80958 CITIES The Case for WITHOUT Incremental Housing SLUMS Patrick Wakely & Elizabeth Riley Cities Alliance Policy Research and Working Papers Series No. 1  |  June 2011 About Cities Alliance Policy Research and Working Papers Series Publications in this series capture rigorous research and analyses of trends and innovations in urban develop- ment. These are designed to inform as well as to stimulate focused debate and discussions on the issues among all city stakeholders, from national and local policy makers to federations of the urban poor, to contribute to the overarching goal of enabling the transformative role of cities in poverty reduction. Papers in the series are normally subject to a robust peer review process by recognised experts in the fields of discussion, and are produced to the highest editorial standards. They are normally published on-line but few available hard cop- ies can be requested for by sending an email to: info@citiesalliance.org. Series Editorial Director: William Cobbett Managing Editor: Chii Akporji © The Cities Alliance, 2011 1818 H Street, NW, Washington, D.C. 20433, U.S.A. www.citiesalliance.org All rights reserved. The material in this publication is copyrighted. Requests for permission to reproduce whole or portions of it should be directed to the Cities Alliance Secretariat located at the above address. The Cities Alliance en- courages the active dissemination of its knowledge and learning. Permission to disseminate will normally be granted promptly and, when reproduction is for noncommercial purposes, without asking for a fee. Cover Photo: Policarpa Salavarietta, Bogota, 1976. © DPU Associates The architectural styles and standard of construction of each floor of this house in Barrio Policarpa Salavarrieta in Bogotá, Colombia clearly reflect the growing fortunes and changing preferences of its owner over the twenty-year period of its incremental construction. Right from the start he/she built a second entrance door with the ambi- tion of being able to sublet part of the house sometime in the future in order to supplement his/her income. The neighbours on either side have chosen not to invest so much in their dwellings, or they have been unable to do so. Design: The Word Express. Contents A b b r e v i at i o n s a n d A c r o n y m s  v Acknowledgementsvii Forewordix Executive Summary xi SECTION 1   T h e i s s u e s  1 1.1 Informal urban housing processes 1 1.1.1  Acquiring land 2 1.1.2 Construction 3 1.1.3  Infrastructure and services 3 1.1.4  Secure title 3 1.2 Strategic approaches to low-income urban housing 4 1.2.1  Public social housing 4 1.2.2  Slum upgrading 5 1.2.3  Sites and services 5 SECTION 2   I n c r e m e n ta l H o u s i n g S t r at e g i e s in Context 7 2.1 The emergence of policies to support incremental housing 7 2.2 Range of sites and services projects 8 2.3  Perceived problems 9 2.4 The shift away from support for incremental housing 12 2.5 Support for incremental housing now (2010) 14 iii T H E C A S E F O R I N C R E M E N TA L H O U S I N G SECTION 3   T h e Ca s e f o r I n c r e m e n ta l H o u s i n g S t r at e g i e s  1 7 3.1  The numbers case 17 3.2  The financial case 18 3.3 The urban management case 19 3.4 The urban development case 21 3.5  The governance case 23 3.6 The social and economic development case 23 SECTION 4   C o m p o n e n t s o f I n c r e m e n ta l H o u s i n g S t r at e g i e s  29 4.1  Land and location 29 4.1.1  Cost and location 29 4.1.2 Land acquisition and law reform 31 4.1.3  Land tenure 32 4.2 Finance 33 4.3 Infrastructure and services 35 4.4  The private sector 37 4.5  Beneficiary selection 38 4.6 Site planning and building controls and supports 40 4.7 Community organisation and asset management 41 4.8  Strategic planning 42 SECTION 5   CONCLUSIONS , C A PA CITY BUILDING & THE W AY F ORW A RD 5.1  Capacity building 46 5.1.1  Human resource development 46 5.1.2  Organisational development 46 5.1.3  Institutional development 47 5.2 Priorities 47 5.3  In conclusion 48 References49 Bibliography55 iv Abbreviations and Acronyms ACHR Asian Coalition for Housing Rights Asian Coalition for Community Action (ACCA) ADB Asian Development Bank CHF Cooperative Housing Foundation International DBP Development Bank of the Philippines DfiD Department of Foreign International Development DPU Development Planning Unit, University College, London GIS Geographic Information Systems GTZ (now GIZ) Deutsche Gesellschaft für Internationale Zusammenarbeit HUDCC Housing and Urban Development Coordinating Council, the Philippines IADB Inter American Development Bank IDRC International Development Research Center, Canada IHC International Housing Coalition IIED International Institute for Environment and Development MIT Massachusetts Institute of Technology NGOs nongovernmental organizations NSDF National Slum Dwellers Federation SDI Shack/Slum Dwellers international SiDA Swedish International Development Agency SPARC Society for the Promotion of Area Resource Centres S&S Sites and Services UN-Habitat United Nations Human Settlements Programme UNDP United Nations Development Program USAID United States Agency for International Development WB World Bank v vi  |  Section vi Acknowledgements The need to re-make the case for public sector sup- ‘what went wrong’ with sites-and-services and port for incremental housing processes in the con- slum upgrading in the 1980s as a starting point text of current urban policy initiatives and prac- for the study. Julian Baskin and Judy Baker, Lead tices was conceived by Billy Cobbett, Manager of Economist in the Finance, Economics and Urban the Cities Alliance, who commissioned the study Department of the World Bank, made valuable in 2009. The scope and structure was developed in comments on early drafts of the study particularly discussions with Julian Baskin and Andrea Merrick, by providing experienced insights into the role and Cities Alliance’ Senior Urban Specialist and Urban potential of the formal private sector in incremen- Specialist respectively. These discussions served to tal housing processes. Chii Akporji, Cities Alliance’ confirm the importance of reviewing the ways in Communications Specialist managed the design which affordable housing is procured informally and editing of the final publication with great by low-income urban households, and of revisiting professionalism. vii viii  |  Section Foreword The study makes an irrefutable six-fold case for re- subsidiarity; assigning actions to the lowest levels at alizing the full potential of a most successful way which they can be competently carried out; the va- of building: ‘informal incremental urban settle- riety of forms required to meet motivating priorities ment’. It describes procedures by which millions of and the economic use of human and material re- low-income people develop their homes and neigh- sources that follow. As these principles are universal bourhoods, often to surprisingly high standards as they raise a very important issue: obviously they ap- the series of excellent photographs prove. Half a ply to incremental housing in low-income contexts, century ago virtually all squatter settlements were so what can be learned from that experience for seen as slums and city cancers. Few saw the differ- adapting that and other community-building ways ence between “Slums of Hope” that have the poten- of creating homes and neighbourhoods in the same tial for development, and “Slums of Despair” which and in other contexts? did not. Failures of eradication and resettlement policies strengthened the view that many informal It’s this last part of the question that drew me back settlements solved more problems than they cre- to England, my own country where I and other as- ated. The emergent policies described in this publi- sociates have had only very modest success in our cation followed international agencies’ recognition efforts to work with community groups. We learned of this fact, reinforced by some pioneering national the hard way that a society with such a deeply eroded government agencies. community base poses greater and deeper challenges than those that have strong reasons to cooperate. The facts and the potential for incremental housing are stated clearly here but they have yet to be learned There can be few better starting-points for learning by many. It is hugely encouraging that the Cities from incremental housing in its ‘informal’ and inte- Alliance is publishing ‘The Case for Incremental grated states. Housing’. It is a ‘must read’ for all concerned with sustainable development and justice, both by those struggling for their rights to do what they are able to do for themselves as well as by those who support them. Translations are essential. John F. C. Turner The way in which the clearly stated series of incre- mental housing strategies are introduced and set out Author of the seminal paper: ‘Uncontrolled Urban identifies the key principles evident in the practices: Settlement: Problems and Policies’ prepared for ix T H E C A S E F O R I N C R E M E N TA L H O U S I N G the United Nations Centre for Housing, Building the Government of the Unites States of America at the and Planning and presented at the Inter-Regional University of Pittsburgh, Pittsburgh, Pennsylvania, Seminar on Development Policies and Planning in USA, 24 October–7 November 1966. This paper Relation to Urbanization, organised by the United was also reproduced in many important subsequent Nations Bureau of Technical Assistance Operations publications. and the Bureau of Social Affairs, in co-operation with x Executive Summary The issues Incremental housing strategies in context Informal settlements accommodate more than 50 percent of the population of many cities. They are Many S&S projects were implemented in the 1970s characterised by flexible, responsive, and affordable and ’80s. They went out of fashion for a variety of housing processes that enable families to extend reasons, however, significant among which was that and improve their dwellings over time. At the same they were not given enough time to mature before time, their legal status is usually insecure, they are being evaluated. The consolidation of low-income under-serviced by urban infrastructure, are often houses and neighbourhoods is a slow process. unhealthy living environments, and in some cases Projects were often sited on the urban fringes where are physically unsafe. land was cheap but isolated from centres of employ- ment and urban social services and networks. The Government efforts to address these problems cost of services, planning standards, and mandatory through the construction of subsidised completed building controls rendered many projects unafford- dwellings for low-income groups are seriously lim- able to their target groups. ited by cost and management capacity. By com- parison, slum-upgrading programmes can provide The case for supporting security of tenure, adequate infrastructure, and incremental housing strategies local management capabilities to households and communities in existing informal settlements, at a With a better understanding of urban poverty and fraction of the cost. In addition, sites and services new approaches to urban planning and manage- (S&S) programmes can redress the growth of new ment, there is a strong case for governments to initi- informal settlements and the proliferation of slums ate and support incremental housing strategies as a (which are growing by 5 percent per year in many major component of integrated urban development. cities) by providing secure access to land and ser- The case rests on six major arguments: vices and enabling households to construct their dwellings incrementally as their resources allow at  The numbers case. By engaging householders in a significantly lower cost than conventional public the production and management of their own housing programmes. dwellings and neighbourhoods, far more legal, xi EXECUTIVE SUMMARY safe, and healthy dwellings affordable to low- public sector support in seven key areas of inter- income groups can be procured than by con- vention: (i) land and location, (ii) finance, (iii) in- ventional approaches. frastructure and services, (iv) beneficiary selection,  The financial case. By providing security of ten- (v) site planning and building controls and sup- ure and access to services, even poor households ports, (vi) community organisation and asset man- are able to invest in housing and neighbour- agement, and (vii) citywide strategic planning. An hood development through saving and borrow- integrated housing policy framework must address ing, thereby sharing the cost of urban develop- each of these components in conjunction with the ment with the government. others.  The urban management case. By recognising the most effective levels of decision-making Conclusions, capacity building & the way and delegating the authority that is required forward for incremental housing development, part- nerships that enhance the efficiency of urban About half the population of the developing world management and the administration of urban live and work in towns and cities and a third of them services can be built. (830 million) in informal settlements or slums.  The urban development case. By planning areas Though there are many ‘slums of despair’—seem- of legitimate low-income housing develop- ingly hopeless neighbourhoods of poverty and en- ment as part of an integrated urban develop- vironmental degradation—the majority are ‘settle- ment strategy, governments can set strategic ments of hope’—informal neighbourhoods and priorities for an entire urban area rather than communities in the process of building their cities resorting to ad hoc measures. through their own endeavours and ingenuity. They  The governance case. By engaging households demonstrate a process that has been shown to be and community leaders in the incremental both effective and efficient in terms of its respon- development of their housing and neighbour- siveness to their occupants’ fluctuating needs and hoods, a system of good governance that helps fortunes. However, they are often constrained by a ensure transparency and accountability in de- lack of official or recognised supports that would cision-making can be created. extend the effectiveness and efficiency of incremen-  The social and economic development case. By tal housing processes for the development of the encouraging cooperation through incremental city as a whole. As pointed out above, the starting development, local communities are built and point for this is the understanding of the principle strengthened. Furthermore, by creating job of subsidiarity and a political will to devolve author- opportunities through the provision of train- ity down to the level of organised urban communi- ing and technical support, household incomes ties, coupled with investment in innovative capacity can be increased. building. Components of incremental housing strategies National or citywide incremental housing strat- * egies entail the adoption of new approaches to UN-Habitat, 2003 (Second Edition, 2010) xii SECTION 1 The Issues This report has two purposes: to demonstrate that governments and international development institutions should support informal incremental housing processes and to explore how best they can do this. In supporting the incremental building and improvement of housing by and for low-income groups, governments must recognise that the majority of low-income households gradually erect and change their homes as their needs evolve and their resources allow. Informal, incremental housing processes are household driven, enabling low-income people to acquire, extend, improve, and service their dwellings and neighbourhoods over time.1 Virtually all permanent and serviced housing is 1.1  Informal urban housing procured as an incremental process that takes place processes over relatively long periods of time. Only a minute segment of any society—that is, the very wealthy— In general, housing procurement processes for low- has the resources to purchase outright or construct income households in the informal sector take place their dwellings as a one-off event. Upper- and mid- in the opposite order of those in the formal sector dle-income households with regular incomes and (figure 1).2 collateral have access to long-term credit—hous- ing loans and mortgages—that may take between There are many variations to each of these stages 15 and 30 years to redeem. Households with low or in the informal process of settlement development. irregular incomes and no access to formally recog- That said, the stages may be usefully characterised nised collateral construct minimal basic dwellings, and summarised as follows: which they extend and improve as resources be- come available and as the need for bigger or bet- ter structures becomes a priority. This process of extension and modification can take decades—or 1 CHF, 2004. may be never ending. 2 McLeod and Mullard, 2006. 1 T H E C A S E F O R I N C R E M E N TA L H O U S I N G the sale or rental of their land, are less likely to evict Sequence of procurement processes for Figure 1 them than private and commercial landowners. In formal and informal housing many cases, informally organised squatting is done 1 Move onto land 4 in plots with some semblance of regularity to allow ßßßß àààà Informal Formal 2 Build house 3 for access ways and the eventual installation of water 3 Install infrastructure 2 and drainage lines. The majority of squatter settle- 4 Obtain title to land 1 ments, however, particularly in South Asian and Source: McLeod and Mullard (2006). African cities, have been irregularly developed with high densities, making servicing difficult and creat- ing serious obstacles to both access and escape in the 1.1.1  Acquiring land case of fire, flooding, or other emergencies. The informal settlement of urban land falls into two broad categories: squatting and unregistered land The other form of informal land settlement—un- sales. registered land sales—differs significantly from squatting. It is based on an exchange of money at Squatting occurs when households or groups of fam- a price that is agreed on by both the landowner ilies settle on vacant land by constructing rudimen- and the purchaser. What renders this land afford- tary shelters. If they are not immediately evicted, they able to the lowest-income groups is that the land gradually consolidate their dwellings. Encouraged by use is not legally recognised. This is either because the success of the original squatters, other households it is unsafe or inappropriate for habitation or oth- join them. The settlements usually exhibit a high de- erwise in contravention of official regulations, or gree of community cohesion borne out of solidarity because the vendor lacks a recognised title to the during the illegal process of squatting. In some situ- land that he is selling. Furthermore, informal de- ations, notably in some Latin American countries in velopers, recognising the potential risk that such the 1960s and ’70s, squatters settled on large tracts transactions entail, tend to sell their land in small of urban and peri-urban land literally overnight, re- plots that are affordable to lower-income groups. lying on the security of numbers to make eviction The process of parcelling land varies, ranging from technically and politically difficult, if not impossible. direct sale by the owners of peri-urban agricultural These large-scale events are known as land invasions, land, to the wholesaling of tracts of land to infor- and they often have politically motivated organisa- mal developers, who then subdivide it and retail tional support. Squatting occurs on any vacant or it to households, sometimes even providing credit undeveloped land within urban areas or on the pe- facilities. A common characteristic of parcelling is riphery of towns and cities. The land may be suit- the high density of saleable plots, which allows only able for development, or it may have been left vacant minimal provision for access ways and little, if any, because it is not safe, such as steep slopes subject to for service buildings (such as schools and clinics) or erosion; land prone to flooding; or road, rail, river, recreational space, though the provision of public and coastal reserves.3 Although squatting occurs on infrastructure and services at a later date is gener- both public and privately owned land, in most situ- ally anticipated, if not informally “finessed” by the ations squatters choose government land. They feel developers in the first place. that public institutions, which have a weaker sense of ownership or opportunity to gain financially from 3 UN-Habitat, 2003. 