the limitations of the data from the case cities, and the absence of uniform definitions and measures, comparisons between cities remain subject to the perceptions of the researchers involved and are likely to be contested. It is also important to stress that the judgements are essentially relative rather than absolute: although a particular city may perform well against the others in the study, that performance may still be quite poor in an absolute sense. It should also be noted that the judgements made relate to the period leading up to and during the study (roughly early/mid-1990s to 2000/01): conditions may have been different before that and may have changed since then. Despite all these caveats, the table does give some indication of the relative performance of the cities, and it is clear that some cities perform rather better than others. Since performance appears to be generally correlated with level of development, the cities have been listed in descending order of national income per capita. These assessments will be commented on in the following section, in relation to our research questions.
Aspects: economic and political 1 City economic growth. 2 Equality: evenness of distribution of income and wealth. 3 Poverty reduction: in absolute terms (NB: such measures are often disputed, notably in Cebu). 4 Poverty reduction in relative terms: reduced inequality and greater inclusion of the poor. 5 Formal political institutions and systems which are, potentially at least, inclusive of the poor. 6 Political processes in practice, informal as well as formal, and the extent to which these are inclusive of the poor. 7 The existence of specific participatory mechanisms, including sub-city levels of government, which are inclusive of the poor. 8 An active and effective civil society (notably CBOs/GROs and NGOs) involved with urban poverty issues, engaging with city government.
Performance (early/mid-1990s to 2000/01) positive; modest; o neutral; poor; bad. Blank indicates no information available from the city study. Colombo o (1980s) o
Recife o
1 2 3 4 5 6 7 8 9 10 11 12 13 14 o
Joburg o o o o
Cebu or o
Visak o o o o o
Mombasa o o
Kumasi o o
CONCLUSIONS 189
What political and institutional systems, processes and mechanisms, both formal and informal, result in inclusive and pro-poor decisions and outcomes?
likely to damage the fragile economic position of those in poverty by displacing them from key locations. For some cities, the story has been one of economic stagnation and even decline, further reducing the economic opportunities for the poor. In such circumstances, labour-intensive public works schemes, such as those adopted by the AGETUR programme in Benin and elsewhere, can provide temporary employment for certain sections of the labour force as well as creating valuable infrastructure. But there are questions about the sustainability of such arrangements in the absence of donor funding, and about the non-involvement of either city government or organizations representing the poor. For those operating in the local economy or informal sector, secure location is a critical factor. This is an aspect in which city governments can make a difference, by ensuring tenure security and access to essential services. In practice, city governments all too often adopt approaches that are counter-productive: failing to tackle the shortcomings of land tenure and administration systems, and ignoring the informal sector or actively harassing it. There are, of course, valid objectives for regulation of business and trading activity, and conflicts of interest over urban space have to be resolved. But all too often, action by city government reflects the interests of the powerful, as well as corruption on the part of those charged with enforcement. Such bad governance can have a devastating impact on those living on the edges of survival. The other critical role for city government is in the provision of basic infrastructure and services for enterprises and residents. Although some of the key services may lie outside the responsibility of city government, there is still much that the city authorities can do to ensure access by the poor. This includes extending environmental services (or negotiating with other agencies to extend services) to where the poor live and work. It also involves ensuring that bulk infrastructure capacity is sufficient for local services actually to work (for example, for water to flow through taps in poor neighbourhoods). Furthermore, it requires that pricing policies for urban services provide sufficient resources to maintain and extend the system without preventing access by the poor.
CONCLUSIONS 191
Notable examples emerged in Cebu (the organization of street vendors and trisikad drivers) and in Johannesburg (both the civics prior to the ending of apartheid, and more recently Community Development Forums (CDFs) in places like Diepsloot), but also to some extent in Ahmedabad and Bangalore. Tactics are also important: recent progress in Cebu seems to have come with a shift from confrontation to negotiation, while the Bangalore study highlights the informal and covert tactics used by poor groups to protect or improve their position, characterized as politics by stealth. But the case studies also make clear the fragility of initiatives by organizations of the poor. Community or grassroots organizations are vulnerable to multiple problems: conflicting interests; unrepresentative and unaccountable leadership; co-option or undermining by other political interests; sudden changes in political support; and exhaustion of those in leadership who themselves struggle daily with acute poverty. NGOs can, in principle, play a valuable role in supporting and articulating the interests of the urban poor. But again the experience is mixed. In Cebu, there are many active and committed NGOs that support poor groups and have at least some influence within city government. A valuable example was the role that a federation of community organizations and NGOs played in scrutinizing and publicizing the policies of mayoral candidates towards the poor. There were also some positive examples from other cities. However, most of the NGOs identified were more concerned with delivering welfare programmes (sometimes with state sponsorship) than with lobbying the institutions of city governance for changed policies towards the poor. NGOs are often regarded by the poor as opportunistic and self-serving, unwilling or unable to offer real solidarity with them, and without much clout within city government. There have, however, been some notable successes by NGOs at the national level in securing programmes that benefit poor people locally. For the poor to bring their influence to bear, there have to be avenues into decision-making processes. These avenues are more open in some cities than in others, due to variations in political culture and history, as well as the way the representative political system is organized. Ward-level councillors can offer a vital avenue, as indicated most clearly in Bangalore, since those representing poor wards have to demonstrate some responsiveness to their constituents. However, the effectiveness of that avenue depends greatly on the influence that councillors have within city government not to mention the powers and resources of city government to do anything about their demands. This influence in turn depends on the nature of the political system in that city. Directly elected mayors unlike appointed city bosses also have to demonstrate a degree of responsiveness to the poor, who generally comprise the majority of the citys population. The poor may be able to make gains through persistent lobbying and pressure. Persistence is potentially an asset for the poor, where they have a vital interest in the issue at stake. Vote banks and vote bargaining, often perceived as forms of political corruption, can provide ways for the poor to organize to ensure that candidates respond to their needs. But the danger with all these arrangements is clientelism. Clientelistic relationships with those in authority can (and do in several of our case cities) deliver some real benefits for the poor, but they do so on the basis of
dependence and the granting of favours rather than on the basis of regular provision and rights. The case cities also provide some examples of participatory processes which give some voice to the poor: the Community Action Planning system in Colombo in the 1980s; the Integrated Development Planning (IDP) process in Johannesburg; and most notably, participatory budgeting (PB) in Recife. While these arrangements have enabled the poor to have some influence over decisions and resource use, they have serious limitations. They are prone to control by more powerful interests (including professionals in the case of Johannesburg) and are vulnerable to changing political regimes (as in Colombo and Recife) or inadequate financial resources (in Johannesburg and Recife). While PB in Recife has shifted resources somewhat in favour of the poor, and has greatly increased their participation, it remains dependent on the commitment of the mayor. Where PB involves the elected councillors, and where it is incorporated into local legislation, as in some other cities in Brazil, it gives greater assurance to the process and helps to formalize the rights of the poor. Another possibility for greater voice and access by the poor is through democratically elected sub-city levels of government. Where the sub-city level is truly local (ie neighbourhood based, like the barangays in Cebu), has a statutory basis (so that it cannot just be discarded like the Community Development Councils (CDCs) in Colombo) and where it has real resources, then the poor have some prospect of being able to influence decisions that affect them. Alas, in other cities local government is very remote and there is either no provision for a sub-city level or it has not been implemented. For example, in Kumasi, an elaborate structure of sub-city governance was never allowed to function.
