Understanding policy implementation at the local government level: the case of the Johannesburg Inner City Housing Implementation Plan (ICHIP) Colleen Orsmond A research report submitted to the Faculty of Engineering and the Built Environment, University of the Witwatersrand, in partial fulfilment of the requirements for the degree of Master of Science in Development Planning. Johannesburg, 2022 Contents Declaration .............................................................................................................................................. i Abstract.................................................................................................................................................... ii Acknowledgements ............................................................................................................................ iii List of Figures ....................................................................................................................................... iv List of Tables .......................................................................................................................................... v List of Acronyms.................................................................................................................................. vi 1 Introduction .............................................................................................................................. 1 1.1 Introduction ..................................................................................................................... 1 1.2 Background ...................................................................................................................... 2 1.3 Problem statement and rationale ............................................................................ 3 1.4 Objectives .......................................................................................................................... 3 1.5 Research questions ....................................................................................................... 4 1.6 Methods ............................................................................................................................. 4 1.7 Outline of the research report................................................................................... 6 2 Policy implementation theory: a work in progress ................................................... 7 2.1 Introduction ..................................................................................................................... 7 2.2 The evolution of implementation theory, models and frameworks .......... 8 2.3 Policy implementation in the Global South and South Africa .....................12 2.4 Minimising the policy implementation gap .......................................................15 2.5 Conceptual framework: the 7-C Protocol and ICHIP ......................................18 2.6 Conclusion.......................................................................................................................20 3 Johannesburg Inner City Housing: it’s complicated ................................................22 3.1 Introduction ...................................................................................................................22 3.2 Framing local government housing policy .........................................................22 3.3 Housing in the Johannesburg Inner City .............................................................23 3.4 Inner city housing in Johannesburg today .........................................................26 3.5 Inner city housing demand and supply ...............................................................28 3.6 City of Johannesburg strategic and governance framework .......................31 3.7 Conclusion.......................................................................................................................33 4 Assessing ICHIP’s implementation ................................................................................34 4.1 Introduction ...................................................................................................................34 4.2 The status of ICHIP, and why it matters ..............................................................35 4.3 Implementation status of the ICHIP programmes ..........................................40 4.4 The continued relevance of ICHIP for the City of Johannesburg ...............48 4.5 Conclusion.......................................................................................................................49 5 ICHIP implementation through a 7-C Protocol lens ................................................50 5.1 Introduction ...................................................................................................................50 5.2 Context: between a rock and a hard place .........................................................51 5.3 Content: nice in theory but too ambitious ..........................................................52 5.4 Commitment: individual agency and political champions ...........................54 5.5 Capacity: levers and leadership ..............................................................................56 5.6 Clients and coalitions, cynicism and distrust ....................................................58 5.7 Communication: the sound of one hand clapping ...........................................60 5.8 Coordination: the whole is less than the sum of its parts ............................62 5.9 Conclusion.......................................................................................................................63 6 Conclusion ...............................................................................................................................65 6.1 Introduction ...................................................................................................................65 6.2 ICHIP implementation: something is happening, but not necessarily ICHIP 66 6.3 Discussion: Where ICHIP implementation meets theory .............................67 6.4 Recommendations: Leadership and Partnership ............................................69 6.5 Conclusion: An ICHIP ‘toolbox’ is not good enough ........................................70 References ............................................................................................................................................72 Appendix A: Personal Communications List ...........................................................................78 Appendix B: Participant Information Sheets ..........................................................................80 Appendix C: Consent Form .............................................................................................................87 Appendix D: Sample Interview Schedule ..................................................................................88 i Declaration I declare that this research report is my own unaided work. It is being submitted for the degree of Master of Science in Development Planning to the University of the Witwatersrand, Johannesburg. It has not been submitted before for any degree or examination to any other University. …………………………………………………………………………… Colleen Orsmond 29th day of November, 2022 Ethics Clearance No.: SOAP138/06/2021 ii Abstract A smooth progression from policy design to implementation and successful realisation of the intended outcomes cannot be taken for granted. Instead, the interaction between policy, policy implementers and the context in which it is implemented create outcomes that can diverge significantly from the original policy intent. As a case study for policy implementation at the local government level, examining the implementation of the City of Johannesburg’s Inner City Housing Implementation Plan (ICHIP) revealed a challenging socio-economic, political and institutional context that has frustrated progress. Public and private sectors recognised ICHIP as a comprehensive albeit ambitious strategy. Piecemeal and fragmented delivery, weak leadership and a failure to cement a partnership between the City and private sector actors has meant that while certain outputs have been delivered in small measure, the policy outcome of a robust inner city housing market that works for the poor has not been realised. iii Acknowledgements To my supervisor, Margot Rubin, thank you for your guidance and insight, your flexibility on deadlines and for pushing me to do better. I’m grateful that you stuck with me. To JP, thank you for your empathy and allowing me the space to do this work. To my family and friends, thanks for kindly listening to my constant complaints about being tired and busy. And to Ash, thank you for your patience and support. Finally, to the people that I interviewed, thank you for sharing your thoughts and perspectives so generously with me. I enjoyed the interactions very much and hope that our paths cross again as actors working to make the inner city a better place for its residents. iv List of Figures Figure 1.1: City of Johannesburg and private sector interviewees .................................. 5 Figure 2.1: 7-Cs Protocol as a framework for assessing ICHIP implementation ......20 Figure 3.1: Organisations, Programmes and Strategies impacting housing in the Johannesburg inner city, 1992 to 2016 .............................................................25 Figure 3.2: Inner city households by annual income band, 2011 ..................................29 Figure 3.3: Sources of inner city housing supply ..................................................................30 Figure 3.4: Inner city housing actors..........................................................................................31 Figure 4.1: Extract of CoJ Council Minutes, 27/28 September 2017 ............................36 Figure 4.2: Header section of Final ICHIP Report to Council, 30 August 2018 .........38 Figure 4.3: ICHIP Delivery and Facilitation Programmes .................................................40 file:///C:/Users/corsmond/Documents/Wits/MScDP/Research%20Report/Draft%20chapters/Research%20Report_Full%20v1.docx%23_Toc99277282 v List of Tables Table 3.1: Overcrowding in five inner city precincts, 2011 ..............................................28 vi List of Acronyms AFHCO Africa Housing Company BASA Banking Association of South Africa BBB-EE Broad-based Black Economic Empowerment BBP Bad Buildings Programme BNG Breaking New Ground CAPEX Capital Expenditure CEO Chief Executive Officer CoJ City of Johannesburg CRU Community Residential Units CRUM Customer Relationship and Urban Management DED Department of Economic Development EMS Emergency Medical Service FLISP Finance Linked Individual Subsidy Programme GDS Growth and Development Strategy GPMA Gauteng Precinct Management Association GTAC Government Technical Advisory Centre HDA The Housing Development Agency ICHIP Inner City Housing Implementation Plan ICPS Inner City Property Scheme ICTR Inner City Transformation Roadmap IDP Integrated Development Plan IMU Inclusive Mixed Use ISP Institutional Subsidy Programme JDA Johannesburg Development Agency JHC Johannesburg Housing Company JICP Johannesburg Inner City Partnership JOSHCO Johannesburg Social Housing Company JPC Johannesburg Property Company JPOMA Johannesburg Property Owners and Managers Association vii JWA Johannesburg Water Authority KPI Key Performance Indicator ME Municipal Entities MES Mould Empower Serve MHA Madulammoho Housing Association MMC Member of the Mayoral Committee NDoHS National Department of Human Settlements NDP National Development Plan NHFC National Housing Finance Corporation Ltd NIMTO Not In My Political Term NPM New Public Management OPEX Operational Expenditure PHPS Public Housing Programme