Land, markets and informal settlement policy in south Africa

Transcription

1 Land, markets and informal settlement policy in south Africa Marchés fonciers et politique de régulation des établissements informels en Afrique du Sud Lauren Royston, Development Works and Leap project Monty Narsoo, independent consultant and Leap project Abstract This paper adopts both a policy and a local lens to interrogate the land market theme. The policy lens reviews the current policy environment in South Africa, adopting a mainly urban and human settlement focus, for how it addresses the land markets question. It provides contextual background to the evolution of the settlement policy in South Africa, emphasising the key policy debates that informed it. It then plots how policy treats the issues of titling, assets and markets. The local lens is a case study approach drawing on recently completed research undertaken in an informal settlement in Johannesburg Zandspruit which addresses the land transaction issue. With the likelihood that at least households will not be accommodated in formalisation plans, the paper argues that the potential for conflict exists around insiders and outsiders, or qualifiers and nonqualifiers, and that the conditions of scarcity are conducive to the operation of a local market in land. It contributes elements of a conceptual framework for describing and understanding the workings of a local land market and identifies some of the characteristics of that market in Zandspruit. The paper concludes by identifying the importance of gathering and synthesising material on local land markets, about which little is generally known and understood at present, in order to assemble a body of evidence significant enough to enter into the public discourse and influence policy in ways that accommodate poor and vulnerable people, households and communities. Key words: Human settlement policy, local land markets, tenure Colloque international Les frontières de la question foncière At the frontier of land issues, Montpellier,

2 Résumé Cet article explore le thème du marché foncier en apposant deux points de vue particuliers: l un politique et l autre pratique. Le point de vue politique examine la manière dont ce thème est abordé par la politique de development urbain en Afrique du Sud, en présentant son évolution, à partir des débats qui en sont à sa source. L article évoque, en suite, la façon dont cette politique aborde les questions de l enregistrement des titres parcellaires, du capital et du marché foncier. Le point de vue pratique est représenté par une étude de cas, tirée d un travail de recherche sur la question des transaction des biens fonciers, dans un quartier informel, en cours de régularisation, de la banlieue de Johannesburg - Zandspruit. Près de 2000 familles, résidentes à Zandspruit, seront exclues du processus de régularisation. L article explique que ce processus présente un risque important de conflit entre ceux qui seront bénéficiaires du processus de régularisation, et ceux qui en seront exclus, en fonction des critères d accès aux subventions associées à de tels processus, mais aussi que ces conditions sont particulièrement propices à la création d un marché foncier informel. Cette étude de cas contribue à la définition d un cadre d analyse pour décrire et comprendre le fonctionnement d un marché foncier informel en identifiant ses caractéristiques telles qu elles sont pratiquées à Zandspruit. En conclusion, l article plaide en faveur du rassemblement et de la synthèse de travaux sur les pratiques associées au marché foncier informel, afin de combler le manque de connaissances et de compréhension à son sujet, et d établir un compte rendu d ordre suffisamment important pour influencer les débats et les décisions politiques de manière a ce qu elles répondent enfin aux besoins pratiques des personnes et communautés défavorisées et vulnérables. Colloque international Les frontières de la question foncière At the frontier of land issues, Montpellier,

3 Part 1: The Human Settlements Policy Process in South Africa Introduction In the current development discourse, land and property relations occupy a central position. For example, Hernando de Soto argues that property title holds the key to making capitalism accessible to the poor. More humble claims, in a different ideological camp, argue that land plays an important role as a livelihood asset, that it has both a use and exchange value. These are the discourses within which many public policies (and research processes) are engaged, including the human settlement policy process in South Africa 1. These positions often sit uncomfortably alongside one another in public policy that seeks to be as much as possible to as many as possible including the dominant ideology with the state and ruling party, the private sector investment interests and the poor themselves (noting that the poor are not a homogenous grouping with a single set of interests). South African Policy Context In the negotiations on the development of Housing Policy in South Africa in preparation for a new democratic dispensation, a number of external models were looked at and in addition a number of foreign agencies proposed particular perspectives on the way housing and settlement policy should be carried forward. This was taking place against the twin political issues of developing a rights-based constitution in the country and the major international changes in relation to the cold war in the early 1990s. Ideology, or the lack of it, became central to the negotiations. These negotiations had to also take into account the peculiar history of South Africa which always hinged on what was dubbed as the `Land Question. The forced removal of indigenous peoples from at first their rural lands and then, when labour was imperative for the economy, they were given temporary sojourner status in the urban areas with no access to ownership, with few exceptions in the urban areas until the mid-1980s. The change in the 1980s was a result of massive resistance, particularly in the urban areas and led to widespread land invasions fuelled by the increased urbanisation process. In South Africa the urbanisation process was managed by influx control using administrative means to limit influx into urban areas by particularly African people and by segregating living areas. However, as these measures broke down and urbanisation gained momentum, huge backlogs emerged in water, sanitation and associated services. 1 In South Africa, Breaking New Ground is the government s comprehensive new plan for sustainable settlements. Colloque international Les frontières de la question foncière At the frontier of land issues, Montpellier,

