42539 W O R L D B A N K W O R K I N G P A P E R N O . 1 3 0 In Search of Land and Housing in the New South Africa The Case of Ethembalethu Stephen Berrisford Dave DeGroot Michael Kihato Ntombini Marrengane Zimkhitha Mhlanga Rogier van den Brink THE WORLD BANK W O R L D B A N K W O R K I N G P A P E R N O . 1 3 0 In Search Of Land and Housing in the New South Africa The Case of Ethembalethu Stephen Berrisford Dave DeGroot Michael Kihato Ntombini Marrengane Zimkhitha Mhlanga Rogier van den Brink THE WORLD BANK Washington, D.C. Copyright © 2008 The International Bank for Reconstruction and Development/The World Bank 1818 H Street, N.W. Washington, D.C. 20433, U.S.A. All rights reserved Manufactured in the United States of America First Printing: January 2008 printed on recycled paper 1 2 3 4 5 10 09 08 07 World Bank Working Papers are published to communicate the results of the Bank's work to the development community with the least possible delay.The manuscript of this paper therefore has not been prepared in accordance with the procedures appropriate to formally-edited texts. Some sources cited in this paper may be informal documents that are not readily available. The findings,interpretations,and conclusions expressed herein are those of the author(s) and do not necessarily reflect the views of the International Bank for Reconstruction and Development/ The World Bank and its affiliated organizations, or those of the Executive Directors of The World Bank or the governments they represent. The World Bank does not guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of The World Bank of the legal status of any territory or the endorsement or acceptance of such boundaries. The material in this publication is copyrighted. Copying and/or transmitting portions or all of this work without permission may be a violation of applicable law. The International Bank for Reconstruction and Development/The World Bank encourages dissemination of its work and will normally grant permission promptly to reproduce portions of the work. For permission to photocopy or reprint any part of this work, please send a request with complete information to the Copyright Clearance Center, Inc., 222 Rosewood Drive, Danvers, MA 01923, USA, Tel: 978-750-8400, Fax: 978-750-4470, www.copyright.com. All other queries on rights and licenses, including subsidiary rights, should be addressed to the Office of the Publisher, The World Bank, 1818 H Street NW, Washington, DC 20433, USA, Fax: 202-522-2422, email: pubrights@worldbank.org. ISBN-13: 978-0-8213-7373-6 eISBN: 978-0-8213-7374-3 ISSN: 1726-5878 DOI: 10.1596/978-0-8213-7373-6 Library of Congress Cataloging-in-Publication Data has been requested Contents Abstract . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . v Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vii Acronyms and Abbreviations. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ix Executive Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xi 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 The Geography of Apartheid . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 The Peri-urban Areas . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 The Case Study. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 2. The Story of Ethembalethu: Our Hope . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 The Vision . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 Not In Our Backyard: Sued and Arrested . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 Enter the Government . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 Bogged Down in Environmental Impact Assessments. . . . . . . . . . . . . . . . . . . . . . . . . 9 3. Key Lessons Learned. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13 Overcoming Reluctance and Resistance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14 Making Land Use Pro-poor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16 Reforming the Land Market . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19 Realigning Planning Processes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20 Designing a Program for Peri-urban Areas . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24 Re-engineering Program Implementation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26 Freeing Up and Building Capacity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29 4. Recommendations and Suggestions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31 5. Postscript . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35 Appendix: Development Facilitation Act, 67 of 1995 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37 iii iv Contents LIST OF TABLES 1. Legislative Compliance. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22 2. Time Lines for Various Applications for Township Establishment: Best-case Scenario . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23 LIST OF FIGURES 1. Apartheid Spatial Planning for a City . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 2. Map Showing Mogale City (formerly Krugersdorp) . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 Abstract his case study analyzes the difficulties a poor community experienced in accessing peri- T urban land in South Africa.A decade ago, this community, composed largely of laid-off farm workers, wanted to buy their own farm in a peri-urban area west of Johannesburg to establish a mixed-use settlement. The name of the village would be Ethembalethu--"Our Hope"--and the about 250 families started their own savings scheme to make their dream a reality. By 1997, they had saved enough money (R125,000 or about US$18,000) to make their first purchase offer. Now, a decade later, the community's dream has still not become reality, due to numerous obstacles, including three canceled sale agreements, wrongful arrest, being sued in court, an out-of-court settlement for which community members were paid R250,000 to not move into the white neighborhood, and large sums of their own money spent on consultants and environmental impact studies.While the community now has at least a confirmed right to eventually occupy the land in terms of an agreement with Mogale City Municipality, it does not yet legally own the land, and is still trying to get permission to build on and work the land. Millions of black South Africans live in the peri-urban areas. However, government programs, development planning, and environmental requirements, and the current land and housing markets do not allow them realize their aspirations. Based on this case study, we suggest the following areas for policy and program reform: (i) overcoming reluctance and resistance by municipalities and prospective neighbors to low-income settlements; (ii) making land use planning in municipalities explicitly pro-poor; (iii) restructuring the land market; (iv) realigning planning processes; (v) designing a land and housing program targeted to peri-urban areas; (vi) re-engineering program implementations; and (vii) freeing up and building capacity. The study proposes the establishment of a national task force to ensure appropriate followup. v Acknowledgments his study was undertaken at the request of Mogale City for World Bank technical assis- T tance on the design and implementation of integrated housing and agriculture projects. The team was composed of Stephen Berrisford and Michael Kihato (consultants), Zimkhitha Mhlanga (Deputy Director Rural Development, Mogale City), Ntombini Marrengane and Dave DeGroot (Urban Development, Southern Africa, Africa Region, World Bank), and Rogier van den Brink (Country Economist, Poverty Reduction and Eco- nomic Management, Southern Africa, Africa Region, World Bank). Peer Reviewers were Hans Binswanger (consultant) and Kate Kuper (Urban Develop- ment, Southern Africa, Africa Region, World Bank). The responsible managers were Ritva Reinikka (Country Director),Glen Thomas (Director-General LandAffairs),Jaime Biderman (Urban Development, Southern Africa, Africa Region, World Bank), Emmanuel Akpa (Poverty Reduction and Economic Management, Southern Africa, Africa Region, World Bank) and Frank Byamugisha (Environmentally and Socially Sustainable Development, Southern Africa, Africa Region, World Bank). A draft report was jointly reviewed by the Department of Land Affairs and the World Bank. Comments were received from Mzwakhe Ndlela (Director Department of Land Affairs Gauteng), Vusi Radebe (Director, Housing and Land, Mogale Local Municipality), George Phiri (Urban and Rural Development, Mogale City Local Municipality, Molefi Sebilo (Chairman, Muldersdrift Home Trust Foun- dation), Thabo Rabapane (Secretary, Muldersdrift Home Trust Foundation), Michael Worsnip (Cradle of Humankind Management Authority,Gauteng Provincial Government), Shamilla Chettiar (Cradle of Humankind Management Authority, Gauteng Provincial Government), Fredah Moatshe (Gauteng Department of Agriculture, Conservation and Environment), and Sheila Hughes (Senior Manager for Intergovernmental Relations, Department of Provincial and Local Government). The team which conducted this study would like to thank all stakeholders involved for their participation and the opportunity it accorded all of us to learn from this extraordinary story.We will continue to update the public about subsequent developments on the following website: www.worldbank.org/southafrica vii Acronyms and Abbreviations BCDA Black Communities Development Act DFA Development Facilitation Act DLA Department of Land Affairs DOH Department of Housing DPLG Department of Provincial and Local Government ECA Environmental Conservation Act EIA Environmental impact assessment GDOH Gauteng Department of Housing GDACE Gauteng Department of Agriculture, Conservation and Environment IDP Integrated Development Plan LAA Land availability agreement LeFTEA The Less Formal Township Establishment Act LRAD Land Redistribution for Agricultural Development MEC Member of the executive council MHTF Muldersdrift Home Trust Foundation NDOH National Department of Housing NEMA National Environment Management Act NGO Non-governmental organization NIMBY "Not In My Back Yard" (a term used to characterize resistance by wealthy homeowners to the establishment of low-income housing in their area) NSDP National Spatial Development Perspective RDP Reconstruction and Development Program SAMDI South African Management Development Institute SLAG Settlement and Land Acquisition Grant VIP Ventilated improved pit-latrine ix Executive Summary hat are the difficulties the poor face in accessing peri-urban land in South Africa? To W answer this question, this case study records and analyzes the experience of a com- munity in the Muldersdrift area of Mogale City Municipality in Gauteng: the Muldersdrift Home Trust Foundation (MHTF). A decade ago, the members of this community, composed largely of laid-off farm workers, wanted to buy their own farm in a peri-urban area west of Johannesburg to estab- lish a mixed-use settlement. The name of the village would be Ethembalethu,"Our Hope." Beginning with about 250 families--each of which saved and contributed R50 per year, later increased to R100 per month--the association aimed to acquire sufficient land in the area to build its own homes. By the end of 1997, the association was incorporated as a "section 21" (not-for-profit company): The MHTF. It had saved about R125,000 and made its first purchase offer. Now, a decade later, the community's dream has still not become reality. This follows numerous obstacles, including three canceled sale agreements, wrongful arrest, being sued in court, an out-of-court settlement for which community members were paid R250,000 to not move into the white neighborhood, and large sums of their own money spent on consult- ants and environmental impact studies.While the community now has at least a confirmed right to eventually occupy the land in terms of an agreement with Mogale City Municipal- ity, it does not yet legally own the land, and is still trying to get permission to build on and work the land. The peri-urban areas are formerly "rural" localities that are now, due to the rapid expansion of South Africa's metros and major towns, directly in the path of urbanization. They lie officially outside of the "urban edge." In the land market in the peri-urban areas, the rich and the poor compete directly with each other, because both prefer to live close to where they work. The preference of the rich is to live in gated housing communities, cre- ated by the redevelopment of farms. The preference of the poor is to live in mixed-use settlements, where they can establish modest houses, raise their children in safety, benefit from having relatively close access to urban schools and health facilities, as well as work opportunities, while having space to venture into farming and small business activities should such opportunities arise. Millions of black1 South Africans live in the peri-urban areas. However, even if they have the financial means to realize their aspirations, as the Ethembalethu community has, government programs, development planning and environmental requirements, and the current land and housing markets do not allow them realize their aspirations. The methodology followed in compiling this case study since early 2005 has been one of direct involvement with the MHTF. We have participated in project task team meetings, as well as in processes such as the negotiation of agreements between the MHTF and offi- cials of the Mogale City Municipality. We have assisted MHTF in its efforts and provided resource materials to the various stakeholders. We have been impressed by the profession- alism and dedication of the various officials involved. This report should in no way be seen 1. The term "black" in this case study refers to South Africans who are African, Indian, or Colored (mixed race). xi xii Executive Summary as an indictment of their hard work. It makes quite a different point: The capacity of these officials to deliver is seriously undermined, not by a lack of training or education, but by the highly complicated and fragmented framework within which they operate. Bureaucracy is exhausting the capacity of communities and local governments to ensure that low-income South Africans of all backgrounds can acquire and develop land and shelter in South Africa's peri-urban areas. Based on this case study, we suggest the following areas for policy and program reform: Overcomingreluctanceandresistancebymunicipalitiesandprospectiveneighbors to low-income settlements; Makinglanduseplanninginmunicipalitiesexplicitlypro-poor; Restructuringthelandmarket; Realigningplanningprocesses; Designingalandandhousingprogramtargetedtoperi-urbanareas; Re-engineeringprogramimplementations;and Freeingupandbuildingcapacity. Below, we briefly discuss each area of intervention, highlighting key issues and conse- quences that arise from such shortcomings. We also make recommendations and identify the main actors who should be responsible for implementing the suggested policies. The various stakeholders now need to discuss these recommendations. Overcoming Reluctance and Resistance Municipalities can be reluctant to provide land and housing to low-income groups because of fears of non-payment of services; loss of income foregone from"high-end"land use, and the unaffordably high costs of complying with municipalities'interpretations of the standard for basic services, as defined in the Constitution. Wealthier residents, most of whom are white, are reluctant to see low-income settlements in their neighborhood, because of the effect on real estate prices and fears of crime and environmental degradation.We recommend that a special task force initiates an advocacy and public education campaign to allay these fears. It would do this by stressing the benefits of undoing the geography of apartheid, com- bating the culture of non-payments, demonstrating the revenue potential of low-income settlements and providing better guidance for the interpretation of constitutional rights to basic services. This task force would presumably be