University of Ghana http://ugspace.ug.edu.gh WATER AND ELECTRICITY ACCESS FOR HOME-BASED ENTERPRISES AND POVERTY REDUCTION IN THE GREATER ACCRA METROPOLITAN AREA (GAMA) THIS THESIS IS SUBMITTED TO THE UNIVERSITY OF GHANA, LEGON BY EBENEZER FORKUO AMANKWAA (10195687) IN PARTIAL FULFILMENT OF THE REQUIREMENT FOR THE AWARD OF DOCTOR OF PHILOSOPHY (PHD) GEOGRAPHY AND RESOURCE DEVELOPMENT DEGREE MARCH, 2017 University of Ghana http://ugspace.ug.edu.gh DECLARATION This is to certify that this thesis is original and the product of a study I personally conducted. Apart from references to existing literature that were used, which have duly been acknowledged, the thesis has not been submitted to this or any other institution in part or whole for the award of any degree. Ebenezer Forkuo Amankwaa (Candidate) .......................................................... ................................. Signed Date We, the undersigned supervisors, certify that this thesis is an original work we supervised the candidate to produce. We are also convinced that it meets all required standards set by the University of Ghana for an award of a Doctor of Philosophy (PhD) Degree. ................................................ ................................. PROF. PAUL. W. K. YANKSON DATE (Principal Supervisor) ................................................. ................................. PROF. MARTIN OTENG-ABABIO DATE (Supervisor) ................................................. ................................. PROF. KATHERINE GOUGH DATE (Supervisor) i University of Ghana http://ugspace.ug.edu.gh DEDICATION To my lovely wife, Christine Amankwaa Esq, and son, Ebenezer Amankwaa Jr., I dedicate this work to you for your kind understanding, support and prayers. ii University of Ghana http://ugspace.ug.edu.gh ABSTRACT Access to water and electricity services is intricately connected to improvements in the viability of livelihoods thereby reducing poverty. Despite research that attests to the significance of services and livelihoods, these issues have often been examined separately. In particular, the impact of water and electricity access on home-based enterprises (HBEs) remains largely understudied and not clearly understood. This study fills this lacuna. It illustrates how the urban poor in Accra use a range of strategies to improve their water and electricity access which in turn enhances the long-term viability of their livelihood activities. The study adopts an urban political economy approach to highlight how complex governance networks are (re)shaping household economies and the water and electricity agenda such that the vested interests of local elites determine ‘whose access matters’. This conceptual approach is accompanied by mixed methods including surveys, focus group discussions, in-depth interviews, and mobile interviews such as follow-along participant observation conducted with residents in four of the city’s neighbourhoods to examine the intersections between services, HBEs and poverty. The thesis finds that access is defined by proximity, reliability, quality and affordability dimensions that produce inequalities, with urban dwellers experiencing unequal geographies through a multiplicity of governance practices. The study argues that the contestations over access to urban services take place across a series of different domains incorporating the everyday geographies of the water and electricity network and household livelihood activities. Examining this services-livelihoods intersection, the study reveals how improving access to water and electricity significantly enhances home-based enterprises and reduces household poverty, particularly through job and income generation and increased affordability of services. It is argued that people’s poverty reduction strategies illustrate how informality is nested within formal service provision but in a subtle manner so as to not compromise the veneer of formality of the service or livelihood activity. It also highlights the agency of the poor and the complex ways in which they negotiate poverty, which are shown to be due to the inefficiencies of urban governance. These complexities suggest the notion of ‘emergent formalizations’, as it helps in categorizing the particular configuration of formal and informal elements and the patterns of interaction that characterize different modes of governance arrangements in the water and electricity sector. The study recommends inclusive pro-poor policies aimed at supporting service–dependent livelihoods. iii University of Ghana http://ugspace.ug.edu.gh ACKNOWLEDGEMENT “Delight yourself in the Lord, and He shall give you the desires of your heart. Commit your way to the Lord, …. and He shall bring it to pass” (Psalm 37:4-5). But for your grace, strength and wisdom during my research period, this thesis would not have materialised. Thank you, Jesus. First and foremost, I owe special thanks to my team of supervisors, Prof. Paul W. K. Yankson (University of Ghana), Prof. Martin Oteng-Ababio (University of Ghana) and Prof. Katherine Gough (Loughborough University, UK) for devoting quality time to supervise this thesis. Their guidance, encouragement and constructive comments have made this thesis possible. As the PhD was an advertised role I thank Prof. Paul Yankson for selecting me to join the RurbanAfrica Project team. His influence has been both thought provoking and challenging but always in a friendly and approachable manner. May God richly bless him. I am greatly indebted to Prof. Martin Oteng-Ababio for believing in me and giving me the opportunity to pursue this dissertation. For his good working relations before, during and even beyond this study, I say may God grant him good health and success in all his endeavours. A warm thank you is due Prof. Katherine Gough. I thank her for arranging for me to visit her Department in Loughborough University, UK during the dissertation writing period. I appreciate her exceptional academic and personal support throughout the PhD process and even beyond. May God reward her. I also owe special thanks to the entire staff at the Department of Geography for making my stay in UK so special. Particular thanks is due to Dr. James Esson for his insight into the initial conceptualisation and encouragement throughout the writing period especially during my research visit to Loughborough University. I am very grateful to the lecturers and entire staff of the Department of Geography and Resource Development, University of Ghana, including the Head of Department, Prof. George Owusu who supported my application and communicated to me about my successful selection as a PhD Student to join the RurbanAfrica Ghana Team. Many thanks to Prof. Jacob Songsore, Prof. Joseph Yaro, Dr. Joseph Teye, Dr. Charlotte Wrigley-Asante, Dr. Alex Barima Owusu, and Dr. Ernest Agyemang for their support and valuable comments particularly during our departmental seminar presentations. I owe special thanks to all those who participated in the survey and focus groups. Their receptivity and generosity to share their time and lived experiences generated invaluable information. I also say thank you very much to all the government officials, experts, assembly iv University of Ghana http://ugspace.ug.edu.gh members and opinion leaders for availing themselves to be interviewed and also for sharing relevant data. I am highly indebted to Mr. Peter Mensah and Emmanuel Acheampong, for their tireless work and commitment and for accompanying me in various stages of the fieldwork, Mr. Peter Ofori-Atta for the maps developed and Mr. George Asare for assisting with statistical analysis. Alex, I am most grateful for your inspiration and help. His constructive comments on initial drafts of my chapters really gave shape to the thesis. Thanks also to all my friends and PhD colleagues in the Geography Department and other departments, including the UG-PADA team, for their support and encouragement and endless conversations that kept me going. I am grateful to the European Commission for granting me a full PhD scholarship under the 7th Research Framework Programme (SP1-Cooperation, FP7-SSH-2011-2), African Rural- City Connections (RurbanAfrica) Project to complete this thesis. Special thanks go to all the PhD Candidates and the entire RurbanAfrica team, particularly Prof. Jytte Agergaard (Principal Investigator) and the Department of Geosciences, University of Copenhagen for hosting me during the PhD course in 2013 and the project conference in 2016. My final appreciation goes to my father, Mr. Kwaku Forkuo Ababio and my mother, Margaret Ofori for their unconditional faith in me and encouraging me to pursue my own interests. I must specially thank Christine my lovely wife for countless hours spent listening, commenting and correcting my work even at odd hours. Thank you so much Princess for sacrificing for me and always making me smile even when the going was tough. May God richly bless you all. v University of Ghana http://ugspace.ug.edu.gh TABLE OF CONTENTS DECLARATION ................................................................................................................. i DEDICATION .................................................................................................................... ii ABSTRACT ...................................................................................................................... iii ACKNOWLEDGEMENT .................................................................................................. iv TABLE OF CONTENTS ................................................................................................... vi LIST OF FIGURES............................................................................................................ ix LIST OF TABLES .............................................................................................................. x LIST OF PLATES ............................................................................................................. xi LIST OF BOXES .............................................................................................................. xii CHAPTER ONE ................................................................................................................. 1 INTRODUCTION .............................................................................................................. 1 1.0. Background to the study............................................................................................ 1 1.1. The Ghanaian case – Statement of the research problem ........................................ 5 1.1.1. Research aim, objectives and questions .............................................................. 8 1.2. Research propositions ........................................................................................... 9 1.3. Structure of the dissertation ................................................................................. 10 CHAPTER TWO .............................................................................................................. 12 LITERATURE REVIEW AND THEORETICAL AND CONCEPTUAL FRAMEWORK 12 2.0. Introduction ............................................................................................................ 12 2.1. Re-conceptualizing urban governance ..................................................................... 12 2.1.1. What is urban governance? .............................................................................. 13 2.1.2. The dimensions of urban governance ................................................................ 15 2.2. Urban governance: Ghana’s situation in the African and Global context ................. 18 2.2.1. Overview of urbanization.................................................................................. 18 2.2.2. Water access situation ...................................................................................... 22 2.2.3. Electricity access situation ............................................................................... 27 2.2.4. Home-based enterprises (HBEs), urban services and poverty reduction ........... 31 2.3. Theoretical underpinnings of the study .................................................................... 40 2.3.1. Competing approaches to urban analysis – urban political economy ................ 41 2.3.2. Approaches to urban political economy ............................................................ 47 2.4. Conceptual framework ............................................................................................ 53 2.5. Conclusion .............................................................................................................. 57 CHAPTER THREE ........................................................................................................... 59 PROFILE OF STUDY SITES, PHILOSOPHY AND METHODOLOGY ......................... 59 3.0. Introduction ............................................................................................................ 59 3.1. Profile of study sites................................................................................................ 59 3.1.1. The Characteristics of GAMA ........................................................................... 59 3.1.2.2. Accra New Town ....................................................................................... 67 3.1.2.3. Abuja ......................................................................................................... 69 3.1.2.4. Gbawe........................................................................................................ 70 vi University of Ghana http://ugspace.ug.edu.gh 3.2. Philosophical orientation ......................................................................................... 72 3.2.1. Ontological perspectives................................................................................... 72 3.2.2 Epistemological perspectives and methodology ................................................. 74 3.3. Research methodology ............................................................................................ 75 3.3.1. Mapping of services and households and HBEs survey ..................................... 77 3.3.2. Sampling methods and data sources ................................................................. 78 3.3.3. In-depth interviews and focus group discussions (FGDs) .................................. 83 3.3.4. Data processing and analyses ........................................................................... 86 3.3.5. Data limitations ................................................................................................ 87 3.4. Negotiating power, positionality, and ethics ............................................................ 88 3.5. Profile of respondents ............................................................................................. 91 3.6. Conclusion .............................................................................................................. 97 CHAPTER FOUR ............................................................................................................. 99 THE DYNAMICS BEHIND WATER AND ELECTRICITY ACCESS ............................ 99 4.0. Introduction ............................................................................................................ 99 4.1. Dynamics of water access ....................................................................................... 99 4.1.1. Spatial dynamics of water access ...................................................................... 99 4.1.2. Reliability and proximity dynamics of water access ........................................ 105 4.1.3. Quality, quantity and usability dynamics of water access ................................ 110 4.1.4. Affordability dynamics of water access ........................................................... 116 4.2. Dynamics of electricity access .............................................................................. 121 4.2.1. Spatial dynamics of electricity access ............................................................. 121 4.2.2. Reliability and quality dynamics of electricity access ...................................... 127 4.2.3. Affordability and usability dynamics of electricity access................................ 133 4.3. Conclusion ............................................................................................................ 138 CHAPTER FIVE............................................................................................................. 140 ASSESSING THE IMPACT OF WATER AND ELECTRICITY ACCESS ON HOME- BASED ENTERPRISES IN GAMA ............................................................................... 140 5.0. Introduction .......................................................................................................... 140 5.1. Residential livelihood activities in the urban economy .......................................... 140 5.2. Occupational geographies and linkages of household economies in GAMA .......... 148 5.3. Home-based activities vis-à-vis water and electricity access in GAMA ................. 160 5.4. Conclusion ............................................................................................................ 169 CHAPTER SIX ............................................................................................................... 171 THE NEXUS BETWEEN ACCESS TO SERVICES, LIVELIHOODS AND POVERTY REDUCTION IN GAMA ................................................................................................ 171 6.0. Introduction .......................................................................................................... 171 6.1. Synergies between access to services, livelihoods and poverty in GAMA ............. 171 6.1.1. Understanding households’ perceptions of poverty ......................................... 172 6.1.2. Impact of water and electricity access on poverty reduction ........................... 181 6.1.3. Role of governance in improving service provision for poverty reduction ....... 190 6.2. Towards effective governance for poverty reduction – a conceptual model ........... 200 vii University of Ghana http://ugspace.ug.edu.gh 6.3. Conclusion ............................................................................................................ 206 CHAPTER SEVEN ......................................................................................................... 208 SUMMARY OF KEY FINDINGS, CONCLUSIONS AND POLICY IMPLICATIONS . 208 7.0. Introduction .......................................................................................................... 208 7.1. Dynamics of access to urban services .................................................................... 208 7.1.1. Beyond access to water and electricity ............................................................ 209 7.2. Implications of water and electricity access for home-based enterprises ................ 212 7.2.1. HBEs vis-à-vis water and electricity access .................................................... 212 7.3. Implications of water and electricity access for poverty reduction ......................... 214 7.3.1. Rethinking poverty – Income, service and governance ‘poverties’ .................. 214 7.4. Summary of main contributions to knowledge ...................................................... 216 7.5. Recommendations – policy and practice................................................................ 219 7.5.1. Community–household collective action recommendations ............................. 220 7.5.2. Planning and organizational recommendations .............................................. 222 7.5.3. Regulatory recommendations ......................................................................... 225 7.6. Areas for future studies ......................................................................................... 227 REFERENCES ............................................................................................................... 229 APPENDICES ................................................................................................................ 249 viii University of Ghana http://ugspace.ug.edu.gh LIST OF FIGURES Figure 2.1: The supply structure of Ghana’s electricity sector ............................................ 28 Figure 2.2: A typical daily load profile of Ghana showing peak demand periods ............... 30 Figure 2.3: A conceptual framework of water and electricity governance and livelihoods on poverty reduction in Ghana ............................................................................. 54 Figure 3.1: Map of GAMA showing its constituent administrative areas............................ 63 Figure 3.2: Map of GAMA showing the selected study sites .............................................. 64 Figure 4.1: Distribution of water sources in the study area ............................................... 101 Figure 4.2: Frequency and percentage distribution of time spent to procure water ........... 109 Figure 4.3: Distribution of water quality concerns in the study area ................................. 111 Figure 4.4: Distribution of household electricity connection across the settlements ......... 124 Figure 4.5: Distribution of household connection to pre-paid and post-paid meters .......... 130 Figure 4.6: Daily duration of household power supply across the settlements .................. 131 Figure 5.1: Distribution of water and electricity dependent HBEs in GAMA ................... 143 Figure 6.1: Households in GAMA below and above the income poverty lines ................. 179 Figure 6.2: Settlement distribution of households below and above the income poverty lines.............................................................................................................. 179 Figure 6.3: Assessing water access for HBEs .................................................................. 183 Figure 6.4: Impact of water access on HBEs and poverty ................................................ 183 Figure 6.5: Assessing electricity access for HBEs ............................................................ 184 Figure 6.6: Impact of electricity access on HBEs and poverty .......................................... 184 Figure 6.7: Service poverty level in GAMA using the World Bank and GLSS poverty lines.............................................................................................................. 185 Figure 6.8: Perception of the impact of improved water and electricity access on poverty 186 Figure 6.9: Framework on governance, urban services, livelihoods and poverty reduction ...................................................................................................... 202 ix University of Ghana http://ugspace.ug.edu.gh LIST OF TABLES Table 2.1: Population of the world and major areas, 2015, 2030, 2050 and 2100 ............... 19 Table 2.2: Types of public-private partnerships (PPP) ....................................................... 25 Table 2.3: Trends in urban water supply sources in Ghana (%) .......................................... 26 Table 2.4: Installed and projected electricity generation in Ghana (2014 and 2015) ........... 28 Table 2.5: GDP growth decomposition, by per capita value added, 1991–2012 ................. 40 Table 3.1: Brief profiles of settlements selected in GAMA ................................................ 65 Table 3.2: Socio-demographic characteristics of respondents ............................................ 92 Table 4.1: Spatiality of water coverage at national, regional, metropolitan and locality levels ............................................................................................................. 100 Table 4.2: Settlement distribution and chi square analysis of water sources ..................... 102 Table 4.3: Providers supplying water to households in the study area .............................. 105 Table 4.4: Percentage distribution of water flow across the settlements ........................... 106 Table 4.5: Percentage distribution of the domestic uses of water sources ......................... 114 Table 4.6: T-test comparing means of monthly cost of using public and private water sources ........................................................................................................... 116 Table 4.7: Comparison of water utility and vendor prices in GAMA ............................... 117 Table 4.8: Spatiality of electricity coverage at national, regional, metropolitan and locality levels ............................................................................................................. 122 Table 4.9: Main providers/sources of power supply in the study area ............................... 127 Table 4.10: T-test comparing means of monthly cost of using prepaid and post-paid meters ............................................................................................................ 134 Table 4.11: Monthly cost (GHȼ) of alternative power sources across settlements ............ 135 Table 4.12: End user electricity tariff in Accra ................................................................ 137 Table 5.1: Occupation of employed population in Accra in 2010 ..................................... 141 Table 5.2: Type of HBEs in GAMA (%) ......................................................................... 144 Table 5.3: Locational characteristics of HBEs ................................................................. 149 Table 5.4: Average monthly incomes from HBEs (GH¢) ................................................. 157 Table 5.5: Settlement distribution of average monthly earnings from HBEs (%) .............. 159 Table 5.6: Generator prices of HBEs that run emergency recharging centres ................... 167 Table 5. 7: Differences in electricity and generator prices of services charged by HBEs .. 167 Table 6.1: Residents ranking of poverty indicators in GAMA .......................................... 173 x University of Ghana http://ugspace.ug.edu.gh LIST OF PLATES Plate 5.1: Environmental conditions of a typical open space kitchen used for HBEs ........ 152 Plate 5.2: A case of multiple households engaging in cooked food and carpentry business ........................................................................................................... 154 Plate 5.3: Purchasing of sachet water bags by cooked food vendors for their HBEs ......... 163 xi University of Ghana http://ugspace.ug.edu.gh LIST OF BOXES Box 6.1: Households summary of poverty and wealth indicators ..................................... 178 xii University of Ghana http://ugspace.ug.edu.gh CHAPTER ONE INTRODUCTION 1.0. Background to the study “Cities are the best path we know out of poverty. They are the best transformers of civilizations. But, there are also demons that come with density” (Glaeser, 2016, p. 1). Urbanization is a catalyst for socio-economic transformation, however if not properly planned it can create a wheel-spinning effect on service delivery, employment, and urban poverty. In developing countries in particular, the problems of urbanisation are enmeshed and do manifest in challenges for urban governance. Prior studies articulating some of these challenges indicate that globally 748 million people lack access to an improved water source (WWAP, 2015; WHO and UNICEF, 2014) while over 1.3 billion have no access to electricity (IEA, 2013). The future looks bleak with electricity demand, for example, projected to grow by 70% by 2035 (IEA, 2013). The scenario in urban areas appears precarious with the number of people without access to potable water increasing from 111 million to 149 million between 1990 and 2012 (WHO and UNICEF, 2014). The situation in sub-Saharan Africa (SSA) where the percentage of urban people who enjoyed piped water on their premises decreased from 43% to 33% between 1990 and 2015 is not different (World Cities Report, 2016). According to the World Bank (2016), for Africa to achieve the United Nations’ Sustainable Development Goals (UN’s SDGs) on water and electricity, the governments will require to make investments in infrastructure that go far beyond what is expected of the sub-region’s current path. This scenario forecasts a cumulative increase in infrastructure spending of $20 billion annually for achieving the water and sanitation SDG and another $1 billion for achieving universal electricity access by 2030. These data highlight how access to water and electricity in recent times have assumed deteriorating and problematic trajectory, a situation exacerbated by the aggregated statistics normally churned out by public officials which 1 University of Ghana http://ugspace.ug.edu.gh conceal spatial contradictions and exclude quality and affordability considerations (Bellaubi and Visscher, 2014; Amankwaa et al., 2014). Besides, the data are averages and elide important differences in terms of actual access to services and equity. This makes it important to learn about the water and electricity delivery situation, particularly in urban areas. Bearing this background in mind, and considering the SDGs, this study relates in particular to SDG 6 (Clean Water), SDG 7 (Affordable and Clean Energy) and SDG 11 (Sustainable Cities and Communities) as well as the New Urban Agenda (NUA). Their explicit emphasis on sustainable cities and communities and on universal access to water and energy implies that the conceptual and policy landscape may be shifting towards a more nuanced approach to development that places more interpretative weight on infrastructure services and on ensuring that everyone has access to services. This suggests that the Millennium Development Goals (MDGs) had unfinished business with regards to progress in urban areas. It further presents practical questions over how to make cities and settlements inclusive, safe, resilient and sustainable (Arfvidsson et al., 2016). In urban areas, the increasing number of people without access to electricity and water is related directly to the growth of informal settlements and slums (WWAP, 2015). The growth of informal settlements is exacerbated by government’s reluctance and inability to extend adequate services to residents in such areas. Such posture mirrors new forms of ‘urban punishment’ by governments to unclog the city and make cities expensive for newcomers. Consequently, the phenomenon of splintering urbanism – how infrastructure networks fragment the experience of contemporary city life – emerges (Graham and Marvin, 2001; Silver, 2013). This point relates to and is articulated in Yankson’s (2008) and Obeng- Odoom’s (2013) call for services to be examined in relation to urban governance. Though there is a rich body of research regarding how good urban governance can improve services and reduce poverty (Mitlin, 2008), other studies have emphasized why decentralised 2 University of Ghana http://ugspace.ug.edu.gh governance has failed to deliver municipal services effectively in Africa (Martinez-Vazquez and Vaillancourt, 2011). Access to services like water and electricity is intricately connected with improvement in livelihood and poverty reduction (IIED, 2010). Global and African–wide poverty levels are falling, but according to The State of African Cities Report 2014, there are “varying and increasing levels of urban poverty, inequality, and inefficiency” in Africa (UN-Habitat, 2014, p. 37). Poverty decreases affordability of urban services. This situation is related to questions of employment (both in types and job access) and income. The urban informal economy is a distinctive feature of cities in SSA, whereby jobs are concentrated in service and value–added enterprises including home-based enterprises (HBEs) like catering and retail (Anyidoho and Steel, 2016). Gough (2010) indicates that HBEs make key contributions to local economies through job creation and income generation. The relationship between informal work and poverty becomes evident when intra-household dynamics like dependency ratio, number and type of income sources, and level of access to services are analysed. Despite a growing body of global research problematizing access to urban services, and HBEs as livelihood strategies, as well as improving infrastructure services to reduce poverty (Gough et al., 2003; Yankson, 2008; Obeng-Odoom, 2013), the impacts of electricity and water access on household enterprises have until recently been a rather neglected area of research. The literature is replete on the health and time–saving dimensions of water and electricity (Stoler et al., 2012; World Bank, 2015a) and business constraints regarding access to finance (Fox and Sohnesen, 2012) while the infrastructure service constraints and productive uses of water and electricity for household businesses is limited (IIED, 2010). Not only does the focus on the negative portrayal of the health cost tell part of the story and cause policies to be framed from a narrow public health perspective, but it also makes it difficult for city authorities to appreciate the positive narrative and incorporate it into planning schemes. Moreover, 3 University of Ghana http://ugspace.ug.edu.gh discourse on domestic and productive uses of urban services remains separate. Meanwhile, understanding the middle–ground or ‘grey area’ in–between is arguably the key to improving services, enhancing livelihoods and alleviating poverty. Consequently, the question becomes: How can improved access to water and electricity enhance HBEs, via provisioning quality, reliable and affordable services, thereby reduce poverty? I contend that the need for improvement in water and electricity services assumes prominence when the linkages with other dimensions of poverty are considered. The study thus examines a deeper conception of poverty as it is related to electricity and water access. It conceives poverty as a livelihood and governance issue. This perspective provides a dense and novel narrative to the literature on urban services by examining the city, service provision, and their relationship to livelihoods and poverty reduction. To address these knowledge gaps, this study draws upon a governance–poverty framework to provide a contextual exploration of the implications of water and electricity access for HBEs and poverty reduction in Accra, Ghana. The study is distinct and complements existing studies in two ways. First, it problematizes water and electricity access to underscore the deeper meanings and everyday lived experiences of the poor in relation to cost, quality, and reliability. Second, it makes a case that access impacts water and electricity-dependent livelihoods and that grasping this interdependent connection is essential in conceptualising the future sustainability of the informal economy and their relevance to poverty reduction within the wider policy context of electricity and water governance. It does so by using an urban political economy approach (which requires that the study of urban issues and policies be situated in the context of broader socio-economic analysis) to focus on four classes of residential categories in the Accra metropolis: indigenous area (Korle Gonno), low-middle class area (Accra New Town), peri-urban area (Gbawe) and informal settlement (Abuja). 4 University of Ghana http://ugspace.ug.edu.gh 1.1. The Ghanaian case – Statement of the research problem The 2015 MDGs report declared Ghana as a successful case by halving extreme poverty and the proportion of its citizens without access to water in 2010, ahead of the 2015 deadline (NDPC, 2015). This success is due to the increase in access to water from 54% in 1990 to 87% in 2012 (WHO and UNICEF, 2014), while access to electricity soared from 45% in 1990 to 80.5% in 2016 (The Presidency of Ghana, 2016). Notwithstanding the overall success with coverage, access remains an issue among urban residents. The World Bank (2015a) reveals a downward trend in the proportion of residents in metropolitan areas with access to piped- borne water within the first decade of the 21st century: Accra was the worst off with a decline of −22.2 percentage points, followed by Kumasi (−7.7 percentage points) and Tema (−5.7 percentage points). The report further estimates that in Accra, the capital and economic hub, the proportion of households with access to piped water dropped from 91 to 69%. This deficit was offset by informal and needs-driven practices including the purchase of sachet water which accounted for 28.6% in 2010. The costs of water from these sources are typically 5–7 times higher than piped water (GSS, 2013a), indicating that access defined by availability alone is incomplete. Deconstructing the notion of access is crucial for two reasons. First, the figures on coverage conceal intra-city differentials and undercount those ‘living under poverty’ in informal and slum communities (where the daily influx of newcomers increases demand for services) in the aggregate statistics of the wider urban context (Obeng-Odoom, 2012a). Second, beyond the statistics are ‘on the ground’ realities of access including reliability, quality and affordability (Amankwaa et al., 2014). Such dynamics compel the rich who wield power to bypass certain groups to extend services, while the poor rely on contaminated water sources or illegally tap power and connect pipelines thereby causing leakages and compromising the quality of service supplied in urban areas – the case of unequal exposure but equal risk, whereby the actions of the poor extend throughout the city. This sometimes accounts for 5 University of Ghana http://ugspace.ug.edu.gh cholera and other ecological hazards reported in rich neighbourhoods (Oteng-Ababio, 2013) and frequent power outages (Doe and Asamoah, 2014). This, of course, is not to say that official statistics are irrelevant in service delivery research. However, while at the macro level the gap between water and electricity supply and demand is widening, the dynamics are more complex at the micro (city and settlement) level where access is inequitable, distribution is spatially biased and future supply is increasingly becoming uncertain as the utility systems continue to degenerate (Cobbinah et al., 2016). In such situations, income and geographical biases are overlooked and the most vulnerable groups within the urban setting are made worse off. Such observations are evident in the study localities, Korle Gonno, Abuja, Gbawe and Accra New Town, where residents often rely on private vendors who sell their water at high prices. For instance, 35 litres of piped-borne water which should cost GH¢0.0145 based on the tariff of the Ghana Water Company Limited (GWCL), is sold for between GH¢0.20 and GH¢0.50 by private vendors (Ainuson, 2010). A study by Awuah et al. (2009) revealed that only 25% of households connected to GWCL have a 24-hour access to piped water supply, 30% have an average 12-hour water service a day for five days a week, 35% have access to piped water only two days a week, with 10% living mainly in poor areas and at the urban fringes completely without access to piped water. Similarly, electricity access is poor in the study areas occasioning widespread illegal connections and meter sharing. Doe and Asamoah (2014) underline how inequities in electricity access have occasioned complementary but costly informal practices among the urban poor to navigate the frequent power cuts, and load shedding. These ‘on the ground’ realities of access affect household experiences in relation to livelihoods and poverty. The access challenge is that, apart from services being too expensive incomes are too low to afford services: because as urbanization has advanced so has poverty. This point calls for urban services to be examined in relation to livelihoods. Although, Ghana’s urban poverty 6 University of Ghana http://ugspace.ug.edu.gh has dropped by 62% between 1991/92 and 2012/13 (Obeng-Odoom, 2016), the overall urban poverty trend is bleak. The Ghana Living Standards Survey (GLSS 6) of 2012/2013 shows that nearly a quarter (24.2%) of Ghanaians are poor whilst under a tenth of the population are living in extreme poverty (GSS, 2014a). The report further estimates that although both total poverty and urban poverty have fallen in the past two decades, poverty rates have increased in the last eight years in urban areas including Accra. The informal sector, denominated by small–scale enterprises including home-based enterprises (HBEs), is an integral part of Ghana’s urban economy employing between 84 and 87% of the population (Baah-Boateng and Ewusi, 2013). Running a business from the home (e.g. catering, retail, hairdressing and dressmaking) is a frequent livelihood activity in urban Ghana. In Accra, Gough (2010) found that one-third of households run a business from their home, while Gough and Yankson (1997) reported that about 80% of the houses in peri-urban Accra have at least one person working in or in front of the house. The sector thrives on several factors including access to workspace, finance, and infrastructure service. Arguably, key among these is access to electricity and water which is critical for the daily operations of HBEs. Therefore, improving people’s access to these services for productive uses allows them to take advantage of a range of livelihood options through which they are able to secure food and income and escape poverty. More so, this will increase people’s willingness and affordability of services and make cost recovery more realistic. Notwithstanding these positives, little is known about how the dynamics of access to these services impact HBEs as most studies separately focus on household consumptive uses, and business constraints regarding access to credit. These realisations make it important to study Ghana’s water and electricity sectors, service-dependent productive activities and household poverty, and explore how the dynamics of water and electricity access affect HBEs and poverty reduction. 7 University of Ghana http://ugspace.ug.edu.gh 1.1.1. Research aim, objectives and questions The overall aim of this study is to examine water and electricity access in GAMA and its implication for home-based enterprises (HBEs) and poverty reduction. Specifically, the study will: i. Identify and examine the (proximity, reliability, quality and cost) dynamics of households’ access to water and electricity. ii. Analyse the effects of water and electricity access on HBEs. iii. Analyse households’ perceptions of poverty, and show how these relate to water and electricity access. This aim is pursued through exploring households’ experiences of urban services, livelihoods, and poverty. The key research questions in relation to these three themes are: i. What are the (proximity, reliability, quality and cost) dynamics of households’ access to water and electricity? ii. How do the dynamics of water and electricity access affect HBEs? iii. How do households perceive poverty, and how do their perceptions relate to electricity and water provision? iv. What practices are needed to be put in place to promote effective electricity and water governance? The study will explore how these issues play out in the four research localities. The questions will be addressed through a complementary mixed methods design comprising a blend of qualitative and quantitative methods (using surveys, focus groups and in-depth interviews) as well as mobile geography methods such as follow-along participant observation (FAPO). Such a methodological approach generates rich spatial observations and theories grounded in lived experience (Finlay and Bowman, 2016). The rationale is to inter alia better grasp the impact of electricity and water on households and their livelihoods and poverty status. Without contesting the priority for domestic uses of water and electricity, this study argues that the domestic focus, which characterizes the operationalization of Ghana’s urban agenda 8 University of Ghana http://ugspace.ug.edu.gh and the SDGs on electricity and water (e.g. a minimum of 20 litres per capita per day (LPCD)), ostensibly neglects the economic importance of household water and electricity. The study emphasizes that the current framing of the SDG 6 (achieving universal and equitable access to safe and affordable water for all) and SDG 7 (ensuring universal access to affordable, reliable and modern energy services) as a priority in the new urban agenda can, and should, go together with the recognition of water and electricity–dependent productive activities, which often overlap with domestic uses and have the potential to improve standard of living. There is a growing scholarship (IIED, 2010; Hall et al., 2014; Abraham et al., 2015) which argues that such binary conception in relation to consumptive (domestic) and productive (economic) uses is insufficient for grasping the complex and fluid relationship between urban services and livelihoods as it elides the tangible benefits of water and electricity for household economies. This study concludes that the increasing informality and inadequacies in electricity and water access are occurring within the context of Ghana’s decentralised governance which is associated with mixed outcomes (Adarkwa, 2012; Obeng-Odoom, 2016). Importantly, the relationships between services and governance in these studies are framed beyond the confines of the dominant decentralized governance framework. The present study contextualises holistic and people-centred approaches such as co–production (Bovaird, 2007; Mistra Urban Futures, 2015), to studying urban services, livelihoods and poverty in Ghana. 1.2. Research propositions This study is premised on the general proposition that GAMA’s service delivery is characterised by proximity, reliability, quality and affordability concerns. Not only are there differences in how households cope with these inadequacies but these differences are also differentiated by communities. Also, the dynamics of service delivery have significant effects on home-based enterprises (HBEs) and poverty. Meanwhile, the potentials of decreased 9 University of Ghana http://ugspace.ug.edu.gh business operational cost, enhanced viability of livelihood activities, and increased ability to pay for services which translates into better quality of life have been underestimated and glossed over. Therefore, the three propositions in the study are stated as: i) There are variations in the dynamics of water and electricity access in the research localities. ii) There is a positive relationship between water and electricity access and HBEs. iii) There is a positive relationship between water and electricity access and poverty reduction. 1.3. Structure of the dissertation The thesis is organised into seven chapters. This first chapter has briefly introduced the theme and research questions. Chapter two reviews the relevant literature pertinent to service provision, household enterprises and poverty reduction. This review considers various urban services theories in the water and electricity sectors, including governance, co-production and urban political-economy approach. It makes a case for the use of a governance–poverty conceptual framework, by demonstrating how the framework provides the unifying thread for operationalising the concepts and research questions as well as the analytical approach used in the study. Chapter three focuses on the contextual background information for the study. It first describes the study area, then moves on to discuss the philosophy, methods and techniques used for collecting data, and analysing and evaluating the data collected. Reflections on data quality, researcher’s positionality, and ethics are also dealt with under this chapter. It wraps up with the profile of the study participants. Chapters four, five and six constitute the presentation of the findings and analysis of the study. Chapter four responds to research question one by illustrating how access defined by coverage (or availability) alone masks a lot of nuances. It demonstrates that reliability, quality and cost are key aspects of households’ access to water and electricity which are differentiated by communities. Chapter five is devoted to research question two which examines the evidence on HBEs and how the dynamics of access to water and electricity significantly affects HBEs. Chapter six considers research questions three and four by highlighting the nexus between 10 University of Ghana http://ugspace.ug.edu.gh water and electricity access and poverty alleviation. Also, it develops a conceptual model that expounds the role of governance in reducing poverty. Finally, chapter seven consolidates the key findings and conclusions made, leading to a discussion of the contributions of the study and implications for policy, as well as reflections on possible areas for future research. 11 University of Ghana http://ugspace.ug.edu.gh CHAPTER TWO LITERATURE REVIEW AND THEORETICAL AND CONCEPTUAL FRAMEWORK 2.0. Introduction This chapter reviews the literature on governance and illustrates how urban political economy is a useful theoretical toolkit to explain the challenges of governance as far as service provision, livelihoods and poverty reduction are concerned. It is organized into four main sections. Section 2.1 provides a conceptual construct of urban governance. Section 2.2 contextualizes urban governance by describing urbanization and the attendant problems pertaining to service provision, livelihoods, and poverty in Ghana and Africa in general. This sets the stage for Section 2.3 which analyses the competing theoretical approaches and makes a case for how urban political economy is suitable for studying urban governance. Section 2.4 summarizes these ideas into a conceptual framework and shows how it is used in the analytical chapters. 2.1. Re-conceptualizing urban governance In his stimulating paper “Urban Governance in Africa Today: Reframing, Experiences, and Lessons”, Franklin Obeng-Odoom (2016) argues that governance is no longer simply an idea about managing people but also a development paradigm that is instrumental in the pursuit of growth, poverty reduction, and inclusive development. He shows how advocates and international development agencies and donors (World Bank and UN-Habitat) have supported and invested resources into developing urban governance (see Resnick, 2014). Yet to what extent urban governance is a successful development paradigm remains in question. Scholars have argued that a major setback to making this determination is framing. Fritz et al. (2009) claimed that to evaluate governance we need to focus on the process of governance. Critics like Boardman (2014) argued that it is far better to concentrate on the outcomes of governance. Others that joined the debate, (Rotberg, 2014), were split between the two 12 University of Ghana http://ugspace.ug.edu.gh polarized views: process versus outcome. These binaries have created much analytical difficulties which have (mis)led researchers to equate governance to the performance of ‘government’ (Holt and Manning, 2014) or decentralization, leaving out a broad sweep of actors and processes in urban governance. Yet, as rightly noted by Obeng-Odoom (2016), the tendency is still to focus on service delivery without engaging the bigger outcomes of livelihoods, poverty reduction, and inclusive development and how diverse arrangements of institutions at differentiated scales combine to shape and be shaped by outcomes. This study seeks to offer this missing link by building on existing effort to theorize urban governance (Obeng-Odoom, 2016). It particularly aims at reframing urban governance, in terms of its dimensions, processes and outcomes, to develop an analytical framework to evaluate it. I argue, along with Obeng-Odoom (2013, 2016), that urban governance can be examined as a cluster of meanings: decentralization, entrepreneurialism and democratization (DED). By repositioning urban governance within a political-economic context and examining its effects in practice, as well as interrogating the feedbacks of citizens to the outcomes of urban governance, its full tenor can be grasped for research purposes (Obeng- Odoom, 2013). 2.1.1. What is urban governance? In 2001 the UN started popularizing its urban governance with the launch of the Global Campaign for Good Governance (Tibaijuka, 2009). The UN’s concept differs from the World Bank’s idea of governance which sought a pro-market state. The UN saw urban governance as inclusionary and gave recognition to members of civil society as co-partners in urban politics (Obeng-Odoom, 2012b). The UN’s approach resonates with what Simon et al. (2016) call ‘a collaborative dialogue’, in which partnership beyond the public and private sectors is encouraged as well as pragmatism in improving the conditions of urban dwellers. This new approach entails a multilateral way of administering cities, drawing on stakeholders at local, 13 University of Ghana http://ugspace.ug.edu.gh national, regional and international levels. These observations bring to the fore how urban governance has been defined differently by scholars. Attempts to define governance have remained difficult leading to as many definitions as there are researchers (Hufty, 2009). The defining feature of urban governance is that the management of cities is neither the preserve of government nor of the private sector but for a wide variety of actors that interact with one another (Maloutas and Malouta, 2004). This broad meaning of urban governance has occasioned several interpretations. On the one hand, urban governance is seen as a progressive, all-inclusive process of taking care of cities. On the other hand, it is regarded as a neoliberal, minimalist state approach of running cities as businesses (Blakeley, 2005). In practice, the concept is considered to be more nuanced, transcending the state–market, socialist–capitalist frameworks (Obeng-Odoom, 2013), and encompassing the regulation of the interplay among markets, civil society, and the state. In this study, urban governance refers to co-partnership in urban development between local governments and other stakeholders such as organized groups of citizens, and traditional authorities. It connotes the shift from governance linked to hierarchies to networks as a mode of coordinating affairs (Newman et al., 2004). Indeed, at any one period, “an urban area may have varied forms of governance relations threading through it, around it and over it” (Healey, 2007, p. 16). Meanwhile, the notion of partnership does not suggest that these actors have equal power and agency. As Brenner (2004) indicates, urban governance is a concept that marries political, economic and social processes that define contemporary capitalism – itself an inherently unequal system. It seems that advocates of urban governance do not presume that the different actors have equal power or that the amount of power each actor wields is unchanging (Nuissl and Heinrichs, 2011). However, the concept is supposed to have an inbuilt ‘thermostat’ to check the abuse of the unequal power relations embedded in it (Obeng- Odoom, 2013). Recognizing these, I argue along with Obeng-Odoom (2013, 2016) and de 14 University of Ghana http://ugspace.ug.edu.gh Oliveira (2015) who argue that urban governance transcends the boundaries of efficient governments, and should be reframed as a cluster of interlocking meanings of decentralization, entrepreneurialism, and democratization (DED). These dimensions are discussed below. 2.1.2. The dimensions of urban governance Urban governance as decentralization emphasizes partnership among different sectors of society, beyond urban government (Davila, 2009). It insists on collaboration rather than a top- down direct management model by agents of local governments (Newman et al., 2004). Decentralization is important, especially if it goes beyond ‘deconcentration’ of functions (the handing over of administrative or managerial responsibility to sub-national governments) and ‘delegation’ of responsibilities (granting powers to public enterprises and other semi- autonomous bodies to manage sectors such as utilities) to ‘devolution’, which entails deeper forms of responsibilities, autonomy, and powers. Yet as Obeng-Odoom (2016) argues, to suggest that devolution is contest free, as many mainstream analysts assume, is an exaggeration. Even under conditions of devolution conflicts are possible. He argues, for instance, that where decentralization is based on partisan politics, a situation called ‘vertically divided authority’ (Resnick, 2014) can be a central aspect of power dynamics. This makes the issues of content and context relevant when analysing decentralization under urban governance. He further adds that decentralization alone is, however, inadequate in grounding the paradigm of urban governance as the tendency of its advocates is to seek participation in the explicitly political processes of decision making. The key assumption that the congruence between what participants want and what the authorities do is the essence or rather a major asset of urban governance defined as decentralization (see Crook, 2003) is too narrow. Urban governance goes beyond the polity to, for instance, the economy. 15 University of Ghana http://ugspace.ug.edu.gh Urban governance as entrepreneurialism entails a shift towards entrepreneurial tendencies in our urban politics. According to Stanciulescu (2009), urban entrepreneurialism is marketing the city to marketize it. While marketing is about branding the city in ways that will attract private investment and business, marketizing the city involves solving the urban problems and driving local economic development through the use of the market (businesses, and private capital investment). Obeng-Odoom (2016) identifies three key features of entrepreneurialism: an emphasis on urban economic growth, the partnering of governments by business, and the conscious courting of the creative class (professionals, and high level innovators). He emphasizes that the theoretical focus here is the opportunity for ‘profit’ and how it attracts private capital either through creating business friendly environment in the city or creating the conditions that will attract the creative class to move into cities and, in turn, attract businesses that will follow the creative class (Florida, 2003). Whether the starting point is attracting firms or people, the market tends to be the emphasis in entrepreneurialism. Yet, the complexities of urban society – featuring the ‘slum entrepreneur’, ‘entrepreneurial slum’, ‘civil society entrepreneurialism’ (McFarlane, 2012), and entrepreneurship in urban livelihoods and craftsmanship – would suggest that it is limiting to simply equate entrepreneurialism to the pure market. This implies, as Obeng-Odoom (2016) argues, looking at different forms of entrepreneurialism when discussing the relationship between the state and the market (as pro-market, neoliberal, or developmentalist). He reveals that because donors often favour neoliberalism, it is tempting to regard urban governance as just that: ‘neoliberal urban governance’. However, whether urban governance is neoliberal is an empirical question. Social enterprises or entrepreneurialism for example are not simply neoliberal. All such forms of entrepreneurialism appear similar in form; but they differ in detail (Obeng-Odoom, 2016). Devas (2001) argues that such features imply that it is what the poor do for themselves that determines their survival and livelihoods, even though the actions of the institutions of city governance can make that more or less difficult. For these reasons, 16 University of Ghana http://ugspace.ug.edu.gh such forms of entrepreneurialism cannot simply be called neoliberal and the whole gamut of urban governance cannot be reduced to neoliberalism, particularly when urban governance also entails democratization. Urban democratization, advocates claim, is the performance accountability of different actors. Urban democracy is typically seen to be multi-party politics, also called democratized urban local governance (see Resnick, 2014; Stren, 2014). The emphasis is on multi-party competition and their relationship to enriching decision making and urban service delivery mediated by donors (Cameron, 2014). Yet, as Obeng-Odoom (2016) suggests, this depiction is only one expression of democracy. He maintains that democracy is an idea; not any political form (e.g., multi-party politics, and one party politics). It is people’s power: it can manifest itself in many ways not just in multi-party politics. So, democratization is bigger than party politics. Urban democracy connotes deeper political economic processes that are bigger than decentralization. For instance, urban democracy extends attention from participation to control of urban resources, enterprises, and services. Harvey (2012) details how Henri Lefebvre’s idea of ‘the right to the city’ can be used to animate urban democracy. The right to the city can be understood as the equal rights to access and use of the urban commons (the total output of work by workers, and the sum total of rent collectively created by the public). Using different channels urban residents can use various human rights – the right to demonstrate, to free speech, and to association – to attain their right to the city. In summary, it is worth noting that these processes of DED are supposed to work in tandem in the pursuit of livelihoods, poverty reduction, and inclusive development, but they can also work against one another. The DED framework is holistic, as Obeng-Odoom (2016) argues, it is unlike conceptions of governance which offers no sharp distinction between the economic and political ends. In DED, the inputs, processes, and outcomes are all interlinked. In turn, those dissatisfied with urban services–in terms of equity (access and control of resources), efficiency (improved service delivery), and poverty reduction (sustainable and decent 17 University of Ghana http://ugspace.ug.edu.gh livelihoods), could use a ‘voice’ or ‘exit’ to seek redress (Obeng-Odoom, 2016). It is within this context that I analyse the urban political-economy framework of the study. The next section contextualizes urban governance in relation to service delivery, livelihoods and poverty. 2.2. Urban governance: Ghana’s situation in the African and Global context In this section, I explore the interrelationship between urbanization, urban services, livelihoods and poverty. The discussion is structured around three themes: 1) overview of urbanization, 2) the challenges cities face in relation to water and electricity services, and 3) home-based enterprises (HBEs) as a livelihood and poverty reduction strategy. These issues show the context of urban governance and the forces of continuity and change that impel the process and outcome of urbanization in Ghana and Africa in general. 2.2.1. Overview of urbanization The world is experiencing an unprecedented urban revolution involving a transition from rural to urban living. In 1950, one-third of the world’s population lived in cities, today cities are home to 54% (3.9 billion) of the world’s population (7.3 billion), and this is expected to reach 5 billion by 2030 and 66% (6.3 billion) by 2050 (UNDESA, 2015). However, this growth in urban population remains uneven throughout the world. Africa and Asia are less urbanised (40% and 48% respectively), although they are urbanising faster than the other regions. Both regions are projected to reach 56% and 64% urbanise status by 2050 (see Table 2.1). Today, about 40% of Africa’s total population live in urban areas (UNDESA, 2014). Africa’s urban population is increasing more rapidly and the highest rate of urban growth is observed in cities such as Cairo (Egypt), Lagos (Nigeria), Accra (Ghana) and Nairobi (Kenya). While during the past decade, the rate of Africa’s total population growth is about 2.3% per annum, the urban population is increasing at an average rate of 3.3% (UNDESA, 2014). 18 University of Ghana http://ugspace.ug.edu.gh Table 2.1: Population of the world and major areas, 2015, 2030, 2050 and 2100 Population (millions) Major area 2015 2030 2050 2100 World 7 349 8 501 9 725 11 213 Africa 1 186 1 679 2 478 4 387 Asia 4 393 4 923 5 267 4 889 Europe 738 734 707 646 Latin America and the Caribbean 634 721 784 721 Northern America 358 396 433 500 Oceania 39 47 57 71 Source: UNDESA, 2015 The narrative that rapid urbanization is occurring across much of sub-Saharan Africa (SSA) has sparked debate. Critics describe urbanization as ‘parasitic’, ‘premature’, or the ‘migration of poverty’ (Ravallion et al., 2007). These labels suggest that urbanization has outpaced economic development. However, a new angle on this debate has now arisen: it is economic development that drives urbanization, not the other way around (Potts, 2016). Studies by Deborah Potts (2009, 2012a, 2012b) show that urbanization in Africa is not as rapid as skeptics claim. The point is not that there is no expansion of urban settlements. Rather, the urban growth rates in most Africa countries are generally not higher than the national population growth rates. These perspectives complicate some of the key policy positions in global reports about rapid urbanization in Africa because a different policy mix is needed to manage the urban population. This is against the backdrop that there is evidence of counter- urbanization (Côte d’Ivoire), slow urbanization growth rates (Ghana) and high net in- migration rates (Kenya) occurring on the continent (Potts, 2012c). Ghana has consistently exhibited rapid urbanisation and is one of the few countries in SSA to have crossed the 50% urban milestone in 2010. This growth is due to rural–urban migration, natural population increase in towns and cities, and the reclassification of rural jurisdictions to urban once they surpass the threshold populations of 5,000 (GSS, 2013a). If current population growth trends persist through 2030, Owusu and Oteng-Ababio (2014) predict that 19 University of Ghana http://ugspace.ug.edu.gh Ghana’s population would likely exceed 33 million, with almost two-thirds residing in urban areas. It is projected that Ghana’s urbanization level will keep rising and reach 65% in 2030 (Farvacque-Vitkovic et al., 2008) and surpass about 70% by mid-century (World Bank, 2014). Ghana’s urbanization is characterized more by continuity than change, with the large cities increasingly serving as migration nodes. According to the 2010 census, over 40% of the urban population reside in the six largest metropolitan areas – Accra (1,848,614), Kumasi (2,035,064), Sekondi-Takoradi (559,548), Tema (402,637), Tamale (371,351) and Cape Coast (183,727). However, the proportion of the national urban population living in these metropolises relative to the total urban population has declined over time, falling from 53.2% in 1970 to 42.7% in 2010 (GSS, 2013a). The dominance of Accra and Kumasi has been underscored, with these two cities accounting for nearly 31% of the national urban population and 52% of their respective region’s urban population. It has been noted that the degree of urbanization in a country parallels its stage of economic development (Ravallion et al., 2007), although Turok (2013) argues that this relationship is far more complicated. Following Cheshire’s (1995) thesis (that the early stages of economic development typify rapid urbanization in the national core area while the growth of intermediate-sized cities and suburban rings become more prominent at the intermediate stage) the future growth of urban population may be slower than anticipated in the metropolitan centres of the core area. Nevertheless, these centres would remain attractive to migrants, but with limited access to infrastructure services and employment opportunities slum urbanism takes root, especially in inner-city zones (Owusu and Oteng-Ababio, 2014). Greater Accra forms the country’s most densely populated and urbanized region with a density of approximately 1,236 persons per square kilometer compared with a national average of 103 (GSS, 2013a). Metropolitan Accra is estimated to attract a daily daytime 20 University of Ghana http://ugspace.ug.edu.gh population of more than 3.5 million in addition to its resident population (Owusu, 2008), a figure higher than the official population estimate. The high population concentration in the capital city has impelled population spillover into surrounding districts, and the speedy conversion of settlements in Accra’s urban field into a conurbation termed as the Greater Accra Metropolitan Area (GAMA) (Songsore et al., 2009). Much of GAMA’s population growth is occurring outside Metropolitan Accra in adjoining municipalities that until recently were largely rural. As noted by Owusu and Oteng-Ababio (2014), government policy response has been to impose the continuous fragmentation of GAMA into smaller local municipalities as part of Ghana’s decentralization programme, rather than expand Accra’s boundaries to accommodate an ever-expanding metropolis. Unsurprisingly, the 2010 census reported that Kumasi had surpassed Accra in terms of the official population count, but the numbers for Accra overlook the fragmentation of the metropolis and thereby undercount its true population size. In effect, the shrinkage of the official metropolis as a result of continuous fragmentation into new local jurisdictions suggests that Metropolitan Accra is likely to grow more slowly or even decline. This development is further fueled by increasing land and housing prices, as well as the process of gentrification which is pushing the urban poor out of the city core into the adjoining municipalities. However, the growth of GAMA is unlikely to slow down both in terms of population and territorial expansion. Owusu and Oteng-Ababio (2014) project that by 2025, the radius of the built-up area of the metropolis will extend outward for over 50 km in every direction. It will cover districts in the Central region (Kasoa and Awutu Municipalities), Eastern region (Akuapim South/Nsawam-Adoagyiri and Aburi Municipalities), and the Dodowa/Dangme East districts in the Greater Accra region. To conclude, in spite of the consensus on the trend of Ghana’s urban future, the spatial pattern of urban expansion and the dynamics of individual metropolitan areas have been viewed with 21 University of Ghana http://ugspace.ug.edu.gh ambivalence. Owusu and Oteng-Ababio (2014) indicate that Ghana is likely to produce an urban society with two faces. On the one hand, there will be world-class cities with quality infrastructure services, that will be inhabited by the middle and upper classes, and, on the other hand, teeming cities characterized by poor housing and informal settlements and the attendant inadequate water, electricity and sanitation that will be inhabited by lower-income groups and the poor. More so, this will be compounded by limited livelihood opportunities as small-scale business activities in low-income communities is not well developed (World Bank, 2015a). The consequence of this unfolding urban scenario is that the neglect of the urban masses will affect the overall city-life of the affluent. Thus, in spite of their appeal to most Ghanaians as the ‘place of choice’ and ‘place of economic opportunity’, cities experience a variety of problems which have political–economic implications (Obeng- Odoom, 2013). The next sub-section considers these problems by focusing on problems of water and electricity, livelihoods and poverty. 