4 Accompanying these programmes was an attempt to introduce piecemeal limited market economies, which was most evident in the transport and land housing markets. So the imperatives that emerged during the first set of negotiations on housing and settlement policy were: The changing international political and economic climate; The restitution of indigenous peoples land rights ; Mushrooming informal settlements as result of urbanisation; The marginalisation of the poor into peripheral areas as result of previous segregation policies; The huge backlogs in terms of services; and The introduction of the private sector into housing and land markets. In trying to resolve these problems a number of international examples were looked at, particularly within the European and Latin American context: The Chilean programme of housing and title; The Peruvian Cofopri project of titling; The Brazilian programme of turning favelas (slums) into bairos (suburbs) with citizenship as its core; and the European models of social housing. The Policy itself The outcome of this policy debate was that: The first issue was to address the housing and associated backlogs and therefore the target of one million houses in five years was introduced; The second issue was to redress the dispossession of people of their land and therefore titling was central to the housing programme both in terms of new and existing stock; The third issue was to bring the market into play so that these new assets would provide the basis for financial leverage; and The final issue would be to reintegrate urban areas in racial and class terms. The debate on tenure and titling has to be seen in the broader context of the nature of the state and the structure of the economy and urban settlements. The issue rests on understanding the interregnum between formality and informality in the state, the economy and settlements. A further two arise in this context: Colloque international Les frontières de la question foncière At the frontier of land issues, Montpellier,

5 How to integrate the city in spatial terms; and How to relate the first and second economies. These two issues relate to a number of sub issues in terms of which tenure plays a central role. The first key characteristic of the implementation of the original housing programme has been to provide tenure in terms of a title deed and provision of services such water, sanitation, and electricity. The two were interlinked; if households did not have title their ability access services, including free basic services, were severely circumscribed. By and large this programme worked in the first seven years. The second key characteristic of this programme was to leverage private sector resources such as construction and housing finance. In terms of construction, the bulk of the delivery was by the private sector. However there was no significant housing finance into the low income market. Although the formalisation of tenure had occurred, the automatic conversion to formality has not translated into assets being marketable. The overall underpinning of the original housing programme was that the subsidisation was based on breadth, that is, give as many households limited support, as opposed to providing fewer households with greater subsidies. The calculation was that the market would take care of the rest. Hernando De Soto s formalisation of assets theory did not work. As an afterthought, social housing and rental accommodation was seen as the integrating housing programme in inner cities and as part of the urban regeneration in terms of the first housing programme. With the possible exception of one institution, most social housing institutions are struggling. No significant integration or regeneration interventions have occurred through housing. What was totally ignored was the thriving rental market of backyard accommodation. The shift in policy in terms of the `Breaking New Ground document 2 is the movement to greater depth in subsidies, the emphasis on making markets and assets more valuable through government investment in public infrastructure, and the formalisation of informal settlements through their eradication. The point of departure of the policy of Breaking New Ground lies in the following thrust in relation to tenure and titling, linked to the concept of assets as well as the market: `Tenure and Title Deeds Access to title is a fundamental principle of the national housing policy. However, a significant proportion of housing (formal and informal) for lower income South Africans has not yet been transferred into the names of the entitled individuals. While security of tenure has 2 In South Africa, Breaking New Ground is the government s comprehensive new plan for sustainable settlements. Colloque international Les frontières de la question foncière At the frontier of land issues, Montpellier,

