Lip service: How voices from informal settlements were sidelined during the first decade of local democracy in South Africa

Chapter 3 in the book 
 Refractions of the National, the Popular and the Global in African Cities.



Introduction
A majority of the South African population has been urban since the 1990s (Kok et al. 2006: 11).Over 22% of this population lives in the province of Gauteng, which, after the Western Cape, was the country's only other province experiencing net in-migration at the time of the last census (StatsSA 2011: 2).Local government in the province of Gauteng includes the three Category A or Metropolitan municipalities, namely the City of Johannesburg, the City of Tshwane, and the City of Ekurhuleni.This chapter provides an examination of the top-down treatment by two of these metros of those living in their informal settlements over the course of the first decade of democratic local government (2000)(2001)(2002)(2003)(2004)(2005)(2006)(2007)(2008)(2009)(2010), during which both cities were governed by the country's liberation movement turned ruling party, the African National Congress (ANC).
The City of Johannesburg Metropolitan Municipality is a single tier local authority incorporating a geographical area formerly governed by 11 different local authorities with disparate capacities.Johannesburg had begun as a tent camp that mushroomed there during the gold rush to the Witwatersrand from 1886 but had grown quickly and sufficiently to support travelling traders and several sectors of industry.For most of the city's history, the majority black population has been confined to overcrowded residential areas on the periphery of the city and largely excluded from the white centre.
The City of Tshwane Metropolitan Municipality governs the city of Pretoria and its surrounds.Pretoria had been founded as capital city of the Boer republic, the Zuid-Afrikaansche Republiek, in 1860.The geographical boundaries of the city have changed a number of times over the course of its history.Coinciding with the date of the local government elections that took place just after the period covered by this study, the incorporation of an additional hinterland municipality, Metsweding, made Tshwane the third largest municipality in the world by surface area, at 6 368km² (City of Tshwane n.d.).
Municipal expenditure per capita, including investment in infrastructure, is much higher in the metropolitan municipalities than in South Africa's peri-urban and rural areas (SACN 2011).The impressive record of house building in the metropolitan municipalities has, however, been 'over-shadowed by the sizeable, expanding population of these areas compared to the rest of the country' (SACN 2011: 50).The building of record numbers of low-income housing has not prevented substantial proportional growth in the informal settlements of the urban areas of Gauteng and Cape Town, compared to a slight decline in the proportion of informality in the rest of the country.
The ways in which the state takes account of, and acts on, informality presents an important area for considering state-society relations (UN-Habitat 2003).The South African challenge has its roots in colonial and apartheid legislation which impacted decisively the incidence, form and distribution across the landscape and population, of informal housing.The tangible decline in recent electoral support of the ANC in South African metropolitan councils, moreover, raises the question of state-society relations under early democratic rule by South Africa's liberation party.As other research has indicated (Huchzermeyer & Karam 2006;Robins 2002), strategies of the post-1994 government have perpetuated and exacerbated informal settlement.
Democratic South Africa, in constitutional terms, endorses a three-sphere (rather than three-tier) system of government -the national, the provincial and the local -so as to offer equal voice to elected leaders in each sphere.In addition, in the local (and metropolitan) sphere, elections produce an equal mix of proportional representation and ward-based direct representation of individual candidates.Both these constitutional principles are aimed at ensuring that voice is offered to individuals at local level, the first through elected councillors by political party and the second through elected ward councillors representing residents' interests at the micro-level.This is the basis for the national government's pledge that public participation in planning and implementation of policy includes the voice of the poor and of residents in informal settlements.
Along with basic service delivery strategies, legislation that obliges local authorities to consult their constituent communities (Republic of South Africa 2000) has accompanied the consolidation of democratic institutions in South Africa.Wardlevel participation feeds into each municipality's Integrated Development Plan (IDP), which becomes a legally binding programme of development and expenditure upon being ratified by a local council (Republic of South Africa 2000).Characteristics of integrated development planning include community participation; a strategic focus to make the most of limited resources; integration between sectors; and to have an outcomes and delivery orientation.The municipalities must prepare five-year IDPs that are accompanied by a financial plan.Every year, the IDPs are also subject to a review process.This pledge of the national government to ensure listening to popular voices from below through its public institutions, both elected as well as appointed, is the focus of this chapter.An overview of analyses focused on bottom-up participation by urban residents is followed by a short summary of the research methodology used.Subsequently, five constraints 'from above' on real devolved power to the metros of Johannesburg and of Tshwane and to their residents will be identified.