2.2.2. Water access situation Globally, great strides have been made in water supply. Indeed, the MDG target for improved drinking water was met in 2010 – well ahead of the 2015 deadline. It is estimated that over 91% of the total world and 96% of urban population currently have access to improved water supply (WHO and UNICEF, 2015). Contrary to the past when SSA countries were particularly not mentioned in such success stories, today, the situation is different. In SSA, an extra 427 million people gained access to water between 1990 and 2015 (WHO and UNICEF, 2015). This contributed to 87% of the urban population having access to water. Meanwhile, the international development discourse captured in the Sustainable Development Goals (SDGs) which prioritizes access to improved water – or widening coverage of infrastructure to water supply – as the main indicator of progress in the water sector is problematic (Cobbinah et al., 2016). Within this framework, it may be argued that many countries in SSA, including Ghana, are success cases because, although percentage of 22 University of Ghana http://ugspace.ug.edu.gh the urban population in Ghana increased from about 31% to 54% between 1984 and 2015 (World Bank, 2015a), the share of the urban population with access to water increased from 54% to 87% (WHO and UNICEF, 2015). Beyond this ‘wide meaning’ of access to water, defined by connection rates, are also ‘deep meanings’, which are often glossed over in the international reports (Amankwaa et al., 2014). Studies on urban water supply in Africa overly focus on macro level statistics without paying particular attention to the dynamics of supply in relation to micro level spatial inequities and socio-economic characteristics (Stoler et al., 2012). This is not to say that official statistics are irrelevant in urban services research. However, over concentration on statistics elides the intricacies and deeper meanings of access, which are critical in addressing inequities in supply. This study problematizes access to interrogate issues such as water sources (availability), distance (proximity), regularity of flow (reliability), usability (quality) and cost (affordability). To appreciate these dimensions of access to water in Ghana, it is important to explore how the supply, management, and pricing of water have evolved over the past decades. The detailed account of these changes has recently been given by Anna Bohman (2010), and is therefore not duplicated here. Rather, what follows is an appraisal of the literature on the evolution in the water sector. Bohman’s (2010) account shows that urban water supply in Ghana began prior to World War 1 in Accra as a coping strategy to deal with an outbreak of a bubonic plague and to prevent future related water borne diseases caused by the reliance on rain water, wells and lagoons. During the independence era, the Ghana Water and Sewerage Corporation (GWSC) was in charge of providing and managing domestic and industrial water supply and sewerage services. The corporation faced difficulties with regards to the provision of water meters, deteriorating infrastructure, failure to meet the rising water demand and 23 University of Ghana http://ugspace.ug.edu.gh rising debt. These setbacks revealed government inefficiencies in terms of corruption, lack of skill, inadequate capital, and human resources (Bohman, 2010). Between 1983 and 1993, consistent with the thinking in the global economy at that time, the government implemented the Economic Recovery Programme (ERP) of the Structural Adjustment Programme (SAP) (for a thorough account of the programme, see Hutchful, 1995, 2002). Fundamental to the ERP was the neoliberal policy of market extension, premised on the view that markets are efficient while governments are corrupt and inefficient managers. Bohman’s account identifies specific policies within this framework, including the introduction of user fees and removal of subsidies. Consequently, GWSC was restructured to a pro-business management style and later metamorphosed into a limited liability company, Ghana Water Company Limited (GWCL) (Fuest and Haffner, 2007). Although the reform was consistent with the market-oriented approach of international development agencies, local institutions such as the National Coalition against the Privatization of Water (NCAP) suggested non-market options. The resistance by civil society organizations (CSOs) against a pure market strategy coerced the neoliberalist ideologues into concessions (Yeboah, 2006). A monolithic public-private model was changed to a multi- variant model of public-private partnerships (PPPs), summarized in Table 2.2. The government opted for affermage– public ownership with private sector management approach which gave considerable power to the private manager to determine price, market strategy and operation capacity (Whitfield, 2006). Prior to adopting affermage, the restructuring process had completed in 1998 with the management of urban water decoupled from rural water (Bohman, 2010). The latter was managed by the Community Water and Sanitation Agency (CWSA) under the Ministry of Local Government and Rural Development (MLGRD). The Water Resources Commission (WRC) was created as a local agency to oversee the restructuring and investments in the 24 University of Ghana http://ugspace.ug.edu.gh sector (Fuest and Haffner, 2007). Although these developments characterize a process of deregulation, they also show, at least to a limited extent, how deregulation can sometimes mean re-regulation (Bakker, 2001), especially when we consider that the Public Utilities Regulatory Commission (PURC) was formed in 1997 to regulate water management. PURC adopted an automatic price adjustment formula to trigger a gradual adjustment to economic cost recovery. Table 2.2: Types of public-private partnerships (PPP) Option Asset Funds and plans Recipient of Income Typical ownership infrastructure income from duration investment tariffs (years) Service contract Gov’t. Gov’t. Gov’t. Gov’t. pays 1-2 private company Management Gov’t. Gov’t. Gov’t. Gov’t. pays 3-5 contract private company Lease or Gov’t. Gov’t. Private Private company 8-10 affermage pays gov’t. Concession Gov’t. Private Private Private company 20 pays gov’t. Build own operate P rivate, later Private Private Gov’t. acquires 20 and transfer gov’t. assets Divestiture Private Private Private One-off payment Permanent for gov’t. Source: Obeng-Odoom, 2012a In the end, the government made some concessions, abandoned affermage, and opted for a management contract following critiques of neoliberal policies and local CSOs protest. Aqua Vitens Rand Limited, a joint venture company formed by Rand Water Services Pty located in South Africa and Vitens International based in the Netherlands won the management contract in November 2005 to manage urban water in Ghana till June 2011. Subsequently the Ghana Urban Water Company Ltd (GUWCL), a subsidiary of GWCL owned entirely by the Government of Ghana, was formed to manage urban water. This narrative of the evolution of urban water management has implications for water access in urban Ghana. 25 University of Ghana http://ugspace.ug.edu.gh Two main urban water supply sources exist: public utility, and private alternative sources. Although the total installed capacity of the urban water systems operated by the utility provider, GWCL, is 737,000m3 per day, only 551,451m3 per day is produced. However, with an estimated demand of 939,070 m3 per day, close to half (41%) of the total national water need is unmet (Cobbinah et al., 2016). In Accra for example, the two main water supply systems, Kpong and Weija produce about 88% and 83% respectively, of their installed capacity (Adank et al., 2011). As shown in Table 2.3, although the proportion of urban population with access to standpipes increased from 16% to 20% between 1992 and 2013, the proportion with access to piped water considerably declined from 27% to 15% in the same period (GSS, 2014b). Indeed, the gap between production and consumption in Accra widened from 74 million gallons per day in 2008 to 170 million gallons per day in 2010 (Essel, 2010). It is thus not unexpected that only 35.6% of urban Ghana actually enjoy GWCL water supply (GSS, 2014b). The inadequate and unreliable water supply have made water rationing a daily experience of urban residents. In Accra, for example, it is estimated that only 25% of households connected to GWCL have a 24-hour access to piped water supply, 30% have an average 12-hour supply a day for five days a week, 35% have access for only two days a week, with 10% completely without access to piped water (Awuah et al., 2009). Table 2.3: Trends in urban water supply sources in Ghana (%) Source of water 1992/93 1998/99 2005/06 2012/13 Piped water 27.0 22.5 23.1 15.0 Stand pipe 16.2 20.3 21.8 20.6 Neighbour/private 12.2 26.0 23.0 13.3 Water vendor 8.3 3.6 5.3 30.8 Borehole 4.6 2.8 9.9 11.2 Well 16.0 12.6 12.4 6.5 Natural sources 15.7 12.2 4.5 2.6 Source: Compiled from the Ghana Statistical Service (GSS) 2014b These inequities in public supply have impelled urban residents to resort to alternative sources – vendors, boreholes and neighbor/private sources – to meet their water needs. A striking 26 University of Ghana http://ugspace.ug.edu.gh phenomenon is the growing influence of water vendors from 8.3% in 1992/1993 to 30.8% in 2012/2013 as illustrated in Table 2.3. Such developments impact the cost of water. Piped water is said to be cheap, inclusive, or pro-poor, if it is supplied at or below the official rate (PURC, 2005, p. 9). However, private providers (usually wealthy individuals) could sell water at rates which are about 12 times higher than the official rates (Ainuson, 2010). So lucrative is the business of supplying water that, in Accra alone, the number of water vans – a proxy for interest in the water business – rose from 5 to over 600 between 1997 and 2009 (Kanin, 2010). From this perspective, households not connected to the utility network adopt several strategies to increase their water access thereby paying the highest cost for water. 2.2.3. Electricity access situation The demand for electricity exceeds supply in many countries in SSA where access rate is estimated at 35% (WB and IEA, 2015). Similar to water, the statistics on access need to be complemented with ‘on the ground’ realities such as reliability, quality and affordability (WB and IEA, 2015). In many SSA countries, including Ghana, electricity supply is erratic and exemplified by blackouts and power rationing that incurs costs associated with damages to equipment and disruption of productive activities (World Bank, 2009). Ghana generates about 64% of its electrical power from hydro sources (GEC, 2015). Figure 2.1 presents Ghana’s electricity supply structure. The limited water inflows have induced the country to rely on thermal power plants fueled by oil and gas. The inability of these plants to operate at full capacity due to limited and expensive fuel supply inhibits frequent power supply. Table 2.4 presents an overview of electricity generation in Ghana. About $9 billion is required between 2014 and 2019 to raise the needed investment capital to finance the numerous initiatives including oil and gas (which require $5.5 billion) (Eshun and Amoako-Tuffour, 2016). While Ghana has committed itself to universal electricity access by 2020, the real challenge is the capacity to meet this target and, most important, to ensure that supply is 27 University of Ghana http://ugspace.ug.edu.gh reliable and affordable. To illustrate why these dimensions of electricity access remain a challenge in Ghanaian cities, it is imperative to explore how the management of electricity have evolved over the past decades. A thorough review of the trends in Ghana’s power sector has been given by Eshun and Amoako-Tuffour (2016), and is therefore not duplicated here. Figure 2.1: The supply structure of Ghana’s electricity sector Generation Transmission Distribution End-users Volta River ECG Ghana Authority (VRA)  Residential Independent Enclave Power Power Producers  Industrial GRIDCo Company (IPP)  Commercial Ghana Electricity NEDCo import Sou rce: Gyamfi et al. 2015 Table 2.4: Installed and projected electricity generation in Ghana (2014 and 2015) Generating Installed capacity, MW Dependable capacity, MW station/plant 2014 2015 2014 2015 Hydro Akosombo 1020 1020 960 900 Kpong 160 160 140 140 Bui 400 400 380 340 Thermal TAPCO 330 330 300 300 TICO 220 330 200 320 TT1PP 126 110 110 100 TT2PP 49.5 49.5 45 45 MRP 80 80 40 70 Takoradi T3 132 132 120 36 Sunon-Asogli 200 200 180 180 CENIT 126 110 110 100 Karpower - 250 - 225 Renewables Solar 25 2.5 22.5 2 Total 2868.5 3174 2607.5 2758 TAPCO- Takoradi Power Company; TICO- Takoradi International Company; TT1PP- Tema Thermal Plant1; TT2PP- Tema Thermal Plant2; MRP-Mines Reserve Plant Source: GEC, 2015, 2016 28 University of Ghana http://ugspace.ug.edu.gh Ghana’s electricity supply dates back to 1914 when the Gold Coast Railways Administration started the first thermal power generation and public electricity supply in the municipalities of Sekondi and Takoradi (ISSER, 2005). Eshun and Amoako-Tuffour’s (2016) account shows that major reforms in the power sector began under the Structural Adjustment Program of 1986 by the World Bank to encourage greater efficiency and private sector participation. The reforms were to regulate the monopoly power of generation, transmission, and distribution towards deregulating the market for competition and open access to transmission lines to permit new market participants to share in the value chain. The reforms also sought to adjust the electricity tariffs that had resulted in low investment, substantial deficits of the utility companies, and poor service quality (GRIDCo, 2010). The restructuring process led to the Energy Commission (EC) and the PURC being formed in 1997 to establish performance standards for utilities and maintain a healthy competitive power sector that balances the interests of utility providers and consumers (Eshun and Amoako-Tuffour, 2016). The Ghana Grid Company Ltd. (GRIDCo) was also formed in 2006 by the Energy Commission Act (541). GRIDCo’s main function is to facilitate electricity transmission via delivery at power distribution centres to all market actors, while Volta River Authority (VRA) continues its operation as a power generation company, and Electricity Company of Ghana (ECG) as a power distribution company. Presently, the Electricity Regulation Policy, 2008 and the National Energy Policy, 2010 are policies being implemented to address the poor electricity delivery and financial performance of the utility companies. The policies aim to expand generation assets by investing in renewable energy resources such as wind, solar PV, and modern biomass and energy efficiency measures (Gyamfi et al., 2015). The policies require an increment in the installed generation capacity to 5000 MW by 2016 to meet the estimated 66% electricity demand (GEC, 2012). This means almost a double of the current generating capacity of 2,946 MW 29 University of Ghana http://ugspace.ug.edu.gh (Eshun and Amoako-Tuffour, 2016). There are a number of issues, however: (a) the Bui Dam currently has an installed capacity of 400 MW; (b) the Aboadze thermal plant has an installed capacity of 550 MW and a maximum capacity of 682 MW; (c) the largest solar plant, the Nzema solar plant is expected to produce 155 MWh; and (d) the mix for renewable energy is also expected to increase to about 10%. Suggesting that an energy capacity gap of 1049 MW still remains. These factors largely account for the recent power crises (2002, 2007 and 2014– 2016) in the country (Gyamfi et al., 2015) and foreground government’s recent intention to enter into a concessionary agreement with a private entity that would invest in and manage ECG. Figure 2.2 shows the daily load profile of Ghana’s electricity sector. Figure 2.2: A typical daily load profile of Ghana showing peak demand periods Source: GRIDCo, 2010 These developments in the power sector have manifested in geographical biases and rising disparities in electricity access. The national access rate is 80.5%, with the access of urban inhabitants at 78% and rural population at less than 30% (The Presidency of Ghana, 2016). However, the World Bank (2015b) indicates that 62% of poor urban households did not have access to electricity in 2012, in contrast to the ‘better off’ who enjoyed centrally provided electricity of 74%. Mindful of these inequities, governments usually target energy policies to benefit the poor, albeit the poor have failed to reap the benefits. Meikle and Bannister (2003) 30 University of Ghana http://ugspace.ug.edu.gh attribute this to the nature of stepped (exorbitant) electricity charges policy, whereby the unit cost of electricity is tied to the amount consumed (i.e. the higher the amount used, the higher unit cost). This strategy fails to take into account the conditions of the poor (many of whom share meters). This means that the combined electricity consumption quickly reaches the higher unit cost rate and thus falls outside the lifeline category. The access dynamics is complicated by the fact that about 40% of electricity is consumed by the residential sector, with consumption growing at an average of 6.3% over the last decade (GEC, 2011). The industrial sector accounts for nearly half of the consumption, with the rest consumed by non-residential customers, which are mainly small-scale businesses (GEC, 2011). The demand from the residential sector is projected to exceed 4400 MW by 2020 (Ayittey, 2012), thereby exceeding industrial consumption. This can be attributed to the rising per capita income, increasing home businesses and penetration of appliances in higher income homes. This is worrying because already Ghana’s urban economy is dominated with electricity-reliant household businesses. As poor (tenant) households continue to grow, they will naturally engage in such businesses, which will ultimately increase residential electricity consumption. The next section turns to consider how these issues relate to livelihoods and poverty reduction. 2.2.4. Home-based enterprises (HBEs), urban services and poverty reduction The issue of livelihood is central to all three aspects (DED) of urban governance: to decentralization because how the polity is organized and whether decentralized units survive hinge on whether citizens can pay for urban services which, in turn, depends on questions of livelihoods and income; to entrepreneurialism because livelihood issues are at the core of productive activities, and to democracy because of citizens’ participation and demand for improved services and accountability. 31 University of Ghana http://ugspace.ug.edu.gh Promoting income-generating activities for the urban poor is essential to inclusive growth. In SSA, most of the labor force in low-income households is engaged in micro and small businesses, commonly called (informal) household enterprises. While some argue that informal economy growth is caused by rapid urbanization and a lack of entrepreneurialism (World Bank, 2000), this explanation ignores the structural roots of the informal economy. For example, during the ‘lost’ decade of the 1980s, the Ghanaian economy underwent macro- economic reforms, such as large-scale retrenchment in the civil service and privatization of some state enterprises, to make it efficient. Such neoliberal policies led to a fall in formal sector employment (Obeng-Odoom, 2013). Neoliberalism became prominent in the 1980s, coinciding with the influence of public choice theory in giving legitimacy to privatization and deregulation in public policy (Stilwell, 2006). The intellectual rationale for neoliberalism has been discussed (Brohman, 1995) and applied to the developing countries’ quest for economic development (Todaro and Smith, 2006). The overarching claim and emphasis of neoliberalism is that a policy orientation that promotes competition among firms and guarantees the freedoms and choices of economic agents, open markets, and profit maximization is good for job creation, equity and poverty reduction (Dollar and Kraay, 2002). The period from 1983 until the present is described as the neoliberal era in Ghana, although the foundations for the implementation of austerity measures under the SAP, were laid from 1983 to 1993. This period was marked by attempts to enhance economic growth and reduce inflation in order to stabilize the macro-economy. Thereafter, policies like the ‘Accelerated Growth Strategy’ and ‘Ghana Vision 2020’ were pursued with the aim of promoting the private sector as the engine of growth to ensure greater access to pro-poor services and resources (Aryeetey and Harrigan, 2000). Since then, successive governments have been pursuing similar austerity measures with the most recent one starting in 2015. 32 University of Ghana http://ugspace.ug.edu.gh These transformations have contributed to structural changes in the urban economy, with corresponding favoruable impact on the private sector with respect to investment, job creation and employment growth (GIPC, 2010). On average, the contribution of the private sector to gross capital formation has been significantly higher than the public sector in Ghana since 1987. Out of a total gross capital formation1 (as a share of real GDP growth) of 1.7 in 2008, the private sector contributed 1.1 or 64.7% (AfDB and OECD, 2008). Contrary to claims that the private sector would expand to absorb the retrenched workforce in the public sector, there was only a slight rise in the share of private formal sector jobs while, in relative terms, there was a sharp rise in private informal sector2 in total employment (GSS, 2008). Thus, we see in neoliberalism a force of ‘creative destruction’, a situation in which one positive – private sector employment growth – tends to undermine another – informal employment conditions (Harvey, 2007). Although some associate informality with illegality, an estimated 70-80% of the non-agricultural workforce in Ghana is engaged in informal activities which are legal, even if they are not registered or not located in compliance with urban zoning (Baah, 2007). Indeed, in Ghana, it is perfectly legal to do business in one’s own name without formal registration (Kwankye and Anarfi, 2011). The literature on informality revolves around different perspectives on whether and how the informal and formal sectors are interdependent (Portes et al., 1989). Three schools of thought have been distinguished: dualist, structuralist and legalist (see Anyidoho and Steel, 2016). The dualistic approach to informality perceives informal economic activities as “what people really do for themselves beyond the reach of state regulation” (Hart, 2008, p. 6). In this view, informal self-employment refers to what people do to earn some income when they are 1 The contribution to the total fixed assets in the country 2 The term informal sector was first used by Keith Hart (1973) to describe how the Frafra people in Accra earned their livelihoods. 33 University of Ghana http://ugspace.ug.edu.gh marginalized by formal structures and unable to obtain wage employment. The legalist school views the existence of informal businesses as a rational response by informal enterprises to over-regulation by government bureaucracies and barriers to formal, legal status (de Soto, 1989). This view presents informality as a way “to avoid the costs, time, and effort of formal registration” (Chen, 2012, p. 5). Contrary to the viewpoint that the informal and formal sectors operate in separate spheres, “the structuralist school sees the informal economy as subordinated economic units (micro- enterprises) and workers that serve to reduce input and labor costs and, thereby, increase the competitiveness of large capitalist firms” (Chen, 2012, p. 5). This view is consistent with the informalization of jobs when businesses outsource activities to private contractors instead of employing workers in-house. Although this approach theorizes linkages between informal workers and formal businesses, it is one in which the informal is dependent on the formal. The voluntarist approach has emerged as a fourth school of thought. This ‘revisionist’ view posits that, workers choose informal employment voluntarily and, given their characteristics, have higher utility in an informal job than in a formal one (Maloney, 2004). The voluntarist school raises doubt about the preferability of formal jobs along the various dimensions mentioned in the literature on labor market segmentation. For example, if formal employment is linked with the provision of social protection benefits (e.g. pension), in SSA such benefits might be undesirable in the eyes of the employed due to low payment amounts, and possibly doubtful nature of fringe benefits connected to formal jobs as well as existing social capital. The concept of informal economy comprises both the informal enterprises and production sector, and informal employment which includes “all employment arrangements that do not provide individuals with legal or social protection through their work…whether or not the economic units they work for or operate in are formal enterprises, informal enterprises, or households” (ILO and WIEGO, 2013, p. 3). In the informal economy, “production, distribution, and employment relations tend to fall at some point on a continuum between 34 University of Ghana http://ugspace.ug.edu.gh pure formal relations (regulated and protected) at one pole and pure informal relations (unregulated and unprotected) at the other….(with) the formal and informal ends of the continuum often dynamically linked” (Chen, 2005, p. 8). In such an integrated approach, not only the behaviors of informal actors but their outcomes are conditioned by formal economic, policy, and institutional environments and actors to which they are linked, whether directly or indirectly (Meagher, 2013). In SSA, the share of informal employment ranges from 33% to 82% and accounts for 35% to 62% of non- agricultural gross value added (ILO and WIEGO, 2013). The informal sector employs between 84 and 87% of Ghana’s workforce (Baah-Boateng and Ewusi, 2013). Evidence indicates that the majority of micro and small enterprises consist of household enterprises; that is, a self-employed individual working alone (72%) or with a family member or casual workers (21%) (World Bank, 2011). The importance of informal sector activities that operate from the home is increasingly being recognized (Strassmann, 1987; Gough, 1996; CARDO, 2000; Gough et al., 2003; Tipple, 2005; Gough, 2010). Recent studies of home-based enterprises (HBEs) reveal a variety of analytical approaches, including factors affecting the incidence and frequency of HBEs in South Africa (Tipple, 2005), longitudinal analysis of economic adaptability, market flexibility and social continuity amongst HBEs in Ghana (Gough, 2010), and the general characteristics and roles of HBEs in local communities and economies of varying scale (Gough et al., 2003). Gough (2010), for example, found that HBEs are a common feature of urban settlements in the global South, although the number of households operating HBEs ranges from a fifth to almost a half, with a slight tendency for the frequency to be highest in African cities. The increasing frequency of HBEs is in line with the increasing informalization of economies in SSA (Haug, 2014). The frequency of HBEs also varies substantially within cities, between various neighbourhoods, depending on local socio-economic levels, settlement age and service levels (Gough, 2010). 35 University of Ghana http://ugspace.ug.edu.gh Generally, those operating HBEs are predominantly women and, for many, HBE is their primary source of income. For example, 73% in South Africa (Gough et al., 2003), and 93% in Bangladesh (Ghafur, 2002) are run by women. This is because the home is perceived as a female space. Gough (2010) further adds that HBEs make an important contribution to household economies, provide needed services, hence it should be supported with infrastructure development, access to credit and business skills training programmes and regulations that will make them formal. However, a notable theme that is rarely analysed in the HBE literature concerns the implications of access to infrastructure service such as workplaces, water, and electricity. Evidence suggests that as settlements become consolidated with improved service delivery, the opportunities for income-generating enterprises have been shown to expand and thrive (Fox and Sohnesen, 2012). A theme which repeatedly emerges in the home-based enterprise literature is the way in which HBEs have been defined differently by scholars (CARDO, 2000), and the fluid and contested nature of the concept of the home space (Blunt and Dowling, 2006). For example, in Latin America, enterprises tend to be operated from rooms within the home (Gough and Kellett, 2001), whereas in African cities, many enterprises are operated outdoors either in front of the house or in interior courtyards (Yankson, 2000a, 2000b). In the Ghanaian context, factors such as compound family houses (Korboe, 1992), and high person-per-room occupancy and tenancy rate (Konadu-Agyemang, 2001) are fundamental to delineating the spatial limits of income-generating activities that take place in the home. In line with Tipple (2005), this study conceptualizes an HBE as one which occurs in or very close to the home (inner room, outdoors in interior courtyards or in front of the house, and nearby public spaces) rather than in a commercial or industrial building or area. It comprises activities that involve both a home-based component (e.g. making food) and a non-home- based aspect (e.g. selling food on the street). In contrast, enterprises located in the settlement but operated by outsiders were not counted as HBEs. Also, business operators who used the 36 University of Ghana http://ugspace.ug.edu.gh home environment to store products or equipment but worked, for example, in their previous place of residence were excluded. The rationale is to understand households’ access to water and electricity services in relation to their home businesses. In urban Ghana, a wide variety of enterprises, such as shops selling daily necessities and cooked food, hairdressers, and tailors/seamstresses, are run from the home as a livelihood activity. For example, in Accra, Gough (2010) found that one-third of households ran a business from their home while Yankson (2000a) reported that 49% of homes housed a business. In a study of peri-urban Accra, Gough and Yankson (1997) found that about 80% of the houses have at least one person working in or in front of the house. Another study by Sinai (1998) found that virtually every compound house in Kumasi was used for income generation by one or more household and, out of 600 households, around 25% used their home for HBEs. Informal activities have increasingly been recognized as a vital element to address household poverty. Charmes (2000) argues that there is an overlap when it comes to working in the informal economy and being poor, as average incomes are lower than in the formal economy. However, recent studies indicate that though incomes of HBEs are low they are very significant in poverty alleviation (Tipple, 2005; Mensah, 2015). At the same time, it is evident from existing studies that income alone is unlikely to result in improvements in well-being (GSS, 2013b). Other factors that have arguably been the particular focus of local governance strategies include the lack of access to services (Mitlin, 2008). Yet, official statistics on urban poverty rarely consider the extent and quality of service delivery. Meanwhile, the deficiencies in services such as water, electricity and sanitation are well documented and these inadequacies do not only affect households but also household enterprises (World Bank, 2011). The literature amply elaborates on the strong linkages between urban services and poverty. The intention here is not to duplicate the review of empirical studies on these linkages (see 37 University of Ghana http://ugspace.ug.edu.gh Franks and Cleaver, 2007; Obeng-Odoom, 2013), but to underscore the fact that these linkages more often than not operate through or are modified by the livelihood activities of the poor. Such a position, as this study argues, requires placing particular emphasis on the developmental significance of domestic and productive uses of household water and electricity services (IIED, 2010; Abraham et al. 2015). This makes water and electricity a socio-economic ‘good’, but it can also become a socio-economic ‘bad’ when the dynamics of reliability, quality and cost characterize service provision. The poor and marginalized population, which with limited resources remain unable to adopt coping strategies, are most affected by consequences of poor access to water and electricity. Improvement in urban services is therefore essential to enhance the viability of household enterprises thereby contribute to poverty reduction. Peoples’ experiences of poverty go far beyond money (or income). The argument is that, there are some residents in low-income areas with higher incomes who are unable to find improved access to services. Consequently, the lack of services in urban localities means that the poor spend a considerable proportion of their income to access water and electricity (McGranahan et al., 2006). A study in Accra by the World Bank (2010) showed that, among households whose monthly income is below GH¢100, only 57% have a GWCL piped water connection. In contrast, over 70% of those within GH¢501–1,000 income bracket have a GWCL connection, while 83% of those with incomes between GH¢1,001–2,000 have a piped water connection. On average, low-income people in Accra consume only half of the amount of water consumed by the rich (Adank et al., 2011). Because of this strong nexus between poverty, income, and the quality of municipal services, there is a corresponding correlation between place and people poverty such that there are fairly distinct groups of poor people located in areas with poor services (Obeng-Odoom, 2012a). This relationship is bidirectional: poor people live in poor neighbourhoods and poor 38 University of Ghana http://ugspace.ug.edu.gh neighbourhoods breed poor people. Such forces of circular and cumulative causation are strongly at play in the livelihoods and services sectors in Ghana’s cities (Obeng-Odoom, 2011). The implication of these dynamics is that the majority of low-income residents of cities, for whom securing water and electricity at higher rates has adverse consequences on the consumption function and livelihood activities, are made worse off. This is the context that the poor have to overcome if they are to move out of poverty. Suggesting that the success of the SDGs revolves around service provision, employment creation, eradication of extreme poverty, and inclusive economic growth (SDSN, 2013). The fundamental prerequisite for this is a hybrid urban governance system to drive a transformation agenda (Simon and Adam-Bradford, 2016). For nearly three decades, the Ghanaian economy has almost always grown more quickly than have the economies of other countries in SSA. Ghana’s economic growth has been associated with a shift of the economy from agriculture to services. According to the World Bank (2015b), the share of agriculture in GDP declined by nearly 50%, from over a third of GDP in 1991 to 23% in 2012 while services accounted for more than half of per capita GDP growth (see Table 2.5). This rapid growth translates into accelerated poverty reduction. The poverty rate fell by more than half between 1991 and 2012, from 52.7 to 21.4% (World Bank, 2015b). In 2012, the poverty rate in Ghana was less than half the African average of 43%, while in 1991, it had been only 10% lower than the African rate. The extreme poverty rate declined even more quickly, dropping from 37.6 to 9.6% between 1991 and 2012 (World Bank, 2015b). What is troubling is that inequality levels have been rising steadily over the years both in terms of income, access to, and the control of urban services and resources. This is worrying particularly because inequality slows down poverty reduction (Fosu, 2009). 39 University of Ghana http://ugspace.ug.edu.gh Table 2.5: GDP growth decomposition, by per capita value added, 1991–2012 Contribution of Contribution Contribution of within-sector of changes in intersectoral Total Sector changes in output employment shifts per worker Agriculture 17.66 -16.97 11.64 12.33 Industry 4.83 8.94 7.40 21.17 Mining, manufacturing, utilities -0.38 4.97 2.90 7.50 Construction 3.57 3.96 6.14 13.68 Services 24.93 27.25 4.04 56.23 Wholesale, retail trade, hotels 4.45 16.03 -6.65 13.82 Transport, storage, communications 1.67 3.84 12.98 18.48 Other activitiesa 13.41 7.39 3.13 23.93 Subtotals 47.42 19.22 23.08 89.72 Demographic component n.a. n.a. n.a. 10.28 Total - - - 100.00 Total % change in value added per capita - - - 92.06 Sources: World Bank, 2015b. Note: n.a. = not applicable. a. Other activities include financial and business services, public administration, education, health care, social protection, and other services. In summary, the outcomes of urbanization in terms of service delivery, livelihoods and poverty do not occur in a vacuum. Patterns of spatial inequalities, social differences and structural processes interact to produce governance challenges. In order to understand this, the subsequent section illustrates how urban political economy could be a useful theoretical toolkit to explain the challenges of urban governance. This approach helps to understand how political and economic factors intersect with people’s access to urban services, livelihoods and poverty. 2.3. Theoretical underpinnings of the study Rapid, chaotic, parasitic are common accompanying buzzwords for urbanization in SSA. This section explores the possibility of developing a unified framework which would provide a more integrated analysis of the contradictions that characterize SSA’s urbanization. The building blocks for this framework are discussed in two sections. The first section presents an overview of the different approaches to urban analysis, focusing specifically on co- production and urban political economy. The second section examines these approaches by focusing on the contributions of David Harvey, Manuel Castells, and David Drakakis-Smith. Based on a review of some of the key theories in urban political economy, the framework 40 University of Ghana http://ugspace.ug.edu.gh illustrates, along with Obeng-Odoom (2013), how a more interdisciplinary analysis of economy, politics, society and culture can be used to explore, analyse and evaluate urban governance. 2.3.1. Competing approaches to urban analysis – urban political economy Scholars have illustrated the strengths and weaknesses of the different theories and approaches to urban analysis. Recently, Smith et al. (2016) have examined the major theoretical and conceptual approaches to urban services, and found that none is sufficiently comprehensive to explain patterns of service provision in all types of cities: public choice theory, critical theory, urban political ecology, collective action theory, and social integration. They concluded that a major challenge is to account for both central administrative control of services and more generative, bottom-up service provision (Smith et al., 2016). Over the years, urban economics approach has been applied to urban analysis. Glaeser (2007, pp. 3-4), for example, calls the urban economics approach to cities, “spatial equilibrium in which if something is particularly good in one location, then, we should expect to see something bad offsetting it”. In urban economics, cities exist because of economics of scale and their problems are seen as the result of externalities (O’Sullivan, 2003). Essentially, urban economics treats the city as separate from the rest of society, and space is conceptualised, metaphorically, as a tree, distinct and separate from the rest of the forest (Sawers, 1984). However, as noted by Soja (1976, p. 29) “… urbanisation is not an isolatable process which can be hacked out of its larger societal context and examined through narrow disciplinary filters.… To understand the dynamics of urbanisation requires a perspective which can encompass the continuous and holistic interaction between social process and spatial form as it unfolds over time”. 41 University of Ghana http://ugspace.ug.edu.gh Dissatisfied with the urban economics approach to cities, political economists have sought to apply broad political economic perspective Soja talks about to the study of spatial issues (Sawers, 1984; Harvey, 2005). However, as Newton (1981) noted over three decades ago: the study of urban political economy needs no justification, for cities are the heart and arguably the soul of our civilisation, and their political and economic conditions are the linchpins of its [sic] existence. The question is not whether to study urban political economy, but how to do so. Urban political economy has borrowed from many established currents in political economy, including institutional and Marxist political economy (see Sawers, 1975, 1984). So, as Stilwell (2000, p. 25) recounts, “attention turns from the individual to class; from exchange relationships to production relationships; from allocative efficiency to the dynamics of capital accumulation; and from the market as a mechanism for harmonious interactions between buyers and sellers to its role in perpetuating alienation, exploitation, conflict and recurrent economic crisis”. Obeng-Odoom (2013) notes that, these characteristics are evident in the Marxist theory of accumulation, so it is not surprising that its application in a spatial context has been an important current in urban political economy. David Harvey’s work has been crucial in establishing this focus (see Harvey, 2001). Urban political economy also draws from the many other different currents in political economy (e.g., classical economics, Keynesian economics, and feminist political economy) and also from concepts and insights from other disciplines such as planning geography and sociology, though economic concerns still dominate the approach (Stilwell, 2000). This pluralist approach to the study of cities does pose a problem of coherence because each of the different currents and disciplines has different approaches and units of analysis (Obeng- Odoom, 2013). Urban political economy conceptualises urban development as a product of the mode and relations of production (Stilwell, 1992). As articulated by Harvey, “the creation of … cities 42 University of Ghana http://ugspace.ug.edu.gh and rapid urbanisation bring the town to rule over the country (simultaneously rescuing the latter from the idiocy of rural life and reducing the peasantry to a subaltern class) (1998, p. 53)”. To urban political economists, the shape of cities is not exogenously determined (as argued by urban economists); preferences are not independent of the economic system (Obeng-Odoom, 2013). Thus, it is the interests of private capital that shape the nature of cities (Sawers, 1975). Cities are a function of the social and geographical assembly of a surplus product (Harvey, 2008) and the urban phenomena should be seen in terms of oppositions rather than homogeneous characteristics. The oppositions between the core and the periphery in cities are not the result of externalities or accidents. Rather, they are the product of the process of capital circulation (Harvey, 2006). What we see as “buildings, piped water systems, shops and so on are socially created fields of relationships between individuals and classes, designed to maintain and perpetuate the dominance of the ruling capitalist class” (Stilwell, 1992, p. 19). By conceptualising urbanisation as a product of a political economic system driven by the interest of capital, the result and mobilisation of surplus product, class analysis becomes central to urban political economy. Urbanisation is seen necessarily as a class-based process (Harvey, 2008). Urban problems are disaggregated into ‘problems in cities’ (e.g. poverty and inequality that have more structural causes than city size) and ‘problems of cities’ (including pollution that may relate to how big a city is) (Stilwell, 1992). Even within these two categories, ‘urban problems’ fall disproportionately on different socio-economic and political classes (Stilwell, 1992, p. 33). There is also emphasis on social movements such as those that may unify the working class to the petty bourgeoisie to demand social justice in the provision of urban goods such as affordable homes and services (Nicholls, 2008). Urban political economic view sees the state as largely controlled by the dominant class, and the role of the urban planner is to plan the city in a way to facilitate the process of capital 43 University of Ghana http://ugspace.ug.edu.gh accumulation (Sawers, 1975). The Marxian view, in its crude form, is that both the state and the planner team up with the capitalist against the working class. But a more sophisticated view of class goes beyond the usual taxonomy of capitalists and working class to encompass the role of landlords and private service providers in urban development and other aspects of competing interests between class fractions (Obeng-Odoom, 2013). Urban political economy, therefore, provides an alternative to the urban economics approach, which sets aside all social, environmental, and political factors, using the assumption of ceteris paribus, and models urban research based on the concepts of ‘economic man’ and ‘rational choice’ (Stilwell, 1992, p. 16). The political economy approach to urban analysis recognises that economic phenomena are in a complex web with social and political processes, the ‘spatial consciousness’ or ‘geographical imagination’ (Harvey, 2005). Although the broader analytical approach of urban political economy is more useful, a recurrent problem is that there is no one method for urban political economic analysis (Obeng- Odoom, 2013). One popular approach that incorporates elements of political economy analysis such as class interest in looking at urban governance is the co-production method. Pioneered by the political economist, Elinor Ostrom and her colleagues at Indiana University, many urban scholars have used this approach to evaluate urban governance in both the Global North and South (Bovaird, 2007; Mitlin, 2008). Co-production, which builds on the idea of people as assets, is an important ‘way of seeing’ and so a useful approach for this study. Originally, the concept of co-production was used to critique dominant theories of urban governance in the 1970’s. The concept appears to have come to prominence in the early 1980s, when the setbacks of traditional ‘provider-centric’ service models of the welfare state had become obvious and a particular set of citizen–state interaction was discovered in US cities (Whitaker, 1980; Ostrom, 1996). However, the limitations of this traditional model are vividly exposed when we consider the potential role of users in services. The most widely 44 University of Ghana http://ugspace.ug.edu.gh cited typology for characterizing this role is Arnstein’s (1971) ‘ladder of participation’. On the lower rungs of this ladder are manipulation, therapy, and placation of the public, then the more positive activities of informing and consultation; on the higher rungs, we find partnership and eventually even delegated power and citizen control. However, this ladder disguises the complexity of provider–user interests and relationships. Ostrom (1996) distinguishes the production functions between service provider and citizen provider. She argues that if they are interdependent (complementary), then some minimum input from one is required for any output to be obtained, and the most efficient (and sometimes only) outcomes will be when both participate in the production process – in this case, service delivery and job creation. Generally, research has focused on the contribution of co-production to service delivery rather than looking at more fundamental political issues such as its implications for the distribution of power between organized citizens and the state. However, there has been some recognition of the implications for co-production in citizen involvement and participation (Marschall, 2004). Hence, although much of the literature discussing co-production is based within the context of a weak and reducing state, this literature identifies and discusses more fundamental and lasting limitations of state power (Mitlin, 2008). Ostrom defines co-production as “the process through which inputs used to provide a good or service are contributed by individuals who are not in the same organization” (Ostrom, 1996, p. 1073). This definition is too wide given the proliferation of partnership arrangements in services. Therefore, Joshi and Moore define a narrower form of co-production as the “provision of public services (defined to include regulation) through regular, long-term relationships between state agencies and organized groups of citizens, where both make substantial resource contributions” (2004, pp. 32-33, 38). They emphasize that such arrangements are not temporary and may not involve formal agreements, rather, potentially 45 University of Ghana http://ugspace.ug.edu.gh being undefined, informal and renegotiated almost continuously (Joshi and Moore, 2004) in a context in which ‘Weberian’ institutional boundaries are blurred. Bovaird (2007) suggests that this definition is too narrow, as it only focuses on state agencies and does not consider other professional providers. In this study, we define co-production, along with Bovaird (2007), as the provision of services through regular, long-term relationships between professionalized service providers and service users or other community groups, where all parties make substantial resource contributions. This definition is not intended to preclude consideration of the different classes and interests that various types of co-producers might have with respect to their co-producing roles (Bovaird, 2007). Rather, the focus is on users, volunteers, and community groups as co- producers, highlighting the complex relationships that exist between each of these groups and public sector organizations and other stakeholders who may play co-production roles. A deeper analysis is helped by Foucault’s work on governmentality and state power (see Dreyfus and Rabinow, 1982). Foucaldian analysis provides a framework through which we can understand co-production (Mitlin, 2008). Co-production, whether promoted by the state, ruling elite or civil society, teaches these partners new ways of working and strengthens collective practice (Abers, 1998). As civil society gains knowledge of the processes of the state, it occupies such spaces of governmentality in its own right (Mitlin, 2008). As Appadurai notes, it may be argued that in so doing, community organizations merely replicate oppressive tendencies of the state. But in his observations of the federation in Mumbai, he draws more positive conclusions: “my own view is that this sort of governmentality from below, in the world of the urban poor, is a kind of counter-governmentality, animated by the social relations of shared poverty, by the excitement of active participation in the politics of knowledge, and by its openness to correction through other forms of intimate knowledge and spontaneous everyday politics” (Appadurai, 2001, p. 33). Drawing conclusions across a range of examples 46 University of Ghana http://ugspace.ug.edu.gh is clearly difficult. In practice, it is likely that in some cases citizen groups succeed in occupying these spaces and making them their own; and in others, such groups take on the interests of state (Mitlin, 2008). As an agent of the state, it might be argued, lower-cost services might be achieved, but little contribution is made towards broader issues of democratic practice. Following Obeng-Odoom (2013), and as with any descriptive oriented approach in the social sciences, what you see hinges on the lens through which you look. Here, the lens is shaped by three intellectual approaches, namely: Harvey’s analysis of economic interest shaping cities; Castells’ analysis of state intervention, class, power and cities; and Drakakis-Smith’s analysis of colonialism and cities. These approaches are discussed to identify which specific concepts and ‘way of seeing’ are useful for the study. 2.3.2. Approaches to urban political economy David Harvey (1973) remains the most pre-eminent contributor to the development of modern urban political economy. Starting in the 1970s, he argued that urbanism is too complicated to have one theory embrace all aspects of it. One could theorize about different aspects of urbanism, but find it difficult to provide a complete theory that explains infrastructure service, employment, and poverty. In his analysis of cities, Harvey provided a theory focusing on one aspect of urbanism – social justice. This theory rest on the proposition that cities are shaped by capitalists in a way that yields profits for them and alienates others. Cities create need and maintain demand for goods and services produced and distributed by the owners of capital. Urbanism is thus a product of the production, appropriation, and circulation of the surplus value created in capitalism (Harvey, 1973). According to Harvey “… space in the modern city has an ideological purpose. In part, it reflects the prevailing ideology of the ruling groups and institutions in society. In part it is fashioned by the dynamics of market forces which can easily produce results which nobody 47 University of Ghana http://ugspace.ug.edu.gh in particular wants” (1973, pp. 309-310). People’s needs, aspirations and plans are fashion based on where they are. Therefore, one way to manipulate people is to (re)shape where they are and create scarcity of what they need to survive wherever they are (Obeng-Odoom, 2013). It is argued that cities are manipulated to create scarcity of jobs and infrastructure services in order to guarantee profit for the hegemonic class (Harvey, 1973). According to Harvey (1973), this manipulation of space to favour the dominant class makes cities inevitably unequal, granting differential access to resources. He noted that this tendency towards inequality is manifested in five ways, namely: access to opportunities, resources and services; the price of accessibility; the cost of proximity; the changing location of jobs and housing infrastructure and service; and the availability and price of resources. For example, access to opportunities and services is a potential source of inequality because it comes at a price which may be in the form of overcoming distance. As such, proximity may increase or decrease income (Harvey, 1973). Similarly, the changing location of jobs and infrastructure services is a cause of urban inequality because the lack of responsiveness of low-income housing to the demand of the poor traps them in the inner city, causing them to seek jobs in the informal economy and relying on private service providers. Finally, Harvey (1973) argued that the availability and price of resources is a source of urban inequality because it is contingent on where it is located, the farther away one is from the resource the more expensive the resource becomes. Since the city is a resource system, the local price of a resource, therefore, depends on accessibility and proximity to the user (Obeng-Odoom, 2013). These themes have been further developed in Harvey’s subsequent works. For instance, Harvey (2006) makes the point about the unequal nature of capitalist urbanization in a slightly different manner. He argues that urbanization is accumulation of surplus value by dispossessing the poor, the working class of every right to the city, through policies such as slum renewal and upgrading. It may be argued that an active political process is needed to 48 University of Ghana http://ugspace.ug.edu.gh more fairly redistribute wealth (Obeng-Odoom, 2013). It is thus important to analyse the process by which the state intervenes in urban development. Harvey has assessed the contradictions in state interventions generally. In his theory of creative destruction Harvey (2007), argues that neoliberalism as the dominant economic ideology has further empowered economically stronger classes and undermined the position of weaker classes, particularly through the destruction of state institutions and pro-social justice discourses. However, Harvey’s other contributions did not analyse the inherent inequality in the process of state intervention (e.g. inequality in the bargaining process) in the urban setting (Obeng-Odoom, 2013). To provide the essential insights into these facets of the state and urban politics, it is useful to consider the contributions of Manuel Castells to understand how class and power shape cities. Castells’ pioneering work in urban studies analytically emphasizes the provision of the means of collective consumption by the class-state as the main source of inequality. His books ‘The urban question’ (Castells, 1977) and ‘City, class and power’ (Castells, 1978) sought to answer why the ‘urban question’ becomes an urban crisis? According to Castells (1978, pp. 169-170) the ‘urban question’ denotes the way in which public goods such as housing, infrastructure services, commerce, and health (the means of collective consumption) are provided in society. As cities grow, accumulation of capital intensifies, and labour is drawn into the centre of this large pool of concentration (Obeng-Odoom, 2013). Concentration requires the provision of collective goods like roads, schools, hospitals, water and electricity. Such collective goods, Castells (2010) noted, serve as the lubricant for the maximization of profit. Yet those who own and control capital would not undertake the production and management of these goods, not because the goods are not essential, but because these goods are difficult to appropriate and run in a profit maximizing way (Obeng-Odoom, 2013). According to Castells, it is the state that tries to mitigate this contradiction by producing, managing, and distributing the means of collective consumption. However, state intervention is the source of further 49 University of Ghana http://ugspace.ug.edu.gh contradiction because, according to Castells (1978, p. 2), “…as an expression of a class society, the state in practice acts according to the relations of force between class and social groups generally in favour of the hegemonic fraction of the dominant classes”. This orientation led Castells to conclude that, when the state provides housing, the motive is to provide a property-less workforce shelter in order to keep the wheels of accumulation turning (Castells, 1978). It may be argued that the poor delivery of infrastructure services by the state is to create the need for private service provision. For this reason, poor public service provision should be understood as a deliberate signal by the state to private operators (who may be cronies of key state officials) to provide private services. This process also perpetuates rent-seeking tendencies among the urban poor who work in the informal economy. As such, the attempt by the state to resolve the urban question ends up sustaining and worsening the class divide. For all these reasons, Castells argued that the urban question is a political one with the “collective equipment and the resulting urban system [being] decisively affected by the role of the state” which prolongs and specifies the social stratification of the class system (1978, p. 34). Castells’ theory is that the urban question should be seen as part of the larger political process and urbanization between the city and power is, therefore, a relationship embedded in society (Castells, 2010). Like David Harvey’s work, Manuel Castells’ analysis is useful in understanding modern urban development. Both implicitly use the notion of capital accumulation and class conflict. However, suffice to say that these views do not necessarily apply directly in the African context. Developed primarily with a focus on the experience of cities in the Global North context, they tell us little about how the historical antecedents of colonialism affected the structures and patterns of urban development in a Global South context. It is imperative to turn to a theory which weaves inequality and class struggle in the historical context of colonialism and post-colonialism (Obeng-Odoom, 2013). It is not a shift from economy to 50 University of Ghana http://ugspace.ug.edu.gh culture, but the shift to a specifically colonial and postcolonial perspective (Ribera-Fumaz, 2009) to make the urban political economy of Harvey and Castells more sensitive to cities in Africa. Here, the pioneering contribution of David Drakakis-Smith is particularly useful. Drakakis-Smith (1987) analysed the effect of mercantile colonialism, industrial revolution, late colonialism and independence on Third World cities to establish their interconnections with contemporary urbanism. The era of industrial revolution following mercantile colonialism, Drakakis-Smith argues, had a more significant impact on urbanization in the Global south through culture, technology, and political structure. Drakakis-Smith (2000) observed that European values shaped the nature of buildings and settlement patterns. Hence, in some countries, only white communities benefited from infrastructural development. This bias marked the genesis of urban primacy in the Global South. Drakakis-Smith notes that, “it is in this period therefore, that sees the genesis of contemporary urban primacy as economic and political power is concentrated on certain cities at the expense of others” (1987, p. 17). According to Drakakis-Smith (1987), the economic recession in the West during colonialism (the inter-war years of the twentieth century) also had implications for urbanization in the Global South. This period was marked by mechanized agriculture which caused many people to lose their land and migrate to cities. More so, there were demographic changes stemming from the influx of Europeans into the Third World to take up middle-class jobs. With this background, Drakakis-Smith indicated that the dawn of independence witnessed a new wave of urbanization because of local perception that many jobs had been left behind by the colonialist. However, the supply of jobs could not match the demand from those migrants, leading to the proliferation of under- and unemployment. The period immediately after independence was marked by another form of colonialism – neocolonialism being the state of commercial dependence (Drakakis-Smith, 1987). This master-servant relationship exists in contemporary cities in the Global South – taking the 51 University of Ghana http://ugspace.ug.edu.gh form of a new international division of labour. The result, according to Drakakis-Smith (2000), is rural-urban migration and a swollen urban population, which is caught in a vicious cycle of under- and unemployment. These dynamics have undermined the household economy and affected the home because women and children, hitherto not active in the labour market, have been forced into it. Particularly, women engage in household enterprises that offer the flexibility to combine reproductive activities. Following Castells, such deteriorating conditions could be expected to trigger some active response from urban social movements (Obeng-Odoom, 2013). However, Drakakis-Smith (1987) takes a different view. According to him, many of the urban poor are indebted so how they react to social problems is determined by the whims of their creditors. Hence, the more likely form of protest for the urban poor is through low-key challenges (Drakakis-Smith, 1987) including contests that take the form of rent defaulting and illegality such as settling on public land (squatting) and other informal activities in slums. Therefore, for the urban poor protest against the status quo would be, in Drakakis-Smith words: any activity designed to influence the redistribution of scarce resources, much of which could be quite informal … and only occasionally would this encompass open insurrection but many are prepared to take public action on a more personal level to gain access to essential resources they would otherwise be denied. The most visible type of action in this context is the seizure of and by squatters … the most common form of political action by the poor is simply to use whatever means they can to cope with the unequal system. In this sense, the entire informal sector with its illegal and semi-legal living and working environment constitutes a form of political protest (1987, p. 52). From this point, it can be argued that spatial underdevelopment in Africa has also been due to the dynamics of colonialism. However, one cannot reasonably conclude that all development problems in African cities were the result of colonialism; more contemporary factors have played a key role too. This nexus between internal and external forces, and its impact on cities in Africa, has been emphasized by Akin Mabogunje (1990). According to 52 University of Ghana http://ugspace.ug.edu.gh him, colonialism was a sine qua non for the incorporation of African economies into the global capitalist system. The predicament of most African countries today, Mabogunje argues, “is the product of the distortions, contra-dictions, and conflicts between the various elements of pre capitalist and capitalist social relations that continue to operate in their national economies” (1990, pp. 122-123). These views need to be framed more systematically to guide the analysis of the study. 2.4. Conceptual framework How can we usefully draw on these various insights to analyse contemporary urban processes, problems, and policies highlighted in Ghanaian cities? The conceptual framework which draws on the analytical contributions stemming from co-production and urban political economy shows how myriad of forces shape cities. It applies a governance-poverty approach to urban political economy analysis by exploring three key components: (1) external and internal factors that shape cities, (2) intervening mechanisms marked by governance arrangements, economic interests and household dynamics, and (3) resulting outcomes of economic growth, poverty reduction and inclusive cities. As outlined in Figure 2.3, the first component entails external and internal factors. External influences are usually in the form of the enduring effects of globalization and (neo)colonialism manifested in development programmes or ideas imposed or negotiated by agencies and networks supposed to be supportive of economic development in Africa. Internal factors operate through pressures such as those coming through local economic interest and social movements and through state policies and practices. For instance, state policy and slow governance actions compel those who wield power to bypass certain groups to extend service provision. In Abuja, residents in the informal settlement have extended water and electricity services through a self-help initiative. 53 University of Ghana http://ugspace.ug.edu.gh Figure 2.3: A conceptual framework of water and electricity governance and livelihoods on poverty reduction in Ghana External influence Mechanisms Outcomes Globalization Governance Inclusive cities Colonialism (Co-production) Neocolonialism Economy Poverty reduction Internal pressure (Livelihoods) Economic interest Social movements Household Policies and practices (Service delivery) Economic growth Source: Author’s construct, 2016 (Adapted and modified: Obeng-Odoom, 2013) The framework approaches service delivery indirectly through the idea of ‘collective consumption’, the notion that urban residents rely on collectively provided goods, such as utilities like water and electricity, and social networks, to survive and reproduce the existing array of (in)formal socio-economic relationship. Collective consumption is interwoven with the contradictions internal to capitalism because, while the provision of services is critical for the reproduction of labour and the continued functioning of markets, such service provision also represents a threat to capitalist profits if resources are diverted to collective (public) goods (Smith et al., 2016). Since collective consumption is “at once in the general interest of the capitalist class and not in their general interest” (Merrifield, 2002, p. 120), government-based service provision is portrayed as a disjointed, historically uneven process usually motivated by crisis and prone to the lowest expenditure (Smith et al., 2016). In this view, non-governmental service co- providers assume a greater burden in cities as governments relinquish their public duties. 54 University of Ghana http://ugspace.ug.edu.gh Thus, there may be a tendency of the state to worsen urban problems by its attempt to ostensibly resolve them, if it acts in a class-biased manner (Obeng-Odoom, 2013). This brings into attention the second component which focuses more substantively on how: (a) the uneven distribution of political and economic power affects water and electricity delivery; and (b) economic interest is often structured to prioritise elites at the expense of marginalised groups. For example, how political economic factors trigger differential access to services occasioning the dominance of private service providers and informal livelihood activities in the urban economy (Swyngedouw, 2004). This component aims at understanding governance arrangements and how these relate to the economy and household outcomes. The different levels of this component clearly overlap. However, they are differentiated in order to emphasize that governance dimensions as well as the household and economic interests need to be explicitly considered. Essentially, the idea is to explore what reforms are feasible from a co-production perspective. This framework is based on a diagnostic rather than a prescriptive approach to governance (Rodrik, 2008). The ultimate goal is an inclusive city with improved service delivery, enhanced livelihood activities and reduced poverty: but the focus is on how to move in this direction rather than on the final goal. Good governance encapsulates the notion of focusing on priorities and improvements that are feasible and affordable rather than trying to reform governance wholesale – whether at the macro or sector level (Grindle, 2007). The economy takes centre stage in my understanding of these constructs. Particularly because it is where polity and individual actions and outcomes converge. As social scientists, we seem to be more successful in developing theories and conceptual frameworks which account for how macro level processes determine or constrain the workings of smaller units - households and communities – than explaining how smaller units influence broader political and economic processes (Beall and Kanji, 1999). For instance, the literature on SAP is rich in 55 University of Ghana http://ugspace.ug.edu.gh frameworks and case studies that explore the impact of liberalization, cuts in employment, public services, public sector spending and subsidies on low-income households. Mindful of this, ‘household’ has been conceptualized here to recognize the inequalities of power and welfare among household members, rather than treating it as an undifferentiated utility optimizing unit. The household is an important arena of policy research. Moore (1994) argues that household ideologies and social identities are constitutive of economic and political processes, not simply determined by them. This point is integral to systems of redistribution, which structure the type and distribution of services and livelihood opportunities within and beyond the household. Issues of governance (co-production), economy (livelihoods) and household (services) are key focal elements for reducing poverty and promoting growth and inclusive cities as depicted in the third layer. Livelihoods have been conceptualized not only in terms of income earning but a much broader array of activities such as improving access to resources, services and opportunities, and managing social networks (Esson et al., 2016). A focus on the livelihood strategies of households is to illustrate the significance of human agency, and how they negotiate structural constraints. In an era of globalization, seeking to rapidly reduce poverty can result in a focus on those poor whom the market can liberate from poverty but neglect the needs of those who need different forms of support, policy changes, or broader changes within society. Urban poverty has been conceptualized to include: (1) limited access to income and employment; and (2) poor access to reliable, quality, and affordable water and electricity. The additional qualitative aspects are needed because income poverty measures provide important but incomplete guidance to redress multidimensional poverty (Amankwaa, 2016). A synthesis of these theoretical insights into a framework is a useful ‘way of seeing’ in the Ghanaian context, where the desire to achieve the SDGs requires successfully integrating sustainable livelihoods into the nexus thinking of good governance and poverty reduction. 