URBAN LIVELIHOODS AND INTRA-HOUSEHOLD DYNAMICS: THE CASE OF MPUMALANGA AND ENHLALAKAHLE TOWNSHIPS, KWAZULU-NATAL, SOUTH AFRICA by Sarah Mosoetsa A dissertation submitted for the degree of Doctor of Philosophy (PhD) School of Social Sciences, Sociology Department University of the Witwatersrand Johannesburg 2005 ii Abstract The focus of the research is on the responses of urban individuals in households and communities to socio-economic crisis in post-apartheid South Africa. In both Enhlalakahle and Mpumalanga Townships, there is a process of diversification of social, political and economic livelihood activities by women and men, young and old. The livelihoods literature together with literature on work and inter-household relations are merged to achieve a holistic understanding of urban livelihoods in South Africa. The study adopts a relational approach to capture the unequal power dynamics shaped by both gender and age, in the process of acquiring and securing livelihoods. The central argument of the study is that women and men, depending on their age, tend to adopt different livelihood activities in responding to their socio-economic crisis in post-apartheid South Africa. Specifically, the study shows that for individuals, households become central sites of production and survival, but only if supported by the state through social grants. At a community level, there are embryonic forms of ?counter-movement?, led by an alliance of popular movements in civil society such as the Treatment Action Campaign (TAC), Concerned Citizens Groups and anti-privatisation groups. The thesis explores diverse micro-level livelihood responses of urban households and communities shaped by a specific structural and historical macro-level context. iii Declaration I declare that this thesis is my own unaided work. It is submitted for the degree of Doctor of Philosophy in the University of the Witwatersrand, Johannesburg. It has not been submitted before for any other degree or examination in any other university. _________________________________ _________ day of _____________________, 2005. iv Acknowledgements I wish to express my deepest gratitude to the following people: My supervisors, Professor Eddie Webster and Doctor Jo Beall, for their intellectual guidance and support. Their commitment and patience is deeply appreciated. It was a great honour and experience to have worked with both. The Department for International Development (DfID), as part of the ?Crisis States Programme? co-ordinated by the Development Research Centre of the London School of Economics (LSE), for their financial contribution to this study. The Sociology of Work Unit (SWOP) for giving me an institutional home and a stimulating academic environment for the entire duration of my study. My colleagues at SWOP and the Sociology Department of the University of the Witwatersrand, to Andries Bezuidenhout, Sakhela Buhlungu, Bridget Kenny, Khayaat Fakier and Shameen Singh. To Professor Michael Burawoy, at the University of Berkeley, whose great ideas shaped my ideas about research and data analysis. To the Reworking Livelihoods Cluster at the University of Natal, thank you all for your insightful contribution to this study. A special mention, to Professors Gillian Hart and Ari Sitas who encouraged me to think big and whose work continues to inspire me. To Karin Pampallis for layout and editing of the thesis. My family for their support throughout the years. I will always be indebted to my parents, my sisters, brothers, aunts, and cousins. To Meiki, Thabang, Tebogo, and Darly, I would not have made it this far had it not been for your unconditional love. This thesis is dedicated to my brother, Jabu Mbatha. My friends, Michelle Williams, Malehoko Tshoaedi, Molly Dhlamini, Thabo v Sephiri, David Khunou, I appreciate all your support and encouragement. To Grace Khunou and Loren Landau for reading numerous drafts of this thesis and their insightful comments. Khunou, thank you for the many years of your unwavering friendship. Most of all, my partner, Mandla Radebe, without whom this study would not have been possible and worthwhile. Finally, to all the women and men of KwaZulu-Natal who agreed to be interviewed and who shared their experiences with me. To the Gama family in Mpumalanga Township and the Mazibuko family in Enhlalakahle Township who welcomed me into their homes and assisted me during my fieldwork process, ?Ningadinwa nangomuso?ukwanda kwaliwa umthakathi?. vi Contents Abstract ii Declaration iii Acknowledgements iv List of Tables x List of Figures xii List of Boxes xiii List of Abbreviations xiv Map xvii Preface xviii PART I 1 Chapter 1 The Household, Work and Livelihood Literature 2 1.1 Introduction 2 1.2 Defining the Household 1.2.1 A sociological definition 1.2.2 Intra-household dynamics 3 3 6 1.3 The Sustainable Livelihoods Approach, Principles and Framework 1.3.1 The sustainable livelihoods approach 1.3.2 