Before concluding, the role that the state plays in the production of urban informal settlements is broached.The conclusion itself will argue that the national state's centralising forces severely limit the extent to which local state institutions are able to incorporate the voices of informal settlement residents into their planning and practice.Accordingly, the national rhetoric of democratic participation, particularly in South Africa's metropolitan areas, is not accompanied by concomitant practice.

Shortcomings of formal participation from the bottom up
State-created mechanisms of participation are what Cornwall (2002) has labelled 'invited' spaces of participation in order to distinguish them from 'invented' spaces of participation in which communities respond on their own terms.Since only genuinely enabled citizens can realise the potential of popular participation to influence government, truly democratic participation spaces are difficult to achieve even in societies with low levels of inequality (Esau 2007).
In the South African context, invited spaces of participation have been criticised by many scholars.These state-created mechanisms of participation may • tie communities to pre-set agendas (Heller 2009;Putu 2006); • be dominated by ward councillors' political allies (Sinwell 2010); • rely on overly technical discourses that ordinary people find difficult to follow (Putu 2006); • treat as aberrations the differences within communities (Friedman 2007); • tend to de-radicalise or 'domesticate' social movements (Sinwell 2008: 245); • privilege the organised over the non-organised (Friedman 2007); • manage to attract particularly the middle class (Bekker & Leildé 2003); and • present barriers to the poor, in particular those in the informal sector (Friedman 2007;Mohamed 2006).
These shortcomings of formal participation processes have been offered as de facto explanations for the slow progress with regard to the social welfare of poor communities.In most cases listed here, the focus is upon difficulties regarding the mobilisation of the agency of the poor, both from below as well as via the local authority.Moreover, this planned mobilisation ought to take place within state-created institutions.

Methodology
This chapter draws on a study of the local state housing sector in democratic South Africa as it responds to informal settlement (Groenewald 2012).The research took the approach of studying up to focus on local state approaches to informal settlement in two major cities for the decade after the local government elections held on 5 December 2000.Until 2000, the continued reliance on racially defined voting districts had extended white control at the local level, in what has been termed a 'delayed transition at the local level' (Bekker et al. 1997: 38-43).The period studied therefore represents the first decade of local democracy in South Africa, during which the character of the post-apartheid local state has emerged, together with the kind of urban citizenship that has been constructed under the ANC.The research sites were the local municipalities in the cities of Johannesburg and Tshwane, which, for the period covered by the project, were two of six metropolitan councils among the 284 South African municipalities.In-depth interviews were conducted with 23 senior appointed and elected officials responsible for housing.More detail on these interviews is available (Groenewald 2012).
During the research process it became clear that in each of these two cities a dedicated unit coordinated the planning process and drew up a coherent IDP, taking into account the needs voiced during this annual community consultation process on the one hand, and the electoral promises of the majority party, on the other.Long-term strategic city priorities and sector-specific inputs were also considered when council made adjustments to the IDP.Actors within the local state felt that the legislated obligation to consult communities was indeed being met through an annual round of community meetings.They further reported routinely going beyond the legal requirements of the IDP process.In both cities, the mayor had arranged opportunities for direct engagement with community members, departmental community meetings were sometimes held, and service delivery satisfaction surveys had been conducted.Every participant in the study mentioned instances where they had personally been involved in community consultation.Involvement was therefore limited neither to elected politicians nor to appointed officials; neither to policy decision-makers nor to those responsible for practical implementation.Both elected and appointed staff spent a substantial amount of time in evening and weekend meetings listening to community members.Further consultation of specific communities had also been a requirement of specific court judgments.
A number of the most influential participants in this study were asked to comment on the relative importance of a range of influences over local state policy and practice, particularly regarding participation.In addition to the internal shortcomings of formal participation listed above, five factors that limit the reach of local participation can be distinguished in the accounts of these senior decision-makers within the housing sector.These five structural constraints are • planning priorities set from above, • the separation of implementation and accountability, • the failure of the courts to provide consistent protection to the poor, • ineffective needs assessment and • inadequate funding.

Priorities from above determine funding and performance targets
Participants indicated that the intentions of the ruling party played an important part in determining local state programmes.The IDP had to balance needs articulated by communities with the priorities and electoral promises of the ruling party.