56 University of Ghana http://ugspace.ug.edu.gh This framework is not meant to suggest that city authorities can simply rely on the initiatives of the urban poor (through co-production) for solutions to urban problems. Rather, the idea is to highlight the implication of household dynamics for urban planning and policy research. In doing so, it links the micro-politics of households into wider urban analysis. Overall, the aim is to analyse the linkages between micro units such as urban households and the macro political and economic processes operating in the city. This framework enables conceptual and practical examination of how the polity, economy, and governance trilogy can be leveraged to make cities and human settlements inclusive, safe, resilient and sustainable as enshrined in SDG 11 (see Arfvidsson et al., 2016; Simon et al., 2016). 2.5. Conclusion Urban governance, as a development paradigm has become highly influential. Through this paradigm, donors have given much support to cities and their governance. Analysing water and electricity access, livelihoods and poverty dimensions, the chapter demonstrates that cities serve as a locomotive for national economic growth which, in turn, support urban development. The levels of urban poverty have been falling partly under the aegis of entrepreneurialism (market), democratization (urban community), and particularly from decentralization (state). What is worrying is the steady rise in inequality levels both in terms of income, access to and control of urban services and resources. This chapter also considers diverse theoretical currents that can inform urban analysis. It explored how currents of urban political economic analysis is appropriate for studying urban governance in cities. The approach postulates cities as the product of a complex web of factors, with the economy being central to analysis and recognized as an integral part of society. It developed a conceptual framework that conceptualized the city as a reflection of the mode and relations of production, highlighting the importance of both historical forces and contemporary dynamics. Hence, the conflicts, and disequilibria are endemic in the urban 57 University of Ghana http://ugspace.ug.edu.gh economy, occasioning various urban problems, making state intervention important, particularly in the provision of the means of collective consumption. However, due to the class biasness of state intervention, there is the recurrent possibility of actually worsening urban problems. It is thus imperative to examine whether urban citizens really benefit from a good society and economy, typified by quality and acceptable urban services and sustainable livelihoods. It is this paradigm that informs my analysis in the rest of the study. 58 University of Ghana http://ugspace.ug.edu.gh CHAPTER THREE PROFILE OF STUDY SITES, PHILOSOPHY AND METHODOLOGY 3.0. Introduction This chapter gives an overview of the study area. It also discusses the philosophical orientation, and methodology adopted for studying services, livelihoods, and poverty of the urban poor. Science is a social activity whose aim is the production of the knowledge of the kinds and ways of acting of independently existing and active things (Bhasker, 1979). It should be noted that the knowledge produced through the data collected is inherently linked to personal and situational characteristics (Limb and Dwyer, 2001). Following Punch’s (2010) call for increased awareness of personal challenges in order to enhance the process of reflexivity and positionality during research, this chapter discusses how these processes influenced the research approach and methods applied. The interpretations and arguments presented in this thesis are value-laden, situated, and socially embedded, which includes my own positionality. The chapter is structured into three parts. The first part describes the contextual background of the study area. It first discusses the reasons for the choice of GAMA and the study area. This is followed by a discussion of some key characteristics of the research localities. The second section discusses the philosophical paradigm (critical realism) and methodology (mixed methods) employed to generate, analyse, and interpret the study findings. It describes how the data was analysed and reflects on the researcher’s positionality and ethical issues. The final part of the chapter examines the profile of the study participants captured during the fieldwork. 3.1. Profile of study sites 3.1.1. The Characteristics of GAMA The Greater Accra Metropolitan Area (GAMA) is selected as the study area. This choice is due primarily to evidence gathered from previous studies, which have examined the drivers 59 University of Ghana http://ugspace.ug.edu.gh of increased poverty in Accra particularly the inefficiencies in its present governance system and basic service delivery (Amankwaa, 2016; Oteng-Ababio and Melara Arguello, 2014; Yankson, 2008). This is because, according to Oteng-Ababio and Melara Arguello (2014, p. 1), the issue of poverty “demands a multifaceted solution that includes more subjective concerns like vulnerability and social exclusion, and objective ones like good urban governance and power relations. Failure to do so can lead to a downward spiral into extreme poverty that can affect the entire society”. Thus, studies have attributed urban poverty in GAMA to the poor service provision, which is an outcome of the inefficiencies of decentralised governance, and have recommended streamlining and reframing urban governance to address the new forms of urban development that benefits the few. So, a thorough understanding of the dynamic forces and systems such as (infra)structural economic issues that drive GAMA’s accumulation of poverty is fundamental in re-ordering and re- working the roles of the state, market, and society as institutions of change, and re-organizing the channels for keeping the attainment of the ends of urban governance in check (Obeng- Odoom, 2016). Prior studies by Oteng-Ababio et al. (2013) and Yankson et al. (2005) have conceptualised GAMA as the much wider area of the Greater Accra region that was originally composed of the Accra Metropolitan Assembly (AMA), the Tema Municipal Assembly (TMA) and the Ga District Assembly (GDA). Currently, GAMA has metamorphosed into 12 autonomous, but spatially and functionally integrated administrative demarcations, as articulated in chapter 2, section 2.2.1 (see Appendix 5). The AMA is the anchor of the larger GAMA and administers what the World Bank (2015a) calls ‘Accra city proper’. This is a 200 km2 stretch of land which hosts Ghana’s most diversified socio-economic and political economy. Much of the 60 University of Ghana http://ugspace.ug.edu.gh city’s important commercial, administrative and cultural locations are found in Accra Central3. In terms of spatial extent, GAMA covers 1,079 km2 (33%) of the Greater Accra region’s total land surface of 3,245 km2 (GSS, 2005). GAMA is home to about 3.7 million inhabitants, representing roughly 96% of the region’s total population of around 4 million (GSS, 2012). This makes GAMA the most densely populated, with a total number of 450,794 households and a household size of 3.7, and the single largest urban conurbation in Ghana (and the 11th largest metropolitan area in Africa). The operational definition of GAMA has been illustrated on the map in Figure 3.1. Urban sprawl and an unsupervised physical expansion from the metropolitan boundaries of Accra during the 1960s have led to the emergence of the peri- urban and physically undeveloped Ga District. The Ga district experienced massive population explosion occasioning the carving of several districts out of it (see Appendix 5). Most of the inhabitants in the Ga district commute to Accra Central to purchase retail goods for their HBEs. 3.1.2. The Study sites This thesis is part of the European Commission funded ‘African Rural-City Connections’ (RurbanAfrica) project coordinated by the University of Copenhagen in Denmark. The RurbanAfrica project explores the connections between rural transformations, mobility, and urbanization processes, and analyses how these contribute to an understanding of the scale, nature and location of poverty in sub-Saharan Africa (SSA). Having selected four countries Rwanda, Tanzania, Cameroon and Ghana, the project addressed four crucial components (Work Packages - WP): 1. agricultural transformations, 2. rural livelihoods, 3. city dynamics, and 4. access to services in cities; with the overarching aim of addressing knowledge and 3 Accra Central is delineated by the Ring Road corridor which runs from the Korle Lagoon in the west, through Kwame Nkrumah Circle and the Independence Avenue to Osu in the east. 61 University of Ghana http://ugspace.ug.edu.gh policy synergies. Within each of the project countries, two cities were selected as fieldwork sites: the largest city and a rapidly growing intermediate-sized city; the rationale being that, it is essential to understand the dynamics of the largest city in order to understand the factors driving urban growth within a country, but studying this city alone would create a misleading picture. Thus, conducting a similar study in a second smaller but still rapidly growing city facilitates an exploration of city dynamics in an intermediate-sized city and enables comparisons with the major city and the other intermediate cities in the project. So, in Ghana both Accra and Sekondi-Takoradi were selected as the study sites. Gough et al. (2015) provide more details of the focus groups and interviewees in each city. Esson et al. (2015) and Amankwaa et al. (2015) provide overviews of the case study settlements in these cities. For this thesis, the focus is on Accra, the capital city. Four residential areas, Korle Gonno, Accra New Town, Gbawe and Abuja, were studied in order to cover a range of neighbourhoods. The selection of these four localities was considered by the researcher as practicable in view of time and resource constraints. They differ in terms of older and newer areas, range of income levels, differing locations within the city, and differing types of access to urban services and livelihood strategies. Thus, the differences provide a useful basis for their comparisons for analytical purposes. Figure 3.2 is a map of the study area showing the study sites. An overview of the case study localities is provided in Table 3.1. 62 University of Ghana http://ugspace.ug.edu.gh Figure 3.1: Map of GAMA showing its constituent administrative areas Source: Author’s construct, 2016 63 University of Ghana http://ugspace.ug.edu.gh Figure 3.2: Map of GAMA showing the selected study sites Source: Author’s construct, 2016 64 University of Ghana http://ugspace.ug.edu.gh Table 3.1: Brief profiles of settlements selected in GAMA Settlement History Location Income- Rate of population Remarks level growth Korle Indigenous/ Western Low-income Stable overall Stable population Gonno traditional edge of though size but settlement centre movement in movement in and and out out. Poor services. Accra Migrant Northern Low/middle Shrinking Residential being New settlement edge of income overall though converted to Town established in centre movement in as commercial; 1940s, mainly well as out subdivision of Muslim existing dwellings Abuja Informal City Low Growing in size Considerable settlement, centre income infilling taking illegal, growing place. rapidly since 1990s. Gbawe Indigenous Western Low/middle/ Growing rapidly Well-functioning core, growing periphery high income land market with rapidly since astute chief. 1990s 3.1.2.1. Korle Gonno Korle Gonno is an indigenous (Ga dominated, albeit becoming ethnically diverse), low- income settlement located along Accra’s coastline and next to the Korle Lagoon. Beginning as a small fishing and farming village, the settlement was once known for its thought out site planning and architecture. The same cannot be said of it today as the area is increasingly characterized by overcrowding and infilling, and declining urban services. Alongside these changes to the built environment, the settlement’s demographics are also changing. The 2000 Census registered 27,826 residents and by 2010 this had reached 30,555 with a total household of 8,599. This is due to both natural increase and in-migration. Meanwhile, some indigenes are also acquiring properties in Accra’s peri-urban locales, such as Kasoa and Weija. Korle Gonno’s growing population coupled with changing macro-economic conditions has had an impact on the spatial architecture. For example, older single-storey buildings made 65 University of Ghana http://ugspace.ug.edu.gh from mud are being gentrified into architecturally unique multi-storey structures, particularly along the coast, which is emerging as a prime real estate. Renting of rooms is a lucrative venture for most landlords who now turn facilities such as bathrooms and kitchens into sleeping rooms. This is particularly visible as one moves away from the Korle Bu Hospital and the commercial districts around Guggisberg Avenue, towards Old Winneba Road via the residential areas. These issues contribute to the downgrading of the settlement which is further manifested in the poor provision of basic services. The majority of houses have infrastructure capable of providing pipe water and electricity but this does not equate to the ability to access water and power; issues of regularity, cost, and quality are widespread and partly explain why the practise of illegal tapping of water and power is rife. The frequent power outages with no prior warning damage electrical items that are plugged in. One of the ways people overcome challenges associated with accessing affordable housing and services is to cohabit with family members and use their home spaces for business activities. Korle Gonno’s local economy functions in a manner similar to other parts of Accra, with informal sector economic activities, such as selling fast-food, provisions and beauty products as well as mobile phone credit kiosks, hair salons and paramedical services, a prominent feature of the landscape. A section of the populace engages in fishing activities. The uniqueness of Korle Gonno’s local economy lies in the spatial distribution of activities, with key economic hubs often located alongside key transport corridors. Beyond these major roads, it is often amidst Korle Gonno’s residential areas with its well- planned street layout that home-based enterprises (HBEs) are most prominent, and they are also an important source of income for residents and a key component of the local economy. The challenges associated with accessing household water and electricity negatively affect business activities. For example, issues around water quality means that those engaged in the selling of cooked food are reliant on treated sachet water, which comes at an extra cost and 66 University of Ghana http://ugspace.ug.edu.gh therefore reduces profits. Also, the poor access to electricity has altered the economic playing field as poorer business owners now lose customers to those who can afford to keep their prepaid meters topped up. 3.1.2.2. Accra New Town Established during the 1940s, Accra New Town is a low/middle-income settlement located on the northern edge of Accra central. Up until the early 1970s, Accra New Town was known as Lagos Town. This name reflects the visible presence of migrants of Yoruba extraction from Nigeria within the settlement. This situation changed dramatically in 1969/70, when Ghana implemented the Aliens Compliance Order, which compelled nationals of the other West African countries who had not regularised their stay in Ghana to leave the country. New Town had the highest population density in Accra by the 1980s (370 people per hectare), and while it still exhibits physical signs of this legacy in the form of gentrification and intense infilling, recent census data indicate the population is declining. In 2000, the population was 45,130 decreasing to 31,363 in 2010, with a total of 8,352 households. A plausible explanation for this observed trend is the changes in land use, most notably the conversion of residential properties into commercial spaces such as retail outlets and printing presses capable of generating higher rental income. Similar to other parts of the city, single rooms and compound houses (which vary in size and can contain as many as 20 rooms) are the most common types of accommodation found in Accra New Town. This is more noticeable as one moves away from the main roads where multi-storey commercial buildings such as printing presses and retail outlets predominate, and towards the more densely populated residential areas behind them. Accra New Town is increasingly synonymous with the printing industry, yet this is a relatively recent addition to the local economy as numerous other income-generating activities take place in the settlement, particularly along the main road where commercial banks, pharmacies and larger 67 University of Ghana http://ugspace.ug.edu.gh retail stores selling electrical items and clothing can be found. Likewise, informal household enterprises such as cooked meals, containers selling provisions and beauty products are highly visible, as are mobile phone credit kiosks, fashion designing shops, hair salons and sachet water stands. As land/space is becoming scarcer in the more densely populated areas, disputes over who is allowed to use the space outside a property for commercial activities are increasing. This is partly due to some landlords preventing tenants from running a business from their property and neighbours being less willing to allow their property to be used by others for home businesses. Besides working space, one key challenge that household enterprises face is the access to water and electricity services. This is because the majority of houses have infrastructure capable of providing piped water, but access to water is a long standing problem in Accra New Town (partly following the construction of a large gutter that separates the settlement from Nima). The area lacks boreholes and protected dug wells. When pipe flows, landlords often place locks on taps and charge tenants for sourcing water in addition to their rent, as a means to cover the monthly water bill and even make profit. Although it is more cost effective to source water from a metered pipe, the majority of residents are unable to do so and, therefore, have to spend more of their income sourcing water from private vendors on a daily basis. Similar to water, the majority of homes in Accra New Town are connected to an electricity supply, albeit illegally in most cases. Power outages are rampant. In a typical week electricity may be unavailable without warning on three to five occasions for varying lengths of time. Despite these problems, the settlement has a relatively quality power supply compared with other parts of Accra which is a key reason why a printing press industry heavily reliant on electronically powered machinery has emerged in Accra New Town in the past decade, with companies relocating from other parts of the city. This is perceived as having a negative 68 University of Ghana http://ugspace.ug.edu.gh bearing, however, on households who consider that the presence of the printing press is reducing the quality of supply to their homes and increasing the number of electrical fires in the area. These challenges associated with accessing water and electricity services negatively impact the economic success of income-generating activities and divert money away from meeting daily household needs. 3.1.2.3. Abuja Abuja is located in the heart of the city of Accra adjacent to the Cocoa Marketing Board and the Accra Brewery on the Graphic road of the Central Business District (CBD). The settlement began as a temporary shelter in the 1980s. The name, Abuja, is historically traced to the Ghanaian migrants who returned from Abuja in Nigeria in 1983 and settled there. Prior to this, workers of the defunct Ghana Railways Corporation the majority of whom relocated following its collapse inhabited the area. An estimated 8000 residents earn their living from business activities within the settlement. Abuja’s growth became prominent throughout the 1990s when it became a popular area for housing and work activities due to its affordable rent; close proximity to the city centre; and central markets that provide income-generating opportunities. Officially, Abuja does not exist. It is not recognized as a residential zone due to land tenure issues. This implies that government-provided services like pipe-borne water supply and garbage collection facilities are absent. The early settlers and those who wield power, however, have managed to extend electricity and water infrastructure to the area. These local elites are connected to the utility providers, and they in turn, supply services to the whole settlement as private shower operators and electricity distributors. Intermittent water supply is the norm in Abuja, therefore shower operators often collect and store large quantities of water during periods when the taps are flowing. Abuja is also subject to what is colloquially known in Accra as ‘light off’ i.e. intermittent power outages. The frequent ‘light off’ 69 University of Ghana http://ugspace.ug.edu.gh alongside meter overloads and the use of candles have contributed to the frequent fire outbreaks in the settlement. The most pressing issue concerning electricity is the introduction of prepaid meters whereby regardless of service quality, customers are always charged. Since meter sharing is the norm, it is also more difficult for families and HBEs to budget, as they are unable to predict when their credit will finish. Abuja hosts a transport terminal and a mix of informal residential and commercial structures, with many hawkers and niche food markets, as well as small home businesses. One of the ways people survive in the area and improve their affordability to services is to rent a container, which can be used for their economic activities and as a place to sleep. This implies that the challenges with accessing services have direct bearing on home businesses. As young men and women are less likely to obtain work in the formal labour market, because of their low educational levels, they engage in entrepreneurial livelihood activities that enable them to be job creators as opposed to job seekers. Some male youth earn income via illegal means, such as cybercrime and dealing in drugs. Very often, income from business activities is sent back to the migrant’s village to be reinvested in a family business, which is deemed more beneficial than sending money purely for subsistence. 3.1.2.4. Gbawe Gbawe is a peri-urban settlement located about 10km west of Accra central, established by a family of hunters from Ga Mashie. In the not too distant past, it was covered in foliage and subsistence farming activities dominated. Gbawe has undergone significant social and physical changes since the early 1990s. Three changes are of particular importance here: 1) Rapid population growth; 2) Drastic decline in the dominance of farming activities; 3) The increasing sale of land for residential use. Concerning Gbawe’s growing population, recent census data indicate that the population in 1984 was only 837. By 2000 the population had skyrocketed to almost 29,000 and a decade later in 2010 it was 67,998 with a total number of 16,624 households. 70 University of Ghana http://ugspace.ug.edu.gh In terms of housing provision, Gbawe displays many of the characteristics commonly found in Accra’s other peri-urban locales. What was formerly arable land is now covered by single family homes of varying sizes, ranging from a single room to multi-storey mansions with swimming pools, at varying stages of completion. Notwithstanding the heterogeneity in the built environment, some basic housing patterns are discernable. Gbawe’s indigenous core is where high-density houses built from mud and populated by indigene homeowners can be found. As one moves away from the old town, houses become more diverse architecturally with a mixture of homeowners and tenants. Gbawe’s proximity to the Weija Dam, one of Accra’s primary water sources, makes pipe-borne water supply rarely unavailable. Homes in the indigenous core make use of communal water sources, and if there is no pipe on the premises, it is likely that a neighbour will have one that can be accessed for a fee. Properties built more recently typically contain infrastructure capable of providing pipe water. With the exception of a handful of properties in the old town, the majority of homes in Gbawe are connected to an electricity supply, albeit in some cases illegally. Power outages are recurrent but when power is available, the current is of acceptable quality. In cases where the electricity supply is being sourced from a neighbour, the introduction of the prepaid meters has resulted in a reluctance to share one’s connection due to fears of spiralling costs and sudden power cuts, leaving some residents without access to electricity. In terms of the local economy, farming still provides an occupation for a handful of Gbawe’s residents, albeit it is declining in importance as the land is increasingly being sold and converted into residential and commercial buildings. Similar to other parts of Accra, informal household enterprises such as selling cooked meals, small containers selling provisions and beauty products, mobile phone credit kiosks, hair salons and a variety of sachet water stands are features of the landscape. In most cases, these activities take place in spaces in and around 71 University of Ghana http://ugspace.ug.edu.gh the home, driveways and front pavements. The challenges with electricity and water access negatively affect income-generating activities in Gbawe. 3.2. Philosophical orientation Underlying every piece of research are assumptions about the nature of reality (ontological assumptions - what is real?), corresponding assumptions about the properties that knowledge should have (epistemological assumptions - how do we know about it?), and therefore which set of theories and concepts can be appropriately applied (methodologies). Currently, the philosophy of science is moving towards some convergence of methods. This converging trend is unlike the 1980s when an antipathy existed between the quantitative and qualitative research paradigms4. The converging trend in the philosophy of science rests within the philosophical paradigm that quantitative and qualitative research approaches should not be treated as contradictions or alternatives to each other, but rather, it is plausible to bring the two paradigms together in order to shed light on social phenomenon (Olsen, 2004). Consistent with the thinking of methodological pluralism and convergence of both quantitative and qualitative approaches, this study employs mixed methods, underpinned by the philosophical approach of critical realism (Bhaskar, 1989; Sayer, 2000) which combines positivism and interpretivism. 3.2.1. Ontological perspectives Moses and Knutsen (2007) separate the myriad of philosophical viewpoints into two main camps, albeit acknowledging a range of stances in-between that are also possible – of which critical realism is one. On the one hand, there is empiricism and positivism (naturalism), the traditional scientific approach, which views reality as distinct from our experience of it, 4 Following the work of Walby (2001), the antipathy between the quantitative and qualitative research paradigm has been referred to as epistemological chasm 72 University of Ghana http://ugspace.ug.edu.gh claiming that regularities and patterns exist in nature that can be observed and described. The opposite view however asserts that, patterns of interest are not firmly rooted in nature but are a product of our own making; each of us sees things differently, and what we see is a complex mix of social and contextual influences and presuppositions (Moses and Knutsen, 2007). This ideology refers to constructivism (interpretivism), as it holds the ontological position that reality is constructed by our experience of it; the role that humans play in interpreting their environment is acknowledged as a crucial aspect of our understanding of reality. Based on these two views of reality, it is clear that the ontological approach chosen for a particular study depends on the research questions to be addressed. For example, if a researcher wants to explore the causal link between service provision and household access, then a positivist approach may largely be appropriate as the reality to be explored is independent of human perception. However, a social constructivist stance is more appropriate to investigate the impact of the dynamics of access on household livelihoods or income, as the reality is constructed by humans and it is their interpretation of that reality that makes the subject of inquiry. This study seeks to understand how socially constructed aspects of electricity and water ‘access’ can be incorporated with the physical aspects of water and electricity ‘delivery’. The study postulates that some realities are best represented through a naturalist ideology (objectivist view) and others through a constructivist methodology (subjectivist view). It intimates that it is the successful combination of these components that will lead to the eventual improvement in the services that city authorities provide to the urban poor. The ontological approach that seems best for the study is critical realism. Critical realism amalgamates the ontological positions of naturalism and constructivism by accepting the existence of realities which are autonomous of our observation, while recognising human agency and the meanings that humans can attribute to reality. The ontological position held by critical realists is that there is a reality that exists independently from human conception 73 University of Ghana http://ugspace.ug.edu.gh of it, but it is buried below layers of socially constructed understanding of reality (Bhaskar, 1978). A critical realist perspective identifies the complex nature of reality and the different realities that different questions seek to explore. Some of the main tenets of critical realists are: the world exists independently of our knowledge of it; our knowledge of the world is fallible and theory-laden; objects have causal powers and susceptibilities; the world is differentiated and stratified, consisting of events, objects and structures and; social phenomena are concept-dependent, hence interpretation of meaning. These views suggest that realists share some characteristics with empiricists, idealists and post-modernist on some ideologies but build bridges between these approaches (Moses and Knutsen, 2007). 3.2.2 Epistemological perspectives and methodology The ontological position that reality is buried below layers of socially constructed meaning has direct implications for the epistemological position (Bhaskar, 1978). Critical realism accepts both the internal mental realities of humans and the objective realities of nature and therefore there is a corresponding epistemological dualism. Put differently, the properties of the knowledge in question will be dependent upon the category of the reality explored. Bhaskar (1978) argues that we have to distinguish the ‘real’ (mechanisms that are causes), from the ‘actual’ (that which happens) and also from the ‘empirical’ (that which we encounter). The result of this engagement for service provision is to disarm scientific reductionism, which is often used by service providers – in conjunction with the authority of natural science – to dismiss citizens’ agency and demands for improved service delivery. The same is true of the methodology of critical realism. Compared to positivism (naturalism) and interpretivism (constructivism), critical realism endorses or is compatible with a relatively wide range of research methods, but it implies that the particular choices should depend on the nature of the object of study and what one wants to investigate (Moses and Knutsen, 2007). The application of critical realism allows the combination of quantitative and 74 University of Ghana http://ugspace.ug.edu.gh qualitative methods, and accepts both forms of data as they correspond to their respective methods. From the perspective of urban geography, merging quantitative with qualitative descriptions allows for a detailed understanding that is simply not possible when purely quantitative data is considered. As Sayer (2000, p. 17) posits “meaning has to be understood, it cannot be measured or counted, and hence there is always an interpretative or hermeneutic element in social science”. Having set out the rationale for the use of critical realism to frame the research paradigm, the next section moves on to look in more detail at the mixed method research approach. Different elements of the mixed method approach are discussed including the use of triangulation through quantitative and qualitative methods. 3.3. Research methodology As articulated above, the choice of mixed methods was informed by the belief that narrow views of the world characterised by positivist orientations can be misleading, thus requiring that researchers approach complex, multifaceted and dynamic research phenomena from different perspectives and paradigms so as to gain a holistic perspective about the phenomena under study (Alvesson and Sköldberg, 2009). Post-positivists argue that by combining methods researchers can overcome the weaknesses or intrinsic biases that come from single- method, single-observer and single-theory studies (Dick, 2005). The literature on research methods indicates that the strength of mixed method design enhances both theory testing (deductive approach) and theory building (inductive approach) through extension, convergence and contradiction of findings (Creswell, 2009; Denzin, 2012). Accordingly, Grafton et al. (2011) note that the lack of use of such methods suggests missed opportunities. The emphasis on multiple sources of data under the mixed methods research strategy has given prominence to the concept of triangulation – a concept borrowed from navigation and military sciences. Just as the application of basic principles of geometry allow navigators to 75 University of Ghana http://ugspace.ug.edu.gh singly view multiple points with greater accuracy, “organizational re-searchers can improve the accuracy of their judgments by collecting different kinds of data bearing on the same phenomenon” (Jick, 1979, p. 602), thereby enhancing the rigour, relevance and validity of their results. Triangulation can take place within and between methods (Thurmond, 2001). Within-method triangulation allows the researcher to collect and cross-check data for internal consistency while the researcher’s interest for using the between-method triangulation is to test the degree of external validity (Jick, 1979; Thurmond, 2001). In term of the methodology, this research was conducted in several phases. The first approach involved a two-phase sequential exploratory mixed method research design which builds on the extensive qualitative fieldwork (comprising focus group discussions, semi-structured and in-depth interviews) conducted under the RurbanAfrica5 project (see Gough et al., 2015), to develop a quantitative instrument (questionnaire survey) for the present study. The choice for the qualitative methods by the broader RurbanAfrica project is supported by the argument that qualitative techniques are effective for understanding the lived experiences, complexities, negotiations, and shared meanings of peoples’ everyday social worlds and realities (Limb and Dwyer, 2001). It is also consistent with the assertion of Johnston et al. (2000) that qualitative methods provide access to the motives, aspirations, and power relationships that account for how places, people, and events are represented. The field experiences alongside the project settlement reports I generated helped me to identify key issues of concern in the study sites which informed the conceptualisation of my research problem and questions. At the next phase, during the main fieldwork for this thesis, the concurrent triangulation strategy which involves the simultaneous collection of quantitative and qualitative data that 5 RurbanAfrica is an EU funded research project (Grant No. 290732 SP1‐Cooperation FP7‐SSH‐2011‐2) on ‘Rural-City Connections in Sub-Saharan Africa’ 76 University of Ghana http://ugspace.ug.edu.gh are integrated at the data interpretation stage was employed (Terrell, 2012). This strategy resonates with the principles of complementary designs, in which the findings of one method are used to elaborate on or modify the findings of other methods (Sale et al., 2002). The complementary mixed methods design used in this study comprised a blend of both secondary and primary data collection instruments. The secondary data sources included the Population and Housing Census (PHC) and Ghana Living Standards Survey (GLSS) reports. The 2010 census data was analysed to ascertain population figures, economic activity levels/types and service provision averages for GAMA and the study localities. The GLSS 5 and 6 reports were analysed to provide the basis for the income poverty analysis. Other policy documents on decentralisation, utilities provision and poverty reduction were used to supplement findings made in the study. The primary data entailed quantitative and qualitative techniques which are discussed below: i. Mapping of services and questionnaire survey for households and home-based enterprises (HBEs); ii. Focus group discussions (FGDs) with residents and HBE operators; and iii. Key informant and expert interviews with stakeholders and community opinion leaders. 3.3.1. Mapping of services and households and HBEs survey Guided by the notion that place matters, mobile interview methods involving mapping of facilities such as standpipes, boreholes, and vendor service points (using hand-held Global Positioning System (GPS) units), transect walks and follow-along participant observations (FAPO), were conducted to obtain, at first hand, the inefficiencies associated with urban services and conditions of HBEs. Such mobile geography methods generate rich spatial observations and theories grounded in lived experience (Finlay and Bowman, 2016). In the next stage, quantitative data involving 400 households and HBEs in four residential areas in GAMA – Korle Gonno, Accra New Town, Abuja and Gbawe were collected with the 77 University of Ghana http://ugspace.ug.edu.gh use of interview-based questionnaires (see Appendix 1). The questionnaire administration was done as cross-sectional survey. The use of interview-based questionnaires facilitated inter-personal contact between the researcher and the researched. This resulted in higher response rates, as articulated in earlier studies (Yansaneh, 2005). The draft questionnaire was piloted prior to the actual fieldwork to help clarify and re-word portions of the research instrument, thereby minimising potential non-response errors during the fieldwork. Besides socio-economic and demographic profiles, the survey inter alia explored households and HBEs and their water and electricity access dynamics, including the role of governance in poverty reduction. The quantitative data generated on households and HBEs was done so as to answer a number of important sub-research questions with the aim of addressing the three objectives guiding the study. Specifically, the interviews with households were meant to answer questions like what are the sources of water and electricity? Why do households use these sources? How do the dimensions of distance, reliability, quality, and cost interact to affect households’ access? How are the dimensions of access differentiated by communities? What is the impact of water and electricity access on household income? How do households perceive poverty and how do their perceptions relate to access to these services? The interviews with HBE operators were also meant to answer questions like what are the water and electricity sources used by business operators? Does the use of household water and electricity for HBE operations affect households? If so, why do HBEs still find it relevant to use household electricity and water for their businesses? Do HBE operators have other opportunities to access these services; and how does poor access impact business income and poverty? 3.3.2. Sampling methods and data sources Previous studies (Behrens et al., 2006) have suggested that ascertaining an appropriate and adequately representative sample size for surveys conducted in African cities, where 78 University of Ghana http://ugspace.ug.edu.gh resources are somewhat scarce, is a huge challenge. Usually, the use of large sample sizes in surveys implies that ‘more is better’ and that the larger the sample size, the closer to reality the outcome will be (Twumasi, 2001). However, Yansaneh (2005) and Agyemang (2015) have challenged this widely held view. It is contended that large sample sizes do not necessarily reflect the actual demographics of the entire population as there may be a correspondingly large number of non-sampling errors. Mindful of this, and on the basis of Yamane’s (1967) simplified statistical table, a total sample size of 400 respondents was considered appropriate for the households and HBEs survey. Yamane’s (1967) simplified statistical table states thus: n =  1+N (e) 2 where ‘n’ is the sample size, ‘N’ is the population size, and ‘e’ is the level of precision. The sample size assumes a sampling error of + 5%, a confidence level of 95% and a 0.5 maximum variability. The use of Yamane’s (1967) simplified statistical table as the statistical basis for determining a sample size for this study is justified in the sense that it assumes a normal distribution of salient characteristics among the sample population being measured (Appeaning-Addo, 2012). In adopting this sample size, the key considerations were the research objectives and population of localities studied in GAMA. Second, to expand on an earlier study in similar communities (apart from Abuja) in Accra, but under different context and scale, by the broader RurbanAfrica project in which I was actively involved (see Gough et al., 2015; Amankwaa et al., 2015). The outcome of the application of the above formula on the sample size in the four localities is illustrated as follows: Korle Gonno 21% representing 87; Accra New Town 23% representing 93; Gbawe 48% representing 195; and Abuja 5% representing 23; altogether 79 University of Ghana http://ugspace.ug.edu.gh making 398: extra 2 samples were added to round up the figure to 400. However, generating a sample frame based on these divisions was complicated logistically and financially especially in Accra New Town which recorded the highest sample size of 195. Moreover, selecting only 23 respondents from Abuja (which recorded the lowest sample size) was worrying as studying informal settlements has always been bedevilled with challenges due to the lack of reliable systematic data and the difficulties in relation to the boundaries of the population. Based on the above reasons, each of the four communities was assigned a predetermined sample size of 100. Similar reasons account for why a sample size of 50 was allocated to each cluster during the multistage cluster sampling strategy. Thus, 100 was practically a reasonable sample size in each locality to form the basis of any statistical analysis (Yansaneh, 2005). Thereafter, a stratified multistage cluster random sampling design was employed to select respondents for the household survey. The multistage sampling design was implemented in two stages. In stage 1, a stratified sampling of residential units into clusters was performed following a reconnaissance survey of the study localities. Stratification assumes that the population within a particular class will exhibit similar socio-economic characteristics. However, the challenge with this logic is that the correlations among units in the same cluster could inflate the variance, implying that the precision of survey estimates could be lowered substantially, compared with a design in which households are not clustered (Yansaneh, 2005). Earlier studies by Appeaning-Addo (2012) and Songsore et al. (2009) have stratified GAMA on the basis of housing density and demographics into low, middle and high-income areas. The stratification of households into any of these three income strata can be problematic, bearing in mind that some wealthy people in Accra live in low-income areas (Songsore, 2008). Mindful of this, the stratification of households, in this study, is based on the current 80 University of Ghana http://ugspace.ug.edu.gh structural conditions and infrastructural service provision. These are used as proxies for delineating the study sites into well-serviced and poorly-serviced clusters. The first cluster, well-serviced area, is generally composed of single family housing units, or several individual households living in different units of the same compound, a phenomenon termed as ‘multi-habitation’ or compound housing system (Appeaning-Addo, 2012). In most cases, the infrastructure network in such areas is well-planned and service provision is somehow reliable. Households generally have the means to complement official water and electricity provision with alternative sources like large storage tanks, boreholes, rechargeable lamps and generators. In sharp contrast, the second cluster, poorly-serviced area, is generally made up of large multi-habited compound houses generally inhabited by large family sizes with high densities especially in the indigenous core. A key defining characteristic of the poorly-service area is the extent to which households rely on shared facilities like in-yard standpipes and pre-paid meters. While some houses may have separate connections, such facilities are usually illegally connected. Households usually cope with access challenges by purchasing water from vendors and using candles and torchlights. These two clusters provide the background to examine the influencing roles of spatial and infrastructural variables which are hypothesized in this study to greatly influence households’ access to services and their livelihood activities. In stage 2 of the multistage sampling design, each cluster in a particular study site is assigned a pre-determined cluster size of 50 as both clusters face similar concerns regarding the dynamics of access to water and electricity albeit with different coping strategies. Following this, a systematic sampling technique was employed in the selection of every 7th house beginning from a random starting point (the first number obtained through a lottery system). On few occasions, it was hard getting the right count because of how houses are located and designed. In such cases the next house immediately after the 7th count was considered. In 81 University of Ghana http://ugspace.ug.edu.gh instances where the household head or potential heads in a selected house were unavailable or declined to participate in the survey, any other accessible and willing adults were considered. Also, where selected housing units had more than one household being resident there, only one household was randomly selected and interviewed. The process was repeated until the said cluster size for a locality has been obtained. For every interviewed household, data on HBEs engaged in by all household members was collected for the entire household per a residential unit. It was found that in terms of the HBEs distribution, the majority of households – 85% (N=340) – engage in HBEs in GAMA. Both landlord/caretaker’s households and tenant households (as well as newcomers and old residents) were targeted to account for the heterogeneity in urban low-income populations, whereby landlords manage utilities for the entire house and houses double as spaces for business activities. The household and HBE interviews made it possible to assess the dynamics regarding access to water and electricity, and also measure the impact of these services on home businesses. Suffice to add that, the greater part of the analysis on livelihoods is based on the 340 HBE operators. The unit of analysis in the survey is the household, which is an important sampling unit in the African context. Meanwhile, the concept lacks a universal definition (Behrens et al., 2006). Following the Ghana Statistical Service (GSS), this study defines a household as a “person or group of persons, who live together in the same house or compound, share the same house- keeping arrangements and recognize one person as the head of household” (GSS, 2012, p. x). Typically, a household may consist of the nuclear and extended family members as well as house helps co-habiting in a house. Household heads in the sampled houses constitute potential respondents, based on their availability and receptivity to participate. As defined by the Ghana Statistical Service (GSS, 2012), household heads as used in this study refer to individuals who wield economic and social responsibility for the entire household. However, 82 University of Ghana http://ugspace.ug.edu.gh the heads are not (in all cases) responsible for the sustenance of the entire (extended) family. In their absence, all accessible adults were considered potential participants in this study. A semi-structured questionnaire (with both closed and open-ended questions) was used to elicit information from respondents in the form of face to face interviews (Wisker, 2008). The local languages (Twi, Fante, and Ga) were mostly used though in some cases people who are literate opted to be interviewed in English. The survey questionnaire was structured into six thematic parts reflecting the various thematic issues under investigation by the researcher (see Appendix 1). Some of these thematic issues include household access to water and electricity; home-based enterprises; services, business activities and poverty nexus; background characteristics among others. Construction of the survey instrument was inspired mainly by the objectives of the research, the research question under investigation and the literature and theoretical orientation of the subject matter. The questionnaires were administered by myself and two experienced male research assistants (who are in their early thirties and MPhil degree holders) between November and December 2015, with each taking about 45-60 minutes to complete. The assistants were trained by the researcher over a period of two days after which the instrument was pretested through a pilot at Ga Mashie area which shares similar characteristics with most of the study localities. In sum, these detailed approaches largely addressed objectives one and two. The qualitative part of the complementary mixed methods design, which mainly focuses on objective three, is presented in the next section. 3.3.3. In-depth interviews and focus group discussions (FGDs) The qualitative approach and its associated research methods are useful for obtaining data capable of advancing our understanding of the “sets of meanings which people use to make sense of their world and their behaviour within it” (Cohen et al., 2011, p. 9), in this case residents’ access to services and livelihood. For this study, the qualitative data employed the within-method triangulation which allows multiple techniques, in this case key-informant and 83 University of Ghana http://ugspace.ug.edu.gh expert interviews and focus group discussions, within a single method (Creswell, 2009). The choice of within-method triangulation is justified as it enabled me to collect and interpret data and crosscheck for internal consistency (Thurmond, 2001). Using both within and between method triangulation approaches in a single study, as this study is designed, provides valuable insights that cannot be gained from a single method (Creswell, 2009). Interviews were conducted with 36 key informants who were purposively selected from the perspective of experts in the fields of governance and service delivery, including: three policy makers (PURC, ECG and GWCL); three planning and governance experts (MLGRD, AMA- TCPD, NCAP); eight service providers at the community level; four assembly members; 12 opinion leaders; and six officials of NGOs and CBOs. Data on interviewee’s gender, educational level, and current designation were collected which provided important context for the interpretation of the study findings in the analytical chapters. Meanwhile respondents were unwilling to provide data on age, years of experience and the number of years served in their current portfolio. The key stakeholders were interviewed on their insights of GAMA’s present decentralised governance system, level of service delivery, priority issues for the urban poor, and strategies to strengthen the linkages between water and electricity access and income-generating activities, and overall poverty reduction. Besides examining what policy makers, industry players and other professionals were doing to address concerns regarding service delivery and livelihood activities, the interviews also focussed on some aspects of the quantitative survey that delved into the perception of households and HBEs on the quality and affordability of services. The interviews complemented the survey and provided a better context to addressing the objectives of the study. At the community level, the key informants were interviewed on issues of local governance, access to and use of services, and poverty. Also discussed were the main challenges confronting district assemblies and other local authorities in discharging their roles and what can be done to address their challenges and strengthen the community interface. 84 University of Ghana http://ugspace.ug.edu.gh Each of the conversations with the key informants and experts lasted an average of 45-60 minutes. The interviews were held in the offices and sometimes homes of the participants. In most instances, the interviews were tape recorded upon the consent of the informants. Interview guides were used to direct the flow of the conversation (see Appendix 2). The researcher was assisted by the two research assistants in conducting the interviews and focus groups. In terms of the FGDs, a total of 8 focus groups were conducted in the four research localities consisting of male and female youth and elderly male and female. Thus, two focus groups, involving eight discussants in each group, were held in each locality with respondents all aged above 18 and engaged in a range of livelihood activities. In all, the FGDs engaged 64 people with the aim of further expanding our understanding of the on the ground realities of water and electricity access, livelihoods and poverty nexus, which is the overall objective of the study. The focus groups were conducted to further probe some of the views already obtained from the survey and interviews. Interesting cases or stories were followed up to ensure that the discussions delved deeper into the everyday experiences of households and HBE operators with respect to water and electricity access and the impact of services on businesses. The focus groups were organised with the help of assembly members and their representatives. They were mostly held in a suitable location such as a community room or on the compounds of assembly members. Discussants were selected through a non- probabilistic convenience sampling technique (where participants are selected based on their convenient accessibility and proximity to the researcher). Suffice to add, however, that the selection was representative of residents in the poorly-serviced and well-serviced clusters, landlords/indigenes and tenants, old residents and newcomers, HBE owners and operators, and association representatives. In a few cases, some of the groups included interviewees who provided unique and insightful cases during the survey. This provided the avenue to validate 85 University of Ghana http://ugspace.ug.edu.gh the reliability of such data and also make sure such issues were known to other community members. Each session lasted about 90 minutes. Refreshments were offered to the participants at the end of the discussions. All of the FGDs were tape recorded with the explicit consent of participants. Essentially, the group discussions afforded the researcher the opportunity to observe the discussants debate, agree or disagree with comments expressed by other discussants until a consensus was reached by the group. This approach was a practical way of ensuring that the conclusions of the study corresponded to the experiences of households and HBEs as well as the overall objectives of the study. The downside, however, is that the phenomenon of social loafing could occur whereby a participant exerts less effort in a group setting. Also, discussants could have acted in a way that could have prevented a member from freely expressing an idea or opinion, which the group collectively did not want to share with outsiders. Appendix 3 presents the guide used for the FGDs. Employing the concurrent triangulation strategy (Terrell, 2012), the qualitative data collection was undertaken alongside with the household survey, usually towards the end of the survey in each locality. In the end, these collaborative approaches with the study participants unravel and respond to challenges in the water and electricity sectors and contribute to knowledge on how policy makers could initiate and implement tailor made interventions for poverty reduction. 3.3.4. Data processing and analyses Data generated during the household survey was processed to ensure that all questionnaires were completely and accurately filled out. Each questionnaire was coded, with the assignment of a unique number, and inputted into a spreadsheet in the Statistics Package for Social Sciences (SPSS 17) and Stata 12. A number of statistical tests were performed, including the descriptive statistics and cross tabulations performed on the quantitative data to 86 University of Ghana http://ugspace.ug.edu.gh analyse the dimensions of access and measure the impact of access on HBEs. Descriptive statistics performed on the dynamics of access and association with poverty are presented as charts and tables for comparative analysis. The Pearson chi-square was performed simultaneously to test whether any observed differences in the data were indeed statistically significant (see Appendix 4). All recorded interviews and FGDs were transcribed verbatim (the ones in the local Ghanaian languages were translated into English). Following the transcription, the data was carefully read and organised. During these processes, codes, which involved reducing the data set into smaller meaningful segments associated with specific study objectives and assigning names to these segments, were generated based on the emergent categories identified within the data. This allowed key themes to emerge from the responses and were organised based on descriptive and analytic codes. The former “reflects themes and patterns that are obvious on the surface and are stated directly by research subjects” (Cope, 2010, p. 283) while the latter reflects “a theme the researcher is interested in or one that has already become important in the project” (ibid, p. 283). Coding the data helped support the study’s theoretical concepts and approaches and generate new insights from the empirical data. Direct quotations from participants were used to support the findings and relate them to their lived experiences. 3.3.5. Data limitations The primary data from which this study is drawn is beset with some limitations. First, similar to Gough et al. (2015), some respondents shied away from sharing their specific personal details, particularly on income, age and years of working experience. There were some HBE operators who stated their incomes as a range, and averages were calculated based on their responses. Others could not readily indicate their incomes, hence their specified daily savings was treated as their incomes. This could have introduced uncertainties in the income data since respondents may be earning more than they reported. In spite of the explanation that 87 University of Ghana http://ugspace.ug.edu.gh this study was purely for academic purposes, some interviewees especially the public officials were cautious in disclosing their age, years of experience, and the number of years served in their current portfolio. Under such circumstances, it is dicey to determine whether respondents purposefully understate or exaggerate their ages and years of work depending on their position and retirement concerns. This could also be a source of uncertainty in respondents’ personal data which provides important context for the interpretation of the study findings. Second, translating survey questions from English language into the Ghanaian languages – Twi and Ga – for the majority of respondents could have introduced some distortions. For instance, attributes which help to understand poverty, such as ‘quality of life’ and ‘standard of living’ were difficult to translate as they could mean different things depending on the context. To overcome this challenge, an in-house demonstration using English and these two local languages was organised prior to the data collection. Also, daily meetings were held to discuss new challenges with the translation and other practical matters related to the conduct of the survey. A discussion of how the researcher negotiated the issue of positionality during the field work ensues in the next section. 3.4. Negotiating power, positionality, and ethics Before the data collection, the study was granted approval by the College of Humanities Ethics Committee of the University of Ghana (see Appendix). The importance of ethics in research cannot be overlooked (Resnik, 2015). Hence, the objectives of the study were thoroughly explained to participants, and what was expected of them, for their oral consent before all surveys and interviews were conducted. Participants were assured utmost confidentiality. Pseudonyms were used to ensure the anonymity of participants. In most cases, however, participants were not concerned about disclosing their identity and wanted their names and faces to be revealed. Efforts were, also, made to carefully ensure that the contacts 88 University of Ghana http://ugspace.ug.edu.gh (assembly members and traditional authorities) who facilitated smooth community entry into the various research localities only provided the general access to all potential respondents, and on the basis of sampling procedures, the researcher and his team led the survey to avoid any plausible biases in the selection of participants for the study. Wadel (1991, p. 75) indicates that “...the collection of data is a discriminating activity, like the picking of flowers, and unlike the action of a lawn-mower...”. This implies that data is not something out there, ready to be collected. The researcher is selective in the research process and has subjective influence when choosing respondents. My position as a young male doctoral candidate, the subjectivity of each participant, and our social interactions also influenced the entire research process. This makes the researcher’s positionality very crucial. The concept of positionality refers to the notion that a researcher’s characteristics (including age, gender, education, class, cultural and ideological background) vis-a`-vis the researched (collaborators) can influence access to participants and the data that are produced (Mohammad, 2001). In social science disciplines like geography, the relevance of situating knowledge in particular contexts and articulating the positionality of the researcher cannot be overemphasised (Teye, 2012). As rightly put by Cormode and Hughes (1999), a dimension of positionality that is particularly important in mixed methods research is power relations between the researcher and the researched (or collaborators). This is because mixed methods often imply collecting data from different groups of respondents transcending techniques, suggesting that the researcher needs to adapt quickly to changing power relations which can be challenging. In this study, the significance of positioning and personality to the research processes was dependent upon a number of factors. First, the data sought is considered to reside with the participants, therefore, the research method is being used as a tool to mine the data. Second, the data is considered to be constructed through the process of the research encounter and therefore participants are attributed with qualities such as agency. 89 University of Ghana http://ugspace.ug.edu.gh In line with a critical realist standpoint, the implications of my positionality varied during the entire research process. The mere fact that the researcher and the two research assistants are Ghanaians, and share similar cultural traits as many of the respondents in the study sites, did not make our entry and acceptance any easier. The positionality ‘address’, whether as insiders or outsiders, has profound impact on the receptivity by study participants and access to information. For example, when sensitive issues such as the practice of illegal connection of power and water, and poverty issues were being interrogated, our positionality became a key concern in the research process. Such cases required a high degree of reflexivity on our part. One way this was handled was to relate to the participants by ‘putting ourselves in their shoes’ and for instance asking questions such as ‘since supply is absent or irregular how do you manage?’ This way respondents could situate their situation within the broader national water-energy and poverty issues. On the contrary, when some of the common governance issues like the state of the implemented lifeline tariff targeted at offering subsidised water and electricity to lower-income households, and the impact of poor services on businesses were explored, power relations played a much less predominant role – although in some instances, government officials became wary of sharing information with the research team. The implications that a particular researcher’s positionality has on research is incredibly complex and equally dependent on the power structures that exists within communities. For example, it is likely that some residents agreed to participate in the fieldwork because of the respect they have for the chief or assembly member who granted us the community entry. Also, it was possible for some participants to be economical with information and present a positive portrayal of the community to remain in the good books of the local authorities. These experiences relate to the theoretical perspectives on the merits and demerits that a researcher may encounter by being perceived as an ‘insider’ or an ‘outsider.’ It is often argued that insiders (i.e., researchers who study a group to whom they belong) have an advantage because they are more likely to be trusted by respondents, and can therefore access 90 University of Ghana http://ugspace.ug.edu.gh information that would not be given to an outsider. A counter argument is that outsider researchers are more objective, open-minded and are able to win trust and gain access to information without distorting the meanings of these responses (Mohammad, 2001). Meanwhile, Mullings (1999) has challenged these traditional notions, and argued that the ‘insider–outsider’ boundary is not stable, and that it is actually subject to the dynamism of positionalities in time and space. For instance, based on our insider status (as Ghanaians), respondents occasionally tended to take our knowledge of the local situation for granted and sometimes reckoned that we were already aware of their challenges. In other situations, particularly during the focus groups, our outsider address (as non-residents, emphasised by geographical distance) allowed us to get more insight into the local situations because the respondents found it more comfortable revealing information pertaining to their lived experiences to ‘strangers/outsiders’. The research process remained fluid, depending on changing circumstances, communities, and individuals. Thus, mixed methods researchers face a bigger challenge of learning to adapt to the dynamism of positionalities (power relations, and insider–outsider boundary dynamics). To this end, the researcher’s disposition to the study should not significantly affect the research process and the validity and reliability of the results. 3.5. Profile of respondents Having discussed the study area and methodological processes, the impetus of this section is to profile the households and HBEs that were interviewed during the fieldwork. The profiles captured in Table 3.2 include: gender, age, educational level, marital status, household size and number of rooms, and length of stay in residence and household enterprise. 91 University of Ghana http://ugspace.ug.edu.gh Table 3.2: Socio-demographic characteristics of respondents Variables Categories Freq. % Female 226 56.5 Gender Male 174 43.5 15 – 29 years 143 35.7 Age 30 – 39 years 161 40.2 40 – 49 years 55 13.8 50yrs and Above 41 10.3 None/No formal education 40 10.0 Educational level Primary/Middle/JHS 205 51.2 Secondary/SHS/ O’Level 82 20.5 Polytechnic/Nursing/Vocational 47 11.8 Tertiary/Higher 26 6.5 Single/Never married 122 30.5 Marital status Married 174 43.5 Widowed/Divorced/Separated 66 16.5 Cohabiting/Consensual union 38 9.5 1 – 3 persons 172 43.0 Household size 4 – 6 persons 169 42.2 7 – 10 persons 45 11.3 Above 10 persons 1 4 3 . 