Support PPPF Preferential Procurement Policy Framework RDP Reconstruction and Development Programme SDF Spatial Development Framework SERI Socio-Economic Rights Institute of South Africa SHI Social Housing Institution SHRA Social Housing Regulatory Authority SMME Small, Medium and Micro Enterprises SPRE Special Programme for the Relocation of Evictees TEA Temporary Emergency Accommodation TEAP Temporary Emergency Accommodation Policy TUHF Trust for Urban Housing Finance UDZ Urban Development Zone 1 1 Introduction ‘The best laid plans of mice and men often go awry’ Robert Burns, “To a Mouse”, 1785 1.1 Introduction A smooth progression from policy design to implementation and realisation of the intended outcomes cannot be taken for granted. Policy-implementation gaps - or differences between policy goals and what is actually realised - are evident at all levels of government today, even more so as the context in which policies are implemented becomes more complex (Makinde, 2005; Ansell, 2017). Brynard (2005:658) noted that implementing public policy is a ‘complex political process, rather than a mechanistic administrative one’ and this much is evident in the case of the Johannesburg Inner City Housing Implementation Plan (ICHIP). As a case study for policy implementation at the local government level, exploring the nature of ICHIP’s implementation revealed a challenging and complex socio-economic, political and institutional context that has frustrated progress, resulting in a limited and fragmented delivery. ICHIP was commissioned by the City of Johannesburg (CoJ) Housing Department in 2015 with the intended outcome of ‘making the inner city housing market work better for people with lower incomes’ (CoJ Housing Department, 2017a:2). With demand for inner city housing far outstripping supply, particularly at the lowest end of the market, ICHIP aimed to stimulate the supply of affordable rental housing through eleven programmes targeting both public and private sector provision. A range of actors from the public and private sectors interviewed during the course of the research project recognised ICHIP as a comprehensive but ambitious strategy, and it was perhaps in trying to do too much that it erred given the implementing capacity of the City. The holistic approach is fundamental to ICHIP however, in that it recognises that all rungs of the housing ladder need to be strong 2 to ensure a healthy inner city housing ecosystem. But limited resources, weak leadership and a failure to cement a partnership between the City and private sector actors has meant that while certain outputs may have been delivered in small measure, the intended policy outcome of a robust inner city housing market that works for the poor has not been realised. 1.2 Background The Johannesburg inner city is a place of opportunity, an entry point for many who come to Johannesburg looking for employment. As the primary trade and transport hub of the metropolitan region, it is under constant pressure to provide housing for urban immigrants, but a lack of decent, cheap rental accommodation has created high levels of informality, stressed buildings, overcrowding, and unsafe living conditions. A study of the 2011 census data and sample inner-city precincts by the team that developed ICHIP estimated that the number of households living in the inner city exceeded the supply of housing available by approximately 92’000 units, and that the bulk of demand was for very cheap accommodation. The study also showed a housing market responding to the housing supply shortage: the ICHIP report states that in the six years prior, 50 000 new apartments had been delivered in the Johannesburg inner city. But these were largely in the gap market, so not priced low enough to serve the majority of the residents seeking accommodation. This left the informal market to cater to the very poor using spaces not designed for housing and putting a number of households in precarious, high-risk housing situations. (RebelGroup, 2016) ICHIP was commissioned in response to a call for an updated housing action plan articulated in the 2013 Inner City Transformation Roadmap (ICTR). It was developed by a multi-disciplinary team who consulted with the Johannesburg Development Agency (JDA), the Johannesburg Social Housing Company (JOSHCO) and the CoJ Departments of Housing and Development Planning (RebelGroup, 2016). In October 2017, a media statement issued by then-mayor Herman Mashaba announced that City Council had approved a plan to ‘tackle the housing 3 challenge within the inner city’ (JICP, 2017: no page number), however the research process revealed that that Council had not, and has still not to this day, approved ICHIP as a housing strategy. This matter of administrative process alongside a number of other factors has significantly impacted its implementation, and are discussed in detail in chapter 5 of this report. 1.3 Problem statement and rationale ICHIP appeared to be a robust, comprehensive strategy that included detailed project implementation plans, set out administrative requirements, actions, timelines, and budgets. It responded to a clear need for affordable inner city housing, particularly for the very poor, identified in the Inner City Transformation Roadmap. Yet in 2021, four years after the CoJ media statement announcing the Council’s approval of the plan, a scan of CoJ reports, presentations and documents revealed only an occasional mention of the word ‘ICHIP'. While this does not necessarily mean that implementation has failed or that the policy has been abandoned, it raises the question as to the embeddedness of the strategy and how much has actually happened since the announcement. Understanding the extent to which ICHIP has been implemented and how any outcomes compare to the original intentions offers insight into policy implementation at the local government level in South Africa’s largest metropolitan municipality. While the research process revealed that the ICHIP was a strategy rather than a policy of the City of Johannesburg, the parallels that can be drawn between policies and strategies as sets of actions identified to address a certain issue or problem render the ICHIP implementation relevant to policy implementation studies. 1.4 Objectives This research aimed to identify and understand the reasons behind ICHIP’s implementation successes and weaknesses and to relate this understanding to the existing body of literature on the policy-implementation gap. Various theories and models of policy implementation have been developed but a number of scholars contend that the discipline lacks a unifying theory, and that there is a need for 4 further research in this regard. In considering how the case-and context-specific data obtained from the study of ICHIP’s implementation relates to different policy implementation theories or models, this study aims to contribute to the body of knowledge relating to policy-implementation gaps. 1.5 Research questions Main research question: To what extent has the City of Johannesburg ICHIP been implemented and what are the reasons for implementation successes or failures? Sub-questions: Which aspects of the ICHIP are being or have been implemented according to the ICHIP plan and by which entities? How successfully are they being implemented as measured against required outputs, timelines and budgets? Which aspects of the ICHIP are not being implemented – at all or to a limited extent - and for what reasons? To what do stakeholders and role players in the ICHIP attribute implementation successes or failures? How do the ICHIP interventions or programmes integrate into existing CoJ housing and development strategies? 1.6 Methods This research project used a single case study approach to explore the extent to which ICHIP has been implemented and to explain the reasons for its successes, failures and weaknesses. The objective is to generate context-dependent knowledge rather than context-independent rational analysis (Flyvbjerg, 2006). In this sense, the goal of the case study is to generate a deep understanding the ‘real- world behaviour’ of ICHIP’s implementation by examining the contextual and other complex conditions that relate to it (Yin, 2011:5). The study employed a largely 5 qualitative method comprising a review of CoJ policies, programmes and Council minutes and any media articles and statements relating to ICHIP implementation, followed by semi-structured interviews with key informants and experts. I used purposive sampling to select initial interviewees and then snowball sampling to identify additional participants. Samples of the participant information sheet, consent form and interview guide are provided in Appendix B, C and D respectively. I interviewed representatives from the CoJ Housing and Planning departments, the CoJ Inner City Office and the Johannesburg Development Agency (JDA). Private sector participants represented Social Housing Institutions (SHIs), private landlords, independent consultants and academics. I did not manage to identify and interview any representatives from JOSHCO or the JPC, municipal entities that are mentioned often in this report and are key actors in a number of ICHIP programmes. A number of the interviewees provided perspectives on their involvement and I used their annual business plans to make additional deductions. A full list of the research participants is shown in Figure 1.1 below and a comprehensive list of the interviewees and interview dates is provided in Appendix A. Figure 1.1: City of Johannesburg and private sector interviewees CITY OF JOHANNESBURG PRIVATE SECTOR Housing Department Housing Finance Social Housing Institutions Developer / Landlord Inner City Office Civil society Development Planning Department Consultants and academics Johannesburg Development Agency Emanuel Sotomi Director: Public Housing Programme Support JDA Official (anonymous) Tanya Zack Urban Planner Renney Plit Chairman: Plitvest (Pty) Ltd Anne Steffny Director: JICP, GPMA (NPCs) Katherine Cox R&D and Development Impact Manager: TUHF SHI Manager (anonymous) Social Housing Institution Msizi Khuhlane Assistant Director: Intergovernmental Relations, Policy & Research Elize Stroebel Chief Executive Officer: Johannesburg Housing Company Housing Department Official 1 (anonymous) Housing Department Official 2 (anonymous) Housing Department Official 3 (anonymous) SHI Executive (anonymous) Social Housing Institution Planning Official (anonymous) Andreas Bertoldi Managing Partner: RebelGroup South Africa Neil Klug Senior Lecturer: Wits University School of Architecture & Planning Thabo Maisela Director: Inner City Office & Special Advisor to Mayor on Inner City Regeneration 6 The interviews were recorded and transcribed, then coded in two ways: the transcripts were first filtered to identify the extent to which different components of ICHIP have been implemented, and then to identify the interviewees’ opinions and perspectives on the reasons behind implementation challenges. I categorised these opinions and perspectives using a conceptual framework based on the ‘7-C Protocol’ described in more detail in chapter 2. Through this process I identified recurring themes and perspectives that informed my assessment of the key reasons for ICHIP’s weak and fragmented implementation. 