Councillors and officials described the mechanisms by which ruling party priorities were incorporated into planning in very similar terms.Elected councillors valued their party caucus as a forum that could be used to ensure cohesion in policy and practice, while the Mayoral Committee provided direction with regard to various policy areas.This was the case in both cities.
The participants also explained how their performance management was informed by political priorities.The mayor of a city would pronounce certain priorities during the State of the City address.These priorities would feed into the IDP, which would in turn be used by city departments to develop their detailed scorecards.A city's Annual Performance Plan would include all these detailed targets.
The system mimicked the target development process described by officials in the provincial and national government.At provincial level, the budget speeches of members of the Executive Council would feed into the targets of provincial departments, while at the national level, goals set in the president's State of the Nation Address would be broken down into various ministers' portfolios, and this would be broken down into targets for various sections of a national department.The targets of provincial governments would of course inform their budget allocations to municipalities.
Over and above this, officials in the national Department of Human Settlements were responsible for ensuring alignment between the IDPs, provincial plans and the national planning framework.The Department of Human Settlements conducted imbizos that gave community members the opportunity to talk directly to staff of the department.Actors in the national Department of Human Settlements said that strategic priorities were driven by articulated demand.They expected municipal IDPs to reflect the priorities expressed in the president's State of the Nation address and national and provincial priorities, along with local demand.Central state participants were frustrated when municipal IDPs did not reflect national and provincial priorities which, they argued, were also informed by articulated demand.Projects needed to be initiated by municipalities, but aligned with provincial plans, otherwise they would simply not be funded by the national department.The devil lies in the detail: due to the timing of the financial year-end at municipal and provincial levels, provincial planning is completed earlier in the fiscal year than municipal planning which places municipalities at a disadvantage.All in all, the metropolitan local state was under significant pressure to align its own plans to those of the central and provincial government.

Separation of implementation and accountability
In the local state housing sector responding to informal settlement, the sway of central and provincial government over local government was further entrenched by the fact that accountability was located at a different level of government to that for primary control over funding and implementation.
Participants' descriptions divulge a practical separation between those who are actually responsible for implementing the bulk of housing projects and those who consult communities.
The central government divided the housing budget among South Africa's nine provinces which then had control over the further division of the budget among allocations to municipalities and the provincial government's own operational and capital expenditure, including housing projects.No downward accountability to specific constituencies is built into the design of South Africa's provincial level of government.Instead, all elected representatives come from a party list.
Provincial officials generally described their role as funded delivery.A technical feasibility study, which included a land suitability survey, was conducted as part of the planning of each provincial housing project, and layout plans were advertised for public comment.Provincial plans would usually be presented to affected communities prior to approval.'Greenfields' projects, in which nobody resided on the land earmarked for development, would involve no community consultation, while other cases might warrant social impact assessment.A dedicated communication division was responsible for handling housing project-related queries from the general public.In some cases, opportunities for direct engagement between the provincial government and the community were arranged by the municipal speaker's office, when the municipality held meetings in informal settlements, or when the central government held imbizos.For example, when the members of surrounding suburban communities had initially objected to a mixed income state-supported development on the grounds that it would affect their property values, these objections had decreased over time, said Ned, a provincial official.Participants in the provincial government sometimes attended community project updates and community consultation opportunities arranged by municipalities so that the relevant background or technical information could be provided.Consultation by the provincial government which exercised the primary control over decision-making about housing was therefore primarily reactive and technical.
City councillors in the study considered their engagement with communities a critical contribution to the housing process, which finds some support in the fact that councillors' knowledge of the grassroots level was both acknowledged and valued by their provincial counterparts.Provincial participants said that they relied on municipalities to conduct in-depth consultation and to provide information about community priorities and preferences.However, the portion of the housing budget controlled by municipalities was small and funded primarily from service delivery, their current account.Although municipalities had the primary responsibility for local planning based on the IDP process, implementation of the priorities identified in the IDP depended on the size of the housing capital budget available at the local level.
Provincial participants experienced this as a functional division of labour; an experience which may well have been moulded by their location within a province where the local and provincial state is governed by the national ruling party which determines the overarching policy framework.The framework removes accountability to grassroots communities from provincial officials who implement housing projects that directly affect those communities, but at the same time, holds accountable local authorities that have limited influence over the distribution of the provincial housing budget.Officials feared that further consultation would add fuel to the fire of already frustrated communities.The leaders who were accountable to the constituencies were therefore not the ones who had real power over implementation.