5 Single room 262 65.5 Number of rooms 2 rooms 77 19.2 3 rooms and Above 61 15.3 Landlord/landlady 103 25.7 Tenure status Tenant 187 46.8 Free habitation/family member 99 24.7 Caretaker/Other 11 2.8 Compound/Multiple shared 257 64.2 Housing type Single family detached 106 26.6 Self-contained/Apartment 37 9.2 1 – 3 households 146 36.5 Number of households 4 – 6 households 98 24.5 7 – 10 households 67 16.7 11 – 20 households 70 17.5 Above 20 households 19 4.8 0 – 5 years 95 23.7 Length of stay in residence 6 – 10 years 63 15.8 11 – 20 years 78 19.5 21 – 30 years 94 23.5 Above 30 years 70 17.5 0 – 3 years 118 30.3 Length of stay in business 4 – 6 years 89 22.8 7 – 10 years 82 21.0 11 – 20 years 79 20.3 Above 20 years 2 2 5.6 Source: Field data, 2015 92 University of Ghana http://ugspace.ug.edu.gh The sample shows a relatively unequal distribution between male and female respondents with 56.5% of them being female and 43.5% male. This finding is consistent with the report by the Ghana Statistical Service that there is an increasing phenomenon of female-headed households (34.7%) in urban Ghana (GSS, 2012). It is possible this high percentage of female respondents is due to the time of the day the survey was conducted, as women in the study area are mostly found in the house as caretakers and owners/operators of HBEs. Traditionally, females are known to be responsible for managing household electricity and water issues leaving them with a lot of lived experiences. The finding further supports studies which indicate that women still dominate the informal sector in Accra (Anyidoho and Steel, 2016; Amankwaa, 2017). In terms of age, the sample is dominated by those in the younger age bracket. The majority (40.2%) of the respondents in all study localities comprise middle-aged people within the age group of 30–39 years, while a little over one-third (35.7%) of them are young people below 29 years. The remaining one-quarter of the respondents fell into the older age groups, with 13.8% of them between 40–49 years and 10.3% being 50 years old and above. The overall dominance by the younger age category is hardly surprising since Accra has a young population (GSS, 2012). The regional average shows a mean age of 26 years (GSS, 2013a). The sample provides the background to illustrate how the young and old perceive and experience access differently, in terms of bill payments, sourcing water or buying electricity top-up, and managing service usage. The sample is also a useful marker for exploring the informal livelihood activities undertaken by both the young and old populations. Prior studies illustrate cases how young people have taken over HBEs from their parents ensuring continuity of the businesses across generations (Gough, 2010). Views of younger generations are important to observe because they hold responsibility for any future policies and societal attitudes. 93 University of Ghana http://ugspace.ug.edu.gh A little over half of the sample (51.2%) have primary and junior high school (JHS) education. Also, one-fifth (20.5%) of them have a senior high school (SHS) or O’Level education. Out of the remaining sample, 11.8% have Polytechnic/Vocational qualification, and slightly less (10%) have no formal education. Those with a tertiary/higher degree constituted 6.5%. This trend reflects the general situation within the Ghanaian urban population where increasingly a number of young people after successful completion of their basic and SHS levels are unable to further higher education due to limited opportunities and financial cost (GSS, 2013a). Suffice to add that, the situation in the study localities is slightly better than the national averages. In Ghana only about 1.7% of the population have attained a University education whiles 17.9% and 7.8% have Junior and Senior high qualifications respectively. As much as 23.4% have no formal education in their lives (GSS, 2012, p. 49). These observations seemingly explain why many find refuge in informal livelihood activities, as by Ghanaian standard such qualification may not even guarantee the minimum public sector work which is poorly paid and hitherto non-existent. At the same time, the overall 70% of the sample having at least SHS education further suggests that informal enterprises are no longer a livelihood strategy for people with little or no formal education. In the light of constraints that people encounter, dwindling formal sector job opportunities and increasing unemployment situation, it has become necessary to transcend education reasons about particular economic activities in order to eke a living in the urban economy (Amankwaa, 2012, 2017). This finding is, thus, not only beneficial for assessing the relationship between education and access to services but also the dynamics of informal livelihood activities. A significant portion (43.5%) of the sample is married. Those who are single and never married constitute nearly one-third (30.5%) while 16.5% are within the divorced/widowed/separated category, and the remaining 9.5% cohabiting or living in consensual unions. Some gender differences were observed: among the females, 56% are married compared to 70% for males. About 13% of females are cohabiting compared to 5.3% 94 University of Ghana http://ugspace.ug.edu.gh for the men. The proportion of the never married among females is 29% compared to 25% for males. Proportionally, therefore, there are more women living in either marital relationships or in union compared to men. This supports the observation made by Langevang (2008) that the present economic circumstances in Ghana make it difficult for many young people to be able to marry and sustain marital relationships and this explains the high proportion of the never married in the sample. This background gives a sense of household headship which directly relates to the resource pool available for household income and the existing arrangements these categories may have in relation to accessing water and electricity. The distribution helps to appreciate married or single-headed households’ motivations to engage in HBEs, and how that contribute either as the main source of income or strategy to supplement household budget. Household size and number of rooms occupied are proxy indicators of a household’s economic status, with direct implication for access to water and electricity. A significant portion (43%) of the sample is within the 1–3 household size group. Slightly less than that (42.2%) have their household size to be 4–6 persons. Households with 7–10 persons constitute 11.3% and only 3.5% have more than 10 persons in a household. This finding is less surprising given that low-income settlements are noted for having larger household size. This is consistent with the city-level mean household size which is 4.75 persons per household (UN Habitat, 2011). Tied to household size is the number of rooms occupied. Expectedly (as together 85% of the sample have a household size of between 1–6 persons), the data show that nearly two-thirds (65.5%) occupy a single room. This is followed by 19.2% occupying two rooms while the remaining 15.3% occupy three rooms or more. Despite the existence of economies of scale in household consumption, larger households tend to be poorer as they may have to support more dependents (World Bank, 2011), resulting in lower per capita income. This observation has implications for access to water and electricity, which tend to 95 University of Ghana http://ugspace.ug.edu.gh represent a large share of household budget, suggesting that strategies to help increase income generating activities are very significant. The issue of housing type and tenure status connects to household arrangements in relation to accessing water and electricity and space for home business. Nearly two-thirds (64.2%) of the respondents live in multi-habitation. Single family detached housing (including wooden kiosks and metal containers) constitutes 26.6% while the remaining 9.2% live in single family self-contained apartments. Also, the majority (46.8%) of respondents are tenants. Landlords and family members with free habitation together make up 50.4% of the sample (25.7% and 24.7% respectively). The remaining 2.8% are caretakers. This background is useful because typically home owners are responsible for managing water and electricity. This may have consequences for the majority of tenant households. In particular, apart from the cost burden this situation may influence how tenants handle the politics of space access to ran a business. Regarding the number of households, more than one-third (36.5%) live in houses containing 1–3 households. Nearly a quarter (24.5%) of houses contain 4–6 households. A slightly balanced distribution was observed between houses with 7–10 households (16.7%) and those with 11–20 households (17.5%). Only a handful (4.8%) live in houses with more than 20 households. Melara Arguello et al. (2012) reported a fairly similar finding that most houses in low-income areas have an average of six households. This distribution tends to come with the phenomenon of ‘space clamoring’ as it may hinder the ability to easily access a workplace to operate a home business. Typically, members of the landlord’s family may have the whip hand in securing spaces than tenant households. Consequently, this unequal access to working space tends to have a positive relationship with the preferred choice and scale of livelihood activity. For that reason, it is unsurprising that HBEs are evolving and witnessing a reconfiguration from the home space to satellite spaces such as roadsides and alleyways. 96 University of Ghana http://ugspace.ug.edu.gh In terms of residence, the majority (23.7%) of the respondents have lived at their current residences for between 1–5 years. Respondents who have lived for 6–10 years constitute 15.8% while those in the 11–20 years category represent 19.5%. Those who have lived for 21–30 years make up 23.5% followed by those who have lived for more than 30 years (17.5%). This ties in with length of stay in business, where nearly one-third (30.3%) of the sample have ran their HBEs for less than 3 years. Those who have ran the business for between 4–6 years followed with 22.8%. Slightly less (21% and 20.3%) have ran their HBEs for the past 7–10 years and 11–20 years respectively. Operators with more than 20 years stay in business constitute only 5.6%. The relatively balanced distribution between length of stay in residence and business is significant because length of stay is a key determinant of strategies for coping with access to water and electricity, especially through established acquaintances and networks which are known to provide the wheels on which home businesses in particular thrive. 3.6. Conclusion In this chapter, the selection and justification of GAMA, and the four research localities, as the study area have been discussed. The philosophical orientation, data used, and the methods employed in obtaining the data have also been presented. Informed by the research problem and objectives, and the fact that the study subjects are poor urban residents operating livelihood activities in the informal sector, a multiple research approach was adopted. This approach involved a combination of qualitative and quantitative techniques under the (critical realist) philosophical assumption that the two paradigms can indeed complement each other. Quantitative primary data was obtained principally through questionnaire surveys. Qualitative primary data was derived from the key informant and multi-participant interviews, focus groups, and other mobile geography methods including follow-along participant observations (FAPO) and mapping of facilities and HBEs. Both methods were designed and implemented to achieve data triangulation and tap their complementary 97 University of Ghana http://ugspace.ug.edu.gh advantages. Various statistical analyses were performed on the quantitative data while the qualitative data was subjected to thematic analyses. Secondary sources of information, particularly Population and Housing Census and Ghana Living Standard Surveys, were relied on in the study. Some data limitations, positionality and ethical issues have also been addressed. Finally, a profile of the households and HBEs has been explored in the chapter. In sum, it has been shown that mixed methods are crucial in investigating and understanding the lived realities of people in low-income households particularly in relation to their access to water and electricity and livelihood strategies. The ensuing analytical chapters provide a detailed exploration of these issues. The next chapter discusses the dynamics of access to water and electricity. 98 University of Ghana http://ugspace.ug.edu.gh CHAPTER FOUR THE DYNAMICS BEHIND WATER AND ELECTRICITY ACCESS 4.0. Introduction This chapter analyses and discusses research findings with respect to objective one, which seeks to identify and examine the dynamics of access to water and electricity in GAMA. The chapter is divided into two main sections. The first part explores the dynamics of water access, while the second is devoted to the dynamics of electricity access. Specifically, the chapter delves into the spatial aspect of access, and situates access within a deeper perspective of proximity, reliability and quality in relation to cost and usability of services. It, thus, demonstrates how these dynamics look beyond access/coverage, and manifest themselves within the Accra metropolis. 4.1. Dynamics of water access 4.1.1. Spatial dynamics of water access Based on the 2010 census data, the population of GAMA has better access to water relative to the national and regional averages, as illustrated in Table 4.1. The data show that 60.2% of households within the Accra metropolis have access to pipe-borne water in a two-tier hierarchy including in-house and in-yard sources, compared with 54.4% at the regional and 33.5% at the national levels. The latter result indicates a reduction from the 2000 census, which stood at 39.9% (GSS, 2013a), largely on account of a surge in the use of vendor standpipes. With regards to settlement variations, access in Korle Gonno (66.7%) and Gbawe (64.2%) appears better than the average for GAMA. Abuja does not exist officially, since it is not recognized as a residential zone, and as such lacks government-provided services including pipe-borne water facilities, thereby consigning them to vendor sources. 99 University of Ghana http://ugspace.ug.edu.gh Table 4.1: Spatiality of water coverage at national, regional, metropolitan and locality levels Selected settlements in GAMA Sources of water National Regional Metropolitan Korle Accra New Abuja Gbawe Gonno Town Total 5,467,054 1,036,370 501,903 8,599 8,352 - 16,624 Pipe in-house 790,493 272,766 159,701 3,006 1,433 - 6,260 Pipe in-yard 1,039,667 291,107 142,303 2,726 2,901 - 4,410 Sampled total (%) 33.5 54.4 60.2 66.7 51.9 - 64.2 Source: Computed from Ghana Statistical Service (GSS), 2014b At the same time, the GSS (2013c) and World Bank (2015a) report that among the six metropolitan areas in Ghana (namely: Accra, Kumasi, Sekondi-Takoradi, Tamale, Tema and Cape Coast) Accra’s population with access to pipe-borne water declined by about 22% between 2000 and 2010. This raises question marks over the sufficiency and dependability of macro-level data, as it provides a fairly positive picture, albeit excluding the quality of service provided. Furthermore, the statistics only provide information on averages, and therefore conceal important differences between and within neighbourhoods in terms of actual access to water service and equity. This necessitates delving into the study settlements to investigate the realities of water access and everyday strategies used to patch up and ameliorate the gaps and deficiencies in water supply. The reportedly used sources of water have been categorised and illustrated in Figure 4.1. The analysis reveals that households’ access to water is mixed, complex and enmeshed in a water service supply network (WSSN) which is characterised by a combination of sources including in-house, in-yard, standpipes, boreholes and water tanker delivery. Thus, because of supply challenges of GWCL, a substantial number of informal actors have emerged to serve residents. These informal players may serve as the only water providers to some users or render a complementary service in areas where the main pipe-borne water supply is rationed. 100 University of Ghana http://ugspace.ug.edu.gh Figure 4.1: Distribution of water sources in the study area 44% 31% 10% 5% 5% 3% 2% Public Stand In-yard In-house Public Borehole Public Water Private Own Private Pipe connection connection Tank Borehole Water Tank Source: Field data, 2015 From Figure 4.1, it is clear that in terms of private water sources, 31% of households share an in-yard connection whereas only 10% have in-house GWCL connection, with another 5% complementing GWCL connection with domestic sources such as private borehole and water tank. Regarding public water sources, 44% of households make use of privately operated vendor standpipes while the remaining 10% rely on public borehole and water tank. Overall, the data indicate that a considerable percentage of households, representing 41%, tend to have infrastructure in their residence, either as in-house or in-yard, capable of providing water via a pipe connected to GWCL. A key point to note is that, sharing a yard tap among households seems a good approach to enhance access (coverage) provided the users can get enough water and do not have to revert to procuring water from vendors. For a greater majority of respondents (44%) who are unable to access pipe-borne water directly from their homes, water can be sourced from public standpipes operated by private vendors with large storage tanks. It is worth mentioning, however, that the presence of pipes and infrastructure capable of providing water does not necessarily equate to the ability to source water. It is one thing having the tap and it’s another seeing water flow. As a result, there was a fairly equal distribution of respondents who rely on tanker services and 101 University of Ghana http://ugspace.ug.edu.gh boreholes/wells as alternatives during interruptions in water supply. Such tendencies have given prominence to the water service supply network (WSSN), albeit with time, quality and cost implications. The dynamics of water access in GAMA is well understood, if looked at from neighbourhood- specific peculiarities. This is because generalized characterizations such as declarations about ‘access to water has increased’ as opposed to concrete evidence based micro-level analysis can be misleading (Obeng-Odoom, 2012a). Table 4.2 summarises the distribution of available water sources used by households to satisfy their water needs. Even some landlords carry gallons to the vendors. A chi-square test was conducted to ascertain whether or not there are significant differences in the number of people who use different sources of water in the different communities. The analysis revealed that there is a significant difference among the number of people who obtain water from all the different sources across the study localities at a 99% confidence interval, with the exception of vendor tank. This implies that any difference in the number of people who obtain water from vendor tank is due to chance. Table 4.2: Settlement distribution and chi square analysis of water sources Vendor Settlement In-house In-yard Private Private Vendor Vendor borehol connection Connection Tank borehole standpipe tank e Korle Gonno 16 60 2 2 48 4 5 New Town 9 53 1 13 44 10 17 Abuja 1 3 0 1 95 3 0 Gbawe 25 44 4 6 42 5 4 X2 Value 28.114 81.417 5.089 17.124 78.398 4.082 26.491 P-Value 0.000 0.000 0.165 0.001 0.000 0.253 0.000 Source: Field data, 2015 The analysis reveals that each of the settlements demonstrates mixed forms of access based on their peculiar socio-economic characteristics. Consequently, households manage uncertainty in water supply and establish themselves in the metropolis by leveraging a 102 University of Ghana http://ugspace.ug.edu.gh patchwork system of basic services that draws importantly from informal systems, practices and supplies. For clarity purposes, the percentage totals are more than 100% because respondents had the option of selecting multiple sources of water, and several households take water from both formal and informal providers. The data indicate that in terms of in- house/in-yard connections the study samples appear to have better access when compared with the community averages data. For instance, 76% of households in Korle Gonno have in- house/in-yard water supply (compared to 66.7%), followed by 69% in Gbawe (compared to 64.2%), and 62% in Accra new Town (compared to 51.9%). In the case of Abuja, only 4% were captured to have connection and they represent the category of private shower/standpipe operators. Another related but key point is that important differences exist in the percentage of households that have pipe-borne water from the utility provider, and those who purchase water from private vendors. For instance, in-yard taps and vendor standpipes are the commonly used water sources in the study area with the exception of Abuja where 95% of respondents depend on private showers/standpipes as the main available water source (p = 0.000). Although exceptional, the case of Abuja is not unexpected since it is an informal settlement where the majority of people do not enjoy water provision from the utility provider and therefore access water from private shower/standpipe operators, some of whom have installed poly-tanks (plastic storage tanks) that serve the public for some days even when supply is interrupted. Gbawe presented a unique case by recording the highest percentage of respondents (25%) who have in-house pipe connections (p = 0.000), as well as private poly-tanks. This revelation is unsurprising as Gbawe is located near the Weija Dam, one of Accra’s primary water supply sources. Again, Gbawe displays many of the characteristics commonly associated with Accra’s peri-urban settlements which is characterised by a growing private home ownership 103 University of Ghana http://ugspace.ug.edu.gh being perpetuated by the middle class who tend to acquire properties in peri-urban areas (Gough and Yankson, 2000, 2011). Such properties are often of a self-contained nature, and typically contain infrastructure capable of providing pipe-borne water from GWCL and sometimes self-made mechanized boreholes that pumped water into poly-tanks (through reverse osmosis). In relation to households who use in-yard connections, Korle Gonno recorded the highest number of respondents: 60% (p = 0.000). This is unsurprising considering that, on average, Korle Gonno's indigenous core has high density houses which tend to be populated by family- occupied households who make use of communal water sources. A key contribution to this finding is the tendency for tenant households to share the facility in the premises either through monthly billing or pay as you fetch system depending on the landlords’ predilection. A possible explanation for this preference is the combined effects of tenancy arrangements, which can be precarious (with tenants evicted at short notice if the landlord receives a more lucrative offer), and the rising cost and delays associated with acquiring separate meters (Arku et al., 2012). For those unable to access pipe-borne water directly from their homes or a neighbor, water can be sourced from vendors, tanker services and boreholes. In this regard, New Town recorded the majority of respondents (10%) using vendor water tanks while 13% and 17% reported private and vendor boreholes respectively. The findings may be explained by the mismatch between the frequent interruptions in the supply of water (a long standing problem) and the increasing demand of water that is being underpinned by the accompanying scenarios of demographics and the doubling of residential properties as commercial centres. Table 4.3 summarizes the water situation in the study area and presents the different water providers. In a nutshell, the findings on the spatial dynamics of access to water showcase the necessity of being able to connect to multiple sources of water at any given time, a possibility 104 University of Ghana http://ugspace.ug.edu.gh spurred by a network of informal vendors whose customer base is fluid and free to go to other providers. More relatedly, it stresses the interplay between formal and informal networks that are essential to water access in a complex, multi-dimensional setting such as the settlements studied in GAMA. Importantly, this recognition of the chain in water service supply is consistent with earlier studies which argue that as the delivery of water is getting more and more complex and unpredictable, residents in GAMA are forced to depend on a labyrinth of strategies (Oteng-Ababio, 2013; Peloso and Morinville, 2014; Amankwaa et al., 2014) both by increasing water security and improving their household livelihood activities. Table 4.3: Providers supplying water to households in the study area Providers/ Settlement Korle Gonno New Town Abuja Gbawe Water utility GWCL GWCL - GWCL Population (HH) 30,555 (8,599) 31,363 (8,352) 5,000 70,000 (16,624) Provision by public utilitya 66.7% 51.9% - 64.2% Provision by private vendors 6.8% 15.2% 0.1% 7.1% (standpipe or private connection)a Provision by water tankersa 0.03% 1.4% - 1.6% Notes: aPercentage of the population using this provider as a main or supplementary source. Source: Computed from 2010 PHC water utility data and HH interviews In essence, this conceptualisation raises the central issues of proximity, reliability, quality and affordability and connects to the day-to-day lived experiences of how households procure water. This is the focus of the subsequent sections. 4.1.2. Reliability and proximity dynamics of water access Respondents were asked to indicate how continuous or reliable the supply from the water sources was and the distance (return journey time) spent to procure water. Reliability or continuity of water supply is understood as an example of everyday practice, both because water is a daily basic need, and because each day, many parts of the metropolis can be cut off from the municipal system. The waterscape of households studied in GAMA is characterised 105 University of Ghana http://ugspace.ug.edu.gh by a mix of areas without water supply, and those with erratic and unreliable supply. Thus, households show discrepancies with the reliability indicator as they receive an intermittent water supply, and at fewer supply-hours. The comment below describes the situation: I have pipe in my house but for 2 months now, the taps have not been running. When it flows, it does so for a short time and then ceases, and whenever the bill comes, it is around GH¢50. Because of that I owe them [utility company]. ... What should we do to the money we borrowed to finance our work since the water is not flowing. How can I pay my lenders? We are pleading that the authorities see to it that the situation is resolved and we get back to the good old days (Participant, male elders focus group). As illustrated in Table 4.4, the study found that only 19% of the overall sampled households have daily access to pipe-borne water supply, while the majority of them (68.3%) have access to pipe-borne water supply four days per week. At the neighbourhood level, Korle Gonno had the majority of its respondents having daily access to pipe-borne water supply while Gbawe had the majority of respondents having four days’ weekly supply. Table 4.4: Percentage distribution of water flow across the settlements Settlements Daily 4 days a 2 days a Once a week week week Korle Gonno 43.4 14.3 57.6 50.0 Accra New Town 18.4 26.4 30.3 22.2 Abuja 28.9 26.7 12.1 5.6 Gbawe 9.3 32.6 0 22.2 Overall sample 19.0 68.3 8.2 4.5 Sample size (N) 76 273 33 18 Source: Field data, 2015 Here it is important to take into account that even on the scheduled days the taps rarely flow and when they do, flow is guaranteed for only a few hours mostly in the evenings or at dawn. It is thus common to find water hose connected to a tap that is ‘grudgingly’ giving out water. For example, although the majority of houses in Korle Gonno had pipe connection at the time 106 University of Ghana http://ugspace.ug.edu.gh of the study, respondents complained that water had not flowed through the taps at weekends, for several months without any explanations from GWCL. Moreover, landlords often place locks on taps and charge tenants for sourcing water in addition to their rent. They also allow neighbours to procure water for a fee. This helps landlords defray water bills, and even make profit (see Melara Arguello et al., 2013). The water situation is reflected in the comment below: With the water issue, formerly even if the pipe stops flowing in the morning, by afternoon time it will flow, but now it’s constant that the pipe will stop flowing on Friday and flow again on Sunday night. At times I wake up at night to fill my gallons. Also, formerly even if the water does not flow here we are able to get water from Korle-Bu but now, we don’t. Some doctors do not even get water. At times people carry water from houses to the wards when they give birth (Participant, female elders focus group). Similarly, during the study, participants complained that water had not flowed through the pipes in certain parts of Accra New Town (‘Zongo’ and central areas) for nearly a month despite no warning or any clear explanation from GWCL. Although this was an unusually long period of time, an intermittent water supply is the norm with water unavailable for days at a time as part of broader water rationing policies in Accra. This has context specific ramifications because the large Muslim population residing in Accra New Town needs water to perform their ablutions before prayer. The situation is summarised by a respondent in the following words: The flow is not constant. You will have no idea if the taps will be opened on Sundays or Mondays, either late evenings or at dawn. In fact, the water situation has now made ‘nobodies’ of big guys who work in big companies or preside over mosques, as you see them in the morning carrying gallons in search for water. The old, young, rich and poor are all the same when it comes to water (Participant, female elders focus group). Gbawe is located near the Weija Dam, one of Accra’s primary water sources. Accordingly, pipe-borne water supply is regular, and in contrast to other areas of the city the utility provider 107 University of Ghana http://ugspace.ug.edu.gh usually informs residents about any disruption to their water supply ahead of time. Households in the old town are more likely to make use of communal water sources, and if there is no pipe on the premises, it is likely that a neighbour will have one that can be accessed for a small fee, and in some cases for free. However, residents explained that females are primarily charged with overseeing the household’s water needs, therefore even though most vendors are nearby, it can be challenging waking up before dawn and carrying large buckets of water across the gutters, as illustrated in the comment below: …Last time I went to fetch water at night and someone had shifted the piece of wood over the gutter so I fell into it with the bucket of water on my head. Someone finally came and helped me out (Participant, female elderly focus group). In the case of Abuja, residents recounted that for three months, supply from the shower operators has been unpredictable and one had to be tactical about the time to visit the showers. Patrons are sometimes forced to use group showers (dormitory-like showers) to manage water interruptions. Also some operators have installed large poly-tanks to secure water and serve customers even when flow is interrupted. However, during prolonged periods of water shortages residents recounted walking to other parts of central Accra to procure water, albeit with time and cost implications. The comment below summarizes the water situation: We are water poor! It's like we use water in everything we do here, I mean to bath, wash, clean, work, and so on. And because most of us don't have gallons to store water we mostly rely on the showers. Even if you do, where will you keep them [gallons]? My neighbor here used to keep his containers in front of his room but they got stolen. Some now store it with the shower operators. (Participant, male youth focus group) The shower operators constitute long-time occupiers who organised themselves in close collaboration with the landlords/owners, Ghana Railway Corporation, and the local assembly to get connected to pipe-borne water. This case can be likened to the newcomers in the developing areas of Gbawe who ensure and improve their access to water by organising themselves, mobilising their resources together and collaborating with the utility provider. As 108 University of Ghana http://ugspace.ug.edu.gh noted by Andreasen and Moller-Jensen (2016) such self-help tendencies are common in the African context due to the slow pace of service delivery by the formal governance structures. With regards to proximity, measured in time spent, Figure 4.2 indicates that a quarter of households (25%) spend less than five minutes to access water while seven out of ten people (71.5%) access water within 30 minutes. Another point to be raised, though expected, is that respondents who are in the 0-5 minutes’ category appear to access water through GWCL connections and domestic sources while the majority of those in the 6-30 minutes’ category tend to access water from the vendor sources. In effect, compared with the utility provider procuring water from private vendor sources tend to increase time usage. Figure 4.2: Frequency and percentage distribution of time spent to procure water 300 70.0% 61.5% Frequency Percent 250 60.0% 246 50.0% 200 40.0% 150 30.0% 25.0% 100 100 20.0% 10.0% 50 10.0% 40 6 1.5% 8 2.0% 0 0.0% 0-5 mins 6-15 mins 16-30 mins 31 mins - 1 More than 1 hour hour Source: Field data, 2015 Another noteworthy point for reflection relates to the fact that neighbourhood configuration and topography are key benchmarks that must be reflective and measured accordingly when analysing proximity dynamics of water access. For example, on average, houses in Korle Gonno and Gbawe (Old Town) fall within 100 meters of at least a commercial water point because of the density particularly in the indigenous core, characterized by nucleated households and the relatively low lying topography. In Accra New Town access to a private vendor point fall within 200 meters and this is largely due to their less nucleated settlements 109 University of Ghana http://ugspace.ug.edu.gh and hilly topographies. In Abuja, the few shower operators are strategically positioned to provide a relatively uniform proximity to residents such that there is some form of regularity in terms of the spatial distribution of access to private vendor points. A related point worth noting is that applying settlement configuration and topography as indicators for defining proximity benchmarks in different locations of a service area of a water utility may show considerable differences between the locations. As noted by Bellaubi and Visscher (2014) in their study on water delivery in Kenya and Ghana, this may encourage the utility provider to take the necessary steps to enhance equity in service delivery. This section illustrated the everyday practices involved in procuring water for day-to-day needs that routinely lead households outside of the official water supply system. As the literature on Accra has acknowledged, rationing embodies the deficit that GWCL is facing (production against demand). Gough et al. (2015) conclude that high and low income areas experience almost the same level of unreliable water supply by the GWCL. However, households in high and middle income areas tend to cope better as they often have storage tanks to store water, compared to poorer households who mostly use barrels and small containers, notwithstanding issues of quality and what such water is used for. This is the subject of the next section. 4.1.3. Quality, quantity and usability dynamics of water access Beyond issues of coverage, regularity of supply and convenience (piped to home or close by), other concerns which require considerations include whether or not water supply is safe, sufficient to meet the per capita water demand, usable and satisfy one’s household needs. The unsatisfactory supply of water, alongside the tendency for households to use stored water during interruptions, influences the quality of water. Figure 4.3 captures respondents’ quality concerns, with the majority of them (62.5%) citing dirt and colour as pressing issues. 110 University of Ghana http://ugspace.ug.edu.gh Figure 4.3: Distribution of water quality concerns in the study area 160 37.5 Frequency Percent 40 140 150 35 120 30 25 100 25 100 19.8 80 20 79 60 15 9.5 40 8.2 10 38 20 33 5 0 0 Dirt Colour Worms Taste Odour Source: Field data, 2015 During the focus group discussions, participants were emphatic about the poor quality of pipe-borne water supplied by both GWCL and small-scale providers. It was revealed that when the pipe flows, the water is foamy in nature, has particles in it and emits an unpleasant odour. Due to frequent interruptions, residents often collect and store water during periods when the taps are flowing. However, stored water has to be used shortly after being sourced, else if left for two days or longer impurities described as ‘oily’ become visible at the bottom of the container, this occurs even if the water is covered with a lid. Such water is, therefore, primarily used for laundry and bathing because it causes stomach upsets and other water related illnesses among residents if consumed without treatment such as boiling. This dissatisfaction is buttressed by a respondent in Korle Gonno who lives in a compound house: If you open the tap, you hear some noise for some time before the water starts flowing. Some children open their mouth and drink directly under the pipe when they are thirsty. But if you collect it in a cup or bowl you will see that the water is too white and after 2 minutes some particles will settle down (Participant, female youth focus group). The quality of water supplied by mobile water tankers is also problematic as operators can cart both borehole and treated water for consumers for their domestic uses. These tankers, often manufactured from scrap metal, are rarely cleaned (usually quarterly by climbing into 111 University of Ghana http://ugspace.ug.edu.gh the tanks and using laundry detergents containing bleach) and instances of growing spirogyra around the tanks raises question marks over the quality of water in the tanks. One of the participants during the elderly focus group in Accra New Town intimated: “You can’t tell where or how the water got to your house, you only have to be wise when using it”. This confirms earlier studies (Ainuson, 2010; Obeng-Odoom, 2012a) that have found that water supplied by tankers typically gets contaminated through corrosion of the tankers or from faecal pollutants in previously-carried water. Likewise, Sarpong-Manu and Abrampah (2006) argue that the quality concerns relate to the fact that water is handed over at various points (network- tanker-vendor-client), which all pose potential points of contamination. It is unsurprising therefore that the Joint Monitoring Program (JMP) of the World Health Organization (WHO) does not consider water tankers as safe water supply source (Kariuki and Schwartz, 2005). The issues discussed above clearly have implications for the quantity of water used. The study found that households depending on alternative sources use less water than those connected to the utility provider. On the average, households consume 18-25 litres per capita per day (lpcd) which is below the 60 lpcd (27m3 per month) average consumption estimated by GWCL for low income households. The finding is consistent with Sarpong-Manu and Abrampah (2006) who found that a poor household uses about 180 litres per day, which would only be about 22.5 lpcd, assuming a household size of 8 persons. For middle and high income areas GWCL estimates 90 lpcd (32.4m3 per month) and 120 lpcd (28.8m3 per month) respectively. When the data is further analysed to account for the differences among the study settlements with respect to their daily water consumption, Korle Gonno uses 19.8 lpcd, whereas Gbawe uses 19 lpcd, followed by Accra New Town with 17.2 lpcd and, finally, Abuja with 14.6 lpcd. Per the WHO classifications of access to water supply (less than 20 lpcd as ‘no access’, 20 - 112 University of Ghana http://ugspace.ug.edu.gh <50 lpcd as ‘basic/minimum access’, 50 - 70 lpcd as ‘intermediate access’ and 100 lpcd as ‘high/optimal access’), the study settlements, on average, only have basic access. A related point for reflection is that the consumption per household found in the study area is lower than the average consumption estimated by GWCL. This seems to suggest that the actual number of users of yard connections is larger than the number estimated by the utility, but it may also imply that more water is lost (non-revenue water) before it reaches the users. Mindful of the fact that households have highlighted numerous water sources, of different quality, and the possibility that these sources are deemed to serve different purposes, the study further analysed the primary domestic uses of water. As illustrated in Table 4.5, availability of in-house or in-yard connections per se does not seem to drive drinking water decisions considering that nearly 74% of respondents consume sachet water as their preferred drinking water source. The rest drink pipe-borne water only after it has been treated (boiled and refrigerated). Thus, juxtaposed by the 41% of overall sample having pipe-borne water supply, it can be inferred that socioeconomic status and municipal supply do not determine sachet water consumption at the metropolitan scale. These findings contradict that by Stoler et al. (2015), which identified sachet water consumption as a proxy for higher disposable income and lack of knowledge about household water treatment methods. Another noteworthy point is that the 74% of households using sachet water as their primary drinking water source is significantly higher (2.5 times more) than the 28% reported for the metropolis by the GSS (2014b), pointing to possible different research questions posed and target household population and the associated reliability problems of data sources. This case also indicates that there has been an increasing trend in the consumption of sachet water over the years. In a related study by Stoler et al. (2012), 50% of households in a sample of Accra’s slum neighbourhoods reported using sachets as their drinking water source. The unifying 113 University of Ghana http://ugspace.ug.edu.gh thread linking this present study and Stoler et al.’s is how sachet consumption in Accra has transitioned from higher to lower income populations. Table 4.5: Percentage distribution of the domestic uses of water sources Activities Sachet Vendor In-yard In-house Vendor P r i v a te Vendor Private water standpipe connection connection tank borehole borehole tank Drinking 73.8 22.3 1.4 1.8 0.5 0.2 0 0 Cooking* 0 54 32.0 8.0 3 1 1 1 Washing 0.0 59.8 27.0 7.5 3.0 1.2 1.0 0.5 Bathing 0.0 60.0 27.0 7.4 3.3 1.3 1.0 0.0 Cleaning 0.0 60.0 26.8 7.5 3.0 1.2 1.0 0.5 Source: Field data, 2015. Notes: * 58 cases who do not cook are excluded. Cleaning as used here refers to other forms of clean-up activities apart from washing. Typically, an average household of five will use 3-5 large bags of sachet water (containing 30 sachets) per week, and the median number of sachets consumed per day is 5. The top reasons cited by respondents for consuming sachet water were varied. They included; better quality (42.2%), convenience (21.2%), because they were chilled (20.8%), preferred (9.8%), and because there was no other option (6%). These factors relate more to social processes and attitudes toward water quality. The interview data revealed that a lot of people perceived the consumption of sachet water as a symbol of modernity. Sachets extend drinking water coverage deeper into low-income areas and alleviate the need for safe water storage, potentially introducing a health benefit over stored tap water (Stoler et al., 2012; Amankwaa, 2017). Quite expectedly apart from drinking, sachet water was not used as a major source for domestic activities although respondents mentioned that they occasionally used it for cooking and bathing during severe water shortage. In terms of domestic activities at the household level such as cooking, washing of clothes, bathing and cleaning, the analysis displayed a fairly equal distribution among the various sources of water. In this regard, there seems to be more reliance on vendor operated standpipes, and this is hardly surprising since it is the commonly 114 University of Ghana http://ugspace.ug.edu.gh accessed source by the majority of respondents. It was observed that 54% of respondents use standpipes for their cooking, with 60% reporting using it as their source for washing clothes, and bathing. This is followed by households which have in-yard connections or source water from a neighbour who has one. The analysis showed that 32% of respondents use the pipe-borne water for cooking, with 27% reporting it either as their washing or bathing source, and 26.8% using it for other household cleaning purposes. Similarly, households which have their own in-house pipe have 8% of the respondents using it for cooking, with 7.5% reporting using it either as their source for washing clothes or cleaning, and 7.4% using it as their bathing source. Also, sources such as separately owned and vendor water tanks and boreholes were reported to be used at varying degrees as alternative water sources for domestic activities during periods of water shortages. Rainwater is also collected and used for bathing and laundry. The foregoing analysis shows that households concerns regarding the quality of water supply range from physical, chemical to bacteriological. A key point for reflection is the fact that a combination of factors account for these concerns. These include, on the part of the utility provider, issues relating to old and broken pipelines, and dry taps owing to the long periods of lack of water flowing through them, as well as household management practices such as storage media and duration, and fetching container, on the side of residents. Consequently, consumption of sachet water is considered more preferable as a source of drinking water. While the sachet water business in effect extends improved water access deeper into low- income and informal settlements, and eases the need in those places for a method of safely storing drinking water, it equally diverts a chunk of water from the utility network, often restricting access of those further down the network. Also there is a growing controversy over the inadequate management of the new waste streams created by plastic sachets, all of which 115 University of Ghana http://ugspace.ug.edu.gh makes extensive multi-disciplinary research and targeted policies in relation to the increasing trend of sachet water use crucial. Problematically, the rising reliance on water supply from private vendors combined with sachet drinking water consumption means that households are incurring a double burden in terms of water cost, which is what the next section examines. 4.1.4. Affordability dynamics of water access The analysis showed wide variation in the monthly cost of water accessed from public and private sources. Table 4.6 indicates that monthly cost of water increases as households procure water from vendors rather than having a GWCL supply. Those who depend on vendor sources on the average pay GH¢69 per month. When compared with having GWCL connection, it is clear that the price that is paid for water if it is bought from vendors is nearly double the GWCL tariff, at GH¢35. With the majority (54%) of the overall sample having no household connection and completely depending on vendor sources, it can be concluded that poor households are paying more than their counterparts receiving water from GWCL (see Van Rooijen et al., 2008; Amankwaa et al., 2014). A t-test comparing means statistic was conducted to ascertain whether or not there is a significant difference in the type of water source and monthly cost of water. The analysis revealed that there is a significant difference between the monthly cost of water procured from the public and private vendor sources, at a 99% confidence interval. Those who depend on private vendors for their domestic water incur a monthly cost which is twice as much as those who obtain water from the utility provider. Table 4.6: T-test comparing means of monthly cost of using public and private water sources Category Observation x ± Std. Dev t-value P-value Public source 181 34.92 ± 26.32 Private vendor sources 219 69.43 ± 51.80 -5.78 0.00 Mean difference 34.51 ± 4.24 Field data, 2015 116 University of Ghana http://ugspace.ug.edu.gh Similarly, it can be argued that the presence or absence of water infrastructure and supply influence the vending prices of water. Table 4.7 illustrates the prices of water in the study area and compares it to the official GWCL tariff. It can be observed that consumers either pay a flat rate or according to an increasing block tariff. For example, a compound house connected to the utility network with 16 people each using 50 litres per capita per day (lpcd), will consume more than 20 m3 per month and will therefore have to pay according to the higher tariff. Table 4.7: Comparison of water utility and vendor prices in GAMA Tariffs category Unit volume (litres) Unit price (GHp) Unit volume Comparison (GHp/m3) GWCL residential 0-20m3 10006 178.33 178.33 (GH¢ 1.78) >20 m3 1000 267.33 267.33 (GH¢ 2.67) Unmetered premises Flat rate per house monthly 1160.70 1160.70 (GH¢11.60) Commercial/industrial Flat rate 380.00 380.00 (GH¢3.80) Kufuor gallon Korle Gonno 20 40 2000 (GH¢20.00) Accra New Town 20 40 2000 (GH¢20.00) Abuja 20 50 2500 (GH¢25.00) Gbawe 20 20 1000 (GH¢10.00) Sachet water Single 0.5 20 40,000 (GH¢400) Bulk (x30) 15 300 20,000 (GH¢200) Note: Sachet producers are supposed to pay a commercial rate. Unit price is in Ghana pesewas Source: Field data, 2015 and PURC water tariff effective July 2015 The analysis becomes revealing in the study area as it is clear that a 20 litre gallon of water if it is bought from private vendors’ costs 5.6 times more than that from a formal utility in 6 Prior to Dec 2015, 1000 litres of water was GH¢1.78. Newly approved tariffs for same quantity goes for GHC 2.98 117 University of Ghana http://ugspace.ug.edu.gh Gbawe, at GH¢1.78/ m3, compared to up to 11.2 times more than the utility price in Korle Gonno and Accra New Town and, finally, up to 14 times more than the utility price in Abuja. In times of scarcity, which is a couple of times every month, the price in Abuja, for instance, goes up to more than 20 times the official price for domestic users. This is illustrated in the quote below: When the tap goes out we really suffer because they double or triple the money you pay for water. For example, the [Kufuor] gallon that we pay 40 pesewas is more than doubled so we pay GH¢1.00. I beg the woman that I have a lot of children so she allows me to pay GH¢0.70. In a day I fetch water at the cost of more than GH¢2.50 pesewas (Participant, female youth focus group, Accra New Town) In essence, prices vary considerably among the study settlements such that places where there is pipe infrastructure and getting a vendor standpipe is easier, competition amongst vendors is higher thereby influencing the price. In Abuja, however, without public water infrastructure, a handful of people can access tap water even to commercialise it, thus pushing up the water prices. An interesting observation made by Van Rooijen et al. (2008) which resonates well with this present study is that water prices per m3 in Sukura and Old Fadama were found to be about 3 and 6 GH¢/m3 respectively, which is 4-9 times the price of pipe- borne water supply. A possible explanation to the difference in price is that Sukura is a formal low income settlement with some pipe infrastructure, while Old Fadama is an informal settlement where connections to the utility are officially not permitted. Cost associated with drinking sachet water also needs to be emphasized as it constitutes an additional burden on household expenditure. Usually purchased from local vendors’, sachet water comes in two sizes; small 500ml bags that can be purchased individually as and when needed for 20 pesewas and a larger bag containing 30 sachets that costs between GH¢3.00- 3.50. Although Table 4.7 clearly indicates that the former choice is more expensive (double the cost), respondents proffered some reasons for explaining why they choose the sachets over the bags, and this has been captured in the comment below: 118 University of Ghana http://ugspace.ug.edu.gh We buy the single sachets because some of us here don't have our own fridges to chill water and you know it is difficult to drink it [water] hot. In my case because we are a group of friends who share [sleep in] this single room it's difficult to monitor who drinks when and who buys next. For others, it’s about the cost of the full sachet bag or because they don’t want their children to misuse it (Participant, male youth focus group, Abuja) The water vendors themselves are often members of the community who have access to pipe- borne water and who see this as an opportunity to commercialise it. There is a monopoly amongst vendors and price is agreed upon by them alone. While the burgeoning water business represents a considerable source of household income, the lax regulation allows vendors to exploit the situation – as they can create artificial water scarcity conditions by hoarding water to be able to charge higher prices. The irony of these observations is that, vendors who are connected to the GWCL network and sell to residents receive water through their pipes while individual households with the same pipe connection in the vicinity do not get water. Such developments epitomise the elite capture of resources by local elites, mostly the affluent, resourceful and powerful, which entrenches power dynamics and further marginalizes vulnerable residents. The implications of this micro politics are clear and numerous; some vendors have to afford the services of small-scale water tankers to stay competitive. Also, there is the potential of a household being disconnected when a landlord fails to pay accumulated water bills. Likewise, large households spend much of their time and income on accessing water and those who have busy work schedules have no option than to contract middle-persons at a fee to collect water from neighbouring communities. In essence, although it is more cost effective to source water from a metered pipe, a large number of people living in GAMA are unable to do so, and thus depend on alternative water service providers. Ironically, the majority of these providers either directly or indirectly 119 University of Ghana http://ugspace.ug.edu.gh depend on GWCL as their source of water. Accordingly, three main types of vendors have been identified. First, community vendors who sell water from 1,000-5,000 gallons or 3.8- 18.9 m3 storage tanks. They buy water from tanker operators and sell directly to the public at GH¢0.20 per 10 litre bucket and GH¢0.40-0.50 per gallon (20-25 litres). Typically, a family of five uses 5-8 buckets per day, making 5 to 8 trips with gallons and can spend between GH¢1.50 – 2 daily and GH¢3 on weekends. But in times of severe shortages the same quantity sells for GH¢1.00. Second, neighbourhood (household) vendors who have pipe connections and sell water between GH¢0.40-0.50 per gallon directly from the yard taps to raise money to pay their monthly bills which is between GH¢60 and GH¢120. In most instances, the facility is owned and controlled by the landlord, and household members together with other members of the neighbourhood access it at a fee. The vending point is only accessible during certain hours when people often have to queue for long periods of time. GWCL does not officially recognise these vendors as they partly account for the inability of some residents to get water supply. Third, the tanker operators supply water to households directly and to other intermediary providers from 1,200 to 4,500 gallons (4.5 to 17.0 m3) tanker trucks. Typically, low income households buy 1,200 to 3,000 gallons while middle-high income households buy 3,000 to 4,500 gallons. Prices paid for tanker services vary across the income strata, with low income households paying about 40 GH¢/m3, and middle and high income households paying 30 GH¢/m3 and 35 GH¢/m3 respectively. The difference is because, in addition to smaller household sizes, high income households have the means to procure water in large quantities than their poor counterparts. Tanker operators have formed associations, that have formal connections with GWCL, and their prices are determined through consultations between 120 University of Ghana http://ugspace.ug.edu.gh PURC. In 2008, PURC launched its ‘Tanker services guidelines’ in an attempt to regulate and ensure safe and affordable supply of water. In summary, whilst government expects that subsidies reach deserving people, the question remains whether a non-targeted subvention like the lower water price does reach the right people, the poor, (Komives et al., 2005), under the present conditions of shared system, unsatisfactory pipe-borne water supply, and reliance on vendors, a situation that exposes the inefficiencies of the decentralized governance system of the city. Also, the discrimination, and politics from below, in water distribution in favour of some parts of the community and high water bills and vendor prices raise key concerns of access that need to be addressed. Ultimately, taking into consideration the links between local conditions and prices, provide useful baseline information which makes a future study of factors influencing vendors’ prices by geographic location and understanding how these factors contribute toward such pricing crucial. 4.2. Dynamics of electricity access 4.2.1. Spatial dynamics of electricity access Electricity coverage is also good in GAMA (see Table 4.8). The 2010 Population and Housing Census (PHC) results show that out of 501,903 households in the Metropolis access to electricity stood at 93.8% whilst the remaining was distributed among different sources including flashlight/torch, kerosene lamps, candles, private generators, solar and gas. National and regional access stood at 72% and 87.1% respectively, although since 2010 the national access rate has increased to 80.5% in 2016 (The Presidency of Ghana, 2016). 121 University of Ghana http://ugspace.ug.edu.gh Table 4.8: Spatiality of electricity coverage at national, regional, metropolitan and locality levels Selected settlements in GAMA Source of National Regional Metropolitan Korle Accra Abuja Gbawe electricity Gonno New Town Total 5,467,054 1,036,370 501,903 8,599 8,352 - 16,624 Electricity (mains) 3,511,065 902,831 470,915 8,001 7,994 - 13,147 Sampled total (%) 72 87.1 93.8 93 95.7 - 79.1 Source: Computed from Ghana Statistical Service (GSS), 2014b Access to electricity in GAMA is better appreciated if viewed from the perspective of intra- city and class dynamics. The analysis presents a fairly good picture of access to electricity among the residential categories in the metropolis - indigenous (Korle Gonno) and low- middle class (Accra New Town). This suggests there is little disparity in the access to electricity among the social and economic classes in the metropolis. The relatively lower rate of access in Gbawe is attributable to the shared system that dominates households in the old- town, coupled with the growing developments in the new site that are yet to be connected to the national grid. Over the years, connection to electricity in Gbawe has been gradual; with older residents explaining that when the population was smaller and farming dominated, the settlement suffered from a lack of electricity infrastructure. The shift from agricultural to residential area instrumentally underpins changing availability and access to electricity. In Abuja, it is striking to note that in the absence of public provision, residents have managed to extend connections from the city network to the settlements through a self-help initiative. Here, the longtime occupiers, in close collaboration with the local assembly, came together and levied themselves to buy electric poles and cables to connect electricity to the area and into their homes. Anybody joining the system later is required to pay a certain amount (deemed to be used to buy electric cables) for future maintenance and upgrades when necessary, especially in the event of a fire outbreak, which is common in the settlement. Recent development suggests that opinion leaders in Abuja are taking steps to erect a new 122 University of Ghana http://ugspace.ug.edu.gh transmission pole to tackle the issue of overloading. This initiative follows the recent fire incidence in 2015 and the caution by the local authority regarding the reconnection of the settlement to electricity. Such self-help approach to improving residents’ access to basic services has been reported elsewhere (see Andreasen and Møller-Jensen, 2016; Daniel, 2014). However, the high access to electricity in the metropolis and the local cooperative efforts implemented to improve access do not equate to equitable access among residents (Gough et al., 2015). A closer look at the study area suggests that households get connected to the electricity grid in a fashion similar to that of water described above. Figure 4.4 illustrates the electricity supply situation. It categorises the different modes of electricity connections available at the household level. The results show important differences in how households actually connect to electricity from the utility provider. Although there was fairly good access (in terms of coverage) across different socio-economic classes, household level access exhibited disparities such that overall 54.5% of respondents access electricity via the shared house metered connection compared to 35% having their own separate household connection. Also, a handful of the respondents, representing 6.5%, tap power from a neighbor outside their residence, albeit in some cases illegally, while the remaining 4% have no connection at all, and as a result resort to alternative sources of lighting including private generator, flashlight/torch, kerosene lamps and candles. 123 University of Ghana http://ugspace.ug.edu.gh Figure 4.4: Distribution of household electricity connection across the settlements 250 Household metered connection House metered connection Outside house connection No connection 218 200 150 140 100 74 67 54 6050 20 2632 30 161 2 24 2 0 17 9 3 5 0 Korle Gonno Accra New Abuja Gbawe Overall study Town sample Source: Field data, 2015 With regards to the settlement dynamics it was only in Korle Gonno, a low-income indigenous settlement, where a large share of households, constituting 67%, had separate meter connections, whereas in the other settlements the majority of households accessed electricity via the shared system - Accra New Town, 74%, Gbawe, 60%, and Abuja, 54%. The peculiarity of Korle Gonno is expected, and the common reason being that residential properties have often passed through multiple maternal lineages with long-time owner- occupiers and as such there is some form of cooperation and balanced autonomy among households, although senior family members often oversee properties and manage available services. Unlike Korle Gonno, Accra New Town typifies a migrant-mixed population, where multi- storey commercial buildings predominate. The settlement has a relatively good power supply compared with other parts of Accra which is a key reason why a printing press industry heavily reliant on electronically high voltage powered machinery has emerged in Accra New Town in the past decade, with companies relocating from other parts of the city. With the rising commercial centres one would expect to see more separate connections, however, most premises tend to access electricity from an already existing meter in the establishment or a 124 University of Ghana http://ugspace.ug.edu.gh nearby source. The focus group discussions further revealed that the practice where outstanding debt is left on tenant-acquired meters on their time of vacating the premises has become rampant, as such, landlords have become less amenable to encourage the acquisition of separate meters, as they have to either settle the debt themselves or clandestinely shift it on to the new occupant, or at worse disconnect such meters. The 60% of households connecting to electricity via the shared system in Gbawe is quite surprising. However, a possible explanation is the predominance of households in the indigenous core who share meters. This is coupled with the heterogeneity of the built environment, characterised by a mix of household frontage shops and the growing commercialisation of properties along the main road which very often connect to a meter available in the house, or a nearby source. Although a significant share of households (17%) in Abuja, mainly the longtime owner occupiers and the local elites, enjoy separate metered connection the majority of the respondents’ (54%) access electricity via the shared system. This is hardly surprising considering that, first, tenure is insecure and secondly the majority of residents are migrants who are struggling to make ends meet. Such a status and orientation leave little motivation for them to secure private meters. Another key point is that there is some form of local influence/politics by the first-time settlers through the self-help initiative, and, in subtle ways, their power determines one’s ability to directly connect to electricity from the mains or not. Such micro politics have consigned a significant majority of residents to the shared system, a clear scenario of how strategies to access services are being created and perpetuated to the advantage of the resourceful and local elites (see Amankwaa, 2016). In addition, a significant proportion (20%) also connects, in most cases illegally, from a neighbour. Such outside connections are widespread and may not be termed illegal per se as there is a recognised arrangement with the provider and they contribute to the payment of the 125 University of Ghana http://ugspace.ug.edu.gh electricity use (Scott and Seth, 2013). The remaining 9% who do not have any form of access to electricity depend on alternative sources mainly candle; this practice is worrying as candles use has been reported to be the main source of incessant fire outbreaks in the settlement. One underlying cause of the increasing trend in access to electricity via shared meter connections is the rising number of tenant households in Accra. The result is a complex interplay of power dynamics and implications for their access to services. This is further perpetuated by the autonomy of landlords to allow a private meter installation in their residence or not, and economic disincentive on the part of tenants (to install separate meters) in the wake of precarious tenancy arrangements (short notice evictions at the landlord’s discretion). A tenant in Accra New Town puts it this way: These days it’s not easy to get a room and your money is small to buy a meter. Even if you have, who are you to tell the landlord you want to go for your own meter, or ask him to show you the bill before paying? You must have secured a new room to do that because he will eject you with the least chance. (Participant, male youth focus group). Table 4.9 summarizes the electricity situation in the study area and presents the different providers and lighting sources. In summary, households’ connections to the electricity grid are in a fashion related to the earlier account on water except that connecting to electricity is slightly more expensive. Consequently, illegal connection of electricity is quite common particularly in Accra new Town and Abuja. It must be added however that, the illegality is excused by some homeowners with installed meters who, on grounds of good- neighbourliness, permit others to tap from their meters and then contribute towards the payment of the bill. This collaborative way of service provision makes it possible for people to get better access to electricity service in GAMA, and this provides clear evidence of private co-production in the study area, as reported elsewhere (Fuseini and Kemp, 2016). 126 University of Ghana http://ugspace.ug.edu.gh Table 4.9: Main providers/sources of power supply in the study area Providers/ sources Korle Gonno New Town Abuja Gbawe Electricity utility ECG ECG - ECG Population (HH) 30,555 (8,599) 31,363 (8,352) 5,000 70,000 (16,624) Provision by electricity utilitya 93% 95.7% - 79.1% Provision by kerosene lampa 2.5% 1.6% - 8.9% Provision by candlea 1.8% 0.7% 3.1% Provision by flashlight/torcha 1.7% 0.9% 6.7% Provision by private generatora 0.3% 0.5% - 0.7% Notes: aPercentage of the population accessing power as a main or supplementary source. Source: Computed from 2010 Population and Housing Census (electricity utility data). The high coverage of electricity in the metropolis and the local cooperative efforts implemented to improve access do not suggest an efficient electricity service delivery. The spatial dynamics of access to electricity demonstrate the predominance of shared meter connection in GAMA. Here, under the supervision of the landlords, households share the facility through the monthly billing system or the pay as you go system depending on the meter type, either post-paid or pre-paid. The type of meter connection is thus at the core of households’ concerns regarding quality of power supply – the focus of the next section. 