1.7 Outline of the research report The following chapter of this report reviews policy implementation literature, and the evolution of the theories and analytical frameworks developed over the past 50 years to better understand policy design and implementation. An explanation of the contextual framework used to analyse the research findings concludes chapter 2. The third chapter provides a context for ICHIP implementation in describing the macro-level and micro-level governmental framework in which CoJ operates and the Johannesburg inner city housing landscape past and present. The fourth chapter examines ICHIP’s journey from inception through to its current status and assesses how much has been achieved against the original objectives, before considering if and how ICHIP is informing CoJ Housing Strategy. In the fifth chapter I analyse the factors that have shaped implementation and possible reasons why it has not progressed according to the original intentions and timelines of the strategy. The sixth and final chapter synthesises the discussion in chapter 5, relating it back to the policy implementation theories and models reviewed in chapter 2, and providing a few recommendations for consideration. 7 2 Policy implementation theory: a work in progress 2.1 Introduction The study of policy implementation is relatively young within the field of social sciences (Kahn, 2016). Harold Laswell first positioned implementation as an important part of policy process in 1956, but it was in 1973 that the subject assumed a higher profile when Pressman and Wildavsky published their seminal book, ’Implementation: How Great Expectations in Washington are Dashed in Oakland’ (Brynard, 2005; Madue 2008; Kahn, 2016). Theories and models of policy implementation have evolved from the rational, scientific approach of classical implementation theory to analytical frameworks that recognise the complex contexts in which policies are delivered and the need for more agile and integrated approaches to policy design and implementation. The proliferation of theories and models on policy implementation demonstrate the varied attempts to understand and predict implementation successes and failures, yet scholars of the discipline contend that it still lacks a unifying theory, and more work is needed in this regard (Mosse, 2004; Brynard, 2005; Madue, 2008; Kahn, 2016). Public policy can be summarised as a set of decisions and actions taken to address a certain issue or matter of concern; policy implementation is the execution of that policy by public and private sector stakeholders to achieve the policy goals (Kahn, 2016). The concept of the policy implementation gap has emerged from observations of implementation failures or weaknesses that result in ‘significant gaps between the planned outputs and outcomes of public policy and what actually occurs’ (Ansell et al., 2017:467). Understanding how and why these gaps emerge is the subject of policy implementation theory. This literature review covers the evolution of the key theoretical positions and frameworks associated with policy implementation and how this evolution has been organised into generations by policy implementation scholars. It then 8 considers policy implementation studies in the context of the Global South and South Africa, particularly in terms of the meaning of the state and its relationship with society and informality. The objective of policy implementation research is to understand implementation processes better, improve them and reduce policy implementation gaps. The analytical frameworks that various scholars have developed to enhance this understanding are reviewed in the third section, before I draw on one of the frameworks, the 7-C Protocol, to describe the conceptual framework applied to this research report. 2.2 The evolution of implementation theory, models and frameworks The originals: top-down and bottom-up theories ‘Classical political scientists assumed that public policies… would be smoothly implemented by efficient public bureaucracies and eventually solve the problems they were meant to solve’ (Ansell et al., 2017: 467). This linear, rational and Weberian approach forms the basis of classical implementation theory and the top-down model that underpins it. Top-down policy implementation is therefore a deductive process triggered by a centrally generated decision and implemented by an efficient administration (Matland, 1995). It was Pressman and Wildavsky who first drew attention to the reasons why implementation failed due to weaknesses that emerge throughout the implementation chain (Ansell et al., 2017). These weaknesses include vague objectives, weak political support, insufficient resources and misaligned stakeholders (ibid.). In this model, the solution to implementation failure was to improve the public administration function (Linder & Peters, 1987). Policy implementation scholars identified a number of weaknesses in the top- down model. Firstly, it fails to consider the social, political economic and administrative context in which policy is applied, nor does it acknowledge the impact of local-level administrators’ interpretations of and attitudes towards the policy (Makinde, 2005, Hudson et al., 2019). In focussing on only the administrative processes, classical implementation theory ignores the political aspects of policy implementation, regardless of whether the policy and its goals are 9 political or not (Matland, 1995). It also ignores the impact of the quality of the policy itself and whether it is appropriate to the problem it intends to address (Ansell et al., 2017). Bottom-up policy implementation on the other hand recognises the importance of local context and local actors for policy design and implementation. Promoted by scholars such as Lipsky (1978, 1980), Berman (1978 and 1980), Hjern and Porter (1981), Hjern (1982), Hull and Hjern (1987), a bottom-up approach demands that policy is determined at the lowest level of the organisation, by the people who will be responsible for implementing it or ‘street level bureaucrats’ (Lipsky,1980 and Yates, 1977 in Linder and Peters, 1987; Matland, 1995). These local-level actors are engaged during the policy formulation process to understand their issues and objectives as well as the networks and coalitions that will determine how the policy will be received when implemented (Matland, 1995). Critics of bottom-up theory say that it confers too much authority on local actors to make decisions about what is important and needed, removing this authority from the officials whom voters elect to make such decisions on their behalf (Linder and Peters, 1987; Matland, 1995). Matland (1995) points out that bottom-up approaches also overestimate the autonomy of the local implementers, because they must still operate within a legal and institutional framework that is determined by central authorities and institutional frames. Hybrids: best of both worlds or a compromise? The shortcomings of the top-down and bottom-up approaches led to the creation of hybrid models that attempted to address their respective weaknesses; Elmore’s forward and backward mapping approach was one such hybrid (Elmore, 1982 and 1985 in Matland 1995). In it, policy is designed in two directions: by mapping outcome criteria to policy objectives (that is, in a top-down manner) but also identifying the behaviour to be changed at the lowest level and working backwards to address it (ibid.). This approach thus deals with the policy objectives while also considering the realities and objectives of target groups and policy implementers 10 (ibid.). Linder and Peters (1987) note that Mazmanian and Sabatier (1983), advanced this approach to implementation studies by recognising that the complexity of the public policy implementation environment impacts how the policy would be realised at the lower levels. Their model was based on three pillars: the tractability of the problem, the implementation structure and how the statute or policy provided for this as well as other resources to implement, and non-statutory variables (Sabatier and Mazmanian, 1980). The New Public Management (NPM) discourse that emerged in the 1990s reinforced classical implementation theories by applying managerialist principles to policy implementation (Ansell et al., 2017). This approach used performance management techniques to address administrative weaknesses, and charismatic leadership to influence and motivate local implementors (ibid.). It also sought to enlist the targets of policy by offering ‘exit and voice’ mechanisms to give them a sense of empowerment and agency (Hirschman, 1970 and Sørensen, 1997 in Ansell et al., 2017). Matland’s (1995) ambiguity/conflict model, embraced all theoretical approaches, suggesting that the approach to implementation depends on the context and circumstances rather than one’s theoretical stance. He noticed that different approaches are appropriate in different situations: top-down in cases where the policy is clear and unambiguous, whereas bottom-up is useful where policy is more uncertain or vague. The model recognises the inherently political nature of policy implementation, particularly in cases where a policy is contested and successful implementation demands significant amounts of bargaining and negotiation between stakeholders. (Matland, 1995) More recently, scholars have focussed attention on policy design as an integral factor in policy execution. This approach was already evident in the evolutionary and backward mapping models proposed by Pressman and Wildavsky (1973) and Majone and Wildavsky (1978) (Linder and Peters, 1987). These models suggested an iterative, interactive form of policy formulation in which policy emerges from 11 interactions between policy makers, implementing agents and targets of the policy (ibid.). Ansell et al. (2017) and Hudson et al. (2019) propose a collaborative approach based on consultation and consensus seeking between the policy makers, implementers and targets. They also recognise that policy implementation is an adaptive, iterative process rather than a two-part process of policy design followed by implementation. Ansell et al. (2017) stress that this approach necessarily blurs the lines between politics and administration, and demands an attitude of mutual learning, flexibility and experimentation from all parties, noting that such an approach is likely to be resisted by politicians but can be overcome. Generations of policy implementation research In tracing the evolution of policy implementation research and thought, some authors make reference to three generations of scholars, however there does not appear to be a consistent interpretation of these generations. Brynard (2005) considers the first generation to be the classical political scientists who assumed that, on proclamation of a policy, it would be executed automatically in machine- like fashion by an efficient administration. The second generation, led by Pressman and Wildavsky, challenged the first-generation view as simplistic, recognising the complexity and political nature of policy and using case studies to demonstrate failed implementation (ibid.). The third generation took it a step further by trying to understand policy implementation better by creating analytical frameworks and models in the search for explanatory theories (ibid.). Kahn (2016) on the other hand considers Pressman and Wildavsky the first- generation of policy implementation scholars, as they challenged the ‘status quo’ thinking of classical political scientists who assumed policy implementation was automatic. The first generation studied policy implementation failures, identifying the complexities, variations and the political nature of implementation, but their work was criticised as atheoretical and too case-specific (ibid.). Kahn’s second generation moved from the descriptive approach of the first generation to a more analytical one, developing frameworks and models in an attempt to understand 12 and explain policy implementation successes and failures (ibid.). This generation includes scholars like Meter and Horn (1975), Sabatier and Mazmanian (1989), Elmore (1979) and Lipsky (1978, 1980). The third generation tried to bridge the gap between the top-down and bottom-up second-generation theories of the second generation by developing hybrid models (ibid.). For Kahn (2016), the onus is on the next generation to develop the grand theory of policy implementation. Brynard (2005) and Kahn (2016) identify a similar evolution of policy implementation thought: from the classical scientific approach, through the recognition of implementation complexity, to trying to understand that complexity using analytical frameworks and finally finding ways to integrate those frameworks into a unifying theory. I consider Kahn’s interpretation to be more consistent with policy implementation literature, but it is perhaps more important to focus on [note] the endpoint reached by both, that policy implementation theory is not yet fully mature and requires further research. 