2 S ECT ION 1: T H E ISSUES 1.1.2  Construction practice is especially dangerous in seismic or un- The building of houses in informal settlements is stable areas such as slopes, river banks, and uncom- invariably an incremental process. Households start pacted reclaimed land. with the most basic and affordable shelters. They then invest in the extension and improvement of 1.1.3  Infrastructure and services their dwellings as their circumstances allow and in Many new informal settlements have only the most accordance with their priorities for the investment of precarious access to water, sewerage, and solid waste their time, energy, and resources and their assessment disposal, and many illegally hijack power. Utility of the risk of eviction. In many circumstances, the agencies and service departments are torn between development of a permanent house is a lower prior- their duty to provide adequate services to all citizens ity than investment in other assets, such as a business and the implications of giving de facto recognition to or a child’s education. Extension or improvement of illegal settlements by providing them with facilities. a house may be confined to times of good fortune (In many cities this dilemma has been inadvertently when a surplus income is available and may be con- eased by the process of privatising service delivery, strained by hard times of unemployment or sickness. since private sector suppliers tend to have little inter- est in the legality of their customers as long as service A universal constraint to investment in informal set- charges are paid). Yet over time—and with political tlements is the insecurity of title to land and prop- pressure—infrastructure is upgraded and services erty. The threat of eviction or demolition accounts are provided: water lines are laid, sewerage disposal for the creation of slum conditions in urban areas. systems are installed, access to power is provided, At the same time, the low quality of housing and ser- streets are paved and lit, and in some cases schools vice facilities in informal settlements keeps the cost of and clinics are built. However, these interventions rental housing affordable for the poorest families. The cost substantially more and inevitably cause greater extent of buildings developed by landlords who rent social and physical disruption than if they had been rooms to poor households in informal settlements is provided at the outset of the settlement process. often under-estimated. The rental housing market is a major source of income, not only for large-scale land- 1.1.4  Secure title lords who may own many slum properties, but also Early slum-upgrading projects tended to be con- for individual subsistence landlords and households fined to the supply, extension, and improvement that rent rooms to supplement their incomes. of physical infrastructure in under-serviced settle- ments. Little or no attention was paid to commu- The standard of construction in informal settle- nity development, engagement in local governance, ments if often cause for concern. Although frequent- or the legal status of the households’ title to their ly this is due to the appearance of non-permanent land and property. Increasingly, however, security and second-hand building materials rather than a of tenure has been recognised as crucial to house- real danger of structural collapse, there clearly is a holders’ sense of ownership and stewardship of their need for technical advice. Local skilled artisans and neighbourhood assets. Unfortunately, where land non-governmental organisations (NGOs) can pro- has been squatted on or sold for development that vide technical advice in some cases, but in others, as does not conform to its statutory use category or when structures are built for rent rather than owner the prevailing planning standards (such as plot sizes occupation, dangerous buildings are erected. This and road widths), formalisation of title in the name 3 T H E C A S E F O R I N C R E M E N TA L H O U S I N G of the incumbent households or even the granting groups, excluded from it by poverty or lack of secure of collective title to a community is difficult, par- collateral, to join it. To do so, the public sector con- ticularly when the original occupation took place structs dwellings and applies subsidies to make them several years (or generations) back. There are many affordable for low-income households to rent or buy examples of complex legal arrangements for the with long-term repayment terms. transfer of title to land involving the claims of long- term absentee landowners, the sharing of property While this approach generally provides good-qual- between de jure owners and de facto users, the estab- ity, fully serviced dwellings of permanent construc- lishment of fair compensation for land that has lost tion, it has many drawbacks, the most prominent its value by being squatted on, and so on. of which is its high cost to the state. For instance, in 1972 during the heyday of conventional public hous- In summary, the incremental construction process ing construction, the Tamil Nadu Slum Clearance of the urban informal sector, while generally be- Board in India set a target to rid the city of Madras ing commercially viable (affordable) and socially (Chennai) of its slums over the following five years. responsive (flexible), is beset by inefficiencies and By 1977 it had built 17,450 subsidised housing units. insecurity that stem from its informality, which is Impressive though this was, a further 200,000 house- generally equated with illegality. holds (1 million people) were left in appalling and unserviced slum conditions. In the same year, the Karachi Development Authority set a target to con- 1.2  Strategic approaches to struct 3,000 minimal apartments to re-house slum low-income urban housing dwellers. By 1980, when the project was abandoned, only 800 units had been constructed.4 More recently, Governments and international agencies have ad- the South African government set a target to eradi- opted three broad approaches to ameliorating the cate slums by 2014 by providing subsidised serviced deficiencies in the informal incremental housing housing for the country’s lowest-income groups. procurement process. Each approach ultimately That ambitious target resulted in the doubling of the aims to extend the formal housing supply to em- country’s housing budget from R5 billion to R10 bil- brace households with lower incomes or with less lion (US$0.7 billion to US$1.4 billion) between 2004 collateral than is conventionally acceptable in the and 2009—an increase of nearly 20 percent per year. private housing market. Although none of them Despite the enormous investment, by 2009 only half has met with outstanding success or universal ac- of the required output of 500,000 new subsidised claim, and some need further development, each housing units per year had been achieved.5 has merits that should not be ignored. Public housing agencies throughout the rapidly 1.2.1  Public social housing growing cities of developing countries have had sim- The first approach that dominated urban public ilar results. Governments and municipal administra- housing policies in the 1960s—and still exists to some tions of rapidly growing cities with large and grow- extent in most countries—is often referred to as con- ing housing deficits simply cannot afford to subsidise ventional or social housing. The approach is based on the idea that the government must intervene in 4 Wakely and Aliani, 1996. the formal housing market to enable lower-income 5 Ndaba, 2008. 4 S ECT ION 1: T H E ISSUES decent housing for all of the low-income urban pop- decent housing for the growing low-income popula- ulation. Furthermore, their efforts are often left open tions of developing cities. to opportunistic abuse in the open market. 1.2.3  Sites and services 1.2.2  Slum upgrading The third approach relies on urban sites and services The second approach to low-income urban hous- (S&S) programmes, which combine elements of the ing is slum upgrading, which international agen- other two approaches. S&S programmes abounded cies began to promote in the 1970s. Slum upgrading in the 1970s and ’80s but have gone out of fashion encompasses public sector support to households for reasons that are discussed in Section 2.3 below. that have been denied access to the formal land and Conceptually the S&S movement sought to mini- property markets and have instead taken the initia- mise the public costs and subsidies required by con- tive to house themselves informally (illegally) either ventional social housing programmes by providing individually or through unregulated developers, as only those components that individual households outlined in Section 1. 2.1 above. The upgrading pro- could not easily procure for themselves—land, infra- cess can entail the granting of secure title to land and structure, and services. The low-income homeown- property to encourage household and community ers to whom serviced plots were allocated had to use investment; the installation or extension of public their own resources, labour, and time to build their infrastructure (such as water, sanitation, and pow- houses. The interpretation of publicly provided ser- er); the provision of services (such as schools, health vices varied widely, ranging from pegs in the ground facilities, and recreation space); and the develop- demarcating plots with access to communal public ment of effective local governance and management water points and latrines, to fully serviced starter mechanisms. homes that the beneficiaries could extend. Slum-upgrading processes have successfully reached The S&S programmes were beset by many problems, many low-income households and stimulated invest- including the imposition of standards of design and ment in the development of regular low- to middle- construction that were unaffordable, speculation income neighbourhoods. However, even where slum and gentrification, high rates of default on loans upgrading has been effective, it has not necessarily and rents, and social problems such as unemploy- been efficient, as most informal settlements have not ment and exclusion. Many of those problems arose been planned with the provision of access and ser- because S&S schemes were located on cheap land on vices in mind, which often makes the installation of the urban periphery, far from jobs, transport, and infrastructure both costly and environmentally dis- social facilities. Poor financial planning and mis- ruptive. Furthermore, many settlements are built on management also caused problems. Nevertheless, peri-urban land that has been informally subdivided the concept of state-supported incremental housing by its former agricultural owners or on inner-city sites remains valid. that have been spontaneously squatted by their oc- cupants, and therefore are not located in accordance The S&S approach has the potential to build upon with the optimal distribution of land uses for the city the affordable and flexible incremental housing con- as a whole. In short, while the upgrading of informal struction and improvement practices of low-income settlements addresses existing housing inadequacies, groups within the framework of planned interven- it cannot accommodate the provision of affordable tions that boost the supply of housing at a cost that 5 T H E C A S E F O R I N C R E M E N TA L H O U S I N G is acceptable to both the governments and the poor. supporting responsive incremental housing construc- With a better understanding of urban land markets, tion and neighbourhood development—have dem- housing finance, urban planning requirements, gov- onstrated many innovative approaches to officially ernance, and poverty, there is considerable scope recognised incremental housing strategies. Providing for contemporary S&S schemes to contribute to the informal neighbourhoods with water supply and housing of low-income people in suitable locations sewerage networks, adequate access ways, schools, at affordable costs and to socially and environmen- and clinics after they have been developed—although tally acceptable standards. technically cumbersome and socially disruptive—has been shown to be a more economical use of public re- In conclusion, liberal democratic governments in sources than the construction of conventional public rapidly urbanising countries (except Singapore and housing. The best S&S projects ensure that adequate Hong Kong in the 1960 and ’70s) have been unable facilities are available and that households have secure to construct conventional subsidised housing that title to land within a framework of strategic urban is affordable to the lowest-income groups in signifi- development plans before incremental construction cant numbers. Regardless, many governments con- begins. Nevertheless, the success of the S&S approach tinue to pursue the political rhetoric of such policies. hinges upon providing the right support for incre- Successful slum-upgrading programmes that provide mental housing processes without over-institution- security of tenure to land and acceptable standards alising and formalising them to the extent that they of service provision—even while respecting and become inflexible and restrictive. 6 SECTION 2 Incremental Housing Strategies in Context 2.1  The emergence of policies to support incremental housing Government projects to assist the poor in self-building have been common since the mid- twentieth century.6 In Nigeria, for example, the colonial government acquired large tracts of land and laid out and installed basic infrastructure before allocating the serviced plots to individuals and corporate bodies for development.7 In the mid-1940s, the government of Mexico authorised private developers to lay out vast residential housing estates (without imposing planning requirements and development controls) and sold plots at costs affordable to very low-income households. For example, Ciudad Netzahuacoyotl, on the outskirts of Mexico City, was developed in 1958 with minimal services, had a population of over half a million by the 1970s, and is now fully developed with a population of over 2 million.8 Early state-supported incremental housing initiatives strategies of poor people in Peru. Charles Abrams tended to be piecemeal, one-off projects. Official also recognised the effectiveness of informal housing programmes for the construction of subsidised com- processes and suggested state support for incremen- pleted housing dominated public housing policies. tal housing projects as early as 1964.11 Encouraging At the same time, urban squatter settlements and and supporting the poor’s approaches to meeting slums grew rapidly on illegally invaded and subdivid- their own housing needs began to be recognised as a ed land, but those highly visible, informal, and incre- way forward for policy makers, who were struggling mental processes were in no way seen as a legitimate to build sufficient public housing units at affordable means to provide housing. It was only in the 1960s prices to shelter rapidly growing low-income urban that this view began to change. Largely due to the re- populations. search and writings of a small group of academics, the incremental approach began to be recognised as 6 UN-Habitat, 2003. a logical and effective strategy employed by the poor 7 Adedeji and Olufemi, 2004. 8 Gattoni, 2009. to house themselves. Notably, John F. C. Turner9 and 9 Turner, 1967. William Mangin10 looked at and analysed the high- 10 Mangin, 1967. ly organised and effective informal urban housing 11 Abrams, 1964. 