led by the Presidency, which is the coor- dinating agency for the National Spatial Development Perspective. Making Land Use Planning Pro-poor The main problem faced by groups like MHTF is the almost complete lack of land avail- able to low-income groups to combine housing needs and agricultural activities. One factor is that the so-called "Urban Edge" policy zones land as purely urban or purely agri- cultural, disallowing for a transition zone of mixed land use typically found in peri-urban areas elsewhere in the world, but almost entirely absent in South Africa. This policy should be reviewed and municipalities encouraged, as part of their Integrated Development Plans (IDPs), to designate and acquire land for peri-urban, mixed-use settlements. This would Executive Summary xiii allow poor people to combine agriculture and housing near their working places. The task force, with guidance from DPLG, should develop IDP models and guidelines to guide municipalities in this. Restructuring the Land Market The key problems are the unavailability of small parcels in the peri-urban land market, and the lack of incentives to dispose of under-used land. The reasons for this can be found in the difficulties in subdividing land and the regressive regime of land taxation in munici- palities such as Mogale City. To address these issues, government should finally implement the 1998 repeal of the Subdivision of Agricultural Land Act (1970) (Subdivision Act) and, led by the Department of Provincial and Local Government (DPLG), develop guidelines for municipal property rates, which, at a minimum, should avoid the reinstatement of the highly regressive rates of 1939. Unifying Planning Processes At present,two parallel processes are required to plan and implement housing development: Township Development and Environmental Impact Assessment (EIA). This wasteful dupli- cation stifles the capacity of all actors and constrains the ability of poor people to acquire and develop land. There is considerable lack of clarity on how communities must go about establishing a township. In addition, the EIA involves too many steps, which leads to a lengthy process and discourages decision-making by officials. Merging these different legal procedures for all projects involving land and housing development in peri-urban areas into a single process is highly desirable, and would streamline and decentralize the approval of the EIA process. Designing a Program for Peri-urban Areas Currently, there is no program for mixed land use and housing for poor people who want to engage in "multiple livelihoods". And the standards set for infrastructure and housing are often inappropriate. Constitutional guarantees for basic service delivery are interpreted to imply overly expensive infrastructure, which pushes low-income housing projects to remote locations, where land is cheap, to bring total costs down. Moreover, poor people are not able to use housing subsidies to build their own houses. Our recommendations include developing an integrated program that helps poor people acquire land, housing, and agri- cultural and other business support and finance. This would allow more affordable stan- dards to promote mixed-use settlements in locations closer to work and commerce, and simplify housing standards, while including an option for subsidized own-construction of houses by the poor. Re-engineering Program Implementations Under the current setup, no single agency bears overall responsibility for the planning of land development and use. This uncertainty leads to lengthy and frustrating processes of land acquisition and development by municipalities and prospective beneficiaries. xiv Executive Summary Decision-making is too centralized, which, for example, leaves the expertise of the organi- zation's lower levels underused. In addition, money flows into sectoral and program"silos," making planning and financing an integrated village community an almost impossible task for municipalities and the poor. The intergovernmental relations unit of DPLG, together with the task force mentioned above, should clarify roles and responsibilities of different departments and levels of government and decentralize decision-making to the local level, following the subsidiarity2 principle.We also recommend that the National Treasury provide incentives for agencies to unify financing and access requirements. Freeing Up and Building Capacity The main constraint on building capacity seems to be the complexity of requirements. The uncertainty created by this complexity reduces the willingness and ability of officials to take action. In addition, there is a specific lack of legal and implementation capacity among public-sector lawyers and project managers. Inadequate involvement of non-governmen- tal organizations (NGOs) and other private-sector providers to support community groups and municipalities further aggravates the capacity constraint. The development of clear guidelines and manuals is, therefore, urgently required. We also recommend that DPLG, with support from the South African Management Development Institute (SAMDI) and other relevant departments, design and implement capacity-building programs for public- sector lawyers and program managers. Municipalities should outsource more support functions to NGOs and other providers, and focus on monitoring activities. Finally, bene- ficiaries should get project preparation grants large enough to hire NGO and other expert support. 2. The principle of devolving decisions to the lowest practical level. CHAPTER 1 Introduction W hat are the difficulties the poor face in accessing urban and peri-urban land in South Africa? To answer this question, this case study records and analyzes the experience of a community in the Muldersdrift area of the Mogale City Municipality in Gauteng Province: the Muldersdrift Home Trust Foundation (MHTF). A decade ago, the members of this community, composed largely of laid-off farm workers, wanted to buy their own farm in a peri-urban area west of Johannesburg to establish a mixed-use settlement. The name of the village would be Ethembalethu: "Our Hope." Now, a decade later, their dream is still not a reality. This follows three canceled sale agreements, the wrongful arrest of nearly 50 community members, a lawsuit against the community, one out-of-court settlement and two formal objections lodged by neigh- bors. In addition, the community has spent a great deal of money on consultants and on two environmental impact studies. While community members now have access to the land, they are not yet the legal owners of the land and are still trying to get permission to build their own houses. The Geography of Apartheid One of the legacies of apartheid is the duality of living spaces. Apartheid gave physical expression to the legal and socioeconomic discrimination that was known as "separate development". During the period of formal apartheid the government legislated where people could live according to their race. Africans, Indians and people of mixed race ("Coloreds") were consigned to designated locations separate from (and invariably inferior to) the "white" residential areas. Under this system black families were to live in rural 1 2 World Bank Working Paper Figure 1. Apartheid Spatial Planning for a City A2 M-H H M M IXIN G M G IN MIX L A1 A2 L A2 A2 A1 White C.B.D. C.B.D. Frame Major road routes Indian C.B.D. Industrial RESIDENTIALAREAS homelands or "satellite" townships, while sending their adults to work in the cities, in the mines and on the farms. Whites lived in low-density neighborhoods close to work, right around the city center and in its immediate suburbs. A "buffer zone" of 15 to 40 km, with very low-intensity land use, further separated the races. The geographic legacy of apartheid persists today. Compared to other cities in the world, South African cities exhibit low population densities where one would expect them to be high (in and around the city centers) and high densities where one would expect them to be low (in and beyond the peri-urban areas). Excessive expenditure by blacks on transportation (often around 40 percent of wages) is the most obvious cost of this geography of apartheid. Fragmented families entail high social costs, including high crime and HIV and AIDS rates, all costs that can be partly linked to parents and children not living together. Spouses may live or work elsewhere, either in other townships, on the property of white households that employ them, or in the rural homelands. This duality of living spaces did not emerge spontaneously. It was the result of an elaborate network of laws, zoning regulations, tax incentives, plot subdivision restrictions, and targeted government programs, which systematically sought to make it impossible for people to mix outside the confines of strictly defined relationships. After 1994, the first democratically elected government started to dismantle this system and to provide land and housing for the poor. Enormous progress has been made in dismantling the complex maze of apartheid laws and regulations, but, as this case study shows, there is still work to be done. Similarly, about 1.8 million free houses for low-income people were built by In Search Of Land and Housing in the New South Africa 3 government contractors. However, most of these houses were built in the existing segregated settlements, far away from work and decent public services, which reinforced the spatial patterns of apartheid despite concerted efforts at every level to redress its effect. The Peri-urban Areas At the same time, after the removal of the apartheid controls on where blacks were allowed to live, many of them desperately tried to find places to live closer to their work and good schools for their children. As a result, South Africa has experienced a massive, unplanned, spontaneous resettlement program in the last decade. Middle- and high-income blacks seem to have done relatively well by using the existing and well-developed private markets for land and housing. However, the poor have fared far worse. The mushrooming of informal settlements has been the most visible expression of the scale of their demands and the failure of government programs and the private markets to meet them.3 The peri-urban areas are formerly "rural" localities that are now, due to the rapid expansion of South Africa's metros and major towns,4 directly in the path of urbanization. They lie officially outside the urban edge. In the peri-urban land market, rich and poor compete directly with each other, because both prefer to live close to where they work. The rich prefer to live in gated housing communities, created by the redevelopment of farms. The poor prefer to live in mixed-use settlements, essentially villages, where they can estab- lish modest houses and raise their children in safety. Here they can benefit from having relatively close access to urban work opportunities, schools and health facilities while being able to also venture into farming and small business activities. Millions of black South Africans live in these areas; but, even if they have the financial means to do so, as the Ethembalethu community has, government programs, development planning and environmental requirements, and the current land and housing markets do not allow them to realize their aspirations. The peri-urban black population in South Africa is composed largely of current and former farm workers and workers in allied service industries. This group, as the term peri-urban suggests, is neither fully rural nor urban in orientation and aspirations. A central aspiration of much of the peri-urban population-- of which the MHTF membership is an excellent example--is to become "small holders," defined as owning a serviced plot that is large enough to support small-scale gardening and animal husbandry. In practical terms, smallholdings may be plots of about 1,000 square meters in size. As urbanization proceeds, this type of smallholding in peri-urban areas is a real asset that will increase in value dramatically over the medium term. The Case Study This study describes the saga of how one community, instead of squatting in an informal settlement or invading a new plot of land, attempted to buy land legally and build their own houses in a peri-urban area, using their own savings. The Ethembalethu story draws attention to the many challenges that poor people face accessing land and housing in South Africa. 3. The National Department of Housing (NDOH) estimates that 2.5 million households reside in informal settlements in SA. 4. South African Cities Network, State of the Cities Report 2004, p. 37. 4 World Bank Working Paper The purpose of this case study is to highlight the complex challenges that face poor communities that attempt to secure their constitutionally mandated rights to adequate housing. These challenges stem from inconsistent or inadequate policies and legislation, confusion between the myriad agencies involved, lack of clarity over responsibilities and accountability, lack of capacity of the implementing agencies, the rising costs and delays in accessing building materials, corruption, and the absence of information and training of both government officials and the housing hopefuls. Finally, the case highlights the Not- In-My-Back-Yard (NIMBY) ferocity with which wealthier citizens may still resist change in the new South Africa. This report draws practical lessons from the case study and makes suggestions for reforms. South Africa's implementation systems are extremely complex and laden with transaction costs, and also, in certain respects, explicitly biased against the poor. This is particularly the case with regard to the subdivision of land, land taxation and, as this case study shows, the discriminatory application of the law against poor black people. One example is that the requirements for EIAs are seldom imposed on rich white people. The methodology followed in compiling the case study has been one of direct involve- ment in the project, since early 2005. We have participated in project task team meetings as well as in processes such as the negotiation of agreements between the MHTF and local officials within the jurisdiction of Mogale City Local Municipality. We have supported MHTF in drafting correspondence and provided resource materials to the various stake- holders. We have been impressed by the professionalism and dedication displayed by the various officials involved: in no way should this report be seen as an indictment of their hard work. This report makes quite a different point: the capacity of these officials to deliver is seriously undermined, not mainly by a lack of training or education, but by the highly complicated and fragmented framework within which they operate. Bureaucracy is exhausting the capacity of communities and local governments to ensure that low-income South Africans of all backgrounds can acquire and develop land and shelter in South Africa's peri-urban areas. This report is organized as follows. We begin with a narrative of events, starting in 1996 and ending in 2006. We draw the key lessons emerging from this experience. We then investigate the main issues, and suggest reforms and improvements in the following areas: (i) improving access to land; (ii) simplifying and aligning legal procedures; (iii) designing a land and housing program for peri-urban areas; and (iv) reforming the land market. We conclude with a call for further participatory action research. CHAPTER 2 The Story of Ethembalethu: Our Hope he story of the Ethembalethu community--now represented by the Muldersdrift T Home Trust Foundation--begins more than 50 years ago, with the arrival of the Mphale family in the Muldersdrift area. Muldersdrift was then a white farming community west of Johannesburg. In the intervening years, the area has undergone rapid change and now lies squarely in the path of commercially-driven residential and commercial land development pushing outward from Johannesburg. Given its agricultural history, the area has a significant number of white smallholdings with farm workers and occupants who want to obtain more secure tenure and housing. The Mphale family lived, in terms of an agreement with a white farmer, on the land surrounding and including the farm known as Portion 78 Driefontein. This farm plays a central role in the story covered in this case study. When the farmer sold his land in 1973 to another white person--as only whites had land-ownership privileges outside designated "black spots"--he gave Mr. Mphale written permission to remain on the land. The new owner, however, changed the basis on which the Mphales could remain, making it more of an informal labor tenancy arrangement rather than the indefinite right to remain on the land implied by the earlier arrangement.5 5. Labor tenancy is an arrangement under which the owner allows occupants to stay on the farm and either graze their livestock or cultivate land in return for a certain number of days' work. It is wide- spread in certain parts of South Africa. Labor tenants are protected from eviction by the Land Reform (Labor Tenants) Act of 1996. The definition of a labor tenant in that act is, however, strict and it is unlikely that the Mphales would have qualified for such protection. In any event this protection only came into effect in 1996. 