In summary, a separation was practised between the provincial government responsible for budget allocation and implementation and the local state responsible for community consultation and local planning.The division of competencies between the provincial and local state left ordinary residents without direct access to or influence over the primary housing decision-makers.

Court rulings give inconsistent guidance
The local state is closest to ordinary people.It has, therefore, become the focus of popular contestations over the content of national citizenship.Compared to the rhythm of state planning processes and ruling party meetings, court judgments were ad hoc and relatively unpredictable.Nevertheless, they had a direct influence on the practice of local authorities.Two court cases against the City of Johannesburg, in which the City of Johannesburg and other government departments were challenged with respect to the provision of water and housing led to limited gains for pure communities (see Mazibuko and Others v City of Johannesburg and Others; Occupiers of 51 Olivia Road, Berea Township and 197 Main Street Johannesburg v City of Johannesburg and Others).
The local state could also seek eviction orders from the courts with regard to habitual non-payers residing in social housing, or in cases where public land had been occupied for longer than six months, or seek ownership of a bad building that had been abandoned by the owner.Although this was not raised by participants, the local state in both cities had successfully applied for several eviction orders issued under various circumstances.In the City of Tshwane, the courts were ostensibly also being used to test in which cases the city would be required to formalise or service informal settlements and when they could be evicted or tolerated while being provided with rudimentary services only.This transpired from a comment by Nellie, a Tshwane councillor, who said that legal informal settlements were 'the ones where the court has instructed us to provide services.'Court judgments, therefore, provided neither reliable relief to the poor, nor lasting guidance to the local state.
Metropolitan municipalities lobbied for greater local state control over housing programmes, partly because they had been taken to court by residents with regard to housing and shelter.But despite a number of limited victories for poor communities, the local state had also successfully applied for eviction orders issued under various circumstances.While officials and councillors in this study declared that they had no choice but to adhere to court judgments, they sometimes acted in contravention of the law by evicting residents without a court order.In the case of informal settlements razed to the ground, there was no immediate possibility of return, regardless of court rulings.The series of court cases between city residents and the local state have, therefore, not consistently resulted in the protection of the poor.
One reason for this inconsistency was apparent in the aftermath of one of the early cases of court-based contestation over citizenship that had faced the post-1994 South African government.In the prominent constitutional court case of Irene Grootboom, the provincial government was instructed to provide the funds for the relevant municipality to provide tents, toilets and water to the informal settlement community where Grootboom lived (Grootboom and Others v The Government of the Republic of South Africa and Others).The judgment also recognised the right to housing, but like other cases in which citizens have taken the state to court over services, failed to resolve the question of fiscal responsibility (Huchzermeyer 2004).While popular voices were often validated by court rulings, this did not lead to consistent improvement of the fate of South Africa's poor.A constitution that is globally renowned for obligating the state to progressively realise rights, has not sufficiently challenged a system designed to privilege central control over local spending.

Inadequate funding restricts local state responsiveness
The claim by participants that the allocated funds from the central state were insufficient for addressing the identified housing need is supported by several pieces of evidence.First, as mentioned above, the two metropolitan municipalities in this study exhausted their housing budgets by means of project implementation.
Second, the number of houses that can be delivered by the local authorities based on the available subsidies constitute a small fraction of the known housing need.Given the strict criteria for subsidised housing stipulated by the national Housing Code, however, estimates of housing need are likely to underrepresent the population unable to access adequate shelter via the open market.As mentioned above, the state has also placed the burden of self-identification on those who wanted to register their need on the housing demand database.Third, the central budget allocation for Housing has never reached the 5% of the national budget that the ANC envisaged during the transition (Pottie 2003).
Local authorities each have an operating and a capital budget that must be prepared together, since commitments in the capital budget incur operating costs.Increases in operating expenditure need to be considered before decisions are made about capital projects to ensure that the budget will be balanced (ETU n.d.).Capital budgets are for longer-term developments or assets and may be funded from a combination of loans, donations, government grants, public-private partnerships or the operating income in a particular financial year (ETU n.d.).Operating budgets, on the other hand, cover the day-to-day costs and income to deliver municipal services and are funded from the rates and taxes that municipalities collect together with their equitable share of national revenue (ETU n.d.).Since the equitable share depends on the number of poor people within the municipality's area of jurisdiction, rural municipalities benefit more than urban municipalities, although the equitable share covers only a small share of municipal operating expenditure (ETU n.d.).Local authorities are therefore effectively expected to be self-sufficient for the purpose of service delivery.