2.3 Policy implementation in the Global South and South Africa Policy implementation in the Global South must be considered through the lens of how the state behaves in contexts increasingly characterised by informality, poverty and limited state capacity. The interface between the state and civil society, the ability of the state to exercise its power and how it does so is of particular relevance for public policy implementation. Lund (2006) writes about the shadow state and twilight institutions, where ‘the state’ is interpreted as both a system and an idea, and as the lines between the state and civil society blur, power and authority become more fluid, enabling non-state actors to exercise a form of state authority, both delegated and assumed (ibid.). This blurring of the lines between state and non-state is echoed in the blurred lines between informality and formality, legality and illegality that abound in Global South urban contexts. Bénit-Gbaffou (2018a) reflects on the work of various scholars studying state intervention in city-making in post-colonial countries like India and South Africa. 13 She unpacks the perspectives of Roy (2009), Te Lintelo (2017), Bénit-Gbaffou (2018), Rubin (2018) and Charlton (2018) on how the state interacts with informality and illegality, often through informal practices of its own (ibid.). Their work moves away from the assumption that state actions as they relate to policy are always coordinated and purposeful, and while state interventions may be experienced as cohesive, they often mask a diversity of rationalities, interests and objectives across departments, agencies and levels of government (ibid.). The way the state deals with knowledge is both inconsistent and political, in some cases deliberately perpetuating a state of confusion or lack of knowledge to allow more flexibility in its interventions (Bénit-Gbaffou, 2018a). Te Lintelo (2017, in Bénit- Gbaffou, 2018a) identifies three types of informal state practices that are relevant in policy implementation: street-level bureaucracy, policy short circuits that contradict or by-pass legislation, and tolerance of informal or illegal practices that still deliver outcomes aligned to state objectives. In other cases, policy outcomes may diverge from the intention but are considered ‘good enough’ for the state, as the practicalities of correcting the outcomes lie beyond its capacity (ibid.). Charlton (2018, in Bénit-Gbaffou, 2018a), cites the example of RDP housing in South Africa, where the effects of the programme may not have been realised exactly as intended but the delivery of housing in itself has been quite remarkable. Turning the focus to South Africa, and in apparent contrast to the informal practices of the state discussed by Bénit-Gbaffou et al. (2018a), public policy implementation in democratic South Africa has been strongly influenced by the New Public Management (NPM) framework that promotes decentralisation, performance management and accountability in the public sector (Munzhedzi, 2020). Le Gales (2016) identifies this evolution not specifically in a global South context, but in relation to increasingly complex, globalised and differentiated societies where the traditional lines of authority between the state and society are blurred. In this context, performance measurement and indicators are not counter- intuitive in the face of informality, but rather provide a different way for the state to exercise control and steer the actions of society. Le Gales (2016:10) notes that ‘measurement and quantification are not neutral’ and the choice of measurement 14 approaches and criteria, generally neoliberal in orientation, can create new inequalities or exacerbate existing ones in societies, reinforcing informality and illegality. Madue (2008:200) suggests that the new South African government has really struggled to ‘[match] intention with outcome, and rhetoric with practice’ resulting in significant policy implementation gaps, particularly in the area of municipal service delivery. Brynard (2007) and Munzhedzi (2020) note that various reasons have been cited for these gaps, including a lack of political will and commitment to implementing policy; over-ambitious targets in some part due to a lack of reliable data to set realistic and clear policy goals; lack of capacity and skills; poor management of existing resources; ineffective government; corruption and maladministration; and a lack of coordination not only between the political and administrative spheres, but also between government departments (). Von Holdt (2010) contends that there is broad agreement that a functioning bureaucracy is essential for a developmental state to set development goals, develop the policies to achieve them and successfully implement those policies. But in South Africa, the post-Apartheid bureaucracy is debilitated by an internal conflict between a nationalistic desire to establish a modern state and the drive for African sovereignty. This diverts attention away from implementing policy and service delivery as focus is directed to issues of elite class formation, hierarchy based on deference rather than merit, an ambivalence towards authority and skills, and a breakdown in discipline (ibid.). The views above provide some insights into the complexities of policy implementation in the global South and the challenges faced in South Africa in particular. In the following section, I look at how different scholars have interpreted successful policy implementation and approaches to avoid policy implementation gaps. 15 2.4 Minimising the policy implementation gap Various scholars have identified factors that they believe are necessary for successful policy implementation, or conversely, minimising policy implementation gaps. If policy implementation gaps are defined as ‘significant gaps between the planned outputs and outcomes of public policy and what actually occurs’ (Ansell et al., 2017:467), then identifying policy implementation gaps requires that the difference between a policy’s goals and what is finally realised is measurable (Makinde, 2005). However, what is measured and how it is measured depends to an extent on the theoretical stance of the measurer. Matland (1995:154) suggests that ‘the pivotal question is whether attention should be focused on fidelity to the designer’s plan or on the general consequences of the implementation actions when determining success’. In this sense the distinction is drawn between implementation outputs and outcomes. Classical or top-down implementation theorists might use a relatively narrow assessment of how well a policy has been executed in terms of process and outputs (fidelity to the plan), whereas bottom-up and hybrid theorists might look at the broader impacts and whether the intention of the policy has been achieved in terms of its outcomes or broader consequences (ibid.). Determining the exact intention of a policy can be challenging however, as policy – and particularly statutory policy, is often vague by design (ibid.). Mosse (2004:639) proposes that good policy ‘legitimises and mobilises political support’, and to secure buy-in from a wide range of stakeholders it needs to be broad enough to address a similarly wide range of interests. Proponents of top-down models like Van Meter and Van Horn (1975), Mazmanian and Sabatier (1983), and Pressman and Wildavsky (1973) advise that policy goals must be clear and consistent, the number of stakeholders and actors involved must be minimised, as should the extent of change needed or implied by the policy, and the implementing agency should agree in principle with the policy’s goals (Matland, 1995). Bottom-up theorists on the other hand shy away from providing specific advice given the inductive nature of the approach, however they do note 16 the need for flexibility in the policy scope to allow for adaptation to the local context and the values of the local actors (ibid.). Makinde (2005) believes that policy implementation hinges on the interaction of many variables, but four are critical: accurate and consistent communication of policy goals and tasks; sufficient resources including human, material and intangible ones like the authority to implement the policy; a bureaucratic structure that facilitates efficiency, cohesion and coordination between departments; and a positive disposition or attitude towards the policy on the part of the implementers. Makinde (2005:65) also notes that the source of policy implementation gaps may be ‘the policy itself, the policy maker or the environment in which the policy has been made’. This highlights that policy implementation gaps are not only attributable to poor execution. Sabatier and Mazmanian (1980) criticised policy implementation literature at the time for being too focussed on implementation as the only reason for policy failures. They identified that policy failure could be caused by implementation failure, by poor policy design or by policy failure (ibid.). Implementation failure occurs when the intended effects are not achieved because of the way in which the policy was implemented, but weak policy design could mean that a policy will not succeed even if the implementation is flawless (ibid.). Policy failure is slightly different in that it occurs when the policy is implemented according to plan and achieves its objectives, but creates other negative and unintended consequences in the process (ibid.). Hudson et al. (2019) also attribute implementation failures to weaknesses in the policy design process, whether insufficient collaboration and consideration of context, or basing policy goals on inadequate information. They refer to the effects of ‘dispersed governance’, where policy formulated by national government and applied by local government encounters is inappropriate for the local context, or is resisted by local implementers who distort policy outputs through street level bureaucracy (ibid.). Ansell et al. (2017) note the importance of collaborative policy design that connects the policy makers to the target groups and implementing agents. This improves policy design by assimilating knowledge from all 17 stakeholders and enhancing innovation (ibid.). It also builds a sense of joint ownership in a policy which improves engagement in and commitment to its implementation (ibid.). Such an approach requires a shift from a government to a governance approach, in which the state recognises the importance of non-state actors. Bénit-Gbaffou (2018b:2152) notes that that the power of the governance model lies in the application of state and non-state resources towards achieving policy objectives, and depends on state’s ‘ability to build and maintain coalitions between state and non-state agents in order to mobilise the resources needed for urban change’. Policy instrumentation – the programmes, agreements and institutional structures that support policy implementation – is also an important determinant of success. Bénit-Gbaffou (2018b) contends that the choice of policy instrumentation is an inherently political act, and thatpolicy instruments deployed by the state represent their policy objectives more accurately than the intentions stated in policy documents. Lascoumes and Le Gales (2007) suggest that analysing choices of policy instrumentation over time can help to identify the evolution of policy intention, particularly where the ideology underlying policy is vague. On the other hand, in cases of ideological vagueness, the process of choosing policy instrumentation can help to identify and confirm the policy objectives (ibid.). The use of performance indicators and measurement by the state is also a form of policy instrument associated with a modernising state and neoliberal policies, and is, according to Le Gales (2016) a political as well as policy instrument that provides the state with a new means of governing an increasingly complex, globalised and differentiated society. Despite the lack of a unifying theory of policy implementation, scholars of the subject offer multiple insights into why policy implementation fails or succeeds. In the following section I describe an approach that assimilates these insights and that underpins the conceptual framework used in this research project. 