CONCLUSIONS 193
relation to directly elected mayors and indirectly elected mayors. What is clear, though, is that the external appointment of mayors, as in Kumasi, breaks the accountability link to local citizens. Even so, the Indian experience of appointed municipal commissioners can have positive outcomes, including the willingness to take necessary but unpopular measures, as in Ahmedabad. It is clear that the detailed arrangements for representative democracy matter, and must be tailored to the particular political and cultural conditions of that city. It is also apparent that periodic elections are not sufficient to ensure that decisions reflect the interests of citizens, particularly the poor. Where elections are supplemented by other, more participatory arrangements (some examples of which have just been mentioned), it is possible to produce more nuanced perceptions of citizens needs and priorities. However, the risk with such arrangements is that the voices of the poor are drowned out by the voices of the more articulate and powerful. In all this, information is critical, both for effective citizen participation and for accountability of elected and appointed officials to those they are supposed to be serving. Yet, in most of the cities, information particularly about the use of financial resources seems to be regarded as something to be withheld rather than disseminated. In some cities, the media (and particularly local radio in some places) play an important role in making citizens aware of issues, although all too often the media lack the capacity for investigation or are co-opted by other powerful interests. However well or badly the democratic decision-making processes work in a city, the outcomes for the poor depend ultimately on the ability of city government and related institutions to deliver on those decisions. In many of the ten cases studied, city governments are heavily constrained: by limited jurisdictions that exclude many of the poor; by fragmentation of responsibilities for infrastructure and services; by conflicts with other levels of government; and by inadequate skills and weak management capacity. Perhaps the greatest constraint is financial. Many of the case cities are in a poor financial condition, with limited and inelastic local revenue sources, weak revenue collection systems and poor financial management. Some cities, such as Cebu, Recife and Kumasi, have benefited from substantial transfers of resources from the centre. Others, like Johannesburg and Ahmedabad, have potentially huge revenue bases but face many difficulties in securing the revenues due to them. Some cities have made progress in improving their financial situation (for example, Ahmedabad), while others, such as Mombasa and Kumasi, remain in a financial mire, unable to improve services to anyone, let alone the poor. On the whole and not surprisingly richer cities are able to finance better services than poorer ones. For these cities, notably Santiago, Recife and Johannesburg, the critical issue is the equitable distribution of resources so as to ensure that disadvantaged sections of the population benefit. This is often hindered by outdated and perverse jurisidictional boundaries, as well as by inherited bureaucratic culture. If cities are really to address poverty and exclusion, they require democratic and participatory processes that move beyond clientelistic arrangements in which benefits are distributed as favours, to a situation in which services are delivered as a matter of routine and right. This requires effective, participatory and accountable democratic systems in which the voice of the poor can be
heard. It also requires an active civil society that can enable the poor to organize and articulate their needs. Furthermore, it requires city level government systems that are properly staffed, managed and financed, with transparent and accountable processes. And last but not least, it requires leadership, both political and managerial, which is responsive to the needs of the poor, has vision and has the capacity to overcome some of the constraints that city governments face. This is a tall order. Among the case cities, none matched up to these requirements. Nevertheless, there were some positive examples of movements in the right direction in several of the case study cities.
CONCLUSIONS 195
be recognized. Formalizing such organizations may reinforce exclusion, especially of minorities and the poorest, and may obscure the competing and conflicting interests within low-income communities. Federating and networking GROs/CBOs at the city level enables them to work more effectively, and helps to prevent officials and politicians from manipulating community leaders, so long as these organizations maintain close links with the people they are supposed to represent. Trade unions, associations of small-scale producers and other civil society organizations can all play a valuable role in aggregating the interests of poor groups in order to lobby or bargain for improvements, or to provide common services; but they often do so in ways which exclude nonmembers and potentially disadvantage the poorest and most vulnerable. 5 NGOs The local knowledge and experience that NGOs possess can make them important partners, but they have limitations in terms of scale and local accountability, and they should not be used as gatekeepers to the poor. While some NGOs have had a beneficial influence on policy at local and/or national level, many are poorly equipped for an advocacy role and may have weak links with the communities for whom they claim to speak. 6 Strategies for GROs/CBOs and NGOs Moving beyond confrontation to the politics of engagement with city government may offer benefits for the urban poor. This may take the form of joint committees of community representatives, NGOs, councillors and local government officials, which offer the potential for developing inclusive and pro-poor policies and practices. But such strategies also carry the risk of cooption and disempowerment. In particular, adopting the role of contractor for government-initiated programmes may create dependence on the state and reduce NGOs freedom of manoeuvre. On the other hand, it has the potential for high-profile advocacy and institutionalizing more participatory approaches to the delivery of such programmes. 7 Political relationships Political relationships at city level are complex, informal as well as formal, and frequently opaque, but they can and often do deliver some benefits for the poor. This is more likely when electoral politics is combined with the practice of vote bargaining. However, political relationships are often clientelistic, reinforcing the dependency of the poor. Interventions to address urban poverty need to recognize these complex political relationships and avoid imposing simplistic solutions that undermine those arrangements that currently deliver at least some benefits for the poor. 8 Democratization and claim-making Democratization at the local level has widened the scope for the urban poor to make their claims, and for urban politics to move beyond clientelism towards more open political bargaining. While, almost by definition, the poor have less influence over the levers of power than the better off, by their sheer numbers
and by their persistence they can and do make their voice heard and thus achieve some gains. 9 The design of the city-level political system The design of the city-level political system is important, including arrangements for electoral representation and executive control. While it is clear that no one model of urban government is necessarily superior to any other, some observations can be made. An elected mayor has to demonstrate a degree of responsiveness to the poor majority in order to get elected, whereas an indirectly elected or centrally-appointed mayor does not. However, conflict between a directly elected mayor and councillors (especially if the majority belong to a different party) can hinder decision-making. A centrally-appointed mayor can over-ride the wishes of locally elected representatives. Elected councillors representing poor wards can provide an avenue for influence for their poor constituents and must demonstrate a degree of responsiveness to them, whereas election by proportional representation (PR) on a party list basis gives power to the party machine. However, PR systems are likely to give a more balanced representation, especially to women and minority groups. Thus, it may be desirable to have a mixed system of ward councillors and councillors elected by PR, or some system of reserved seats for particular groups. Multi-partyism can offer opportunities which civil society groups, including the urban poor, can use to bargain for improvements. However, in many contexts, multi-partyism reduces local politics to factionalism, reinforcing ethnic or religious rivalries. The detailed design of the democratic system matters, including the spatial scales of representation, the checks and balances between executive and legislature, and the relationship with higher levels of government; but it also depends on the particular local context, so that specifying general rules is inappropriate. 10 Mechanisms of participation and accountability Periodic elections alone are not a sufficient mechanism to ensure that decisions reflect the needs and priorities of local citizens. They need to be complemented by a range of mechanisms of direct, deliberative and participatory democracy that enable citizens to have a voice in the decisions that affect them. Participatory budgeting (PB) offers a valuable example, although not without its weaknesses. The design of these mechanisms needs to ensure that the poor are not excluded or drowned out by the better off and more articulate. Elections also need to be accompanied by effective and accessible mechanisms for holding elected representatives and officials accountable. In this, the availability of information, for example about the resources available and their use, together with the transparency of decision-making, are crucial for challenging patronage and dependency. 11 The media The media can play a vital role in disseminating and interpreting information, and local radio can be particularly important in providing accessible information to a non-literate population. However, the lack of investigative reporting skills, combined with control by powerful interests, compromise its effectiveness.