The severe shortage of funds limits the capacity of the local state to respond meaningfully to the priorities identified by community consultation.These priorities and those identified in the most recent ruling party election manifesto(s) together inform the Integrated Development Plan (IDP) of each local authority.The priorities captured in the IDP determine the order in which the proposed projects for a specific financial year are funded until the budget has been fully allocated.Once the allocations have been depleted, no further projects can be approved.Compared to a properly funded local state where various models of meaningful democratic control over the budget can be pursued, inadequate funding of the local state limits the influence of community consultation to that of prioritising expenditure which essentially pits one poor community against another.
The value of prioritisation is severely diminished by the fact that only a small proportion of the population in need of assistance can be assisted with the available budget.The waiting periods that result from this relationship of need to provision vary widely but have been roughly estimated at a decade.The burden of self-identification placed on the poor has therefore been deepened by the obligation to retain the status of a qualifier for more than a decade by remaining in the lowest income category and continuing to have dependents.Should the ten-year period be sufficient for the state to provide that person's needs, but the dependents reach majority age after nine years, the prospective beneficiary would be disqualified from receiving subsidised housing.The prospective beneficiaries are effectively expected to be patient indefinitely while adequate shelter continues to elude them.
The challenge cannot be expected to dissipate, as contemporary South African cities reflect the trend of previously unparalleled urbanisation taking place across the developing world (UN 2016).In the meantime, local authorities are left with little choice but to turn to partnerships, giving the private sector exaggerated influence compared to ordinary people.

Ineffective needs assessment hinders planning and burdens the poor
In the province of Gauteng, both the high rate of in-migration and natural growth contribute to the need for housing.Local state officials and councillors were aware of projections with regard to the urban growth rate in general, including the rate of household formation in the lowest income categories and the rate of in-migration.Participants in Johannesburg mentioned that in-migration together with the rate of new household formation meant that backlogs were constantly growing.Maureen, a Johannesburg official, gave this account: If you look at the City's Growth and Development Strategy, it talks about all these parameters.We have got a Department of Economic Development in the City where we have got senior economists who will tell you that in terms of population growth we are going to be growing at 1.4% year in year out and because Johannesburg is the key economic hub of Africa, the City is supposed to grow at this X amount.But 9 out of 10 (times) that is not necessarily matched with available resources and that is one big challenge.… So we are going to end up here and we hope by some miracle that we will have a blip.Because Local Authorities' funding is based on the ability to get more revenue for your Council growth and if unemployment goes up and up and up, our ability to increase our revenue base gets eroded as well.... It sounds like an excuse for a government official to say maybe we must raise the rates and taxes by 70%, but the truth is if you look at migration into the cities, it is higher than what we can afford.
The degree to which systematic needs assessment and projections influenced planning was explored in interviews.Census data and household surveys conducted by Statistics South Africa would inform national planning priorities and the distribution of funds, but the frequency of the census has been decreased to once every ten years.In addition, the South African census is plagued by questions about accuracy, including undercounting in excess of international benchmarks (Berkowitz 2013;Schultz 2013).Whether accurate or not, these tools appeared to be largely ignored when it came to planning at both the provincial and the local level.Instead, all participants considered principal source of information on housing need to be the housing demand database.
In 2007, the approach to housing needs assessment in the province of Gauteng had changed from that of a waiting list with limited information about prospective beneficiaries and their needs, to a demand database that disaggregated need in terms of the kind of housing that the prospective beneficiary wanted and in terms of the income category of beneficiaries.The participants in government were visibly animated by the cleaning of the demand database which also represented a way to ring-fence the vast and growing need for housing.The original post-1994 housing waiting list had been problematic because it simply contained a list of people who had indicated that they needed state subsidised housing and failed to disaggregate need.The list had not been cross-checked against other government databases to exclude non-qualifiers who had either claimed to earn lower salaries than they actually earned or who had claimed to have dependents who did not actually qualify as dependents, and people who received houses elsewhere had not been removed from the list over time.