6 been achieved in principle, this is not always supported legally with formal title. The consequence of such titling backlogs is that residents are not able to participate in the housing market with the trade of their housing. This undermines their mobility (and their ability to access work where it is available), their choice (and their ability to find housing that corresponds with their specific needs), and the asset creation potential of the state s investment in their housing. It is estimated that 878,000 houses were developed as Old Township Stock. Of these, 47% have been transferred to date (413,006) at a selling price of R3 billion and an estimated value of R24 billion. In addition, a large number of houses constructed under the existing housing programme have not yet been transferred to households. In order to address these problems, Government will implement a range of measures to stimulate a renewed uptake in the Discount Benefit Scheme in order to transfer the balance of the free-standing public housing stock and will establish a high priority focus on completing the transfer of housing stock constructed under the existing housing programme. This needs to be contrasted with following statement: `Prohibition on Resale Section 10A of the Housing Act prohibits the sale of government-subsidised property for a period of eight years following occupation. This section was enacted to protect subsidy beneficiaries from downward raiding. However, the provision has also had the unintended consequence of undermining beneficiary choice and housing mobility, and creating a significant barrier to formal secondary transactions. For this reason, an amendment to Section 10A of the Housing Act, 1997, is to be introduced to reduce the period to one year following occupation. Whilst this will protect against households using the subsidy as a means to access cash quickly, it will also allow a deliberate sale arrangement to be pursued if the owner so desires. What these statements suggest is that while there is movement to formalise so that assets can be realised, the latter amendment indicates that prohibition was an attempt to stop informal sales because the issue was not title but livelihoods and people sought to cash in formal assets to survive. The issue of unregistered sales is therefore critical in understanding the nature of the land and housing asset to beneficiaries of housing subsidies. These unregistered sales also have a lot to tell policy makers about local land markets. There is a third relevant area that the BNG begins to address and that is of backyard shacks, which is an important arena of understanding local tenure arrangements and markets. Tenure tends to be understood in terms of parcels of land but also has to take into account landlord tenant relationships in highly densified areas where location is primary. The BNG makes the following point: Colloque international Les frontières de la question foncière At the frontier of land issues, Montpellier,

7 Backyard rental accommodation is increasingly recognised as an important component of the overall private rental sector and is plays a significant role in the housing market. Little public policy research has however been undertaken in this realm. In order to develop appropriate policies, the Department will initiate investigate the introduction of several interventions including greater regulatory authority over the landlord-tenant relationship, the development of minimum norms and standards and the provision of home improvement grants linked to counter-funding by landlords. Space, the built environment and their economies do not seem to be integrated in policy terms. Equally, the relationship between rights and power on the streets are not connected. Here again forms of tenure and their relationships to the above issues are driven by the formal understanding of markets. There is a whole set of relationships that operate in the interface of formality and informality, which is also the interface between technocrats and politicians; understanding formal power and informal power. These dilemmas relate to the ability to secure tenure against the twin onslaught of the state s intention to govern and the realities of whose power rules in the urban settlements and the twin economies in operation there. To understand these realities at local level requires a different lens that takes into account what happens on the ground where the local state attempts to manage a myriad of lived experiences and perceptions of what constitutes security of tenure and the subversion and use of state administrative documents to create a place for people in the urban economy. This case study examines relationships in a particular informal settlement and begins to suggest the complexity of how formal governance has to deal with informal practice. Part 2: Case study local expressions of land relations and land rights Introduction A thorough understanding of the complex processes involved in land relations and land rights is still emerging in South Africa, especially the way that these processes are expressed locally (often referred to as informally or extra-legally). Invisibility and lack of recognition carry the risk of sidelining many vulnerable people, households and communities from development opportunities (Leap, 2005). Colloque international Les frontières de la question foncière At the frontier of land issues, Montpellier,