18 2.5 Conceptual framework: the 7-C Protocol and ICHIP The objective of this research project is to establish the extent to which the City of Johannesburg has implemented ICHIP and possible reasons for implementation successes or failures. Despite the range of viewpoints on policy implementation, some authors believe that certain factors or variables are common to most, if not all, theoretical approaches. Brynard (2005) identified six variables to which Munzhedzi (2020) added a seventh. Together these variables constitute the ‘7-C Protocol’ (ibid.) and I have used this as a conceptual framework for analysing the ICHIP implementation. While the 7-C Protocol is not claimed as a definitive explanation of policy implementation, based on the literature I reviewed I believe it provides a comprehensive framework for evaluating policy implementation. Unpacking the 7-C Protocol The 7-C Protocol comprises seven variables that determine strong or weak policy implementation. They are context, content, commitment, capacity, clients and coalitions, communication and coordination. Both Brynard (2005) and Munzhedzi (2020) emphasise that these factors are interrelated and do not function in isolation. Context refers to the social, economic, political, legal and institutional environments in which a policy is both developed and implemented (Brynard, 2005; Munzhedzi, 2020). According to Brynard (2005), content is relevant to policy implementation in two ways. Firstly, the policy typology, classified as distributive, regulatory or redistributive, dictates whether the policy aims to generate public goods, set rules of behaviour or redistribute a finite quantity of public goods between different constituencies (ibid.). According to Munzhedzi (2020:95), ‘governments function by coercion’ so the policy typology determines the amount of coercion that will be needed as well as appropriate methods of applying it. A second consideration 19 about content is the appropriateness of the policy goals themselves as well as the methods chosen to achieve them (Brynard, 2005). Commitment indicates the willingness of the implementer to carry out the policy tasks (Brynard, 2005). Brynard (2005) notes that while mostly associated with bottom-up theories and the attitudes of the administrative staff to the policy, commitment is also applicable in top-down approaches where it is more commonly referred to as ‘disposition’. This translates into political will at all levels of the implementation hierarchy. Capacity is the ability to deliver policy based on the resources that are available and can be mobilised to support the required outputs (Brynard, 2005.). In the context of public policy implementation, capacity is determined by the structural, functional and cultural ability of the organisation, and it includes tangible resources like budget, materials and staff, and intangible resources like leadership, motivation, commitment and resilience (ibid.). Clients and coalitions are the ‘interest groups, opinion leaders and other outside actors who actively support a particular implementation process’ (Brynard, 2005:661). The state must be strategic when choosing coalition partners, and invest sufficient time and energy to identify the appropriate partners and build strong relationships with them (ibid.). Effective communication of policy goals and tasks is central to policy implementation, as the implementers need to understand what they have to deliver and the consequences of non-delivery (Brynard, 2005; Munzhedzi, 2020). An evaluation of communication must also include the organisational structures and processes that are put in place to support it (Brynard, 2005). Finally, coordination refers the degree to which municipal officials are aligned and support one another in executing policy (Munzhedzi, 2020). It is determined by the strength and nature of the relationships between the policy implementation 20 stakeholders, and influences how resources as well as information flows between them (ibid.). The 7-C Protocol and ICHIP In applying the 7-C Protocol to ICHIP’s implementation, I have analysed how each of the seven factors described above may have affected implementation based on a review of various documents and the perspectives shared in the interviews conducted. Where possible I have applied the analysis at the level of the individual delivery and facilitation programmes, however the granularity of the information provided did not always allow for this. Figure 2.1 below illustrates how the 7-Cs protocol has been applied as a framework to assess ICHIP implementation at a policy and programme level. Figure 2.1: 7-Cs Protocol as a framework for assessing ICHIP implementation 2.6 Conclusion In this chapter I reviewed the theories and analytical frameworks developed over the past fifty years to better understand public policy design and implementation and the reasons that policy implementation gaps develop. Despite the numerous 21 models put forward, scholars of the discipline maintain that it is still lacking a common or unifying theory and that further research is needed. However, it is possible to identify seven factors that contribute to or detract from successful policy design and implementation that are common to most policy implementation models and theories. These have been synthesized into a 7-Cs protocol which underpins the conceptual framework of this research study. But before assessing the research findings, it is useful to understand the Johannesburg inner city housing context that gave rise to ICHIP. This is the subject of the next chapter. 22 3 Johannesburg Inner City Housing: it’s complicated 3.1 Introduction As the entry point to Johannesburg and the trade and transport hub of the metropolitan region, the inner city is a cosmopolitan and dynamic urban centre. Despite an active private sector housing market, there is still a serious lack of decent, affordable housing, particularly for the very poor. The City of Johannesburg (CoJ) has over the years implemented a range of programmes to try to address this issue, with the Inner City Housing Implementation Plan (ICHIP) the most recent, having been finalised in 2017. The context within which a public policy is to be implemented is highly relevant for its design and the way in which it is implemented, and ICHIP responded to the Johannesburg inner city housing context at that time. In this chapter I explore this context by firstly looking at the macro-level housing landscape within which CoJ operates. I then review Johannesburg’s inner city housing history with a focus on the various programmes and initiatives that have been implemented since 1994, before assessing the current situation and its key actors. Finally, I describe the high-level governance structure of the municipality and the strategic imperatives that direct its work. 3.2 Framing local government housing policy The 1996 Constitution places an obligation on the state to ensure access to adequate housing for all its citizens. Adequate housing includes ‘adequate privacy, adequate space, adequate security, adequate lighting and ventilation, adequate basic infrastructure and adequate location with regard to work and basic facilities - all at reasonable cost’ (Yitay, 2011 in Mashiane and Oduko, 2020:99) and suggests a broader interpretation of housing as ‘human settlements’ rather than simply physical housing structures. The Constitution also requires that the three spheres of government – national, provincial and local – work together in a system of 23 cooperative governance in which each sphere enjoys a level of autonomy in the areas designated to them. This means that while CoJ is delegated authority to facilitate housing delivery according to its own policies within its area of jurisdiction, it must do so within the legal and regulatory framework established by national and provincial government. CoJ is also required to pursue the housing policy objectives and principles as set out in the National Housing Act 107 of 1997 and in policies such as the National Housing Code (2009), the Comprehensive Plan for the Development of Sustainable Human Settlements or “Breaking New Ground” (BNG) (2004), the National Development Plan 2030 (2012) as well as provincial housing policy (Tissington, 2013). National housing policy identifies inner city regeneration, the needs of the poor, equitable access to housing, densification, and ensuring a range typologies and tenure options as priorities (RSA, 1997; Department of Human Settlements, 2009). It also recognises the need for private sector participation in housing delivery, and to empower individuals and communities to fulfil their own housing needs (ibid.). This positions the role of the state as both deliverer and facilitator of housing development. BNG (2004) signalled an important shift in national housing policy away from increasing housing ownership towards rental housing as a way to meet the demand for housing in the low-income segment (Tissington, 2013). A number of housing policy instruments were established via the National Housing Code to support the delivery of affordable rental housing and social housing by local government. These programmes include the Institutional Subsidy Programme (ISP), the Social Housing Programme and the Community Residential Units (CRU) Programme (ibid.). 