The principles of the sustainable livelihoods approach 1.3.3 The capital assets framework 1.3.4 Assets: from labour to household relations 1.3.5 Critique of the capital assets framework 11 11 15 16 23 31 1.4 Reconceptualising Work 31 1.5 Conclusion 39 Chapter 2 The Macro Context: South Africa?s Policies 41 2.1 Introduction 41 2.2 Social Welfare Policies 2.2.1 The social welfare net 2.2.1 The crisis of poverty and inequality in South Africa 42 42 49 2.3 Labour Market Policies 2.3.1 Introduction 55 55 vii 2.3.2 The Labour Relations Act of 1995 2.3.3 The Basic Conditions of Employment Act of 1997 2.3.4 Labour law review 56 57 59 2.4 Economic Policies 2.4.1 From RDP to GEAR 2.4.2 The crisis of unemployment in South Africa 61 61 66 2.5 South Africa?s Civil Society 71 2.6 Labour and the Basic Income Grant Debate 78 2.7 Conclusion 80 Chapter 3 Research Questions and Methodology 84 3.1 Introduction 84 3.2 Research Questions 84 3.3 Research Methods 3.3.1 Qualitative research methods and techniques 3.3.2 Sampling methods 3.3.3 Choosing the research sites 3.3.4 Interviews as a research technique 3.3.4.1 Introduction 3.3.4.2 Pilot Interviews 3.3.4.3 Semi-structured Interviews 3.3.4.4 Oral History Interviews 3.3.4.5 Key Informant Interviews 85 85 88 88 89 89 91 93 94 97 3.4 Archival Research 98 3.5 Access to Households and Respondents 99 3.6 The Researcher and the Respondents 100 3.7 Qualitative Data Analysis 102 3.8 Problems and Limitations 104 3.9 Conclusion 105 PART II 106 Chapter 4 Background of South Africa, KwaZulu-Natal and the Two Research Sites: Mpumalanga and Enhlalakahle Townships 106 4.1 Introduction 106 4.2 South Africa?s Triple Transition 4.2.1 The nature of the triple transition 4.2.2 Socio-political transition 4.2.3 Economic transition 106 106 108 113 viii 4.3 Enhlalakahle and Mpumalanga Townships 4.3.1 Background 4.3.2 Structural context 4.3.3 The two townships: convergence and divergence 119 119 124 127 4.4 Conclusion 129 Chapter 5 Urban Household Economies: Access to Resources and Income 131 5.1 Introduction 131 5.2 A Process of ?Clustering? or ?Overcrowding?? 132 5.3 Hiding their Poverty, Hiding their Disease: The Power of Rural Ties 137 5.4 Household ?Income Pooling? and Expenditure 140 5.5 Essential Services: Household Access and Expenditure 151 5.6 Beyond Capitalist Commodity Production 5.6.1 Introduction 5.6.2 Social reproduction: women vs. men 5.6.3 Shift from wage goods to subsistence production 5.6.4 Petty commodity production: working in the informal economy 156 156 158 163 164 5.7 The Fluidity of Poverty 5.7.1 The well-being of households 5.7.2 An ?improving? household: the Zondi family 5.7.3 A ?coping? household: the Mazibuko family 5.7.4 A ?declining? household: the Mhlongo family 167 167 168 170 172 5.8 Conclusion 173 Chapter 6 Intra-household Dynamics: Power and Control over Resources 175 6.1 Introduction 175 6.2 Re-imagining the Domestic as a Site of Struggle 6.2.1 Shifting gender roles: the ?end of patriarchalism?? 6.2.2 ?De facto? heads of households and breadwinners: the micro- political relations of control 177 177 180 6.3 Fractured Social Relations I: Income and Time Conflict 185 6.4 Alcohol and Drug Abuse 189 6.5 Fractured Social Relations II: Domestic Violence 192 6.6 Inter-generational Conflict 194 6.7 Conclusion 198 Chapter 7 Beyond the Household: State and Civil Society Relations 199 7.1 Introduction 199 ix 7.2 The Post-apartheid State: From Labour to Welfare Policies 201 7.3 Post-apartheid Townships 7.3.1 Introduction 7.3.2 Membership of community organisations 7.3.3 The end of monetary contributions 7.3.4 The crisis of representation 7.3.5 Faith-based organisations 7.3.6 New community-based organisations and the new crises 7.3.7 Engaging the state: the Treatment Action Campaign 7.3.8 Home-based care: volunteerism in the community 204 204 205 207 209 209 222 228 232 7.4 Conclusion 237 Chapter 8 Conclusion ? Linking the macro and the micro: The State, individuals in households and communities 240 8.1 Introduction 8.2 Households as fragile sites of stability 8.3 Community Organization: Social resources beyond the household 8.4 The role of the state: implications of the micro-level policy framework 240 241 242 246 8.2 Livelihood Activities: Mpumalanga Township vs. Enhlalakahle Township 251 8.3 Theoretical Considerations 255 8.5 Conclusion 258 References 260 x List of Tables 1.1 Determinants and associated indicators of urban vulnerability at household and community level 21 1.2 Work and production 34 2.1 South Africa?s welfare system, 2004 42 2.2 Social grants, details and amounts, 2003-3006 48 2.3 South Africa?s welfare budget per province, 1998-2003 49 2.4 Department of Social Development budget, 2001-2004 49 2.5 South Africa?s poverty rate and poverty gap, 2002 50 2.6 Trends in number of social-grant beneficiaries, 2000-2003 55 2.7 Summary of selected Labour Law Amendments 2002: LRA and BCEA 61 2.8 Projected and actual targets of GEAR, 