8 Case study background 3 In Johannesburg, households live in informal settlements and backyard shacks. The city s housing master plan maps out a coordinated and realistic course of action to create housing opportunities for the indigent and homeless by 2009 ( The Zandspruit informal settlement is located north of Johannesburg, east of Beyers Naude Drive and north of the Zandspruit River near Honeydew. The settlement consists of a transit camp 4 (1 200 households), private sites 5 (3 600 households invaded land) and an official sites and service project (440 households in the Mayibuye Project) 6. The city estimates that households need housing and has earmarked a portion of land for future housing development. According to the city s initial projections, the formalised Greater Zandspruit will yield housing units. Approximately households are unlikely to be accommodated. This creates potential for conflict around insiders and outsiders, or qualifiers and non-qualifiers, as well as conditions of scarcity conducive to the operation of a local market in land. The Zandspruit picture is not unique to Johannesburg, neither is it unique in sub Saharan African terms, painting as it does an image of local land arrangements in the context of urbanisation and land demand that far exceeds formal market or public supply. Zandspruit then provides a particular local lens of land access by the urban poor, and very possibly segments of low and middle income groups, who do not have access to land provided by the public and the formal private sectors. Findings from recent research in Zandspruit 7 provide insight into how people accessed and continue to access, and transact in, land. This research was undertaken by Development Works for Planact, an urban NGO in Johannesburg. These findings are important in the context of the national strategy s emphasis on markets and assets, as they indicate the existence of a local (or informal) land market about which very little is generally known. This section therefore presents some of these findings in order to make visible the local practices that people have instituted in this particular place. This research is by no means the only material available on local land markets but unless additional material is collected and synthesised into a significant enough body of evidence with which policy makers can engage, unintended policy consequences are likely to be as varied as they are harmful to the local practices that they affect. 3 Research undertaken for Planact by Development Works 4 An area with official but temporary recognition generally provided with some services. In reality a transit camp is seldom expected to be temporary. 5 Four privately owned pieces of land over which the current tenure arrangements are ambiguous 6 All figures are official estimates 7 Undertaken by Development Works for Planact Colloque international Les frontières de la question foncière At the frontier of land issues, Montpellier,

9 Conceptually classifying land holding patterns An assumption underlying an endeavour to reveal or make visible local realities in order to better understand them, develop appropriate recommendations and influence policy intervention, is that the conceptual framework exists to name the practices that are identified and reference them to a larger body of evidence. This is arguably not the case in the public discourse on local land markets in South Africa (if such a discourse even exists). Furthermore, if influencing policy in this respect is initially a process of communication, then a starting point must be the language that is used to engage in dialogue. The common language will derive in turn from the conceptual framework. As a starting point use, is therefore made of a three part conceptual categorisation of informal settlements (Durand- Lasserve and Royston, 2002), although there is considerable diversity between countries and cities. Unauthorised land development or informal subdivision is the main type. Unauthorised land developments are a widespread phenomenon on the fringes of most developing cities. Most often, such settlements have developed on private agricultural land, frequently outside the municipal boundaries. In most sub Saharan African cities, customary owners are the main providers of land for housing, even if their right to the land is not formally recognised by the state. Squatter settlements are a second type of informal settlement, found on the urban fringes or in centrally located areas, mostly on public land but also less frequently on private land, especially when disputed. These can be the result of an organised invasion, or a gradual occupation. Contrary to common belief, access to squatter settlements is rarely free. An entry fee must generally be paid to an intermediary, or to the person or group who exerts control over the settlement, and sometimes also rent. Informal rental housing is a third category and covers a wide range of situations and levels of precariousness. Rental is the most common form of tenure in formal as well as in informal settlements. Tenants and subtenants form a heterogeneous group. They can be found in unauthorised land developments, in squatter settlements or in dilapidated buildings in city centres. Backyard shacks, prevalent in some of South Africa s cities, are an example of informal rental housing. This conceptual categorisation is particularly helpful in the case of Zandspruit for two reasons. Firstly, such great variety exists in residents responses about the manner in which land is currently held in the transit camp and the private plots that it was extremely difficult to paint any picture at all. Secondly, having adopted this conceptual categorisation, the research found evidence that at least two have been in existence in Zandspruit. Colloque international Les frontières de la question foncière At the frontier of land issues, Montpellier,