3.3 Housing in the Johannesburg Inner City Over the past 50 years the Johannesburg inner city has witnessed massive transformation with periods of dramatic decline and regeneration (Tissington, 24 2013; Robb, 2018). As Murray (2011, in Robb, 2018:33) states, it went from being the ‘finance capital and large-scale corporate enterprise of South Africa, and sub- Saharan Africa at large, to becoming the residential home for the urban poor, both formally and informally’. After two decades of so-called ‘white flight’ in the 1960s and 1970s to the northern suburbs, the inner city became a magnet for poorer South Africans seeking employment opportunities in the economic centre of the country, but overcrowding, poor building management and a deterioration of public services led to an inner city characterised by ‘crime and grime, urban decay and anarchy’ (Tissington, 2013:31). The late 1990s saw the start of a regeneration narrative that sought to attract investment back to the inner city as a driver of economic development, and the development of housing stock – albeit not necessarily affordable housing – was a consistent element in that narrative (ibid.). Since the late 1990s, the inner city has seen a number of housing strategies and programmes linked to this regeneration narrative that are set out in Figure 3.1 below. These programmes have been driven by CoJ and partnerships between the City, the private sector and community organisations. They have made use of housing and economic development policy instruments like the ISP and Urban Development Zone (UDZ), and to a large extent reflected the strategic goals and development paradigm of the City and its mayor at the time. In the early 2000s for example, the long-term vision of the Joburg 2030 Master Plan of a “World Class African City” focused on investment and economic growth. The Inner City Regeneration Strategy adopted in 2003 mirrored this strongly neoliberal stance with a focus on attracting middle and upper-class residents, gentrification and the eradication of slums (Tissington, 2013). The shift in strategic focus in the Joburg 2040 Growth and Development Strategy (GDS) towards inclusion, sustainability and eradicating poverty was echoed in the inner city regeneration strategy of the 2013 Inner City Transformation Roadmap (ICTR). The ICTR recognised the need for developing new housing stock in a non-exclusionary way, with an emphasis on providing housing for the poor and very poor households resident in the inner city (CoJ, 2013.). However, despite the plethora of programmes, policies and plans, various authors suggest that the ‘world class African city’ paradigm continues to 25 dominate city thinking and that little progress has been made towards a more inclusive inner city (ibid.; Robb, 2018). Figure 3.1: Organisations, Programmes and Strategies impacting housing in the Johannesburg inner city, 1992 to 2016 (Source data: Tissington, 2013; Robb, 2018) Legend:Johannesburg Transitional Metropolitan Council / City of Johannesburg Programme Private sector /PPP Programme Organisation / agency Policy or strategy document 1992 2001 1993 1994 1995 1996 1997 1998 1999 2000 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 Transitional Housing Programme Seven Buildings Programme Inner City Shelter Forum (ICSF) Central Johannesburg Partnership (JCP) Inner City Office (ICO) JDA Bad Buildings Programme (BBP) Better Buildings Programme Inner City Property Scheme (ICPS) Urban Development Zone (UDZ) Inner City Position Paper Joburg 2030 Master Plan Inner City Regeneration Strategy Inner City Housing Action Plan Bad Buildings Strategy Joburg 2040 Growth & Development Strategy (GDS) Inner City Transformation Roadmap (ICTR) Inner City Housing Implementation Plan (ICHIP) 26 ICHIP was created as a response to the ICTR demand for an update of the Inner City Housing Action Plan of 2007 that would provide for a range of housing options and typologies to suite the range of needs and means of inner city households (CoJ, 2013). According to the final ICHIP report, it ‘takes into account all sectors of the housing market and the needs of all income groups with an emphasis on the very poor… [and builds] on the extensive work that has already been undertaken and is underway’ in the inner city (RebelGroup, 2016:7). 3.4 Inner city housing in Johannesburg today Despite the rise of the Sandton economic node, the inner city remains the core of Johannesburg and an active commercial, retail and residential centre that ‘still functions as a significant economic focus of the city’ (CoJ, 2021a:48). The role of the inner city as a trade hub and gateway to Africa contributes to its cosmopolitan nature. Its economic and residential opportunities and its transport infrastructure make it the entry point to Johannesburg for urban migrants from elsewhere in South Africa as well as other African countries (RebelGroup, 2016). A 2017/2018 Quality of Life survey found that over 15 000 migrants enter the city of Johannesburg every month, roughly one third of them from other countries (CoJ, 2021b). The inner city is home to large numbers of non-governmental, faith-based and community-based organisations, significant cultural and heritage assets, educational, artistic and sporting institutions as well as SMME incubators and creative industries (ibid.). From a spatial planning perspective, this physical, social and economic infrastructure and the potential to optimise these assets position it as the ‘heart of the new compact polycentric city [that] acts as the connector, bridging the historical north-south divide’ (CoJ, 2021a:81) and as one of five transformation zones in CoJ’s 2040 Spatial Development Framework (SDF). While the inner city’s economic and social importance as well as its potential for development is recognised, rapid growth and urbanisation are straining its urban management systems, placing pressure on infrastructure and creating more demand for service delivery (RebelGroup, 2016). Weak urban management, ageing 27 and under-maintained infrastructure, and overcrowding have created ‘an inner city under stress’ (ibid.:8). The lack of affordable housing is recognised as one of the main challenges of the inner city in multiple publications and strategy documents, including the Johannesburg 2040 GDS, the 2040 SDF and the City’s Integrated Development Plan (IDP). High poverty levels, a stressed built form, and weak property ownership and management systems all contribute to a vicious property cycle (RebelGroup, 2016). The range of housing supplied does not adequately respond to the demands for welfare and emergency accommodation of economically vulnerable households needing a more flexible rental model. These households often have no option but to rent from unscrupulous landlords and live in very poor conditions that exacerbate their vulnerability (ibid.). Mayson and Charlton (2015) note a complex range of management systems that have emerged across the formal and informal inner city housing markets to respond to the changeability and precariousness of inner city households’ circumstances. They suggest that, while some of the management practices and housing conditions expose residents to unacceptable levels of risk, others provide decent accommodation and fill a gap that public housing strategies are failing to fill (ibid.). The challenge for municipal housing strategists is therefore to find a balance between intervening in cases of unacceptable risk and assimilating other strategies that – while illegal under the current legal and regulatory framework – are responding adequately to the realities of many city dwellers (ibid.). While the challenges of inner city housing are many, RebelGroup (2016) also identified a number of opportunities. There is large scope for repurposing properties in the inner city and an apparent willingness in the private sector to do so. The private sector is responsive to housing demand in both the formal and informal markets. New funding mechanisms and institutions are opening up the market to investors, particularly to small scale landlords and black entrepreneurs. 28 And the transport and infrastructure base is significant, even if ageing and undermaintained in certain respects. (ibid.) 3.5 Inner city housing demand and supply Establishing current demand and supply of inner city housing, particularly on the lower end of the rental scale, is frustrated by a lack of recent, reliable and consistent data. The information in this section is derived largely from the ICHIP report prepared by RebelGroup in 2015. The data in that report is based on the 2011 Census findings together with the project team’s own fieldwork in a number of inner city areas, and is presented as indicative. Housing demand The nature and scale of demand for housing can be inferred from population and household statistics. Between 2001 and 2011, the Johannesburg inner city population grew 23% and household size grew by 6%, both above the national averages (RebelGroup, 2016). This reflects a high rate of urbanisation, but the increasing household size also suggests a housing supply shortage that forces sharing arrangements between households. Indeed, RebelGroup found that in five precincts, flats were oversubscribed by 100% and residential houses by 300%. Table 3.1 below shows an analysis of overcrowding in five inner city precincts based on the number of households versus number of properties registered in the City’s rates database. (RebelGroup, 2016) Table 3.1: Overcrowding in five inner city precincts, 2011 (RebelGroup, 2016:40) 29 Available data indicates that the majority of inner city households (89%) rent, and that most of these households are low- to middle income, earning less than R7000 per month (RebelGroup, 2016). This translates to a housing budget of under R1600 per month, if a guideline of 25% is applied (SERI, 2016). The household income graph in Figure 3.2 below shows that more than 22’000 households earned less than R9600 per annum in 2011, and of those households, 75% reported earning no income at all (RebelGroup, 2016). Figure 3.2: Inner city households by annual income band, 2011 (RebelGroup, 2016:32) In 2016, the estimated shortfall of housing just for those households in the inner city considered to be at high or very high risk, was 30,084 units. These households were earning less than R3200 per month and were thus in need of a public sector housing intervention, as the private formal sector was (and is) not able to deliver housing in this rental price bracket. (RebelGroup, 2016) Housing supply The range of housing typologies on offer in the inner city is wide, from shared rooms and spaces through to 3-bedroom flats. Subletting and shared living arrangements are common and indicative of both low incomes and the lack of 20 000 18 000 16 000 14 000 12 000 10 000 8 000 6 000 4 000 2 000 0 No income R1 - R4 800 R4 801 - R9 600 R9 600 - R19 600 R19 601 - R38 200 R38 201 - R76 400 R76 401 - R153 800 R153 801 - R307 600 R307 601 - R614 400 R614 401 - R1 228 800 R1 228 801 - R2 457 600 R2 457 601 or more 30 affordable options. Building management systems are well-established in both formal and informal rental markets and access to services is relatively good but costly. Figure 3.3 below provides an overview of the various sources of housing supply in the inner city as well as their pricing ranges and estimated units provided (where known). While there is no reliable, consolidated information on housing delivery, RebelGroup (2016) estimated based on interviews and various reports that the private sector was delivering between 2000 and 4000 units in the R1500 to R3000 per month rental range, and less than 1000 units in the above- R3000 per month category. Social housing institutions were delivering only 200 communal /shared facilities units for less than R1000 per month, and 500 units in the R1050 to R2100 rental range. Their study found no units being delivered by the public sector for less than R600 per month. This analysis demonstrates the clear need for a strategy and concerted effort to increase the delivery of affordable housing if the supply shortfall is to be addressed. (RebelGroup, 2016) Figure 3.3: Sources of inner city housing supply (RebelGroup, 2016:42) • Large formal private rental providers (e.g. JPOMA members) • Estimated 55’000 (??) units under management • Rentals R750 – R5’000+ / month (ex. services charges) • Social Housing Institutions (JHC, JOSHCO and Madulammoho) • Estimated 4'000 units under management • Rentals R750 - R2'500 / month (generally incl. of service charges) • Generally semi-formal sub-letting arrangements (mainly sectional title units) - sub-letting of space • Number not clear • Rentals R500 - R2'000 /mth (bed, doorway, shared rooms, etc.) • Formally owned but poorly managed and over-crowded rental stock • Number not known • Rentals not known, but likely in line with formal private (noting significant sub- letting prevalent) • Occupied non-residential spaces - may have some community structure • Number not known • Rentals not known but case studies suggest very low i.e. < R500/ mth Formal Private Rental Social Housing Small Scale Landlords Slumlords Informal 31 The actors Figure 3.4 below provides an overview of the organisations and institutions in public and private sector that touch or impact inner city housing. While the view is not exhaustive, it gives a sense of the number of actors that have an interest in housing and whose involvement should be taken into consideration in any housing strategy implementation. Figure 