CONCLUSIONS 197
12 Traditional authorities Chiefs and other traditional authorities are still important in certain places. Although generally part of a system of hierarchical and patriarchal social relationships, and frequently resented as being not democratically accountable, they can provide a countervailing force and an alternative avenue of influence for the poor where local government performs badly. In certain cases, they may facilitate access to land for the poor. 13 Decentralization Decentralization in many countries has focused attention on city government for addressing the needs of the poor. However, most city governments face severe constraints in terms of both capacity and vision about how to achieve this. Decentralization of responsibilities and resources, combined with democratization, can in the right circumstances result in greater efficiency and responsiveness of city government, bringing benefits for the urban poor. But there is nothing automatic about this, and cities need the right political and institutional arrangements, powers and resources if they are to achieve such improvements. National governments remain responsible for ensuring the right governmental structures and the macro-economic framework in which city economic development and poverty reduction can take place. 14 Functions and boundaries of city government City governments should be strengthened, rather than assigning important functions to higher levels of government where accountability is more remote, or to separate agencies over which there is little or no democratic control. Where operational efficiency requires a separate agency, lines of accountability to democratically elected local representatives should be clear. The boundaries of city government should include the peripheral areas where the poor reside, so that city resources can be shared equitably (although there may need to be more than a single tier of city government, especially in large cities). 15 Sub-city levels of government Sub-city levels of government can help to bring decision-making closer to citizens and be of benefit to poor neighbourhoods, providing that such units are at a sufficiently local level, are democratically elected, have an equitable share of resources, and have some constitutional protection from sudden changes in political favour. 16 Addressing the constraints on city government If city governments are to be effective in addressing the needs of the poor, it will be necessary to tackle a number of constraints on their operations. Specific reforms include: better city-level information on poverty, environmental conditions and access to services; overcoming the institutional and legal obstacles, such as out-dated and inappropriate bylaws and regulations that inhibit informal sector businesses and livelihood opportunities for the poor and prevent services being provided in informal settlements; upgrading technical and managerial skills among staff, improving management systems, and encouraging more responsive
and pro-poor attitudes and practices on the part of local officials; extending local revenue sources, improving systems of revenue collection and providing equitable inter-governmental transfers; establishing more transparent budgeting systems, expenditure management and monitoring, so as to make better use of available resources; and improving relationships with higher levels of government so that the latter can be supportive rather than undermining, and can enhance local level performance as well as ensuring local accountability. 17 Civic leadership The personal qualities of civic leaders both elected and appointed can make a significant difference to achieving effective and pro-poor city governance. This requires an environment that encourages dynamic, responsible and responsive civic leadership to emerge, and an institutional framework that ensures that the poor benefit not just through favours but as a matter of routine. This in turn requires civic education for all those involved: officials, elected representatives and citizens. 18 Access to land, infrastructure and services The ways in which city governments can have the greatest impact on poverty reduction are generally through ensuring access to land for housing and economic activities, and through the direct provision of infrastructure and services. This includes the maintenance of infrastructure something too often neglected as well as initial capital investment. Such interventions can increase income opportunities for the poor as well as reducing costs, time losses and risks to health. Even where municipal governments do not have direct control (for instance, in relation to land), they have levers of influence (eg willingness to legalize settlements and regularize tenure, as well as adopting realistic zoning, land development regulations and building codes). Since city governments have limited resources, they need to redirect those resources towards the infrastructure and service needs of the poor, while at the same time ensuring that the bulk capacity is there to support the services at the point of delivery. 19 Implications for funding agencies A number of specific implications for funding agencies emerge from the study. These include: the need to engage with the institutions of city governance, not just with national governments, and to channel resources through local-level institutions not just national institutions; the need to build on whatever is already happening locally that is positive rather than bypassing existing institutions, processes and networks (while at the same time recognizing that community organizations may not be fully representative or inclusive); the importance of undertaking a political appraisal as part of project design, in order to understand the complexity of local political processes and competing interests; the need to fund citywide infrastructure rather than just isolated projects, and to design such interventions in ways which take account of the interests and livelihood systems of the urban poor; and finally, the importance of long-term engagement in order to facilitate the transformation of city governance.
CONCLUSIONS 199
It must, of course, be acknowledged that changes such as those envisaged here are neither easy nor automatic, and are likely to take time to achieve. They may require radical changes in the local and national political climate. They will also require greater international resources to be directed at reducing urban poverty. For those operating in particular cities in the South at the present time, these changes and requirements may seem unrealistic. Nevertheless, as at least some of the city case studies show, progress is possible.
Annex 1
Research Methodology
This research was carried out between the beginning of 1998 and early 2001. The research questions identified in the initial project document were wideranging: How can urban governance influence the conditions for urban economic growth? How do urban governance institutions seek to distribute the benefits of that growth, how this works in practice, and who benefits from these processes of distribution? How can and do the poor influence the agenda of the institutions of city governance in their interests?
Stage 1 of the research (1998 to mid-1999) involved: a review of the literature and international experience according to seven key themes identified as being critical to the research area (see the list of Theme Papers in Annex 2); a series of nine city case studies, carried out by local researchers in each of these cities, according to a research guide developed by the UK team; the research involved assembling and analysing available data, together with selected key-informant interviews, rather than substantial original research;
ANNEX 1 201
an analysis of this material across the nine city case studies according to the key themes; a workshop with all the researchers at which the results were discussed.