10 Informal sub-division The settlement referred to as the private plots started with pockets of informal structures on the four privately owned sites some of which were named after their original land owners: Mbele (Erasmus), Woolf, Vuku Zenzele and Breaker Brothers. Initially it seems that people were renting sites from the landowners for R150 to R250 per month a process know in South Africa as shack farming. Demand grew, invasion spiralled and the settlement mushroomed. The practice of a farmer or smallholder renting out sections of the farm or plot as sites for settlement, or shack farming, qualifies as informal subdivision in the classification being offered. In the consolidated oral history of the transit camp that was gathered in the research, the residents interviewed identify a period of shack farming when one Mabaso and Son are said to have sold sites for R200, although they disappeared when it was discovered that they were selling stands that they did not own. This version of informal subdivision differs from that prevalent in the private sites because the legal owner (i.e. the council) was not informally subdividing, but an individual or company ( Mabaso and Son ) was doing so without the legal owner s knowledge. This is probably why non-resident respondents in the research simply refer to this period as one in which people were squatting illegally on the land, although this is not strictly accurate. Informal rental A rental perception dominates in the understanding of how people currently hold their land in the private plots, although the precise nature of the rental arrangement varies depending on who is perceived as the primary owner. In some plots, the period of shack farming has lasted longer than in others with the original owner continuing to extract a levy of some sort through an intermediary. In other private plots, the tenure form can be characterised as a sub-letting arrangement from a shack owner who is living elsewhere (who in turn is either informally renting from an original owner or someone else). This represents an informal market within an informal market. A third informal rental variant is present in the public site, commonly understood as backyard shacking in terms of which officially recognised occupants are renting out backyard space (or shacks) to third parties. The notion of informal rental in the classification being offered belies this complexity but it also aids in describing the tenure arrangements on the ground. The notion of landlord seems to be a fluid concept in the private plots, pushing perceptions of what is recognised as ownership. When respondents refer to landlords, it appears that they are most commonly making reference to a plot owner from whom they are renting their site (most easily recognised by convention), a shack owner who may be letting a shack on a site that is owned by someone else (the distinction between shack and land ownership emerges as significant), or an informal or perceived owner from whom they may be renting their site (least easily recognised and accepted conventionally). Colloque international Les frontières de la question foncière At the frontier of land issues, Montpellier,

11 There is also considerable fluidity in the notion of ownership in the public site. Proof or evidence of land rights correlates highly with security and if they are secure people may be referring to themselves as owners. Officially, council owns the transit camp land. However, many although not all residents interviewed perceive that they own their sites, and have affidavits and receipts as proof of purchase. Site registration with the committee is an important factor in securing sites. Further conceptual work is probably required in the classification employed here, in order to make it more applicable to the local land markets issue. In particular, as it currently stands the issue of land markets is implicit in the classification, as evidence of land transactions demonstrates. Understanding the role of municipal registration in order to understand the local land market Property (shacks and/or land) is being transacted in Zandspruit, whatever the official position may be. An understanding of the nature of those transactions is dependent on an understanding about how the interventions of council, in the form of registration, created a particular set of conditions that increased scarcity and provided several instruments with which to transact. Essentially, registration is an official response to managing a tidal wave of urbanisation in a context of scarcity of formal and affordable land supply. It is a municipal strategy to register informal settlements in the city, as well as to register their inhabitants. This is therefore not registration in the sense of title deed registration, although it does confer some kind of tenure security as it indicates official acceptance of the settlements and their current occupants. Implicit in this approach to managing informal settlements is the idea that an intervention at a particular point in time indicates acceptance of the settlement at its current size at the point of intervention, acceptance of the occupants of sites and shacks as the time of registration and perhaps a promise of relocation or upgrading at some time in the future. Often community representative take the responsibility for making sure that the settlement does not grow in exchange for this acceptance. There is mostly also an expectation on the part of council that the current residents will not sell the rights that this registration has granted them (the right to be there and stay and the right to some kind of development in the future). Successive registrations of this nature, often without an obvious logic that can be understood by residents, creates the hope that development will come at some point. Registration also provides evidence of land rights (however precarious) which aids the operation of a local land market as buyers and sellers have written records on which to trade. Evidence of tenure is in the public plots, including B-forms, green shack numbers, receipts of payment, committee record of land holdings, and oral witnessing. In the private plots evidence is less formal, examples cited being verbal agreements and receipts of purchase containing stand numbers. Non-resident respondents cite written records only including the register which resides with council as Colloque international Les frontières de la question foncière At the frontier of land issues, Montpellier,