1999-2000 65 2.9 Unemployment in South Africa, 1999-2004 67 2.10 Formal and informal employment, 1990-2002 69 2.11 Number of registered trade unions and their membership levels, 2000- 2003 70 2.12 Number of South African non-profit organisations, by legal status 72 2.13 2.14 Associational life in Gauteng Conditions that contribute to an enabling environment for civil society 73 74 2.15 Apartheid and post-apartheid legal conditions of civil society organisations 75 3.1 Summary of interviews conducted in Mpumalanga and Enhlalakahle Townships 91 4.1 Size of households in South Africa, 1996-2001 109 4.2 Output, employment and wages in the textile and footwear sectors in South Africa, 1993-2002 118 4.3 Key indicators of Enhlalakahle Township (Ward 1), 1996-2001 120 4.4 Key indicators of Mpumalanga Township (Ward 6), 1996-2001 122 5.1 Size of households in Mpumalanga and Enhlalakahle Townships 134 5.2 Type of shelter in Mpumalanga and Enhlalakahle Townships 135 5.3 Size of house in Mpumalanga and Enhlalakahle Townships 136 5.4 Main sources of household income in Mpumalanga and Enhlalakahle Townships 140 5.5 Total household monthly income in Mpumalanga and Enhlalakahle 143 xi Townships 5.6 Monthly expenditure on food in Mpumalanga and Enhlalakahle Townships 145 5.7 Viable and survivalist livelihood activities 148 5.8 Monthly expenditure on education in Mpumalanga and Enhlalakahle Townships 149 5.9 Monthly expenditure on health in Mpumalanga and Enhlalakahle Townships 150 5.10 Monthly expenditure on electricity in Mpumalanga and Enhlalakahle Townships 152 5.11 Sources of energy in Mpumalanga and Enhlalakahle Townships 152 5.12 Television and radio usage in Mpumalanga and Enhlalakahle Townships 153 5.13 Electrical appliance usage in Mpumalanga and Enhlalakahle Townships 153 5.14 Source of household water in Mpumalanga and Enhlalakahle Townships 154 5.15 Monthly expenditure on water in Mpumalanga and Enhlalakahle Townships 154 5.16 Work and employment in Mpumalanga and Enhlalakahle Townships 157 6.1 Productive and non-productive activities and mean minutes for men and women, South Africa, 2001 187 7.1 Associational life in Mpumalanga and Enhlalakahle Townships 205 7.2 Number of full-time equivalent (FTE) employees in the non-profit sector 232 7.3 Condition of patients seen by community caregivers 234 8.1 Summary of inter- and intra-household and community livelihood activities in Mpumalanga and Enhlalakahle Townships 254 xii List of Figures 1.1 The sustainable livelihoods framework 16 2.1 South Africa?s Gini coefficients, 1995 and 2001 51 2.2 South Africa?s Human Development Index by province, 1990 and 2003 52 2.3 Unemployment rate (official definition) by race and gender, 2004 68 4.1 Household type by race in South Africa (percentage), 1996-2001 110 4.2 Real value added by the manufacturing sector in South Africa, 1960- 2002 115 4.3 Labour productivity in the manufacturing sector in South Africa, 1990- 2002 116 4.4 Employment in the manufacturing sector in South Africa, 1967-2002 117 5.1 Summary of livelihood responses to income and food insecurity in Mpumalanga and Enhlalakahle Townships 148 xiii List of Boxes Box 7.1 The Concerned Citizens Forum 225 Box 7.2 Interviewing Mrs Gwala 234 xiv List of Abbreviations ABE Affirmative business Enterprise AID Acquired Immune Deficiency Syndrome ANC African National Congress APF Anti-Privatisation Forum ASSA Actuarial Society of South Africa BBC Black Business Council BCEA Basic Conditions of Employment Act BIG Basic Income Grants BSA Business South Africa BUSA Business Unity South Africa CBO Community-based Organisation CCG Concerned Citizens Group CCMA Commission for Conciliation, Mediation and Arbitration COFESA Confederation of Employers of Southern Africa COSATU Congress of South African Trade Unions CPI Consumer Price Index CSO Civil Society Organisation DfID Department for International Development [UK] DSF Durban Social Forum FBO Faith Based Organisation FEDUSA Federation of Unions of South Africa FOSATU Federation of South African Trade Unions FTE Full-time Equivalent GATT General Agreement on Tariffs and Trade GDP Gross Domestic Product GEAR Growth, Employment and Redistribution [strategy] GCIS Government Communication and Information System GNP Gross National Product HDI Human Development Index HIV Human Immunodeficiency Virus xv HSRC Human Sciences Research Council HAYCO Hammarsdale Youth Congress IDC Industrial Development Corporation IDP Integrated Development Plan IFP Inkatha Freedom Party LFS Labour Force Survey LPM Landless People?s Movement, LRA Labour Relations Act MCCG Mpumalanga Concerned Citizens Group NACTU National Council of Trade Unions NEDLAC National Economic Development and Labour Council NGO Non-governmental Organisation NPO Non-profit Organisation NQF National Qualifications Framework OHS October Household Survey. QDA Qualitative Data Analysis PGIEP Policy Guidelines for Integrating Environmental Planning