7 T H E C A S E F O R I N C R E M E N TA L H O U S I N G The origin of state involvement in incremental hous- the construction of core houses) meant lower costs, ing strategies was therefore the reluctant acceptance which allowed S&S projects to target low-income that informal housing delivery systems performed groups, including those already in squatter settle- much better than public attempts to build dwell- ments.13 However, many of the earlier state-support- ings in a number of respects: they were affordable ed incremental housing initiatives tended to impose without recourse to public subsidy, they were flexible relatively high standards that were unaffordable to and responsive to the changing needs and unstable the lowest-income groups, many of whom sold up- fortunes of poor urban families, they were self-man- market to middle-income groups. In response, the aged and made few demands on hard-pressed public incremental housing initiatives that emerged in the administrations, and they met the needs of the rap- 1980s assumed that the poor could only afford to idly growing urban populations of developing towns spend 15 to 20 percent of their disposable income and cities.12 As a result of this acceptance, the early on housing and infrastructure costs. Projects there- 1970s saw a conceptual shift from slum clearance to fore had to keep standards to a minimum, striking a slum upgrading. The destruction of dwellings was balance between affordability and the negative per- replaced with the recognition of the value of the in- ception of the “officially sanctioned construction of formal housing that constituted a major proportion new slums.”14 of many cities’ total housing stock. Slum-upgrading projects were initially confined to the provision of physical infrastructure and service buildings. Only 2.2  Range of sites and services later was the importance of security of tenure and projects the development of a collective sense of ownership and community asset management properly under- In the early 1970s, multilateral donors and banks, stood. The upgrading process almost invariably en- most notably the World Bank, began to support tailed the demolition of some dwellings to make way incremental housing by launching a series of S&S for infrastructure runs, sites for schools and clinics, projects (and, soon after, squatter-settlement-up- and the removal of dangerously located dwellings. grading programmes). Other significant multilater- Even in the best projects, land had to be made avail- al lenders and donors in the housing sector includ- able for re-housing adjacent to the upgraded area ed the Inter-American Development Bank (IDB), in order to minimise the disruption of community which funded S&S projects in Central America, cohesion. As serviced land was made available to and the Asian Development Bank (ADB), which re-located families, S&S projects began to develop funded projects in Asia. The United States Agency independent of the slum-upgrading programmes, for International Development (USAID) was also though many remained linked. heavily involved in S&S but later shifted to funding slum-upgrading projects in Africa and the Middle S&S projects stimulated the self-building processes East.15 that were evident in squatter settlements in two ways. First, they gave residents control over their houses. More importantly, they gave the state control over 12 Van der Linden, 1992. the location of settlements and the standards of de- 13 Cohen, 2007. velopment within them. Lower standards (in terms 14 Cohen, 2007. of infrastructure, residential densities, services, and 15 IHC, 2008. 8 S ECTI O N 2 : I NCREMENTAL H OUSING ST RAT EGIES IN CONT EXT By 1983 the World Bank had supported more than 2.3  Perceived problems 70 S&S projects.16 The key components of each project were a plot of land, infrastructure (roads, Many of the early S&S projects were judged to be water supply, drainage, electricity, or a sewage net- problematic. This was partly because the length of work), and sometimes part of the superstructure of time for the projects to mature was not understood a house. Government agencies were thus involved in and so they were evaluated on the basis of unre- acquiring land, dividing it into plots (leaving land alistic short-term objectives. But there were also available for the construction of service and ame- very real problems on the ground. The failure of nity buildings), providing basic infrastructure, and the first World Bank-funded S&S project, Parcelles setting up the financial mechanisms to sell or lease Assainies in Dakar, dampened the initial enthu- the land and its services to the intended beneficia- siasm of donors and governments for S&S and ries. The beneficiaries were responsible for building pushed them into the upgrading of existing slums the house (sometimes with the exception of a pre- and the wider-focus, “softer” approach of urban constructed core) using their own resources, such as policy and management. informal finance or family and community labour. Occasionally, the project provided loans to cover at Some projects did not reach lower-income levels as least part of the cost of construction. intended, for a variety of reasons. First, corruption in the beneficiary selection process allowed upper- Some S&S projects extended the provision of servic- income speculators to benefit from subsidised de- es even further. For example, some included a “wet veloped land. The fragility of household budgets services” core on each plot with connections for wa- and the inability of households to meet the cost of ter, drainage, sewerage, and electricity, while others construction on top of paying the basic land and had a full sanitary core consisting of a bathroom/ infrastructure costs were also under-estimated. toilet and a kitchen to foster household and commu- Mandatory house designs and construction stan- nity health. Still others provided beneficiaries with dards often required the use of house plans and a complete room or a plinth on which to build.17 high-quality building materials that were not af- Projects with more extensive provisions of a super- fordable. Furthermore, projects were located on structure resulted in more expensive units and there- the urban fringes, where land prices were low but fore either required greater subsidies or demanded a transport costs to centres of employment and social higher income level from the project beneficiaries. facilities were high, if available at all. Finally, ben- Similarly, plot sizes ranged enormously, from 25 eficiaries could not meet deadlines for construction square metres of bare land with ungraded access because they were either unable to raise funds or roads, communal water points, and community la- had higher investment priorities. trines in the big World Bank schemes of Chennai (Madras) in India; to 250 square metre plots with A World Bank evaluation of S&S projects in El bathrooms, kitchens, and two rooms in a World Salvador, the Philippines, Senegal, and Zambia Bank project in Amman, Jordan. In Tanzania and claimed that they were affordable down to the 20th Zambia, experimental “sites-without-services” proj- ects were carried out to keep costs low.18 The boxes 16 Cohen, 1983. below provide examples of four S&S programmes in 17 Srinivas, no date. the developing world. 18 Choguill, 1995. 9 T H E C A S E F O R I N C R E M E N TA L H O U S I N G Parcelles Assainies, Dakar, Senegal BOX 1 The first urban World Bank S&S project, Parcelles Assainies, began in Dakar, Senegal in 1972. The site for the project was a vacant area north of Dakar dominated by sand dunes. The Office d’Habitat de Loyers Modérés (OHLM) planned to use a 50-year, $8 million interest-free loan from the World Bank to develop 14,000 house plots of 150 square metres each on 400 hectares of land, with an average occupancy rate of 10 people per household. The project included minimal infrastructure and sufficient public spaces, schools, health clinics, and community centres to serve 140,000 people. Beneficiaries would buy their plots and infrastructure through their savings or a 15-year loan at 7 percent interest, and then build their home. The project was intended for the very poorest families, but initially it only reached down to the 47th income percentile. After some adjustments, the majority of plot purchasers fell between the 20th and 65th income percentiles. The project start was delayed due to disagreements between the government and the World Bank. By 1978 only 20 households had moved in. Major cuts to the project budget were made in 1976, reducing the size of the site to 300 hectares and cutting back on education, health, and community facilities. The political need for plots to be occupied, coupled with corruption, meant that better-off occupants dominated the site—1,000 families never even went through the selection process. The project wound up in 1981–82, five years later than planned. There were two main reasons behind the project’s difficulties. First, the project designers “did not explicitly examine the projected level of density of the project from the perspective of the medium or long term time frame; and secondly, [they] did not consider the planned settlement in terms of the wider patterns of land use existing in the city at the time. Rather decisions on density were project-specific [and] disconnected from the urban context as a whole.”a a Cohen, 2007. Las Presita, San Miguel, El Salvador BOX 2 The S&S Las Presita project in San Miguel, El Salvador, began with the supply of 900 units by the implementing agency FUNDACIÓN SALVADOREÑA DE DESARROLLO Y VIVIENDA MÍNIMA (FUNDASAL) in the 1970s with a World Bank loan. It had an innovative clustered land development pattern (36 clusters, each with a large central yard surrounded by 12 to 15 houses), legal land tenure, phased infrastructure, and core starter housing. Core starter options varied in design and had water, sewerage, and electricity. Each core house was built by 15 to 20 families through officially organised mutual aid. Participant families were selected through a long vetting process. Community facilities included a primary school, park/playground, and community centre. 10 S ECTI O N 2 : I NCREMENTAL H OUSING ST RAT EGIES IN CONT EXT Dandora, Nairobi, Kenya BOX 3 The Dandora project was the first major World Bank S&S venture in East Africa and the first to give major emphasis to community development as an integral part of the project’s design and management. It grew out of the far-sighted Nairobi Metropolitan Growth Strategy formulated within the Nairobi City Council (NCC). The project site was on the eastern fringes of the city with easy access to its main industrial area. The project’s first phase made 6,000 plots available, each with a “wet services” core for three different income groups, the majority of which comprised just a toilet, tap, and shower. Loan financing for the construction of two rooms using unpaid self- help labour was provided. Community facilities included six primary schools, two health centres, markets, two community centres, and a sports hall. Trunk infrastructure included paved roads with street lighting, water, and a waterborne sewerage system. The project included 330 plots sold at market prices to subsidise the cheapest plots. The agreement with the World Bank stipulated that a special project department be set up within the NCC to manage the project. The department comprised four divisions: administrative, legal, technical, and community development. The community development division was responsible for all aspects of public participation, including working with and providing training for allottees before, during, and after the occupation of their plots. It also coordinated independent non-governmental organisations working with the project. income percentile.19 However, an independent review and uniform housing estates within just a few years reported that the bottom 40 percent of households of initiation. Many government officers were still were excluded from most World Bank projects.20 This wedded to the unrealistic image of conventional led to criticisms that S&S projects mostly benefited public housing, and politicians, wary of being the better-off within squatter settlements, leaving a branded as public slumlords, could not cope with “sharply defined underclass” without housing.21 supporting incremental housing that could take years or decades to mature. Furthermore, there was Low rates of cost-recovery were frequently cited as a lack of trust in the construction skills of the poor a failure of S&S projects. This was in part due to people that the projects sought to target. As a re- the imposition of short-term cost-recovery regimes sult, officials, unable to devolve responsibility to the that belied the very nature of incremental develop- most effective levels of authority (that is, the ben- ment and that low-income households were un- eficiary households and communities), meddled able to meet. Other causes included delays in the and provided too much unsought guidance from provision of services, inadequate or unmanageable above.23 collection methods, lack of sanctions for non-pay- ment, and the absence of political will to enforce payment.22 19 Kearne and Parriss, 1982. 20 Choguill, 1995. 21 Peattie, 1982. In sum, there was an expectation that S&S projects 22 Srinivas, no date. would produce orderly, well-serviced, well-built, 23 Choguill, 1995. 11 T H E C A S E F O R I N C R E M E N TA L H O U S I N G Khuda-ki-Bastee, Hyderabad, Pakistan BOX 4 In 1986 the Hyderabad Development Authority (HDA) set out to simplify the provision of affordable housing to the lowest-income groups by imitating the informal housing process. It subdivided land on the city’s urban fringes into 3,000 plots and advertised them. Eligible applicants were then housed in reception areas for two weeks under very basic conditions to test their seriousness. Those who passed the test were then allocated an 80-square-metre plot with no infrastructure for a payment of US$30. Water was delivered by tanker. Beneficiaries had to live on their plots, but no construction standards were imposed. To avoid speculation, plots that were not occupied were repossessed by the HDA. Beneficiaries were encouraged to pay instalments into a savings account to cover the cost of infrastructure, which was provided once payments had been made. Non-governmental organisations provided organisational and managerial support and training. After 8 years, the project had achieved strong results. Around 2,800 plots had been allotted with a total population of around 18,000. Five doctors provided health services in a permanent health- care unit of the Family Planning Association of Pakistan. Private buses took residents to central Hyderabad and industrial locations every 30 minutes. The development had 110 shops, and more than 247 carpet looms provided jobs to at least 600 people. In terms of infrastructure, residents had collected and spent some US$1.5 million on water supply, sewerage, and electrification, and every house had electricity and indoor water supply. Finally, more than 2,000 houses had been built with permanent materials (nearly all families had started with reed huts), and around 70 collateral-free loans for enterprise development with a total value of $27,000 had been disbursed to some 150 families. The approach to low-income housing provision in Hyderabad had several significant characteristics. First, the programme required only minimal capital outlay and management overheads by HDA. By mimicking the informal settlement process, the programme was affordable and sustainable. Finally, the scheme was most successful in the blocks where non-governmental organisations had helped to organise community leadership and management. The HDA used several instruments to ensure the success of the programme: the incremental nature of the scheme; the continuous availability of plots; the issuing of dwelling permits so that only vacant houses and plots could be reallocated; and a simple, single-window bureaucratic procedure, which was performed on the spot. 2.4  The shift away from support Although the World Bank has maintained a fairly for incremental housing consistent level of funding for housing, in 1992 it shifted from S&S and upgrading loans to large- The perceptions and problems that plagued the S&S scale policy-related loans for housing finance, ad- projects of the 1970s and ’80s triggered policy shifts justment, and the privatisation of public services away from direct support to low-income urban (table 1).25 housing procurement. By the mid-1990s, only Sudan and Tanzania included S&S as part of their national 24 Choguill, 1995. housing policies.24 25 IHC, 2008. 12 S ECTI O N 2 : I NCREMENTAL H OUSING ST RAT EGIES IN CONT EXT Regional breakdown of World Bank shelter loans by type, 1992–2005, in 2001 US$ millions TABLE 1 Sites & Slum Housing Housing Disaster Percentage Region services upgrading policy finance relief Total in region Sub-Saharan Africa 16.32 42.42 2.47 17.04 2.92 81.26 1.2 East Asia and Pacific 35.80 40.78 36.12 439.05 33.92 585.66 8.6 Europe & Central Asia 16.46 10.61 311.34 235.20 305.37 878.98 12.9 Latin America & Caribbean 00.0 128.97 656.73 1,584.89 397.34 2,773.29 40.8 Middle East & North Africa 358.26 94.42 48.37 290.43 549.82 1,341.31 19.7 South Asia 79.00 21.18 2.37 145.32 883.95 1,131.81 16.7 Total 505.84 338.38 1,057.40 2,711.93 2,173.32 6,792.31 100 Percentage by programme 7.4 5 15.6 39.9 32 100 type Source: Buckley and Kalarickal (2006) Currently, the IDB spends 47 percent of its hous- Assistance to housing rarely constitutes more than 2 ing sector loans on upgrading and titling, 18 per- percent of bilateral funding (table 2).26 cent on the development of long-term mortgage credit, 20 percent on one-off demand-led subsidies According to the International Housing Coalition or vouchers to individual households, 3 percent (IHC),27 the reasons for the decline in donor coun- on subsidised core home construction, 3 percent try funding for urban housing since 1990, including on institutional reform, and only 2 percent on ser- incremental S&S and slum upgrading, include the viced sites. following: The ADB has implemented many housing projects  Housing is a long-term investment, so visibil- across Asia through loans and technical assistance ity is hard to maintain. programmes that have supported mortgage systems,  Donors want short-term results, but housing the reorganisation of housing authorities, housing development is slow and complicated. finance strategy development, market-based hous-  Housing initiatives are riskier to implement ing finance loans, and upgrading. than other types of programmes.  Housing programmes do not have a large sup- The USAID, the Swedish International Development porting constituency in the donor countries Agency (SIDA), and the UK Department for such as that for HIV/AIDS programmes. International Development (DFID) have all down-  Land titling and local politics make housing graded or dismantled their housing and urban de- programmes difficult. partments and units over the past decade. Since 2003 housing has been a minor component of their activi- 26 IHC, 2008. ties, often confined to disaster reconstruction efforts. 27 IHC, 2008. 13 T H E C A S E F O R I N C R E M E N TA L H O U S I N G Housing as a percentage of all official development assistance and official aid, 1995–2003 TABLE 2 Institution 1995 1996 1997 1998 1999 2000 2001 2002 2003 Multilateral 0.001 0.177 0.233 0.072 0.023 0.015 1.254 0.314 0.253 Bilateral 0.247 0.147 0.508 0.461 0.344 0.160 0.168 0.180 0.164 Source: IHC (2008) from unpublished UN-Habitat report.  