5 6 World Bank Working Paper Figure 2. Map Showing Mogale City (formerly Krugersdorp) The Vision In the mid-1990s, Mr. Molefi Selibo, a community leader and long-time resident in the same area was motivated by the widespread tenure insecurity and poor living conditions of farm workers and farm dwellers in the Muldersdrift area, and began to organize fam- ilies into a housing association. Beginning with about 250 families, the association started a savings scheme aimed at acquiring land in the area upon which to build their own homes and carry out small-scale farming. Their vision was clear: a village with houses, some individual farming fields, a grazing area and some common areas for recre- ation. The painting done by them and reproduced on the cover of this report illustrates their dream. Beginning with around 250 families--each of whom saved and contributed R50 per year, later increased to R100 per month--the association aimed to acquire sufficient land in the area to build their homes. By the end of 1997, the association was incorporated as a "section 21," or not-for-profit company, the MHTF. In Search Of Land and Housing in the New South Africa 7 Not In Our Backyard: Sued and Arrested In 1997 the association made its first offer to purchase land, which was successful.6 In March 1997 the MHTF entered into a purchase agreement with the owner of Portion 120 Rietfontein, a smallholding of 8.5 hectares in Muldersdrift. The deed of sale was also concluded. However, during the course of negotiations with the seller on the method of payment, the sale was suddenly canceled. A group of neighboring landowners had made a successful counter-offer to purchase the land, which began MHTF's long battle with the NIMBY syndrome which plagues the spatial distribution of human settlements throughout South Africa. The next year the MHTF made an offer on a similarly sized piece of land, Portion 77 Rietfontein, adjacent to the first plot. This offer was also initially accepted and the MHTF entered into a purchase agreement. However, the MHTF was unable to pay the full purchase price immediately and sought help from a commercial bank for a loan to cover the balance. Despite the organization's proven saving record, they could not get a top-up loan from the bank to cover the purchase price over and above their accumulated saving. Fortunately, the seller agreed to an installment sale for the balance of the purchase price: MHTF was to pay him a deposit of R50,000, with the balance of R104,000 to be paid in 15 monthly install- ments of roughly R7,000. Aware of the need for technical advice on how to establish their village, the MHTF used its other savings to procure the services of engineers, geo-technical consultants and a planner to prepare their application for planning approval for a combined residential and agricultural production project. In particular, the MHTF had to hire engineers and other consultants to prepare the application for "township establishment."7 They also submitted an application for housing subsidies to the Gauteng Provincial Department of Housing's (GDOH) People's Housing Process (PHP) program,8 which was approved in 1998 for 263 households. The MHTF, in fact, did not want to establish a typical low-income "township," but rather a low-density "village" where they could build their own houses, live in uncrowded conditions, have some common spaces for recreation and grazing of animals, and have the opportunity for some households to engage in small-scale farming. However, because this form of settlement is not legally recognized and the national housing subsidies are only available to formally established "townships," the MHTF had to plan for a "township." In late 1998, with 82 percent of the purchase price paid, the MHTF received a letter from the landowner canceling the second sale agreement and attempting to return the sum of R126,000, which had already been paid by the MHTF towards the R154,000 purchase price. It was clear the same neighbors who had thwarted the previous purchase 6. Section 21 of the Companies Act (1973) provides for not-for-profit companies to be established. 7. Township establishment is the process in which undeveloped land is prepared for urban develop- ment through surveying, registration of individual sites and basic town planning. It is regulated by a host of different laws and is a legal precondition for the supply of municipal services: a municipality may not extend municipal services to a settlement that has not been proclaimed as a township. 8. The People's Housing Process is a variant of the national capital subsidy scheme for providing low- income housing and its distinguishing feature is that it relies on the individual beneficiaries to carry out much of the construction work themselves. 8 World Bank Working Paper had influenced this seller too. The MHTF now had to engage lawyers to protect their con- tractual rights to the land. This proved to be a wise decision, as in March 1999, the landowner applied to the High Court to have the sale canceled on the basis of alleged "mis- representation" by the MHTF. Trespassing charges were laid against the members of the MHTF as well and, on March 5, 1999, 47 MHTF members were arrested. Eventually, the South Africa Police Service acknowledged that these arrests had been wrongful and released the MHTF members. Two years after the landowner's letter, in November 2000, the matter of the contract came before the High Court. It became clear that the landowner had no legal basis upon which to cancel the sale and he, together with the representative of the white neighbors who had preempted the first purchase attempt by MHTF, proposed an out-of-court settlement. After some negotiations the MHTF accepted a settlement in terms of which the sale agree- ment would be canceled and the white neighbors would put R270,000 into a trust fund, with a representative of their choice and the chairman of the MHTF as the trustees. In addi- tion to this, the landowner refunded the R126,000 already paid by the MHTF towards the purchase price. A condition of the contract was that the MHTF could only use the R270,000 in the trust fund for the purchase of alternative land for housing, provided that this land did not fall within a prescribed circumference around the designated land of the white neighbors who had contributed to the fund, as depicted on a diagram appended to the court papers. In other words, the community would be paid R270,000 for agreeing not to acquire land in the area. Enter the Government . . . Bruised but not beaten, the MHTF again looked for alternative land. In 2001 the MHTF, still intact despite the difficult experiences of the past five years, made an offer on a much bigger piece of land, further away from the racially hostile environment where their previ- ous purchase agreements had been frustrated, but still in the Muldersdrift area.9 In 2001 the MHTF entered into its third purchase agreement: for Portion 78 Driefontein, (roughly 31 hectares) for a price of R650,000. At the same time, GDOH agreed in principle to transfer the MHTF's 1998 successful application for PHP capital subsidies to this land. By now, the MHTF had further refined their vision of how they wanted to live. They decided to build an eco-village to be called Ethembalethu.10 However, the land they now made an offer for was three times the size of the previous two farms and the purchase price of R650,000 was beyond their means. After paying for consultants and lawyers the MHTF only had about R400,000 available, R270,000 of which was tied up in the trust established after the settlement of their court case. They appealed to the GDOH for advice and the matter was referred to the Gauteng Provincial Office of the National Department of Land Affairs (DLA). During the course of negotiations between DLA, GDOH and Mogale City, the white neighbors of this new site got wind of the proposed project and, as had the neighbors surrounding the previous farms, began to raise objections. The spurious basis of their 9. The remainder of portion 78 of the farm Driefontein, 179 IQ Krugersdorp; which is a farm of just over 30 hectares. 10. Ethembalethu means "our hope." In Search Of Land and Housing in the New South Africa 9 objections was that an environmental impact assessment (EIA) had not been completed. This was irrelevant as, under South African law, an EIA is not required for the mere act of purchasing land.11 Nevertheless, the officials of DLA and GDOH seem to have been alarmed by these objections and notified Mogale City that they would not be contributing towards the purchase of the land after all. After protracted correspondence between the three spheres of government, the DLA finally agreed to purchase the land. The full price of R650,000 was paid by DLA to Mogale City, which then acquired the land in the name of the municipality in March 2002. DLA was, however, unable to single out any one of its available programs to suit this project: the establishment of a village. So the DLA officials devised a hybrid approach, which combined two existing programs. These were the settlement and land acquisition grant (SLAG) and the Commonage Program.12 Under this hybrid, DLA bought the prop- erty and transferred the land to the municipality. Unfortunately, there was no contract or agreement concluded between the DLA and Mogale City as to how the land should be used once it had been transferred to Mogale City. Neither has it been possible to obtain the project file from DLA, which would shed more light on the precise nature of the Department's approach to either the question of which program's funds were used to purchase the land or the terms agreed upon by DLA and Mogale City. The understanding was that it should be used to accommodate the MHTF community, and the community's claim and rights to this land should be confirmed in a "land availability agreement" (LAA)13 signed between the MHTF and the municipality. Such an agreement was eventually concluded between the MHTF and Mogale City. In terms of this agreement individual households will obtain title to their individual residential plots. Business plots would be released on a case-by-case basis in terms of business plans to be submitted to Mogale City. Bogged Down in Environmental Impact Assessments Once the land purchase was final the MHTF had to obtain two separate legal permissions through two separate legal processes to develop the land. First, a "scoping" EIA needed to be carried out in terms of the Environment Conservation Act (ECA) (1989). Second, a DevelopmentFacilitationApplication(DFA)neededtobewrittentotheGautengDevelopment 11. Under the Environment Conservation Act, 73 of 1989, EIA approval has to be obtained from the provincial government's department responsible for environmental management, in this case the Gauteng Department for Agriculture, Conservation & Environment, where a specified activity or land use change is envisaged. Simply changing ownership of land is not one of these specified activities, and so is not on its own a "trigger" requiring an EIA process. Should the owner of land wish to carry out one of these activities, then an EIA would have to be done, but this requirement is separate from the question of land ownership. 12. When a final draft of this report was presented to the National Director General of Land Affairs in early 2007 he said that there would not have been any reason why the SLAG could not have been used alone to cover the costs of both land acquisition and housing development. He could not understand why the departmental officials at the time did not make that choice. One possible reason is that the SLAG can only be used where the department has compiled a comprehensive list of potential beneficiaries and this had not been done in relation to the MHTF members by the department. 13. A land availability agreement (LAA) is a comprehensive agreement or contract concluded between an organ of state, typically, and a group of beneficiaries or, sometimes, a developer, in terms of which a specified land parcel is secured for the benefit of the beneficiaries or developer while the ownership resides with the organ of state. Normally, the LAA will include a process by which the ownership of the land will pass from the organ of state to the other party once a specified set of criteria have been met. 10 World Bank Working Paper Tribunal to obtain planning permission, in terms of the Development Facilitation Act (1995) (see the Appendix). These two processes were to run simultaneously. The reason for this is that there are two separate legal requirements. First, the ECA requires an "authorization" by the relevant environmental authority (in this case the provincial Department of Agriculture, Conservation & Environment) if certain specified "land use changes" occur.14 Second, municipalities are unable to extend municipal services to an area that is not a "proclaimed township." This can be achieved either through the DFA processes (an application to the provincial Development Tribunal), or through the old Transvaal Provincial Ordinance for Town Planning & Township Establishment, 15 of 1986 (which entails an application directly to the municipality). The scoping EIA which was conducted during 2002 revealed more robust opposition to the proposed development from the neighbors. The EIA scoping report submitted to the Gauteng Department of Agriculture, Conservation & Environment (GDACE) was formally accepted by them in 2003. However, department officials indicated that they were not pre- pared to support the project as it stood then, and required that a full EIA be done in addition to the completed scoping report. The department also stipulated that a number of specific aspects be dealt with in the full EIA, primarily through additional specialist studies. Where a housing development takes place within an already developed urban area an EIA is seldom required, as there is no "change of land use," but it is a common requirement wherever "greenfields" housing development is undertaken, especially outside already built- up areas.15 In all cases the cost of the EIA has to be carried by the applicant. A portion of the housing capital subsidy is earmarked to cover these costs, up to a predetermined ceiling. In May 2003, the DFA application for planning approval of the proposed project was submitted to the Gauteng Development Tribunal. At the tribunal hearing in October 2003, the application had to be withdrawn due to various procedural and technical flaws. There was no evidence that the tribunal members were unhappy with its content. Nevertheless, this represented a substantial setback to the project. Following this withdrawal of the DFA application, the project's momentum dwindled because of lack of funding for the various studies required: a full EIA and a resubmission of the DFA application for planning permission. Worse, the environmental and planning consultants who had done the bulk of the preparatory work for the EIA and DFA processes could not be paid by the Gauteng Department of Housing, because, legally speaking, after the withdrawal of the DFA application, the MHTF housing project was no longer a "hous- ing project." Moreover, the GDOH said they would not be able to make these payments until there was a proper LAA signed between the Mogale City Municipality (the landowner) and the MHTF. This agreement was only signed in mid-2005. Subsequently, the project has enjoyed renewed support from the GDOH, presumably on the grounds that the provincial government was satisfied that the land rights of the intended beneficiaries were now secured. The consultants were paid later in 2005 for their work on the earlier applications and reappointed to resume work on new applications. 