12 well as resident evidence such as A-forms, B-forms and green cards in respect of the transit camp. A council register is also evident in the private plots. It seems, in both cases, simply to be a list of names with ID numbers referenced to shack numbers. The lists / registers are also held by the community structures. In the private sites, there seems to be something of a dual system as mention is made of the existence of landowner and committee records. The registration narrative The first registration in the public site is said to have been conducted by council in 1997, at which time the council also gave residents IDs (A-forms and green cards), fenced the area and hired a security guard to control it. A-forms are settlement registration forms containing a dwelling number, name, and ID or passport number. Green cards are access cards to display to gate security. Original stand-holders have the A-forms and green cards of the first registration in 1997 and B-forms were issued in a subsequent registration to those who came later. A-forms appear to register an entitlement to in situ upgrading for the first group of occupants, while the B-forms are associated with relocation letters to the adjacent extensions 9 and 10 (Plots 52 and 70 with 402 and 121 sites respectively) for subsequent settlers. Both forms therefore represent evidence of a right to formalisation and upgrading, either in situ or via relocation. C-forms were issued to residents of the private plots by council in the 1997 registration too. At this time it appears that residents were also included on the region 5 housing waiting list. C-forms therefore appear to confirm recognition of the then current residents to be there as well as a place in the housing queue. The forms reference shack numbers to the names and ID number of the shack owner. Access to the space freed up by relocation, either vacant stand or backyard, appears to be a tradable commodity, with the community leadership allegedly playing a role in sales and possibly also in rentals in the private plots. Registration is a critical intervention because it creates insiders and outsiders: those whose access is officially accepted and those who are not; those who stand to gain greater access to secure tenure and development opportunities and those who do not qualify; those who have an interest in working with council and those whose interests are not recognised let alone served by council; those whose interests are represented by the committee and those who are not. Registration brings with it evidence of tenure as well as the stratification of rights (A, B and C forms, green cards and relocation letters leading ultimately to title deeds), and evidence of tenure creates an instrument for local land transactions. Land transactions: multiple methods and meanings Local registration co-exists with council registration. On one hand, council management via registration relies on the community structure keeping a record too. On the other hand, local rules and Colloque international Les frontières de la question foncière At the frontier of land issues, Montpellier,

13 procedures for sale and transfer, in the context of an apparently vibrant local market, have been developed and are applied and enforced, albeit unevenly. This is despite the official position that the various forms cannot be transferred. Local rules may mimic council rules in some respects, but they do not operate within what council considers to be legitimate bounds. For example, sale of sites and transfer of A-C forms is officially unacceptable but widely practiced and both appear to be locally acceptable. However, evidence seems to exist of local rules being broken too (such as leaders selling sites themselves). One factor in this respect may be accountability and oversight, or the lack thereof. Another may be the very real difficulty of organising, mobilising and retaining organisational coherence is a context where there is very little to offer. This would apply within both the transit camp in Zandspruit and to the private plots. The case study findings raise the question are leaders acting as agents or brokers or facilitators in the sale of sites and forms or are they acting out of individual interests to gain personally and materially from the transaction. This distinction is important as it demonstrates whether local rules are being developed, applied and adhered to or whether rules are being developed but ignored by leadership. Are local rule-makers being rule-breakers too? It also enables a reflection of whether the situation is simply less systematic than this, which does appear unlikely given the rules and procedures cited. In practice site sales and B-form transfers are taking place. To external parties this is perceived in negative terms illegal, informal, impossible in fact whereas in community terms these may simply be reflections to an outsider of local rules and procedures which are locally systematic and understood. Residents perceptions in both the transit camp and the private sites are that people can sell their sites with approval from the committee, unless they are renters. In the public sites, mention is also made of permission from the police being necessary, which seems to be a reference to the role of the police in verifying affidavits. Stakeholder responses distinguish between stands and shacks, indicating that while shacks may be sold, the sale of stands is not permitted. Neither is the sale of forms allowed. However, most respondents also recognise that sales do occur and that there is no official response. This activity seems to be linked to community leaders in the perception of some and it is labelled as corrupt. In the case of the private plots, there is recognition that the sale of stands is widespread, although unrecognised by the landowners and council. Conflict: Access to land and the urban economy Zandspruit experienced a great deal of conflict which involved vigilante groups violently driving out foreign nationals from the government-designated transit camp area. Although most people attribute the conflict to Zimbabweans at first, on closer interrogation crime seems to be an underlying factor, as well as lack of response on the part of the police to the occurrence of crime. A specific criminal Colloque international Les frontières de la question foncière At the frontier of land issues, Montpellier,