into Land Reform RDP Reconstruction and Development Programme SACC South African Council of Churches SACP South African Communist Party SACTWU Southern African Clothing and Textiles Workers Union SADF South African Defence Force SANCO South African National Civics Organisation SANGOCO South African Non-governmental Organisation Coalition SAP South African Police SECC Soweto Electricity Crisis Committee SMME Small, Medium and Micro Enterprises SNA System of National Accounts SPSS Statistical Package for Social Sciences TAC Treatment Action Campaign TB Tuberculosis UDF United Democratic Front UNDP United Nations Development Programme xvi VAT Value Added Tax WTO World Trade Organisation xvii MAP of KwaZulu-Natal showing Greytown and Mpumalanga Township (2004) xviii Preface In the course of my fieldwork for my masters research report, I identified the growing informalisation of footwear production in Pietermaritzburg, KwaZulu- Natal, formerly the centre of South Africa?s footwear industry. I had begun my research examining the response of the industry to global competition, but soon realised that the really interesting sociological question was how the growing number of retrenched women workers, households and communities were surviving without regular formal employment. I undertook interviews with the women working in the footwear sector in Pietermaritzburg and visited them in their homes in Enhlalakahle Township in Greytown. In subsequent research, a similar pattern of retrenchment among women textile workers in Mpumalanga Township in Hammarsdale is identified. Both communities were struggling with the crisis of unemployment and poverty. They were also responding to the crisis through various intra- and inter-household and community livelihood activities. The central aim of this research became, therefore, to examine these changing livelihood activities of individuals, households and communities in Mpumalanga Township in Hammarsdale, eThekwini municipality, and Enhlalakahle Township in Greytown, Umvoti municipality, both in the province of KwaZulu-Natal, South Africa. The study explores the impact of macro-economic changes and micro-level responses of households, different household members, and the two communities in KwaZulu-Natal. This follows from Colin Murray?s (1999:3) argument that ?multiple livelihoods cannot be comprehended at the macro-level and must be investigated empirically at the micro-level.? The study examines the social resources on which people rely and whether these are confined to urban areas or extend to rural areas, whether they are confined to the family or extend to wider social networks, and whether they are confined to areas of current livelihood xix activities and neighbourhoods. While many studies of household and community livelihoods exist, these have primarily focused on rural areas, and those with an urban focus often missed or ignored the urban-rural linkages. This is central given the history of long-term migration and strong links between town and country in KwaZulu-Natal. The aim of this research was to overcome this gap and conduct empirical research on the urban poor by looking at urban households, without losing sight of rural-urban linkages. The study considers gender and age as two key variables as both point to differences among household members in relation to life-cycle stages and inter- generational relationships. The two translate into differences in the ability to cope with economic difficulties within and outside of households (Pearson, 1997). Gender and age are most importantly shaped and influenced by social relations within and between households. They determine how urban household livelihood activities are examined and understood. Hence, the research offers a gendered and intergenerational analysis of urban household livelihood activities and intra- household dynamics. It assumes that women and men, depending on their age, tend to adopt different livelihood activities that bring with them specific rewards. Therefore, the question that this research explores is not just how resources are allocated, shared and exchanged among household members, but how gendered rights and obligations are constructed, maintained and made to appear ?natural?. What are the conditions and modalities through which particular groups of women interrogate and overtly challenge prevailing gender ideologies and arrangements (Hart, 1997:19)? Furthermore, does women?s access to income-earning opportunities have any implications for gender relations within the household? Is economic dependency a major factor in structuring bargaining inequalities between women and men (Kabeer, 1997)? The thesis explores the ways in which lack of income by men and income access by women leads to serious tensions in households, not only between men and women but also between young women xx and elders. The thesis explores the following assumptions. First, it is assumed