Many agency staff retain a rural focus. million slum dwellers. Poverty Reduction Strategy  Most donors are unable to maintain a long- Programmes, promoted by the World Bank and the term focus. United Nations Development Programme (UNDP) are beginning to distinguish between urban and ru- This section has emphasised the changing role of ral poverty and to recognise the importance of ur- international donors because of their influence on ban housing, not only to alleviate the social impacts national and municipal housing policies and strate- of poverty but also as a proactive poverty reduction gies in the 1970 and ’90s. Nevertheless, the impact strategy. The effects of climate change on urban of donor funding on the scale of urban housing areas and particularly the urban poor are already is limited. From an operational perspective, only beginning to be understood and have influenced national and local governments can provide the approaches to securing housing and infrastructure necessary support for housing the lowest-income for lower-income urban populations. Examples of urban groups. Regardless, the international com- current urban-housing programmes in developing munity still has an important role in supporting countries can be found below. policy change and helping to develop local innova- tions, initiatives, and capacities to tackle both the The ADB is working with the Development Bank of complexity and scale of urban housing and related the Philippines (DBP) and the Housing and Urban problems. Development Coordinating Council (HUDCC) to house 20,000 poor urban families outside of Metro Manila through a US$30 million loan and US$1.5 2.5  Support for incremental million technical assistance grant. This includes a housing now (2010) component for local government to develop sites and to distribute land titles to informal settlers Despite the shift away from housing projects and so that they can develop incrementally, as well as programmes to investment in urban management credit facilities for shelter finance and small en- and governance, support for incremental housing terprise development and capacity building for has continued on an ad hoc basis. Much of it has communities, local governments, and government been implemented without donor backing and as housing agencies to decentralise shelter deliv- component projects of broader poverty alleviation ery. The shelter finance component of the project programmes. The Millennium Development Goals provides incremental financing for plot purchase all have implications for policies that address urban on terms appropriate to the circumstances of the poverty. In particular, Target 7D aims to achieve a poor and through financing mechanisms linked significant improvement in the lives of at least 100 to the formal financial system. To make the plots 14 S ECTI O N 2 : I NCREMENTAL H OUSING ST RAT EGIES IN CONT EXT affordable, the project avoids distorting interest that demonstrate good design and construction. One rates. It instead uses targeted land cost subsidies such project, the Ashraya Urban Housing Scheme, while retaining an incentive for the poor to pay and provides free plots of land (up to 60 square metres) for financial institutions to be proactive in ensur- to low-income people registered on the municipal- ing repayment.28 ity’s “siteless/houseless” list. Title deeds are generally registered in the name of female household members, In Zimbabwe, housing cooperatives are officially and each family is responsible for building a house on sanctioned housing delivery mechanisms for low- the plot with financial and technical support from the income earners in Harare. Government statistics corporation.30 show that 11,500 housing plots have been provided to cooperatives. Over the past 10 years, 2,350 hous- In South Africa, Ivory Park is a planned, informal, ing units have been built in the city through S&S low-income settlement established in 1991 to pro- programmes and incremental housing schemes. The vide S&S accommodation for overflow from the S&S programmes include serviced plots with super- Alexandra and Tembisa areas in Johannesburg. It structure and title deeds, and the incremental hous- houses 240,000 people and features a variety of ing schemes provide unserviced plots with layout housing projects ranging from contractor-supplied plans.29 For example, Nehanda Housing Cooperative capital subsidy houses to People’s Housing Process in Dzivaresekwa has 5,300 planned residential plots schemes that promote, organise, and support com- of which more than 600 have been developed with munity-based house construction and unassisted houses ready for occupation. self-help housing. Private companies, NGOs, and cooperatives are central to the delivery of housing In Central America, the SIDA is working with a and housing services. Since 1996 the government has range of NGOs to deliver housing microfinance. built 3,600 houses in the Kaalfontein area, of which Some NGOs are actively involved in developing new 600 were built through S&S projects managed by the sites and supporting incremental housing process- People’s Housing Process.31 es. As part of its urban poverty and housing strat- egy, the SIDA provides micro-credit for the repair In addition to projects on undeveloped land, and extension of houses with loans in the range of there are even more examples of upgrading proj- US$500–US$1,500 over 18–36 months, with flexible ects that support ongoing incremental housing guarantees. processes by “formalising” informal settlements. Current initiatives tend to be small in scale, and In India, housing corporations have been established delivered through partnerships with local gov- in some states to cater to the housing needs of the ernments, NGOs, cooperatives, and the private rural and urban poor. In Karnataka, for example, sector. There are no examples of national, metro- the Rajiv Gandhi Housing Corporation was set up politan, or municipal housing polices or strategies in 2000 to harness government funds, administer to support incremental household or community loans, organise the manufacture or bulk procurement of building materials, and implement new housing projects founded on self-help housing construction. 28 ADB, 2003. 29 The Herald, 2009. These projects set up savings organisations to assist 30 http://ashraya.kar.nic.in. residents in repaying loans and to build model houses 31 Omenya, 2004. 15 T H E C A S E F O R I N C R E M E N TA L H O U S I N G initiatives for the procurement of secure access to of larger poverty alleviation programmes. This is urban housing. But unlike their predecessors in the evidence that some of the failings of early S&S and 1970s and ’80s—most of which were stand-alone upgrading projects have been heeded and that un- initiatives divorced from the wider urban con- derstanding of housing processes and poverty has text—recent and current projects tend to be part advanced over the past 30 years. 16 SECTION 3 The Case for Incremental Housing Strategies The overall objectives of public sector engagement in the procurement of housing by the lowest income groups are to:  Stimulate the development of formally recog- low-income communities themselves. This applies nised and affordable urban housing in ap- to both the upgrading of existing informal settle- proved locations and in sufficient quantities ments and the incremental development of new to reduce urban housing deficits and meet the urban areas and communities. The photographs in needs of urban growth. this chapter illustrate that with the right supports  Build partnerships between the local govern- (and sometimes without them), low-income urban ment and private sector landowners and devel- settlements can become stable communities with opers to bring affordable serviced plots to the good-quality houses and adequate services. market.  Optimise public sector investment in the pro- vision of infrastructure and services to emerg- 3.1  The numbers case ing low-income neighbourhoods, including the development of partnerships with private Between 20 percent and 70 percent of the urban sector infrastructure and utility providers. population in developing cities already produce  Develop and support local organisational ca- their housing incrementally, the majority of them pacity for incremental development and com- informally (illegally) with little or no security of ten- munity asset management and maintenance, ure. If government strategies are not put in place to including the development of partnerships recognise and support this process, the world’s ur- with NGOs. ban slum population will more than double in the next 30 years—from 925 million to 2 billion.32 The six cases discussed below are based on the un- derstanding that to be effective and sustainable, pub- Experience over the past 40 years clearly shows that lic sector support of incremental housing produc- neither countries nor cities have the financial or tech- tion must include technical and financial assistance, nical resources to consistently produce affordable income generation opportunities, and community subsidised housing for more than a minute fraction and individual capacity-building and community of their low-income urban populations. Therefore, facilities, delivered through partnerships with gov- ernment agencies, NGOs, private bodies, and the 32 UN-Habitat, 2003. 17 T H E C A S E F O R I N C R E M E N TA L H O U S I N G the only realistic strategy for public housing agen- little impact. The government has largely reverted to cies is to build upon and formalise the informal pro- subsidising the building of finished homes for a small cesses that are already in place. These have already proportion of the middle- and upper-income groups, been shown to produce enough affordable housing, and by 2007 squatter settlements housed more than even though it is often imperfect in terms of security, 60 percent of the metropolitan population.37 health, safety, and amenities. In sum, many governments still need to be con- The case of Lagos, Nigeria, demonstrates the con- vinced that investment in housing is a critical re- sequences of a lack of state support for incremental sponse to urbanisation that can lead to national eco- housing—indeed, the lack of any effective hous- nomic growth and reduce poverty. However, hous- ing policy. Between 1976 and 1985, approximately ing policies based on the construction of subsidised 850,000 new housing units were required in the conventional dwellings in the formal sector have Lagos metropolitan area alone, but only 82,000 for- virtually no impact on urban housing needs, which mally recognised units were actually built. Although are instead met by informal construction. In other around 100,000 people arrive in Lagos each year, words, if you can’t beat ’em, join ’em—and in doing building plans for 1991 anticipated just 4,800 new so improve them. housing units.33 The National Shelter Strategy of Ghana states that 3.2  The financial case very few attempts have been made to harness and supplement the population’s own non-conventional The financial case for support to incremental hous- strategies for procuring shelter. It goes on to point ing is based on securing household investment in out that the underlying problem that besets the housing and community facilities. Secure housing is country is that although the government has realised the greatest financial asset available to most urban the significance of non-conventional strategies of families. Better-quality dwellings have greater ex- housing procurement, articulating such a complex change values, more value as collateral for borrow- issue into a refined process that can be promoted ing, and a higher price when sold. At the same time, and implemented on a nationwide basis is a mon- home ownership provides social security and status strous task.34 As a result of the lack of support for to its owners and occupants. Therefore, it is in the incremental housing, approximately 90 percent of best interest of householders to invest in their hous- the housing stock in Ghana is produced informally.35 ing as long as their investment is secure. Few people will invest in property if there is any ambiguity in Even in countries that have supported incremental the legality of their title or the physical safety of their housing, the scale of the support has been so limited asset. To a large extent, it is the insecurity of title as to barely make a dent in housing needs. In Senegal, and the location of informal settlements that create for example, in addition to the Parcelles Assainies project, S&S projects were developed in Thiès (1980) and Kaolack (1992) along with the Malika-Keur 33 Adedeji and Olufemi, 2004. 34 CHF, 2004. Massar project in Dakar (1989). In 1998 a third major 35 CHF, 2004. project was started in Dakar comprising 23,000 plots 36 ENDA-RUP, 2005. planned on 624 hectares.36 These initiatives have had 37 Cohen, 2007. 18 S ECTI O N 3 : THE CA S E F OR INCREMENTAL H OUSING ST RAT EGIES slums. Experience shows that where security is en- 3.3  The urban management case sured, all but the poorest will invest time, energy, and resources to extend and improve their house and its The urban management case for incremental hous- surroundings. Thus, by securing land titles, govern- ing strategies is premised on the principle of sub- ments can create good urban environments at a rela- sidiarity—in other words, the recognition of the tively low cost to the state as well as enhance public most effective level or location of decision-making revenue in situations where local taxation is based and authority for each component of an activity. It on property values. entails the identification of all public, private, and community sector actors and their competencies, It is estimated that small loans to low-income bene- and the casting of their roles accordingly. ficiaries of the Sri Lanka Million Houses Programme in the 1980s generated returns 4 to 6 times their val- In this context, the construction, maintenance, and ue through the investment of family savings and in- management of dwellings are logically the respon- formal borrowing, in addition to the value of labour sibility of households, not of government. Only the and other material inputs by the beneficiaries. In owners and users of housing understand their own Kenya the credit provided by the Dandora S&S proj- priorities for investment, their available resources, ect was augmented by a factor of 4 to 5 by individual and their changing housing needs. The government’s household inputs.38 In the Parcelles Assainies proj- role in the procurement of housing is that of facilita- ect in Dakar, the World Bank’s Project Completion tor, banker, or guarantor, providing those inputs that Report found that for every US$1 of Bank money households and communities cannot effectively or provided, US$8.2 of private funds were invested on- efficiently provide for themselves. NGOs often have site.39 By 2006 many houses in the project had two key roles as advisors to, organisers of, and interme- floors each with six rooms and often had a third floor diaries between low-income households and vari- under construction, evidence of the willingness and ous departments of local and central government. capacity of households to invest in their homes.40 In turn, different levels and departments of govern- ment have key roles in the supply and subdivision Security of tenure is not necessarily sufficient to gen- of land and the extension and management of infra- erate the return on public sector inputs cited in the structure and urban services. above examples. Many financial issues affect house- holders’ capacity to develop their dwellings. In many The process of supporting incremental housing early S&S schemes, households were unable to afford strategies is a unique vehicle for rationalising many the plots allocated to them—in addition to increased urban management processes by devolving or del- transport, infrastructure, and food costs, as well as egating many of the government’s traditional re- the resources needed to build a new house. However, sponsibilities to the most appropriate actors and other evidence suggests that residents on appropri- developing partnerships with private and civil soci- ately located plots with cheaper and easier access to ety actors. A key role for government is the acquisi- services and financial support packages can mobilise tion of suitably located land for low-income groups. enough resources to build and improve their homes over time. This is not possible when poor families are 38 Lee-Smith and Memon,1988. committed to fixed-term payments on conventional, 39 Rowbottom, 1990. fully completed dwellings. 40 Cohen, 2007. 19 T H E C A S E F O R I N C R E M E N TA L H O U S I N G GUACAMAYAS, BOGOTA, COLOMBIA A sites and services project started in 1976 by the Caja de Vivienda Popular on the city fringes but with good access to the city centre and industrial areas. 1976 Core service units with one room on each plot were provided. Many households moved into the site with second-hand building materials and components to start extending their dwellings immediately. Patrick Wakely 1977 Within one year several families had extended their houses to include a second floor. Patrick Wakely 2009 Barrio Guacamayas has become fully urbanised, with traffic confined to perimeter roads and pedestrian precincts in the interiors of the blocks. Houses continue to be extended and improved. Guacamayas has its own community website, http://www.barrioguacamayas.com. Maria Victoria Echeverri Governments have the critical ability and authority corporations—such as railways, ports, and airport to allocate public land for housing, acquire it from authorities and nationally owned industries—own private (illegal or legal) owners, or work with own- large tracts of often vacant or under-utilised urban ers to develop the land for low-income housing. land. Based on the principles of subsidiarity, how- Most significantly, governments are often large-scale ever, poor households also have a potential role in urban landowners themselves. Public agencies and finding and negotiating the transfer of land. For 20 S ECTI O N 3 : THE CA S E F OR INCREMENTAL H OUSING ST RAT EGIES example, with NGO support squatters in Mumbai informal urban development processes and to en- have mapped vacant land and successfully negoti- sure adequate and relatively efficient provision for ated its transfer so that they may resettle away from infrastructure and service delivery and the rational dangerous railway trackside slums.41 use of urban land. They allow the government to shape the development of towns and cities in ac- Management of the incremental extension and up- cordance with strategic priorities developed for an grading of infrastructure is not easy, especially since entire urban area, rather than engaging in small- it is generally undertaken by different utility agen- scale fire-fighting. Supported incremental housing cies and enterprises. Nevertheless, as examples such can be a means to reduce uncontrolled low-density as the Orangi Pilot Project in Karachi, Pakistan, urban sprawl in favour of high-density compact have shown, by co-ordinating organised community development. groups with different levels of government, effective and responsive systems can be developed that are af- In Dakar, the absence of concerted, large-scale sup- fordable in both managerial and operational terms.42 port for incremental housing has resulted in low- density urban sprawl, which causes high infrastruc- In sum, state-supported incremental housing can ture, energy, and transport costs; environmental catalyse the decentralisation of government in ways problems; and low levels of urban productivity and that improve urban management and the adminis- economic growth.44 By contrast, in Aleppo, Syria, in- tration of urban services, including the development formal settlements housing some 45 percent of the of partnerships with the private sector and com- population cover some 30 percent of the developed munity actors. Examples exist in which the private area of the city, embracing both centrally located sector, both large companies and small and local and peri-urban land. While much of it serves the enterprises, have been contracted for the installation occupants well in terms of location, it distorts land of infrastructure and the delivery of some services, markets and presents a significant problem for the while NGOs frequently take responsibility for social city’s planners and departments responsible for the and community development. management of urban services. To address these con- straints, the city established a municipal Informal Public-private partnerships have been formed to Settlements Department to begin to formalise the legally develop and manage land for low-income settlements within the framework of a city develop- incremental housing. NGOs have also worked with ment strategy. private enterprises to promote the effective use of in- frastructure and support cost-recovery procedures, Over time, the concern that government-supported as in the Favela Bairro programme in Rio de Janeiro, incremental housing would amount to “building Brazil, which featured partnerships between local official slums,” particularly in central locations, has NGOs and the private electricity company, Light.43 largely been refuted. Three decades after the first S&S projects were implemented, many settlements are 3.4  The urban development case 41 Burra, 1999. 