14. It is important to note that the ECA's EIA regulations have been repealed with effect from July 1 2006, to be replaced by new EIA regulations in terms of the National Environmental Management Act, 107 of 1998, and the requirements for land use change EIAs has fallen away. 15. As indicated above, the requirements for an EIA have changed but this reflects the position at the time this issue was decided. In Search Of Land and Housing in the New South Africa 11 In the second half of 2005, the preparation of the EIA and DFA applications restarted in earnest. In August 2005, it became clear that provided both processes moved without interruption, they would probably only be completed by about September 2006. Then, the GDOH raised concerns about the budget and fees for consultants, despite having accepted these projections two months earlier. The percentage of the total allocation from the NDOH set aside to cover these professional fees had been nearly exhausted in the earlier applications. There were, as a result, insufficient funds left to cover the costs of the full EIA report and the new DFA application. These developments caused concern with the municipal officials and consultants working on the project. The consultants said that they would not be prepared to continue with the project if they could not bill for the amounts of time required to complete the two processes, and the municipal officials felt that the process of appointing new consultants-- at lower fees--would both delay and compromise the quality of the two applications. This impasse was eventually broken and the application for an environmental autho- rization was submitted in 2006. However, it was rejected by the relevant department (see paragraph below), and the DFA application is still to be heard by the Development Tribunal in late August/early September 2007. Another unexpected obstacle that arose in early 2007, however, was the issue of the proposed buffer zone around the Cradle of Humankind World Heritage Site. The World Heritage Site was confirmed by UNESCO in 1999 and the Gauteng provincial government was designated as the Management Authority in terms of the World Heritage Convention Act (1999). Subsequently, the Gauteng gov- ernment embarked on a master planning exercise for the World Heritage Site. This resulted in the demarcation of a proposed buffer zone around the designated World Heritage Site, in which "township establishment" would be prohibited. The buffer zone does not yet have any legal status, although the Gauteng government is seeking to change this. Now that the two outstanding authorizations (the EIA and the DFA applications) both must be decided by organs of the Gauteng government, the project may still be denied the necessary per- missions to realize the hopes and aspirations of the MHTF community. The final EIA submitted to the provincial Department of Agriculture, Conservation & Environment in 2006 was rejected by them in 2007, and they requested yet another set of reports and studies. An exasperated community and their environmental consultant have now pursued political channels and have set up a meeting with the provincial minister for Agriculture, Conservation & Environment. CHAPTER 3 Key Lessons Learned olicies and programs for land reform and housing delivery in South Africa tend to P have either an urban or a rural focus. The peri-urban areas have largely slipped through the cracks.16 The inability of people to implement land reform and housing projects in these areas, as the Ethembalethu case shows, is a result of this policy and program gap. In peri-urban areas, housing need not be as densely constructed as in an urban setting, and economies of scale in public utility provision need not drive plot sizes and housing patterns as strictly. Similarly, farm sizes need not be held to "viability" norms found in commercial farming. Peri-urban farming will typically be a part-time activity, constituting a small, but significant, share of a household's income. In the absence of more flexible and integrated policies and programs for land devel- opment, the prevailing models of land use and land development will remain those that were introduced under the apartheid era, effectively (although unintentionally) reinforced now by measures such as the urban edge policy and environmental regulation. Challeng- ing the spatial patterns of apartheid is difficult in any circumstances, but it is particularly difficult in the absence of a supportive policy framework and program. A number of lessons and recommendations emerge from this case study. Improving access to land for low-income households in the peri-urban areas to allow for the type of integrated housing and agriculture settlement aspired to by the MHTF needs to be explicitly 16. Peri-urban areas in South Africa surround the larger cities, and while previously designated as rural, are now becoming increasingly but unsystematically urbanized. This report's focus is specifically on those peri-urban areas which have been targeted by commercial developers for "edge city" type land development. They are characterized by high levels of speculative activity and, on the whole, rapidly rising land prices. 13 14 World Bank Working Paper facilitated. This also follows international practice suitable for peri-urban areas. This facilitation should include: Revisiting the urban edge policy, which creates too strong a dichotomy between residential and agricultural land; Overcoming reluctance from municipalities to the settlement of low-income families, due to fear of non-payment for municipal services and unwillingness to forego income from alternative, "high-end" land development; MakingtheIntegratedDevelopmentPlansexplicitlypro-poorforbothdemand-led and supply-driven land acquisition and housing development; Simplifyingandaligninglegalprocedures; Clarifyingintergovernmentalrolesandresponsibilities; Unifyingrelevantsubsidiesanddecentralizingthemtothelocallevel; Introducingkeyreformstoenablemoreeffectiveparticipationinthelandmarket by low-income South Africans; Identifyingandimplementingappropriateinfrastructureservicestandards;and Buildingand/orfreeingupcapacitytosupportdemand-drivenhousingandland reform proposals and proponents. The following sections provides an analysis of each of the challenges that frustrate the poor as they try to access land for housing in peri-urban areas. It offers lessons learned from this case study and gives corresponding recommendations. It also proposes concrete approaches for assessment and transformation of a fragmented legal framework into a pro-poor and developmental instrument to facilitate the accelerated delivery of such housing. Overcoming Reluctance and Resistance Overcoming Reluctance of Municipalities Municipalities in South Africa are unwilling to acquire and develop, or support the development of, well-located land for low-income groups for three main reasons: Income foregone from alternative, "high-end" uses; fear of non-payment for municipal services; and cost implications of the constitutional guarantees for basic services. Income Foregone from "High-end" Land Development. The first reason municipali- ties are often reluctant to set good land aside for low-income groups is that in booming spec- ulative land markets, such as the West Rand where the Ethembalethu project is located, promoting high-end land development, which caters for the high-income market, is a lucrative and crucial source of income for the municipalities. For instance, the Mogale City IDP expressly targets this market for expansion. For a cash-strapped municipality such as Mogale City, the short-term advantages of rapidly promoting high-income land devel- opment are self-evident. However, international experience demonstrates that low- and middle-income fami- lies can also constitute a substantial source of revenue to municipalities over time. Well- located, low-cost housing developments--as is the case in Ethembalethu--appreciate in value, generating wealth for residents. This asset wealth is gradually leveraged into new income-earning investments, and as property values and incomes increase, municipal In Search Of Land and Housing in the New South Africa 15 property tax revenues will theoretically grow accordingly. In the rapidly growing economies in and around South Africa's metropolitan areas, the locations that will show the most asset-value appreciation over the medium term are precisely the peri-urban edge areas near transportation routes and growing sources of employment. Remote RDP hous- ing sites that maximize capital investment by selecting the cheapest possible land are not connected to the emerging patterns of investment and growth. This disconnection from near- and medium-term asset appreciation will burden municipalities with operation and maintenance subsidies for decades to come. Non-payment for Municipal Services. The current and sad reality is that the residents of many low-income areas in South Africa are a drain on municipal resources. This is due to their inability (and in numerous cases, unwillingness) to pay for municipal services including water, electricity, sanitation and solid waste removal. The great majority of arrears owed to urban municipalities firstly arise from previous debts from township rent boycotts under apartheid and secondly from non-payment for services by residents of rapidly growing informal settlements and subsidized low-income housing (the so-called "RDP houses," provided through NDOH grants). This failure to pay for municipal services has resulted in the accumulation of huge arrears across all municipalities, estimated to exceed R20 billion in aggregate across the 283 local authorities. The Mogale City municipality already faces arrearages of about R400 million against an annual budget of R640 million, of which the capital expenditure portion is R70 million. Because of this, many municipalities are reluctant to support development of housing for low-income citizens who are likely to add to service payment arrears. To help munici- palities resolve this problem, the National Treasury has introduced the "free basic services" grant. This grant subsidizes the consumption of water, sanitation and electricity up to certain thresholds set sufficiently high to meet the basic needs of low-income households (although the adequacy of these thresholds is the subject of ongoing debate.) Recovery of costs above these minimum thresholds remains the responsibility of municipalities. Failure to pay by poor consumers is difficult to address, because it would ultimately rely on the politically sensitive legal process of eviction. Thus, it is particularly ironic that even projects like Ethembalethu, with its positive track record of members' own-savings and investment, have found themselves getting bogged down. Cost Implications of Constitutional Rights to Basic Services. Municipalities also feel constrained by the constitutional guarantees of basic services for all citizens. At a practical level, these guarantees have been translated--particularly in RDP housing in urban areas-- into development standards that are often unaffordable in terms of operations and main- tenance. Large plot sizes require extensive road and drainage networks, water-borne sewerage is very expensive to maintain, and 36m2 fully finished houses have become the expected standard. To remain within the overall subsidy limits, the setting of high infra- structure standards with high capital costs has resulted in locating RDP housing in remote areas where land is relatively inexpensive. But remoteness from sources of employment and commerce limits the ability of RDP residents to improve their incomes and conse- quently their ability to pay for services. Completing this vicious cycle, municipalities must subsidize from their own resources the unaffordable operation and maintenance costs of such RDP housing developments, and then become reluctant to accept new liabilities. 16 World Bank Working Paper The design of the Ethembalethu project directly addressed many of the issues that have afflicted RDP developments. First, the project site is well located near transportation routes to business and employment areas, yet is still sufficiently peri-urban that the cost of land acquisition is relatively low. Second, the members have consciously chosen to accept more affordable standards for capital investments and correspondingly reduced operation and maintenance costs. These include septic tanks or Ventilated Improved Pit-latrines (VIPs), narrower internal roads in keeping with low projected volumes of traffic, and flexibility on finishing standards of houses. This is on the assumption that beneficiaries can and will make further improvements through their own resources over time. Third, the MHTF members have selected lower capital investment standards precisely to allow for larger plot sizes and communal areas that will enable small-scale farming activities and provision of rental accommodation, both of which will add to incomes and affordability. Developments like that proposed by MHTF, which meet the constitutional service- delivery mandates in a manner which minimizes recurrent cost obligations for the munic- ipality, make sense on numerous levels. However, it seems that government officials would benefit from more practical guidance in interpreting the constitutionally guaranteed rights to basic services. Making Land Use Pro-poor Revising the Urban Edge Policy The current definition of the urban edge is not appropriate for combined housing and agri- cultural uses. As a consequence of this policy, almost no land is available for such purposes. In 2002 the MEC for Development Planning & Local Government in Gauteng pro- claimed an "urban edge" around the urban areas of Gauteng, which was approved as a provincial policy by the full Executive Council of the Province. The urban edge is effec- tively an urban growth boundary beyond which the provision of infrastructure and indeed any form of "urban" development is now prohibited unless there are compelling excep- tional circumstances.17 The municipalities refined this boundary where it affected their areas and there is not always complete congruence between the provincial and the local urban edges. Nevertheless, most stakeholders have taken this boundary seriously and it has become increasingly difficult to obtain planning approvals for projects beyond the edge. The Gauteng Development Tribunal, as a provincial body, feels especially obliged to respect the urban edge, although it has on occasion approved projects that fall outside this boundary. The provincial urban edge policy permits the approval of development beyond the edge in certain cases, which include the case of "rural residential uses or agricultural holdings in specified areas." The density of the MHTF proposed project, however, makes it unlikely that it will fall within either of these categories, especially in the light of the municipality's concerns with a compact layout to minimize infrastructural costs. There is a clear need to revisit the urban edge policy to allow for the type of integrated housing