The list was shortened by means of data cleaning exercises that removed people who were listed with incorrect ID numbers, who had obtained or applied for housing elsewhere, or who did not meet or no longer met the criteria to qualify for subsidised housing -for example, if their dependents had become adults or their salaries had increased.In addition, prospective beneficiaries on the waiting list were asked to show themselves at state housing offices.Such an indigent approach that shifts the burden of proof of poverty onto marginalised households ignores their lack of resources to prove that they qualify for assistance (McDonald & Pape 2002).For this reason, the burden of self-identification could contribute to a further underestimation of need.Following the clean-up, the total number of prospective beneficiaries in the province of Gauteng on the database had decreased from around 800 000 to over 600 000.
While the new demand database was a more reliable indication of the number of people who had self-identified and qualified for subsidised housing at a given point in time, it could not be expected to accurately reflect the need for assistance with access to shelter.In addition, because the list was not publicly available and because programmes did not respond to prospective beneficiaries in the order in which they had registered, communities had no way of knowing whether they had been bypassed and whether nepotism or corruption played a part in beneficiary selection.Moreover, the demand database was perceived to be applied in a corrupt and arbitrary manner (Rubin 2011).
While the local state broadly relied on regular community consultation to inform the IDP, the demand database where the poor had self-identified their need of housing assistance was the primary mechanism used to assess housing need.
With regard to informal settlements, shack registration exercises were conducted that could also identify where there was a need for intervention.The provincial administration had conducted an audit to identify and record the location of all informal settlements in 2005, an exercise that was mentioned and considered helpful by all the participants in understanding the demand for state-funded housing.However, information on needs assessment was held by the provincial office and did not feature prominently in the planning processes in which provincial administrators were directly involved.Since the participants were animated by the provincial audit, they appeared not to recognise the need for a method that would measure and predict need over time, rather than to provide a once-off snapshot of housing need.
To avoid shacks being used as a 'queue jumping mechanism', the allocation of subsidised housing was linked to a shack number in the city's database when the occupant of a shack became a beneficiary of a government programme, said Maureen, a Johannesburg official.This local state approach therefore served to strengthen the Housing Code focus on individual beneficiaries rather than geographic areas or communities, which in turn limited the scope of state housing assistance.Planning and budget allocations responded to a housing demand database that reflected an overly static understanding of need on the part of the central and provincial state, whose practices placed the burden of identifying need primarily on poor households rather than on the state.
In both cities, the list of people on the housing demand database did not overlap closely with the residents of informal settlements, as indicated by participants and corroborated by staff of Statistics South Africa.A substantial proportion of people living in informal settlements did not qualify for subsidised housing.In addition, many temporary workers and migrants did not want RDP housing in the city, whereas candidates for family units did not want social housing, which could entail a lifetime of paying without owning anything.Participants had taken cognisance of the fact that a one-size-fits-all approach to housing was unlikely to succeed, as many migrants did not require family housing and a substantial proportion of people without adequate housing did not qualify for fully subsidised housing.
Participants in the local state knew that the estimated need for housing was increasing at a high rate due to both internal growth and in-migration.The state actors were sufficiently familiar with the results of needs assessment exercises to be able to summarise the scope of housing backlogs in the two cities.They were aware that the bulk of housing need in the province was concentrated in the two cities and participants in each city knew in which specific parts of their city need was concentrated.Yet, projections did not play an important role in planning, as the budget allocated to the local state did not even allow it to clear housing backlogs.The number of people in need of fully or partially subsidised housing far exceeded the number that could be assisted, given the funds allocated to address informal settlement and housing.The local state did not have at its disposal any housing instrument that could adequately respond to known need or urban growth.
The obliteration of procedural justice is most starkly accentuated by participants' comments about community participation.Housing need was acute and demand was articulated at every community consultation opportunity at the local level, said Gladys, a Johannesburg official.In Johannesburg, where there was no indication of a failure to spend allocated funds, some participants feared that further consultation would serve merely to raise expectations, although no surplus remained to respond to any additional consultation.The phrase 'social exclusion,' used when multiple deprivations associated with poverty push people to the margins of society to the extent that the advantages of citizenship elude them (Abercrombie et al. 2000), can therefore be applied to the residents of informal settlements in contemporary South Africa.