In Stage 2 (late 1999 to early 2001), the research focused on a more specific question: What sorts of political and institutional processes, systems and mechanisms, both formal and informal, result in pro-poor and inclusive decisions (and outcomes)? (Outcomes were included in brackets since it is outcome, not just decisions, which are ultimately of interest, but recognizing that it would not be possible to demonstrate causal relationships in terms of outcomes.)
From this, a series of more detailed research questions were formulated. Stage 2 included: identification of four of the original city cases (Bangalore, Cebu, Johannesburg and Kumasi) which appeared to offer the most interesting and fruitful insights for the research agenda; a further case was added (Recife), focusing on the participatory budgeting programme; a workshop with city researchers to discuss the research questions and methodology, and to identify key sectors and key groups for detailed study in each city; field research in the five cities, in each case involving the collection of aggregate data, analysis of city finances, interviews with key informants in various sectors, and focus group discussions, using techniques such as community profiling, oral histories, time lines, critical incident analysis and Venn diagrams. In each city, detailed case studies were carried out in two or three poor residential neighbourhoods and with two or three identified poor or excluded groups (eg street traders). Particular attention was paid to three sectors in each city: land and housing, water and sanitation, regulation of the informal sector; a concluding workshop at which the second round case studies were discussed and a cross-city analysis presented according to the main themes.
Some additional studies were also commissioned in Stage 2: on the AGETUR/AGETIP model of labour-intensive public works in Benin; an overview of participatory budgeting in Brazil, focusing on three cities, for comparison with the Recife case; a review of another series of case studies carried out by IIED of the experiences of innovatory urban poverty reduction programmes in eight municipalities in Africa, Asia and Latin America.
Annex 2 lists the theme papers, city case studies, cross-city analyses and additional cases published as Working Papers under this research.
Annex 2
Urban Governance, Partnerships and Poverty Research Working Papers
Theme papers
1 2 3 4 5 6 7 8 City Economic Growth Elizabeth Vidler Urban Economic Growth and Poverty Reduction Philip Amis Households, Livelihoods and Urban Poverty Jo Beall and Nazneen Kanji Who Runs Cities? The Relationship between Urban Governance, Service Delivery and Poverty Nick Devas Civil Society and Urban Poverty Diana Mitlin The Urban Environment Fiona Nunan and David Satterthwaite RuralUrban Interactions Cecilia Tacoli Urban Governance, Partnerships and Poverty: A Preliminary Exploration of the Research Issues Carole Rakodi
ANNEX 2 203
Additional studies
28 Participatory Budgeting in Brazilian Cities: Limits and Possibilities in Building Democratic Institutions Celina Souza 29 Poverty Reduction and Employment Generation: The Case of AGETUR, Benin Blandine Fanou with Ursula Grant 30 Urban Politics and Governance: A Review of the Literature Carole Rakodi 31 Reducing Urban Poverty: Drawing Lessons from Recent Urban Programmes David Satterthwaite
These papers can be ordered from: The Publications Office School of Public Policy University of Birmingham Birmingham B15 2TT United Kingdom Tel: +44 (0)121 414 4986 Fax: +44 (0)121 414 4989 Email: C.A.Fowler@bham.ac.uk Price: 5 for the theme papers (papers 18), 10 for all others. Alternatively, the papers can be viewed on our website: www.idd.bham.ac.uk/research/Projects/urban-governance/urbgov.htm Articles based on the Stage 1 city case studies were also published in Environment and Urbanization 21 (1), April 2000. Articles reviewing and comparing the results of Stage 1 studies were published in International Planning Studies 6 (1), 2001.
References
Abers, R (1998) Learning democratic practice: distributing government resources through popular participation in Porto Alegre, Brazil, in Douglass, M and Friedmann, J (eds) Cities for Citizens, John Wiley and Sons, Chichester Abugre, C and Holland, J (1998) A Poverty Analysis in Support of Project Preparation in Kumasi. Ghana: Water Sector Improvement Project Amis, P (1999) Urban economic growth and poverty reduction. Urban Governance, Partnership and Poverty Working Paper 2, IDD, University of Birmingham, Birmingham Amis, P (2001) Rethinking UK aid: lessons from an impact assessment study of DFIDs Slum Improvement projects. Environment and Urbanization 13(1): 101103 Amis, P (2002) Municipal government, urban economic growth and poverty reduction: identifying the transmission mechanisms between growth and poverty, in Rakodi, C with Lloyd-Jones, T (eds) Urban Livelihoods: A People-centred Approach to Reducing Poverty, Earthscan, London, pp97111 Amis, P and Grant, U (2000) Urban economic growth, governance and poverty: a comparative city review. Urban Governance, Partnership and Poverty Working Paper 18, IDD, University of Birmingham, Birmingham Amis, P and Grant, U (2001) Urban economic growth, civic engagement and poverty reduction. Journal of International Development 13: 9971002 Amis, P and Kumar, S (2000) Urban economic growth, infrastructure and poverty in India: lessons from Visakhapatnam. Environment and Urbanization 12(1): 185196 Amis, P and Rakodi, C (1995) Urban poverty: concepts, characteristics and policies. Habitat International 19(4): 403405 AMREF and OVP/MPND (African Medical Relief Fund and Office of the VicePresident/Ministry of National Planning and Development) (1997) The Second Participatory Assessment Study: Kenya, Nairobi, AMREF and OVP/MPND Arrossi, S, Bombarolo F, Hardoy, J E, Mitlin, D, Coscio, L P and Satterthwaite, D (1994) Funding Community Initiatives, Earthscan, London Bangura, Y (2000) Democratization, equity and stability: African politics and societies in the 1990s, in Ghai, D (ed) Reviewing Economic and Social Progress in Africa: Essays in Memory of Philip Ndegwa, Macmillan Press, Basingstoke Barrera, M (1999) Political participation and social exclusion of the popular sectors in Chile, in Oxhorn, P and Starr, P K (eds) Markets and Democracy in Latin America: Conflict or Convergence, Lynne Rienner, Boulder CO, pp81102 Beall, J (1995) Social security and social networks among the urban poor in Pakistan. Habitat International 19(4): 427445 Beall, J (1997) Social capital in waste: a solid investment? Journal of International Development, 9(7): 951961 Beall, J (2000a) Valuing social resources or capitalizing on them? Social assets and the limits to pro-poor urban governance. Urban Governance, Partnership and Poverty Working Paper 19, IDD, University of Birmingham, Birmingham Beall, J (2000b) Life in the cities in Allen, T and Thomas, A (eds) Poverty and Development (2nd edn), Oxford University Press in Association with the Open University, Oxford and Milton Keynes, pp425442 Beall, J (2001) Valuing social resources or capitalizing on them? Limits to pro-poor urban governance in nine cities of the South. International Planning Studies 6(4): 357375