4.2.2. Reliability and quality dynamics of electricity access Reflecting the national power supply situation, electricity delivery in the study area is subject to what is colloquially known as ‘light off’ i.e. intermittent power outages as part of the broader energy supply management at the city level. Access to electricity is characterized by intermittent supply due to high demand amidst low generation capacity and technical challenges triggered by inadequate investment and maintenance of the power plants which are typical of Ghana’s energy supply systems (World Bank, 2015b). The most pressing issue concerning electricity in GAMA is the introduction of prepaid meters. The new system replaces the old post pay monthly billing system. Under the previous 127 University of Ghana http://ugspace.ug.edu.gh system, a monthly bill is sent to the property and the occupants divide the costs among themselves. The amount each person pays relates to the electrical items they use (i.e. those with more paid more). A notable problem with this system was that regardless of service quality, customers were always charged. Yet the monthly billing cycle was popular, as it provided a sense of order because provided the bill was paid, the company would supply power. In contrast, the prepaid system requires consumers to preload a card or remote with credit in order to access electricity. Recently, SMS meters have been introduced where text messages regarding credit use and top-ups are sent to users’ mobile phones. The main problem with the pre-paid system is that it is more difficult for families and home-based enterprises to budget, as they are unable to predict when their credit will finish. In a context where residents live in compound houses, they either contribute together to top up credit or households’ top-up in turns. However, disputes arising due to the lack of suitable formula to distribute the cost or regulate overconsumption and who is next to top-up a prepaid credit, and with how much, when the credit finishes and power is suddenly cut off make such arrangement unsustainable in the long-run among the sharing households. This is further worsened by the financial unpreparedness of certain households to make payment for the next top-up and aspects of who bears responsibility for traveling to top-up and fixing faulty meters. This situation is illustrated by a tenant who lives in a compound house that connects power from a nearby house: This prepaid is causing quarrels, some have even ended up in the court. Tenants feel cheated because others use more credit and sometimes too landlords do not buy the right amount of credit. …You have to roam about before you get credit to buy. I for instance think buying prepaid credit from elsewhere lasts longer than buying from the vendor located in the vicinity. So when it’s my turn, I don’t buy until I go to town. (Participant, male elderly focus group, Accra New Town). 128 University of Ghana http://ugspace.ug.edu.gh Figure 4.5 captures the distribution of pre-paid and post-paid meters among households. Overall, there is a very high penetration of prepaid metered connection of up to 78.5% in the study settlements in GAMA. A chi-square statistic was conducted to determine whether or not there are significant differences in the number of people who use different types of meter in the different study localities. The analysis revealed that there are significant differences in the use of prepaid and post-paid meters in the four settlements at a 99% confidence interval (see Appendix 4). This is a reflection of the successful transition process from the monthly metered billing system of electricity access to the prepaid metered system as part of the broader attempt by government to ensure efficiency and quality energy service delivery in the country. Reliability was measured by the duration of electricity supply (or power outages). On average, the results show that more than two-thirds (88%) of the overall sample receive electricity supply half of the day (12 hours), while 9% receive 6 hours of power supply, followed by 2% receiving 18 hours, and finally, only 1% experience 100% constant supply of power. In other words, on average an outage lasts for 12 hours on days that households experience power interruptions. This finding ties in with the recent load shedding exercise by the utility provider where households often receive power supply half of the day, alternating between day and night time (6am-6pm or 6pm-6am). 129 University of Ghana http://ugspace.ug.edu.gh Figure 4.5: Distribution of household connection to pre-paid and post-paid meters Prepaid meter Post-paid meter 350 314 300 250 200 150 82 92 100 7664 70 32 50 16 20 2 0 Korle Gonno Accra New Abuja Gbawe Overall study Town sample Note: The remaining 16 cases of the overall sample constitute those without connection Source: Field data, 2015 The time dimension of the load shedding exercise further relates to the broader power dynamics experienced at the community level such that certain places receive more hours of electricity supply in a day than others. Although these irregularities were mainly highlighted by residents, authorities on the other hand discounted such claims indicating that the disparities could be explained by the voltage (kilowatt hours) consumption of a particular block or cluster, particularly if there is a health care or social centre and a large commercial establishment. As intimated by Scott and Seth (2013), a political economy analysis in electricity supply offers the potential for understanding how governance might be strengthened to improve efficiency and equitability of services. Importantly, the time dynamics have implications for household businesses in the metropolis (discussed further in the next chapter). Delving deeper into the settlement specifics, Figure 4.6 illustrates that the majority of households (91%) in Gbawe receive electricity supply half of the day (i.e. 50%), followed by 88% each in Accra New Town and Abuja, and 85% in Korle Gonno. For the least number of hours of power supply of between an hour and 6 hours, 11% each of the sample were recorded 130 University of Ghana http://ugspace.ug.edu.gh in Korle Gonno and Abuja, while 9% and 7% were recorded in Accra New Town and Gbawe respectively. Only 4% of households in Korle Gonno receive 100% constant supply of electricity, and this could be households having a double-face meter connection. Figure 4.6: Daily duration of household power supply across the settlements 100% 4 3 1 2 90% 80% 70% 60% 85 88 8850% 91 Full 40% Half 30% Quarter 20% 10% 11 9 11 7 0% Korle Gonno Accra New Abuja Gbawe Town Source: Field data, 2015 The analysis revealed that in a week, power outages are experienced without warning on four to five occasions and this is alternated between day and night time depending on the load shedding schedule. Cumulatively, this could equate to between 8 days and 16 days of power outages in a month. Even when power is available there can be as many as 3 to 6 blackouts in a day at varying lengths of time. Such sudden power cuts (on and off) without prior notice are worrying as they can cause damages to plugged in electrical items that are frequently used by households such as fridges or TVs, and may result in fires - in one such case four children were killed in Accra New Town, with several child deaths recorded in Abuja. One respondent in Abuja related well with the impact of frequent power outages and commented: The dumsor [light off] is like disco lights. If you don’t put the switches and every appliance off when they start the on and off, and you try to be switching them off as and when they put the lights off, you are dead because your bulbs and appliances can get faulty when the light comes with high current (Participant, elderly focus group). 131 Percentages (%) University of Ghana http://ugspace.ug.edu.gh With the power cuts, households can no longer practice weekly cooking as food items stored in fridges and freezers go bad, and resulting in additional expenditure on food. This situation is recounted by a tenant who could not hide his experience: Last I quarreled with my wife because she left the stew and stuff in the fridge to go bad. She was supposed to be heating them every morning. There is no money, so when these little issues come up they add to your stress and you become fed up (Participant, elderly focus group, Accra New Town). The reliability issues discussed above are further worsened by quality of service associated with the introduction of the prepaid system. Quality was measured by the stability of current or the appropriate voltage supplied. The analysis shows that 50.3% of respondents rated the electricity current to be of acceptable quality while 49.7% rated it to be of unacceptable quality. Despite the seemingly good statistics, the qualitative data provide useful insights regarding residents’ dissatisfaction with quality of service, as illustrated in the comments below by a participant during the male youth focus group in Gbawe: In the night the current gets so low that your TV and fridge can go off. They say it’s the corn mill shop nearby but we don’t believe it since other places also complain. It even destroyed my sister’s laptop charger. …And the prepaid when your credit is exhausted you run into a negative balance. They say some of us still owe on the old meter. But I am wondering if they are not done deducting. For me, if they ask us to demonstrate against the new meter I will do it. However, stakeholders, particularly the ECG officials, appear to have different and somewhat opposing perspectives on the adoption of prepaid meters for the city. This is reflected in the comments below: The pre-payment metering system is the surest way to bring the services of the company [ECG] to the doorsteps of customers, while at the same time improving revenue collection for the company. (ECG official in charge of public relations, Gbawe) 132 University of Ghana http://ugspace.ug.edu.gh The introduction of the prepaid meter, though having problems in the initial stages, people will soon get abreast with how it operates. …It will possibly ensure that no bills are left by tenants for landlords when tenants are ejected or they leave the premises. (ECG officer in charge of the pre-paid meter faults, Korle Gonno) While government officials expressed optimism about the success of the switchover process of the pre-paid meters, it stands to reason that the main focus and intentions of the proposed system is the financial benefits. The new system facilitates cost recovery and eases bill distribution by workers as it pertains to the old system. Commensurate efforts with respect to public education and effective user friendly platforms must accompany the new system. Significantly, the variability in the level of access combined with reliability and quality concerns play out differently to impact affordability. Respondents were unanimous in perceiving that prepaid meters read faster and consume more current as compared to post- paid meters. The cost implications and related usability issues are the focus of the remainder of the chapter. 4.2.3. Affordability and usability dynamics of electricity access Households were asked to state their average monthly expenditure on electricity. We checked the monthly bills of post-paid customers and receipts of pre-paid users to verify the genuineness of the figures they stated. As shown in Table 4.10, the mean monthly cost of electricity of households who use pre-paid is GH¢86 while those with post-paid meters pay GH¢57. In effect, customers using pre-paid meters pay on average GH¢29 more on their electricity bills than direct debit customers. The minimum cost was GHȼ5 and the maximum cost was GHȼ 350. A t-test comparing means statistic was conducted to determine whether or not there is a significant difference in the type of meter used and monthly cost of electricity. The analysis revealed that there is a significant difference between the monthly cost of electricity accessed through the prepaid and post-paid meter types at a 99% confidence interval. Implying that those who depend on prepaid meters for their domestic power incur a 133 University of Ghana http://ugspace.ug.edu.gh monthly cost which is more (nearly one-third) than those who obtain power from the post- paid meters. Table 4.10: T-test comparing means of monthly cost of using prepaid and post-paid meters Category Observation x ± Std. Dev t-value P-value Prepaid 314 85.99± 191.51 Post paid 70 57.42± 66.83 2.19 0.01 Mean difference 28.57 ± 12.99 Source: Field data, 2015 Although initially preloading a card with GH¢10 could allow a small household to access electricity for five days, currently the same amount cannot guarantee 3 days’ power supply. For instance, the average household in Accra New Town spends between GH¢70 and GH¢120 a month on electricity and this is far more than they used to pay under the previous system, GH¢35-60 with more flexible instalment payments. Furthermore, the main concern regarding the rise in costs of electricity in turn has an impact on availability and differential access among households. For instance, in the indigenous core areas of the study settlements, the financial inability of some residents to reload the card explains why they do not enjoy continuous power supply and this has occasioned the emerging practice where some households and frontage shops tap power illegally from their neighbors. Likewise, those in the remaining parts of the settlements are sometimes unable to access power due to the unavailability of the prepaid credit vendors during certain periods of the day (especially evenings and on weekends). This is particularly worrying for the middle- class residents who leave home early in the mornings for work and return late in the evenings. The situation has been highlighted in the quote below: I came home one day and realised that they had changed our meter to pre-paid, and disconnected it because of the debt on the old meter. Now we tap power from our neighbor. We are 5 in all who share the prepaid meter. We don’t know what appliances people use. If we take GH¢5 from each person and buy GH¢25, it’s not able to even 134 University of Ghana http://ugspace.ug.edu.gh last 1 week but before it could last for nearly 2 weeks. So I went to the prepaid vendor and I called the ECG for answers. They told me when you buy the credit they deduct GH¢3 as charges. (Participant, female elderly focus group, Accra New Town). Alongside the rising cost, the reliance on alternative sources of power to make up for the frequent power outages also dominates concerns over electricity (Komives et al., 2005). The key alternatives used by households include: flashlight/torch, kerosene lamps, candles, and private generators. Table 4.11 presents the additional cost associated with the alternative sources of power. The findings show that on average households spend GH¢40 as monthly cost on alternative sources. By comparing with the regular cost of electricity from ECG, households who use the prepaid meters together spend GH¢126 monthly while those who are connected to the post-paid system spend GH¢97 monthly. Thus, the impact of power outages is that households (with pre-paid meters) spend nearly 50% more to access alternative sources of power. The minimum cost was GHȼ 2.50 and the maximum cost was GHȼ 2000. Table 4.11: Monthly cost (GHȼ) of alternative power sources across settlements Settlement Minimum Maximum Mean Std. Deviation Korle Gonno 2.50 200.00 13.1136 27.13933 Accra New Town 3.00 360.00 31.5698 60.14901 Abuja 5.00 240.00 24.5121 44.45626 Gbawe 3.00 2000.00 79.9803 363.36251 Overall sample 2.50 2000.00 39.61 201.66 Source: Field data, 2015 A related point worth noting is that the increasing use of these strategies also illustrates the character and complexity of the metropolis’ neighbourhood differences. For instance, households in low income and informal settlements (Korle Gonno and Abuja) generally rely more on less expensive rechargeable lamps, battery torches and candles, a possible explanation for the relatively low monthly mean cost of GHȼ13 recorded for Korle Gonno and GHȼ24 for Abuja. Contrarily, in Accra New Town and Gbawe where the middle class dominates it has been observed that households tend to spend on average GHȼ31 and GHȼ79 135 University of Ghana http://ugspace.ug.edu.gh (increasing to as high as GHȼ360 and GHȼ2000) respectively on alternative sources. A likely reason for this is the significant portion of households who are able to afford private generators that come with expensive fuelling cost. Thus, alternative sources show more promise, arguably, in making power accessible to poor households, yet they add to the cost pressures in managing access. Another point for reflection relates to the fact that, low-income earning jobs that mostly characterise low-income households complicate the cost dynamics of access to power in the metropolis. Problematically, people who live from hand to mouth become vulnerable as they sometimes have to choose between spending the little income on food or paying for water and electricity services. Specifically, the poor class complained of spending more of their (daily) income accessing electricity, almost on a daily basis. This in part accounts for the illegal tapping of power. Tapping, which usually occurs at night hampers access and availability, and in some cases affects quality/stability of the power (Kariuki and Schwartz, 2005). Closely related to electricity cost is the number of kilowatt hours (KWh) of electricity consumed by households. The analysis revealed that only 12 of the respondents, representing 3%, fell into the lifeline category by consuming less than 50 KWh of electricity every month. The next consumption bracket of 51 KWh – 300 KWh was the modal bracket taking up 298 respondents comprising 74.5% of the sample. Another significant proportion, 19.8% of the sample (constituting 79 respondents) consumed between 301 KWh and 600 KWh of electricity monthly. The number of people who consumed between 601 KWh and 1000 KWh of electricity monthly was 11 representing 2.7% of the sample. Many respondents indicated that they restrict/conserve electricity as much as possible to minimize cost. This may be partly due to incremental tariffs that punish people who share a connection, but also because people may need to complement their power supply with alternative sources at an extra cost. The 136 University of Ghana http://ugspace.ug.edu.gh results are consistent with Scott and Seth (2013) who noted that connection charges and tariffs can be a barrier to access for poor households even when an electricity supply is physically available. Table 4.12 illustrates the end user electricity tariffs of the utility provider. It can be observed that consumers pay according to an increasing block tariff. For example, a compound house connected to the electricity network with 10 people using 3.5 KWh per day, will consume about 98 KWh monthly which is more than the lifeline target of 50KWh per month and will therefore have to pay according to the higher tariff (i.e. GH¢65.98 plus GH¢31.33 extra charges, with breakdown as follows – service charge GH¢6.33, national electrification levy GH¢4.45, street light GH¢4.45, VAT GH¢14.66, NHIL GH¢2.44, subsidy GH¢1). Table 4.12: End user electricity tariff in Accra Tariff category Unit price/ Unit volume Monthly price comparison (GHp/KWh) (GHp/KWh) in study area ECG residential 0-50 (Exclusive) 33.56 1,678 (GH¢ 16.78) 51-300 67.33 3,433 – 20,199 (GH¢ 34.33 - 201.99) 301-600 87.38 26,301 – 52,428 (GH¢ 263 – 524.28) >600 97.09 > 58,351 (GH¢ 583.51) Service charge (GHp/month) 633.17 633.17 (GH¢ 6.33) ECG non-residential 0-300 96.79 – 301-600 102.99 – >600 162.51 – Service charge (GHp/month) 1055.29 – Source: Computed from PURC electricity tariff effective December 2015 Per the utility data from PURC, it is a little wonder that, in the study area, only 3% of respondents fell within the lifeline (exclusive) category while the majority of households, representing 74.5% were within the 51-300 bracket. This finding on electricity cost provides useful insights for policy makers in planning tariff schedules for low-income residents. Similar to water, it questions the degree to which such subsidies (delivered through increasing 137 University of Ghana http://ugspace.ug.edu.gh block tariffs) actually benefit the poor (Scott and Seth, 2013; Komives et al., 2005). Especially, the fact that most households share meters in multi-occupied compound houses and combine multiple sources to improve access suggests that the poor are paying more for electricity. In sum, the key households’ concerns regarding the dynamics of access to electricity relate to the phenomenon of shared meter connections and introduction of pre-paid meters due to fears of spiralling costs and sudden power cuts, leaving some residents without access to electricity. Residents would much prefer prepaid meters to be allocated to individual household units. The study thus recommends that the introduction of multi-tenant pre-paid meters may be the way forward for multi-habited compound houses as depicted in the study area. 4.3. Conclusion With regards to spatial patterns of access to water, it was revealed that water coverage at the metropolitan level was above the national and regional averages. At the community levels coverage compare favourably with the metropolitan level, albeit with some disparities. Beyond coverage, a functional interaction was established between households’ access from the utility provider, which serve less than half of the sample, and a complex water service supply network established through informal community and neighbourhood vendors and tanker operators who are major actors in the water supply chain because of limitations in the performance of the utility provider. Reliability, quality, cost and usability were found to be examples of everyday practices that are key determinants/benchmarks for unravelling multi- dimensionality of access in the metropolis, as access defined by coverage alone hide a lot of details. Sachet water consumption is growing among households suggesting that socioeconomic status and municipal pipe-borne water supply may not suffice as standalone reason for sachet consumption. The interaction between formal actors and informal networks in water service delivery was further exhibited in the micro-politics particularly between 138 University of Ghana http://ugspace.ug.edu.gh landlords and tenants, and long-time occupiers and newcomers. The study concludes that water supply networks show more promise, especially when combined with official recognition, regulation and complementary approaches to making services accessible and affordable to poor households. Like the case of water, the spatiality of electricity coverage indicates a positive situation, albeit huge access gaps in relation to neighbourhood dynamics in GAMA. Beyond coverage, there is a strong tendency for households to access electricity through shared metered connections, as opposed to separate households, even for premises that double as commercial centres. Besides mode of access, the introduction of pre-paid meters for residents in the metropolis was found to dominate concerns over electricity access, as the meters were perceived to run fast and the practice of meter sharing was associated with overconsumption; budgeting for next top-ups; and issues concerning who fixes faulty meters. These issues coupled with dynamics of reliability, quality, cost and usability of electricity make the shared system unsustainable in the long-run. The outcome is that it has led to the increase in illegal connections, and supply has to be complemented with alternative sources including private generator, flashlight/torch, kerosene lamps and candles, despite the associated cost. Using these strategies was found to vary based on the residential and socio-economic status of households. The findings stress the interplay between formal actors and informal networks that are essential to improve water and electricity access in a complex, multi-dimensional setting in the study area. These dynamics expose the inefficiencies of the decentralized governance system of the metropolis. The next chapter thoroughly discusses the implications of the above findings on income generating activities for the Greater Accra Metropolitan Area. 139 University of Ghana http://ugspace.ug.edu.gh CHAPTER FIVE ASSESSING THE IMPACT OF WATER AND ELECTRICITY ACCESS ON HOME-BASED ENTERPRISES IN GAMA 5.0. Introduction This chapter examines how access to water and electricity enhances income-generating activities. The chapter first gives an overview of residential livelihood activities in the urban economy of Accra. Secondly, it analyses the occupational geographies and linkages of household economies as well as the ramifications of GAMA’s present water and electricity delivery on home-based enterprises (HBEs). Specifically, the chapter examines the prospects of HBEs for generating employment and income, and demonstrates how poor access to services is negatively affecting the fortunes of residential enterprises. 5.1. Residential livelihood activities in the urban economy Data from the 2010 Ghana Population and Housing Census show that GAMA is not very different from the other five metropolitan areas when it comes to livelihood activities. About 93% of the population aged 15 years and above are employed. The private sector is the dominant sector employing 91% of the labour force, though the majority of these (74%) are in the private informal category (service, sales, craft and related trades) with only 17% employed by the private formal sector. In terms of employment status, 35.2% of the working population are employees while 48.2% are self-employed without employees with only 7.4% having employees. Also, 3.9% work as apprentices, 2.2% are engaged as contributing family workers, 2.0% are casual workers and 0.9% are house helps (GSS, 2014b). The working population of the research localities are engaged in various occupations as displayed in Table 5.1. Overall, it is evident that the majority (38.5%) of the labour force in GAMA are service and sales workers while 20.1% are engaged in craft and related activities. Those employed as managers, professionals, and technicians constitute 17.2% of the labour 140 University of Ghana http://ugspace.ug.edu.gh force with only 1.7% engaged as skilled agricultural, forestry and fishery workers. The key occupations driving the employment sector are wholesale and retail (35.2%), manufacturing (14.1%) and accommodation and food services (10%). The same patterns can be observed in the research localities. The above analysis suggests that the contribution of the informal sector to household economies cannot be overemphasized. Table 5.1: Occupation of employed population in Accra in 2010 Percentage employed Occupation Korle Gonno New Town Gbawe GAMA Managers 4.1 3.8 6.8 5.3 Professionals 7.7 5.7 11.1 7.9 Technicians and associate 4.5 3.9 4.1 4.0 professionals Clerical support workers 4.2 3.4 4.0 3.7 Service and sales workers 39.2 40.7 34.7 38.5 Skilled agricultural, forestry 1.7 1.1 2.5 1.7 and fishery workers Craft and related trades 20.4 22.2 22.6 20.1 workers Plant and machine operators, 4.9 6.8 7.2 6.0 and assemblers Elementary occupations 13.1 12.2 6.8 12.6 Other occupations 0.2 0.2 0.2 0.2 Sample size 13,847 15,421 27,978 772,877 Source: Compiled from GSS District Analytical report and unpublished community level data Home-based enterprises (HBEs) have become an important part of the urban landscape providing livelihoods for people in the informal economy. This recognition has been widely documented by scholars (Strassmann, 1987; Tipple, 1993; Gough, 1996; CARDO, 2000; Yankson, 2000b; Gough et al., 2003; Gough, 2010; Oosterbaan et al., 2012). This study conceptualizes HBEs to include activities where part, but not necessarily all, of the activity takes place in the home and immediate environment; inner room, outdoors in interior 141 University of Ghana http://ugspace.ug.edu.gh courtyards or in front of the house, and public spaces including streets, alleyways and lorry stations. Therefore, the study broadly refers to such activities as residential livelihood activities as a way to be able to embrace the continuum of enterprises that start from the home but end up elsewhere in the vicinity and also appreciate their contribution to the local economy. This is important because home businesses are increasingly spreading out of the core domain of the conventional definition of HBEs because of unavailable spaces and to some extent operators’ optimism to enjoy economies of scale. Thus, women preparing koko7 and kenkey8 in the home but selling elsewhere in public spaces were classified as HBEs. In contrast, enterprises located in the settlement but operated by outsiders were not counted as HBEs. Similarly, business operators who used the home environment to store products or equipment but worked, for example, in their previous place of residence were excluded. The rationale is to understand the impact of households’ access to water and electricity on the viability of home businesses. To provide the relevant context within which the discussion is situated, the study categorized HBEs into water and electricity businesses. In this regard, it was found that 62% of the enterprises are water-dependent while nearly 80% of them are electricity-driven, as depicted in Figure 5.1. This suggests the significance and implication of water and electricity access for enhancing the performance of HBEs, which shall be discussed later in this chapter. Based on the conception of HBEs above, this section investigates the following questions: Are the businesses productive and perceived as profitable by their operators? What is their contribution to household income? How does access to work space and services affect HBEs? 7 Koko is a soured and spicy corn/millet smooth porridge usually served with fried bean cake (koose) 8 Kenkey is a staple dish made with corn flour (or corn flour and grated cassava) and wrapped in maize or corn husks, usually served with pepper sauce and fried fish. 142 University of Ghana http://ugspace.ug.edu.gh The fieldwork unraveled these dimensions. It established that an HBE type is not simply defined by the product or service it provides. Instead, it is conceived as a holistic entity that involves the operator (male-female, head-spouse-other), the market (household, neighbourhood sales), and the location (home, inner-locality, etc). The categories of Table 5.2 have been set up in part with these criteria in mind. To bring these constituents out and yet avoid excessive sample fragmentation, some product/service types have been combined and categorised. For example, brick making, welding, aluminium fabrication, woodworking, bag making, soap and detergents making as ‘manufacturing’; internet cafes, typing and photocopying/printing centres as ‘secretarial services’ etc. Figure 5.1: Distribution of water and electricity dependent HBEs in GAMA 100 15% 20.5% 80 37.8% 60 85% No40 79.5% 62.2% Yes 20 0 Residential Water-dependent Electricity-dependent enterprises enterprises enterprises Source: Field data, 2015 The study found that residential livelihoods are becoming increasingly widespread in GAMA. More than 8 out of 10 households (85%) in the study area engage in HBEs which provide crucial resources and access to basic goods and services through informal networks. The spread of HBEs is remarkably similar across the research localities: Korle Gonno (88%), Abuja (87%), Gbawe (85%) and Accra New Town (80%) (see Table 5.2). The slightly higher percentages in Korle Gonno and Abuja may be due to the greater share of households which tend to work from within their multi-occupied compounds, compared with Gbawe and Accra 143 Percentages (%) University of Ghana http://ugspace.ug.edu.gh New Town where there is a rising number of single family houses with many people working outside the settlement. It is worth adding that although there is a substantial number of enterprises in Accra New Town, the vibrant local economy is fueled largely by outsiders who trek daily to work in the area rather than HBE operators. Table 5.2: Type of HBEs in GAMA (%) Type of HBE Korle Accra New Abuja Gbawe Overall Gonno Town sample Food/drink processing 10.4 11.8 13.7 10.5 12.1 Food/drink sale 30.1 29.8 30.5 20.5 27.7 Retailing and personal services 19.7 24.7 30.5 20.7 23.8 Hairdresser (beautician)/barber 11.4 11.3 4.6 15.3 10.4 Seamstress/tailor 11.4 6.2 11.5 5.9 8.6 Manufacturing 2.3 3.8 4.6 11.8 5.6 Repairs 4.5 3.8 3.4 10.5 5.6 Secretarial services 8.0 6.2 - 2.4 4.1 Traditional healers 1.1 1.2 1.2 1.2 1.2 Others 1.1 1.2 - 1.2 0.9 Sample size 88 80 87 85 340 Source: Field data, 2015 These findings are consistent with earlier studies. In a longitudinal study of Madina, a low- income area of Accra, Gough (2010) found that two-thirds (66%) of houses and one-third of households (33%) had an HBE. Similarly, Gough and Yankson (1997) reported that around 80% of houses in peri-urban Accra have at least one person working in or in front of the house. The unifying thread of these earlier studies and this present one is the continuity and adaptability in HBEs (see Gough, 2010), which tend to be the outcome of structural and broader policy changes and processes. More specifically, the findings are not unexpected in view of the fact that formal sector employment has become an increasingly remote prospect for many urban residents. This process has occasioned a rise in entrepreneurial ingenuity of people, which is underpinned by factors such as multi-habitation and densification of 144 University of Ghana http://ugspace.ug.edu.gh households that naturally create the economic environment (including ready markets) needed by HBEs to thrive. In terms of the distribution of the various types of HBEs in the research areas, overall Abuja recorded the highest number of enterprises engaged in activities such as retailing, processing and sale of food and drinks, and tailoring. This is not surprising because Abuja contributes to Accra’s economy in two key ways: 1) it is a hub of remaking and rebranding clothing which have been purchased from the nearby Kantamanto market; 2) it is a host to a well-networked food market that serves the CBD. Abuja is followed by Gbawe where the majority of hairdressing, manufacturing and repair activities abound. Korle Gonno and Accra New Town dominate when it comes to secretarial services. The Korle Bu hospital and the printing press centres present in Korle Gonno and Accra New Town respectively might have contributed to the rise in these services. The remainder of the HBEs is fairly distributed across the study areas. Overall, GAMA displays a wide range of HBEs with the most common types being sale of cooked food and drink (27.7%) and retailing of items such as recharge units, stationery, clothes, consumables and other daily needs like toiletries (23.8%). Retailing activities are popular because they cater to local demand, entail limited skills and require little start-up capital. The retail outlets sell groceries and daily needs products like gari, soap, toilet roll, canned food and drinks. There are also several avenues where operators specialize in only a selection of these, for instance fresh produce and local-made drinks. Together, these HBEs account for nearly 52% of the sample. Consistent with previous studies, the results highlight how HBEs are gradually shifting from food processing to the sale of food and general merchandising. Gough et al. (2003) found that 43% of HBE operators in Mamelodi, Pretoria were involved in the sale of food and drinks. This observation can be attributed to the increase in tenant households where people usually do not have the means to follow the catering path 145 University of Ghana http://ugspace.ug.edu.gh and instead engage in easygoing trading activities. The study gives credence to the national statistics which indicate that more than a third of the population (35.2%) is involved in the wholesale and retail (sales and services) activities in GAMA (GSS, 2014b). It illustrates how the frequency of HBEs is increasing, in consistency with the growing informalization of urban economies (Haug, 2014). The processing and repackaging of food (12%) is a common enterprise which entails a chain of activities where operators buy in bulk quantities, process and/or repackage the product, and create value-addition. Such enterprises include chop bars where women make staple foodstuff like kenkey, fufu9 and banku10 which is heavily reliant on water, as well as repackaging different products like crunchy and other snacks in small bags for people to buy. With such businesses, operators are able to minimize their operational cost by relying on the household pool of labour and resources. For example, children fetching water and running errands before and after school hours, and spouses providing logistic support and managing business profits. Similar to food processing, trade in alcoholic beverages entails a recognizable number of drinking spots making the local gin (akpeteshie) and assorted bitters while others sell beer. Residents expressed worries about the rise in the number of such spots because of the noise level, crime, and also due to the impact on the youth particularly in relation to lifestyle and poverty. This result chimes with the findings made by Mensah (2015) who found that the increasing number of drinking spots and its attendant impacts is due to youth unemployment in Ga-Mashie. 9 Fufu is a staple food made with cooked and pounded cassava and green plantain usually served with light soup 10 Banku is a fermented corn/cassava dough mixed proportionately and cooked in hot water into a smooth, whitish consistent paste. Served with soup, stew or pepper sauce with fish. 146 University of Ghana http://ugspace.ug.edu.gh The food-related HBEs vary in size from ones that occupy a container shop and offer a range of dishes such as waakye11, plain or jollof-rice, banku and fufu with a seating or eating space for customers to ones that operate from a makeshift table and sell specific food for customers who usually eat from arranged tables in a corner space or carry away for privacy reasons. Likewise, the retail outlets vary greatly in size from ones that occupy a whole room-size shop and sell a wide range of products to ones which operate from a table top and sell a limited range of items, such as gari, sugar and water, usually in very small quantities. The advantage of the smaller HBEs lies in their selling in small quantities, offering convenience of proximity and long working hours, and sometimes offering credit to regular customers and known neighbors. The findings reveal an increasing number of HBEs that offer services to residents. This includes hairdressers and barbers as well as seamstresses and tailors, who make up to 19% of the sample. These enterprises largely depend on water and electricity. Even though these two occupational categories have existed for a long time, a noticeable trend among them is the customizing of clothing from African fabrics by fashion designers; and, home delivery service by beauticians who specialize in pedicure, manicure and make-up. This emerging fashion scene has been articulated by Langevang (2016) as a way for fashion designers to pursue their passion, change the associations ascribed to ‘Africanness’, and to revitalize failing clothing industries. Manufacturing of items for sale to a wider market is another noticeable enterprise in GAMA. This constitutes 5.6% of HBEs, with operators engaging in brick making, welding, aluminium fabrication, woodworking, bag making, and soap and detergents making. A few operators 11 Waakye is a dish of cooked rice and black eye beans usually served with (hot) pepper sauce and fish 147 University of Ghana http://ugspace.ug.edu.gh collect plastic bottles, aluminium cans and metals (electronic waste) to sell to recyclers. Repair work by computer specialists, electricians, car mechanics and carpenters is also common, making up 5.6%. As computer ownership and internet penetration is low among low-income populace, cafes and other commercial centres offer the convenience of internet access, and typing, printing, and photocopying services. Such secretarial services represent 4.1%. A couple of herbalists or traditional healers also operate in the study areas, constituting 1.2%. Finally, the ‘other’ category of services include laundry, car washing bay, licensed chemical shops, etc. In most cases, such HBEs are owned and operated by men. In sum, despite the fact that HBEs are widespread, much remains to be uncovered about their spatio-temporal aspects which is key in understanding how households maximize their spaces to obtain optimum outcomes. The argument arises as to who oversees the use of home space? What mechanisms are in place to guarantee equity in access to space? Accordingly, this chapter contributes to knowledge by providing evidence on the geographies of residential livelihood activities, their implication on service delivery, and their diverse contributions to and linkages with household economies. This is crucial because such complexities and perspectives are treated separately in recent literature on informal livelihoods (Anyidoho and Steel, 2016). Put simply, we know little about the degree of dependence on these enterprises: the proportion of household employment and income which is generated from HBEs, how the income generated is expended by the households, and the gender differences. These are some of the critical questions and issues the next section examines in the research localities in GAMA. 5.2. Occupational geographies and linkages of household economies in GAMA A key role of residential enterprises is that they enable the production and supply of a wide range of goods and services that are readily required for day-to-day domestic living. The study identified three fundamental reasons explaining why households engage in HBEs: proximity 148 University of Ghana http://ugspace.ug.edu.gh or convenience, household support, and possibilities of combining productive and reproductive work. Further, operators cited diverse reasons for selecting or settling on their current premises. The outcome of their locational decisions has contributed to the modifications in the spatial and economic landscape of the metropolis (Oosterbaan et al., 2012) as well as on the development and sustainability of the enterprises (Gough, 2010). The majority of HBE operators live in compound houses usually occupying only one or two rooms, resulting in them usually having 12–24 m2 of internal space. This may even be smaller especially in the indigenous core areas. The small size of the dwellings and available internal space have resulted in the spreading out of businesses to satellite spaces outside the home. Table 5.3 illustrates the locational characteristics of HBEs in the study area. The majority of enterprises are located at either the frontage of the houses of operators (33.5%), in public spaces (30.2%) or in the interior courtyards (21.2%). Another 10.4% are operated from rooms including kitchens, halls and verandas. Only a small fraction (4.7%) are operated in workshops attached to the houses and in backyards. Such agglomeration of HBEs allows the integration of business activities into domestic chores which has implications for access to services. Table 5.3: Locational characteristics of HBEs Enterprise locations Percent Room 10.4 Courtyard 21.2 Frontage of houses 33.5 Detached from houses 4.7 Public spaces 30.2 Sample size 340 Source: Field data, 2015 In relation to the categories of HBEs, more than a third of businesses engaged in the production and sale of food is located in the frontage of houses. Whereas retailing and 149 University of Ghana http://ugspace.ug.edu.gh personal and secretarial services are typically sited in public spaces including alleyways and lorry stations. Dressmaking, barbering and hairdressing shops, metal and carpentry shops, and workshops for general repairs and miscellaneous activities are either attached to or detached from houses as well as in room spaces such as balcony and store rooms. These locational diversities connect to the issue of space which is required by HBEs for production, showcasing, storage and seating customers. With respect to the quality and acceptability of space utilization, two-thirds (66%) of the operators indicated they were satisfied with the space for their work while 34% indicated otherwise. The majority of the latter category explained that they would need about double or more of their current space to expand their scale of production. These concerns regarding space were more crucial for HBEs such as food, tailoring and metal-based activities. The results reveal that although there are times when no formal regulatory mechanisms are in place with regards to access to space for residential livelihood activities, there exists a wide range of informal arrangements and regulatory mechanisms. It was found that length of stay and occupancy status (long-time occupiers’ vs. newcomers, landlord’s family vs. tenants), and social networks are key factors that influence access to space. The quote below by a 35- year-old single parent in Korle Gonno illustrates the ‘who you know’ dimension characterizing the privatization of and access to space, suggesting that operators cannot operate at will. When I first came to live in this house, I wanted to sell fried egg with bread and fruits in the evenings after my cleaning job. But the landlord told me he doesn’t allow tenants to sell in-front of his house. Meanwhile his wife and in-laws sell in the house. So I begged the owner of the next three houses, she is my church member, and she gave me the space to sell. I pay GH¢20 to her every month. In another instance, the quote below by a middle-aged caretaker of a family multi-habited house in Gbawe demonstrates how the ‘first come first serve’ principle is shaping the spatial architecture of HBEs. She opined: 150 University of Ghana http://ugspace.ug.edu.gh As for my area getting a space to work is hard: it is based on first come first serve. There is no space so you will see that a lot of the businesses operated by the new comers like hairdressing and tailoring shops are located outside the home in detached shops and open spaces. Because the old businesses, I mean the food sellers and table stores, have occupied the inner yard and frontage spaces of houses. Still, other operators strongly underscored that, beyond affinity, the arrangements regarding who gets to use a particular space are underpinned by economic enticement. This view is captured in the voice below by an elderly tenant in Accra New Town, who owns and operates a small provision shop. Before, you could just speak with the landlord and start a business in or in front of the house. But now it’s not easy like that. Sometimes you can be there but when a new tenant comes he or she will be given a space to do a business in front of the house. The landlords will tell you that their case is different. But we think they give them [landlords] big money and get other help from them. Two key points are of significance here, first, there is the privatization, and a subtle politicization, of space by landlords and long-occupied residents who act as overseers (space- lords) and determine who accesses what space and for what purpose. Second, unlike HBEs operated in rooms and courtyards, those operated from the frontage of houses and those sited in public spaces come at a price which ranges between GH¢15-50 monthly depending on the type and scale of business and social networks. In addition, such operators are at the mercy of city authorities when it comes to daily tolls and eviction harassment. These observations are a manifestation of the consequences of the increasing saturation and density of enterprises. More so, the concepts of ‘near’ and ‘distant opportunity spaces’ proposed by Sumberg et al. (2014) are useful here. Sumberg et al. (2014) define an opportunity space as the spatial and temporal distribution of work options that young people may exploit as they attempt to establish an independent life. ‘Near opportunity spaces’ are those that are available in the vicinity of where they are living. ‘Distant opportunity spaces’ – norms and expectations – are 151 University of Ghana http://ugspace.ug.edu.gh influenced by characteristics of social difference (such as gender, age, class, and ethnicity) as well as by social relations and networks. In a related point the analysis reveals that unlike in the newly built-up landscapes of Accra New Town and Gbawe where overcrowding of space is minimal, the densely populated areas of Korle Gonno and Abuja are experiencing a lot of overcrowding and encroachment on public spaces. In most cases, the spaces between the houses and the frontage space are only large enough for a person to pass through aslant, as people have erected table shops on every available space to sell or set up ‘open space kitchens’ as illustrated in plate 5.1. This presents a potential hazard as the layout may not permit entry by emergency services especially in the case of Abuja. Plate 5.1: Environmental conditions of a typical open space kitchen used for HBEs Source: Author’s field data, 2015 A section of the respondents who did not engage in any HBE expressed worry about the uncontrolled erection of structures on the little space in the houses by operators. They indicated strongly how such a phenomenon affected the spatial landscape and beautification of the home environment as well as ventilation since some enterprises contributed to the 152 University of Ghana http://ugspace.ug.edu.gh concentration of smoke and heat. The voice below of a 25-year-old male SHS leaver who works as a shop keeper in Accra New Town shares his view on the situation. Is here a living place or a cooking space? People cook on every space they find. Now we have to prepare our family meals at the entrance to our rooms because the compound is always busy on fire; it is either the koko and kenkey sellers cooking as early as 4:30am, or the chop bar people cooking from 6am to 6pm. You are even shy to bring a friend to your room because the environment is dirty and noisy. I have to at times stay outside till 11pm before coming to sleep in the room because of the heat and noise. Re-echoing this point, a 28-year-old male tenant from Abuja who could not hide his frustration lamented: You can’t have your privacy in this house. You open your door and it’s a cooking pot been washed close by or wastewater running in front of your door. Every minute the children are knocking at your door for a match stick or for a helping hand to carry a pan of food(stuff) meant for the street. It is a problem but what can you do. At times when you arrive home late from work you are able to get some last minute food to buy. This implies that although residents expressed these concerns they underscored the benefits of HBEs and emphasized the fact that people need to make a living. One common characteristic that the settlements studied demonstrate is how the house type influences the number and mode of operation of HBEs. Thus, the phenomenon of one house, multiple households and many home enterprises was notable (Gough, 2010). In the indigenous core areas of Korle Gonno and Gbawe and also in migrant settlements like Accra New Town and Abuja, most of the houses are compounds inhabited by multiple households hence there are often several HBEs within one house operated by members of different households. Similar to the findings made by Gough (2010), although the HBEs are independently operated there are cases where operators running similar businesses cooperate closely. In a particular case in Abuja, in a compound house occupied by ten separate 153 University of Ghana http://ugspace.ug.edu.gh households, the women in six of the households (mostly from the north) process and cook food for sale. Although they make and sell their own food they share the cooking space and equipment like cooking stove and utensils, often help each other, and agree on the different timings and locations to sell their food. This is however not the case in Gbawe’s newly built- up landscape as the houses are predominantly single family dwellings with a few frontage shops usually dealing in retailing and dressmaking, showing how house type (social status) influences the nature and operation of HBEs. There are households who have more than one HBE, albeit often operated independently. This could be different HBEs, for example: a woman selling koko and her husband collecting and refurbishing scraps; or similar HBEs: a husband making soap and detergents for his wife to sell; or a woman and her partner running a fashion designing business where they share similar equipment but maintain separate accounts. Plate 5.2 illustrates a situation where women from different households engage in cooked food business while their partners use the same yard for their carpentry business. Plate 5.2: A case of multiple households engaging in cooked food and carpentry business Source: Author’s field data, 2015 154 University of Ghana http://ugspace.ug.edu.gh Similarly, there are occasions where one household member operates several HBEs. For instance, a water vendor who runs a public shower and a bar; a hairdresser wholesaling and retailing sachet water and running a public shower; a young lady owing a small provision shop and running a hairdressing salon where she does make-up and sells wigs and ladies underwear. In other instances, one enterprise could be subdivided into multiple businesses. For example, a mother running a provision shop and her two daughters separately owning freezers to wholesale and retail soft drinks and sachet water. A few individuals combine both formal and informal jobs; working as a teacher and operating an internet café cum photocopying and printing shop, or working as a cleaner in a company and selling fruits and egg with bread in the evening. This observation of multiple-modes of livelihoods or what is popularly called moonlighting – a phenomenon of multiple job- holding– have been reported in the literature. For instance, Baah-Boateng et al. (2013) reveal that moonlighting is practiced by a large number of salaried employees. They indicate that the desire to moonlight is a function of varied factors, including personal and household characteristics: changing labour market outcomes and opportunities in the urban economy; the nature of formal (under)employment (working hours and earnings) and the need for additional income due to low wages (see also Owusu, 2007). Gender differences, in relation to ownership and management, were observed. The study found that whereas ownership of HBEs was dominated by males who owned 51%, females accounted for 49%. On the contrary, nearly two-thirds (65%) of females were in charge of managing the businesses as against 35% being males. A key point to note is that whereas ownership is slightly in the hands of men, day-to-day management is more for females. This finding must be put in context as some HBEs are owned and operated by the household as a whole, but since in Ghana and most African societies, men are recognized as household heads, they are often reported as the ‘owners’ (Owusu, 2001). A wife would typically describe 155 University of Ghana http://ugspace.ug.edu.gh herself as the ‘owner’ of an enterprise only when it is operated by her and her children without the husband's assistance. This revelation emphasizes the argument that HBEs are more attractive for females since it conveniently enables the combination of domestic and economic activities. Undoubtedly, this has crucial implications for the urban planning of employment activities for women. A primary contribution that HBEs make to the household economies of low-income people is the generation of employment and income. The job creation capacity of HBEs is appreciably high: 70.5% of all the study participants were somehow doing some form of work for the HBEs either on full-time or on part-time basis while the remaining 29.5% were not involved at all in any activity. Among the household members employed by HBEs 35% were full-time employees while 65% were part-time workers. Analytically, this implies that every single HBE out of the total of 340 surveyed employs one full time worker and two part-time workers. This infers that the job creation capacity of each HBE is three persons. In contrast, a total of 310 households, representing 77.5%, of all the respondents surveyed have people working outside the home environment. With a total of 440 outside workers counted, this works out roughly to an average of one external worker per household. Juxtaposing this figure against the three persons working in each HBE, it implies that home businesses offer two times more employment opportunities than businesses that exist outside the home. This revelation confirms Afrane (2003) and Strassmann (1987) who reported that employment shares of HBEs could be twice as much than outside business, especially in low-income urban neighbourhoods. Comparative analysis of average monthly earnings from HBEs and businesses outside the home confirm the high economic potentials of residential livelihood activities. It is evident from Table 5.4 that manufacturing provides the highest average income returns of over GH¢1500, followed by dressmaking and fashion designing; retailing and personal services (including private water vendors and phone recharging centres); food and drink business; 156 University of Ghana http://ugspace.ug.edu.gh hairdressing and barbering; repairs and secretarial services, in that order. Traditional medicine which provides the lowest average monthly income yields GH¢380. The average monthly range of incomes from HBEs was between GH¢196 and GH¢2,616 with an overall mean of GH¢890. On the other hand, the average income from household members working outside the home was estimated to be GH¢600 per worker. It must be added that the income figures from HBEs are technically turnover returns and not profit per se. Due to poor note-taking and accounting inadequacies many of the operators could not determine their actual profit margin with a high degree of accuracy. This phenomenon is documented in the literature on the informal sector (Osei-Boateng and Ampratwum, 2011; Anyidoho and Steel, 2016). Most importantly, since the two income streams are precisely different, it will be incongruent and misleading to compare their relative contributions to the total household income. Table 5.4: Average monthly incomes from HBEs (GH¢) Average Type of HBE income Minimum Maximum Food/drink processing and sale 1000 120 4200 Retailing and personal services 1120 100 3800 Hairdresser (beautician)/barber 750 200 1500 Seamstress/tailor 1250 400 3000 Manufacturing 1550 300 5500 Repairs (electrical, mechanical) 700 150 1800 Secretarial services 650 140 1400 Traditional healers 380 100 750 Others (PA hiring, washing bay) 610 260 1600 Total average 890 196.6 2,616.6 Source: Field data, 2015 Again, it was observed that more than two-thirds (68%) of HBE operators do not distinguish the business finances from their domestic banking transactions. The explanation to this phenomenon is that most of the enterprises are directly linked with basic economic survival 157 University of Ghana http://ugspace.ug.edu.gh of the household. This was more in the case of women (78%) as compared to 60% in the case of men. More especially, this observation is understandable because women are often responsible for minor domestic purchases in the home while the men take care of larger expenditures. Consequently, women handle household business coffers. In a related observation, it was evident that business activities are inseparably linked with normal family chores. For example, a customer is not denied a service simply because the household is having supper. Typically, a household member suspends the activity for a while to render the service. This work-lifestyle continues from morning till night especially for households which operate chop bars and retail and personal services, which cumulatively constitutes the majority of HBEs. Obviously, this serves the interest of females as such household integration and arrangements facilitate mobility and allows the combination of domestic and income generating activities (see Esson et al., 2016). Besides financial contributions, a substantial share of HBE outputs (good and services) are consumed by the household members. The results reveal that most households consume some of the goods produced or sold, albeit the fraction consumed is relatively minimal. For example, 85% of households engaged in the processing and sale of food and drink consume 8% of their output. While those engaged in retailing and personal services consume 12.7% of their output. In sum, it can be said that the enterprises that provide basic needs particularly daily products and consumables have some proportions of the output being used by households. More significantly, the analysis shows that income generated from HBEs is spread over a number of household expenditure items including food, utilities, sanitation, transport, housing and education, in the order of importance. Although this pattern of income uses is not uniform in the study areas, nonetheless, it is consistent with the expenditure patterns of low-income households at the national level. Comparable to the national ranking by the Ghana Living 158 University of Ghana http://ugspace.ug.edu.gh Standards Survey (GLSS 5), food expenditure takes the largest share of household income. This present study found that the fraction of HBE income committed to food is about 30% as compared with 40% at the national level (GSS, 2008). Virtually, a similar proportion (26%) goes into the payment of utilities, particularly water and electricity. Further, the results reveal spatial variations in the average income ranges of HBEs across the research localities, as shown in Table 5.5. The monthly income from Gbawe is relatively higher than that of Accra New Town, Abuja and Korle Gonno. In monetary terms, this is reflected in the following order GHȼ1,372, GHȼ1,212, GHȼ855, GHȼ742 respectively. These differences reflect the disparities in scale and density of HBEs across the settlements. Also, it is clear from the table that the majority (73%) of operators in Abuja are within the lower income bracket of less than GHȼ600. This is followed by 42% from Gbawe and 35% from Accra New Town who are within the middle income bracket of between GHȼ600 and GHȼ1200. A similar pattern is observed in the high income category where 31% of operators in Gbawe and 28% in Accra New Town recorded an average income of more than GHȼ1200. Table 5.5: Settlement distribution of average monthly earnings from HBEs (%) Income Korle Accra New Abuja Gbawe categories Gonno Town Below 600 36 37 73 27 600 – 900 43 4 8 11 901 – 1200 8 31 7 31 1201 – 1500 3 12 4 12 Above 1500 10 16 8 19 Average income (GHȼ) 742.98 1,212.20 855.60 1,372.70 Source: Field data, 2015 In a nutshell, the study established that the enterprise operators demonstrate a high degree of resourcefulness and flexibility with regards to carving out a space for their enterprises and using the same space for multiple purposes. Overall, greater HBE employment was generated in Korle Gonno and Abuja (9 in 10 households), compared to Accra New Town and Gbawe 159 University of Ghana http://ugspace.ug.edu.gh (8 in 10 households). The reverse is, however, the case when it comes to income levels, with Gbawe and Accra New Town recording higher income earnings than Korle Gonno and Abuja. Put differently, the share that HBEs contribute to household income in Korle Gonno and Abuja is lower than in Gbawe and Accra New Town. Among the conceivable explanations is the much lower density but relatively higher purchasing power of the latter as compared to the higher density but lower purchasing abilities in the indigenous core areas of the former. All of this has policy implications. While several factors explain these complexities, it is undisputed that the dynamics of access to services, particularly water and electricity, significantly add a nuanced dimension to our understanding of HBEs. This is the focus of the next section. 5.3. Home-based activities vis-à-vis water and electricity access in GAMA In consonance with the literature, this study found that residential livelihood activities have been characterised by the concentration of businesses in the home (rooms, courtyards, and frontage of houses) and the rising enterprise growth resulting in encroachment on space outside the immediate home environment (along roads and lorry stations). Despite the burgeoning scholarship that attests to the significance of HBEs, this study advances the discourse on household economies by examining the impact of access to water and electricity on HBEs and how the outcome contributes to poverty reduction in GAMA – a seemingly neglected aspect in the literature. The rationale being that the processes shaping enterprise growth are complex, dynamic and consistently produce new urban entrepreneurial landscapes and ‘poverties’. Two key findings are worth mentioning: First, access to quality water and electricity is unsatisfactory (as discussed in chapter 4); Second, 62% of the enterprises are water-dependent while 80% are electricity-dependent (as illustrated in figure 5.1). This high share of HBEs relying on water and electricity implies that improving access can have significant implication for business success. Suffice to add that the opposite can also pose major challenges to HBEs. 160 University of Ghana http://ugspace.ug.edu.gh The analysis shows that about two-thirds of HBEs are connected to sources of electricity (65%) and around half have water supply (52%), albeit an unreliable and unsatisfactory supply. Similarly, their access to other services is also poor: toilet facilities (35%), garbage removal (19%) and waste water drains (12%). Quality supply of water and electricity was indicated by the operators as the most direct and crucial infrastructural service needs for their enterprises. The outcome of the disparities in access to water and electricity manifests in two ways: 1) impact household livelihoods, 2) modify the entrepreneurial landscape. In the ensuing paragraphs below, the analyses take a deeper look at the impact of water and electricity access on HBEs in the study area. The discussions are carried out along the reliability, cost, and quality dynamics of access which have been articulated in chapter five. In the case of water access for business, the fact that the majority of operators depend on private vendor sources indicate that the dynamics of water access has tangible economic bearing on HBEs. The analysis reveals that nearly half of the operators (48.6%) use between 1 to 20 gallons of water daily for their operations while one-third (33.4%) use between 21 to 40 gallons. Only the remaining 18% of operators have direct access to in-house/yard piped- borne water. In terms of cost, this translates into 48.9% of the operators who rely on private vendors spending between GH¢1-10 as their daily business expenditure on water while 32.3% spend between GH¢11-20. Assuming an average of 6 working days in a week, the enterprises spend around GH¢25-250 and 251-500 as monthly cost of water. This high cost of water incurred by the majority of HBEs is a reflection of the water supply situation in GAMA where most operators rely on private vendors. The cost of water is significantly higher when compared with their counterparts who have access to in-house/yard piped-borne water. This latter category pays an average amount of between GH¢45-100 as monthly bill—approximately GH¢2-4 daily. The significance of this finding for planning and policy making is not far-fetched. First, private water vendors who often draw heavily from 161 University of Ghana http://ugspace.ug.edu.gh municipal supply systems (Van Rooijen et al., 2008), and yet pay only a small fraction of their profit to the utility provider (Amankwaa, 2016), are enjoying a greater share of revenue which government could have benefited from directly. This is largely because of the discrepancies in relation to water consumption and per unit charge by GWCL, and the cases of flat rate billing which is common among households who sell water to their neighbours. Second, government, by improving access, can reduce expenditure on treating sanitation- related diseases. Another related but key concern is that during periods of water scarcity operators who depend on water such as the food and drink traders bear extra cost to obtain water for cooking. This reduces their profit margins and diverts human resource and working time to non-profit activity. In most cases, the long distances to the water sources lead women and children to spend much of their time fetching water at the expense of their businesses and schooling. Thus, large water-driven HBEs spend sizeable amounts of their income buying water and those whose operations do not allow them to spend time fetching water contract intermediaries to collect water at a cost. In the face of irregularities in water supply and for quality concerns, some operators resort to the use of sachet water for their food business (as captured in plate 5.3) which is quite expensive for those who operate on a small scale. The quote below by a middle-aged female chop bar operator in Abuja explains the situation better. I pay GH¢ 0.50 for each gallon of water, and in a day I use over 20 gallons for my food business so you can imagine the cost. …Even during hard times, I sometimes cook with sachet water. I call the truck driver to bring it here because customers think during such times we cook with borehole water. Even though I win their trust, this really affects my profit. But it is better than not cooking and losing your customers. In all the study areas, it became clear that operators mix or use household water for business operations. The operators say frequent interruptions in water supply, lack of space and storage media, poor quality of stored water, and the financial inability to procure separate water for 162 University of Ghana http://ugspace.ug.edu.gh household and business which is often associated with the misuse of water are responsible for this trend. This phenomenon explains the difficulty on the side of operators in accounting for their water expenditure. In most cases, food vendors tend to use less quality water sources for activities such as hand washing, and cleaning plates and cooking utensils. They also reuse water in order to be able to ensure environmental sanitation. The voice below buttresses this point: Most of us use the home water for our business. I am a tenant, as you can see the space here is too small to fetch water and store it for your business alone. It’s not easy. Especially when the pipes don’t flow I use the little water in the house for cooking and rather ask the children to go and look for some [water] elsewhere before they go to school. It is the money I get from cooking that I spend on their schooling (Participant, female focus group, Accra New Town). Plate 5.3: Purchasing of sachet water bags by cooked food vendors for their HBEs Source: Author’s field data, 2015 With respect to electricity access for business, the introduction of the prepaid system combined with meter sharing and the frequently unannounced power outages creates three key outcomes on HBEs: 1) increased expenditure 2) interrupted business activities 3) damage 163 University of Ghana http://ugspace.ug.edu.gh to electrical equipment. The analysis shows that 70.5% of HBEs jointly share meters with households or other enterprises while 29.5% have their own separate meters. In terms of cost, this translates into 74% of the operators spending on average between GH¢40 and GH¢400 on electricity monthly. They also spend between GH¢25 and GH¢800 as monthly expenditure on alternative sources due to power outages. The operators were emphatic about conflicts arising from the high charges accompanying meter sharing since bills are usually allocated and users are not privy to the details regarding consumption and the actual amount of credit purchased. It was evident that HBEs having their own separate meters relatively pay less electricity bill as they are able to control and monitor credit usage. Additionally, the rising cost of recharging prepaid credits and frequent power outages have a direct and tangible impact on informal activities in GAMA. For some HBE categories such as dressmaking, barbering, hairdressing, baking, and frozen fish and ice water trading, the lack of electricity supply results in complete operational shutdown which leads to loss of income. Although, some workers have devised coping strategies by travelling to nearby areas to access power or working overnight to complete a job order or buying iced-blocks to chill water and drinks such strategies come at extra cost which in the end affects profit margins. The implication is that customers also bear the brunt of the power situation as they end up paying more because of the cost of recharging prepaid meters and express delivery service by operators such as dressmakers who maneuver to meet deadlines. A 48-year-old single parent who doubles as a seamstress and provision shop owner in Accra New Town explains: Electricity is of concern to us, because most of the self-employed depend on it. I mean the hairdressers, provision shops and tailoring business. We the seamstresses work overnight to meet our deadlines. But now you can't predict when the lights will go off to plan your working schedule. You can be in your shop and the light will go off for a whole day but you still have to pay your rent and utilities. During Christmas, Easter and funerals customers end up fighting with us because you can try and try but you still can’t deliver. 