42 Pervaiz et al, 2008. Incremental housing strategies provide the gov- 43 Riley and Wakely, 2005. ernment with an opportunity to regulate ongoing 44 Cohen, 2007. 21 T H E C A S E F O R I N C R E M E N TA L H O U S I N G NAVAGAMPURA, COLOMBO, SRI LANKA A sites and services project initiated by the National Housing Development Authority as part of the Sri Lanka Million Houses Programme in 1985 on an inner-city site. 1985 The project was laid out as terraced housing with shared “party walls” between each dwelling—an innovation in Sri Lanka, where detached houses on individual plots were the norm. Patrick Wakely 1986 The uniform roof level was spontaneously maintained in the initial construction stages. Patrick Wakely 2009 Navagampura has become a regular part of the urban fabric of Colombo and is still being developed by its residents. Kumudu Jayaratne indistinguishable from “regular” neighbourhoods.45 incremental housing, on the other hand, provides for While 30 years may be enough time for houses in long-term service planning and the anticipation of unregulated squatter settlements to also reach high future needs. It can provide the financial and techni- standards without any state assistance, many are still cal assistance and security needed to accelerate the characterised by disparities in housing standards, un- local development of regular neighbourhoods within safe construction, illegal land tenure, and precarious infrastructure and service levels. Officially supported 45 Gattoni, 2009. 22 S ECTI O N 3 : THE CA S E F OR INCREMENTAL H OUSING ST RAT EGIES the context of planned urban growth. Incremental working with the city’s ward commissioners, the low- housing can therefore contribute not only to better est level of public administration. The BOSC struc- urban development on a local scale but also to the ture is a hierarchy of elected committees, the base of development of cities as a whole, provided that they which are primary committees, which represent some are planned as integrated components of the urban 50 to 1,000 households. The primary committees are fabric rather than isolated entities on the city fringes. then represented on the 90 ward committees, which in turn send representatives to the 29 thana com- mittees and then on to the Dhaka City Corporation 3.5  The governance case (DCC). The ward commissioner, who is elected by his/her constituents, is the link to the city’s service The organisation and management of incremental providers. This interface between government and building processes—and particularly engagement organised representative non-governmental bodies in the installation and extension of neighbourhood has become a widely accepted model that works well infrastructure—provides a means to the develop- in many of the city’s wards, reducing corruption and ment of decentralised participatory decision-making giving voice (and confidence) to residents.46 and governance. Good governance not only helps to ensure transparency and accountability in the Slum/Shack Dwellers International (SDI) in Africa management of the financial and physical aspects and India and the Asian Coalition for Housing Rights of housing and neighbourhood building, it is also a (ACHR) are two examples of international federa- vehicle for community development that can stimu- tions of NGO-supported slum dwellers. They both late a wide range of local capacity-building activities. have a strong message of developing mutual under- Participation helps to create a sense of ownership and standing between urban low-income communities pride in the local environment among residents that and government by nurturing mutual appreciation can engender a feeling of responsibility for the main- of the aims, ambitions, strengths, and constraints tenance and management of community assets (such faced by the other.47 Ideally, such appreciation leads as streets, drains, street lighting, public open space, to mutual understanding, the development of trust, schools, and clinics). It also reinforces the advantages and—eventually—active city-building partnerships. and power of collaboration over competition for ac- These are the foundations of progressive urban gov- cess to resources within urban low-income groups. ernance that grow out of direct government engage- ment in incremental housing policies.48 Appropriate engagement in incremental housing pro- cesses, either through slum-upgrading programmes or S&S projects, can introduce mutual understand- 3.6  The social and economic ing into the normally antagonistic relationship be- development case tween local leaders and government officials. For ex- ample, the Busti Baseer Odhikar Surakha Committee Closely related to the fostering of good governance, (BOSC) was set up in Dhaka, Bangladesh, in the late incremental housing processes can be important 1990s by the Coalition for the Urban Poor (CUP), an alliance of 53 NGOs. The committee has established 46 Banks, 2008. a citywide network of “accountability mechanisms” 47 D’Cruz and Satterthwaite, 2005. to include the urban poor in urban governance by 48 Riley and Wakely, 2005. 23 T H E C A S E F O R I N C R E M E N TA L H O U S I N G CIUDAD BACHUE, BOGOTA, COLOMBIA (Low-rise housing) Begun in 1977 by the government housing agency Instituto de Crédito Territorial (ICT), the incremental housing project has two components: high-rise apartments (see next page) and low- rise terraced housing of concrete post and beam construction with pre-cast wall and floor slabs, which householders can extend by building a second floor. 1979 Pre-fabricated core houses. At this time, households were already assembling building materials for extensions, replacing doors, and installing security grilles. Patrick Wakely 1985 Many of the pre-fabricated panels were replaced with conventional block and brick construction. Patrick Wakely 2008 Roof terraces and third-floor extensions are being added. Patrick Wakely 24 S ECTI O N 3 : THE CA S E F OR INCREMENTAL H OUSING ST RAT EGIES CIUDAD BACHUE, BOGOTA, COLOMBIA (High-rise housing) Apartment blocks with pedestrian access in the front provide two-story maisonettes in which the ground floor can be extended in the rear; second-story flats, accessible by an open gallery, can be extended by building on the roof. 1978 Column and beam construction. 1979 Maisonettes and flats ready for occupation. Patrick Wakely 1981 Extension of flats on roof and maisonettes extending on plots in the rear. 1987 Two-story extension on roof and extensive personalisation. Patrick Wakely 2008 Owners select the paint colour. Patrick Wakely 25 T H E C A S E F O R I N C R E M E N TA L H O U S I N G CHINAGUDALI, VISAKHAPATNUM, INDIA A sites and service project on the urban fringes meant to resettle squatters from the centre of Visakhapatnam. Only communal water points and individual pit latrines are provided for each plot. 1988 Private building material suppliers arrived at the site on the first day of occupation. Patrick Wakely 1989 Pour-flush pit latrines were supplied on each plot. Construction still utilised temporary materials. Patrick Wakely 2009 Chinagudali has developed into a thriving suburb of Visakhapatnam. P. Rambabu and effective catalysts for the social and economic together in a “common cause.” This is especially development of poor households and communi- important in S&S projects in which the beneficia- ties. By organising themselves (or being organ- ries are randomly selected (that is, projects that are ised) to engage in developing their housing and not the result of slum re-location in which existing local environment, households inevitably come communities are moved as a whole). This presents 26 S ECTI O N 3 : THE CA S E F OR INCREMENTAL H OUSING ST RAT EGIES an opportunity to develop and consolidate social of community-based savings groups and provides solidarity and to introduce and support local en- loans and grants for land acquisition, upgrading, terprise initiatives and employment, notably in the house building, income generation, and food pro- infrastructure and housing construction activi- duction; it also aims develop better relations with ties of the project itself. They can also build social the government.53 By 2007 the Sri Lanka Women’s capital around issues that are not related to the im- Bank, which began as a small housing-related sav- mediate urban environment, such as sporting or ings scheme in Colombo in the early 1990s, had over cultural activities that engage the youth and NGO 60,000 active low-income members and total assets programmes for anything ranging from women’s of over Rs. 800 million (US$8 million), managed literacy to environmental health, nutrition, and through 160 branches in 22 of the country’s 25 ad- home economics. ministrative districts.54 There are numerous successful examples of such Both the location and the provision of facilities strategies. For example, central to the Dandora for income-generating activities in incremental S&S project in Nairobi, Kenya, was the establish- housing projects can have an important impact ment of a Community Development Division of on poverty reduction and householders’ capac- the Project Department that managed the organi- ity to improve their housing. For example, work- sation of house builder groups and made links shop facilities were planned as part of the Shivaji to other citywide welfare organisations and so- Park S&S project in Alwar, near New Delhi, India. cial programmes.49 Specially trained community Together with the project’s central location, these workers played a similar “citizen-building” role in facilities allowed many women to start new enter- both the S&S and upgrading components of the prises—mainly tailoring and other dress-related Lusaka housing projects of the mid-1970s50 and activities—working from home and supplying lo- in Indonesia’s extensive Kampong Improvement cal stores. This increased the number of working Programme.51 family members from less than 1 to an average of 2.5. Household incomes increased by 6 to 8 times The strategic importance of locally managed savings between 1985 and 2000.55 groups as a vehicle for community building and the nurturing of solidarity, particularly among women In conclusion, support of incremental housing engaged in incremental house building, is becom- processes can provide the basis for the wider so- ing well understood. Not only can regular savings of cial and economic development of low-income very small amounts of money generate significant households and communities. However, it also re- capital funds for borrowing, but the process of col- quires many government and municipal depart- lecting, banking, lending, and recovering loans is ments responsible for housing and works to take a powerful means of community building and the development of grassroots collective management 49 Lee-Smith and Memon, 1988. capacities. Such savings schemes underpin the ac- 50 Jere, 1984. tivities of the National Slum Dwellers Federation 51 Silas, 1984. 52 Mitlin, 2008. (NSDF) and Mahila Milan in India and their 53 Phonphakdee et al, 2009. partners in SDI.52 In Cambodia, the Urban Poor 54 Sevanatha, 2007. Development Fund supports a growing number 55 Lall, 2002. 27 T H E C A S E F O R I N C R E M E N TA L H O U S I N G on or develop their capacity to provide social and Community Development Division. Similarly, to economic support and acquire new skills and pro- implement the national Million Houses Programme fessional competencies. For example, the 1970s in the early 1980s, the Sri Lanka National Housing Dandora S&S project in Nairobi, mentioned above, Development Authority had to retrain its cadre of led the City Council to establish a new permanent building technicians to enable them to take on new Housing Development Department with a strong roles as community support advisors. 28 SECTION 4 Components of Incremental Housing Strategies 4.1  Land and location Land is perhaps the principal component of government support for urban, low- income housing. Government influences availability of land through allocating and using publicly owned land; expanding infrastructure and services to new sites, enabling their development; administering rules and regulations such as planning laws, building permits, and developmental controls that impact land price and availability; and ensuring efficiency and transparency when any of these actions are taken.56 Government should consider three basic aspects of any land it plans to make available for incremental development: its location, its price, and the title conditions under which it is transferred. 4.1.1  Cost and location squatter in an illegal and insecure shanty in Mexico Selecting the right land for the incremental develop- City, who made a living from casual employment as a ment of low-income housing is difficult, but it is cru- semi-skilled mason (at building sites in the city cen- cial to the success of any incremental housing policy tre), was awarded a secure title to “affordable” and or project. Peri-urban land has often been acquired well-serviced—though basic—housing on the urban for this purpose because of its relatively low price, periphery for him and his wife, who ran a kiosk serv- but the cost of extending infrastructure to it renders ing tourists. Living in the new residence, however, in- it unaffordable to the target groups. When projects creased his transport and housing costs from 5 per- have been built in poor locations, as was the case with cent of household expenditure to 55 percent, forcing many S&S schemes in the 1980s, the lack of affordable him to abandon his new dwelling and seek rental ac- transport links to employment centres, commerce, commodation in another centrally located informal and community facilities has led to empty plots and settlement.58 Similar stories come from cities as far- rapid turnover of residents.57 flung as Nairobi, Kenya; Johannesburg, South Africa; There are many anecdotal accounts illustrating the 56 E&U, 2009. importance of location to low-income families living 57 Van der Linden, 1992. on minimal and precarious incomes. For instance, a 58 Turner, 1976. 29 T H E C A S E F O R I N C R E M E N TA L H O U S I N G Phnom Penh, Cambodia; and Ahmedabad, India. Thus, evidence indicates that successful new incre- Despite having secure title on their dwellings, house- mental housing projects—ones that generate stable holds commonly abandon serviced sites and subsi- and vibrant low-income communities—are sited dised housing on the urban fringes and return to the within existing city conurbations and near existing precarious existence of squatting in shacks with no employment sources, services, and infrastructure. infrastructure because the latter are closer to job op- Moreover, they thrive despite the cost constraints portunities and offer more community ties. and concerns about the relatively long-term “un- sightliness” of low-income settlements in the process In Guyana, one of the failings of the IDB-supported of construction. Low Income Settlements Programme (LISP) was the location of the land—sites were on agricultural The successful Shivaji Park project in Alwar, close to land in distant locations. People could not afford to New Delhi, is located just 3 kilometres (km) from build houses so far away from population centres, major office complexes (where all major services and because of the high cost of extending trunk and social and physical infrastructure are easily ac- infrastructure to the sites, support services such as cessible), 4 to 6 km from schools, and 2 km from a schools, clinics, and playing fields were planned for main hospital. Retail and wholesale markets are also but not built. There was no commerce on-site and nearby, and high- and medium-income public and no employment nearby.59 Food was more expensive private housing complexes surround the settlement. in these peripheral locations and added to over- Those working in micro-enterprises stimulated by the all increases in household expenditure.60 As a con- project travel less than 1 km to work. Those working sequence, occupancy rates were low. Similarly, in for home-based manufacturers travel no more than El Salvador, FUSAU-Integral, a company that pro- 3 km to fetch raw materials from central neighbour- moted integrated housing solutions, found that the hoods. The maximum workplace commute, travelled main constraints to scaling up its work were the lack by about one-third of the residents, is 6 km. Thus the of suitable and affordable land, the high cost of basic project has sparked local economic development not services, and the distance of available land from em- only by fostering workplaces within the settlement ployment sources.61 but also by being near the city centre. The central lo- cation of Shivaji Park also enables greater access to S&S projects have often ignored the social impact of educational facilities, resulting in a much lower drop- settling thousands of people far from urban centres out rate among students 7 to10 years of age than that and have failed to support the building of new com- found in other resettled low-income communities munity networks or to foster social cohesion and located on the outskirts of Delhi.63 local management capacities. Re-location involves the loss of socio-spatial support networks, a loss that S&S projects within urban areas usually entail devel- the poor are least able to absorb.62 New settlements oping relatively small parcels of land, in contrast to near existing centres of commerce, employment, the vast projects of the 1970s and ’80s that benefited and housing have almost invariably done better than distant ones. For example, many of those moving to 59 Gattoni, 2009. 