REFERENCES 207
Neo-Liberalism and Urban Strategies in Developing Countries, Zed Books, London Burgwal, G (1995) Struggle of the Poor: Neighbourhood Organization and Clientelist Practice in a Quito Squatter Settlement, Centre for Latin American Research and Documentation, Amsterdam Carroll, T F (1992), Intermediary NGOs: The Supporting Link in Grassroots Development, Kumarian Press, West Hartford Castells, M (1979) The Urban Question: A Marxist Approach, Edward Arnold, London Castells, M (1983) The City and the Grass Roots: A Cross-cultural Theory of Urban Social Movements, Edward Arnold, London Castells, M (2002) Reader on Cities and Social Theory, edited by I Susser, Blackwell, Oxford Chalker, L (1991) Good government and the aid programme, Speech by the Rt Hon Lynda Chalker MP, Minister for Overseas Development, at the Royal Institute of International Affairs, 25 June Chambers, R (1995) Poverty and livelihoods: whose reality counts? Environment and Urbanization 7(1): 173204 Chambers, R and Conway, G (1992) Sustainable Rural Livelihoods: Practical Concepts for the 21st Century, IDS Discussion Paper No. 296, Institute of Development Studies, Brighton Chant, S (1998) Households, gender and ruralurban migration: reflections on linkages and considerations for policy. Environment and Urbanization 10(1): 521 Chant, S (2002) The informal sector and employment in Desai, V and Potter, R (eds) The Companion to Development Studies, Edward Arnold, London, pp206213 Chazan, N, Mortimer, R, Ravenhill, J and Rothchild, D (1992) Politics and Society in Contemporary Africa, Lynne Rienner, Boulder CO Choguill, C (1996) Ten steps to sustainable infrastructure. Habitat International 20(3): 389404 Christian Aid (1993) A review report of Christian Aid/Joint Funding Scheme Urban Community Development Projects in Thailand, Chile and Bolivia. Submitted to the ODA Joint Funding Scheme, November 1993, Christian Aid, London Clark, J (1991) Democratizing Development: The Role of Voluntary Agencies, Earthscan, London Craig, D and Porter, D (2003) Poverty reduction strategy papers: a new convergence. World Development 31(1): 5369 Crankshaw, O (1997) Shifting sands: labour market trends and union organisation. South African Labour Bulletin 21: 2835 Crook, R and Manor, J (1998) Democracy and Decentralisation in South Asia and West Africa: Participation, Accountability and Performance, Cambridge University Press, Cambridge Crook, R and Sverrisson, A (2001) Decentralization and Poverty Alleviation in Developing Countries: A Comparative Analysis, or is West Bengal Unique?, IDS Working Paper 130, Institute of Development Studies, Brighton Dahl, R A (1961) Who Governs? Democracy and Power in an American City, Yale University Press, New Haven CT Das, V (1996) The spatialization of violence: case study of a communal riot , in Basu, K and Subrahmanyam, S (eds) Unravelling the Nation, Penguin, New Delhi, pp157203 Davey, K (1996) The structure and functions of urban government in Davey, K with Batley, R, Devas, N, Norris, M and Pasteur, D, Urban Management: The Challenge of Growth, Avebury, Aldershot, pp47102 Davey, K, with Batley, R, Devas, N, Norris, M, and Pasteur, D (1996) Urban Management: The Challenge of Growth, Avebury, Aldershot
REFERENCES 209
Etemadi, F (2001) Towards inclusive urban governance. Urban Governance, Partnership and Poverty Working paper 25, IDD, University of Birmingham, Birmingham Evans, P (1996) Government action, social capital and development: reviewing the evidence on synergy. World Development 24(6): 11191132 Fainstein, S and Hirst, C (1995) Urban social movements, in Judge, D, Stoker, G and Wolman, H (eds) Theories of Urban Politics, Sage, London, pp181204 Fanou, B and Grant, U, (2001) Poverty reduction and employment generation: the case of AGETUR, Benin. Urban Governance, Partnership and Poverty Working Paper 29, IDD, University of Birmingham, Birmingham Farvacque-Vitkovic, C and Godin, L (1998) The Future of African Cities: Challenges and Priorities for Urban Development, World Bank, Washington, DC Fernando, A, Russell, S, Wilson, A and Vidler, E (1999) Urban governance, partnerships and poverty in Colombo. Urban Governance, Partnership and Poverty Working Paper 9, IDD, University of Birmingham, Birmingham Foley, M and Edwards, B (1999) Is it time to disinvest in social capital? Journal of Public Policy 19(2): 141173 Fowler, A (1997) Striking a Balance, Earthscan, London Freire, M and Stren, R (eds) (2001) The Challenge of Urban Government: Policies and Practices, World Bank Institute, Washington, DC Friedmann, J (1998) The new political economy of planning: the rise and fall of civil society, in Douglass, M and Friedmann, J (eds) Cities for Citizens: Planning and the Rise of Civil Society in a Global Age, Wiley, Chichester, pp1935 Frigenti, L et al, (1999) Local Solutions to Regional Problems: The Growth of Social Funds and Public Works Employment Projects in Sub-Saharan Africa, World Bank, Washington, DC Gatabaki-Kamau, R with Rakodi, C and Devas N (1999), Urban governance, partnerships and poverty in Mombasa. Urban Governance, Partnership and Poverty Working Paper 11, IDD, University of Birmingham, Birmingham Gazzoli, R (1996) The political and institutional context of popular organizations in urban Argentina. Environment and Urbanization 8(1): 159166 Gilbert, A (2000) Housing the Third World cities: the critical issues. Geography 85: 145155 Gilbert, A and Gugler, J (1992) Cities, Poverty and Development: Urbanization in the Third World, Oxford University Press, Oxford GJMC (1999) iGoli 2002: Making the City Work, Greater Johannesburg Metropolitan Council, Johannesburg Goetz, A M and Gaventa, J (2001) Bringing Citizen Voice and Client Focus into Service Delivery, IDS Working Paper 128, Institute of Development Studies, Brighton Gorman, R F (1984) Private Voluntary Organizations as Agents of Development, Westview Press, Boulder and London Government of Ghana (1995) Ghana Living Standard Surveys, Accra Grindle, M S (1980) Politics and Policy Implementation in the Third World, Princeton University Press, Princeton NJ Grindle, M S and Thomas, J A (1990) After decision: Implementing policy reforms in developing countries. World Development 18(8): 11631181 Grootaert, C, Kanbur, R and Oh, G-T (1995) The dynamics of poverty: why some people escape from poverty and others dont. An African case study. World Bank Working Papers No 1499, Washington, DC Gupta, S, Davoodi, H and Alonso-Terme, R (1998) Does Corruption Affect Income Inequality and Poverty? Fiscal Affairs Department, International Monetary Fund, Washington, DC Haddad, L, Ruel, M T, and Garrett, J l, (1999) Are urban poverty and undernutrition growing? Some newly assembled evidence. World Development 27(11): 18911904