164 University of Ghana http://ugspace.ug.edu.gh Moreover, the case of meter sharing by different HBEs particularly in the indigenous core areas is bedeviled with issues of overconsumption, uncertainty about next top-up and financial unpreparedness of some operators during top-ups, which negatively affect business operations and income levels. Similarly, operators join long queues to top up their prepaid meters to avoid running out of credit, which again takes time away from business activities. The few fortunate ones who have the financial means to acquire generators to keep their business running complain of profit loss owing to high overhead cost. The quote below by a 55-year-old timber and printing press businessman in New Town elucidates these issues surrounding electricity. You will be shocked when you go to the timber market area now, generator everywhere and the whole place is noisy. With 6 power cuts in a day, why won’t people see the need to get a generator to enjoy constant supply of power for their business. My nephew is a repairer, and he tells me all the money he is saving now is to buy a generator. Regarding the impact of power on employment and income, the voice of a middle-aged female university graduate who operates a fashion design business from the first floor of a storey building in Korle Gonno with a workforce of 25 workers, summarises the situation. We are managing the electricity situation. I have to put some workers on a shift system because we can spend the whole day here without power to work with. When I use the generator like this I spend a lot on fuel. For the past two months my husband has been supporting my business with some money meant for the home. Even at my boutique, I have asked 2 attendants to go home: and they don’t switch on fans and TV. It was striking to note that the power situation has necessitated a practice whereby customers make prior arrangements with operators to ascertain their availability and power situation before making any trip. To avoid transport cost and save time, most of such arrangements are done on the phone. The mobile phones, apart from enabling operators to stay in touch with their customers, also solve the issue of having to temporarily leave one’s business for making 165 University of Ghana http://ugspace.ug.edu.gh errands like top-ups and purchases. The voice below by a young beautician who perches at the frontage of her mother’s provision shop in Gbawe further explains these dynamics. My shop was closed for 2 days because of the continuous black-out. Now customers feel reluctant to come because they are not sure the shop is opened. Because of the power situation we have our phone numbers written outside the shop. Some customers call to get their hair fixed as early as 5am before the 6am black out while others come at 9pm after work. Anyway I am able to do this because I live just close by. With these dynamics, it is not surprising that some operators are establishing emergency phone charging facilities where people pay to recharge their phones to be able to stay active with their business partners. Tables 5.6 and 5.7 illustrate some of the coping strategies of certain HBEs and their accompanying cost. The experience of a young unemployed polytechnic graduate who helps in his father’s tailoring shop in Abuja reveals some of the challenges with the backup recharging centres. The mobile phone plays a central role in his father’s business since it allows virtual business deals with the retailers in Kantamanto and beyond. He reported: I used to send my battery to the charging shop nearby, but after a few times I realized that it would die very fast. When I asked the operator what could be the problem since I was using it for the same length of time, he responded that my battery was faulty. But I suspected that he was not charging it fully even though he charged me the full price. I decided to send it to the one in town and now my battery serves me longer as before. The only problem is I spend more time commuting to send it and also going for it. Finally, the analysis reveals that power interruptions damage appliances such as hair dryers, shaving machines, and freezers that are used for business. Sometimes the TVs and radios that are used to entertain and retain customers also get damaged. Commenting on how the situation is impacting businesses, a 34-year-old female provision shop owner who also sells cooked food to children in a nearby school in Korle Gonno exclaimed: 166 University of Ghana http://ugspace.ug.edu.gh We do have light but the dumsor [black-out] is too much. Sometimes up to 4 times in a day. It has destroyed my TV and fridge that I use in my provision shop to sell water, ice block and ice cream. I haven't fixed it because it will get damaged again. Another issue too is the prepaid. It runs too fast, and sometimes I have to rush and queue to top up leaving my shop with a caretaker. It's really disturbing. Table 5.6: Generator prices of HBEs that run emergency recharging centres Equipment/Cost (GH¢) Customer Settlement Phone Mobile Chargeable Tablet/ base per day battery phone lamp I-pad Korle Gonno 0.70 1.00 1.00 1.50 75 Accra New Town 1.00 1.50 1.50 2.00 105 Abuja 1.50 2.00 2.00 2.00 140 Gbawe 1.00 1.00 1.20 1.50 95 Source: Field data, 2015 N/B: Operators price the charging of a mobile phone over a battery because of the perceived inherent security they provide. Table 5. 7: Differences in electricity and generator prices of services charged by HBEs Cost (GH¢) Type of enterprise Electricity Generator Category Dressmaking shop* 50 70 Negotiable Tailoring shop* 40 55 Negotiable Hairdressing shop** 15 20 Negotiable Barbering shop 5 6 Fixed Electrical repairs shop*** 40 55 Negotiable Café 1.00 1.50 Fixed Printing and photocopy shop Fixed Typing (1 page) 1.00 1.50 Printing 0.70 1.00 Photocopy 0.50 0.80 Scanning 1.00 1.50 Entertainment centre Fixed Movie (full) 1.00 2.00 Game 1.00 1.50 Soccer (full match) 1.50 2.50 Source: Field data, 2015 * Cost of sowing a straight dress or shirt is estimated at GH¢50 and GH¢40 respectively ** Cost of fixing (perming and drying) a hair is estimated at GH¢15 *** Cost of fixing an electrical gadget (e.g. TV) is estimated at GH¢40 Besides the cost of repair of equipment, these happenings are shifting customers to other operators who have such facilities functioning. The quote below by a couple who run a 167 University of Ghana http://ugspace.ug.edu.gh drinking spot in the veranda of their family house in Gbawe Old Town elucidates the observation: I have realized that when the lights go off the customers who come here reduce, because there is another drinking spot in the next two lanes where the woman has bought a generator so she is able to serve chilled drinks and also play her sound system in her loud speakers. In this business it is the music that attracts and keeps people around. The longer they stay the more they drink and you get your money. Besides cost, interruptions in work and damages to appliances, the prepaid meters pose a lot of technical problems to HBEs. Notwithstanding such difficulties, technicians at ECG appear unresponsive to operators who complain. Most of them expressed frustrations with the situation where they have to contribute to transport ECG officials to service a faulty transformer or meter. In most cases they are forced to use private electricians because according to them ‘the value is the same’ and they are more responsive. These concerns of operators are reminiscent of the poor state of electricity access at the household level discussed in chapter five. In essence, the analyses and discussions have so far emphasized the impact of the interplay between formal and informal networks that are crucial to guarantee water and electricity access for HBEs. Such intricate and interdependent entanglement of the formal-informal, legal-illegal, and public-private dichotomies regarding the water and power landscape relates broadly to the concepts of disaggregation and co-production which have been highlighted in the literature on informality. For instance, Ahlers et al. (2014) highlight how a seemingly legal service connection in an officially established neighbourhood may incorporate obscure extra-legal components, such as resale to unconnected households. In such cases, informality is nested within formal service provision but in a subtle manner so as to not compromise the veneer of formality of the service (see Komives et al., 2005). In other words, although daily negotiations and struggles over water and electricity involve contestation, they also entail the 168 University of Ghana http://ugspace.ug.edu.gh request for a favour, or gift, or a temporary connection, or buying a small amount, agreeing on a delay in payment of the bill or top-up, organizing labour to fetch, mend, fix, or pay. With each act, social and business relations are created, consolidated, and contested (Ahlers et al., 2014), and the ‘urban service–entrepreneurial’ space is reconfigured. In sum, all of these factors invariably have implications on livelihood strategies and longstanding expectations of what household poverty should look like in poor urban settlements. Consequently, the next chapter turns attention to how the findings discussed so far relate to the poverty situation in GAMA. 5.4. Conclusion This chapter examined the role of water and electricity access in enhancing livelihood activities of low-income people in GAMA. The findings revealed that a wide variety of enterprises are run from the home and immediate surroundings, with the majority providing for the daily needs of the local economy. Shops selling basic essentials, preparation and sale of food, hairdressers, barbers, tailors and seamstresses tend to be the most common types of enterprise. In all, 85% of households engage in HBEs; 80% are electricity-dependent while 62% are water-dependent. Although the phenomenon of HBEs in GAMA is not recent, the number and geographical spread of the enterprises have increased significantly with the majority of them being compelled to move out of the home to satellite spaces such as alleyways and lorry stations. This was attributed to the infilling with new rooms in compound houses, clamouring for space in response to the growing saturation of HBEs, the shrinking public sector jobs and increasing informalization of the urban economy, which according to the World Bank (2015a) is central to the challenges of urban governance in Ghana. The findings revealed that a complex but well defined relationship exists between landlords or ‘space-lords’ and the operators in terms of working space and type of HBE. The study established that HBEs are not merely marginal economic activities but rather, they are 169 University of Ghana http://ugspace.ug.edu.gh important sources of employment and income. They offer two times more employment opportunities than businesses that exist outside the home. The study has demonstrated that the classification of income should not be merely in terms of type of products or services but should take into account locational and household traits. All of these have policy implications. Most significantly, the study revealed that the survival and increased opportunities of HBEs are often predicated on improved infrastructure and service provision. Because of the concerns about electricity access, few HBEs have acquired generators to keep their businesses running constantly and this is subtly shifting customers from poorer operators who cannot afford generators or top up their prepaid meters on time. Besides, it has contributed to the rising costs of goods and services and, more recently, the emergence of generator prices. As pointed out by Gough (2010) the frequency of HBEs varies not just between cities but also within cities, and within one neighbourhood will change over time, partly due to changes in the physical infrastructure. Similarly, Muhoro (2010) has reiterated how improved access to services does not only lead to positive impact on business performance, but also translates into improved quality of life and overall poverty reduction. The next section turns to these dimensions by examining the implication of these findings on poverty reduction. 170 University of Ghana http://ugspace.ug.edu.gh CHAPTER SIX THE NEXUS BETWEEN ACCESS TO SERVICES, LIVELIHOODS AND POVERTY REDUCTION IN GAMA 6.0. Introduction This chapter examines the implications of the research findings on the nexus between water and electricity access and household enterprises for poverty reduction in GAMA. The chapter first analyses how households themselves perceive poverty. It then discusses how households’ analytical expressions of poverty relate to income and access to water and electricity to establish the state of ‘income and service poverty’. Finally, the chapter examines the role of governance in improving service delivery and reducing poverty – ‘governance poverty’. It wraps up with a model that envisages how a functional decentralized governance system can contribute to poverty reduction through improved service delivery and livelihood activities. 6.1. Synergies between access to services, livelihoods and poverty in GAMA Accounting for how access to services impacts households and business operators requires considering how the outcome contributes to their poverty situation. This justifies the third research objective focusing on analysing people’s perceptions of poverty and how their perceptions encompass water and electricity access. In the literature, no simple definition exists for poverty as any definition has to be understood in its particular socio-economic and socio-cultural contexts (Lister, 2004). Saunders (2005) notes that how we define poverty is crucial as it is bound up with value judgements which have implications for research and policy. This chapter explores households’ definitions of the poor using an inductive approach to characterize and bring out dimensions of poverty that are important to poor people. This requires effacing personal prejudices and assumptions about the significance of particular factors and sectors in reducing poverty and about the best conceptual approach for 171 University of Ghana http://ugspace.ug.edu.gh understanding poverty. Hence, the organization of the following sections and the concepts used are determined by what emerged from the analysis of households’ definitions of poverty. The analysis is discussed by assessing the relevant arguments on the drivers of poverty. In other words, conventional knowledge of poverty is (re)examined and contextualised with the intention to comprehend how people perceive poverty in GAMA. In doing so, the study reports the views of respondents through qualitative and statistical analyses to contribute to debates on which factors influence poverty including perceptions of the role of governance in improving access to water and electricity. The study challenges existing conceptions of poverty and contributes to the debate in three thematic areas – income poverty, service poverty, and governance poverty. 6.1.1. Understanding households’ perceptions of poverty Don’t ask me to tell you about my situation [poverty], because you can see it around me. Just look at me, my children, the clothes we are wearing, and the food we eat, there is no future, no support. My small room has nothing valuable inside, no kitchen, no drains, things are just scattered. No one will tell you. So you write what you see, that is poverty (48-year-old poor single parent of 3 children, who is a food vendor in Abuja). At my age see yourself and compare with me, the clothes you are wearing even tells you the kind of job and opportunities you have. If someone sees you they will respect you, they won’t talk to you anyhow like you are nobody because of the small money or help they will give you (34-year-old male unemployed university graduate, who helps with his mother’s frontage provision shop in Accra New Town). The voices above suggest that poverty is a multidimensional social phenomenon which means different things to different people. Three main observations emerge here. First, the faces of poverty vary between and within settlements. Second, the causes of poverty vary by age, gender, education, household size, occupation type as well as other social and economic 172 University of Ghana http://ugspace.ug.edu.gh contexts including assets ownership status. Third, poverty has both income and non-income aspects, which include infrastructure service and general well-being. The multidimensionality of poverty was assessed by asking the participants to identify the factors that influence their poverty situation. The analysis revealed six key aspects: jobs (or income), water access, electricity access, sanitation facilities, education and governance. These six indicators were used as determinants of poverty, and they were assessed by asking the respondents to rank them in the order of priority. Table 6.1 displays the results for the four study areas. The data indicate that overall, access to job or a source of income was the highest priority indicator for defining poverty. Access to electricity and access to water were the second and third priorities respectively followed by sanitation facilities and education. Governance was ranked last as the indicator with the lowest priority: in most cases, residents linked it to their dissatisfaction with government’s inability to provide jobs and the poor delivery of urban services which is making life unbearable. In addition to these indicators, residents also point out housing and healthcare as related factors for measuring poverty. Table 6.1: Residents ranking of poverty indicators in GAMA Settlements Job/ Electricity Water Toilet/ Education Governance income Sanitation Korle Gonno 1 2 3 4 6 5 Accra New Town 2 1 3 4 5 6 Abuja 4 2 3 1 5 5 Gbawe 1 3 4 2 5 6 Overall 1 2 3 4 5 6 Source: Field data, 2015 Importantly, the overall fair representation of electricity and water access complicates the conventional knowledge of poverty, which is often linked to having jobs or money. A tailor in Abuja noted, “if you have a job but there is no light to work with how can you make money?”. This suggests that, defining poverty by income alone can be problematic as poverty also seems to be synonymous with an acceptable level of quality delivery of services which 173 University of Ghana http://ugspace.ug.edu.gh has a multiplier effect on businesses and income. Also, it is discernible that taking the household as a micro unit of analysis, access to electricity and water appears to be a private concern that residents have to manage themselves on a day-to-day basis. The findings illustrate that the first four indicators, jobs, sanitation, electricity and water access, have direct consequences for quality of life. A 51-year-old self-employed man in Korle Gonno indicated during the focus group: We work to have some self-fulfilment, but when you always spend your income on paying bills then you are at a standstill; it’s like you don’t see any change in your life. You become worried because you can’t acquire any valuable asset for the family. This clearly illustrates how people give meaning to life – having a sense of self-worth and avoid becoming trapped in the downward spiral of poverty (Sen, 1999). The spatial variations in the definitions of poverty within the research localities have many root causes and manifold expressions. For example, residents in the indigenous core areas of Korle Gonno associated poverty with the basic needs of life, whereas those in the central and peri-urban locations such as Accra New Town and Abuja, where the built-up environment is characterized by a migrant-mixed population, linked poverty to the lack of acquisition of assets and human dignity; equal choices and opportunities for living a dignified life. In Korle Gonno, the issue of not having enough to eat arises frequently and is captured in the following voices by some poor residents: Poverty is when in the evenings I eat kenkey, sleep, wake up in the mornings eat kenkey again. Sometimes I sleep hungry because I do not have kenkey at home. I wake up at night because of a stomachache and drink water to ease my hunger (19-year-old hairdressing apprentice, who is a mother of two children). Here in KG we don’t have much hope to live better with what we earn. There are many needs; food, children’s fees, and bills like water, light, toilet and rent. The family people also worry you with loans and other issues like funeral donations (52- 174 University of Ghana http://ugspace.ug.edu.gh year-old commercial driver, who shares a room with his three children in a family house). In the case of the peri-urban area, Gbawe, relatively different views were observed within the settlement between the lower and upper classes of the population. The poor households tended to cluster into the indigenous core area (Old Town) with particular spatial and demographic characteristics: multi-habited compound houses, communal living, single- headed and densely-populated households, and several HBEs. The middle class, however, are dominant and visible in the surrounding built-up and newly developing areas, with a lot of single family apartments and where the majority of them work outside the settlement. The quotes below exemplify the divergent positions; it appears that the common theme linking them is the cost of living, low salaries, lack of jobs, and lack of self-worth. The rich have one stable job but the poor are rich in many jobs. To be poor is to be straining your back always: you work hard yet you bring little home. You get tired but you are never satisfied. Because your little money can’t buy you anything. (45- year-old seamstress, who doubles as a foodstuff trader in Gbawe Old Town). There are people who live worse than I do. For some I am poor, for others not. But compared to my former state, I am just surviving. My grandfather once said, ‘if we fail to make our brains work, our body suffers’. But if I openly admit myself to be poor, life will become psychologically harder (38-year-old university graduate who runs a family fashion designing shop in Gbawe New Site). These quotes reveal how people tend to think about their economic position both synchronically and diachronically (Urminsky et al., 2014). Thus, they compare their situation with that of others, and they compare their present state with their earlier standard of living. Both are ways of trying not only to comprehend the changes in their social status, but also to psychologically mediate their experience and externalize the responsibility for the current state. 175 University of Ghana http://ugspace.ug.edu.gh In the more central locations of Accra New Town and Abuja, the lack of access to urban services, self-worth and motivation, and immediate cash needs interact to keep residents poor. In Abuja, the case is illustrated by the story of a 25-year-old migrant who shares a group accommodation with over 30 young males in a wooden dormitory-like structure. He sells fan milk products in a push-cart bicycle in Accra central on a work and pay basis; where he collects the products on credit and makes sales at the end of the day. Such pre-financing agreement is common among migrants, especially newcomers, but it is often accompanied by a collateral to the pre-financier (see Amankwaa et al., 2016, Amankwaa, 2013; Tutu, 2014). Even though living and working conditions appear precarious, as he recounted his good old days when he used to travel to Accra to sell his uncle’s local made shoes from Kumasi, he is optimistic and motivated about the future. He noted: Poverty is when you know good but you cannot do good: you know life but you can’t live it. It’s when you keep looking up to people and forget to keep going. It’s when you are not free, unhappy and fear what the future will bring. The quote shows how people point to general economic situations that impoverish everybody as a way to look beyond their personal circumstances. It brings to light issues of psychological effects such as insecurity from not knowing where the next meal will come from and shame at having to go without food or borrow from friends which tend to have strong symbolic value. In Accra New Town, the voices below typify and contextualize how households and business operators perceive poverty. Cash speaks, but the absence of it keeps you silent. You need to work to get cash; and most businesses here require good power and water supply. So poverty is when you don’t have a job or when you have it but don’t have water and light to work with. You can’t take care of yourself; pay your rents, bills and children’s school fees (36-year- old businessman, who serves as a secretary to a local NGO). 176 University of Ghana http://ugspace.ug.edu.gh Money is blood, and blood is life. So if you have money it’s like having blood in you; you can be mobile and be outspoken. You will not be stuck indoors and you can give donations at funerals and people will respect you. But if you don’t have money, it’s like you are dead (58-year-old timber contractor and a former assembly member). The quotes bring out the fact that poverty may not completely be the outcome of the lack of one thing but from various interlocking factors that cluster in people’s experiences and how they define poverty (Lister, 2004). The analysis reveals several cross-cutting issues, which suggest that the poor are not a homogeneous group, but experience poverty differently. They attribute their poverty condition to several factors depending on their backgrounds and personal circumstances. For example, typically, in the research areas, men associated poverty with a lack of material assets, whereas for women, poverty is defined as lack of food and money for daily living. Generational differences emerged as well. Younger men and women consider the ability to further education and earn income as the most important asset, whereas older people cite as most important the status connected to the acquisition of properties and living a more respectable and dignified life(style). Viewed more holistically, these interlocking dimensions of poverty are visible in the criteria or standards people use to define poverty and differentiate between categories of who is poor and who is better off (Narayan, 1999). Suffice to add that, there is not one universal account sufficient to structure a singular story(line) explaining poverty and the everyday life struggles of urbanites in GAMA. Nonetheless there are common threads that are woven through many experiences which are highlighted here. See Box 6.1 for households’ summary of poverty. To give this micro analysis a broader context, the study complemented the qualitative assessment with quantitative analysis. In doing this, the study analysed households’ income as a measure of poverty. It adopted both the Ghana Living Standards Survey (GLSS 6) and the World Bank’s definition of poverty as national and international poverty lines respectively to enable comparisons with studies elsewhere. The World Bank defines income poverty as 177 University of Ghana http://ugspace.ug.edu.gh living on less than $1.90 per day. In cedi terms, this is equivalent to about GH¢6.8412 per day (GH¢205 monthly). National income poverty lines vary greatly across the world, and in Ghana the GLSS pegs absolute poverty line at $1.83 per person per day ($1.10 for the extreme poverty line) (GLSS 6, p. 8). In cedi terms, this is equivalent to GH¢6.58 per day (GH¢198 monthly). Box 6.1: Households summary of poverty and wealth indicators Poor households  Family owned/inherited, or live in unstable and wooden houses  Lack adequate food and household care  Own no assets like land or TV and/or radio  Do not have easy access to water and electricity; use standpipes and share pre-paid meters  Are unable to save money or invest  Some have children who can’t go to school, or have to leave school prematurely  Depend on social relations for their sustenance or help Well-off households  Own solid and stable houses that are usually passed onto children and dependents  Feed their family well and provide good care  Own assets like land, cars, TV and/or radio as well as savings and investments  Have easy access to water and electricity; in-house pipe and separate pre-paid meter  Can send their children and dependents to school  Self-dependent and able to assist others Source: Interviews The proportion of households that are below and above the national and international poverty lines is presented in Figure 6.1. The distribution across the research localities is also displayed in Figure 6.2. It is clear that more than half (52%) of households surveyed in GAMA are below the national income poverty line (GLSS 6) while 56% are below the World Bank poverty line. This implies that five out of every ten households are poor according to the national and World Bank poverty lines. This finding confirms earlier studies (Coulombe and Wodon, 2007; Annim et al., 2012) that have found a growing spatial inequality and household 12 The cedi – dollar conversion rate used is $1 = GH¢3.6 178 University of Ghana http://ugspace.ug.edu.gh poverty in Ghana. Although the study areas display poverty levels relatively lower than the overall average, important differences can be observed. Using the national income poverty level, poverty incidence was much higher in the indigenous and informal settlements compared with the low-middle income and peri-urban areas. Figure 6.1: Households in GAMA below and above the income poverty lines Proportion of households in GAMA below and above national and international income poverty lines 100% GLSS 6 Income Poverty line = 80% 48% 44% ¢198 per head per month 60% World Bank 40% Income Poverty line = 52% 56% ¢205 per head per 20% month 0% GLSS 6 World Bank Below Income Poverty line Above Income Poverty line Source: Field data, 2015 Figure 6.2: Settlement distribution of households below and above the income poverty lines Proportion of households below the World Proportion of households below the Bank income poverty line across the GLSS 6 income poverty line across settlements the settlements Gbawe Gbawe 14% 15% Korle Korle Gonno Gonno 33% 31% Accra New Accra New Town Town 26% 23% Abuja Abuja 29% 29% Source: Field data, 2015 As shown in Figure 6.2, based on the GLSS 6 poverty level, Korle Gonno and Abuja recorded the highest average poverty rate of 33% and 29% respectively compared with 23% and 15% 179 University of Ghana http://ugspace.ug.edu.gh in the case of Accra New Town and Gbawe. Further, the trend in the proportion of households under the national poverty line does not vary much from what is observed under the World Bank poverty line, since Korle Gonno and Abuja again was the highest contributor to poverty incidence in GAMA (31% and 29% respectively). This result is consistent with the findings of Melara Arguello et al. (2013) which indicate that indigenous neighborhoods like Korle Gonno encounter many of the same dimensions of urban poverty as the more famed slums or informal settlements in the city: in this case Abuja. The study concludes that the income poverty rate recorded in the households sampled from GAMA is more than twice the national poverty rate of 21.4% reported by the World Bank (2015b) or the 24.2% reported by the GLSS 6 (GSS, 2014a). Put simply, two out of the four settlements had their rates of poverty incidence lower than the national average while the remaining half had rates higher than the national average. Four key observations are worth underlining here. First, the statistics reveal growing inequalities in household poverty. This complicates the account that Greater Accra is Ghana’s least poor region. It confirms the qualitative storylines of poverty described by households. Second, not all households in areas characterised by ‘poverty indicators’ are chronically poor. Although there is some clustering of poor households, they are often surrounded by others that manage to remain above the bottom line, or even rise out of poverty in some cases, signifying that geographical factor(s) are not wholly determinants of poverty. This observation is in consonance with earlier studies by Burke and Jayne (2010) in Kenya and also Annim et al. (2012) in Ghana that revealed that one’s poverty status and especially one’s ability to escape from poverty are not clearly defined by location. Third, income poverty has many dimensions and is characterized by low earning household enterprises, high cost of living, poor access to services, and high household dependency. The impact of the different factors combines to keep households, and sometimes whole communities, in poverty. Fourth, 180 University of Ghana http://ugspace.ug.edu.gh the findings show consistency with earlier studies which claim that there is rising intra inequality and household poverty in Ghana (Owusu and Yankson, 2007; Oteng-Ababio and Melara Arguello, 2014; World Bank, 2015b; Amankwaa, 2016); and indicates the need for inclusive governance in order to achieve sustainable national poverty reduction. In view of these observations, it is discernible that although Ghana’s economic growth in the past two decades accelerated poverty reduction, cutting the poverty rate from 52.6% to 21.4% between 1991 and 2012, sustaining poverty reduction requires a commitment to reducing inequality and improving access to opportunities for all citizens (World Bank, 2015b). As the World Bank (2015b) points out, the challenge is to ensure that prosperity is shared across the entire population and this must be addressed through a multifaceted, well-targeted and fiscally sustainable package of policies that balances the needs of the poor with that of the economy. So far, this section has outlined a focused approach to defining poverty in terms of the inability to live the kind of life a person values or to participate in society – involving both a low-income and a poor quality of living. Such formulation of poverty reminds us that income (or money) is just a means to an end and the goods or commodities it buys or services it renders are equally crucial aspects of achieving self-satisfaction. The next section explores the role of water and electricity access on poverty dynamics in GAMA. A detailed understanding of the nuances will give a better appreciation of the relationship between improved access and enhanced HBEs. 6.1.2. Impact of water and electricity access on poverty reduction Improving access to electricity and water supply has a multiplier effect and positive relationship with poverty. This is because improved reliable, quality and affordable access enhances business performance, affordability of services, acquisition of assets and living a dignified life open to opportunities and choices. In other words, poverty is also about consumption of state-provided infrastructure and services – ‘commodities’, or what some 181 University of Ghana http://ugspace.ug.edu.gh scholars refer to as ‘social wages’ (Baulch, 1996). Throughout the study, households and business operators emphasized the importance of electricity and water services. A housewife who runs a small beauty parlor in Accra New Town noted: “Water is life, but because we have no water life is miserable. …No light means no life, because you are alive but dead: you can’t work, you can’t watch TV and entertain yourself… All you do is to sleep all day”. To be able to place the qualitative interpretations of the relationship between access to services and poverty at the household scale within a wider scholarship, the study statistically measured the extent to which water and electricity access actually impacts on enterprises and poverty in GAMA. On a five-point Likert scale where 1 = Strongly disagree and 5 = Strongly agree, business operators were asked to rate their responses to the statement ‘There is good access (proximity, reliability, quality and cost) to water and electricity for your household business’; ‘Improved access to water and electricity positively affects your livelihood and poverty’. The responses were scored and recoded into ‘Poor/Low’ for business operators who selected the options ‘Strongly disagree’ and ‘Disagree’ while the responses of those who indicated ‘Strongly agree’ and ‘Agree’ were recoded as ‘Good/High’. However, operators who selected the option ‘No idea/Indifferent’ were recoded as ‘Average/Moderate’. The results are illustrated in Figures 6.3 to 6.6. The majority of operators whose HBEs depended on water, whether private or public sources, were dissatisfied with the current level of water supply in relation to quantity, quality, reliability and cost. The result displayed in Figure 6.3 shows that 64%, 35% and 1% of the HBE operators rated their water access for business as ‘Poor’, ‘Average’ and ‘Good’ respectively. In measuring the impact of water access on poverty, 84%, 12% and 4% of the operators indicated that poor access to water has ‘High’, ‘Moderate’ and ‘Low’ impacts respectively on their livelihoods and poverty levels, as shown in Figure 6.4. 182 University of Ghana http://ugspace.ug.edu.gh Figure 6.3: Assessing water access for HBEs 180 70% 160 64% 60% 140 156 50% 120 100 40% 80 35% 30% 84 60 20% 40 20 10% 3 1% 0 0% Poor Average Good Scores Percent Source: Field data, 2015 Figure 6.4: Impact of water access on HBEs and poverty Impact of water access on household livelihoods and poverty 250 90% 84% 80% 200 70% 60% 150 205 50% 40% 100 30% 50 20%12% 10 4% 10%28 0 0% Low Moderate High Scores Percent Source: Field data, 2015 Likewise, the majority of electricity-driven HBE operators were dissatisfied with the current level of electricity access, with particular emphasis on quality, reliability and cost. Figure 6.5 illustrates that 70%, 21% and 9% of the operators rated their electricity access for business as ‘Poor’, ‘Average’ and ‘Good’ respectively. In quantifying the impact of electricity access on poverty, 85%, 11% and 4% of the operators indicated that poor access to electricity has 183 Frequency Frequency Percentage of HBEs Percentage of HBEs University of Ghana http://ugspace.ug.edu.gh ‘High’, ‘Moderate’ and ‘Low’ impacts respectively on their livelihoods and poverty levels, as displayed in Figure 6.6. Figure 6.5: Assessing electricity access for HBEs 250 80% 70% 70% 200 227 60% 150 50% 40% 100 30% 21% 20% 50 67 9% 10% 28 0 0% Poor Average Good Scores Percent Source: Field data, 2015 Figure 6.6: Impact of electricity access on HBEs and poverty Impact of electricity access on household livelihoods and poverty 300 85% 90% 80% 250 274 70% 200 60% 50% 150 40% 100 30% 20% 50 11% 10% 14 4% 34 0 0% Low Moderate High Scores Percent Source: Field data, 2015 In short, poor access to services impacts 85% of electricity and water-dependent enterprises with negative outcomes on household poverty levels. Thus, a direct and positive relationship exists between access to services and poverty reduction, such that an increase in the cost of accessing services, due to poor service delivery, results in an increase in the cost of doing business which translates into poverty. These findings support the research propositions that 184 Frequency Frequency Percentage of HBEs Percentage of HBEs University of Ghana http://ugspace.ug.edu.gh poor access to services has negative repercussions on businesses, and poor business outcomes contribute to household poverty. Drawing on the income poverty analysis, the study assessed the ‘service poverty’ level of households in GAMA. In doing this, the GLSS and World Bank poverty lines were juxtaposed vis-à-vis households that use private water sources and the shared pre-paid electricity meters, the rationale being that the majority of HBEs captured in this study rely on private sources of water and electricity connections, and these modes of access come at an extra cost as compared to households having direct provision from the utility providers, GWCL and ECG. The result is presented in Figure 6.7. Figure 6.7: Service poverty level in GAMA using the World Bank and GLSS poverty lines Water and electricity service poverty level of households below the income poverty lines 80 69 70 70 60 51 50 50 40 30 20 10 0 GLSS 6 World Bank % Households Using Prepaid Power Source % Households Using Private Water Sources Source: Field data, 2015 For all households below the World Bank and GLSS poverty lines, 70% of them access electricity through the shared pre-paid meter system while 50% access water through private sources. Thus, with regards to ‘service poverty’, half of households are categorized as ‘water poor’ whereas more than two-thirds are categorized as ‘electricity poor’. The findings evoke special attention because, as earlier emphasized, private mode of access to water and electricity is associated with several concerns – lack of trusted sources, poor quality, waste of productive time, higher costs, inability to monitor users and credit usage, meter failures and 185 Percentages (%) University of Ghana http://ugspace.ug.edu.gh technical faults. Such concerns result in increased business operation cost, diversification of businesses, loss of jobs and redundancy among workers, and people’s inability to live meaningful lives. The findings confirm the research propositions that improving water and electricity access in terms of quantity, reliability, quality, and cost has a positive impact on HBEs and poverty reduction in GAMA. This assertion is further corroborated when the business operators were asked to rate their responses to the question ‘To what extent do you think improved access to water and electricity can enhance your livelihoods and reduce poverty’? It was striking to find that the majority of operators (93%) who depended on water and electricity, whether on private water sources or shared pre-paid connections as well as those who presently have direct supply from the utility providers, were optimistic that improving access to services would have a positive (55%) and moderate (38%) impact on their livelihoods and thereby reduce their poverty levels. The result is displayed in Figure 6.8. Figure 6.8: Perception of the impact of improved water and electricity access on poverty Perception of the impact of improved water and electricity access on livelihoods and poverty 140 60% (55%) 120 50% 132 100 (38%) 40% 80 91 30% 60 20% 40 (7%) 20 10% 18 0 0% Negative Moderate Positive Scores Percent Source: Fieldwork, 2015 To complement these quantitative observations, the qualitative analysis further revealed that community and household poverty are often connected to infrastructure service provision. As 186 Frequency Percentage of HBEs University of Ghana http://ugspace.ug.edu.gh Charmes (2000) points out, the relationship between informality and household poverty becomes evident only when people’s voices and intra-household dynamics are analysed. For example, in Abuja, participants in the focus groups indicated that every resident is poor precisely because of the absence of government provided basic services such as water, electricity, sanitation, and other necessities like school, health facility, and security in the settlement. Specifically, some residents also made a distinction between community and individual-household poverty, whereby the latter refers to the ability of an individual or household to cope with the rising cost associated with accessing services from private vendors (see Amankwaa, 2016). A middle-aged migrant who operates a chop bar bemoaned: Every day you have to share your little money with the shower operators to get water. The prepaid meter runs like phone credit, every other day the woman [meter owner] asks us for money. We pay for something you can’t tell how much you have used. A noticeable observation in the remaining study areas was that, residents pointed out how particular segments of the settlements were identified as the most isolated and located farthest from key services. In Gbawe (Old Town) for example, old women define people in poverty as “those without drinking water”. The implication is that such sections of the settlement spend more time, energy and money to actually access services, as captured in the quote below: Rich people are those who have piped-borne water in their houses. They have big poly-tanks and you don’t see their children carrying gallons in long queues, they don’t sleep in light off and they use generator for their home business. It was further noted that when respondents complained and refused to pay for these services officials responded, “the power and water sometimes come in the night and the meter reads”, implying that residents have to pay for services they have not actually used. In most cases this explains why some households refuse to pay their utility bills and end up getting disconnected. 187 University of Ghana http://ugspace.ug.edu.gh Additionally, differences in perspective between the youth and elderly were discernable. For example, historical comparisons were frequently made by the older population, usually aged 40 years and over, who often looked back on the post-independence era and the early 2000s with nostalgia, loss, and regret. Comparing their current situation with that before the early 2000s, a 63-year-old pensioner who lives in a family house in Korle Gonno recounted: …Then there were not much differences between people. There was no poverty, the pipes were flowing and the lights were always on. Businesses were doing well. Things were moderate and we did not owe anyone anything. It is interesting how residents are able to recount and connect their present circumstances to broader political economy issues such as national governance. Anaman (2006) reports how the government’s structural adjustment programme, embarked on from mid-1983 to 1991, although modestly successful with increased growth rates (due to increased total investment) also led to increased poverty particularly among the lower and middle classes. Hitherto (from 1966 to 1983), governance was marked by military rule, eroding human rights, public sector debt and corruption, economic hardship, and a cash and carry system (where patients had to pay upfront before receiving health care) (Anaman, 2006). However, the period from 1992 to the millennium era has witnessed a turn-around with political stability, democratic governance, economic development, and social intervention policies and programmes, albeit with mixed outcomes of prosperity and sustainability when carefully analysed. Among the youthful population, especially those who engage in food processing and sale, hairdressing, dressmaking, as well as the aluminum fabricators and electricians, access to quality and dependable water and electricity emerged as a major factor defining poverty. Out of necessity, households which engage in these daily wage jobs, with no guarantee of supply from public utility providers, pay more to the informal providers to keep their business ongoing. A youth leader during the male focus group in Gbawe stated: 188 University of Ghana http://ugspace.ug.edu.gh The poor are those with widespread absence of the public utility provision. They often have numerous small, dangerous and illegal sources of water and electricity and job rather than a single source. Stepping back and looking at the bigger picture, these dynamics are occurring at a time where the economic climate is characterized by declining job opportunities and overcrowding in the informal sector due to the rising number of people trying to carve a niche in this sector. Government is therefore expected to bolster up efforts aimed at creating the required business climate for household economies to thrive, one of which is improving business infrastructure like electricity and water supply. In this respect, the case of a security man from Accra New Town who suffered retrenchment at Fan Milk Company Limited during the peak of the load shedding exercise readily comes to mind. His marriage failed soon afterwards and he was abandoned by his wife and children from the family house. He now stays with a friend in a wooden kiosk and finds work (weeding, cleaning) that earns him a free meal, or sometimes a bit of money. In another case, an unemployed teenage mother who lives in a container store by the roadside with her two children in Accra New Town stated: The water and electricity situation is affecting us. The light comes during the day but it is in the evenings that the shop gets busy after people have closed from work. The small tips from my madam on a good day have stopped coming. She has laid some workers off and put some of us on part time, it’s not her fault, because we can sit at the shop for almost a week and there will be no work to do. I find it uneasy staying home for people to see I now work on shift basis so I leave at my regular time in the mornings and sometimes come home late when the children are sleeping like I worked overtime. But all the busy schedules do not reflect in my pay as sometimes I even find it difficult bringing food home for the children. The above voices and experiences support the claim that poor quality service provision translates into the crippling of businesses and retrenchment of workers which increases unemployment and poverty. In some cases, workers are asked to work on a shift-basis, as 189 University of Ghana http://ugspace.ug.edu.gh part-time workers, to reduce business overhead costs. The findings tie in with Danso (2007) who indicates that the need for these business services becomes critical if HBEs have to compete at the immediate local area and community level where people from outside the settlement, with different economic backgrounds and capital, also have their businesses. In essence, the study has provided compelling evidence to support the argument that when household connections are reliable and affordable, the cost and time spent obtaining water or electricity pre-paid units are significantly reduced and household members can engage in more productive activities (Kariuki and Schwartz, 2005). Improvement in quality and affordable utility services can lower costs and expand market opportunities for businesses and, thereby, increase productivity and investment that drive economic growth (de la Fuente, 2004). For example, multi-country studies in Latin America and Africa report that although the growth of output attributable to improvements in infrastructure and services is relatively small (0.15 – 0.5%) it is associated with significant reductions in poverty (Estache, 2002). On the contrary, underinvestment in infrastructure and services reduced annual growth by 1–3% (Estache, 2002). In sum, besides income, nearly half of the recommendations from households and business operators regarding poverty alleviation centred on the need to improve the delivery of basic services by government. The next section assesses the role of governance in ensuring improved delivery of services to boost the performance of household economies. 6.1.3. Role of governance in improving service provision for poverty reduction Government is like a parent which should take care of its children. But what kind of government do we have? One hand gives and the other takes away (Resident, Gbawe). Understanding the multidimensionality of poverty brings to light the significance of exploring the causes of household poverty, and addressing settlement-specific problems based on the perceived needs and priorities of the poor. One essential pre-requisite of institutionalizing this 190 University of Ghana http://ugspace.ug.edu.gh approach to understanding poverty, however, is the existence of a decentralized governance system (Hansson et al., 2014). In this study, decentralized governance is defined as the negotiated outcome of interacting actors – users, volunteers, community groups and professionals – as co-producers. Recognizing that each of these groups has a quite different relationship to the service sector can introduce participatory development in which citizens at the grassroots are involved in planning, formulating and implementing projects for themselves. To demonstrate this, the section analyses both residents and stakeholders’ perceptions of government’s role in reducing poverty through improved service delivery within the context of Ghana’s decentralized local government system. Drawing from the previous understanding that poverty equates with household and individual opportunities and tendencies, here we extend poverty as a phenomenon associated with broader political and economic processes where government is expected to provide for the welfare of all. The quote below by a 58-year-old opinion leader in Gbawe underscores the point: Before, people ascribed poverty to laziness and incompetence, but I think it has to do more with structural forces; the strings that pull and limit how far you can go, what you can do, and determine what opportunities are available to you. In Accra New Town, this perspective was captured by a middle-aged woman, who operates a public shower and a video and game centre at the timber market, as follows: I feel very uncomfortable when you ask me how I would describe government’s role in my present situation. I cannot, but I do know the government is the main cause for our poverty. They make the decisions and manage our national purse. These quotes highlight how public perception of the government’s role in the provision of services results from a complex combination of factors. There is widespread dissatisfaction at what is perceived as incompetence and mismanagement of state resources that has caused, in some cases, total neglect and inequality in the distribution and provision of basic services. During the interviews, a retired banker from Gbawe indicated: 191 University of Ghana http://ugspace.ug.edu.gh Now we have oil but things are the same. Government is building schools and roads, but without the water and electricity how can we work and be able to pay our children’s fees or pay for our transport to use the road to the marketplace? Residents believe government officials have over the years failed to allocate funds to the water and electricity sectors and this lack of prioritization has triggered the longstanding challenges in service delivery. One of the youth leaders pointed out during the focus group in Abuja: Governments keep shifting blames. We are cleaning the mess by the previous government. They politicize everything, that’s our bane. Politicians don’t care about the suffering masses, because they have mechanisms to enjoy 24/7 access to services. Business operators alluded the situation to government’s failure to recognize the contribution of household businesses to the local economy. A female food vendor underscored this point during the elderly focus group in Korle Gonno: “If every household was doing some business, in one way or the other, the unemployment pressure on the government will reduce”. For these reasons, many residents express concern at what they perceive as ineffectiveness in the decentralized governance system, particularly at the local assembly level, which is expected to be the first point of call for addressing community development issues, and ensuring government’s projects and programmes are properly implemented. This concern is reported by a graduate teacher who lives with his family in a family house in Korle Gonno as follows: When the pipes do not flow and you tell them [assembly members], they tell you they will report it. Even if you report a broken pipeline, faulty street light, or an illegal connection, there is little they are able to do. We thought the assembly members were agents of change here but they are like us, their voices do not reach far. This quote highlights issues relating to decentralized urban services such as fiscal devolution, bureaucracy, maintenance culture, interactive interface and self-organization and community 192 University of Ghana http://ugspace.ug.edu.gh ownership which have been emphasized in the wider literature as key aspects of governance in practice (Connell, 1997; Obeng-Odoom, 2013; Oteng-Ababio and Amankwaa, 2014). In terms of context, the ways in which people define ‘governance poverty’ in the research localities varies according to a number of factors like personal circumstances, economic status, and gender which in most cases was related to the position of respondents in the socioeconomic ladder before or after the social and economic transformations of the late 1980s through the early 1990s and 2000s. In Accra New Town, most residents recounted circumstances where people gained access to services through connections and often paid for favours with a gift. A retired public servant, now a unit committee member and a part-time teacher, who lives in a family house where he also works as a water vendor noted: Recently, at the water company office I was asked to pay some money and they will bring me flat bill every month. This was after I complained about the rising bill. The officer in-charge was my former student so he was trying to help; it’s your personal and market relations. Participants during the focus groups indicated how the practice of offering gifts to people in authority has become an accepted and expected way of facilitating water and electricity connection processes and getting things done. An opinion leader noted: “If you don’t commit something the field officers won’t come”. Two observations are clear here. First, the practice has become more pervasive and extreme, posing hardships above and beyond what most people can afford. Second, it is contributing to the marginalization and exclusion of the greater majority of residents whose service needs are not adequately catered to and efficiently addressed. These findings point to issues of nepotism and corruption which are pervasive in Ghanaian society (see Agbele, 2011). Likewise, in Abuja while the informal status of the settlement excludes residents from enjoying government-provided services, this position was understood by residents as deliberate neglect and politicization of the governance process, which hampers the freedom 193 University of Ghana http://ugspace.ug.edu.gh and quality of life of people. A middle-aged businessman who is a member of a local NGO commented: Are we not part of the city? Don’t people here work? Don’t we pay daily tolls to AMA? Don’t we pay to get water and electricity? So what is the point that those here are not officially recognized? Don’t we vote? Tell me, during elections don’t they [politicians] come here? Governance should be inclusive; different solutions for different people. WEhat makes here informal? Is it not about the land tenure? Can’t government arrange to supply us with services? Here is a busy centre but we can’t develop the area because you don’t know what tomorrow brings. We can’t live freely! Despite AMA’s policy of not providing services to informal settlements and buildings without land tenure security, the early settlers and local elites who wield influence have managed to connect pipelines and an electricity substation in the settlement. Subsequently, they act as private ‘service-lords’ – shower managers and meter connection providers – who supply services to households and businesses. Such self-help approach to improving residents’ access to services are common in low-income societies in African cities as reported in Dar es Salaam, Tanzania by Andreasen and Møller-Jensen (2016) and in Jos, Nigeria by Daniel (2014). In Korle Gonno and Gbawe the situation is similar, especially in the indigenous core areas. For many households and business operators, their lack of participation in designing, deciding, and implementing government projects underpin the failure of the local government system, with respect to service delivery. A self-employed computer technician in Gbawe emphasized: “We must decide what we want, where we want it and how to manage it; that is local governance”. Residents believe grassroots involvement in development issues such as the siting and management of water pump station, community water tanks, as well as electricity substations and vending operation centres is critical to achieve an interactive decentralized governance interface. Yet, throughout both settlements people complain about the lack of representation, slow response and humiliating treatment from officials in their 194 University of Ghana http://ugspace.ug.edu.gh attempt to get their voices heard and demand for improvement in service delivery. Many become discouraged by bureaucracy and unresponsive officials, as well as withheld information. An opinion leader in Korle Gonno retorted: “Is it government to the people or people to the government”? He further added: You have to queue a lot and go from one desk to another… The officers are unfriendly [and] do not inform as they should. All they say is go there, go and come next month. You can spend more than a month chasing after them to get your issue addressed. So if you are not patient or you don’t have time you may not return. A 62-year-old pensioner from Ghana Commercial Bank who is an opinion leader and serves as a board member of a local NGO in Gbawe noted: Our leaders announce they are going to commission a new borehole and then leave us to the mercy of fate, not asking whether we are prepared to accept that source of water and at what price we are prepared to pay. It is the same way they did with the transition to the new pre-paid meters and even with the relocation to the new market. …You heard the government intends to bring prepaid meters for water? That can’t happen at all, we must vote against it. For electricity, you can delay reloading your prepaid meter but water is different because the water you drink affects your health. You’ll be killing the poor people if you do that. In fact, water should be free for all. The evidence provided regarding the ‘governance poverty’ situation in the four research localities connect to wider issues of disconnect between leaders and the people – the phenomenon of poor urban communities and municipal interface, which has been articulated by Owusu and Afutu-Kotey (2010) in their comparative study in Accra and Sekondi- Takoradi. This has resulted in the weak participation and poor provision of services which is manifested in the rise in illegal connections, estimated billing, unattended to broken pipes and faulty meters, and outstanding bill debts. Stakeholders, including government officials responsible for service provision, assembly members and NGOs alike were in favour of the need to improve GAMA’s public services through a hybrid decentralised governance system. According to them, this may trigger 195 University of Ghana http://ugspace.ug.edu.gh positive multiplier effects on designing poverty alleviation interventions. For instance, commenting on the elements of good governance, a development consultant and a private businessman who is the secretary to an NGO in Accra New Town noted during an interview: Effective accountable governance is vital to building inclusive, prosperous, and safe cities. Effective decentralized governance requires policies that work towards ensuring all citizens, especially the poor and other vulnerable groups, have access to basic services: policies and practices that prevent corruption, through accountability and transparency. It also involves efforts that encourage effective stakeholder participation in governance by the private sector, NGOs, CBOs and the poor. In a related point, a development planner at the AMA’s Town and Country Planning Department (TCPD) pointed out that: A central feature of a developmental state is for government to ensure citizens have equal access to improved basic services by encouraging broad stakeholder participation. However, factors such as rapid population growth, uncontrolled development leading to informal settlements and structures, and ineffective monitoring on the part of some government officials largely due to inadequate funds have contributed to the missing links in the current decentralised governance system and the resultant poor provision of services being witnessed. This perspective is reiterated by a governance expert at the Ministry of Local Government and Rural Development (MLGRD) as well: Now in most developing countries, governments are encouraging PPP [public private partnerships], as the private sector actors are able to mobilize enough funds and extend access to remote and underserved areas to complement government’s efforts. …The problem is with pricing, but this is why government’s role as a mediator and regulator in any PPP must be clearly articulated and followed to the letter. In contrast to this view, a governance expert at the National Coalition Against the Privatisation (NCAP) of water who doubles as a co-founder of an NGO commented: The issue of high cost associated with the private sector is because PPPs lack a people centred-approach. Money rather than people is often the motivation. If we talk about 196 University of Ghana http://ugspace.ug.edu.gh PPPs, it should involve public, private and people partnerships. If we leave the ‘people’ part out, then private actors will always take advantage of the masses. The above viewpoints suggest how in practice the tenets and approaches of PPP are fundamentally inconsistent with inclusive metropolitan governance. It resonates with Obeng- Odoom’s conceptualisation about the possibility to “re-interpret and change public-private partnership to public-people partnership”; a model that replaces the private sector with the ‘people sector’ in a continuum (2012a, p. 1143). Meanwhile, this study shows that realising the ‘people’ component in the ‘public’ and ‘private’ as a tripartite partnership rather than replacing the private sector with the people sector seem more laudable in the Ghanaian case. These viewpoints link up with the concepts of co-creation and co-production which focuses on the systematic pursuit of sustained collaboration between government agencies, non- government organizations, communities and individual citizens. Such conceptualizations place the citizen at the centre of policymakers’ considerations, not just as target, but also as agent (see Mistra Urban Futures, 2015; Smith et al., 2016). The stakeholders’ analysis reveals that, not only are there differences in how stakeholders perceive governance vis-à-vis urban services and poverty reduction but these differences are also differentiated by communities. For example, the assemblyman for Korle Gonno argued that the underlying reasons for the disjuncture between decentralized governance in theory and actual existing urban governance in practice can be traced to the lack of effective coordination and revenue mobilization on the part of metropolitan authority to encourage wider consultations and ensure effective community sensitization and support. He noted: The challenge confronting local assemblies is because we [government officials] do not involve the local people in the process of designing and deciding community projects, and so we find it very difficult defending its implementation and also encouraging participation or adherence. Day after day I have to deal with apathy from residents. Some feel they are not enjoying their portion of the national cake. 197 University of Ghana http://ugspace.ug.edu.gh The assembly member whose jurisdiction includes Abuja bemoaned the negative perception by the general public and the consequent ‘I don’t care’ attitude exhibited by residents in the settlement. She emphatically stated: We need to continue dialoguing with government to discuss the practical ways that can be adopted to address the issue of the lack of provision of basic services to the people. It’s tricky; how can we implement smooth decentralized governance without involving the people. …After all they know the area well, they know the issues, and they know houses with illegal connections and broken pipes... They know how and when the services are most needed. These differences emerge because urban development challenges are complex and of concern to different sectors, actors and cultures, and therefore responses and solutions need to be found through cooperation and collaboration. Accordingly, governance should provide an arena for co-creating knowledge and understanding that promotes sustainable urban futures (Polk, 2015). The policy makers’ analysis found that ECG and GWCL officials were aware of the challenges bedevilling GAMA’s water and electricity delivery system. However, they were unanimous in cautioning against any hyped expectations in an overnight solution as the government was instituting restructuring and cost recovery mechanisms to improve delivery and ensure value for money. There was a significant level of consensus among policy makers about the need for the public to be moderate and conservative in their use of water and electricity, and show more commitment to paying bills and reporting leakages and illegal cases. As to whether the utility companies were ready to support any targeted profitability initiative that seeks to ensure regularity and improvement in service delivery, through a process of mapping and clustering households which engage in water and electricity- depended enterprises, an official at the GWCL responded: The current policies and tariff regime do not support or encourage us to enrol any system that is based on settlement or sector specific dynamics like the home-based 198 University of Ghana http://ugspace.ug.edu.gh enterprises you are talking about. Even though they are residential based, such small- scale businesses are supposed to be captured under the commercial billing rate system. The idea of using residential enterprises as a middle ground for reducing poverty through improved service delivery was underscored in the policy circles at the Public Utility Regulatory Commission (PURC). One of the officials at the policy unit stressed: One surest way for the utility companies to ensure cost recovery is by improving service delivery among the core densely populated areas. Because such areas demonstrate high density and the majority of the households engage in some form of small-scale businesses. That means, apart from the numbers people are able to mobilise funds to pay their utilities once their jobs are doing well. Although the evidence suggests that policy makers do not have any immediate policy intervention aimed at targeting household businesses, they were unanimous when it came to the contribution of such businesses to household economies considering its growth, versatility, and longevity. This study asserts that the call for a targeted policy may be worth (re)considering in the future in view of the fact that recently there have been calls by the Association of Ghana Industries (AGI) and National Board for Small Scale Industries (NBSSI) on the government to review the current utilities tariff regime which makes industries subsidise for residential users. The stakeholder analysis presents a challenge for socio-economic development and poverty reduction in GAMA, as effective citizen engagement through hybrid decentralized governance are prerequisite for improving urban infrastructure and service delivery (Adarkwa, 2012). In expounding this nexus between governance and poverty, Rondinelli et al. (1989) and Adarkwa, (2010) point out that the provision of infrastructure services is fundamental in enhancing labour productivity, ensuring the effective functioning of markets, and the creation of opportunities for employment and entrepreneurship which leads to improvement in the income status of the citizenry. More significantly, philosophical understanding of neoliberal ideology suggests improvement in infrastructural services such 199 University of Ghana http://ugspace.ug.edu.gh as water and electricity is a means to facilitate production, capital accumulation and promotion of individual freedom through availability of alternatives for choice making (Obeng-Odoom, 2016). Thus, good access to water and electricity contributes to poverty reduction (World Bank, 2015a, b), through innovative and tailor-made policy interventions. Recognizing this dilemma, the next section summarizes the findings and discussions presented so far into a conceptual model that explicates the synergies, strengths and successes that can be leveraged in a hybrid decentralized system that focuses on poverty reduction through improved service delivery and livelihood activities. 