3.4: Inner city housing actors (created by author; source data CoJ Housing Department, 2021a) 3.6 City of Johannesburg strategic and governance framework The long-term strategic priorities of the City of Johannesburg are articulated in the 2040 GDS (published in 2011) and are aligned to the 2012 NDP as well as provincial strategies like the Gauteng Transformation, Modernisation and Reindustrialisation strategy of 2015 and the Growing Gauteng Together 2030 strategy published in February 2020. The City’s IDP provides a 5-year strategic plan and the current IDP for 2021 to 2026 identifies 12 strategic priorities, one of NATIONAL & PROVINCIAL GOVERNMENT CoJ MUNICIPAL OWNED ENTITIES NON-PROFIT COMMERCIAL GENERAL / CROSS FUNCTIONAL HOUSING AUXILLIARY SERVICES / URBAN MANAGEMENT * Citizen Relationship & Urban Management HOUSING FINANCE PUBLIC SECTOR PRIVATE SECTOR INNER CITY OFFICE CRUM* (Region F) EMERGENCY MANAGEMENT SERVICES (EMS) HOUSING Human Settlement Policy, Planning & Research Public Housing Programme Public Housing Support Programme Public Housing Stock Management Finance Business Management Support Small landlords Slumlords PLANNING 32 which is Integrated Human Settlements. These strategic priorities are supported by strategies like the the Joburg City Safety Strategy (approved in 2003 and revised in 2016), the SDF 2040, the Inner City Transformation Roadmap of 2013, and the Joburg Human Settlement Strategy articulated within the 2021 to 2026 IDP. The Human Settlement Strategy recognises the inner city’s housing challenges as well as the City’s constitutional mandate to ensure access to adequate housing for its residents. It acknowledges ICHIP as part of the Inner City Revitalisation Programme (CoJ, 2021b) The governance structure of the City provides for the separation of the legislative and executive functions. The City Council is made up of elected councillors who are responsible for policy formulation and overseeing policy implementation. The Council elects an Executive Mayor who provides strategic direction for the City, supported by a Mayoral Committee comprising 10 councillors or Members, each responsible for a portfolio. Section 79 Portfolio Committees are chaired by councillors and have an oversight function that includes monitoring the delivery and outputs of departmental programmes, reviewing CoJ plans and budgets, and holding the Executive accountable for performance against City policies and priorities. Finally, the CoJ administration, led by the City Manager, comprises departments and is staffed by employees responsible for delivering the vision and strategies of the Council. (CoJ, 2021b) Municipal Entities (MEs) are the City’s service delivery agents. They are wholly- owned by the City and governed by service level agreements. There are 13 MEs, however the key ones for this report are the Johannesburg Development Agency (JDA), responsible for supporting area-based economic development initiatives; the Johannesburg Property Company (JPC), which develops and manages all CoJ- owned properties to optimise their social and commercial value for the City; and the Johannesburg Social Housing Company (JOSHCO), which is responsible for providing and managing affordable rental housing for the City. (CoJ, 2021b) 33 3.7 Conclusion Under the 1996 Constitution, the City of Johannesburg as the local arm of the State is obliged – alongside its provincial and national counterparts - to ensure its residents have access to adequate housing. The Constitutional interpretation of adequate housing implies more than physical structures, but points to adequacy in terms of location, pricing, infrastructure and services (Yitay, 2011 in Mashiane and Oduko, 2020). In formulating its approach to housing policy and strategy, the City enjoys a level of autonomy but must still work within the national and provincial legislative frameworks and support their policy objectives, which, as they relate to housing, include prioritising the needs of the poor, inner city regeneration and densification (RSA, 1997; Department of Human Settlements, 2009; Tissington, 2013). They also recognise the role of the private sector in housing provision and the agency of individuals to provide for themselves; in this sense the state is a facilitator of housing provision rather than purely a provider. A number of housing strategies and programmes have been implemented in the Johannesburg inner city since the 1990s as part of various inner city regeneration drives. Despite these programmes and a dynamic private sector housing market, there is still a serious lack of decent, affordable housing in the inner city, particularly for the very poor. This has created a situation in which the most vulnerable households are forced into precarious informal housing situations that further increase their vulnerability. Housing developers and landlords in the formal private sector are unable to provide housing at rental prices accessible to the very poor, so this responsibility falls to the City. An inner city housing situation faced with multiple challenges and involving a large number of actors requires a strategy that embraces complexity and facilitates cooperation between the public and private sectors. The following chapters analyse the extent to which ICHIP has been able to achieve this. 34 4 Assessing ICHIP’s implementation 4.1 Introduction In February 2017, the City of Johannesburg’s (CoJ) Housing Department presented the draft Inner City Housing Implementation Plan (ICHIP) to the Inner City Partnership Forum and in May of the same year it secured approval for the plan from the CoJ Mayoral Committee. Eight months later on 1 October 2017, the CoJ published a media statement in which Mayor Herman Mashaba announced that ‘the City’s Council has approved our plan for tackling the housing challenge within the inner city and creating safe, clean and connected communities with access to economic communities’ (eNCA, 2017; JICP, 2017). The announcement referred to the need for 30’000 accommodation units to address the housing shortfall in the inner city, the important role of public-private partnerships in the strategic approach, and said that ‘[t]hrough the Inner City Housing Implementation Plan, we are set to make the inner city housing market work better for the poor’ (ibid.). ICHIP was premised on an approach that was contextually appropriate, prioritised the poor, and promoted public-private partnerships. It addresses the key strategic areas of providing temporary emergency accommodation to meet the City’s Constitutional Court obligations, delivering a pipeline of subsidised rental housing, enhancing the delivery of affordable rental stock by the private sector, providing emergency services to buildings in critical condition, dealing with ‘bad buildings’, and reversing the decline in owner-occupied housing through a sectional title rehabilitation programme. ICHIP proposed an institutional arrangement to enable implementation as well as a budget breakdown for the five-year period from 2017/18 to 2021/22. The total projected programme cost was R2,069 billion, comprising approximately R1,9 billion in capital costs, R102 million in operating costs and R58 million in programme management costs. It was foreseen that the CoJ would carry R1,32 billion of the total budget, with the balance coming from 35 state grants and subsidies as well as the private sector. (RebelGroup, 2016; CoJ Housing Department, 2017b) Five years after the mayor’s media statement in October 2017 the question is asked how much has been achieved. The answer from a diversity of stakeholders has ranged from ‘not much’ to more qualified responses identifying pockets of activity related to a few of the programmes. The interviews also revealed that ICHIP was in fact never approved by the City Council. Despite this, there is evidence that ICHIP as a strategy is still being used by the CoJ Housing department to inform its strategy. In the following sections I discuss ICHIP’s journey from its inception through to its current status. I assess how much has been achieved against the original objectives and possible reasons why. Finally, I consider how ICHIP continues to inform CoJ Housing strategy, even if not in the manner initially intended by its authors. 4.2 The status of ICHIP, and why it matters On 5 May 2017, the CoJ Housing department submitted a report to the CoJ Mayoral Committee requesting approval for an Inclusive Mixed Use (IMU) Inner City Housing Implementation Plan (ICHIP) and its associated R2 billion budget (CoJ Housing Department, 2017a). The project was initiated by the Housing department who collaborated with the Development Planning department, the Johannesburg Social Housing Company (JOSHCO) and the Johannesburg Development Agency (JDA) (ibid.). The JDA led the project from an operational perspective, drawing up the terms of reference and hiring consultants RebelGroup to develop the strategy (JDA Official, interview 10 December 2021; Sotomi, interview 22 December 2021). Emanuel Sotomi, Director of the Housing department’s Public Housing Programme Support (PHPS) directorate, explained that his directorate had approached the JDA for their assistance because the JDA had a panel of consultants that the Housing department did not, and is also subject to less challenging procurement mechanisms for project budgets of less than R1 million (interview, 22 December 2021; Housing Department Official 1, interview 8 December 2021). 36 RebelGroup prepared a draft ICHIP report that included an overarching strategy as well as more detailed action and precinct plans. Using RebelGroup’s draft report, the Housing department consulted various internal and external stakeholders to check ICHIP’s viability before taking it to the Mayoral Committee (Housing Department Official 1, interview 8 December 2021). The minutes from the Mayoral Committee meeting were not available at the time of writing, but Sotomi confirmed that the plan was approved (interview, 22 December 2021). The budget was not approved in full however; instead ‘nominal amounts were approved year on year’ (Sotomi, personal correspondence 10 March 2022). In saying that ‘the City’s Council has approved our plan for tackling the housing challenge within the inner city’ (eNCA, 2017; JICP, 2017), the media statement on 1 October 2017 suggested that ICHIP had been approved by the Council as well as the Mayoral Committee. However, every CoJ official that I interviewed including Sotomi stated that ICHIP was not approved by the Council. It had apparently been tabled at a Council meeting on 27 and 28 September 2017, but as the extract of the Council Meeting Minutes in Figure 4.1 below shows, the item was withdrawn by the MMC for Housing, Councillor Ntuli (CoJ, 2017). This begs the question why Mayor Mashaba announced in a media statement three days after the Council meeting that ICHIP had been approved. The media statement also referred to the Council approving the release of 12 CoJ-owned properties for Figure 4.1: Extract of CoJ Council Minutes, 27/28 September 2017 (CoJ, 2017) 37 low-cost housing, and this leads me to believe that the media statement conflated the approval on the 12 buildings by Council with Council approval of ICHIP; whether this was deliberate is not clear. According to Sotomi, Council approval was neither needed nor requested by the Housing department because ICHIP is a strategy not a policy (interview, 22 December 2021). He explained the difference between the two as: ‘Policy speaks to how you do things, it’s about rules. Strategy is saying we have these instruments, this is how we’d like them to work together’ (ibid.). This was relevant because ICHIP was not proposing an entirely new set of initiatives, as it included existing Housing programmes in an overall, integrated strategy. Sotomi confirmed that ‘it was approved by the Mayoral Committee and that was all that was required of it, however, Mayor Mashaba insisted that it go to the Council’ (interview, 22 December 2021). Any strategy, programme or policy that is put to the Council is subject to a rigorous process that includes approval by the relevant Section 79 Committee, an oversight body responsible for vetting and monitoring programme implementation (Housing Department Official 2, interview 30 November 2021; Sotomi, interview 22 December 2021). Figure 4.2 reflects this process; it shows the header of the Final ICHIP Report that was submitted to the Council on 30 August 2018. The header lists the entities to whom the report has been presented and when, and it is evident that by the end of August 2018, the report had been put before the Mayoral Committee a second time, the Section 79 Housing Committee twice, and the Council three times. Sotomi clarified however that the header in the report in Figure 4.2 indicates that there were three attempts to put ICHIP before Council prior to the one that was withdrawn in September 2018, but they were not successful. To this day, ICHIP has not been approved by the Council. 