REFERENCES 211
Hyden, G (1992) Governance and politics in Africa, in Hyden, G and Bratton, B (eds) Governance and Politics in Africa, Lynne Rienner, Boulder CO, pp126 Hyden, G and Bratton, B (eds) (1992) Governance and Politics in Africa, Lynne Rienner, Boulder CO Hyden, G, Oluwu, D and Ogendo, H (eds) (2000) African Perspectives on Governance, Africa World Press, Trenton NJ Imbroscio, D L (1997) Reconstructing City Politics: Alternative Economic Development and Urban Regimes, Sage, Thousand Oaks CA International Labour Organization (1972) Employment, Incomes and Equality in Kenya, ILO, Geneva Jacobs, J (1961) Death and Life of Great American Cities, Random House, New York Jacobs, J (1984) Cities and the Wealth of Nations, Penguin, Harmondsworth Jenkins, P (2001) Strengthening access to land for housing for the poor in Maputo, Mozambique. International Journal of Urban and Regional Research 25(3): 629648 Jerve, A M (2001) Ruralurban linkages and poverty analysis, in Grinspun, A (ed) Choices for the Poor: Lessons from National Poverty Strategies, United Nations Development Programme, New York, pp89120 Johnson, C and Start, D (2001) Rights, claims and capture: understanding the politics of pro-poor policy, Working Paper 145, Overseas Development Institute, London Jones, S (1999) Multi-layered responses to urban poverty: possible donor agency action, in Jones, S and Nelson, N (eds) Urban Poverty in Africa: From Understanding to Alleviation, Intermediate Technology Publications, London Jones, S and Nelson, N (eds) (1999) Urban Poverty in Africa: From Understanding to Alleviation, Intermediate Technology Publications, London Judge, D (1995) Pluralism, in Judge, D, Stoker, G and Wolman, H (eds) Theories of Urban Politics, Sage, London, pp1334 Judge, D, Stoker, G and Wolman, H (eds) (1995) Theories of Urban Politics, Sage, London Kanji, N (1995) Gender, poverty and economic adjustments in Harare, Zimbabwe. Environment and Urbanization 7(1): 3756 Kantor, P, Savitch, H V and Haddock S V (1997) The political economy of urban regimes: a comparative perspective. Urban Affairs Review 32(2): 348377 Karaos, A M A, Gatpatan, M V and Hotz, R V S J (1995) Making a difference: NGO and PO policy influence in urban land reform advocacy. PULSO (Pagsusuri Ukol Sa Lipunan At Simbahan), Monograph No. 14, Institute on Church and Social Issues, Manila Kasfir, N (1993) Designs and dilemmas of African decentralization in Mawhood, P (ed) Local Government in the Third World: Experience of Decentralization in Tropical Africa, African Institute of South Africa, pp2448 Kelly, R and Devas, N (2001) Revenue or revenues? An analysis of local business licences, with a case study of the Single Business Permit reform in Kenya. Public Administration and Development 21: 381391 Khan, A M (1997) Shaping Policy: Do NGOs Matter? Lessons from India, PRIA (Participatory Research in Asia), New Delhi King, K (1996) Jua Kali Kenya: Change and Development in an Informal Economy 19701995, James Currey, London King, R, Inkoom, D and Abrampah, K M (2001) Urban governance in Kumasi. Urban Governance, Partnership and Poverty Working Paper 23, IDD, University of Birmingham, Birmingham Klaarhamer, R (1989) The Chilean squatter movement and the state, in Schuurman, F and van Naerssen, T (eds) Urban Social Movements in the Third World, Routledge, London and New York, pp177197
REFERENCES 213
Mills, A, Bennet, S and Russell, S (2001) The Challenge of Health Sector Reform: What Must Governments Do? Basingstoke, Palgrave Mills, E S and Pernia, E M (1994) Introduction and overview, in Pernia, E M (ed) Urban Poverty in Asia: A Survey of Critical Issues, Oxford University Press, Oxford Mitlin, D (1999) Civil society and urban poverty. Urban Governance, Partnership and Poverty Working Paper 5, IDD, University of Birmingham, Birmingham Mitlin, D (2000) Civil society and urban poverty: overview of stage 1 cases. Urban Governance, Partnership and Poverty Working Paper 22, IDD, University of Birmingham, Birmingham Mitlin, D (2001a) Addressing urban poverty: increasing incomes, reducing costs, securing representation. Development in Practice 10(2): 204216 Mitlin, D (2001b) Civil society and urban poverty: examining complexity. Environment and Urbanization 13(2): 151174 Moore M (1999) Politics against poverty: Global pessimism and national optimism. IDS Bulletin 30(2): 3346 Moser, C O N (1978) Informal sector or petty commodity production: Dualism or dependence in urban development? World Development 6(910): 10411064 Moser, C O N (1996) Confronting crisis: a comparative study of household responses in four poor urban communities. Environmentally Sustainable Development Studies and Monograph Series no.8, World Bank, Washington, DC Moser, C O N (1998) The asset vulnerability framework: reassessing urban poverty reduction strategies. World Development 26(1): 119 Moser, C O N and Holland, J (1995) A Participatory Study of Urban Poverty and Violence in Jamaica, Urban Development Division, World Bank, Washington, DC Moser, C O N and Holland, J (1997) Household responses to poverty and vulnerability (Volume 4): confronting crisis in Chawama, Lusaka, Zambia. Urban Management Programme Policy Paper No. 24, UMP, Nairobi Moser, C O N and McIlwaine, C (1997) Household responses to poverty and vulnerability (Volume 3): confronting crisis in Commonwealth, Metro Manila, The Philipines. Urban Management Programme Policy Paper No. 23, World Bank, Washington, DC Moser, C O N and McIlwaine, C (2001) Violence and social capital in urban poor communities: perspectives from Colombia and Guatemala. Journal for International Development 13: 965984 Moser, C O N and Peake, L (eds) (1987) Women, Human Settlements, and Housing, Tavistock, London Moser, C O N, Herbert, A J and Makonnen, R E (1993) Urban poverty in the context of structural adjustment: recent evidence and policy responses, TWU Discussion Paper DP4, Urban Development Division, World Bank, Washington, DC Natrajan, I (1998) Market Demographics, National Council of Applied Economic Research, New Delhi Nickson, A (1995) Local Government in Latin America, Lynne Rienner, Boulder CO Nunan, F and Satterthwaite, D (1999) The urban environment. Urban Governance, Partnership and Poverty Working Paper 6, IDD, University of Birmingham, Birmingham Nunan, F and Satterthwaite, D (2000) Governance and the urban environment: a comparative analysis of nine city case studies. Urban Governance, Partnership and Poverty Working Paper 21, IDD, University of Birmingham, Birmingham Oates, W (1972) Fiscal Federalism, Harcourt Brace Jovanovitch Inc, New York OECD (Organisation for Economic Co-operation and Development) (1988) Voluntary Aid for Development: the Role of Nongovernmental Organizations, OECD, Paris