Informal settlement as a product of state policy and practice
Debates about the definitions of informality will be familiar to scholars in this field.
Where Huchzermeyer (2004) has argued that informality should be understood not in terms of its lacking physical or top structure, often contrasted to formal dwellings, this argument has not relied on romanticising living conditions in informal settlements.
Instead, it has represented a careful and crucial shift of emphasis away from the inadequacy of the dwelling built by the poor, towards the vulnerability contributed by the state (Huchzermeyer 2011).This approach held the key to shifting the state response away from eviction and towards the improvement of tenure security and living conditions of residents of informal settlements.Nevertheless, it is often because of their gains and aspirations that residents of informal settlements exercise a choice for a precarious place in the city over no place in the city at all (Groenewald et al. 2013).Defining informal settlement in terms of its vulnerability, in particular to eviction, avoids degrading descriptions that trample on this agency, and is accurate in locating responsibility with the state.If residents of informal settlements are unable to extract recognition and responsiveness from the local state, this cements their position of precarity and in particular, extends their risk of eviction.The additional state contributions to precarity identified here, therefore, constitute the state (re)production of informality.The theory that the state produces informal settlement has therefore been strengthened.The centralising tendency of the South African state, therefore, contributes to continued informal settlement.
This interpretation resonates with what I would like to call a Fanonian understanding of the state.Fanon (1963) locates the cause of the native with the coloniser.His argument that colonial administrations are responsible for the existence of the colonised, systematically links a series of characteristics associated with natives to a series of ways in which indigenous peoples are subjugated under colonialism.He demonstrates how colonialism dehumanises its subjects by removing the minimum requirements for a humane existence, setting the example of acting in an inhumane manner and casting the subjects as less than human (Fanon 1963).Fanon illuminates the direct causal relationship of colonial subjugation with the failure of postcolonial governments to improve the plight of the majority of indigenous people.While this perspective took time to permeate the mainstream, when it did so, it caused three critical paradigm shifts: social scientists began to include non-physical factors such as economic capacity into their concepts of race; they began to conceive of these nonphysical characteristics not as inherent, but as inherited from past discrimination; and there was a growing recognition that race discrimination was the cause of the distinct positions of 'race' groups in society, rather than the other way round (Groenewald 2004).The same advance has yet to materialise in the dominant discourse on informality.

Conclusion
Alongside the various internal constraints of formal participation that have been described in the literature, a variety of factors from above limit the potential of participation processes at the level of the city or local state in South Africa.The influence of participation was severely curtailed by the structural constraints on the autonomy and budget of the local state, as compared to other spheres of the South African state.Consultation has therefore not translated into the ability of those in informal settlements to make their voices and priorities count.Poor communities often take to the streets in an attempt to evoke a degree of urgency over their desperation.
The legislated entrenchment of democratic participation at the local level offers scant comfort, as the little influence that residents are able to exert serves to prioritise project expenditure.Even full participatory budgeting could leave residents frustrated, as only a small portion of priorities can be addressed with the available budget.In sum, the political right to self-management and participation in decision-making by residents of informal settlements has been stripped of substance.Under these conditions, neither the reform of invited participation spaces, nor their abandonment in favour of insurgent revolt, can realise social welfare.The meaningful participation to which the Constitution of South Africa pays lip service, requires a national 'will' to build local authorities with greater autonomy and fiscal clout.
The decentralising rhetoric of the national state diverges substantially from its centralising practice, which helps confine the residents of South Africa's informal settlements both to the periphery of urban areas and to the footnotes of the economy.They form part of a global underclass that is being abandoned, forgotten, or wished away on the outskirts of world cities, disconnected entirely from the potential opportunities of a fourth industrial revolution.
In the context of a declining economy, pro-poor activists will have to weigh up the potential of different strategies to respond to this dilemma.Support of local entrepreneurship, including the informal sector, can help, but may not be enough to counter the effects of rising unemployment.Entering the fierce competition for a larger slice of a shrinking national budget pits poor communities against government departments that already serve them inadequately.Competition for scarce resources at the local level can also pit one poor community against another, whether in formal, invited participation spaces, or on opposite sides of barricades and burning tyres.All of these strategies are important for strengthening local communities, but their respective pitfalls make it more urgent to consider a Basic Income Grant as a viable policy option for giving ordinary people more control over their own livelihoods and dignity, including their options for shelter.