and agriculture development which would be suitable for low-income South Africans in peri-urban areas. A line on a map is not a useful policy in this context and a 17. The Urban Edge is not a statutory instrument, but a policy, which is applied by the provincial gov- ernment members responsible for land development decision-making. In Search Of Land and Housing in the New South Africa 17 more nuanced policy framework for peri-urban development is required, one that accepts the complex and unequal nature of the market for land development rights in those areas. Even if, for argument's sake, the urban edge vision is accepted, the availability of sufficient land for low-income housing within the urban edge must be closely interrogated. The GDOH has conducted an extensive mapping exercise of land availability in the province. This information is not yet publicly accessible. However, based on the department's presentation at the 2005 Provincial Housing Indaba, it appears that there are no more than 50,000 hectares of vacant land within the urban edge suitable for residential development. This land is obviously not guaranteed for residential purposes, let alone low-income residential or land reform purposes, and competition for it from other land uses will only become fiercer as the supply dwindles. Given the fierceness of competition in the peri-urban areas, as demonstrated by the MTFH experience, competition within the urban edge no doubt will be even more severe. In other words, low-income citizens are effectively prevented from owning freestanding houses and plots within the urban edge. Therefore, they are condemned to: a) renting, which prevents asset accumulation; b) buying houses in remote areas with low land values, again constraining asset appreciation and perpetuating the apartheid dual human settle- ment patterns; or c) establishing informal settlements. We therefore suggest revision of the urban edge policy to take account of the desire and need of poor people to combine housing and agriculture near jobs. This could include allowing municipalities to propose and implement such changes via their IDPs. A task force led by the Presidency, responsible for the National Spatial Development Program, could undertake this revision and oversee its implementation. Making the Integrated Development Plan Explicitly Pro-poor The review of the urban edge policy proposed above should be executed in collaboration with municipalities to ensure maximum congruence between the provincial and municipal urban edges. The municipal IDP18 is meant to be the primary planning instrument for guiding decisions on local development issues. In Mogale City the IDP emphasizes that housing is a municipal priority of the highest order but identifies Muldersdrift as an area for "demand-driven"housingasopposedto"needs-driven"housing.Itseemsthat"needs-driven" housing is the term used to describe low-income housing. The assumption underlying this argument is that the poor cannot drive demand, but simply have needs that must be met. However, in practice the poor do drive demand for land and shelter, as can be seen by the mushrooming of informal settlements and the emergence of organizations such as MHTF. The Mogale City IDP fails to clearly support the poor in their demand for good land.19 For instance, whereas the Muldersdrift Spatial Development Framework, which is formally 18. The IDP is a mandatory plan required of all municipalities in terms of the Local Government: Municipal Systems Act (2002). 19. In all fairness, the Mogale City IDP does include a "Muldersdrift" housing project of 250 "planned erven" on its list of proposed projects. The project is scheduled for implementation in 2006/07 and these 250 "planned erven" have to compete with 22 other projects representing 14,350 units, most of which were scheduled to commence construction in the 2004/05 or 2005/06 financial years. The IDP also accepts that the major challenges to providing housing for the poor in the municipality are "[the] delay of EIA approval by GDACE, objectors and land availability in rural areas". 18 World Bank Working Paper part of the IDP, identifies Muldersdrift as an area "in transition," it does not specifically pro- vide any support for a project such as this one. In the inevitable face of Not-In-My-Backyard neighbors' opposition to projects such as Ethembalethu, an IDP that is ambivalent on the issue of land for the poor is one that will, by default, inexorably promote segregated pat- terns of land use and restrict the poor to the margins of the urban land market. Although the IDP does not provide a basis for rejecting the proposed project outright, it also does not provide a categorical argument supporting this or similar projects. In short, any IDP that is not explicitly pro-poor is de facto anti-poor. The process of land identification in this case was carried out entirely by the MHTF members. Through their efforts, spanning many years and traversing many obstacles, they identified the land, engaged in successful negotiations with the seller to acquire it and then engaged with the municipal, provincial and national authorities to ensure that it was actually purchased. They have clearly and unambiguously been shown to be perfectly capable of negotiating a demand-driven land identification and acquisition process. However, government processes have, so far, not been able to actually transfer the title deed of the land to the community itself. The land is being made available to the MHTF through a land availability agreement, but the MHTF still lacks formal rights to the land, or planning permission for its development. Demand-driven land identification, in which the primary responsibility for identifying, acquiring and preparing the land for development falls on the proposed beneficiaries, has not been followed by an effective state "supply" response to transfer the land to the community. Instead, the state acquired--and still holds--the land itself. Some would argue that the process of land identification and acquisition for the poor should be carried out entirely by the state. However, South Africa's experience with a purely state-led approach does not corroborate this proposition. For example, a large state- driven housing project--"Cosmo City," not far from the site of the MHTF project--has been driven with considerable difficulty by the City of Johannesburg. Officially announced in 1997, this project has only recently (2006) begun to be implemented, after many years of legal conflict with neighboring objectors. When completed, Cosmo City will only satisfy a small portion of the housing needs existing within the informal settlements which the project is supposed to replace. In other words, Cosmo City demonstrates that the supply-led approach can be as slow and lacking in effective delivery as the demand-led approach exemplified by the MTHF case. What seems clear, is that both demand-driven and state-led processes face consider- able implementation obstacles. Neither approach will have much success if these obstacles are not removed. Could an improved IDP process assist in alleviating these implementa- tion problems? For instance, a process of broad land identification (zoning), could be car- ried out as part of the IDP process, in consultation with the GDOH and the provincial office of DLA, and taking into account the needs of the various communities in Mogale City. Such an exercise could guide the decisions by beneficiaries and government by defin- ing areas in the municipality for low-income housing and the type of mixed development that the MTHF plans. Such zones or areas could then be granted special status by already "pre-qualifying" them for "planning permission" and EIA scopings, and be made immune to unnecessary law suits. This approach would facilitate both demand and state-led approaches. In Search Of Land and Housing in the New South Africa 19 Reforming the Land Market Another crucial factor in peri-urban areas, especially during the recent property boom, has been widespread land speculation, which drives up prices and impedes the ability of low-income beneficiaries to acquire assets. Speculation is biased strongly against the poor, who typically can only afford to pay the productive value of the land since they have no access to credit. Scrapping the Subdivision Act Faced with very high prices, the poor typically opt for small plots, if land markets make such small pieces of land available. Land markets in rural and peri-urban areas are, however, subject to subdivision restrictions in terms of the Subdivision of Agricultural Land Act (1970). Although this law was officially repealed by the South African parliament in 1998, the Department of Agriculture has opted to postpone the date on which this repeal comes into effect. In this way, the law, which was originally designed to assist the state's policy of maintaining minimum farm sizes, continues to be enforced in a completely different policy environment. While all the current government policies point towards encouraging small-scale and intensive agricultural development by "emerging" black farmers, the Subdivision Act remains as a formidable obstacle to the implementation of these policies. The law itself does not prescribe minimum farm sizes but it does require any subdivision of agricultural land to obtain the prior, written consent of the Minister of Agriculture. Increasingly, the minister has exercised her discretion generously, rather than restrictively, and has not directly impeded the subdivision of land, especially where the intended landowners are small-scale black farmers. In practice, however, this law remains as a fur- ther bureaucratic obstacle--well past its sell-by date--in a process already top-heavy with rules and regulations. Avoiding Regressive Land Taxation The historical absence of a land tax on agricultural land has increased the attractiveness of holding land as an asset without productive use. To date, the land wealth of large farms has either not been taxed at all, or, as in Gauteng province (based on the old Transvaal province's Local Government Ordinance, 17 of 1939) has been taxed at a rate 100 times less per hectare than that which applies to a 1 hectare small farm. This policy is in place in Mogale City and makes subdivision unattractive. The Local Government: Municipal Property Rates Act (2004), establishes a new legal framework for land taxation and empowers municipalities to tax agricultural land accord- ing to locally defined policies. This new law came into operation on July 1 2006. It gives municipalities considerable freedom to determine their approaches to land taxation, including agricultural land. Mogale City municipality has an opportunity to design a policy for land taxation that discourages speculation and encourages subdivision, should that be the municipality's intention. Unfortunately, in the absence of guidelines to implement the new act for agri- cultural areas, some municipalities, including Mogale City, are tempted to copy the old 20 World Bank Working Paper 1939 system, even though this ordinance was scrapped with the passing of the new act. If this were to happen, land taxation would continue to be highly regressive and anti-poor. Realigning Planning Processes Merging Township Establishment and EIA Processes Under existing procedures in South Africa, two processes--EIA and "township establish- ment"--that are critical to the implementation of any housing development, including MHTF, run simultaneously. This practice is far from ideal, creating confusion among stakeholders, and providing additional opportunities for project opponents to voice their opposition. Especially for well-resourced objectors this provides them with the opportunity to have "two bites at the cherry." It also places considerable additional strain on officials in the relevant departments. No matter how well the planning and environmental consultants may coordinate their activities or how effectively various officials interact with each other, the parallel and over- lapping legal procedures for obtaining planning and environmental permissions have the effect of: Confusingstakeholders,whodonotalwaysunderstandtheoftenarcanedifferences between the two processes; Compromising the quality of each process--the decision-makers are always able to "pass the buck" for difficult decisions and so neither process needs to deal with all issues precisely and clearly; Increasingthecostoftheapplicationfortheapplicant;and Slowingtheoverallprocess,astheparticipantsineachprocessawaittheoutcomes of the other process. The duplication of effort and resources in subjecting projects to both planning and envi- ronmental assessment, when the two processes increasingly deal with identical issues, is wasteful. Streamlining the legal processes through which these projects pass will present fewer opportunities for legal challenge, and will also increase the likelihood of more sustainable and well-considered projects. Concerted effort on the part of both the Department of Environmental Affairs & Tourism (DEAT) and the DLA is required to achieve rationalization of these two processes. Currently the DEAT is responsible for the EIA process, now regulated in terms of the National Environmental Management Act's (NEMA) new EIA regulations. Planning authorizations for township establishment are currently issued in terms of either the provincial town planning laws dating back to before 1994 or the 1995 Development Facil- itation Act. The DFA was intended as an interim measure until a comprehensive overhaul was done of the entire legal framework governing land development and land use plan- ning. So far this overhaul has not happened. The lessons that emerge from the Ethembalethu experience demonstrate clearly the urgent need for these laws to be rationalized and, in the process, to establish a coherent and efficient relationship between environmental and planning authorizations. The 2001 In Search Of Land and Housing in the New South Africa 21 White Paper on Land Use Management & Spatial Planning is a commitment by cabinet to such a process, but legislative efforts in the intervening years have not achieved the objec- tives of that document. Very few, if any, of the prescribed legal requirements are unreasonably complex, if viewed in isolation from the others. In practice, though, the range of laws and regulations effectively overwhelms officials, and the failure of the laws to acknowledge other, parallel, sometimes conflicting laws makes coordinated implementation virtually impossible. The range of different requirements, exacerbated by the absence of clear coordinating mecha- nisms for the different spheres of government responsible for their implementation, makes the problem of overlapping procedures very disempowering especially for comparatively under-resourced, demand-driven actors like MHTF. Therefore, a comprehensive set of legal requirements or checklist for a land and housing project should be compiled and merged into one, single procedure. The table below sets out the range of laws that have to be complied with in the pursuit of land and housing. In addition to the legislation, the MHTF must navigate five different sectors and no less than 10 government institutions. The minimum time needed to com- ply with the legislation regarding housing is nine months (see Table 1). Simplifying Township Establishment Procedures There are three possible routes to obtain legal permission for land development, or "township establishment". Although comprehensive in many respects, "township establishment" does not represent all legal approvals that are required, but is the most important step required before the national Registrar of Deeds will issue title deeds to the individual pur- chasers or beneficiaries. So township establishment tends to be central to the overall process of obtaining all the other permissions and approvals. The three possible routes to township establishment are via the following laws: The Development Facilitation Act (1995). The DFA was enacted to speed up the approval of land development projects. It provides for each province to establish its own Development Tribunal, an appointed body consisting of equal numbers of government officials and private sector experts with wide powers to ensure that land development projects receive speedy approvals. The underlying principle of the DFA is that the tribunals operate in parallel to other existing structures: an applicant for land development permission or, in this case, township establishment, has a choice to follow either, for example, the applicable provincial planning ordi- nance or the DFA. Instead of the phrase "township establishment" the DFA uses the phrase "establishment of a land development area." TheLessFormalTownshipEstablishmentAct(1991).LeFTEAwaspartofapackage of laws approved by the apartheid government in its last years as a belated attempt to expedite land delivery to blacks who previously had been denied this opportu- nity. LeFTEA is unconstrained by the constitutional strictures of post-1994 legisla- tion and so provides for minimal public participation and gives officials a wide discretion to approve township establishment regardless of issues such as plot size, service standards and environmental considerations. 