REFERENCES 215
Rakodi, C (1999a) A capital assets framework for analysing household livelihood strategies. Development Policy Review 17(3): 315342 Rakodi, C (1999b) Urban governance and poverty: A preliminary exploration of the research issues. Urban Governance, Partnership and Poverty Working Paper 8, IDD, University of Birmingham, Birmingham Rakodi, C (2001) Urban politics and governance: a review of the literature. Urban Governance, Partnership and Poverty Working Paper 30, IDD, University of Birmingham, Birmingham Rakodi, C (2002) A livelihoods approach: conceptual issues and definitions, in Rakodi, C with Lloyd-Jones, T (eds) (2002) Urban Livelihoods: A People-centred Approach to Reducing Poverty, Earthscan, London Rakodi, C with Lloyd-Jones, T (eds) (2002) Sustainable Urban Livelihoods: A People-centred Approach to Reducing Poverty, Earthscan, London, pp322 Rakodi, C, Gatabaki-Kamau, R and Devas, N (2000) Poverty and political conflict in Mombasa. Environment and Urbanization 12(1): 153170 Rao, S L and Natrajan, I (1996) Market Demographics, National Council of Applied Economic Research, New Delhi Rashid, A (1998) Ghaziabad, Orangi, in Arif Hasan (ed) Community Initiatives: Four Case Studies from Karachi, City Press, Karachi Ravallion, M (2001) On the Urbanization of Poverty, Policy Research Working Paper 2586, World Bank, Washington, DC Rhodes, R A W (1997) Understanding Governance: Policy Networks, Governance, Reflexivity and Accountability, Open University, Buckingham Roberts, B (1995) The Making of Citizens, Edward Arnold, London Robinson, M and White, G (1998) Civil society and social provision: the role of civic organisations, in Minogue, M, Polidano, C and Hulme, D (eds) Beyond the New Public Management: New Directions in Latin America, Asia and Africa, Edward Elgar, Cheltenham, pp228245 Rodrguez, A and Winchester, L (1999) Urban governance, partnerships and poverty in Santiago. Urban Governance, Partnership and Poverty Working Paper 14, IDD, University of Birmingham, Birmingham Rodrik, D (2000) Institutions for high quality growth: What are they and how to get them? Comparative International Development 33(3): 231 Roy, A (2001) The Algebra of Infinite Justice, Penguin, New Delhi Ruble, B A, Tulchin, J S and Garland, A M (1996) Introduction: globalism and local realities: five paths to the urban future, in Cohen, M, Ruble, B A, Tulchin, J S and Garland, A M (eds) Preparing for the Urban Future, Global Pressures and Local Forces, Woodrow Wilson Center Press, Washington, DC Rland, J (1992) Urban Development in Southeast Asia: Regional Cities and Local Government, Westview, Boulder CO Rland, J (ed) (1996) The Dynamics of Metropolitan Management in Southeast Asia, Institute of Southeast Asian Studies, Singapore Russell, S (1999) Community Development Councils: grassroots testimony on Community Development Councils: local activists experience and perspectives, in Fernando, A, Russell, S, Wilson, A and Vidler, E, Urban governance and poverty in partnerships. Colombo Urban Governance, Partnership and Poverty Working Paper 9, IDD, University of Birmingham, Birmingham Russell, S and Vidler, E (2000) The rise and fall of governmentcommunity partnerships in urban development: grassroots testimony from Colombo. Environment and Urbanization 12(1): 307316 Rutherford, S (2000) The Poor and their Money, Oxford University Press, New Delhi
REFERENCES 217
Stren, R E and White, R R (eds) (1989) African Cities in Crisis: Managing Rapid Urban Growth, Westview, Boulder CO Stren, R with Kjellberg Bell, J (eds) (1995) Perspectives on the City: Volume 4 of Urban Research in the Developing World, Centre for Urban Community Studies, University of Toronto, Toronto, pp1946 Swilling, M (ed) (1997) Governing Africas Cities, Witwatersrand University Press, Johannesburg Tacoli, C (1998) Ruralurban interactions: a guide to the literature. Environment and Urbanization 10(1): 147166 Tacoli, C (1999) Ruralurban interactions. Urban Governance, Partnership and Poverty Working Paper 7 IDD, University of Birmingham, Birmingham Tanzi, V (1995) Fiscal federalism and decentralization: a review of some efficiency and macroeconomic aspects. Annual World Bank Conference on Development Economics 1995, World Bank, Washington, DC Tendler, J (1997) Good Government in the Tropics, Johns Hopkins University Press, Baltimore and London Thomas, J (1992) Informal Economic Activity, Harvester, Wheatsheaf, London and New York Thorbek, S (1991) Gender in two slum cultures. Environment and Urbanization 3(2): 7181 Thurman, S (1994) Community development in Kenya: a review of the issues and proposals for action, Mimeo, Ford Foundation, Nairobi Tokman, V (1991) The informal sector in Latin America: from underground to legality in G. Standing and V. Tokman (eds) Toward Social Adjustment: Labour Market Issues in Structural Adjustment, International Labour Office (ILO), Geneva, pp141157 Tomlinson, R (1996) The changing nature of Johannesburgs economy in Harris, N and Fabricius, I (eds) Cities and Structural Adjustment, University College Press, London Tovey, K (2002) The institutional responses to the water needs of the urban poor: A study of collective action in Delhi slums, India, Dissertation submitted to the University of Cambridge Turner, B (ed) (1988) Building Community: A Third World Case Book from Habitat International Coalition, Habitat International Coalition, London UNCHS (1996) An Urbanizing World: Global Report on Human Settlements 1996, United Nations Centre for Human Settlements (Habitat)/Oxford University Press, Oxford UN-Habitat (2001a) Cities in a Globalizing World: Global Report on Human Settlements 2001, United Nations Centre for Human Settlements (Habitat)/Earthscan, London UN-Habitat (2001b) Good Urban Governance: A Normative Framework, Summary for the Preparatory Committee for the Special Session of the General Assembly for an Overall Review and Appraisal of the Implementation of the Habitat Agenda, Nairobi, February UNDP (1996) Human Development Report, 1996, United Nations Development Programme, New York UNDP (1997a) Corruption and Good Governance, Management Development and Governance Division, Bureau for Policy and Programme Support, United Nations Development Programme, New York UNDP (1997b) Reconceptualising Governance, Discussion Paper 2, Management Development and Governance Division, Bureau for Policy and Programme Support, United Nations Development Programme, New York UNDP (2000a) Overcoming Human Poverty: UNDP Poverty Report 2000, United Nations Development Programme, New York