6.2. Towards effective governance for poverty reduction – a conceptual model Based on the analysis of residents’ and stakeholders’ perceptions of the role of governance in poverty reduction this section presents a framework of the interconnections between governance, provision of urban services, livelihoods and poverty. The section highlights how good governance provides opportunities for citizens to participate actively in shaping their world (Holmes, 2011). This perspective is fundamental to any discourse on engaging citizens in policymaking and in designing and providing urban services (Smith et al., 2016). Undoubtedly, as the stakeholder interviews indicates, such participatory approach can be effected in multiple ways and at various levels, from informal local and community settings, through incorporated entities, NGOs and CBOs, to key government institutions and public officials. Thus, the theory and practice of decentralized governance is increasingly concerned with placing the citizen at the centre of policymakers’ considerations, not just as passive targets, but also active agents whose agency matter and whose right to participate directly or indirectly in decisions that affect them should be actively facilitated (see Polk, 2015). Such an approach honours the fundamental principle of a decentralized governance system – where citizens’ welfare, empowerment and overall socio-economic development is the primary goal. 200 University of Ghana http://ugspace.ug.edu.gh In many governance systems, particularly in developing countries, citizen participation in policymaking and service design and provision has been debated and poorly attempted with mixed outcomes (Holmes, 2011). Although there have been some notable achievements, there is extant literature and policy briefs advocating for a methodical, in-depth and bottom-up collaboration (Bardhan, 2002; Shah, 2006). Meanwhile, critics of good governance have argued that genuine engagement in the design and co-production of policy and services necessitate major shifts in the structure, culture and operations of government agencies and local assemblies (McFarlane and Desai, 2015). The value of this perspective can be realized in two ways. First, one that demands of citizens to orient to the public good, and demonstrate the willingness and capabilities needed for effective participation. Second, one that demands of government officials and public servants’ new skills and orientation as enablers, negotiators and collaborators (Holmes, 2011). For the most part, populations in low-income settlements and certain groups within them are disengaged, disconnected and disenfranchised. Ghana is not an exceptional case. As the analytical discussions have so far revealed the pursuit of an active citizen-centric urban service engagement demands political will and commitment for the actual devolution of power and decision-making to frontline government officials and technocrats – and to the citizens and stakeholders with whom they engage (Owusu, 2012). In other words, recognizing how the missing links in governance translate into real life outcomes is crucial, thereby justifying strengthening the linkages between governance and poverty reduction. Using the water and electricity scenario presented in this present study as an exemplary case study, and illustrating a more nuanced analysis, the study proposes a conceptual model that emphasizes the nexus between effective decentralized governance, improved urban services, enhanced livelihoods and reduced poverty. Figure 6.9 illustrates the model. At its broadest, the model frames the process as a cluster of interlocking components of the public provider (state), public user (citizen not consumer), and household business (middle 201 University of Ghana http://ugspace.ug.edu.gh player) – with overlapping dimensions, as shown in Figure 6.9. Underlying and connecting this trilogy is a hybrid decentralized governance system which adopts a co-production viewpoint in policymaking and urban service design and provisioning. In the ordinary sense and at a more prosaic level, the model portrays decentralized governance as the substructure whose commitment is manifested in practical initiatives that various local governments have pursued to ensure citizens’ involvement in decisions and development issues that affect them. Figure 6.9: Framework on governance, urban services, livelihoods and poverty reduction Household increased access/affordability: Feedback on Utility cost recovery quality of service Urban services: water and electricity Enhanced livelihoods Reduced poverty Equity, inclusivity Quality of life and accountability Home-based Poverty reduction: enterprises citizens not consumers Hybrid Metropolitan-community decentralized interface: report broken Improved service delivery: governance pipes, faulty meters, illegal quality, cost, reliability connections, etc. Linking governance to improved services, enhanced livelihoods and poverty reduction Source: Author’s own construct, 2016 As illustrated at the right-hand side of the model, it is expected that citizens who make up households will become more proactive and demonstrate their commitment as co-producers and co-creators of services. This will create a strong metropolitan-community interface where citizens will, first, act as owners and managers of services and facilities, and second, report issues of broken pipes, meter tampering and illegal connections. Thus, the citizen appears 202 University of Ghana http://ugspace.ug.edu.gh twice: once as a customer and again as stakeholder of the service delivery system. This process is expected to generate positive feedback from the GWCL, ECG and other stakeholders involved in the chain of service delivery, once the existing thread connecting the two sides is premised on the principles of inclusivity, equity, accountability, profitability and sustainability. The outcome of these processes is illustrated in the chain of outputs which is depicted at the left-hand side of the model. Thus, a hybrid decentralized governance system will greatly benefit HBE operators, who represent a dominant majority in low-income settlements, through improved quality, reliable and affordable services. This will boost their business performance, through decreased operational cost and increased income, with a multiplier effect on poverty reduction at the household and community levels. More especially, such an interface will translate into better quality of life of people because first of all, with the satisfaction in service delivery, households can increase their demand (or level of access) and secondly, their chances of being able to pay for these services. From the supply side, cost recovery and profitability of the utility companies is also guaranteed through the process – improved services and the connection with the local people in whose communities these services are co-produced. The framework may seem simplistic or optimistic. But this integrative model is not intended to suggest an overly simplified conceptualization of the governance architecture in Ghana as it does recognize the challenges of co-production especially when engaging with different sectors and marginalized groups – a thorough assessment of which is beyond the remit of this study. It is discernible that problems may arise including conflicts resulting from differences in the values of the co-producers (Taylor, 2003), incompatible incentives to different components or sectors, unclear divisions of roles, free-riders (Mayo and Moore, 2002), and the undermining of capacity of the voluntary sector (citizens) to lobby for change. It is 203 University of Ghana http://ugspace.ug.edu.gh envisaged that although these problems have the potential to undermine the model’s success, with effective governance mechanisms, the model will yield successful outcomes. The optimism is bolstered by experiences and lessons drawn from Kimani catchment (Tanzania), Swahili city (Zanzibar), and Tikal (Guatemala), where the desire to achieve the SDGs has successfully integrated sustainable livelihoods into the nexus between good governance and poverty reduction (Biggs et al., 2015; see also Franks and Cleaver, 2007). Essentially, the motivation here is to highlight the expectations of residents, as citizens, to be more effectively engaged in policymaking and service co-production, and to explore the capacities of government functionaries to initiate and facilitate such engagement. In this sense, co-creation is conceived as co-design, co-production and co-implementation combined (Mistra Urban Futures, 2015; Polk, 2015). Such a focus considers human beings as people with agency who have to be free in making choices about what they want to be and do, and about how they deploy the resources and opportunities available to them. In a broader sense, this nuanced analysis of this hybrid governance system is critical for the sustainable transformation of the socio-economic and environmental aspects of Accra’s growth in line with Ghana’s urban policy and decentralization programme. The local government system, implemented since 1988, aims to encourage a greater degree of local autonomy by devolving central administrative authority to the district level, and bringing government services and functions closer to the citizenry as enshrined in the Local Government Law of 1988 (PNDC 207) and 1993 (Act 462 now Act 936, 2016). After two decades of implementation, the Ministry of Local Government and Rural Development (MLGRD) developed the decentralization policy framework in 2010 with the theme ‘Accelerating Decentralization and Local Governance for National Development’ in a quest to close the loops on the negatives associated with the programme while building on the positives. 204 University of Ghana http://ugspace.ug.edu.gh Specifically, objective four seeks to facilitate economic growth, employment and income generation in order to promote household welfare and alleviate poverty. Overall, the mandate of the policy is to achieve sustainable, equitable economic growth and poverty reduction through citizen participation and accelerated service delivery at the local level within a decentralized environment (MLGRD, 2010). Otherwise stated, there is a general consensus among analysts of Ghana’s decentralization programme, both from interviews and from the literature that the programme has led to improvement in the flow of resources and increased access of people to central government services, albeit these positive results are marginal and the process needs to be accelerated (Ayee, 1998; Owusu, 2005; Kumi-Kyereme et al., 2006). It is perhaps against this backdrop that the World Bank (2015a) indicated that the existence of urban challenges in Ghana suggests an inefficient decentralized governance system which, if not well understood and managed, could cause ‘grave danger’ for national development. In effect, the model advocates practical approaches and envisages the outcome of a functional interface between communities and metropolis, in which local knowledge and participation leverages stakeholder financing. This would require a step-up in the content and context of devolution of the local government system, particularly in the direction of fiscal devolution (Martinez-Vazquez and Vaillancourt, 2011). By extension, the experiences and perspectives of this endearing interface must be considered against the SDGs; explicitly taking step-by- step measures to decentralize the basic services whose delivery is highly linked to the achievement of SDGs which have the overall goal of contributing towards poverty reduction. The model epitomizes the tenets of development in practice which entails comparing and contrasting different households, settlements and cities, providing leverage for the successful exemplary case studies, and illustrating that urban services are a response to the agency and purposive actions of elites and commoners alike. Smith et al. (2016, p. 1587) put it this way: that “whether produced through governance in response to collective democratic action, contentious grassroots protest, or regional structures affecting competition and options, 205 University of Ghana http://ugspace.ug.edu.gh widespread urban services are rarely provided without the active agency of individuals and groups”. 6.3. Conclusion This chapter has untangled the multidimensionality of household poverty. Someone is ‘poor’ when they have a low income, low standard of living, poor access to water and electricity as well as poor interface with government – all of which translate into the inability to live the kind of life people value (Sen, 1987, 1999). Such a formulation of the relationship reminds us that income, and more generally money, is a means to an end rather than an end in itself. The study reveals five key findings. First, numerous factors converge to make poverty a multidimensional, dynamic and location-specific phenomenon. Second, poverty is understood as the absence of what is required for material well-being – particularly income/money and food but also welfare indicators including assets. Third, the absence of basic infrastructure services – particularly water and electricity – emerged as critical elements of poverty. Fourth, poor people’s definitions reveal important political and psychosocial aspects of poverty. Often, they are mindful of their lack of voice, power, and independence, which submit them to humiliation and exploitation by their better off neighbors and politicians. Finally, poor people focus on assets rather than income and link their lack of physical, human and social assets – quality of life – to poor governance, that has exacerbated their vulnerability and poverty. Similarly, the analysis of residents’ and stakeholders’ perceptions of the role of governance in poverty reduction through improved service provision highlights two points. First, it illustrates how ideologies and attitudes towards accessing services have changed as well as how informal networks of reciprocity are constructed in the delivery of services. One person’s willingness to bend the rules allows another to access water or electricity purely by capitalizing on the acquaintance. Second, maintaining connections to people in government 206 University of Ghana http://ugspace.ug.edu.gh and business, and in general to those with some control over scarce resources, is essential to avoiding poverty. However, residents indicated that they are losing connections they once had as societies become more divided between rich and poor. In a move to make urban governance more relevant, the study (re)examined the decentralised governance system to comprehend the impact of access to water and electricity on livelihoods and the outcome for poverty reduction. To illustrate this, it presented a model that reveals the synergies that can be leveraged in a hybrid decentralized governance system oriented towards poverty reduction through improved service provision. The overarching goal of the model’s approach relates closely to the drive towards a paradigm shift in the meaning of governance and poverty. First, decentralized governance – that is concerned with mere political inclusion rather than devolution which does not translate into egalitarian socio-economic change (Obeng-Odoom, 2016). Second, poverty – that tends to focus on income rather than on the denial of choices and opportunities for living a meaningful life (Vizard, 2001). In particular, the model illustrates how genuine and functional decentralized governance that emphasizes co-creation can improve service delivery and livelihood activities. In short, governance and poverty have been discussed in a more holistic sense, based partly, on the increased recognition given to the views of the poor themselves. The poor are not a homogeneous group as they experience poverty in different ways, which require a range of policy responses including neighbourhood development planning. 207 University of Ghana http://ugspace.ug.edu.gh CHAPTER SEVEN SUMMARY OF KEY FINDINGS, CONCLUSIONS AND POLICY IMPLICATIONS 7.0. Introduction This study has examined the implications of the relationship between access to water and electricity and home-based enterprises (HBEs) for poverty reduction in Accra, Ghana. It expands current understandings of the developmental significance of domestic water and electricity services through exploring narratives of households and HBEs. The study challenges the hitherto existing homogenisation of access and poverty in urban communities and highlights participants’ agency and complex ways in which they negotiate different modes of access, which is shown to be intricately linked to water and electricity-dependent enterprises. This chapter consolidates the key findings of the study and highlights its major empirical and conceptual contributions. The chapter is divided into six further sections which address the research aim and questions, recommendations and possibilities for further research. Section 7.1 outlines the contributions of this study to current understandings about poor people’s access to urban services by demonstrating how key aspects of reliability, quality, and cost impact access. Section 7.2 sheds light on how this study has enhanced understandings of HBEs and how access to services impacts HBEs. Section 7.3 outlines the conceptualizations of poverty and how these relate to access to services; it also reflects on the role of governance in reducing poverty. Section 7.4 summarizes the study’s main contributions. Section 7.5 discusses key policy recommendations. Finally, Section 7.6 suggests potential areas for future research. 7.1. Dynamics of access to urban services The provision of water and electricity has been a principal focus of national water and energy policies. Consequently, water and electricity are considered pivotal in enabling economic 208 University of Ghana http://ugspace.ug.edu.gh growth and improving people’s standards of living (Komives et al., 2005). This study fills a major gap by deconstructing access as a means of interrogating how proximity, reliability, quality and cost interrelate as key dimensions for defining people’s acceptability of urban services in Ghana. Using urban political economy as an analytical tool, the dynamics of water and electricity access challenge dominant assumptions that urban settlements experience uniform access or exclusion. It has illustrated the ramifications of socio-spatially embedded conceptualizations of access and underscores people’s individual strategies to increase their access to water and electricity. In doing so, this study has established that expansion in the quantity and improvement in quality and affordable utility services can lower costs and expand market opportunities for businesses and, thereby, increase productivity and investment that drive economic growth and poverty reduction (see also de la Fuente, 2004). Studying these nuances offers valuable insights into service delivery needs of urban communities and the underlying urban political economy context. 7.1.1. Beyond access to water and electricity Urban waterscapes and powerscapes are understood to be a tapestry of formality–informality (Misra, 2014). But the conceptual relevance and empirical viability of this categorization need closer interrogation. The conceptual and empirical interrogation of access based on coverage, which is being used to define service provision in international development discourse (WHO and UNICEF, 2014), reveals that it is untenable in the context of the multifaceted dimensions of access observed in Accra and, arguably, other cities in sub-Saharan Africa. Access to water and electricity has been constructed through an assemblage of formal systems and informal strategies. The spatiality of water and electricity coverage in the study localities depicted a positive situation. However, beyond coverage, the study found huge gaps and disparities at the household and community levels, with each settlement demonstrating mixed forms of access based on their peculiar characteristics. Proximity, reliability, quality, and cost 209 University of Ghana http://ugspace.ug.edu.gh are found to be key aspects of access and benchmarks for unravelling the multi- dimensionality and realities of access in the Accra metropolis. This parallels findings of previous studies (Amankwaa et al., 2014; Bellaubi and Visscher, 2014; Obeng-Odoom, 2012a; Ainuson, 2010), that access defined by coverage (availability) alone masks a lot of nuances. For instance, Bellaubi and Visscher (2014) indicate that the quality of service delivery by the water utilities in Kenya and Ghana, in places characterized by medium-high population density and low-middle income households, is lower than the service level reported for the system as a whole. Consequently, a strong linkage was established between households’ access from the formal utility providers and a complex supply network established through informal private water and prepaid vendors. Understanding the heterogeneity in water and electricity access through the exploration of everyday practices by which informality is operationalized, helps to unravel the needs of the urban poor, in ways that may have previously been overshadowed. This focus on everyday practices illustrates how social heterogeneity facilitates the occurrence of power and water gifting. This resonates with Zug and Graefe’s (2014) account that the phenomenon of water gifting is better understood when the socio-cultural meaning of water is recognised. Drawing on the participants’ narratives this study complicates recent literature on Ghana that emphasizes that municipal water rationing and lower socioeconomic status tend to drive sachet water consumption at the metropolitan scale (Stoler et al., 2015) as urban poor communities with some reliable piped water supply still exhibit taste for sachet water. The study also departs from the thesis that complementary water and electricity sources were the preserve of households with socio-economic status (Stoler et al., 2012; Adam et al., 2013) as it has shown that even low income households are complementing their regular formal supply with polytanks and generators – which is often facilitated by informal providers. 210 University of Ghana http://ugspace.ug.edu.gh The study has established that the so-called informal is not only resilient but also often the dominant mode of supply (Kooy, 2014), as evident in Abuja, the informal settlement, and the newly developing areas of Gbawe. Doe and Asamoah (2014) underline how inequities in access occasion informal practices to navigate the frequent power cuts and load shedding. This study concludes that although services priced outside of the formal utility providers is usually higher per unit, greater water and power security may be obtained from smaller repetitive transactions as well as having the flexibility to pursue multiple sources of power and water on a day-to-day basis. By highlighting these dynamics of access this study has illustrated how informality and formality are interdependent and permeate throughout the electricity and water provisioning process creating a meshwork of service provisioning (see Schwartz et al., 2015). The understanding of access within the informal-formal context in the development literature, as different elements that characterize all service provisioning systems, is both conceptually rigorous and empirically relevant in capturing this categorization. The dynamic nature of formality, which is produced by the interplay of the formal and informal elements, are hence best captured by Misra’s (2014) notion of ‘emergent formalizations’, that characterize different modes of arrangements. This study argues for use of this conceptualization in analysing water and electricity provisioning systems, as it helps in categorizing the particular configuration of formal and informal elements – and the patterns of interaction – that make any system effective in its peculiar context. As Misra (2014) elaborates, this perspective recognizes that the diverse systems in operation are not captured in a formal-informal binary (or a singular formal-informal continuum) but inhabit a three-dimensional conceptual space of ‘emergent formalizations’ which facilitates a more conceptually rigorous and empirically robust investigation. These dynamics expand our understanding of urban political economy analysis. 211 University of Ghana http://ugspace.ug.edu.gh 7.2. Implications of water and electricity access for home-based enterprises This study has provided two key contributions to understandings of household economies of urban residents within the Ghanaian context. First, it has revealed that HBEs contribute to urban livelihoods and should be supported to become more proactive toward facilitating, rather than denying, infrastructure, support services and adequate space for business operators (Anyidoho and Steel, 2016). Second, it has illustrated how water and electricity access affects HBEs in terms of business performance, and shaping the types and scale of home businesses households engage in. The study generates new insights into the burgeoning work in geography on the significance of HBEs and access to water and electricity for household economies. 7.2.1. HBEs vis-à-vis water and electricity access Realizing the production of formalities as a process towards reliable and affordable service provisioning allows a more useful system of differentiation, and helps in deriving suitable interventions regarding the productive uses of water and electricity. This study has contributed to discourse on how to promote income earning opportunities for the rapidly growing labor force in Ghana, by providing empirical evidence on the role that improved access to water and electricity plays in meeting the employment challenge. The HBEs urban households engage in are mostly within the informal sector and dependent on water and electricity supply. The study has revealed that eight out of every ten households engage in HBEs. The study shows a higher frequency of HBEs and this trend is expected to continue. This is particularly because HBEs display a remarkable ability to expand, with their operators being highly adaptable to constantly changing circumstances and some taking over from their ageing parents. The study supports Strassmann’s claims from 30 years ago that the HBE sector is neither disappearing nor in transition to modernity because it is needed and 212 University of Ghana http://ugspace.ug.edu.gh viable as it is (Strassmann, 1987). It also corroborates recent studies in Accra by Gough (2010) who reported that two-thirds of houses run a business from their home. The growth of HBEs has occasioned the spreading out of enterprises from the conventional home space to satellite spaces including roadsides, alleyways and lorry stations. The study has further shown that the saturation of HBEs and clamour for space is a function of the infilling with new rooms in compound houses and increasing informalization of the urban economy, which according to the World Bank (2015a) is central to the challenges of urban governance in Ghana. This signifies the need to effectively manage urban spaces in Accra and successfully integrate informal economic activities into the city’s urban governance processes. Relatedly, contrary to popular belief, some HBE owners are registered, licensed, or in other ways known to the local assembly. A focus on local governance for HBEs is justified also by the finding from the earnings analysis which shows that local factors play an important role in determining earnings, in addition to individual characteristics. This may be because in a more dynamic local economic environment, opportunities for earning income are higher, or it may reflect differences in governance and infrastructure services (Fox and Sohnesen, 2012). The profitability of the home businesses which households in GAMA engage in was shown to be significantly dependent on their access to reliable, quality, and affordable water and electricity supply. The study has revealed that enjoying direct supply from the utility providers results in positive outcomes such as reduced cost of operations, increased productivity and income. This shows that analyses of household enterprises need to incorporate the socio-spatial contexts of infrastructure service (Muhoro, 2010). As pointed out by Gough (2010), the frequency of HBEs varies not just between cities but also within cities, and within one neighbourhood will change over time, partly due to changes in the physical infrastructure. However, having official connection did not necessarily mean 213 University of Ghana http://ugspace.ug.edu.gh enjoying access as operators often complemented it with private vendor sources. Although this has been observed in current discourse on how services affect livelihoods (Komives et al., 2005), this study has illustrated how improved access to reliable, quality, and affordable services propel the agency of HBE operators to expand productivity, enjoy economies of scale and improve their quality of life. 7.3. Implications of water and electricity access for poverty reduction The thesis has contributed to recent calls within service delivery research to explore interactions at a household level and other micro-level relations to advance insights into the links between service provision and poverty reduction (Komives et al., 2005). Through investigating the conceptualizations of poverty in different socio-spatial contexts, this study has generated fresh insights into how urban poverty is being influenced by locally specific factors including household livelihood/income and water and electricity access. Further, households’ experiences in the area of urban governance, which according to Simon (2008) is a crucial aspect of eradicating poverty in low-income and peri-urban areas, are also highlighted. 7.3.1. Rethinking poverty – Income, service and governance ‘poverties’ Poverty consists of multiple, interlocking dimensions, and as such its definition based on income alone does not provide a full picture of the ‘on the ground’ reality of poor people. This shows that understandings of poverty are socio-spatially and socio-culturally embedded and different people emphasize varying aspects (Saunders, 2005). This study found that securing a livelihood and income are fundamental for defining poverty. Two out of the four localities sampled from GAMA, Korle Gonno and Abuja, recorded income poverty levels lower than the national average. This complicates the account that Greater Accra is Ghana’s least poor region. It reveals growing inequalities in household poverty and also questions the normative criteria used to define poverty. 214 University of Ghana http://ugspace.ug.edu.gh What is striking, however, is the extent to which poor access to services emerges as a fundamental factor of poor people’s definitions of poverty. This result ties in with the GSS (2013b) account that there is much higher incidence of poverty using non-monetary estimates than income poverty measurements derived from the Ghana Living Standards Survey. It further gives credence to the fact that non-monetary poverty is a complementary measure to income poverty that gives a complete picture of poverty levels (GSS, 2013b). This evokes special attention because service poverty manifests in increased business operating costs, diversification of businesses, loss of jobs and workers’ redundancy, and people’s inability to live meaningful lives. These findings extend current understandings of poor people’s definitions of poverty by highlighting how households and business operators’ agency is shaped by political economic factors such as the provision of water and electricity services which offers new avenues to understand poor households’ access routes and livelihood options. In view of these observations, it is discernible that rising inequality and household poverty in Ghana indicate the need for inclusive governance in order to achieve sustainable national poverty reduction. Residents, business operators and stakeholders alike, underscored the role of governance in improving service delivery and reducing poverty. Although this was similarly observed by Yankson (2008) and Adarkwa (2010) in their studies on governance, infrastructure service delivery and poverty reduction in Ghana, this thesis further conceptualised these findings into a governance–poverty model. Exploring these complex interactions between livelihoods/income, services and governance among urban households has thus enhanced current understandings of the synergies that can be leveraged in a functional decentralized governance system oriented towards urban services, livelihoods and poverty reduction. These conclusions need to be considered carefully from a policy perspective because the study provides evidence that for many urban poor, monetary income is only part of a much 215 University of Ghana http://ugspace.ug.edu.gh broader array of potential assets. The study has demonstrated that implicit in definitions are explanations of poverty and its outcome, which reflect Lister’s (2004) individualistic or structural perspectives. As articulated by Lister (2004), the former attributes the main responsibility for poverty to the poor themselves while the latter points to how social and political economic structures and processes create and perpetuate poverty. The study has shown how the definitions and conceptualizations combine to shape policy responses to poverty. 7.4. Summary of main contributions to knowledge Through a focus on access to services, the study has examined the private and public sources of procuring water and connecting to electricity, how the implication of the various forms of access feeds into livelihood strategies, and how the multi-activity and locality of livelihood strategies employed by residents are influencing poverty dynamics. The study concludes that home-based enterprises contribute significantly to household economies through job creation, income generation and increased affordability of services. In terms of theoretical relevance, this study has examined major theoretical and conceptual approaches to urban services – governance, co-production and urban political economy. The study is consistent with Obeng-Odoom’s (2013) account of how an urban political economy ‘way of seeing’ does not isolate urbanism, urbanity and urbanization from the broader political economic and socio-cultural contexts – nationally and globally. This perspective has illustrated how (in)formality works and how concepts of (in)formality are deployed strategically to label particular practices and service provisioning modalities. This study has shown that informality is best understood as a process of co-production and collective action. It is consistent with Bovaird (2007, p. 858) who elaborates that “co-production helps to achieve change through recognizing the need to re-conceptualize service provision as a process of social construction in which actors in self-organizing systems negotiate rules, 216 University of Ghana http://ugspace.ug.edu.gh norms and institutional frameworks rather than taking the rules as given”. However, Bovaird’s argument that co-production helps to build a consciousness of autonomy and self- worth among people who become aware of their own central contribution to local governance practice must not go unquestioned, either. This view privileges the idea of ‘people’s power’ and ‘micro politics’. The analysis of access to services and workspace reveals how the vested interests of local elites (affluent and powerful) and longtime occupiers (landlords and caretakers) determine ‘whose access matters’. The study has highlighted the importance of agency which often better explains how and why services are provided than the structural necessities of governance (Giddens, 1984; Parker, 2010). As Smith et al. (2016, p. 1587) elaborate, “whether produced through governance in response to purposive democratic action, contentious grassroots protest, or structures affecting competition, widespread urban services are rarely provided without the active agency of individuals and groups”. The key lesson here is that theories do not comprehensively explain case-specific experiences. Rather, as argued earlier (Chapter 2), their strength is twofold: First, they provide a ‘way of seeing’ and, second, they generate concepts for empirical analysis. This work advances development studies by arguing that governance, as a development paradigm, is a fluid concept that should be concerned with creating the enabling mechanisms for living a meaningful life, as articulated by Sen (1999) in his work on development as freedom. This study, consistent with IIED (2010), calls for the need to bring a wider conception of poverty into the water and electricity sectors; one which addresses poverty as a governance, service and livelihood issue. From this angle, I make a case for HBEs and how they can be used as a measure to justify improvement in service delivery and poverty reduction. By so doing, the study provides a counter proposition to research on Ghana which tends to examine the consumptive and productive uses of water and electricity separately or focus on the water–sanitation–health links as opposed to the water–livelihoods–poverty links. 217 University of Ghana http://ugspace.ug.edu.gh This position brings fresh perspectives to studies on HBEs (such as Mensah, 2015; Gough, 2010, Gough et al., 2003). This study advances urban geographic research by deepening our understanding of the notion of access. It departs from the dominant logic in development discourse where access is defined mainly by extensive coverage. First, it questions the reliability of official data as it conceals the inadequacies and lived experiences of urban poor communities in the aggregate statistics of the wider urban context. Second, it illustrates how delayed and slow governance actions and inequities in access compel those who wield power to bypass certain groups to extend service provision in peri-urban areas (Simon et al., 2006). This causes the urban poor to illegally tap into pipelines thereby leading to leakages that eventually compromise the quality of water supply to rich neighbourhoods. This chimes with Oteng-Ababio’s (2013) study on ‘unequal exposure but equal risk’. This study is consistent with Obeng-Odoom (2012a) as it illustrated how reliability, quality, and cost dynamics, which frequently overlap, have implications for poverty reduction outcomes. Further, the study has demonstrated that not only are there differences in how urban services are accessed, how livelihood activities are experienced, and how poverty is conceived, but the social disparities are also spatially differentiated – a manifestation of the concept of geographies of social and areal differentiation. In terms of policy relevance, this study complicates Ghana’s poverty reduction strategies in the water and electricity sectors that seem to focus mainly on extending coverage, boosting consumptive uses and addressing health impacts. It underlines the benefits of promoting an integrated governance and poverty reduction policy framework that emphasizes key aspects of improved urban services and household enterprises. The study exemplifies how policy initiatives that cater to the service and market needs of cities could potentially reduce poverty (in the water and electricity sector) in SSA, as other researchers have articulated (Simon, 2012, 2015). Consistent with Mistra Urban Futures (2015), this perspective calls for 218 University of Ghana http://ugspace.ug.edu.gh governments to engage residents in co-constructing their own livelihoods and ‘poverties’ rather than accepting one constructed by experts. Overall, what is needed is a new urban service ethos whereby government’s role in policy making is to support, encourage, and coordinate the co-production capabilities and entrepreneurial ingenuities of service users, business operators and the communities in which they live. This ties in with Adarkwa’s (2012) account that the face of cities and towns in Ghana should reflect orderly development, adequate environmental management, and effective local planning to make cities more livable and sustainable. This study highlights the benefits of using mixed methods in studying social phenomena that are embedded in a multiple thematic context. This resonates with Cohen et al. (2011) and Kanbur et al. (2001) who argued that a mixed methods approach is adept at obtaining richer and more textured understanding of effective governance and the everyday politics of urban life (in this case improving water and electricity access to enhance livelihoods) than presently exist. Further, unlike this study that measures a deeper understanding of how households’ themselves define poverty, most prior studies measured in quantitative terms already pre- defined variables of poverty, which blurs our understanding of people’s perceived and lived realities of poverty. Finally, the proposed governance–poverty model typifies an important integrating mechanism, bringing together a variety of stakeholders to highlight processes and outcomes that are often hidden, frequently ignored, and usually underestimated in their potential to raise the effectiveness of public policy. The framework symbolizes how co- production among public service professionals, service users, and their communities can take place through the stages of planning, design, delivery, monitoring, and evaluation. 7.5. Recommendations – policy and practice Whilst policies on Ghana’s water and energy sectors focus on domestic (consumptive) uses, this study establishes that there are visible manifestations of productive uses of household water and energy which can be leveraged to promote livelihoods and reduce poverty. Yet, the 219 University of Ghana http://ugspace.ug.edu.gh sectors have persistently failed to meet required demand resulting in negative impacts on households and businesses. Based on the study findings, this policy section advocates for functional integration of Accra’s water and electricity sectors into its decentralized governance system, as the livelihood potential is intricately linked to the informal sector and warrants official recognition and support. Also, the practice where development precedes planning for urban services should be addressed. Considering the SDGs and their focus on inclusive community–people centred development, discernable recommendations based on three themes are applicable to this study. These include: 1) community–household focus, 2) planning and 3) regulation, which all have significant implications for researchers, policy makers, city planners and funders of new governance–poverty projects in Ghana and other SSA countries. 7.5.1. Community–household collective action recommendations Form associations and involve residents in the provision of services and design of spaces for businesses: Residents and business operators should form associations. This will make their collective voice very effective by demanding better services and working conditions. The utility providers and regulators –ECG, GWCL, PURC– together with AMA should recognize these associations and their agency, and engage them to plan, design, and implement appropriate modes of water and electricity service delivery for settlements in GAMA. This will not only reduce the incidence of abandoned projects, but also create a sense of ownership among community members. Together they can build shared facilities like boreholes and extend meter connections. Similarly, because the home environment is an important space for informal trading, HBEs should be integrated into residential schemes by city planners. AMA and the district assemblies together with landlords and the associations should adopt a bottom-up approach from the outset of the planning, creation and management of business operating space. The benefits of a clearly defined and agreed upon designated areas by all parties are four–fold. First, fines by authorities to HBEs for operating in 220 University of Ghana http://ugspace.ug.edu.gh unauthorized places will be minimal. Second, environmental issues associated with uncontrolled erection of structures in public spaces which affects the city’s landscape and aesthetics will be regulated. Third, the privatization of space and the exploitation by landlords will be curtailed. Fourth, the politics of public urban space which rarely provides for the voices of the poor will be addressed. Facilitate meter acquisition process and install subsidized multi-tenant meters and community standpipes: Beyond coverage, households have identified key deficiencies in access, which entail sharing of pipes and prepaid meters. First, utility providers should facilitate the process of meter acquisition to avoid the situation of elite capture by landlords and longtime occupants. Second, subsidized multi-tenant pre-paid meters (individual household units) should be introduced for multi-habited compound houses. Third, interventions such as free water meter connections for households must be targeted as a lot of them are unable to pay monthly bills and end up being disconnected. Instead, installing public standpipes and storage tanks in different segments of settlements may prove more beneficial. In this regard, UN-Habitat’s installation of storage tanks under the participatory slum upgrade programme (PSUP) is commendable. Through this, households that are willing and able to afford reasonable bills to get connected to the utility provider can be targeted and catered to. This way the politics of access between landlords and tenants will be substantially controlled. Implement genuine pro-poor subsidies: Implementing a targeted pro-poor subsidies regime that is based on household peculiarities is paramount to ensure success. This is because each settlement has pockets of poor and well-to-do people. Even within a particular house there are individual differences in terms of the degree of poverty ranging from very poor, moderately poor, and better off poor. Realising this reality promises new directions for planning – targeting vulnerable and disadvantaged populations and blocking opportunities for 221 University of Ghana http://ugspace.ug.edu.gh elite capture. The present government subsidies, which are based on a volumetric consumption threshold, tend to exclude the actual poor people and rather benefit better off households and landlords that have direct access and manage the GWCL and ECG facilities/meters. In this regard, the Ghana National Household Register (GNHR), which is currently ongoing (to collect data on household members) should be a primary guiding mechanism for selecting beneficiary households and setting pro-poor subsidies. The oversight responsibility of PURC in this regard will be paramount. 7.5.2. Planning and organizational recommendations Form a council between PURC and the private service providers: The composition of this proposed council should include representatives from PURC, ECG and GWCL and the association of water and prepaid vendors and tanker operators. The council must regularize and regulate activities of private actors to improve efficiency, quality and ensure stable and uniform prices. This is against the backdrop that residents are aware of the high price differentials between public and private providers and also within the fraternity of private vendors and meter-sharing agents, as well as the tradeoffs in travel time that can affect attendance at school or performance at work. The overall goal of the council should be the establishment of strong links between the players and promote complementary roles that are necessary for improving access and fostering private investments and spending in the utilities sector. Form a cooperative council between business operators/associations and assemblies: This proposed council should coordinate investments in basic infrastructure and access to social intervention programmes, and advise on the coordination and regulation of the HBEs sector. Also, in line with government’s policy of supporting the private sector to function as the ‘engine for growth’, the council should provide HBEs with institutional credit support and skills training, and possibly enroll operators on the national health insurance scheme (NHIS). 222 University of Ghana http://ugspace.ug.edu.gh This could be developed through the micro-finance credit programme. Further, official recognition and assistance to the HBEs can help rope them into the tax net of AMA and increase revenue. Organized and cooperative entrepreneurial opportunities and health benefits will legitimize the council in the eyes of the entrepreneurs and lay a foundation of trust between the council’s various members. This council must include business owners and associations (with broader membership to efficiently negotiate with and wield influence), homeowners and AMA. Integrating business operators into scale up programmes through the cooperative council: Informal economic activities, like catering services, dressmaking and hairdressing, should be successfully integrated into the private sector. Attempts to scale up the HBEs sector must acknowledge existing local knowledge and leverage it to incorporate the entrepreneurs – recognising that interventions can affect people's wellbeing, and mitigate any disruption of livelihoods. Entrepreneurs should be given the chance to build individual and collective capabilities (social organization) in order to wield influence for accessing opportunities and developing links with the wider urban economy. The National Social Protection Policy (NSPP) initiative of the Ministry of Gender, Children and Social Protection (MGCSP), is doing a good work through the Livelihood Empowerment against Poverty (LEAP) program, which is a cash transfer system to poor households; and Local Enterprise and Skill Development Programme (LESDEP). But the initiatives can be expanded and leveraged through coordination with the cooperative council. Similar effort by the Council for Technical and Vocational Education and Training (COTVET) and its Development of Skills for Industry Project (DSIP) is laudable. However, DSIP’s umbrella initiative, the Skills Development Fund (SDF) which specifically excludes entrepreneurs involved in commerce must be re-engineered to be more, if all inclusive. These policy programmes should be grounded in Ghana’s national development planning agenda with the ultimate aim of achieving the SDGs and new urban agenda (NUA). 223 University of Ghana http://ugspace.ug.edu.gh Pursue an agenda for improved service delivery and new tariff regime for residential economic activities: City planners and service providers should map and identify clusters of residential livelihood activities and come up with a tariff regime that will cater to such a mixed residential–commercial category. The benefit of this approach is two-fold. First, it will decrease the dominance of private providers as the residential–commercial establishments will be captured under a new light industrial category to ensure regular supply of service. Second, it will control the practice whereby vendors and business operators draw heavily from the municipal services but pay bills as residential users (mostly flat rates). Public utilities will thus benefit directly in terms of full payments for units consumed. Although such a tariff regime may come with higher billing, the opportunity cost may still be worth pursuing considering how much business operators pay to private vendors. This proposal is justified in the wake of recent calls on government by the Association of Ghana Industries (AGI) to review the utilities tariff regime which makes industries subsidise residential users. Establish a provider–consumer standards board interface: The utility regulators should establish a networked-interface between service providers and users. This will, first, on the supply side, ensure providers introduce reforms to facilitate effective volumetric pricing based on meter readings, control leakages and reduce wastage in water and electricity usage, and increase revenue of the utilities. GWCL should introduce an electronic billing system to ensure efficiency in revenue mobilisation. Second, on the demand side, a consumer standards board should be institutionalized to fight for the rights of consumers. This interface should endeavour to promote quality assurance and consumer trust (thereby possibly controlling the rising spate of sachet water consumption), provide mechanisms for redress and facilitate public education on pre-paid meter use and system charges. The cooperation and active involvement of the ministries of power and water is key. For instance, they should work through their decentralised district assemblies to harmonise their activities with the 224 University of Ghana http://ugspace.ug.edu.gh employment and gender ministries and integrate the concerns of households and entrepreneurs with planning and governance. 7.5.3. Regulatory recommendations Harmonize and implement laws on governance to improve service delivery: The study finds the Local Governance Act of 2016 (Act 936), which seeks to provide for a National Development Planning System that defines and regulates planning procedures of District Assemblies, laudable. The bill is a one-stop legal document for all local governance matters. Consequently, it is expected that the bill’s emphasis on popular participation, and establishment of an Inter-Ministerial Coordinating Committee (IMCC) for implementing decentralized programmes would be enforced to its entirety, ensuring that the agency of individuals is harnessed to actualize local development. Relatedly, the utilities’ tariff regime must be revisited to capture the productive uses of water and electricity as this is key to establishing sustainable livelihoods. To this end, reliable, quality, and affordable supply of electricity and water should be the principal focus of national water and energy policies, due to its role in enabling economic growth. District-level policy implementers should focus on minimizing inequality in access to services within GAMA to ensure overall improvement in quality of life. Harmonize laws on enterprises and employment to reduce poverty: The Ministry of Employment and Labour relations (MELR), and MGCSP should harmonize the legislations on small medium enterprises (SMEs) development and private sector participation to make them applicable to practitioners. Initiatives to address unemployment and poverty reduction by providing financial support and expert advice to entrepreneurs should be consolidated and coordinated. These may include Youth Employment Agency (YEA), National Youth Employment Program (NYEP), Ghana Youth Employment and Entrepreneurial Development Agency (GYEEDA), Youth Enterprise Support (YES), and Youth Enterprise Development 225 University of Ghana http://ugspace.ug.edu.gh Fund (YEDF). In this regard, the study endorses the National Social Protection Policy (NSPP) objective which involves a well-coordinated, inter-sectoral protection system enabling people to live in dignity through income support, livelihoods empowerment and improved systems of access to basic services. In addition, a common national data-base (e.g. national household register) will be useful to facilitate informed decision-making on poverty or vulnerability status and beneficiary selection as well as integrate the key social protection initiatives. Innovative financing intervention by government and development partners remains key and funding resources should be managed through a Social Protection Trust or an Enterprise Support and Development Financial Scheme. This process will ensure inclusive and sustainable business practices. Other laudable policy initiatives may comprise ensuring that entrepreneurship development programmes such as LESDEP and SDF are reoriented to target more HBEs; ensuring the entrepreneurship policy streamlines various business ventures and empowers entrepreneurs to expand and grow their ambitions; and strengthening partnerships between YEA and the Ghana Centre for Entrepreneurship, Employment and Innovation (GCEEI). The National Board for Small-Scale Industries (NBSSI) and AGI should be adequately resourced to carry out their oversight mandate of coordinating and supporting private enterprise activities. Adopt a vertically integrated investor-owned model in the electricity sector: The study recommends that Ghana considers a vertically integrated investor-owned model. This proposal emphasises that building, operating, and maintaining a national electricity grid is a capital-intensive venture that should ideally be entrusted into a large, vertically-integrated entity to promote economies of scale benefits. For instance, there could be a merger of Volta River Authority (VRA) and ECG to create a vertically integrated entity rather than maintaining them as independent establishments. This way, the power producers, grid operators, regulated utility companies and other market participants can work closely in a 226 University of Ghana http://ugspace.ug.edu.gh coordinated rather than a fragmented fashion to ensure efficiency. Ultimately, this model will promote a more diversified fuel mix so that the lights can be kept on at all times. This is a counter proposal to the successive approaches that unbundle the energy generation sector and open it up to competition with the goal of attracting private investors who will compete with VRA to build up generation capacity. This will inject a sense of ownership and impose market discipline, as necessary ingredients to facilitate operational efficiencies in the sector. The proposal contextualizes government’s intention to enter into a concessionary agreement with a private entity that would invest in and manage ECG. Such a management contract or concession approach is unlikely to survive because the problems are much deeper and interlinked that simply putting a different entity in charge of the distribution network is not enough. The sector’s dysfunction is a function of leadership failure, policy inadequacies and gross mismanagement that has occurred over several decades. Further, the penchant for power barges may provide temporary relief, but it will take many years of sustained investment and disciplined management to make the dream of reliable electricity supply a reality. Finally, the multiplicity of renewable energy resources such as wind, solar, mini and small hydro, wave, and modern biomass should be exploited and piloted for electricity production. In this regard, the recently passed Renewable Energy Act to generate 10% of Ghana’s electricity from modern renewable energy sources by 2020 is commendable, as it is expected to promote renewable energy businesses. 7.6. Areas for future studies This study has interrogated the linkages between access to services, livelihoods and poverty. Just like every other research this study has got its limitations, largely due to time constraints. Hence the areas discussed below could be explored by future research. First, since the spatiality of access found a discrepancy between national coverage and community level realities, future studies should consider establishing the determinants of access using (institutional) political ecology approach. Similarly, differences observed amongst national 227 University of Ghana http://ugspace.ug.edu.gh and local level data open up a broader discussion on whether government officials and public utilities managers could have an interest in not providing empirical statistics. On the one hand, reporting lower efficiencies could justify new investments in system extensions. On the other hand, providing performance data that show progress and improvement, especially regarding the SDGs, can justify funds from the donor community. Hence, it is relevant to explore which reasons underpin the discrepancy, whilst also looking at who may be benefitting from the situation. Second, it is envisaged that future residential enterprises study should be extended to include household businesses run by residents who are located outside the settlement, as well as businesses located within settlements which are operated by others than those who live in the house. This would allow for a more comprehensive analysis of the contribution of household enterprises to the wider urban economy, and enable appropriate comparison with studies on private sector enterprises. Considering the earning potentials of HBEs, it would be interesting to study how sustainable the contributions of informal enterprises are to the national economy. This would provide a good ground for generalizing the implications of livelihood activities on poverty reduction. Other research areas might include: examining the dynamics of renting out rooms as a livelihood strategy, and the extent to which the use of mobile telephones and social media among entrepreneurs’ influence livelihood activities and poverty reduction. Third, it would be useful to provide a detailed but more disaggregated assessment of the multidimensionality of household poverty in future research. The full potential of the urban governance–poverty nexus, which inspired much of the philosophy and methods adopted in this study, as applied in various cities in the developing world, should be explored in future studies. Linked to this is the need for more critical evaluation of both the role of residential enterprises in shaping the wider political economies of developing countries, and the role of the wider political economy in shaping residential enterprises policy. 228 University of Ghana http://ugspace.ug.edu.gh Fourth, future studies could benefit immensely from data on respondents’ expenditure and assets ownership. For example, developing suitable data collection techniques that will encourage business entrepreneurs to keep simple diaries in which they record their daily income and expenditure. In this regard, researchers would follow up on their subjects at regular intervals to monitor progress being made, and address potential challenges arising in working with the research instrument. Finally, future studies should explore the feasibility of renewables such as wind and solar as alternative power sources for households and businesses, and their impact on poverty reduction. In summary, this thesis challenges the dominant logic that characterises development discourse whereby access is defined mainly by extensive coverage alone as incomplete since it elides a lot of nuances. The study has also argued that concentrating on consumptive as opposed to productive uses of water and electricity suggests that the developmental significance of household water and electricity has been overlooked. The study has shown that improving access to water and electricity significantly enhances home-based enterprises and reduces household poverty, particularly through job and income generation and increased affordability of services. In doing so, the study has made valuable empirical and conceptual contributions to geography and broader social sciences research on urban services, livelihoods and poverty reduction which is expected to inform the formulation of inclusive policies and programmes targeting water and electricity–dependent livelihoods in Africa. 229 University of Ghana http://ugspace.ug.edu.gh REFERENCES Abers, R. (1998). From clientelism to cooperation: Local government, participatory policy and civil organizing in Porto Alegre, Brazil. Politics and Society, 26(4), 511–537. Abraham, E. M., Martin, A. M., Cofie, O. and Raschid-Sally, L. (2015). Urban households’ access to water for livelihoods enhancement in Accra, Ghana. Waterlines, 34(2), 139- 155. Adam, F. W., Brew-Hammond, A. and Essandoh, E. O. (2013). Relationships between energy use and income levels, for households in Ghana. European Scientific Journal, 9(16), 233-245. Adank, M., Darteh, B., Moriarty, P., Osei-Tutu, H., Assan, D. and van Rooijen, D. (2011). Towards integrated urban water management in the Greater Accra Metropolitan Area. Current Status and Strategic Directions for the Future, Accra, Ghana: SWITCH/Resource Centre Network Ghana. Adarkwa, K. K. (2010). Housing as a strategy for poverty reduction in Ghana. UN Habitat. Adarkwa, K. K. (2012). The changing face of Ghanaian towns. African Review of Economics and Finance, 4(1), 1-29. AfDB and OECD (African Development Bank and Organisation for Economic Co-operation and Development) (2008). Economic Outlook 2007/2008. AfDB and OECD, Tunis and Paris. Afrane, S. (2003). The place of home-based enterprises in the household economies of the low-income families in Kumasi: Analysis of contributions and linkages. Journal of Science and Technology, 23(1), 34-42. Agbele, F. (2011). Political economy analysis of corruption in Ghana. European Research Centre for Anti-Corruption and State Building. ERCAS Working Paper No. 28. Agyemang, E. (2015). Transportation system as a climate mitigation strategy for the Greater Accra Metropolitan Area (GAMA). Doctoral Thesis. University of Ghana. Ahlers, R., Cleaver, F., Rusca, M. and Schwartz, K. (2014). Informal space in the urban waterscape: Disaggregation and co-production of water services. Water Alternatives, 7(1), 1-14. Ainuson, K. G. (2010). Urban water politics and water security in disadvantaged urban communities in Ghana. African Studies Quarterly, 11(4), 59-82. Alvesson, M. and Sköldberg, K. (2009). Reflexive methodology: New vistas for qualitative research. Sage Publications Ltd. Amankwaa, E. F. (2013). Livelihoods in risk: Exploring health and environmental implications of e-waste recycling as a livelihood strategy in Ghana. Journal of Modern African Studies, 51 (4), 551-575. Amankwaa, E. F. (2016). Poverty penalty: Strategies for coping with water access problems among the urban poor in Abuja, Accra. Loughborough: WEDC. 230 University of Ghana http://ugspace.ug.edu.gh Amankwaa, E. F. (2017). Women and men at the traffic lights: The (re)configuration and (re)gendering of street water vending in Ghana. GeoJournal, 82(2), 329-344. Amankwaa, E. F., Owusu, A. B., Owusu, G. and Eshun, F. (2014). Accra’s poverty trap: Analysing water provision in urban Ghana. Journal of Social Science for Policy Implications, 2(2), 69-89. Amankwaa, E. F. (2012). Livelihoods in risk: Exploring health and environmental implications of e-waste recycling as a livelihood strategy in Ghana. Master’s Thesis. University of Ghana. Amankwaa, E. F., Yankson, P. W. K., Esson, J., Gough, K. V., Mensah, P. and Oteng-Ababio, M. (2015). Ghana: Settlement reports for Sekondi-Takoradi (New-Takoradi, Kwesimintsim, Kojokrom, Assakae, Anaji). Copenhagen: Department of Geoscience and Natural Resource Management, University of Copenhagen, Denmark. Amankwaa, E. F., Tsikudo, A. K. A. and Bowman, J. (2016). Policy brief: Recyclers at risk? Analysis of e-waste livelihoods and blood lead levels at Ghana’s recycling hub, Agbogbloshie. IGC, LSE: UKAID. Anaman, K. A. (2006). What factors have influenced economic growth in Ghana? Policy Analysis, 2(2), 1-13. Andreasen, M. H. and Møller-Jensen, L. (2016). Beyond the networks: Self-help services and post-settlement network extensions in the periphery of Dar es Salaam. Habitat International, 53, 39-47. Annim, S., Mariwah, S. and Sebu, J. (2012). Spatial inequality and household poverty in Ghana. BWPI Working Paper 161. The university of Manchester, Brooks World Poverty Institute. Anyidoho, N. A. and Steel, W. F. (2016). Perceptions of cost and benefits of informal-formal linkages: Market and street vendors in Accra, Ghana. WIEGO Working Paper No. 35. Cambridge, MA, USA: WIEGO. Appadurai, A. (2001). Deep democracy: Urban governmentality and the horizon of politics. Environment and Urbanization, 13(2), 23–43. Appeaning Addo, I. (2015). Assessing residential satisfaction among low income households in multi-habited dwellings in selected low income communities in Accra. Urban Studies, 53(4), 631–650. Arfvidsson, H., Simon, D., Oloko, M. and Moodley, N. (2016). Engaging with and measuring informality in the proposed Urban Sustainable Development Goal. African Geographical Review, DOI: 10.1080/19376812.2015.1130636. Arku, G., Luginaah, I. and Mkwandawire, P. (2012). You either pay more rent advance or you move out: Landlords/ladies’ and tenants’ dilemmas in the low-income housing market in Accra, Ghana. Urban Studies, 49(14), 3177-3193. Arnstein, S. (1971). The ladder of citizen participation. Journal of the Royal Town Planning Institute, 57(1), 176–182. Aryeetey, E. and Harrigan, J. (2000). Macroeconomics, structure and growth. In Aryeetey, E., Harrigan, J. and Nissanke, M. (eds.) Economic reforms in Ghana, the miracle and 231 University of Ghana http://ugspace.ug.edu.gh the mirage. James Currey Ltd, Oxford: Woeli Publishing Services, Accra; and African World Press, Trenton, NJ, pp. 5-31. Awuah, E., Nyarko, K. B. and Owusu, P. A. (2009). Water and sanitation in Ghana. Desalination, 248(1-3), 460-467. Ayee, J. (1998). Assessing the progress of democracy and good governance in Africa: The Ghanaian case. Republic of Ghana: Accra. Ayittey, S. (2012). Ghana’s blue print for a sustainable energy future. Ghana: Environmental Science and Technology. Baah, Y. A. (2007). Organizing the informal economy: Experiences and lessons from Africa and Asia. Accra: Ghana Trades Union Congress. Baah-Boateng, W. and Ewusi, K. (2013). Employment: Policies and options. In Ewusi, K. (ed.) Policies and Options for Ghana’s Economic Development (3rd Edition). Accra: Institute of Statistical, Social and Economic Research, University of Ghana. Baah-Boateng, W., Adjei, P. and Oduro, A. D. (2013). Determinants of moonlighting in Ghana: An empirical investigation. African Review of Economics and Finance, 4(2), 176-202. Bakker, K. (2001). Paying for water: Water pricing and equity in England and Wales. Transactions, 26(2), 143–164. Bardhan, P. (2002). Decentralization of governance and development. The Journal of Economic Perspectives, 16(4), 185-205. Baulch, B. (1996). The New Poverty Agenda: A disputed consensus. Editorial in IDS Bulletin, 27(1), 1-10. Beall, J. and Kanji, N. (1999). Households, livelihoods and urban poverty. Background Paper for the ESCOR Commissioned Research on Urban Development: Urban Governance, Partnership and Poverty. Behrens, R., Diaz-Olvera, L., Plat, D. and Pochet, P. (2006). Collection of passenger travel data in Sub-Saharan African cities: Towards improving survey instruments and procedures. Transport Policy, 13, 85-96. Bellaubi, F. and Visscher, J. T (2014). Water service delivery in Kenya and Ghana: An area- based assessment of water utility performance. Water International, 39(7), 952-968. Bhaskar, R. (1978). A Realist Theory of Science (2nd edition). Hassocks, The Harvester Press. Bhaskar, R. (1979). The possibility of naturalism. A philosophical critique of the contemporary human sciences. Brighton, Harvester Press. Biggs, R., Rhode, C., Archibald, S., Kunene, L.M., Mutanga, S. S., Nkuna, N., Ocholla, P. O. and Phadima, L.J. (2015). Strategies for managing complex socio-ecological systems in the face of uncertainty: Examples from South Africa and beyond. Ecology and Society, 20(1), 52. 