22 World Bank Working Paper Table 1. Legislative Compliance Department Sector and legislation Minimum time needed Environment Gauteng Department of National Environment The new regulations have notably Agriculture, Conservation Management Act, put ceilings on some departmental and Environment 107 of 1998 response times. At best, the application will take 6 months.20 SA Heritage Authority National Heritage Heritage impact assessment is Resources Act, often linked to the broader EIA. 25 of 1999 Gauteng Department of Environmental These EIA provisions have been Agriculture, Conservation Conservation Act, repealed. However, in previous and Environment 73 of 1989 applications, typically the minimum time is 6 months where the application: Is uncomplicated, for instance the specialist studies are straightforward and raise few objections from neighbors; Provides sufficient information upfront. Development planning The Ministry of Subdivision of Generally, no response time ceilings Agriculture & Agricultural Land required of authorities for responding to Land Affairs Act, 70 of 1970 various applications. However, on average 2 months for best-case scenarios. Gauteng Department of Town Planning and Minimum 6 months. Generally no Development Planning Townships Ordinance, 15 time limits for authorities to respond. & Local Government of 1986 (the ordinance) Gauteng Department The Less Formal Minimum 4 months. Generally, no of Finance & Economic Township time limits for authorities to respond. Affairs Establishment Act, 113 of 1991 (LeFTEA) Provincial Development Development At best 4 months. Tribunals Facilitation Act, 67 of 1995 (DFA) Mogale City21 Local Gauteng Removal At best 3 months. Generally no time Municipality of Restrictions Act, limits for authorities to respond. Often 3 of 1996 linked to Town Planning establishment. Land reform National Department of The Provision of Land At best 2 months. Land Affairs and Assistance Act, 126 of 1993 National Department of Land Survey Act, In a best-case scenario, 1 month. Land Affairs 8 of 1997 Source: http://www.environment.gov.za/soer/reports/Mogale%202003/background.pdf accessed June 8, 2006. 20. See Table 2. 21. The Mogale City Local Municipality (MCLM) was established after the democratically held Local Government elections on December 5, 2000. Mogale City is the area that is formerly known as Krugersdorp Local Council, the Magaliesberg Rural Council and a portion of the Magaliesberg Local Area Committee. In Search Of Land and Housing in the New South Africa 23 Table 2. Time Lines for Various Applications for Township Establishment: Best-case Scenario Time Line (months) 1 2 3 4 5 6 7 8 9 Land allocation from government (Provision of Land Assistance Act) Environmental Impact Assessment (ECA, NEMA Heritage, Transport Acts) Town planning establishment (Ordinance, DFA or LeFTEA) Removal of restrictions (Gauteng Removal of Restrictions Act) Division of agricultural land (Subdivision of Agricultural Land) Land registration (Land Survey Act) The Town Planning & Townships Ordinance. This is a provincial law dating back to the old Transvaal province and was enacted in 1985. After 1994 this law was assigned to the provinces that made up the old Transvaal (Gauteng, Limpopo, Mpumalanga and North West) and each province now administers the law as if it were a law passed by its legislature. This law was used to regulate township estab- lishment as well as land use management (rezoning, land use change, and so forth) in the formerly white areas of the Transvaal. Now it applies to the entire province of Gauteng. In evaluating these options for proceeding with Ethembalethu it is important to remember that the EIA process has to be accommodated in all cases. As the provincial environmental authorities require a reasonably final description of the project, in terms of layout, density and coverage, it is not possible to process the EIA and the land devel- opment application in isolation from each other and the two processes must run in parallel for at least a portion of each process' own likely time-frame. Another parallel process that has to be undertaken is the application for subdivision approval in terms of the Subdivision Act. In terms of the Subdivision of Agricultural Land Act (1970) the subdivision of land previously designated as agricultural must be approved by the national Minister for Agri- culture. This process takes a minimum of three months to complete and requires a 30-day period during which the surrounding community and other stakeholders may comment on the proposed subdivision. Note that stakeholder consultation is duplicated, as it is also required under the environmental impact assessment. Of the three options it seems reasonably clear that LeFTEA will be undesirable in the present circumstances given its origin in apartheid law. Despite the apparent attractiveness of its speediness, in practice it is likely to result in a more acrimonious and ultimately protracted process, probably involving a constitutional challenge. It is also unlikely that 24 World Bank Working Paper the approval of development beyond the urban edge will be given in this case as the provin- cial officials responsible for enforcing the urban edge have to assent to any approval. This then leaves the choice between the DFA and the ordinance. A wide range of public and private sector experts consulted on this matter agreed that there would not be a great deal of difference between the two, in terms of speed, cost or likelihood of approval. The key issues that will affect a choice between them are: TheDFAprocessdependsagreatdealontheplanningconsultantforeffectiveand efficient compilation and presentation of the application, whereas the ordinance allocates a wider range of responsibilities to municipal officials. The relative capac- ity of the planning consultant and the municipality will be crucial factors to con- sider when making this decision. In the light of the inevitability of a strident and well-resourced objector lobby, as well as the inevitability of an appeal being lodged against any decision that is finally taken, it is important to consider the advantages that the DFA brings as an inde- pendent decisionmaker, unlike the municipality in the case of the ordinance route, where the accusation of inherent bias is more easily made. In any event, in the case of the MHTF, it was decided that the DFA would be the most suitable approach, but there are no overwhelming or compelling reasons why the ordinance should not be suitable. However, to prevent a repeat of the 2003 withdrawal of an incomplete DFA application it will be essential that the planning consultant and the municipal officials work very closely to ensure that a high quality application is submitted. The above narrative serves to demonstrate the complicated nature of the process and how difficult it is to understand. The recommendation is to improve the township estab- lishment processes by eliminating LeFTEA, adjusting the two remaining options to make room for mixed-use villages, and simplifying their requirements. Designing a Program for Peri-urban Areas Accommodating Mixed Land Use and Multiple Livelihoods As mentioned earlier, the peri-urban areas have not benefited from the current policies and programs. The case of Ethembalethu applies: the DLA was unable to single out any one of its available programs to suit this project, the establishment of a village. So the department officials devised a hybrid approach. The DLA administers specific programs through which it may purchase land for redistribution, for example, land reform for agricultural devel- opment (LRAD), the settlement and land acquisition grant (SLAG) and the commonage programs.22 To accommodate the establishment of a village, however, the purchase on behalf of MHTF became a unique combination of the SLAG and commonage grant for- mulated on an ad hoc basis by the Gauteng Provincial Office of DLA.23 The purchase could not have been made in terms of any one of the specified programs alone, for these reasons: 22. All three of these programs have their legal basis in Act 126 of 1991. 23. There is an internal DLA memorandum that was used to authorize the payment, and this document may shed more light on the nature of the transaction. It has, however, not been possible to get hold of it. In Search Of Land and Housing in the New South Africa 25 LRAD is designed primarily for agricultural development, while in this case, even though the project was to include an agricultural component, it was to be mainly residential. SLAG could be used in principle, but it would have prevented using the housing grant. An inter-departmental agreement between the national Departments of Land Affairs and Housing prohibits a SLAG beneficiary from also benefiting from a housing subsidy. Using the commonage grant would have precluded the municipality from later transferring the full ownership of the residential plots to the individual beneficiary households. A commonage remains municipal property, with the municipality issuing leases for its use. The hybrid approach devised by the DLA officials combined the SLAG and the common- age grant. In this way the benefits of the SLAG (which can be used to acquire land for subdivision for housing purposes and transferred to individual households) was combined with the benefits of the commonage grant--for which there was no need to identify each of the individual beneficiary households. The downside of this approach was that part of a SLAG grant is a planning grant, which is calculated as a percentage of the total land acquisition grant. The planning grant is set aside to cover the costs of the land use planning, environmental planning and business planning needed to promote a sustainable and viable human settlement. Because the approach was not strictly or exclusively a SLAG approach, but a hybrid, the project was deemed to be ineligible for the planning grant. Another impediment to the transfer of land ownership to the community was that no contract or agreement was concluded between the DLA and Mogale City as to how the land should be used once it had been transferred to Mogale City. The understanding was that it should be used to accommodate the MHTF community, and the community's claim and rights to this land should be confirmed in a "land availability agreement"24 signed between the MHTF and the municipality. The case study demonstrates that neither the mass-produced housing estates promoted to date by the national housing subsidy program, nor the agriculture-led land reform program designed for rural areas are applicable or appropriate in the peri-urban context. A new program needs to be put in place, making the existing programs more flexible and integrated to accommodate the need for integrated housing and agriculture settlements in the peri- urban areas. Setting Affordable and Appropriate Standards for Infrastructure and Housing As mentioned earlier, municipalities often feel constrained by the constitutional guarantees of basic services for all citizens. This is because these guarantees have often been interpreted in terms of development standards that are economically unaffordable for the poor and the municipality. Worse, to remain within the overall subsidy limits, low-income housing 24. A land availability agreement is a comprehensive agreement or contract concluded between an organ of state, typically, and a group of beneficiaries or, sometimes, a developer, in terms of which a spec- ified land parcel is secured for the benefit of the beneficiaries or developer while the ownership resides with the organ of state. See footnote 12. 26 World Bank Working Paper is then often moved to unattractive locations where land is relatively inexpensive, repli- cating the old geography of apartheid. However, the design of the Ethembalethu project shows that an appropriate balance can be found between location, plot size and cost of land, on the one hand, and affordable standards for infrastructure and housing (including septic tanks or VIPs, narrower internal roads, and flexibility on finishing standards of houses), on the other. Low-income households should have the option to build their own houses. As the Ethembalethu community demonstrated locally (and as has been demonstrated in innu- merable cases internationally) many poor people can generate their own funds to build houses, if they can access land. Title could be granted provisionally and transferred fully after a window (say 5­10 years) by which time a house must have been built to a minimum set of standards. In this way, government can allocate more resources to land acquisition and servicing the land and less to building houses. Access to both land and subsidized hous- ing are needed since not all of the poor will be able to mobilize the resources to build and not all will want to buy their own land or house in the urban areas. Re-engineering Program Implementation Clarifying Inter-governmental Roles and Responsibilities While comprehensive manuals and guidelines exist for delivering land and housing, officials in all three spheres of government struggle to match their prescriptions to the situations faced by prospective beneficiaries "on the ground." In practice, officials tend to adjust their activities to suit the immediate challenges facing them rather than to fit in with every requirement of the official documents, where these exist. The officials generally are aware of the shortcomings in this approach, but they also understand that it is necessary to ensure implementation takes place. If they wait for everything to be done according to the law they fear they will never be able to deliver and implement projects. They are aware that their meth- ods are not always strictly by the book. As public officials though, they would obviously pre- fer not to have their approach subjected to too much scrutiny by other officials. This leads to an inevitable reluctance to engage with the representatives of other spheres of government, or indeed any other stakeholders, for fear of exposure, censure, or disciplinary procedures. This approach leads to increasingly weak lines of communication between officials, delayed implementation and frustrated stakeholders (including both beneficiaries and objectors). These hesitant, and even secretive, inter-governmental relations are compounded by the problem of unclear assignment of functions relating to land and, to a lesser extent, housing in the South African Constitution. Functional assignments are set out in Schedules 4 and 5 of the Constitution. Schedule 4 sets out areas of concurrent legislative competence, that is, where the competence is held concurrently by both provincial and national government. Schedule 5, on the other hand, lists the areas of exclusive provincial legislative competence.25 25. It is beyond the scope