38 Figure 4.2: Header section of Final ICHIP Report to Council, 30 August 2018 (CoJ Housing Department, 2018) No clear reason other than politics was forthcoming from any of the interviewees as to why ICHIP was never approved by the Council. Director of the Inner City Office, Thabo Maisela, recalled that the report had been sent from one committee to the other in an administrative process that was influenced by the political parties involved (interview 10 December 2021), and this process is evident in the report header in Figure 4.2. I consider the nature of the politics further in the following chapter, but it appears that it was not for lack of trying that ICHIP did not get Council approval. The JDA Official confirmed that the JDA was actively involved in helping the Housing department to complete ICHIP and get it approved by Council, but they did not succeed (interview, 10 December 2021). It does seem that Council approval would have been useful however, particularly in terms of securing the full budget needed to implement the strategy. Sotomi noted that Housing had not wished to put ICHIP to the Council because the only ‘new’ budget needed for ICHIP related to the TEA delivery programme and this budget could be approved at the level of the Mayoral Committee (interview, 22 December 2021). According to Sotomi, the budgets for the other programmes were already accounted for since they were existing programmes of the respective implementing agents (ibid.). This doesn’t explain why the full budget was submitted to the Mayoral Committee for approval in May 2017 however, and I have been unable to ascertain the reasons why it was not approved. But as the following section will show, the lack of budget appears to be a key cause of weak 39 implementation, and a number of the CoJ officials that I interviewed felt that Council approval would have addressed this. Housing official Msizi Khuhlane commented that there was never any money set aside to cover ICHIP’s R2 billion projected cost and it had to be funded from the existing Housing budget (interview, 2 December 2021). He believes that Council approval would have given the Housing department the footing to ask for the additional budget (ibid.), a view corroborated by Housing Department Official 2, who said that once any strategy has been approved by Council, budget must be allocated to implement it (interview, 30 November 2021). Khuhlane also felt that Council approval would have ensured a higher prioritisation of the strategy within the Mayoral Priorities related to the inner city, which should have unlocked additional funding (interview, 2 December 2021). Council approval was also important in relation to securing the operational support for ICHIP across the CoJ. The Planning Official interviewed remarked that it is ‘difficult to get everyone pulling in the same direction’ (interview, 29 November 2021), a view echoed by Housing Department Official 1 who reflected on the challenge of getting departments to work together or getting City officials to deliver anything that falls outside of their job description or Key Performance Indicators (KPIs) (interview, 8 December 2021). The reality is that within the City, different departments can have KPIs that are diametrically opposed (Klug, interview 15 November 2021), and in this respect, having a policy or programme approved by the Council ensures that the programme’s outputs are translated into KPIs for all departments implicated in its execution (Housing Department Official 1, interview 8 December 2021). But even though ICHIP was not approved by the Council, it is evident that certain elements of it have been implemented to some extent. I examine what has been achieved in each programme in the next section. 40 4.3 Implementation status of the ICHIP programmes ICHIP is based on two pillars of implementation: housing delivery and housing facilitation (CoJ Housing Department, 2017a). Each pillar comprises a number of programmes, where the housing delivery programmes focus on increasing the supply of accommodation and the housing facilitation programmes address issues with existing housing or attempt to provide a better platform on which housing delivery can occur (ibid.). There are 11 programmes in all: six delivery programmes and five facilitation programmes that are shown in Figure 4.3. Each programme has a responsible department or municipal entity (ME) and an implementing agent or agents. Figure 4.3: ICHIP Delivery and Facilitation Programmes (Adapted from CoJ Housing Department, 2017a) Before discussing the implementation status of the programmes, it is relevant to consider the institutional arrangements proposed for ICHIP and whether these were realised. The 2016 ICHIP Strategy and Programmes Report noted the need to allocate sufficient human resources to the programme (RebelGroup, 2016). It also recommended that an Inner City Programme Management office be set up, identifying the JDA as suitable for the task given its role as the City’s development Delivery Programmes Facilitation Programmes Programme Responsibility Implementing agent Programme Responsibility Implementing agent JOSHCO JOSHCO, external service providers e.g. MES CRUM ICHIP Programme Office Social Dev; EMS; JOSHCO etc. JOSHCO JOSHCO JDA ICHIP Programme Office JDA Private sector e.g. Jozi Housing, AFHCO DED (ICPS) Support: ICHIP Programme Office; Joburg Water JOSHCO; JDA JOSHCO (muni. SHI); Independent SHIs e.g. JHC, MHA; HDA JDA; Housing; Planning JDA; Ext. service provider; Planning; Joburg Water JDA TUHF; HDA CRUM Legal & Compliance Region F inspectors; External service providers CoJ Housing CoJ Housing; GPF Facilitation Programme 1: Relocation of Evictees Facilitation Programme 2: Sectional Title Rehabilitation Facilitation Programme 3: Bad Buildings Facilitation Programme 4: Municipal Support Facilitation Programme 5: Information Systems 41 facilitation unit (RebelGroup, 2016). The CoJ Inner City Office, located in the Office of the City Manager, would ensure strategic alignment with other Inner City programmes, in particular the Inner City Transformation Roadmap (ibid.). Various interviews confirmed however that the Programme Management Office was not set up, and although the JDA was certainly involved in driving some ICHIP programmes, it did not take on the coordinating role as intended (JDA Official, interview 10 December 2021; Sotomi, interview 22 December 2021). Assessing progress on ICHIP’s Delivery Programmes As the name suggests, the delivery programmes aim to increase the supply of housing in the inner city. The objective of the first delivery programme, Temporary Emergency Accommodation (TEA), is to provide appropriate accommodation for inner city evictees. This is not only an obligation placed on the CoJ by the Constitutional Court, it is also an important piece of the housing supply chain in that it enables the release of buildings for redevelopment as affordable housing (RebelGroup, 2016). There was broad consensus among the interviewees that the TEA programme is being delivered albeit at a slow pace (Maisela, interview 10 December 2021). Other opinions are that it is poorly managed (Khuhlane, interview 2 December 2021) and not being realised in the manner intended by ICHIP (Housing Department Official 2, interview 30 November 2021). A 2021 TEA Readiness Report prepared by the Housing department confirms that four new TEA buildings yielding 412 rooms were completed over the past five years (CoJ Housing Department, 2021a). It also notes 12 projects that were presented to the Mayoral Committee for the five-year period are stalled or have been cancelled for various reasons, and a further 11 new projects are earmarked for a combination of TEA, Affordable Rental Accommodation and old age rental units (ibid.). The report emphasises that the TEA programme has been and continues to be significantly under-funded: against a budget requirement of R561 720 000 projected by ICHIP for the five-year period of 2017 to 2022, only R173 812 000 was received, a shortfall of R387 908 000 or 69% (ibid.). 42 At the time of the 2021 TEA Readiness Report, more than 10 000 TEA beds were required for evictees based on court eviction notices (CoJ Housing Department, 2021a), implying an urgent need for TEA facilities that is not matched by the pace of delivery or the budget allocation. According to Sotomi, securing buildings for TEA facilities from the JPC is critical, but while Housing works closely with the JPC, it is not an easy relationship and they struggle to access the JPC’s asset registers (interview, 22 December 2021). Sotomi suggested that the JPC’s orientation towards maximising profit is fundamentally misaligned with the Housing department’s developmental objectives, and this is where a large part of the disconnect between the two entities originates (ibid.). The second delivery programme aims to increase the delivery of subsidised municipal rental accommodation for very poor households through the expansion of the existing JOSHCO social housing model (CoJ Housing Department, 2017a). There is some overlap with the TEA programme as three of the 11 buildings listed in the 2021 TEA Readiness Report are earmarked for old age rental accommodation (CoJ Housing Department, 2021b). This is a focus of the Housing department as no new old age facilities had been developed by the City since 1994 (ibid.; Sotomi, interview 22 December 2021). One of the buildings currently being used for TEA may be used for old age rental accommodation as well (Sotomi, interview 22 December 2021). Beyond this focus on old age rental accommodation and the conversion of TEA into municipal rentals, there was no other indication of delivery on programme 2. Delivery programme 3 encourages the private sector to deliver large numbers of cheap rental rooms by providing capital subsidies and incentives that make it financially viable for them to do so (RebelGroup, 2016). Six interviewees were aware of an initiative by National Treasury to develop an incentive model for private developers. The JDA and Andreas Bertoldi of RebelGroup were involved in the initiative, and an email exchange in February 2018 between Anne Steffny on behalf of the Johannesburg Inner City Partnership (JICP) and a JDA official confirmed that ‘the incentives discussion this afternoon, with GTAC [Government 43 Technical Advisory Centre] colleagues comes from one of the programmes in ICHIP, and it specifically tries to explore incentives to increase the affordable / inclusionary housing stock’ (Steffny, personal correspondence 5 February 2018). However, according to Bertoldi and Steffny, the model was never approved (interviews 23 November 2021 and 8 December 2021). The objective of the fourth delivery programme is to increase the delivery of social rental housing,