232 University of Ghana http://ugspace.ug.edu.gh Blakeley, G. (2005). Local governance and local democracy: The Barcelona model. Local Government Studies, 31(2), 149-165. Blunt, A. and Dowling, R. (2006). Home. London and New York, Routledge. Boardman, C. (2014). Assessing governance: The importance of evaluating policy outcomes in national mission areas. Governance: An International Journal of Policy, Administration, and Institutions, 27(3), 519-526. Bohman, A. (2010). Framing the water and sanitation challenge: A history of urban water supply and sanitation in Ghana 1909-2005. Doctoral Dissertation in Economic History Umea University, Umea. Bovaird, T. (2007). Beyond engagement and participation: User and community co- production of public services. Public Administration Review, 67(5), 846–860. Brenner, N. (2004). Urban governance and the production of new state spaces in western Europe, 1960-2000. Review of International Political Economy, 11(3), 447-488. Brohman, J. (1995). Universalism, eurocentrism, and ideological bias in development studies: From modernization to neoliberalism. Third World Quarterly, 16(1), 121-162. Burke, W. J. and Jayne, T. S. (2010). Spatial disadvantages or spatial poverty traps: Household evidence from rural Kenya. ODI Working Paper 327, CPRC Working Paper 167. Cameron, R. (2014). Vertical decentralization and urban service delivery in South Africa: Does politics matter? Development Policy Review, 32(S1), s81–s100. CARDO (2000). Housing, work and development: The role of home-based enterprises. Conference proceedings: University of Newcastle upon Tyne. Castells, M. (1977). The urban question: A Marxist approach. Edward Arnold, London. Castells, M. (1978). City, class and power. MacMillan Press Ltd. London and Basingstoke. Castells, M. (2010). Globalization, networking, urbanization: Reflections on the spatial dynamics of the information age. Urban Studies, 47(13), 2737-2745. Charmes, J. (2000). Informal sector, poverty, and gender: A review of empirical evidence. Background paper for the 2001 World Development Report, World Bank, Washington D.C. Chen, M. (2005). Rethinking the informal economy: Linkages with the formal economy and the formal regulatory environment. Helsinki: United Nations University, World Institute for Development Economics Research (WIDER), Research Paper No. 2005/10. Chen, M. (2012). The informal economy: Definitions, theories and policies. WIEGO Working Paper No. 1. Cambridge, MA, USA: WIEGO. Cheshire, P. (1995). A new phase of urban development in Western Europe? The evidence for the 1980s. Urban Studies, 32, 1045-1063. Cobbinah, P. B., Okyere, D.K. and Gaisie, E. (2016). Population growth and water supply: The future of Ghanaian cities. In Benna, U. G. and Garba, S. B. (eds.), Population Growth and Rapid Urbanisation in the Developing World. IGI Global, pp. 231-252. 233 University of Ghana http://ugspace.ug.edu.gh Cohen, L., Manion, L. and Morrison, K. (2011). Research Methods in Education. Routledge, London. Connell, D. (1997). Participatory development: An approach sensitive to class and gender. Development in Practice, 7(3), 248–259. Cope, M. (2010). Coding qualitative data. In Hay, I. (ed.) Qualitative research methods in human geography, (3rd edition), Ontario: Oxford University Press, pp. 281–294. Cormode, L. and Hughes, A. (1999). The economic geographer as a situated researcher of elites. Geoforum, 30, 299-300. Coulombe, H. and Wodon, Q. (2007). Poverty, livelihoods, and access to basic services in Ghana. Ghana CEM: Meeting the Challenge of Accelerated and Shared Growth. World Bank. Creswell, J. W. (2009). Research Design: Qualitative, Quantitative and Mixed Methods Approaches. (3rd edition). SAGE Publications Ltd, London. Crook, R. (2003). Decentralization and poverty reduction in Africa: The politics of local- central relations. Public Administration and Development, 23(1), 77-88. Daniel, M. M. (2014). Local and community driven development approach in the provision of basic facilities in Jos, Nigeria. Cities, 39, 99-108. Danso, K. A. (2007). Leadership concepts and the role of government in Africa: The case of Ghana. Philadelphia, PA: Xlibris Corporation. Davila, J. (2009). Being a mayor: The view from four Colombian cities. Environment and Urbanisation, 21(1), 37-57. de la Fuente, A. (2004). Infrastructure productivity and growth. Unpublished paper, World Bank Institute Governance, Regulation, and Finance Division, World Bank, Washington, DC. de Oliveira, J. A. P. (2015). Review of ‘Governance for pro-poor urban development: Lessons from Ghana’. Public Administration and Development, 35(3), 219-220. de Soto, H. (1989). The Other Path: The Economic Answer to Terrorism. New York: Harper Collins. Denzin, N. K. (2012). Triangulation 2.0. Journal of Mixed Methods Research, 6(2), 80-88. Devas, N. (2001). The connections between urban governance and poverty. Journal of International Development, 13(7), 989-996. Dick, B. (2005). Grounded theory: A thumbnail sketch. [Online]. Available at: http://www.scu.edu.au/schools/gcm/ar/arp/grounded.html (April 20, 2010). Doe, F. and Asamoah, E. S (2014). The effect of electric power fluctuations on the profitability and competitiveness of SMEs: A study of SMEs within the Accra Business District of Ghana. Journal of Competitiveness, 6(3), 32-48. Dollar, D. and Kraay, A. (2002). Growth is good for the people. Journal of Economic Growth, 7(3), 195-225. Drakakis-Smith, D. (1987). The Third World City. Methuen and Co. Ltd, London. 234 University of Ghana http://ugspace.ug.edu.gh Drakakis-Smith, D. (2000). The Third World City (2nd Edition). Routledge, London and New York. Dreyfus, H. L. and Rabinow, P. (1982). Michel Foucault: Beyond Structuralism and Hermeneutics. Harvester Wheatsheaf, Hemel Hempstead. Eshun, M. E. and Amoako-Tuffour, J. (2016). A review of the trends in Ghana’s power sector. Energy, Sustainability and Society, 5, 1-9. Essel, I. (2010). Accra water: Production 80m; demand 250m. Available at http://news.myjoyonline.com/news/201002/42113.asp (December 4, 2010). Esson, J., Amankwaa, E. F., Gough, K. V., Mensah, P., Oteng-Ababio, M. and Yankson, P. W. K. (2015). Ghana: Settlement reports for Accra (Korle Gonno, Accra New-Town, Gbawe, Ashaley Botwe). Copenhagen: Department of Geoscience and Natural Resource Management, University of Copenhagen, Denmark. Esson, J., Gough, K. V., Simon, D., Amankwaa, E. F., Ninot, O. and Yankson, P. W. K. (2016). Livelihoods in motion: Linking transport, mobility and income-generating activities. Journal of Transport Geography, 55, 182-188. Estache, A. (2002). Argentina’s 1990s privatization: A cure or a disease? Unpublished paper, Infrastructure Vice Presidency, World Bank, Washington, DC. Farvacque-Vitkovic, C., Raghunath, M., Eghoff, C. and Boakye, C. (2008). Development of the cities of Ghana: Challenges, priorities and tools. Africa Region Working Paper Series No. 110. Finlay, J. M. and Bowman, J. A. (2016). Geographies on the move: A practical and theoretical approach to the mobile interview. The Professional Geographer, DOI: 10.1080/00330124.2016.1229623. Florida, R. (2003). Cities and the creative class. City and Community, 2(1), 3-19. Fosu, A. (2009). Inequality and the impact of growth on poverty: Comparative evidence for sub-Saharan Africa. Journal of Development Studies, 45(5), 726-745. Fox, L. and Sohnesen, T. P. (2012). Household enterprises in Sub-Saharan Africa: Why they matter for growth, jobs, and livelihoods. Washington, D.C.: World Bank Policy Research Paper 6184. Franks, T. and Cleaver, F. (2007). Water governance and poverty: A framework for analysis. Progress in Development Studies, 7(4), 291–306. Fritz, V., Kaiser, K. and Levy, B. (2009). Problem-driven governance and political economy analysis. Good Practice Framework. The World Bank. Fuest, V. and Haffner, S. (2007). PPP – Policies, practices and problems in Ghana’s urban water supply. Water Policy, 9(2), 169–192. Fuseini, I. and Kemp, J. (2016). Characterizing urban growth in Tamale, Ghana: An analysis of urban governance response in infrastructure and service provision. Habitat International, 56, 109-123. GEC (Ghana Energy Commission) (2011). Energy supply and demand outlook for Ghana. Accra: Energy Commission. 235 University of Ghana http://ugspace.ug.edu.gh GEC (Ghana Energy Commission) (2012). Energy supply and demand outlook for Ghana. Accra: Energy Commission. GEC (Ghana Energy Commission) (2015). Energy supply and demand outlook for Ghana. Accra: Energy Commission. GEC (Ghana Energy Commission) (2016). 2016 Energy supply and demand outlook for Ghana. Accra, Energy Commission. Ghafur, S. (2002). Gender implications of space use in home-based work: Evidences from slums in Bangladesh. Habitat International, 26, 33–50. Giddens, A. (1984). The constitution of society. Cambridge, Polity. GIPC (Ghana Investment Promotion Council) (2010). First quarter 2010 investment report. The GIPC Quarterly Update, 6(1), 1-5. Glaeser, E. (2007). The economic approach to cities. Discussion paper presented at the Harvard Institute of Economic Research and John F Kennedy School of Government, Paper No. RWP08-003. Glaeser, E. (2016). Managing the challenges and opportunities of urbanization in Africa. Annual Bank Conference on Africa. In Siba, E., Barofsky, J. and Grabinsky, J. (eds.), Can rapid urbanization in Africa reduce poverty? Causes, opportunities, and policy recommendations. Africa in Focus. Gough, K. V. (1996). Home-based enterprises in low-income settlements: Evidence from Pereira, Colombia. Danish Journal of Geography, 96, 95–102. Gough, K.V. (2010). Continuity and adaptability of home-based enterprises: A longitudinal study from Accra, Ghana. International Development Planning Review, 32(1), 45-70. Gough, K.V., Tipple, A. G. and Napier, M. (2003). Making a living in African cities: The role of home-based enterprises in Accra and Pretoria. International Planning Studies, 8, 253–277. Gough, K. V. and Kellett, K. V. (2001). Housing consolidation and self-help settlements in two Colombian cities. Cities, 18, 235–247. Gough, K. V. and Yankson, P. W. K. (2000). Land markets in African cities: The case of peri- urban Accra, Ghana. Urban Studies, 37(13), 2485-2500. Gough, K. V. and Yankson, P. W. K. (1997). Continuity and change in peri-urban Accra: Socio-economic and environmental consequences of urbanisation. Final Report to the Danish Council for Development Research. Gough, K. V. and Yankson, P. W. K. (2011). A neglected aspect of the housing market: The caretakers of peri-urban Accra, Ghana. Urban Studies, 48(4), 793-810. Gough, K. V., Esson, J., Andreasen, M., Mainet, H., Namangaya, A., Yankson, P. W. K., Agergaard, J., Amankwaa, E.F., Kiunsi, R., Moller-Jensen, L. and Yemmafouo, A. (2015). City dynamics: Mobility and livelihoods of urban residents. Copenhagen: Department of Geoscience and Natural Resource Management, University of Copenhagen, Denmark, 25pp. Grafton, J., Lillis, A. M. and Mahama, H. (2011). Mixed methods research in accounting. Qualitative Research in Accounting & Management, 8(1), 5-21. 236 University of Ghana http://ugspace.ug.edu.gh Graham, S. and Marvin, S. (2001). Splintering urbanism: Networked infrastructures, technological mobilities and the urban condition. Routledge, London. GRIDCo (2010). Ghana wholesale power reliability assessment: 2010 final report. Power Systems Energy Consulting, GRIDCo. Grindle, M. (2007). Good enough governance revisited. Development Policy Review, 25(5), 553–574. GSS (Ghana Statistical Service) (2005). 2000 Population and Housing Census: Greater Accra Region. Analysis of district data and implications for planning, Republic of Ghana. GSS (Ghana Statistical Service) (2008). Ghana Living Standards Survey: Report of the fifth round. GSS, Accra. GSS (Ghana Statistical Service) (2012). 2010 Population and Housing Census: Summary report of final results. GSS, Accra. GSS (Ghana Statistical Service) (2013a). 2010 Population and Housing Census National Analytical Report. GSS, Accra. GSS (Ghana Statistical Service) (2013b). Non-monetary poverty in Ghana. GSS, Accra. GSS (Ghana Statistical Service) (2013c). Millennium Development Goals in Ghana. 2010 Population and Housing Census Report. GSS, UNDP. GSS (Ghana Statistical Service) (2014a). Ghana Living Standards Survey Round 6 (GLSS): Poverty profile in Ghana (2008-2013). Accra, Ghana. GSS (Ghana Statistical Service) (2014b). 2010 Population and Housing Census District Analytical Report, Accra Metropolitan Area. Ghana Statistical Service, Accra. Gyamfi, S., Modjinou, M. and Djordjevic, S. (2015). Improving electricity supply security in Ghana – The potential of renewable energy. Renewable and Sustainable Energy Reviews, 43, 1035–1045. Hall, R. P., Van Koppen, B. and Van Houweling, E. (2014). The human right to water: The importance of domestic and productive water rights. Science and Engineering Ethics, 20, 849–868. Hansson, H., Mwango, F., Otieno, J. and Nyström, M. (2014). A transdisciplinary research approach: Challenges and benefits of co-production. Cumulus Johannesburg Conference. Hart, K. (1973). Informal income opportunities and urban employment in Ghana. Journal of Modern African Studies, 11(3), 61-89. Hart, K. (2008). The informal economy and opportunities for a new African revolution. The New Legon Observer, 2(18), 4-8. Harvey, D. (1973). Social justice and the city. John Hopkins University Press, Baltimore, MD and Edward Arnold, London. Harvey, D. (1998). The geography of class power. The Socialist Register, 34, 49-75. Harvey, D. (2001). Spaces of capital: Towards a critical geography. Routledge, New York. 237 University of Ghana http://ugspace.ug.edu.gh Harvey, D. (2005). The sociological and geographical imaginations. International Journal of Politics, Culture and Society, 18(3-4), 211-255. Harvey, D. (2006). The limits to capital (3rd edition). Verso, London. Harvey, D. (2007). Neoliberalism as creative destruction. The Annals of the American Academy of Political and Social Science, 610(21), 22-44. Harvey, D. (2008). The right to the city. New Left Review, 53, 23-40. Harvey, D. (2012). Rebel cities: From the right to the city to the urban revolution. New York: Verso. Haug, J. (2014). Critical overview of the urban informal economy in Ghana. Friedrich Ebert Stiftung FES Ghana office. Healey, P. (2007). Urban complexities and spatial strategies: Towards a relational planning for our times. Routledge, London. Holmes, B. (2011). Citizens’ engagement in policymaking and the design of public services. Research Paper No. 1, 2011–12. Holt, J. and Manning, N. (2014). Fukuyama is right about measuring state quality: Now what? Governance: An International Journal of Policy, Administration, and Institutions, 27(4), 717-728. Hufty, M. (2009). The governance analytical framework: Preliminary version for comments. http://repository.graduateinstitute.ch/record/16577/files/GovernanceFrameworkE.pd f (retrieved 23 November, 2016). Hutchful, E. (1995). Why regimes adjust: The World Bank ponders its “star pupil”. Canadian Journal of African Studies, 29(2), 303–317. Hutchful, E. (2002). Ghana’s Adjustment Experience, The Paradox of Reform. Oxford, UK: UN Research Institute of Social Development. IEA (International Energy Agency). (2013). World Energy Outlook 2013. Paris, OECD/IEA. IIED (International Institute for Environment and Development) (2010). Channels for change: Private water and the urban poor. IIED Briefing Paper May 2010, IIED. ILO and WIEGO (International Labour Office and Women in Informal Employment: Globalizing and Organizing) (2013). Women and men in the informal economy: A statistical picture (2nd edition). Geneva: ILO and WIEGO. ISSER (Institute of Statistical, Social and Economic Research) (2005). Guide to electric power in Ghana (1st edition). Resource Centre for Energy Economics and Regulation (RCEER), ISSER. Jick, T. D. (1979). Mixing qualitative and quantitative methods: Triangulation in action. Administrative Science Quarterly, 24(4), 602-611. Johnston, R. J., Gregory, D. and Smith, D. M. (2000). The dictionary of human geography (4th edition). Oxford, England: Blackwell. Joshi, A. and Moore, M. (2004). Institutionalized coproduction: Unorthodox public service delivery in challenging environments. The Journal of Development Studies, 40(4), 31– 49. 238 University of Ghana http://ugspace.ug.edu.gh Kanbur, R., Chambers, R., Petesch, P., Uphoff, N. Ravallion, M. et al. (2001). Qualitative and quantitative poverty appraisal: Complementarities, tensions and the way forward. Poverty, Inequality and Development Initiative, Cornell University. Kanin, Y. (2010). Partnership between Ghana Water Company Limited and Tanker Associations for water service delivery in Accra. Master’s Thesis, Department of Civil Engineering. Kwame Nkrumah University of Science and Technology, Ghana. Kariuki, M. and Schwartz, J. (2005). Small-scale private service providers of water supply and electricity: A review of incidence, structure, pricing and operating characteristics. World Bank Policy Research Working Paper 3727. Komives, K., Foster, V., Halpern, J. and Wodon, Q. (2005). Water, electricity, and the poor: Who benefits from utility subsidies? The World Bank, Washington, DC. Konadu-Agyemang, K. (2001). A survey of housing conditions and characteristics in Accra, an African city. Habitat International, 25, 15-34. Kooy, M. (2014). Developing informality: Jakartaʼs urban waterscape. Special Issue on Informal Space in the Urban Waterscape, Water Alternatives, 7(1), 35-53. Korboe, D. (1992). Family-houses in Ghanaian cities: To be or not to be? Urban Studies, 29(7), 1159-1172. Kumi-Kyereme, A., Yankson, P. W. K. and Thomi, W. (2006). Challenges of decentralization in Ghana: District Assemblies’ responsiveness to community needs. Ife PsychologIA: An International Journal of Psychology in Africa, 14, 60–76. Kwankye, S. O. and Anarfi, J. (2011). Migration impacts within the West African sub-region. Paper prepared for the Migrating out of Poverty Research Programme Consortium, University of Sussex, Brighton, United Kingdom. Langevang, T. (2008). We are managing! Uncertain paths to respectable adulthoods in Accra, Ghana. Geoforum, 39, 2039–2047. Langevang, T. (2016). Fashioning the future: Entrepreneuring in Africa’s emerging fashion industry. Centre for Business and Development Studies, CBDS Working Paper No. 1. Limb, M. and Dwyer, C. (2001). Introduction: Doing qualitative research in Geography. In Limb, M. and Dwyer, C. (eds.), Qualitative methodologies for geographers: Issues and debates. London, England: Arnold, pp. 1-20. Lister, R. (2004). Poverty. Cambridge: Polity. Mabogunje, A. (1990). Urban planning and the post-colonial state in Africa: A research overview. African Studies Review, 33(2), 121-203. Maloney, W. F. (2004). Informality revisited. World Development, 32, 1159-1178. Maloutas, T. and Malouta, M. (2004). The glass menagerie of urban governance and social cohesion: Concepts and stakes/concepts as stakes. International Journal of Urban and Regional Research, 28(2), 449-465. Marschall, M. (2004). Citizen participation in the neighbourhood context: A new look at co- production of local public goods. Political Research Quarterly, 57(2), 231–244. 239 University of Ghana http://ugspace.ug.edu.gh Martinez-Vazquez, J. and F. Vaillancourt, (eds.) (2011). Decentralization in developing countries: Global perspectives on the obstacles to fiscal devolution. Chaltenham: Edward Elgar. Mayo, Ed. and Moore, H. (eds.). (2002). Building the Mutual State: Findings from the Virtual Think tank. London: New Economics Foundation/Mutuo. McFarlane, C. (2012). The entrepreneurial slum: Civil society, mobility and the co- production of urban development. Urban Studies, 49(13), 2795-2816. McFarlane, C. and Desai, R. (2015). Sites of entitlement: Claim, negotiation and struggle in Mumbai. Environment and Urbanisation, 27(2), 1–14. McGranahan, G., Njiru, C. Albu, M., Smith, M. and Mitlin, D. (2006). How small water enterprises can contribute to the Millennium Development Goals. Loughborough: WEDC. Meagher, K. (2013). Unlocking the informal economy: A literature review on linkages between formal and informal economies in developing countries. Manchester, UK: WIEGO Working Paper No. 27. Meikle, S. and Bannister, A. (2003). Energy, poverty and sustainable urban livelihoods. Working Paper No. 126 ISSN 1474-3280. Melara Arguello, J. E., Grant, R., Oteng-Ababio, M. and Ayele, B. (2013). Downgrading – An overlooked reality in African cities: Reflections from an indigenous neighborhood of Accra, Ghana. Applied Geography, 36, 23–30. Melara Arguello, J. E., Oteng-Ababio, M., Ayele, B., Grant, R., Kang, S., Sahu, M. and Alhassan, A. R. (2012). AMA urban community profile: Korle Gonno. Millennium Cities Initiative–Earth Institute/Columbia University–University of Ghana. Accra Metropolitan Assembly, Columbia University, New York. Mensah, P. (2015). Home-based enterprises as an informal economic activity: A case of Ga Mashie in Accra, Ghana. Master’s Thesis, Department of Geography and Resource Development, University of Ghana. Merrifield, A. (2002). Metromarxism: A Marxist tale of the city. New York: Routledge. Misra, K. (2014). From formal-informal to emergent formalization: Fluidities in the production of urban waterscapes. Water Alternatives, 7(1), 15-34. Mistra Urban Futures (2015). Mid-term evaluation 2015 Mistra Urban Futures. MISTRA: The Swedish Foundation for Strategic Environmental Research. Mitlin, D. (2008). With and beyond the state–co-production as a route to political influence, power and transformation for grassroots organizations. Environment and Urbanisation, 20(2), 339–360. MLGRD (Ministry of Local Government and Rural Development) (2010). Ghana’s decentralization policy framework. Government of Ghana. Mohammad, R. (2001). Insiders and/or outsiders: Positionality, theory and practice. In Limb, M. and Dwyer, C. (eds.), Qualitative methods for geographers: Issues and debates. London, England: Arnold, pp. 101- 117. 240 University of Ghana http://ugspace.ug.edu.gh Moore, H. (1994). A Passion for Difference. Policy Press, Cambridge. Moses, J. and Knutsen, T. L. (2007). Ways of Knowing: Competing Methodologies in Social and Political Research. New York, Palgrave Macmillan. Muhoro, P. N. (2010). Off-Grid Electricity Access and its Impact on Micro-Enterprises: Evidence from Rural Uganda. A Doctor of Philosophy dissertation submitted to the University of Michigan. Mullings, B. (1999). Insider or outsider, both or neither: Some dilemmas of interviewing in across-cultural setting. Geoforum, 30, 337-350. Narayan, D. (1999). Can anyone hear us? Voices from 47 countries. Poverty Group, PREM, World Bank. NDPC (National Development Planning Commission) (2015). Ghana Millennium Development Goals 2015 Report. Accra, NDPC. Newman, J., Barnes, M., Sullivan, H. and Knops, A. (2004). Public participation and collaborative governance. Journal of Social Policy, 33(2), 203-223. Newton, K. (1981). Urban political economy. Frances Pinter Ltd, London. Nicholls W. (2008). The urban question revisited: The importance of cities for social movements. International Journal of Urban and Regional Research, 32(4), 1468- 2427. Nuissl, H. and Heinrichs, D. (2011). Fresh wind or hot air – does the governance discourse have something to offer to spatial planning? Journal of Planning Education and Research, 31(1), 47-59. O’Sullivan, A. (2003). Urban economics (5th edition). McGraw-Hill Irwin, Boston, MA. Obeng-Odoom, F. (2011). Ill health unleashed? Cities and municipal services in Ghana. Review of African Political Economy, 38(127), 43–60. Obeng-Odoom, F. (2012a). Beyond access to water. Development in Practice, 22(8), 1135- 1146. Obeng-Odoom, F. (2012b). On the origin, meaning, and evaluation of urban governance. Norwegian Journal of Geography, 66(4), 204-212. Obeng-Odoom, F. (2013). Governance for pro-poor urban development: Lessons from Ghana (1st edition). London: Routledge. Obeng-Odoom, F. (2016). Urban governance in Africa today: Reframing, experiences, and lessons. Growth and Change, DOI: 10.1111/grow.12164. Olsen, W. K. (2004). Triangulation in social research: Qualitative and quantitative methods can really be mixed. In Holborn, M. (Ed.), Developments in Sociology: An Annual Review, Ormskirk, Lancs, UK, Causeway Press. Oosterbaan, C., Arku, G. and Asiedu, A. B. (2012). Conversion of residential units to commercial spaces in Accra, Ghana: A policy dilemma. International Planning Studies, 17(1), 45-66. Osei-Boateng, C. and Ampratwum, E. (2011). The informal sector in Ghana. Friedrich Ebert Stiftung FES Ghana office. 241 University of Ghana http://ugspace.ug.edu.gh Ostrom, E. (1996). Crossing the great divide: Co-production, synergy and development. World Development, 24(6), 1073–1087. Oteng-Ababio, M. (2013). Unscripted (in)justice: Exposure to ecological hazards in Metropolitan Accra. Environment and Planning A, 45(5), 1199-1218. Oteng-Ababio, M. and Amankwaa, E. F. (2014). The e-waste conundrum: Balancing evidence from the North and on-the ground developing countries realities for improved management. African Review of Economics and Finance, 6(1), 181-204. Oteng-Ababio, M. and Melara Arguello, J. E. (2014). Paradigm of mediocrity: Poverty and risk accumulation in urban Africa – The case of Korle Gonno, Accra. Journal of Urbanism: International Research on Placemaking and Urban Sustainability, 7(1), 45-61. Oteng-Ababio, M., Melara Arguello, J. E. and Gabbay, O. (2013). Solid waste management in African cities: Sorting the facts from the fads in Accra, Ghana. Habitat International, 39, 96-104. Owusu, F. (2001). Urban impoverishment and multiple modes of livelihood in Ghana. The Canadian Geographer, 45(3), 387-403. Owusu, F. (2007). Conceptualizing livelihood strategies in African cities: Planning and development implications of multiple livelihood strategies. Journal of Planning Education and Research, 26(4), 450-465. Owusu, F. (2012). Organizational culture and public sector reforms in a post-Washington consensus era: Can Ghana’s reformers learn from Ghana’s good performers? Progress in Development Studies, 12(2&3), 131-151. Owusu, G. (2005). Small towns in Ghana: Justifications for their promotion under the decentralisation programme. African Studies Quarterly, 8, 48-69. Owusu, G. (2008). “Indigenes” and “migrants” access to land in peri-urban areas of Ghana’s largest city of Accra. International Development Planning Review, 30, 177-198. Owusu, G. and Afutu-Kotey, R. L. (2010). Poor urban communities and municipal interface in Ghana: Case study of Accra and Sekondi-Takoradi Metropolis. African Studies Quarterly, 12(1), 1-16. Owusu, G. and Oteng-Ababio, M. (2014). Moving unruly contemporary urbanism toward sustainable urban development in Ghana by 2030. American Behavioral Scientist, DOI: 10.1177/0002764214550302. Owusu, G. and Yankson, P. W. K. (2007). Poverty in Ghana is basically a rural phenomenon: Are we underestimating urban poverty? Ghana Journal of Development Studies, 4(1), 87-105. Parker, J. (2010). Structuration theories. Historical developments and theoretical approaches in Sociology, 2. http://www.eolss.net/sample-chapters/c04/e6-99a-27.pdf (retrieved 2 September, 2016). Peloso, M. and Morinville, C. (2014). ‘Chasing for water’: Everyday practices of water access in peri-urban Ashaiman, Ghana. Water Alternatives, 7(1), 121-139. 242 University of Ghana http://ugspace.ug.edu.gh Polk, M. (ed.) (2015). Co-producing knowledge for sustainable cities: Joining forces for change. Oxon and New York: Routledge. Portes, A., Castells, M. and Benton, L. A. (eds.) (1989). The informal economy: Studies in advanced and less developed countries. Baltimore, MA: The Johns Hopkins University Press. Potts, D. (2009). The slowing of sub-Saharan Africa’s urbanization: Evidence and implications for urban livelihoods. Environment and Urbanisation, 21(1), 253-259. Potts, D. (2012a). Whatever happened to Africa’s rapid urbanization? Africa Research Institute Counterpoints Series, February, London. Potts, D. (2012b). What do we know about urbanization in sub-Saharan Africa and does it matter? International Development Review, 34(1), v-xxi. Potts, D. (2012c). Challenging the myths of urban dynamics in sub-Saharan Africa: The evidence from Nigeria. World Development, 40(7), 1382-1393. Potts, D. (2016). Debates about African urbanisation, migration and economic growth: What can we learn from Zimbabwe and Zambia? The Geographical Journal, 182(3), 251– 264. Punch, S. (2010). Hidden struggles of fieldwork: Exploring the role and use of field diaries. Emotion, Space & Society, 5(2), 86–93. PURC (Public Utilities Regulatory Commission) (2005). Urban water tariff policy. PURC, Accra. Ravallion, M., Chen, S. and Sangraula, P. (2007). The urbanization of global poverty. World Bank Research Digest, 1(4). Resnick, D. (2014). Urban governance and service delivery in African cities: The role of politics and policies. Development Policy Review, 32(S1), s3–ss17. Resnik, D. B. (2015). What is ethics in research and why is it important? National Institute of Environmental and Health Sciences Research report. Ribera-Fumaz, R. (2009). From urban political economy to cultural political economy: Rethinking culture and economy in and beyond the urban. Progress in Human Geography, 33(4), 447-465. Rodrik, D. (2008). Spence christens a new Washington consensus. The Economists’ Voice, 5(3), 1-3. Rondinelli, D. A., McCullough, J. S. and Johnson, R. W. (1989). Analysing decentralization policies in developing countries: A political-economy framework. Development and Change, 20(1), 57-87. Rotberg, R. I. (2014). Good governance means performance and results. Governance, 27(3), 511–518. Sale, J. E. M., Lohfeld, L. H. and Brazil, K. (2002). Revisiting the quantitative-qualitative debate: Implications for mixed-methods research. Quality & Quantity, 36, 43-53. Sarpong Manu, K. and Abrampah, K. M. (2006). Small water enterprises in Africa 4 – Ghana: A study of small water enterprises in Accra. Loughborough: WEDC. 243 University of Ghana http://ugspace.ug.edu.gh Saunders, P. (2005). The poverty wars: Reconnecting research with reality. New South Publishing. Sawers, L. (1975). Urban form and the mode of production. Review of Radical Political Economics, 7(52), 52-68. Sawers, L. (1984). New perspectives on the urban political economy. In Tabb, W. and Sawers, L. (eds.), Marxism and the metropolis: New perspectives in urban political economy (2nd edition), Oxford University Press, New York, pp. 3-17. Sayer, A. (2000). Realism and Social Science. London, Sage. Schwartz, K., Luque, M. T., Rusca, M. and Ahlers, R. (2015). Informality: the meshwork of water service provisioning. WIREs Water, 2(1), 31–36. Scott, A. and Seth, P. (2013). The political economy of electricity distribution in developing countries: A review of the literature. ODI Politics and Governance. SDSN (Sustainable Development Solution Network Thematic Group on Sustainable Cities) (2013). The urban opportunity: Enabling transformative and sustainable development. Background paper for the High-Level Panel of Eminent Persons on the Post-2015 Development Agenda. Sen, A. (1987). The Standard of Living. Cambridge: Cambridge University Press. Sen, A. (1999). Development as Freedom. Oxford University Press. Shah, A. (2006). Local governance in developing countries. World Bank. Silver, J.D. (2013). Reconfiguring electricity infrastructures in Accra and Cape Town: Understanding the political ecologies of networked urbanism. Doctoral theses, Durham University. Simon, D., Arfvidsson, H., Anand, G., Bazaaz, A., Fenna, G., Foster, K., Jain, G., Hansson, S., Marix E. L., Moodley, N., Nyambuga, C., Oloko, M., Chandi Ombara, D., Patel, Z., Perry, B., Primo, N., Revi, A., Van Niekerk, B., Wharton, A. and Wright, C. (2016). Developing and testing the Urban Sustainable Development Goal’s targets and indicators – a five-city study. Environment and Urbanization, 28(1), 49-63. Simon, D., McGregor, D. and Thompson, D. (2006). Contemporary perspectives on the peri- urban zones of cities in developing areas. In Simon D. and Thompson D. (eds.), The Peri-Urban Interface – Approaches to Sustainable Natural and Human Resources Use. Earthscan, London, pp. 3-17. Simon, D. (ed.) (2016). Rethinking sustainable cities: Accessible, green and fair. Bristol: Policy Press. Simon, D. (2015). Uncertain times, contested resources: discursive practices and lived realities in African urban environments. City, 19, 216–238. Simon, D. (2012). Climate and environmental change and the potential for greening African cities. Local Economy, 28(2), 203-217. Simon, D. (2008). Urban environments: Issues on the peri-urban fringe. Annual Review of Environment and Resources, 33, 167–185. 244 University of Ghana http://ugspace.ug.edu.gh Simon, D. and Adam-Bradford, A. (2016). Archaeology and contemporary dynamics for more sustainable, resilient cities in the peri-urban interface. In Maheshwari B. Singh V. P. and Thoradeniya B. (eds.), Balanced urban development: Options and strategies for liveable cities. Water Science and Technology Library, Springer International Publishing, pp. 57-83. Sinai, I. (1998). Using the home for income-generation: The case of Kumasi, Ghana. Cities, 15, 417–427. Smith, M. E., Dennehy, T., Kamp-Whittaker, A., Stanley, B. W., Stark, B. and York, A. (2016). Conceptual approaches to service provision in cities throughout history. Urban Studies, 53(8), 1574–1590. Soja, E. (1976). Spatial inequality in Africa. University of California, CA. Songsore, J. (2008). Environmental and structural inequalities in Greater Accra. The Journal of the International Institute, 16(1), 8-13. Songsore, J., Nabila, S., Yangyuoru, Y., Avle, S., Bosque-Hamilton, E. K., Amponsah, P. E. and Alhassan, O. (2009). Environmental health watch and disaster monitoring in the Greater Accra Metropolitan Area (GAMA), 2005. Ghana Universities Press, Accra. Stanciulescu, G. (2009). The role of urban marketing in the local economic development. Theoretical and Empirical Researches in Urban Management, 10(1), 114-135. Stilwell, F. (1992). Understanding cities and regions: Spatial political economy. Sydney: Pluto. Stilwell, F. (2000). Ways of seeing: Competing perspectives on urban problems and policies. In Troy, P. (ed.), Equity, environment, efficiency: Ethics and economics in urban Australia, Melbourne University Press, Victoria, pp. 13-37. Stilwell, F. (2006). Political economy: The context of economic ideas (2nd edition). Oxford University Press, Melbourne. Stoler, J., Fink, G., Weeks, J.R., Appiah Otoo, R., Ampofo, J. A. and Hill, A. G. (2012). When urban taps run dry: Sachet water consumption and health effects in low income neighbourhoods of Accra, Ghana. Health & Place, 18, 250–262. Stoler, J., Tutu, R. A. and Winslow, K. (2015). Piped water flows but sachet consumption grows: The paradoxical drinking water landscape of an urban slum in Ashaiman, Ghana. Habitat International, 47, 52-60. Strassmann, W. P. (1987). Home-based enterprises in cities of developing countries. Economic Development and Cultural Change, 36, 121–144. Stren, R. (2014). Urban service delivery in Africa and the role of international assistance. Development Policy Review, 32(S1), s19–s37. Sumberg, J., Anyidoho, N. A., Chasukwa, M., Chinsinga, B., Leavy, J., Tadele, G., Whitfield, S. and Yaro, J. (2014). Young people, agriculture, and employment in rural Africa. WIDER Working Paper 2014/080. UNU/WIDER. Swyngedouw, E. (2004). Social power and the urbanisation of water: Flows of power. New York: Oxford University Press. 245 University of Ghana http://ugspace.ug.edu.gh Taylor, M. (2003). Public policy in the community. Houndmills, UK: Palgrave Macmillan. Terrell, S. R. (2012). Mixed-methods research methodologies. The Qualitative Report, 17(1), 254-280. Teye, K. J. (2012). Benefits, challenges, and dynamism of positionalities associated with mixed methods research in developing countries: Evidence from Ghana. Journal of Mixed Methods Research, 6, 379. The Presidency of Ghana (2016). State of the nation address presented to parliament of Ghana by His excellency John Dramani Mahama, President of the Republic of Ghana. Accra. Thurmond, V. A. (2001). The point of triangulation. Journal of Nursing Scholarship, 33(3), 253-258. Tibaijuka, A. (2009). Building prosperity: Housing and economic development. Earthscan, London. Tipple, A. G. (2005). The place of home-based enterprises in the informal sector: Evidence from Cochabamba, New Delhi, Surahay and Pretoria. Urban Studies, 42, 611–632. Tipple, A. G. (1993). Shelter as workplace: A review of home-based enterprises in developing countries. Labour Review, 132(4), 521–539. Todaro, M. and Smith, S. (2006). Economic development (9th edition). Pearson Addison Wesley, London. Turok, I. (2013). Securing the resurgence of African cities. Local Economy, 28, 142-157. Tutu, R. A. (2014). Dilemmatic experiences of young migrants in Accra, Ghana: The merciless hands of pirate urbanization and landlordism. Journal of Housing and the Built Environment, 29(4), 637-656. Twumasi, P. A. (2001). Social Research in Rural Communities (2nd edition). Ghana Universities Press, Accra. UNDESA (United Nations Department of Economic and Social Affairs) (2015). International decade for Action “Water for Life” 2005-2015. New York, United Nations. UNDESA (United Nations Department of Economic and Social Affairs) (2014). World Urbanization Prospects: The 2014 Revision, Highlights. (ST/ESA/SER.A/352). New York, United Nations. UN-Habitat (2014). The state of African cities: Re-imagining sustainable urban transitions. London: Earthscan. UN-Habitat (2011). Housing profile in Ghana. UN-Habitat, Nairobi. Urminsky, O., Bartels, D. M., Giuliano, P., Newman, G., Puntoni, S. and Rips, L. (2014). Choice and self: How synchronic and diachronic identity shape choices and decision making. 9th Triennial Choice Symposium, Marketing Letters. Van Rooijen, D. J., Spalthoff, D. and Raschid-Sally, L. (2008). Domestic water supply in Accra: How physical and social constraints to planning have greater consequences for the poor. 33rd WEDC International Conference, Accra, Ghana. Vizard, P. (2001). Economic theory, freedom and human rights: The work of Amartya Sen. ODI Briefing Paper. London: ODI. 246 University of Ghana http://ugspace.ug.edu.gh Wadel, C. (1991). Feltarbeid in their own culture- a innforing in qualitative oriented together discovery research. Flekkefjord: SEEK. Walby, S. (2001). Against epistemological chasm: The science question in feminism revisited. Signs, 26, 485-509. WB and IEA (World Bank and International Energy Agency) (2015). Sustainable Energy for All 2015: Progress Toward Sustainable Energy. World Bank, Washington, DC. Whitaker, G.P. (1980). Co-production: Citizen participation in service delivery. Public Administration Review, 40(3), 240–246. Whitfield, L. (2006). The politics of urban water reform in Ghana. Review of African Political Economy, 33(109), 425-448. WHO and UNICEF (World Health Organization and United Nations Children’s Fund) (2015). Progress on sanitation and drinking water: 2015 Update and MDG Assessment. New York, WHO/UNICEF. WHO and UNICEF (World Health Organization and United Nations Children’s Fund) (2014). Progress on drinking water and sanitation: 2014 Update. New York, WHO/UNICEF Joint Monitoring Programme (JMP) for Water Supply and Sanitation. Wisker, G. (2008). The postgraduate Research Handbook (2nd edition). Palgrave Macmillan. World Bank (2000). World Development Report 1999/2000: Entering the 21st century. Oxford University Press, Oxford. World Bank (2009). Underpowered: The State of the Power Sector in Sub-Saharan Africa. Africa Infrastructure Country Diagnosis. Background paper 6. The World Bank, Washington. World Bank (2010). City of Accra, Ghana: Consultative citizens’ report card. World Bank, Washington, DC. World Bank (2011). Raising Productivity and Reducing Risk of Household Enterprises in Ghana: Synthesis Country Report. Accra: Prepared by MEL Consulting Ltd. World Bank (2014). Ghana urbanization review (GUR). A presentation made at the Review Workshop of GUR, Ministry of Local Government and Rural Development, Accra, Ghana. World Bank (2015a). Rising through cities in Ghana: Ghana urbanization review overview report. Washington, DC. World Bank (2015b). Poverty Reduction in Ghana: Progress and Challenges. World Bank, Washington, DC. World Bank (2016). Eight things we know about water and electricity utilities in Africa. The Water Blog. The World Bank. World Cities Report (2016). Urbanisation and Development: Emerging Futures. UN-Habitat. WWAP (United Nations World Water Assessment Programme) (2015). The United Nations World Water Development Report 2015: Water for a Sustainable World. Paris, UNESCO. 247 University of Ghana http://ugspace.ug.edu.gh Yamane, T. (1967). Statistics: An introductory analysis (2nd edition). New York: Harper and Row. Yankson, P. W. K. (2000a). Accommodating informal economic units in the urban built environment: Petty commodity enterprises in the Accra Metropolitan Area, Ghana. Third World Planning Review, 22, 313–334. Yankson, P. W. K. (2000b). Houses and residential neighbourhoods as work places in urban areas: The case of selected low income residential areas in Greater Accra Metropolitan Area (GAMA), Ghana. Singapore Journal of Tropical Geography, 21, 200–214. Yankson, P. W. K. (2008). Decentralization and poverty reduction in the Gomoa District of Ghana. Norsk Geografisk Tidsskrift - Norwegian Journal of Geography, 62(3), 230- 240. Yankson, P. W. K., Kofie, R.Y. and Moller-Jensen, L. (2005). Monitoring urban growth: Urbanization of the fringe areas of Accra. Bulletin of the Ghana Geographical Association, 24, 1-12. Yansaneh, I. (2005). Household sample surveys in developing and transition countries. A United Nations Statistics Division publication. Yeboah, I. (2006). Subaltern strategies and development practice: Urban water privatization in Ghana. The Geographical Journal, 172(1), 50-65. Zug, S. and Graefe, O. (2014). The gift of water: Social redistribution of water among neighbours in Khartoum. Water Alternatives, 7(1), 140–159. 248 University of Ghana http://ugspace.ug.edu.gh APPENDICES Appendix 1: Questionnaire used for household survey DEPARTMENT OF GEOGRAPHY AND RESOURCE DEVELOPMENT UNIVERSITY OF GHANA Improving water and electricity access for poverty reduction in urban Ghana. This is a research project being undertaken by a PhD student of the Department of Geography and Resource Development, University of Ghana. The information is being gathered for academic purposes only and will be held in confidentiality. The identity of participants will not be disclosed and their privacy will be strictly observed. (Please read full consent language for verbal consent prior to conducting the survey). Please tick the appropriate answers where applicable. Verbal consent given [ ] Interviewer ______________ Start time _______ Respondent ______________ End time ________ Q.1 Questionnaire No. [ ] Q.2 Code [ ] Q.3 Neighbourhood i. Korle Gonno [ ] ii. New Town [ ] iii. Gbawe [ ] iv. Abuja [ ] A. ACCESS TO WATER A.1. From which of the following sources does this household get its water? (Tick all that apply) Source Tick 1. In -house connection 2. In -yard connection 3. O wn private water tank 4. P ublic/communal/vendor water tank 5. P ublic/communal/vendor standpipe/tap 6. O wn private borehole or well 7. P ublic/communal/vendor borehole or well 8. S achet water 9. O ther (specify): A.2. Among the list mentioned, which type of water do you use for each activity? Water usage Water source Reason 1. Drinking 2. Cooking 3. Washing 4. Laundry 5. Cleaning A.3. If you drink sachet water, how many sachets/bags do your household members use per week? .. If piped water ask A4 to A7, if no piped water jump to A8 A.4. How often does the tap water flow? a. All the time [ ] b. Five days in a week [ ] c. Three days in a week [ ] d. Two days in a week [ ] e. Once a week [ ] f. Other ............................. A.5. On the day it flows, which periods of the day does pipe water flow? a. 24 hours [ ] b. 06:00 am - 12:00 noon [ ] c. 12:00 noon - 06:00 pm [ ] d. 06:00 pm - 00:00 midnight [ ] e. 00:00 midnight - 06:00 am [ ] f. Other ........... A.6. On the day it doesn’t flow, how do you get water? a. Stored water [ ] b. Vendors [ ] c. Other ............................... A.7. How much is your monthly bill? GHC .......................... 249 University of Ghana http://ugspace.ug.edu.gh A.8. If you should have pipe connection how much are you willing to pay monthly? GHC ............. A.9. Who primarily fetches the water for your household? a. Household head [ ] b. Spouse [ ] c. Children [ ] d. Other relative [ ] f. Other ................ A.10. How many minutes/hours (per day) do your household members spend in getting water for household consumption (include time spent at source)? a. 0-15 mins [ ] b. 16-30 mins [ ] c. 31-45 mins [ ] d. 46 mins - 1 hour [ ] e. More than1 hour but less than 2 hours [ ] f. 2 hours and more [ ] A.11. What is the estimated total quantity of domestic water (in gallons - Yellow Gallon) used by your household per day? a. Between 1 and 3 [ ] b. Between 4 and 6 [ ] c. Between 7 and 9 [ ] d. Between 10 and 15 [ ] e. More than 15 [ ] A.12. How often does your household buy water from a vendor? a. Every day [ ] b. At least once a week [ ] c. At least once a month [ ] d. Very occasionally [ ] e. Never uses water from vendors [ ] f. Others…………… A.13. If you buy water from public tap or vendors, how much do you pay per use (gallon/bucket)? Water source Amount per gallon/bucket Total Amount Daily 1. Public/vendor standpipe/tap 2. Public/vendor water tank 3. Public/vendor borehole or well A.14. Who pays for water for this household? a. Household head [ ] b. Spouse [ ] c. Children [ ] d. Other relative [ ] e. Others ............ A.15. Is the cost of water expensive? a. Yes [ ] b. No [ ] c. Not sure [ ] A.16. How does the cost of water affect your household? ................................................................. ............................................................................................................................................................. A.17. What are the other key challenges associated with accessing water for this household? ................................................................................................................................................................. A.18. What do you use to store water? a. Buckets [ ] b. Gallons [ ] c. Jerrycans [ ] d. Polytanks [ ] e. Other ................ A.19. Are there any problems with the quality of stored water? a. Yes [ ] b. No [ ] A.20. Please explain ..................................................................................................................... ..... A.21. Rank on a scale of 1-5, which of the following factors best explains how you access water? 1= Strongly Disagree; 2 = Disagree; 3 = Indifferent; 4 = Agree; 5= Strongly Agree Item 1 2 3 4 5 1. Cost of water influences the type of water used? 2. Source of water influences the type of water used? 3. Distance to water source influences the type of water used? 4. Reliability of water supply influences the type of water used? 5. Household size influences the type of water used? A.22. Have you or any household member suffered from any of these diseases as a result of poor quality water? (Tick all that apply) Disease Tick Disease Tick 1. Diarrhoea 3. Skin rash 2. Stomach aches 4. None A.23. Rank on a scale of 1-5, how much you agree with the following statements. 1 = Strongly Disagree; 2 = Disagree; 3 = Indifferent; 4 = Agree; 5 = Strongly Agree 250 University of Ghana http://ugspace.ug.edu.gh Item 1 2 3 4 5 1. Water related diseases reduce/ interrupt school attendance 2. Water related diseases reduce/ interrupt working time/hours 3. Water related diseases reduce household productivity 4. Water related diseases increase household poverty B. LIVELIHOOD ACTIVITIES AND ACCESS TO WATER B.1. Do you or any household member operate any income generating activity? If No jump to C.1 a. Yes [ ] b. No [ ] B.2. Which business activities do you or any household member do? Relationship to Type of business Location When business No. of workers household head started 1 2 3 4 Location = a. Within the house/compound b. Frontage of house c. Roadside/market but within the community d. Outside the community [ ] e. Other [ Specify ] If business is located in the community continue, if not jump to C1. B.3. Do you pay rent for the working space? a. Yes [ ] b. No [ ] B.4. If yes, please state the amount per month/year GHC .......................... B.5. Do you make any other payments (e.g. tax) to the city authorities? ....................................... B.6. Do you use water in your business? a. Yes [ ] b. No [ ] If no skip to C1 B.7. If yes, please indicate the source, quantity and cost of water used by your household members Relationship to Activity Source of Quantity of Cost of water household head water water used (daily/weekly) 1 2 3 4 B.8. What percentage of your income do you spend on water for income generating activities? a. 1-10% [ ] b. 11-30% [ ] c. 31-50% [ ] d. 51- 70% [ ] e. 71-100% [ ] B.9. Who primarily fetches the water for your business? a. Myself [ ] b. Apprentice [ ] c. Other worker [ ] d. House help [ ] e. Other .............. B.10. How many minutes/hours (per day) do you spend in getting water for your business (include time spent at source)? a. 0-15 mins [ ] b. 16-30 mins [ ] c. 31-45 mins [ ] d. 46 mins - 1 hour [ ] e. More than1 hour but less than 2 hours [ ] f. 2 hours and more [ ] g. None [ ] B.11. Rank on a scale of 1-5, how much you agree with the following statements? 1 = Strongly Disagree; 2 = Disagree; 3 = Indifferent; 4 = Agree; 5 = Strongly Agree Item 1 2 3 4 5 1. The quantity of water is adequate for my business 2. The quality of water is good enough for my business 3. The cost of water is affordable for my business 251 University of Ghana http://ugspace.ug.edu.gh 4. Improved access to water can enhance household livelihoods 5. Water related livelihoods can help reduce household poverty B.12. In which ways does access to water affect your business? ........................................................... ............................................................................................................................. .................................... . B.13. In which ways could improved access to water enhance your business? ............................................................................................................................. .................................... C. ACCESS TO ELECTRICITY C.1. From which of the following sources does this household get its power? (Tick all that apply) Source Tick 1. Household metered connection 2. House metered connection 3. Outside house connection 4. No connection 5. Others (specify) If there is connection ask C2 to C8, if no connection jump to C9 C.2. Which meter type do you use? a. Pre-paid [ ] b. Post-paid [ ] C.3. How much do you pay per month? GHC .......................... C.4. What arrangements are in place for paying for power by your household? a. Household private meter payment [ ] b. Household shared meter payment [ ] c. Household meter payment in turns [ ] d. Others ................................................................ C.5. How often does the power go off? ............................................................................................... C.6. Which periods of the day does the power most often go off? ..................................................... C.7. When there is a power cut, which alternative sources of power do you use? a. Generator [ ] b. Rechargeable lamp [ ] c. Battery lamp [ ] d. Torch [ ] e. Candle [ ] f. Other C.8. How much do you pay per month for your alternative sources? GHC .......................... If no connection, ask C9-11 C.9. If no connection, which alternative sources of power do you use? a. Generator [ ] b. Rechargeable lamp [ ] c. Battery lamp [ ] d. Torch [ ] e. Candle [ ] f. Other C.10. How much do you spend on power per month? GHC .......................... C.11 If you should have power connection how much are you willing to pay monthly? GHC .......... Ask all C.12. Who pays for power for this household? a. Household head [ ] b. Spouse [ ] c. Children [ ] d. Other relative [ ] e. Others .............. C.13. Is the cost of power expensive? a. Yes [ ] b. No [ ] c. Not sure [ ] C.14. How does the cost of power affect your household? ........................................................... ........................................................................................................................................ ..................... C.15. What are the other major concerns regarding electricity in this household? ............................................................................................................................................................... 252 University of Ghana http://ugspace.ug.edu.gh C.16. Rank on a scale of 1-5, which of the following factors best explains how you access power? 1= Strongly Disagree; 2 = Disagree; 3 = Indifferent; 4 = Agree; 5= Strongly Agree Item 1 2 3 4 5 1. Cost of power influences how you use power? 2. Reliability of power supply influences how you use power? 3. Household size influences how you use power? D. LIVELIHOOD ACTIVITIES AND ACCESS TO ELECTRICITY If no livelihood activity jump to E1 D.1 If business is located in the community, do you use power in your business? a. Yes [ ] b. No [ ] D.2 If yes, please indicate the source, reliability and cost of power used by your household members Relationship to Activity Source of power Reliability of Cost of power household head supply (weekly/monthly) 1 2 3 4 D.3. When there is a power cut, which alternative sources of power do you use for your business? a. Generator [ ] b. Rechargeable lamp [ ] c. Battery lamp [ ] d. Torch [ ] e. Candle [ ] f. Other D.4. Do you have a separate meter for your business? a. Yes [ ] b. No [ ] D.5. How much do you pay monthly for power for your business? GHC .......................... D.6. What percentage of your income do you spend on power for income generating activities? a. 1-10% [ ] b. 11-30% [ ] c. 31-50% [ ] d. 51- 70% [ ] e. 71-100% [ ] D.7. Rank on a scale of 1-5, how much you agree with the following statements? 1 = Strongly Disagree; 2 = Disagree; 3 = Indifferent; 4 = Agree; 5 = Strongly Agree Item 1 2 3 4 5 1. The quality of power is good enough for my business 2. The cost of power is affordable for my business 3. Improved access to power can enhance household livelihoods 4. Electricity related livelihoods can help reduce household poverty D.8. In which ways does access to electricity affect your business? ................................................... ............................................................................................................................. .................................. D.9. In which ways could improved access to electricity enhance your business? ............................................................................................................................. ................................. E. IMPROVING WATER AND ELECTRICITY ACCESS FOR ENHANCED LIVELIHOODS AND POVERTY REDUCTION Water E.1 At which level do you think it is best to try to resolve any water problems? a. Household level [ ] b. Community level [ ] c. Other .................................. E.2. What intervention has the government (e.g. water boards, local assembly) made to resolve any water issues? ............................................................................................................... ........................ ................................................................................................................................................................. 253 University of Ghana http://ugspace.ug.edu.gh E.3. What initiative has non-government organisations (NGOs) taken to resolve any water issues? ................................................................................................................................................................. E.4. Do you think something else should be done? a. Yes [ ] b. No [ ] E.5. If yes, who should do it? ............................................................................................................... E.6. If I were a leader in the community, regarding water issues, I would ......................................... ................................................................................................................................................................. Electricity E.7. At which level do you think it is best to try to resolve any power problems? a. Household level [ ] b. Community level [ ] c. Other .................................. E.8. What intervention has the government (e.g. water boards, local assembly) made to resolve any power issues? ....................................................................................................................................... E.9. What initiative has non-government organisations (NGOs) taken to resolve any power issues? ………...................................................................................................................................... E.10. Do you think something else should be done? a. Yes [ ] b. No [ ] E.11. If yes, who should do it? ............................................................................................................. E.12. If I were a leader in the community, regarding electricity issues, I would ................................. ............................................................................................................................. .................................... Services and poverty E.13. On a scale of 1- 5, rank in order of importance which are the most serious concerns and your household expenditure regarding services. (1 being the most serious concern) Services Rank by concerns Rank by expenditure a. Water b. Electricity c. Sanitation d. Transport e. Other E.14. If we talk about poverty, which factors contribute to it? .......................................................... ............................................................................................................................................................. . E.15. What would you recommend should be done to reduce household poverty? ............................................................................................................................................................... F. RESPONDENT DETAILS/ SOCIO- DEMOGRAPHIC DATA F.1. Gender of Respondent a. Female [ ] b. Male [ ] F.2 Age of Respondent a. 18-29 [ ] b. 30-39 [ ] c. 40-49 [ ] d. 50- 59 [ ] e. 60+ [ ] F.3 Ethnic Origin a. Akan [ ] b. Ewe [ ] c. Ga-Adangbe [ ] d. Northern extraction [ ] e. Foreigner [ ] f. Others ……………… F.4 What is your highest educational attainment? a. None [ ] b. Non-Formal Educ. [ ] c. Primary [ ] d. Middle/JSS [ ] e. SSS/O’Level [ ] f. Comm/Voc/Technical [ ] g. Post Sec./Nursing/Polytechnic [ ] h. University [ ] i. Other (specify) [ ] ________________________________ F.5 Religion of Respondent a. Christian [ ] b. Muslim [ ] c. Traditional [ ] d. Others ...... F.6 What is your marital status? a. Married [ ] b. Separated [ ] c. Co-habiting /Consensual Union [ ] d. Widowed [ ] e. Divorced [ ] f. Single/ Never married [ ] 254 University of Ghana http://ugspace.ug.edu.gh F.7 Where did you (mainly) grow up? ......................................................................................... ... F.8 How long have you lived in this community? ............................................................................. F.9 How long have you lived in this house/dwelling? ....................................................................... F.10. What is the nature of your tenancy arrangement? a. Landlord/landlady [ ] b. Tenant [ ] c. Caretaker [ ] d. Free habitation/family [ ] e. Other [ ] F.11. Which housing type do you live in? a. Compound [ ] b. Detached [ ] c. Semi-Detached [ ] d. Apartment [ ] e. Others ......… F.12. How many rooms do you occupy? ..................................................................................... F.13. How many households are in this house? ................................................................................... F.14. What is the size of your household? ............................................................................ F.15. Are you the household head? Yes [ ] b. No [ ] If yes jump to F 17 F.16. If No, what is your relationship with the household head? ........................................................... F.17 Which of the following applies to you in terms of your employment status? (Tick all that apply) a. Unemployed [ ] b. Self-employed [ ] c. Full time for wages [ ] d. Part time for wages [ ] e. Informal worker [ ] f. Study full time or part time [ ] g. Other (specify): _______________ F.18. What is the occupation and level of income of your household members (approximate)? Relationship to Main occupation Average Alternative Average household head monthly livelihoods monthly income income 1 2 3 4 5 F.19. Have you previously run a business? a. Yes [ ] b. No [ ] Please specify ..................... F.20. Have you changed business? a. Yes [ ] b. No [ ] F.21. If yes, what type ............................................ and where .............................................. ....... F.22. Why did you stop? ................................................................................................................ F.23. Do you have any plans to set up a business in the near future? a. Yes [ ] b. No [ ] F.24. If yes, what type ............................................ F.25Where would you like to locate your business? a. Within community [ ] b. Outside community[] F.26 Please explain your answer ..................................................................................................... F.27 Does any member of your household own any of the following physical assets? Physical assets Yes/No Physical assets Yes/No Physical assets Yes/No House Refrigerator Desktop/laptop Land Microwave Television Car Electric iron Radio/tape Motor bike Fan Mobile phone Thank you for participating in this survey… 255 University of Ghana http://ugspace.ug.edu.gh Appendix 2: Interview guide for in-depth interviews Experts and Service providers Please indicate your name and position in this organisation _______________________________ 1. Water and electricity access and governance  What is the state of water and electricity services delivery? o What in your opinion accounts for the challenges in service delivery? o What improvements are needed? What are you doing about the situation (drinkable water)?  What factors are considered in the provision of water and electricity for low-income areas? o What are the connection procedures? o Any pro-poor initiatives? How is it implemented? Any specific challenges? o What municipal-community interface exist in the delivery of services? 2. Livelihoods and poverty reduction linkages  In what ways can service delivery enhance household businesses in low-income areas?  What is your view on the use of domestic water and electricity supply for productive economic activities? o How does this relate to quality of service delivered? o Any special intervention to improve service delivery and promote the local economy?  Do you share the view that improving services for household business activities increases the willingness and affordability of services?  In what ways can improved service delivery reduce poverty in low-income areas? o What strategies are needed? What must be done differently?  What contribution have you made in your capacity to help solve the water and electricity situation in GAMA? How do you want the state of urban services to be in the next 10 years? Key Informants Please indicate your status/ position in this community ________________________________ Services  Since you lived in this community, how has water and electricity access improved or deteriorated? o What do you think accounts for this situation? o What initiatives have been taken to address the situation (assembly, chiefs, NGO/CBO, individuals)? o What improvements are needed? Can community participation be a panacea? Livelihoods and poverty  Does the issue of water and electricity form a major part of the district assembly?  What should be done to ensure that access to water and power is improved to enhance livelihoods? 256 University of Ghana http://ugspace.ug.edu.gh o Any problems about the domestic vs. productive uses of water and power? o How are livelihood activities changing because of poor service delivery?  In what ways can improved access to water and electricity reduce household poverty?  Which of the problems faced in this community do you think should be the priority focus of leaders in this community? o What mechanisms are available for redressing challenges regarding urban services?  What contribution have you made in your capacity to help solve the water and electricity situation in your community? o If you were the assembly man/woman (or person in authority), what would you do to make life better for the people living here? 257 University of Ghana http://ugspace.ug.edu.gh Appendix 3: Guide for Focus Group Discussions 1. Access to services  How do residents in this neighbourhood access the following services?: o Water? (sources/types, shared pipes, bill, pay for connection) o Electricity? (sources/types, shared meters, bill, pay for connection)  How does cost, distance, quality, and reliability influence access and choice? o Are there differences in access (Old town and new site)?  Potential household problems, conflicts and stress? What improvements are needed?  Are people moving to/from the neighbourhood because the services are good/poor? 2. Livelihood  How do people living in this neighbourhood make their money (main economic activities)? o Has this changed over time? How and why? o Do you think people are diverting into other forms of trade? (coping strategies)  How does service availability affect income-generating activities in this neighbourhood? o Any benefits and challenges? o If you use water or power in your business are you willing to pay more to get access (e.g. get metered pipe/pre-paid and pay at a commercial rate (domestic and productive uses of water/power)?  What influences choice of water and power sources for economic activities? 3. Poverty  How do people perceive poverty/wealth (multi-dimensionality) in this neighbourhood? o What are their coping strategies? o Are there any associations? How have they evolved? Any membership requirements?  How does service availability relate to household poverty in this neighbourhood?  What improvements are needed (vulnerable groups) 4. Governance  How do people perceive governance in relation to water and electricity in this neighbourhood? o Has this changed over time? How and why?  How can community participation improve access to services and reduce poverty? o Possible role of local assembly, CBOs, NGOs etc.?  What are the available mechanisms for residents to address their service needs? o Who has more influence to advocate for change? (individuals, groups, local assembly)? 5. Conclusion  In what ways can the water and electricity situation of this neighbourhood be improved to reduce poverty? If possible rank from 1 to 5 the priority suggestions for improvement. 258 University of Ghana http://ugspace.ug.edu.gh Appendix 4: Statistical analysis of household survey data Pearson chi-square test showing the relationship between settlements and meter type Settlement type Prepaid Post-paid Total Korle Gonno 97 1 98 Accra New Town 33 65 98 Abuja 89 2 91 Gbawe 95 1 96 Total 314 69 383 Chi-square = 2.082; df = 3; probability = 0.000 259 University of Ghana http://ugspace.ug.edu.gh Appendix 5: Characteristics of GAMA Composition and demographic characteristics of GAMA Political Administrative Capital Total % Share % % No. of Areas Population of Total Urban Rural households Population Accra Metropolitan Accra 1848614 46.1 100 - 501956 Area (AMA) Sub-Metros in AMA • Ablekuma South 213914 5.3 57913 • Ablekuma Central 268424 6.7 72230 • Ashiedu Keteke 117525 2.9 34964 • Osu Klotey 121723 3 35508 • La 183528 4.6 51155 • Ayawaso East 183498 4.6 43755 • Ayawaso Central 142322 3.5 39116 • Okai Koi South 121718 3 34800 • Ablekuma North 197024 4.9 53039 • Okai Koi North 228271 5.7 64567 • Ayawaso West 70667 1.8 14909 Wuogon Tema Metropolitan Tema 402637 10 97 3 97597 Area(TMA) Sub-Metros in TMA • Tema West 127807 3.2 31154 • Tema East 160213 4 38508 • Kpone Katamanso 114617 2.9 27935 Ga South Municipal Gbawe 485643 12.1 89 11 118846 Area(GSMA) Ga West Municipal Amasaman 262742 6.6 69 31 66706 Area(GWMA) Ga East Municipal Area Abokobi 259668 6.5 87 13 66286 (GEMA) Adenta Municipal Adenta 78215 2 64 36 20478 Area(AdMA) Ledzokuku / Krowor Nungua 227932 5.7 100 - 60859 Municipal Area (LEKMA) Ashaiman Municipal Ashaiman 190972 4.8 100 - 49936 Area(AshMA) La Dade-Kotopon La - - - - - Municipal area13 13New municipality carved from AMA in 2012. The 2010 Population & Housing survey however enumerated it as part of the AMA; 3New municipality carved from TMA in 2012. The 2010 Population & Housing survey however enumerated it as part of the TMA; 4 New municipality carved from Ga East in 2012. The 2010 Population & Housing survey however enumerated it as part of Ga East 260 University of Ghana http://ugspace.ug.edu.gh Kpone Katamanso Kpone - - - - - Municipality3 La-Nkwantanang- Madina - - - - - Madina Municipality4 Ga Central District5 Sowutuom - - - - - Source: Generated from the 2010 Population and Housing Census (GSS, 2012) district;5 New district carved from Ga South in 2012. The 2010 Population & Housing survey however enumerated it as part of Ga South district. 261