60 Cohen, 2007. Khuda-ki-Bastee near Hyderabad in Pakistan had 61 SIDA, 2007. relatives nearby, suggesting that this influenced their 62 Van der Linden, 1992. decision-making. 63 Lall, 2002. 30 S ECTI O N 4 : CO MPO NENTS OF INCREMENTAL H OUSING ST RAT EGIES both from the relatively low cost of land on the urban in “open application” programmes and projects, only fringes and from economies of scale. Higher costs are the development agency—the government or a part- associated with working at a small scale on more cen- nership between government and landowners and/or tral sites, but these are offset by relatively easy con- developers—may identify potential sites. Which ones nections to existing trunk infrastructure and access are chosen depends on their “marketability,” a con- to existing services. While individual land parcels cept that embraces all the criteria and priorities of the may be small, the number of suitable sites within potential takers and the beneficiary selection process. cities is usually large, enabling overall programme scales to be significant. 4.1.2  Land acquisition and law reform Well-located, privately owned land within an existing In sum, identifying land on which to develop S&S built-up area is likely to command a high price, nor- for low-income housing requires a rigorous analysis mally well beyond the means of local governments. of its location and benefits—its initial price and the Compulsory acquisition for low-income housing cost of servicing it are merely starting points. It also requires a level of political will that few politicians entails an assessment of the social and economic costs possess and involves more risk than most can afford of the intended beneficiaries in a context of often to take. So governments have to negotiate with pri- wildly fluctuating family fortunes, insecure incomes, vate landowners using incentives to encourage them and changing household priorities. Attempts may be to make land available or to develop it themselves on made to model such variables, but it is unlikely that terms that are affordable to low-income household- any such exercise would have much use in practice. ers, even if doing so yields suboptimal market returns. However, more down-to-earth analyses methods do For instance, the Maharashtra Slum Rehabilitation exist and have proved highly successful. For instance, Authority in India administers an incentive scheme of with the assistance of the NSDF/Mahila Milan/ transferable development rights through which land- Society for the Promotion of Area Resource Centres owners and developers can obtain attractive terms for (SPARC) coalition, organised groups of pavement the development of commercially marketable land in dwellers in Mumbai, India, have themselves identified exchange for the release of their land for low-income vacant land that meets their requirements, as far as lo- housing development. The Malaysian government cation is concerned, for a low-income settlement and requires private sector developers to make land or have negotiated with the government on the terms for houses available to low-income households at under- its acquisition and development.64 A similar approach market prices as a condition for the issue of develop- has been adopted in Phnom Penh, Cambodia, with ment permits for commercial housing. the support of the Asian Coalition for Housing Rights. Here, squatter households not only have sought out Through local-level strategies such as “land shar- appropriate land for the government to acquire but ing,” the government negotiates with landowners also, with the assistance of young architects, have pre- who possess land that has been made almost worth- sented proposals for its subdivision into plots as well less by squatters. Landowners transfer a portion of as a financial development plan.65 the squatted land to the government in exchange for having the other portion cleared of squatters Because the examples above involve the relocation of existing urban communities, these communities may 64 Patel and Mitlin, 2004. participate in the development process. By contrast, 65 ACHR, 2004. 31 T H E C A S E F O R I N C R E M E N TA L H O U S I N G and thus made available for commercial develop- able to demonstrate workable strategies and pressure ment. The other part of land-sharing negotiations governments to introduce reforms in this politically is with the squatters to get their agreement to trade sensitive and volatile area. space, resulting in higher residential densities, or relocation, for secure tenure to their portion of the Many cities have extensive “reserves” of under- “shared” land. Such land-sharing strategies were utilised urban land in public ownership. While the highly successful in Bangkok, Thailand, in the early central location of such land may once have been 1980s66 and have also been employed in Chennai, important to its nominal owner—which is often an India, among other places. agency, such as a port or railway authority or the military—that is rarely the case anymore, and much The lack of planning and budgeting for future low- of it could now be used for low-income housing. income settlements remains a persistent problem. However, such are the complexities of inter-agency Working with community organisations to develop relations that negotiating such transfers is rarely easy. land cadastres can provide powerful information to help push an agenda for land market reforms, which Strategies such as city land banking—in which cen- are usually complicated and often inequitable. Not tral or municipal governments buy peri-urban agri- only do poor households drive incremental house cultural land at low prices and then sell or develop construction and improvement, but they are of- it once the city has expanded and the land’s price ten active agents in getting land for housing, either has risen—can be used to profitably control future through negotiating tenure for the land they occupy growth. The city of Stockholm, which acquired ex- or through negotiating new sites on which they can tensive rural land at the beginning of the twentieth build, as illustrated by the NSDF in India (and de- century, famously employed this long-term strategy. scribed above). Federations of community organisa- But there are others. For example, in the 1970s Syria tions can make cadastres of land, conduct surveys, conducted an extensive programme of compulsory and estimate the cost of the work that needs to be peri-urban land acquisition that left municipalities done to develop new sites. In Bangladesh it has been with vast reserves of state-owned urban land. By argued that NGOs should take a more active part in default, it also brought large areas of land onto the S&S schemes, help negotiate land deals, advocate for informal market at prices affordable to the lowest- the decoupling of land titles from service provisions, income groups.69 promote techniques such as plot reconstitution and guided land development, verify land records, super- 4.1.3  Land tenure vise procurement, and validate transactions, as well as In many societies individual freehold ownership provide housing microfinance and support the devel- of land and property is the only form of title con- opment of building skills and materials production.67 sidered absolutely secure. However, it has been criticised for enabling households to sell their plots The nature and scale of the challenges of national upmarket, in effect making a profit off government land law reform, especially in urban areas that have grown in a spontaneous and chaotic fashion, are 66 Angel and Chirathamkijkul, 1983. complex and daunting, but they are becoming bet- 67 Rahman, 2005. ter understood by programmes such as the UN- 68 www.gltn.net. HABITAT Global Land Tool Network,68 which is 69 GTZ, 2009a. 32 S ECTI O N 4 : CO MPO NENTS OF INCREMENTAL H OUSING ST RAT EGIES subsidies. Collective titles such as housing associa- 4.2  Finance tions, co-operatives, and condominiums not only are acceptable alternatives to individual freehold In early S&S projects government was generally not ownership but they contain mechanisms to pre- expected to provide financing to enable the construc- vent such speculation. These forms of ownership tion process. (Indeed, in many slum-upgrading pro- do not necessarily prevent individual householders grammes, government financing for construction is from transferring their property, but they can de- still not included in the package of support to house- lay such transfers until the property’s full cost (or holds.) Yet the lack of such financing sometimes de- value) has been redeemed. Such alternative forms of layed construction, as households did not have the tenure and collective management can provide low- resources to build after paying their contributions income groups with protection from market forces to the cost of land and infrastructure, and such de- and support the building of a community, as well as lays, in turn, drew criticism. As pointed out above, improve affordability.70 however, the opposite was frequently true in other projects: households were able to raise considerable Long and renewable leasehold titles to land have oc- amounts of funding independently. Policy makers casionally been used for incremental housing, but should thus think carefully before offering access to they are rarely socially acceptable. The Sri Lanka credit in a support package, and if it is included, it Million Houses Programme in the 1980s issued should take the form of small short-term loans. plots on 20-year leases, as it was argued that the government should be able to use the land for other The credit needs of low-income families engaged in purposes if future circumstances so dictate. Initially the incremental construction of housing differ sig- there was no indication that this deterred house- nificantly from those seeking funding (a mortgage) holds from investing in the construction, extension, in conventional supply-driven housing. The large, and improvement of their houses. However, lease- long-term loans that enable conventional borrow- holders later pressured the government to extend ers to buy or build a home—and commit them to the tenure periods first to 30 and then to 50 years, sustained debt—are a burden for low-income fami- until finally all were transferred to freehold title in lies, who are making their way in the urban econ- 2006.71 omy and society. Incremental house construction requires flexible, relatively small short-term loans In conclusion, all the evidence to date suggests that a that are responsive to the intermittent demands of sustainable supply of well-located, affordable land is households’ changing fortunes and priorities. For essential for successful state-supported incremental instance, many years may elapse between the differ- housing initiatives. Innovative tools and techniques ent stages of house building—small loans may be are available to acquire land, while plot sizes and needed to waterproof a roof and larger ones may be densities can be manipulated to bring down costs. needed to build a second floor. Yet such financing is But ensuring the supply of land for low-income rarely available. The closest approximations tend to housing also requires a political will to reform urban be schemes that provide small-scale credit for enter- land markets. Community organisations and NGOs prise development, assessed on the risk rating of the can play a significant role in building and sustain- ing that political will over time and across changes 70 Payne et al, 2009. in political regime. 71 Wakely, 2008. 33 T H E C A S E F O R I N C R E M E N TA L H O U S I N G proposed business returns and not on the collateral To ensure that monetary support is invested in provided by the borrower’s property. house construction as intended—and as an alter- native to financial credit—building materials have Even the most ‘socially responsible’ banks are cau- been bought in bulk and passed on to authorised tious of treating a small plot in the middle of an house builders at or below cost. However, it has of- informal settlement or low-income housing estate ten been found that such schemes are open to ex- as collateral for a loan. The cost of repossession and ploitation. For instance, in the 1970s Camplands resale may well outstrip the value of the default. S&S project in Kingston, Jamaica, many household- However, there are indications that some progres- ers sold their project-allocated cement, steel, timber, sive banks are prepared to experiment with small and roofing sheets on the open market for a profit housing loans secured by evidence of household- and constructed their dwellings with new or second- ers’ ability to save money on a regular basis over a hand materials that they could acquire more cheaply sustained period of time, reinforced by community on the informal market. In other instances, this has organisations that can provide peer group support not been a problem and on-site organised depots of or pressure to underwrite repayments. The effec- building materials (who buy and store in bulk) have tiveness of this approach is widely demonstrated been able to lower the cost of construction materi- by the Asian Coalition for Community Action als and serve as one-stop shops for materials, advice, (ACCA) programme that is supported by the Asian and housing loans, though they have been known to Coalition for Housing Rights (ACHR) in 150 cit- drive out small-scale local suppliers who are a source ies 15 countries in the region. It is also the basis on of local employment and incomes.73 which successful ‘community banks’ such as the Sri Lankan Women’s Bank work with considerable In more recent programmes, independent credit success. facilities and management have been shown to be more efficient and effective than those administered There are financing (security) advantages to mak- by government. For example, the micro-credit pro- ing even initial starter loans for construction avail- grammes supported by the SIDA and managed by able on an incremental basis, requiring borrowers NGOs in Central America provide credit specifi- to “qualify” for the next stage of credit by complet- cally tailored to housing and infrastructure needs, ing the first. For example, the Sri Lanka Million with their administration providing technical advice Houses Programme, referred to above, issued credit to house builders along the lines described above.74 for construction in three instalments (foundations While NGOs in Bangladesh provide little credit spe- and floor slab, structure and eaves, and the roof) cifically targeted at housing, the loans they admin- that would cover the cost of a basic dwelling; any ister to foster income generation are often used to additions had to be financed by the house builder.72 fund housing improvements.75 The plethora of mi- This worked well even though, in the early stages cro-finance initiatives (both small and large in scale, of the programme, monitoring construction and and either community managed or supported by authorising successive instalments put a heavy ad- ministrative burden on already over-stretched gov- 72 Government of Sri Lanka, 1983. ernment housing officers. This responsibility was 73 Goethert, 2009. subsequently devolved to local community-based 74 Stein and Vance, 2008. organisations. 75 Rahman, 2005. 34 S ECTI O N 4 : CO MPO NENTS OF INCREMENTAL H OUSING ST RAT EGIES NGOs or banks) attests to the ability of poor people on.77 In the Khuda-ki-Bastee project in Hyderabad, to mobilise resources and apply long-term financial Pakistan, residents decided the extent to which they strategies at low risk to lenders. However, credit is were prepared to save in order to continue to up- rarely offered specifically for the construction, ex- grade neighbourhood services. In several blocks, the tension, and improvement of housing. funds accumulated exceeded requirements, which sped up the development process.78 4.3  Infrastructure and services But such decision-making may not always be prac- ticable where new communities are being formed. The success of incremental housing initiatives in Where decisions on the levels and type of infrastruc- large part depends on the timing, standard, and ture provision have to be made without community level of infrastructure and services provision. Where consultation, it is a mistake to assume that service projects have provided infrastructure and services at standards should always be low or that, unlike the too high a level, costs have proved unaffordable for dwellings they serve, services necessarily have to be low-income households, who have been bought out upgradeable over time. There are strong arguments by middle-income groups. Where infrastructure and in favour of providing high standards right from the service standards have been too low or their instal- start of an S&S or upgrading project. This has been lation delayed, plots have remained empty and have shown to stimulate good-quality construction and a failed to attract any income group. So a careful bal- sense of pride in the neighbourhood, which in turn ance has to be struck. motivates care and maintenance of public facilities.79 In addition, many low-income neighbourhoods need This is best done by adhering to the principle of sub- relatively large public spaces and children’s play ar- sidiarity. In theory, the level of infrastructure pro- eas, in part because of the low levels of private space vision can only be equitably and effectively decided in homes. This is particularly in cultures and cli- by the community of users, provided that they fully mates that value social interaction and outdoor liv- understand the trade-offs of initial capital costs, use ing. Furthermore, security lighting, bus-stop shelters, and maintenance costs, and the tenets of environ- and police posts all tend to be needed more in these mental health and safety. Community action plan- communities than in upper- and middle-income ning, pioneered in Sri Lanka in the 1980s, is one of a residential areas, where security is less of an issue.80 series of techniques to engage low-income commu- nities in establishing priorities and setting standards The costs of a high standard of initial provision for infrastructure and service provision based on a may be recovered in several ways. These costs can be thorough understanding of the costs and benefits of spread beyond the confines of the project through their decisions.76 local taxation schemes, or recovered over the long term via user charges. It has been observed that the In the S&S resettlement components of the 1970s project in Lusaka, Zambia, infrastructure layouts were planned by groups of 25 households sharing a 76 Hamdi and Goethert, 1997. 