14 act, the murder of a South African by a Zimbabwean, was the proverbial last straw for residents in a context already beset with crime, widely attributed to Zimbabweans. In addition, the violent conflict was clearly linked to what residents perceived as lack of response to the murder by the police. The police defend themselves citing that the layout, overcrowding and lack of street lights make it difficult for them to police at night and describing Zandspruit as a dangerous place. However, as much as crime and policing (or the lack thereof) features prominently in perceptions about the conflict in Zandspruit, the research demonstrates also that access to resources, especially secure land tenure, is a significant factor in creating community division. The manner in which development intervention or formalisation responds to this potential for conflict is critical. In the current context, there seems to a fundamental contradiction between the management strategies of the state (evident in the ongoing council registration process and the current provincial registration programme, as well as the allpervasive queue jumping mindset) and conflict avoidance approaches. Official registration creates insiders and outsiders, qualifiers and non-qualifiers, recognised and unrecognised, accepted and unaccepted categories of people. It also co-opts leadership into taking responsibility for containing settlement growth on behalf of council. In some cases this buy-in or common interest is based on the quid pro quo of access to a formal development opportunity, which immediately puts the committee in the service of some, not all. The community organisation thus ends up reinforcing the insiders/outsiders division by taking on a job that no-one can perform effectively on ensuring that the settlement does not grow. This is not an isolated experience and, in the context of housing backlog and land hunger and the failure of both the formal market and the state, is a deeply cynical position for council to have adopted. In the context of housing demand unmet by formal supply, local solutions will be found by people in their efforts to secure access to the city. People will make use of whatever livelihood strategies they have at their disposal to survive, including land transactions. Part 3: Conclusion policy implications and research issues The research demonstrates how access to a good location and the scarcity of formal land supply for poor people combine to contribute to informal land access arrangements. It unpacks the nature of these arrangements, using land transactions, and develops the argument that a local land market exists, about which very little is known and understood. This emerging understanding is critical to policy intervention, if unintended consequences are to be avoided. Indeed, a larger body of evidence is needed than is currently available in the mainstream policy discourse. Assembling a body of evidence significant enough to enter this discourse is critical in the current policy environment due to the significance accorded to notions of assets, titling and the property market. Colloque international Les frontières de la question foncière At the frontier of land issues, Montpellier,

15 In this respect, some questions for further research include what the nature of the systems and procedures that govern local land markets is? What governs the behaviours of households engaging in these markets? What impact does the informal settlement management intervention of the local state have on the local market? Can and should the local market be integrated in some manner with the formal market? In addition, a much wider research agenda emerges from the policy analysis and case study presented here including: Trends in formal and informal backyard accommodation and the social arrangements between landlords and tenants City informal management strategies and approaches including the management of public spaces, governability in relation to secure and safe environments, the accessibility of transport to economic opportunities and the registration of households The provision of services to households with less than formal tenure arrangements Township residential property markets: besides investment in public infrastructure, the role of informal property brokers, whether public representatives or individuals has been largely ignored by the state in terms of SMME development. The understanding of their role in both formal and informal trading in property is underestimated. In conclusion, in the South African context rights and the market are seen as paramount to developing a viable development trajectory. However, a gap exists between the formal instruments of the market and the state and the local, or informal, practices in communities. This gap is an ongoing challenge as it contributes to the reproduction of the dual economy and inequity. However, the solutions are not to be found in the simplistic application of extravagant promises that capitalism can be made to work for the poor, when in fact there is already evidence that local markets do in fact operate and that titling is not a prerequisite to deliver access to credit and services for those who want them. Once the existence of local markets is known and better understood, then feasible and appropriate solutions to integration can be developed and tested. Colloque international Les frontières de la question foncière At the frontier of land issues, Montpellier,

16 References: Development Works, Conflict Management in Development Interventions in Informal Settlements: research report synthesis of results, findings, analysis and recommendations, report for Planact, March 2005, Johannesburg. LEAP, Perspectives on Land Tenure Security in Rural and Urban South Africa: An analysis of the tenure context and a problem statement for Leap. RSA, Breaking New Ground Human Settlement Plan: a comprehensive plan for sustainable human settlements, Pretoria. Colloque international Les frontières de la question foncière At the frontier of land issues, Montpellier,