of this report to examine these constitutional provisions in detail, but obviously this split between concurrent and exclusive powers is circumscribed by a number of principles and precedents. In addition, each schedule is divided into two parts, with Part B of each schedule setting out the legislative functions of local government as well. The meaning of concurrence in the South African Constitution is also the subject of ongoing deliberation by constitutional experts and the courts. In Search Of Land and Housing in the New South Africa 27 Where an aspect of governance is not listed in either schedule, it falls under the national government's exclusive competence. Land is one such aspect. Housing, however, is listed as an area of concurrent competence, shared between national and provincial government. In practice, this has translated into a scenario in which both land reform and housing delivery are regulated by national legislation, but in the case of housing the implementation is done by provincial government. Also, in the case of housing there is a de facto assumption of various actual housing delivery functions by local government. However, the funding stream flows through provincial government26 and the process of formal accreditation of municipalities to receive housing funds directly and so to implement housing projects remains incomplete. Clearly it is very difficult to manage housing programs effectively in isolation from the questions of land acquisition and land reform. Similarly, it is difficult for individual pro- jects to be managed effectively by a municipality when the budget sits with provincial gov- ernment. Rationalizing these inconsistencies is a necessary, if not sufficient, prerequisite to speed up housing delivery. The introduction, in 2004, of the "Breaking New Ground" housing policy delegates housing delivery competence to "accredited" local governments. This added another layer of complexity to the concurrent competence functional defini- tion (although no local governments have been accredited to date). A further factor undermining the capacity of the state to deliver land and housing pro- jects effectively and speedily is the treatment of the planning function in Schedules 4 and 5 of the Constitution. While municipal planning is an area of concurrent national and provincial responsibility (and one in which local government has a specified role as well) it is not clear to what extent this function includes the regulation of land use and land development. In practical terms, it is difficult to ascertain whether these aspects of regula- tion fall in the area of legislative competence of municipal planning, or under that of land. Consequently, there is ongoing uncertainty as to whether national or provincial government is the correct sphere to regulate land use and land development. This uncertainty adds considerable additional complexity to an area in which apartheid-era laws still prevail, with different norms, standards, and procedures applicable in each of the apartheid race zones. Further complicating this area of governance is the question of environmental regu- lation. The power to legislate over environment is held concurrently by both provincial and national government. In practice, environmental permissions for land development pro- jects or major land use changes are obtained in terms of national laws, but the permissions are given by provincial governments. Inevitably, in the case of a large project such as Ethembalethu, both planning and environmental laws have to be followed and the two sets of permissions have to be obtained from different authorities, leading to uncoordinated decision-making. In summary, substantial work needs to be done to ensure that the constitutional framework for assignment of powers and functions--for the legislative areas of land and planning especially, but also environment and housing--is rational and appropriate. This is not a "quick fix": the processes either of getting interpretations of the relevant constitu- tional provisions from the Constitutional Court or of amending the Constitution through 26. A process of accrediting municipalities that meet certain criteria is currently in place. Once a municipality has been accredited it will then have the power to receive funding directly from national government. 28 World Bank Working Paper parliament are extremely lengthy. Nevertheless it is essential that this process begins. Numerous senior lawyers have provided opinions to provincial and local government, especially in the provinces of KwaZulu-Natal, Western Cape and Gauteng. A first step should be an evaluation of these different opinions, as well as the few court judgments that deal with the issue. On the basis of this evaluation, a proposal to the national cabinet would need to be made for agreement, in principle, to re-open this issue. This clarification of the legal and constitutional responsibilities is a precondition for the effective alignment of the numerous programs and policies applicable to projects such as Ethembalethu to enable more effective implementation in future. The absence of a clearly defined political responsibility for the implementation of integrated land and housing projects, especially in peri-urban areas, however, inevitably erodes confidence within the sector that this will actually happen. Decentralizing and Unifying Budgets The complexities introduced by the different budget and subsidy flows further complicate implementation. In the case of housing there is a de facto assumption of various actual housing delivery functions by local government, although funding flows through provincial government and the formal accreditation of municipalities to receive housing funds directly and so to implement housing projects remains incomplete. Clearly it is very difficult to manage housing programs effectively in isolation from the questions of land acquisition, land reform and ultimate responsibility for the recurrent costs of operations and maintenance. Land reform budgets sit with the national govern- ment, although, fortunately, decisions on most, but not all, land reform programs have been decentralized to the provincial directors of Land Affairs. A further decentralization to district and municipal levels is clearly required. Similarly, it is difficult for individual projects to be managed effectively by a municipality when the budget sits with provincial government. Again, further decentralization to the municipality is required, as envisaged but not yet realized through the 2004 "Breaking New Ground" comprehensive plan of the Department of Housing. However, as budgets get more decentralized, the municipality should not become merely the disbursement window of all the various programs. The land reform and hous- ing subsidy programs should be combined in unified budgets at the municipal level, and these unified budgets must be based upon medium range municipal housing and land reform strategies that are effectively expressed through "locked in" capital investment and financing plans. Prior to 2004, municipalities were not officially assigned housing delivery functions and so no municipalities in South Africa have yet developed medium-term housing strategies (although Tshwane and Ekurhuleni metropolitan municipalities are in the process of doing so supported by the Cities Alliance). In the case of land reform, municipalities are not legally assigned any functional responsibilities and do not, therefore, have any reason to define localized strategies. This situation must be rectified with clear functional respon- sibilities being assigned to local governments, as is now the case in the housing sector. Rationalizing inconsistencies, decentralizing and unifying various budgets, and clari- fying delivery responsibilities are necessary, but not sufficient, prerequisites to speed up land and housing delivery. Local governments are hamstrung by weak capacity and regu- latory and policy frameworks that fail to meet the needs of the local context. In particular, In Search Of Land and Housing in the New South Africa 29 there is neither dedicated support to municipalities to engage actively in the urban land markets on behalf of the poor, nor are there effective financial and legal instruments to enable municipalities to do so. The process of accrediting municipalities to manage housing delivery is one important step towards strengthening and clarifying local government's role. However, it will be insuf- ficient on its own until such time as the regulatory and policy frameworks are rationalized and new, effective instruments developed. Freeing Up and Building Capacity There are capacity constraints at all levels of government in South Africa. This is widely acknowledged and is receiving attention. However, one of the main conclusions of this report is that the complex nature of government interventions creates a self-imposed capacity constraint. Government officials could work faster, and deliver better, if they could implement more flexible and integrated programs with more decentralized deci- sionmaking. Such reforms would free up existing capacity. Currently however, as this report shows, government officials are unable to provide appropriate advice on the ser- vices, products and programs available to meet the needs of communities such as the MHTF. They are also unable to correctly interpret the relevant policies, legislation and principles applicable in cases such as this. A fresh set of step-by-step manuals outlining the typical decision-tree in such projects would be very useful to guide officials in all three spheres of government in the future. In the meantime, key areas for capacity building are legal skills and project management. In both cases it will be necessary to develop training materials and courses specifically designed for officials and consultants working on land and housing projects. The particular complexities and dynamics of these projects demand that practitioners have more than a general understanding of either the law or project management; they need specialist knowl- edge of issues related to land and housing projects. A particularly important capacity gap is that of legal skills. Lawyers are not especially useful for the implementation of projects, but in the context of planning for land and housing projects it is essential that each sphere of government has professionally sound, carefully considered, and confident legal advice. In many cases, the "legal" arguments thrown up by objectors' lawyers are incorrect or spurious, but the diffidence of in-house legal advisors in the face of more aggressive and confident private sector lawyers undermines efforts to change patterns of land ownership. Where government lawyers are uncertain, officials without legal training understandably develop idiosyncratic legal interpretations of their own, leading to peculiar outcomes. This has a profoundly confusing effect on land devel- opment and land reform processes as different officials are often acting in terms of different understandings of what constitutes the correct legal position. In the Ethembalethu case, most of the "implementing" or "project" officials in all three spheres of government appear to be competent, driven and diligent. But, in the absence of sound legal arguments or clear political support for a particular project they tend to shift their focus to other, less legally and politically complex projects. Land and housing projects are inherently complex and politically charged. Therefore, they require exceptionally strong project management and leadership. Officials often avoid this sort of project; they are seen 30 World Bank Working Paper as more trouble than they are worth, and subject to high risk of a career-damaging project failure. For the same reasons it is increasingly difficult to get high-caliber consultants to support this sort of project. The problems of government capacity are compounded by the weakness of the NGO sector. There is no ready reference point for a project such as MHTF among existing NGOs, and the capacity within these organizations is inevitably overstretched. The past decade has seen a shrinking of the development NGO sector, especially in Gauteng. Without such sup- port, the poor are increasingly left to fend for themselves within a hostile and inadequate legal framework should they wish to obtain land for housing or other purposes. The NGOs in the land and housing sector should not be ignored in capacity-building efforts. Reinvigoration of NGO capacity to deliver would lead to a more dynamic and diverse housing and land reform development sector, in which inconsistencies and ineffi- ciencies could be identified and addressed more effectively. The experience of the MHTF shows that even where a community has stable and committed leadership as well as the ability to engage landowners, government authorities and technical experts with skill and confidence, it is still unable to access land. The support of NGOs certainly assisted the MHTF in the early stages of its existence, but since the weakening of the NGO sector the MHTF is increasingly dependent on its own resources. The MHTF could have been sup- ported by a strong NGO movement. For example, NGOs could have helped the associa- tion to obtain access to information. Under the Promotion of Access to Information Act (2000), the MHTF is entitled to information in the DLA's project file, yet they were unable to access it and at first seemed unaware that they had a right to it. CHAPTER 4 Recommendations and Suggestions he fact that the Ethembalethu community is still uncertain of accessing land after T more than a decade of struggle is not only the result of apartheid. It also stems from the current policy vacuum in peri-urban land reform and land development, as well as a system of government laws, regulations, programs, and implementation proce- dures that is too complex and in which key institutions are uncertain of their roles and responsibilities. Policy development alone will therefore be insufficient. It will have to be supplemented by the development of effective financial and legal instruments to implement the new pol- icy as well as effective capacity to do this. Improved policies and programs should enable the government to guide, and intervene in, the land market to ensure that land is identi- fied and acquired for land reform and housing delivery. In the following table, we have made suggestions for a number of key action areas to address the main issues emanating from this case study. These suggestions will need to be discussed with the various stake- holders. 