77 Schlyter, 1995. common sanitary facility. This joint decision-making 78 Siddiqui, 2005. was effectively used by community workers to devel- 79 Cotton and Franceys, 1991. op a sense of collective identity and solidarity early 80 UN-Habitat, 2007. 35 T H E C A S E F O R I N C R E M E N TA L H O U S I N G full cost-recovery required at the level of individual are rarely satisfactory and do not involve significant projects by aid agencies, notably the World Bank, has cost savings. Such attempts have, on occasion, even disadvantaged the poor beneficiaries of many S&S prompted bad political press and accusations of and upgrading projects, who are expected to bear unpaid “slave labour” practices. Nevertheless, many the capital costs of their infrastructure and service organisations continue to recommend that access to installations. Meanwhile, the costs of capital works basic services be generated largely through self-help in formal upper-income neighbourhoods is spread efforts and the use of community builders, as in the across the city as a whole, through the property tax case of Cooperative Housing Foundation (CHF) in system or levies on user charges. Ghana.84 Another way to keep the cost of infrastructure low is to A strategy to minimise the initial capital costs of in- depend on community labour for construction. This frastructure and services is to improve them incre- practice takes many forms. In some cases, volunteers mentally, alongside the construction of dwellings. organise (usually with NGO managerial assistance) The rationale for this approach is that while basic to provide “sweat equity,” contributing their time and services must be provided right from the beginning skills to community works.81In other cases, formal of an S&S project, they should be kept to a mini- contractors engage local labour in their workforces, mum—including, for example, communal water a strategy as much aimed at income generation as the points and pit-based sanitary facilities, unpaved procurement of infrastructure. Some communities roads, unlined drains, and so on. Such services may establish neighbourhood design and management then be upgraded over time in response to the de- and implementation committees, such as the highly velopment of a community’s cohesion, demand, successful “lane committees” that effectively serviced and increasing ability to pay for higher standards. the vast settlement of Orangi in Karachi, Pakistan, However, despite the savings in initial costs, there with virtually no public sector intervention.82 are drawbacks to this approach: (1) maintenance requirements are high where a lack of community Such participatory approaches to the procurement cohesion results in neglect, rapid deterioration, and of public infrastructure and services vary widely in vandalism; (2) basic public facilities do not instil the terms of efficiency, product quality, and cost sav- sense of pride needed for community development ings. Casual volunteering can work well in upgrad- and investment in house building; and (iii) new ing projects where an established community ex- neighbourhoods are launched with the stigma of be- ists and there is a strong, locally perceived need for ing “substandard.” the installation of infrastructure. In Aleppo, Syria, the occupants of many informal settlements have In sum, the planning of infrastructure works has to be successfully installed relatively sophisticated wa- managed carefully. A lack of coordination among the terborne sewerage systems with no formal techni- various implementation agencies responsible for in- cal assistance and at no capital cost to the govern- frastructure often leads to delays in service provision ment.83 However, where strong community ties do not exist, as in many S&S projects, and where there 81 Cotton and Sohail, 1997. is an entrenched perception that infrastructure pro- 82 Hasan, 2009. vision is a government responsibility, attempts to 83 GTZ, 2009b. mobilise voluntary labour and organise self-help 84 CHF, 2004. 36 S ECTI O N 4 : CO MPO NENTS OF INCREMENTAL H OUSING ST RAT EGIES and in the installation of water supply, sanitation, An interesting example of incremental service deliv- and access.85 Reduced service provision, as in the case ery is the ‘Aguateros’ of Asunción in Paraguay who of the Parcelles Assainies project in Dakar, stores up are relatively small scale entrepreneurs who provide problems for the future and on occasion justifies the potable pipe-born borehole water to some 400,000 fear of officials that their projects are little better than inhabitants of low-income settlements ( approx. 17 slums. In this project, because of cuts in the provision percent of all connections in the city) who would of education facilities and a lack of long-term plan- otherwise not be served by the municipal supply ning, today there are only 22 public primary schools system. Anticipating the informal or officially sanc- for a primary-school-age population of 87,000, with tioned occupation of peri-urban land, an Aguatero 80 students per classroom.86 sinks a borehole and starts to extend a polythene pipe network, making connections to client households at costs that are comparable to the subsidized official 4.4  The private sector public sector suppliers. They also provide credit on mutually agreed terms. A typical Aguatero serves up The formal private sector can have an effective role to around 100 customers on a single borehole. They in the installation of infrastructure and service deliv- are subject to official water quality certifications every ery in both upgrading and sites and services projects six months that guarantees their safety and constancy through conventional subcontracting arrangements. of supply. Aguateros have joined forces and created Virtually all official agencies are legally required to an association to protect their interest, strengthen contract formally registered companies or operate their public image, and prevent attempts by the large through recognized NGOs. Community groups and utility companies to drive them out of business87 individual households, however, are at liberty to em- ploy informal contractors who are not constrained Another innovative example of the formal private by the obligations of ‘formality’ (conditions of em- sector specifically addressing the construction needs ployment, quality control, capital recovery, tax ob- of low-income incremental development is that of ligations, etc), giving them a commercial advantage the Mexican cement and construction materials over their formal sector competitors. conglomerate Cemex through its ‘Patrimonio Hoy’ programme. This is a commercially profit-making Not only is the formal private sector’s involvement scheme in which householders pay approx. US$14 in low-income development inhibited by its com- per week over a period of 70 weeks for which they mercial disadvantage, it also tends to be intimidated receive architectural and engineering advice and by perceptions of the high risk of such involvement, building materials for the construction or exten- stoked by the reputation of slums as harbingers of sion of their dwellings delivered to their properties crime and extortion—perceptions that are often un- at stable prices as and when they are needed. If a justified. Nevertheless, there are opportunities for family’s fortunes change so that it cannot make pay- formal sector enterprises to participate productively ments or use or store building materials they can in incremental development processes, particularly ‘pause’ the process and ‘bank’ their materials. In the those that are prepared to the responded to the often relatively slow, irregular and unpredictable nature of 85 Srinivas, no date. incremental development by low-income communi- 86 Cohen, 2007. ties and households. 87 MIT (2001). 37 T H E C A S E F O R I N C R E M E N TA L H O U S I N G decade since its start in 2000 Patrimonio Hoy has urbanising countries, particularly those where the been used by over 250,000 low-income households indigenous private sector is weak, it has not seen the to construct 160,000 10m2 rooms to a total value of opportunities presented by low-income housing de- US$135million with a default rate on payments of velopment or has shied away from the perceived risks. less than one percent. Patrimonio Hoy beneficiaries Thus its role has tended to remain on the safe ground consistently claim that their building is easier, cheap- of government contracts or partnerships that do not er and of a better quality than if it had been done it require great entrepreneurial initiatives. Where they on their own. Cemex has benefited extensively from have innovated, as in the examples described above, a previously untapped market, creating solid and there is significant evidence of win-win outcomes expanding ‘brand loyalty’ and exploiting the com- that benefit both low-income households and com- pany’s reputation for social responsibility.88 munities and private sector enterprises. A third area of formal private sector engagement in affordable urbanisation, though not specifically 4.5  Beneficiary selection confined to incremental development, lies in the complexities of land markets for low-income hous- The selection of beneficiaries for existing settlement ing. Universally, there is a blurring of boundaries upgrading and S&S on new land presents a plethora between formal and informal dealings in land and of challenges as policy makers seek to balance priori- the role of agents and brokers in land and property ties, issues of political patronage, and the potential title transfers. An interesting solution to the dilem- for corruption. ma of the conflict between affordability and legal- ity is provided by ARGOZ a private for-profit firm The support given to upgrading existing informal established in 1977 in San Salvador, El Salvador. settlements should be based on an analysis of fac- ARGOZ identifies and purchases privately owned tual evidence such as geographic information sys- peri-urban land at agricultural prices, subdivides tem (GIS) indicators of “housing stress” and the it and lets plots on a ten-year rent-to-own basis to potential return on investment. Of course, a com- householders with monthly incomes of c.US$170. plex range of political pressures inevitably influences Payments range from US$5–20 per month, depend- the rationale of any such exercise, but it should not ing on the amortisation period. In addition ARGOZ negate it. While housing stress is relatively easy to provides purchasers with design and technical ad- appraise, social and economic indicators that can vice and helps secure the provision of urban in- be addressed by slum upgrading are less easy to frastructure and services. In the twenty three years evaluate. Similarly, potential returns on investment to 2000 over 300,000 low-income households had should be estimated with an eye to the complex web obtained secure title to affordable properties and of social costs and benefits as well as physical and ARGOZ’ assets had grown from US$50,000 to over environmental gains. Meanwhile, some settlements $140 million.89 are destined for demolition and their inhabitants to relocation due to danger to life and limb or totally In summary, the formal private sector has attributes inappropriate land use. of managerial expertise, access to capital and com- mercial networks that can contribute to low-income 88 Business in the Community (2010). incremental housing processes. However, in many 89 Ferguson and Navarete (2003); also MIT (2001). 38 S ECTI O N 4 : CO MPO NENTS OF INCREMENTAL H OUSING ST RAT EGIES S&S projects fall into two broad categories: (1) those changing the way in which its partners had to work, that re-house existing urban communities that have requiring shifts in conventional definitions of the to be relocated, and (2) open-access projects in household and eligibility criteria. With fixed abode, which any eligible household may apply for a plot. proof of residence, and income verification all in- The first is, in theory, relatively straightforward— creasingly difficult to obtain, there was a demand for people evicted from their previous dwelling or that new, more flexible (and perhaps less accountable) have had to abandon their homes due to disaster re- ways of thinking.92 ceive a plot and whatever other compensation has been agreed upon. But there are invariably winners Some S&S projects have been designed to accommo- and losers in this process, not only within the relo- date a range of low- and middle-income beneficiaries cated community but beyond it. For instance, disas- and to make plots available on the open market. With ter victims may end up better off than many who the hopes of encouraging a social mix and avoiding were more deserving but not affected by the disaster. the creation of a single-class neighbourhood, in some But this is a general dilemma and one not specifically cases upmarket plots are used to subsidise those related to the beneficiaries of incremental housing meant for lower-income groups. But such projects, programmes and projects. even those that meet their goals, have proved diffi- cult to administer and prone to exploitation. In the Eligibility criteria for beneficiary selection for open- Dandora project in Kenya, different plot sizes were access S&S projects are generally based on house- designed for a corresponding range of eligible in- hold income levels and assets. Defining and justify- come groups. The subletting of rooms was encour- ing such criteria can be difficult and have often led aged to supplement household incomes and provide to excessively complicated application and verifica- accommodation for families not willing or unable to tion procedures that are cumbersome to administer. get on the property ownership ladder. However, it As described above, the Khuda-ki-Bastee project was found that the lower-income householders tend- in Pakistan overcame such complications through ed to be squeezed out of the subletting market by the stringent processes of self-selection and by flood- better-off owners, many of whom developed their ing the market with small plots to satisfy demand. larger plots exclusively for commercial subletting.93 However, this process excluded many low-income families who were less desperate and less able to fend In sum, selecting the beneficiaries of open-access for themselves in the informal sector.90 Thus, while incremental housing projects is inevitably a delicate administration should be kept to a minimum and process that involves both commercial and political controls should be as flexible as possible, restrictions interests. Because of this, every effort has to be made may have to be imposed to ensure that government- to develop clear indicators, transparent procedures, supported incremental housing does not primar- and accountable management processes. Projects for ily benefit middle-income groups. In Senegal’s S&S the resettlement of established informal settlement programme, measures taken against speculation communities are somewhat easier, particularly as (included prohibiting the sale of plots for five years plus a fee for changing ownership) increased the to- 90 Hasan, 2000. tal plot cost to more than the purchase price of land 91 ENDA-RUP, 2005. in the open market.91 In the context of its work in 92 SIDA, 2007. Central America, the SIDA found that migration was 93 Lee-Smith and Memon, 1988. 39 T H E C A S E F O R I N C R E M E N TA L H O U S I N G the selection and plot-allocation process can, and sought to impose un-realistic (and un-affordable) should be, undertaken in close collaboration with standards on S&S projects—insisting, for example, communities and their leaders, if not exclusively led on large plot sizes, mandatory standard house de- by them. signs, high-quality construction materials, and low densities. In some cases, the private sector has also become involved in projects, further pushing up stan- 4.6  Site planning and building dards and costs.97 Some S&S schemes even prohibited controls and supports income-generating activities on residential plots, in- cluding the renting of rooms, thereby limiting the op- Site planning is almost invariably undertaken as a cen- portunities for residents to earn additional income to trally controlled technical service, though on a micro help cover the cost of their plots and houses.98 level it has occasionally been done with the participa- tion of project beneficiaries. In the Lusaka project, Planning for higher densities of land use can also groups of 25 households made decisions on the dis- reduce costs. Smaller plot sizes have been recom- tribution of public and private land and the layout of mended as part of a reform programme in Ghana99 common sanitary facilities.94 The distribution of land to reduce the cost of servicing land by achieving use, plot sizes, and access layouts are normally deter- greater economies of scale and enabling smaller in- mined by prevailing norms and regulations. However, fill sites that are already served by trunk infrastruc- incremental housing projects, which are treated as ture to be used. Changing building codes to enable experimental exercises, have often been used to test additional floors to be added over time can increase and/or demonstrate the rationalisation of excessively densities considerably (as seen in Dakar), but this generous planning standards, even while maintaining should be achieved without compromising safety, adequate health and safety conditions. The Sri Lanka especially in earthquake-prone areas. High-rise in- Million Houses Programme is an example. cremental housing is also a possibility, as demon- strated by the vast Ciudad Bachue project built in Allowing for mixed land use, both at the outset of Bogotá, Colombia, in the 1970s, but this can pose incremental housing projects and in their future long-term challenges as communities seek to expand development, is an important principle that applies and maintain their structural safety. to all low-income settlements. The extent to which low-income groups depend on home-based indus- As has been stressed in the earlier sections, building tries for their livelihoods and for their integration controls should be confined to those necessary to into the urban economy at large is now well under- ensure the health and safety of households and the stood.95 The provision of dedicated workshop space wider community, and to facilitate incremental con- within settlements is similarly important and invari- struction processes. Many early S&S projects were ably used by local residents, as in the Shivaji Park not successful because their target groups could not S&S project in Alwar, India.96 94 Jere, 1984. Planning and building standards also tend to present 95 Tipple, 2004. 96 Lall, 2002. problems. Unable to break away, either psychologi- 97 Van der Linden, 1992. cally or legally, from long-established planning and 98 Srinivas, no date. building codes, government officers have on occasion 99 CHF, 2004. 