that the household will become central and the first means of survival for many poor and unemployed people. Does this result in, as Burawoy, Krotov & Lytkina, (2000) argue of Russia, more reliance on women given their traditional role as caregivers and therefore an elevation of their status in the household and community? Also, because of the changing nature of the labour market (with a shift towards the informal economy), there will be more benefits for women than men. Will the elevation of women challenge gender roles and masculine identities? What becomes the status of men, given their challenged roles as ?breadwinners? and therefore heads of households since they are now unemployed? Does this mean a role reversal in the household and an ?end of patriarchalism? as Castells (1997) suggests? Second, there is likely to be a range of similar livelihood activities pursued in both communities. These include an increased reliance on community organisations (formal and informal) such as faith-based organisations, burial and saving societies, as well as non-party-political groups such as the Treatment Action Campaign (TAC) and anti-privatisation groups. Could these organisations provide the basis for a backlash against liberalisation, as Karl Polanyi (2001) argues in his notion of a ?counter movement? led by an alliance of popular movements in civil society? Thirdly, my impression is that there is less dependence on political parties, since they are often blamed for some of the problems experienced by the community and given the bitter history of political violence in the province. The reconfiguration of the relationship between political organisations and individuals and communities signals a new trajectory in the two communities. Trade unions are also likely to be less significant because of their traditional role of only organising in the formal economy. This representational gap will largely be filled by the emerging community organisations. xxi Fourth, given the increase in unpaid work, wage employment continues to be the most important but unstable source of income for the majority of the households. However, there will be growing expectation that the state provides for the poor. Lastly, the contradictory role of the state as both an agent of redistribution and cost recovery has altered state-citizen relations in both communities. At most, it is characterised by conflict and mistrust. State social welfare and labour policies have been limited in cushioning the impact of liberalisation and consequent unemployment and poverty. Thus, this study is about how individuals, households and communities in post- apartheid South Africa are responding to the crisis of unemployment and poverty. It has two parts, each focusing on different aspects of urban livelihoods in the two research sites of Mpumalanga and Enhlalakahle Townships. Part One introduces the research project and its context. Chapter 1 examines the various literatures on household, work and livelihood frameworks. The chapter engages with the livelihoods approach and identifies various shortcomings in its understanding of how poor people survive. In Chapter 2, South Africa?s social security framework is explored, covering social, labour and economic policies. State and civil society relations within South Africa are also discussed. Chapter 3 analyses the various methodological issues, research design and techniques employed in investigating urban household livelihood activities pursued at the two research sites. Part Two presents the empirical data, starting with the background of the two research sites in Chapter 4. The chapter concludes that the macro, historical and current context shapes the outcomes of livelihoods pursued in the two communities. The historical analysis of the country, province and communities seeks to demonstrate that government policies have varied over time and have influenced the creation and sustainability of livelihoods of the poor at the two research sites. Chapter 5, 6, and 7 present an analysis of a series of interviews conducted in 2002, 2003, and 2004. They explore intra- and inter-household and community xxii livelihood activities. Chapter 5 provides descriptive and statistical information of the 29 identified households, including the economic and material conditions and livelihoods of the households. Chapter 6 is a critical discussion of the households, identifying unequal power relations and resource allocation based on gender and age. The chapter concludes that household livelihood activities are linked to community livelihood activities. Chapter 7 discusses community-level livelihood activities through an examination of various formal and informal community organisations. In Chapter 8, the dissertation concludes with a summary of the key findings of the study, its theoretical and practical implications, strengths and weaknesses of the study and suggestions for future research.