Index
absolute poverty 16, 17 access city comparisons 188 employment 57 financial accounts 118 informal politics 77, 78 land 55, 151, 164185, 198 opportunities 46 services 2, 51, 54, 164185, 198 well-being 3 accountability decision-making 163 democratization 3334 elections 176177 mechanisms of 117118 policy 196 public 24, 81, 121 advocacy 86, 137139 Ahmedabad, India city overviews 67, 20, 188 civil society 128, 180182 contracted-out services 179 informal sector 50, 66, 155 labour markets 59, 147 legitimacy 7374 peripheral settlements 97 spending patterns 108 see also India alliances 8889 apartheid 71, 101 assets 22, 149152 attitudes 188 authority aspects 28, 33, 7071, 118, 171, 196197 bad governance 190, 194 Bangalore, India access 151, 168, 169170 accountability 177 city overviews 78, 20, 188 civil society 128, 182 decentralization 74, 7778, 79 economy 4647, 48, 105, 108 high-tech industry 44, 47
informal sector 63, 156157 labour markets 57, 147 responsibilities 98 see also India barangays 8, 84, 86, 108, 119, 120, 141, 163, 167, 192 basic needs 16 see also services Benin 45 boundaries 80, 97, 193, 197 Brazil 19, 31, 7071, 7476, 114 see also Recife, Brazil budgetary processes 34, 110112 see also expenditure; participatory budgeting capacity 4, 100, 176, 188, 190 capital 4, 22, 46, 6466, 149, 150 care economy 6061 casual labour 4446, 59 Cebu City, Philippines access 167168, 173, 174, 175176, 177 city overviews 7, 8, 18, 20, 188 conflicts 114 decentralization 74, 7576, 119, 120 financial resources 105, 108 informal sector 47, 128, 138, 154, 155156 labour markets 43, 147 NGOs 180 participation 141142 political context 89, 161, 163 societal representation 83 chief executives 8182 Chile 7172, 73, 76, 81, 173174 see also Santiago, Chile citizenship 73 civil society advocacy 86 city overviews 6, 182, 188 defining 26 democratization 33, 7172, 93 financial resources 112113
INDEX 221
executive level 8182, 111, 112 see also leadership; mayors expenditure 60, 107112 see also participatory budgeting; spending patterns family support systems 6061 federations 65, 194195 financial resources 39, 40, 100102, 103113, 118, 188, 193 see also revenue formal sector 89, 111, 167, 188 funding 182183, 184 see also support gender 135 Ghana 19, 31, 39, 72, 98, 172 see also Kumasi, Ghana Gini index of income distribution 17 globalization 2728 grassroots organizations (GROs) 124, 127, 129132, 133137, 143, 194195 see also civil society, community-based organizations GROs see grassroots organizations growth coalitions 4951 headcount poverty measures 16, 1819 health 46, 107, 109 high-tech industries 44, 47 home-based enterprises 47 household livelihood strategies 5761 housing access 166167, 169171, 173, 175, 180, 188 livelihood strategies 55 participation 85 spending patterns 109 human capital 46, 149 Hydens framework of governance 2425 inclusivity 3, 188 income 16, 17, 41, 5759 incremental approaches 157158, 170 see also politics by stealth India 19, 3132, 34, 9899, 114115, 161162 see also Ahmedabad; Bangalore; Visakhapatnam indicators of poverty 1617, 167, 187 inequality 2728, 29, 41, 42, 55 informal sector access 167, 184185 alliances 89 government attitudes 188 housing 180 labour markets 4648, 57 location 190 partnerships 50, 138 politics 7778, 79 regulation 145163, 155 relationships 26, 66, 139 responsiveness 118 services 54, 166, 167 settlements 135137, 169, 175 support networks 6164 see also enterprise sector: small-scale enterprises information 100, 192, 194 infrastructure access 170, 198 city comparisons 188 enterprises 150 government responsiveness 176 improvements 110 investments 183 NGOs and CBOs 182 provision 165166, 190 institutional aspects 25, 192 international aspects 4, 16, 2734, 4849 investment 2728, 4849, 183 Johannesburg, South Africa city overviews 7, 9, 20, 188 civil society 128, 135137, 182 decentralization 74 financial resources 100101, 103, 105, 108 informal sector 150, 158159 labour markets 42, 43, 57, 147 livelihood strategies 57, 59 participation 115 peripheral settlements 97 see also South Africa Kenya 19, 31, 39, 7273, 76, 99 see also Mombasa, Kenya Kumasi, Ghana access 168169, 171 accountability 177
INDEX 223
planning 140 policy 141142, 196 political 9193 poverty assessments 18 private sector 177180 participatory budgeting (PB) 104, 111, 115117, 140142 partnerships economic growth 4951 institutionalization 143 private sector 178, 179, 181, 189 responsiveness 9596 social capital 66 patronage 94, 159, 173, 177 PB see participatory budgeting perceptions 17, 18, 5556, 139140, 162 peripheral areas 97, 135137 Philippines 19, 31, 71, 74, 115, 172173 see also Cebu City, Philippines planning 45, 48, 67, 140141, 150151 policy aspects 3031, 4344, 67, 8788, 172175, 194199 political aspects accountability 177 change 6, 2734 city comparisons 188 commitment 91, 113114, 121 community-based organizations 66 conflicts 40, 9899, 196 design 195196 GROs 131132 literature 5 local level 161162, 163 participation 141 relationships 6894, 192, 195 politics by stealth 78, 130131, 157158, 160162, 191 population 2, 7, 1819, 2021 poverty gap indexes 16 poverty lines 16 power 26, 33, 74, 75, 76, 81, 89 see also responsibilities PR see proportional representation pricing policies 172 priorities 30, 5456, 66 private sector 159, 171, 177180 productivity 39, 4446, 51, 55, 150 profitability 148149 property tax 104, 106 proportional representation (PR) 33, 8081, 192, 196 public sector acceptability 162 accountability 24, 81, 121 employment 44, 45 safety 64 space 5152 Recife, Brazil accountability 177 city overviews 7, 1011, 21, 188 financial resources 105, 108 land regularization 179 participatory budgeting 111, 115117 responsiveness 176 see also Brazil reciprocity 24, 6164, 67 regulation 145163, 150, 152153, 185 relationships civil society 94, 126127, 129132, 139, 140, 142143 exploitative 63, 64 household 57 informal 26, 27, 145163 livelihood systems 189 political 192, 195 relative poverty 16, 17 religious organizations 9091 representation 7883, 8384, 133137 reserved seats 33, 81, 192 resilience strategies 56 resources 49, 61, 63, 8486 see also economic aspects; financial resources responsibilities abdication 65, 67 access 171 financial resources 110, 111 governments 7476, 184 services 9798, 165 see also power responsiveness constraints 96103 decentralization 28 democracy 8184, 190, 191 economic growth 43 election of representatives 8184 governments 113120, 175176, 188 increased 93 leadership 163, 193, 198 opportunities 9596, 189 political 196