31 Main action areas Key issues Consequences Recommendations Responsibility Overcome reluctance "Not In My Municipality" Municipalities are reluctant Combat culture of non-payment A task force led by the and resistance resistance to set aside suitable land for through media campaigns Presidency that is responsible Associated with non-payment low-income groups Demonstrate revenue potential for NSDP, and involving of services; loss of income of low-income settlements DPLG, DLA, DEAT, DOH and foregone from "high-end" Agriculture, and other Provide guidance for land use, constitutionally relevant departments interpretation of mandated service standards. constitutional rights "Not in my backyard" Poor communities encounter Mount a public information GCIS (NIMBY) resistance of land resistance from neighboring campaign about benefits of owners and neighbors communities to acquisition of undoing geography of apartheid suitable land and establish racially integrated neighborhoods Use expropriation of land where DLA NIMBY resistance is insurmountable 32 Make land use "Urban Edge" definition is Almost no land available for Revise the urban edge policy to A task force led by the planning pro-poor not appropriate for such combined uses take account of the desire and presidency that is responsible combined housing and need of poor people to combine for NSDP, and involving agricultural uses agriculture and housing near jobs DPLG, DLA, DEAT, DOH and Enable municipalities to Agriculture, and other propose and implement such relevant departments changes via their IDP No use of IDP or other Almost no land available for Develop models and guidelines Presidency and DPLG instruments for proactive low income groups for IDPs that are explicitly pro- designation or acquisition of poor with respect to land use land for low income groups and housing Restructure the land Difficulties in subdividing Insufficient land available in Implement the 1998 Repeal of Department of Agriculture market land small parcels suitable for the 1970 Subdivision Act poor people Regressive land taxation Poor incentives to dispose of Develop guidelines for DPLG and the National under-used land municipal property rates act Treasury Regressive taxation of that promote access to land agricultural land in former and housing for the poor Transvaal Realign planning Two parallel processes are Wasteful duplication and Merge legal procedures for all A task force led by DPLG and processes required to plan and complexity of processes stifles projects involving land and involving DLA, DEAT, DOH implement housing capacity of all actors and housing development in and Agriculture, and other development: Township reduces ability of poor people peri-urban areas into a single relevant departments Development and to acquire and develop land process Environmental Impact Assessment EIA involves too many steps Leads to a lengthy process Streamline the EIA process, and DEAT and discourages further decentralize approval of decision-making by officials most EIAs Difficulties of establishing a Lack of transparency of how Eliminate the "Less formal A task force led by DPLG and township: There are three to go about establishing a Township Establishment Act" involving DLA, DEAT, DOH options with implications township Adjust the two remaining Agriculture, and other rele- that are difficult to options to make room for mixed vant departments understand use villages, and simplify their requirements 33 Design a program for There is no program for Communities such as the Develop an integrated program A task force led by DPLG and peri-urban areas mixed land use and housing Muldersdrift Home that helps poor people acquire involving DLA, DEAT, DOH, for poor people who want to Foundation Trust cannot land, housing and agricultural Agriculture, and other engage in "multiple develop their own "villages" and other business support and relevant departments livelihoods". finance Inappropriate standards for Constitutional guarantees for Set more affordable standards Presidency and DPLG as part infrastructure basic service delivery are to promote mixed use of the NSDP interpreted to imply settlements in locations closer excessively costly to work and commerce infrastructure, which pushes low-income housing projects to remote locations, where land is cheap, so as to bring total costs down Inappropriate standards for Poor people are not able to Simplify housing standards DOH housing use housing subsidies to Include option for own- build their own houses construction by poor people (Continued) Main action areas Key issues Consequences Recommendations Responsibility Re-engineer program It is unclear who has overall Confusion and overlapping Clarify roles and responsibilities Intergovernmental relations implementation responsibility for land devel- mandates lead to lengthy and of different departments and unit of DPLG, together with opment and use planning: frustrating processes of land levels of government the task force mentioned DLA, DPLG, DEAT, or DOH? acquisition and development above for poor people Over-centralization of Expertise of lower levels of Using subsidiarity principle, Intergovernmental relations decision-making government not exploited decentralize decision-making to unit of DPLG, together with lowest possible level the task force mentioned above Money flows in "silos" Planning and financing of an Unify financing and access The National Treasury integrated village community requirements for the new pro- becomes an impossible task gram in peri-urban areas and for poor people other housing and land reform programs 34 Free up and build The complexities of Existing capacity is Develop easily understandable Everyone capacity requirements overwhelmed and guidelines and manuals disempowered, giving a false Implement the recommenda- appearance of lack of tions in this matrix capacity and unwillingness to help the poor Lack of legal and Uncertainties reduce Design and implement capacity DPLG with support from implementation capacity willingness and ability of building program for public- SAMDI and other relevant among public sector lawyers officials to take action sector lawyers and program departments and project managers managers Inadequate involvement of Poor communities requiring Beneficiaries should get grants Designers of the programs NGOs and other private sector land do not get adequate large enough to hire NGO and and the National Treasury providers to support groups support other expert support seeking mixed-use land Municipalities should outsource Municipalities, with support more support functions to NGOs from DPLG and others and other providers, and focus more on monitoring CHAPTER 5 Postscript O n March 7, 2007, the World Bank Country Director in Pretoria hosted a Review Meeting of the documentation produced to date on the MHTF project. This meeting was well attended by senior representatives of the Department of Land Affairs, Mogale City Municipality, GDACE, the World Heritage Site Management Authority and the Independent Development Trust. In addition, two World Bank peer reviewers shared their views on the report. At this meeting vigorous and constructive discussion resulted in widespread agree- ment with the suggestions and recommendations set out above. One difference, however, was that the Director General of Land Affairs requested that his department take respon- sibility for the proposed task force, rather than the DPLG, as he said that the focus of the task force's work is urban land policy, which falls squarely within the DLA's mandate as per the 1997 White Paper on Land Reform. 35 APPENDIX Development Facilitation Act, 67 of 1995 his law, known as the DFA, is significant because it represents the only post-1994 T legislation that expressly sets out to facilitate integrated land and housing projects. Somewhat unusually, however, the DFA operates "in parallel" to other laws. The applicant for planning permission to proceed with a land development project has the option either to follow the conventional route (that is, the town-planning ordinance) or to apply to the Development Tribunals specially set up in terms of the DFA. The act provides for nationally uniform procedures for the subdivision and develop- ment of land in both urban and rural areas. It includes within its ambit the provision and development of land for residential, small-scale farming and other needs and uses. "Land development" means any procedure aimed at changing the use of land for mainly resi- dential, industrial, business, small-scale farming, community or similar purposes. This includes the development of small-scale farming and land tenure matters.27 An integrated land and housing project will have to comply with a number of require- ments under the following headings. Principles There is a long list of principles to be complied with by any development contemplated under the act. Some key ones are: Promotetheintegrationofthesocial,economic,institutionalandphysicalaspects of land development. Promoteintegratedlanddevelopmentinruralandurbanareasinsupportofeachother. 27. Section 1. 37 38 World Bank Working Paper Promote the availability of residential and employment opportunities in close proximity to or integrated with each other. Optimizetheuseofexistingresourcesincludingthoserelatingtoagriculture,land, minerals, bulk infrastructure, roads, transportation and social facilities. Promoteadiversecombinationoflanduses,alsoatthelevelofindividualervenor subdivisions of land. Encourageenvironmentallysustainablelanddevelopmentpracticesandprocesses. Encourage active participation by members of communities affected by land development. Develop the skills and capacities of disadvantaged persons involved in land development. Promotetheestablishmentofviablecommunities. Policy,administrativepracticeandlawsshould: Promotesustainablelanddevelopmentattherequiredscale, Promotetheestablishmentofviablecommunities,and Ensurethesafeuseoflandbytakingintoconsiderationfactorssuchasgeological formations and hazardous undermined areas. Land development should result in security of tenure, and provide for the widest possible range of tenure alternatives, including individual and communal tenure. Development Applications A provincial land development tribunal has been established.28 Among its functions is to approve or refuse a land development application. The choice of whether to bring an appli- cation to this tribunal or use other channels lies with the applicant. An extraordinary power held by the development tribunal is the power to suspend the operation of a wide range of laws that might otherwise obstruct the implementation of the proposed project. In practice, tribunals are hesitant to invoke this power, especially as it may be subject to constitutional challenge, but it does have the positive effect of encour- aging applicants as well as decisionmakers to think more broadly than the options per- mitted by other legislation. Development Procedures Chapter VI of the DFA is probably the most relevant one for our purposes. It is titled "Land development procedures including procedures relating to the development of small-scale farming." The development application shall be subject to the decision of the tribunal under this section. The final decision can provide for conditions of development among them:29 Onsubdivisionofthelandinthelanddevelopmentarea. Ontheownershipandadministrationofthesettlementofpersonsonsuchland. 28. Section 16. In addition section 24 provides for the development appeal tribunal, a provincial body. Appeals from the Development Tribunal lie in the development appeals tribunal. 29. Section 51. In Search Of Land and Housing in the New South Africa 39 On the question whether the use of land in the land development area is to be regulated by: TheConservationofAgriculturalResourcesAct(1983), Provisionsrelatingtotheuseoflandoutsidelocalgovernmentareaswhichhave been prescribed generally for that purpose, or Specific provisions relating to special or strategic projects, which have been prescribed. The question whether the provisions of any of the following shall apply in respect of the land development area in question: Anylawonphysicalplanning, Anylawrequiringtheapprovalofanauthorityforthesubdivisionofland, Sections 9A and 11 of the Advertising on Roads and Ribbon Development Act (1940), and/or Section12andtheNationalRoadsAct(1971). Theenvironmentorenvironmentalevaluations. The manner in which members of any community residing in a settlement shall be consulted during the process of land development whenever land development takes the form of the upgrading of an existing settlement. Applicants can, however, be exempted from the development conditions as prescribed above if the area is: Already settled by persons and is intended to be upgraded into a fully established land development area over a period of time, or Ifitisintendedtobesettledbypersonsonanurgentbasispriortocompletingthe establishment of a land development area in that area, with the intention that such area shall be upgraded over a period of time into a fully established land develop- ment area.30 In arriving at such decision, the tribunal will take into account:31 Thefeasibilityhousingpersonsintemporarybuildingserectedbythemselvesinthe area; Thefeasibilityofprovidingrudimentaryservicesintheareaconcernedandofthe upgrading of such services over a period of time; Thefeasibilityofhousingpersonsintemporarybuildingserectedbythemselvesin the area; The suitability of the area for small-scale farming, taking into account its natural resources and location in relation to agricultural facilities; Thefeasibilityofprovidingoccupantsoftheareawithappropriatesecurityofland tenure; Thefeasibilityoferectingpermanentdwellingsoveraperiodoftime; 30. Section 48. 31. Section 57 (4). 40 World Bank Working Paper The feasibility of establishing an appropriate local government body or including the area within the local government area of such a body and of providing municipal services to the area; The possibility of persons settling in the area being able to acquire sites which are affordable to them, taking into account their likely income and other means of finance, including finance provided by the state; Thefeasibilityoftheareabeingfullyestablishedasalanddevelopmentareaovera period of time; Therightsofanypersoninorinrespectoftheareaand,ifnecessary,thefeasibility of such area or rights being expropriated or otherwise acquired for the purpose of establishing a land development area; and Theenvironmentalsustainabilityofdevelopingorpermittingsmall-scalefarming in the area. Eco-Audit Environmental Benefits Statement The World Bank is committed to preserving Endangered Forests and natural resources. We print World Bank Working Papers and Country Studies on 100 percent postconsumer recy- cled paper, processed chlorine free. The World Bank has formally agreed to follow the rec- ommended standards for paper usage set by Green Press Initiative--a nonprofit program supporting publishers in using fiber that is not sourced from Endangered Forests. For more information, visit www.greenpressinitiative.org. In 2007, the printing of these books on recycled paper saved the following: Trees* Solid Waste Water Net Greenhouse Gases Total Energy 264 12,419 96,126 23,289 184 mil. * Pounds Gallons Pounds CO2Equivalent BTUs 40' in height and 6­8" in diameter In Search of Land and Housing in the New South Africa is part of the World Bank Working Paper series. These papers are published to communicate the results of the Bank's ongoing research and to stimulate public discussion. This study outlines the difficulties poor communities face in access- ing peri-urban land in South Africa that could have implications and lessons for similar communities in other countries facing spa- tial segregation issues. The study focused on one community, com- posed largely of laid-off farm workers that wanted to buy their own farm in a peri-urban area west of Johannesburg. Their dream was to establish a mixed-use settlement. They wanted to call the village Ethembalethu--"Our Hope." About 250 families started their own association and savings scheme to make their dream a reality. By 1997, they had saved enough money to make their first purchase offer. A decade later, the community's dream is still not a reality. The families have faced numerous obstacles: two cancelled sale agree- ments, wrongful arrest, being sued in court, an out-of-court settle- ment for which community members were paid to not move into the white neighborhood, and large sums of their own money spent on consultants and environmental impact studies. In an agreement with the Mogale City Municipality, where the land is located, the community now has at least a confirmed right to occupy the land. But it does not yet legally own the land, and is still trying to get permission to build on and work the land. The case of Ethembalethu is not unique. Millions of black South Africans live in the peri-urban areas. Yet, government programs, development planning and environmental regulations, and the cur- rent land and housing markets do not support realization of their aspirations to become homeowners on sites of their choice. World Bank Working Papers are available individually or on stand- ing order. Also available online through the World Bank e-Library (www.worldbank.org/elibrary). ISBN 978-0-8213-7373-6 THE WORLD BANK 1818 H Street, NW Washington, DC 20433 USA Telephone: 202 473-1000 Internet: www.worldbank.org SKU 17373 E-mail: feedback@worldbank.org