40 S ECTI O N 4 : CO MPO NENTS OF INCREMENTAL H OUSING ST RAT EGIES afford the cost of meeting the conditions set for their But there is still a need for development control func- development. The Dandora project required at least tions to police illegal development and unsafe build- two rooms to be completed using permanent mate- ings put up by unscrupulous speculators. Squatting rials (concrete or dressed stone) within 24 months of on land reserved for service buildings or public open plot allocation.100 By contrast, in the Khuda-ki-Bastee spaces is common. Speculators planning buildings scheme, the Hyderabad Development Authority de- to sublet often have little regard for the quality of liberately sought to free householders from planning construction—to the extent that low-income rental controls. Only the layout of schemes was fixed and housing constructed in official S&S projects as well absolutely no standards were imposed on the plan as in informal settlements can be unsafe. Ideally, or quality of the houses to be built.101 It was rightly the primary functions of development control po- assumed that self-builders are conservative and risk licing should be undertaken by the community of averse, tending to avoid new construction materials residents, though enforcement must remain the re- and methods.102 They naturally aspire to high stan- sponsibility of the state. While NGOs may be best dards and would not wilfully build dwellings that placed to provide neighbourhood-level planning are unsafe or a threat to the health of their families. and building advisory services and to strengthen the However, to achieve their aims, self-builders needed capacity of community-based organisations that ad- technical information and advice on how to evolve minister first-stage development control functions, their homes (often over very long periods of time), it is important that the government has the capacity which was largely provided by NGOs. to ensure their complicity. In sum, revised planning and building codes and procedures to support self-builders are needed in 4.7  Community organisation and many countries. These should be proscriptive, set- asset management ting the limits of good practice, and not the more common prescriptive legislation that stipulates what That a sense of ownership over local community fa- has to be done in detail, leaving little room for inno- cilities engenders a degree of collective responsibil- vation. Effective outreach is essential for households ity for their maintenance and management is now to be able to make informed and technically sound well understood. What is often less clear is the link choices and to achieve good returns on investment. between this sense and the participation of house- Thus, there are strong arguments for a shift from the holds in all stages of the project-planning process. concept of official developmental control to the es- Few of the upgrading and resettlement projects of tablishment of planning and building advisory ser- the 1970s and 1980s engaged with the communi- vices that provide technical guidance on good prac- ties served, least of all at the appraisal and planning tices specific to incremental housing programmes stages of project implementation. User needs and and projects. A building clinic that performed this demands for land and services were assumed with function was set up and staffed by architectural and little or no consultation (with the notable exception engineering students in the vast Thawra City incre- mental housing project in Baghdad, Iraq, for a short 100 Lee-Smith and Memon, 1988. time in the early 1970s.103 Some NGO-run urban re- 101 Siddiqui, 2005. source centres also provide this sort of on-site tech- 102 Goethert, 2009. nical assistance to house builders. 103 Wakely, et al, 1974. 41 T H E C A S E F O R I N C R E M E N TA L H O U S I N G of the Lusaka upgrading and S&S project referred to 4.8  Strategic planning above).104 In new open-access S&S projects, where the beneficiaries are not identified until after the For state-supported incremental housing initiatives site-planning stage, such participation is obviously to have a significant impact on the enormous low- not possible, so community building should be a income housing deficits of most cities and towns, high priority right from the first days of occupation. they must be located within a broader framework. At The administration of the Dandora S&S project in the national level, there needs to be a clear poverty Nairobi, Kenya, was conducted by the Community reduction strategy109 that recognises the detrimen- Development Department especially established tal significance of urban poverty on national and for the project.105 In the Khuda-ki-Bastee project regional development and the role that housing can in Pakistan, the most successful neighbourhoods in play in reducing it. At the level of the city, small-scale terms of developing and maintaining public infra- incremental projects that are divorced from the wider structure and services were those that received NGO housing market will become subject to speculation community development and management support and rapid gentrification. This was the case in Dakar, from the first days of settlement.106 where plots were quickly sold to higher-income groups because these groups had no safer investment Over time, community-based management organi- opportunities. The same problem is also evident in sations may develop to cater to local collective needs. Pakistani cities, where higher-income groups bought In Dakar, for instance, insufficient infrastructure plots intended for low-income housing development was installed and future growth was unplanned, but as a hedge against inflation. By contrast, in India’s cultural and religious organisations that developed Shivaji Park, one of the reasons for the high retention within the community gradually built the capacities rate of original low-income residents was Alwar’s required to deal with deficits in education, health fairly balanced housing market.110 Unfortunately, care, and rubbish and sewage collection.107 This was, most urban housing markets are far from balanced. however, a slow process that, with more support in the initial stages of the project, would probably have In Ghana, it has been recommended that non-con- been more effective. ventional strategies should be developed in each of the four different market segments to improve In short, it is essential that the capacity to support housing supply and finance. It has also been argued local organisations in the management and main- that within each sector there are different groups tenance of community assets be present right from of developers, households, and communities, each the start of the incremental development process. In of which drives housing supply—and that compre- most cases, this entails educative processes that in the hensive strategies must therefore involve all these long run lead to permanent local governance struc- actors. Projects should encompass the high- and tures.108 Clearly, these should be targeted around specific issues and user groups—for example, par- 104 Jere, 1984. ents’ associations concerned with school facilities 105 Lee-Smith and Memon, 1988. 106 Hasan, 2000. and committees concerned with the maintenance of 107 Cohen, 2007. roads, access paths, and drains—whose membership 108 Max Locke Centre, 2005. and roles will change as the incremental develop- 109 Gattoni, 2009. ment process evolves. 110 Lall, 2002. 42 S ECTI O N 4 : CO MPO NENTS OF INCREMENTAL H OUSING ST RAT EGIES upper-middle-income market to maximise the use Situating incremental housing initiatives firmly of local building materials and formal sector bor- within the context of broader land market reforms rowing. Recommendations have also been made for and programmes to facilitate housing supply for all the provision of S&S for lower-middle- and low- income groups is essential to their long-term success income groups, informal settlement upgrading, and and their potential to have a significant impact. wholesale lending among commercial banks, state banks, and microfinance institutions.111 111 CHF, 2004. 43 44  |  Section SECTION 5 Conclusions: Capacity Building & The Way Forward Experience and analysis has demonstrated the efficiency and efficacy of incremental housing for and by urban low-income households and communities. John Turner coined the phrase “housing as a verb” in 1972112 to focus attention on the processes by which urban low-income families house themselves, and away from the prevailing preoccupation with housing solely as a product. Yet, four decades later this understanding is yet to be absorbed by many national and municipal housing authorities where conventional attitudes to the clearance of ’slums’ and the unrealisable ambition to replace them with ready-built public housing persist. As a result informal settlements in many towns and passing responsibility to those people, communities cities continue to expand113. The principle of subsid- and enterprises to whom efficiency, effectiveness and iarity, set out in Section 3.3, and the need for the em- sustainability really matter. powerment of low-income communities and house- holds is often not understood, or is resisted because Enabling, on the other hand, is about ensuring that it is erroneously perceived to undermine the author- those who are empowered have the information, ity of established political interest groups. technology, skills, finance and supports to exercise their new authority responsibly. It is also about However, in many situations the dominant con- roles—who should do what and in partnership with straint to devolving responsibility for the produc- whom. It is building capacity—a process of equip- tion, maintenance and management of affordable ping all actors to perform effectively, both in their housing to urban low income groups is the lack of own job and by working in collaboration or partner- appropriate technical and professional resources ship with others who operate in other fields and at rather than the failure to devolve. It is more a prob- other levels of housing and urban development. lem of ‘enabling’ than one of ‘empowering’. Empowering is about devolving authority. It is about increasing the efficiency, enhancing the effectiveness 112 Turner, 1972. and ensuring the sustainability of development by 113 UN-Habitat, 2003. 45 T H E C A S E F O R I N C R E M E N TA L H O U S I N G 5.1  Capacity building At the start of the Sri Lanka Million Houses Pro- gramme in 1983, the National Housing Develop- To many, capacity building means only training or ment Authority had to re-train and ‘re-tool’ its en- human resource development. Certainly this inter- tire decentralised professional and technical staff. pretation is a very major component of it. However, Construction managers became the directors of if decision-makers, managers, professionals and support and supply teams providing advice, finance technicians are to operate at full capacity, they need and building materials to community-based endeav- to rely on more than just their own abilities. They ours; housing officers became the field administra- need an institutional and organisational environ- tors of incremental loan funds; technical officers ment conducive and supportive of their efforts, became the supervisors and trainers of household energies and skills. Institutional and organisational and community building processes. This was largely constraints present as great an impediment to the ef- a learning-by-doing-and-sharing process of capac- fective management of support to incremental hous- ity building that included the development of new ing processes as the incapacity of professionals, tech- job descriptions, formal training, and the opening of nicians and householders. Therefore to be effective new career opportunities114. capacity building must embrace all three aspects— human resource development, organisational devel- 5.1.2  Organisational development opment, and institutional development. Organisational development is the process by which things get done collectively within an organisation, 5.1.1  Human resource development be it a central government ministry, a local authority Human resource development is the process of department, a private sector enterprise, a non-gov- equipping individuals with the understanding, skills ernmental organisation or community group. It is to and the access to information and knowledge to en- do with management practices and procedures; rules able them to perform effectively. Because of the often and regulations; hierarchies and job descriptions— unpredictable nature of devolved informal settle- how things get done. It is also to do with working ment upgrading or the participatory development of relationships; shared goals and values; team-work, sites and services projects, the traditional boundaries dependencies and supports—why things get done. between professional and technical disciplines (for In many situations flexible and responsive man- example, architecture, planning, engineering, com- agement styles are needed, requiring entirely new munity development) become blurred and overlap. organisational structures, particularly within local Interdependent multi-disciplinary team work is es- government. It also often calls for the establishment sential. Therefore, in addition to acquiring new skills of new relationships between different organisa- that are needed to support incremental housing tions, for example those responsible for poverty re- approaches in their own discipline, managers and duction programmes, community development, en- technical professionals must also acquire a broad vironmental health, adult education, enhancing the understanding of the full range of issues and activi- role and opportunities for women that hitherto have ties outlined in Section 4 above concerning: land; had little engagement with housing departments or finance; infrastructure and services; beneficiary se- authorities. lection; planning and building controls; community organisation and asset management; and city-scale strategic planning. 114 Lankatilleke, 1986. 46 S ECTI O N 5 : CO NCLUS I O NS : CAPACIT Y BUILDING & T H E WAY FOR WARD As described in Section 2.2, in 1975 Nairobi City The ‘City Statute’ promulgated by the Federal Council in Kenya established a Project Department Government of Brazil in 2001, supported by a new (up-graded to a Housing Development Department Ministry of Cities and National Cities Council, es- (HDD) in 1978) to design and manage the vast tablished two years later, provides a ‘Toolbox’ of legal Dandora sites and services project to the east of the instruments that enable municipal governments to city. The creation of a whole new department was manage their own affairs within the tenets of Brazil’s deemed necessary because the management and pro- progressive 1988 national constitution. Emphasis is fessional skills and relationships that were required given to the social use of urban land with a particular differed significantly from those of the Council’s ex- focus on informal settlements and the land needs of isting Housing and Social Services Department and low-income households and communities. Though other related departments, notably Engineering and the City Statute has been subjected to some politi- Water & Sewerage. A pivotal component of the HDD cal opposition from entrenched conservative inter- was its Community Development Division that took est groups at all levels, there is little question that it on functions and professionals that had not existed has had, and continues to have, a significant impact in the City Council before. The HDD built its own at the urban grassroots level. Processes such as mu- capacity as it developed with considerable success— nicipal participatory budgeting, the regularisation of again, learning-by-doing—even though political tenure to land and property in informal settlements, rivalries between the new organisation and the lon- state-private-civil society partnerships in urban de- ger-established Council departments to some extent velopment and management have been enabled by obstructed its operation in the early stages115. legislation emanating from it116. 5.1.3  Institutional development Institutional development encompasses the legal 5.2  Priorities and regulatory changes that have to be made in order to enable organisations and agencies at all As indicated above, capacity needs to be built at ev- levels and in all sectors to enhance their capacities. ery level and across all fields of activity that impinge It embraces such issues as: regulations controlling upon the development and management of cities the financial management and the borrowing and and settlements. However, in every situation there trading capacity of government agencies and mu- are priorities which, for reasons of urgency or defi- nicipal authorities; the ability of local government ciency, take precedent over others in their need for to negotiate contracts and form partnerships with attention and resources. These vary with the particu- private enterprises and community organisations; lar circumstances of any specific country or region, land management, tenure and use regulations; though it has become increasingly apparent that the statutory building standards and other develop- weakest link in the chain is generally at the level of ment controls; and democratic legislation that al- local government and municipal administration. lows, enables and encourages communities to take responsibility for the management of their own Municipal governments and administrations are the neighbourhoods and services. Such institutional key actors in the management of towns and cities. Yet, and legal issues generally need the political and leg- islative authority of national government to bring 115 Lee-Smith & Memon, 1988. about effective changes. 116 Fernandes, 2010. 47 T H E C A S E F O R I N C R E M E N TA L H O U S I N G over the last 40 years, in all but a handful of countries medium scale enterprises’ and contractors’ clients as they have been starved of authority and resources. their own competitive ability. They have tended to be constrained by obsolete leg- islation, restrictive practices, outmoded equipment and inappropriately trained staff. Many of their tradi- 5.3  Conclusion tional development and management roles have been usurped or bypassed by central government corpora- Almost half the population of the developing world tions and utility companies—as well as by low-income live and work in towns and cities and a third of them households and communities that they have failed to (830 million) in informal settlements or slums117. serve. But the paradigms are changing and calling for Though there are many ‘slums of despair’—seem- an urgent and massive exercise in re-building the ca- ingly hopeless neighbourhoods of poverty and en- pacity of local government and administration. vironmental degradation—the majority are ‘settle- ments of hope’—informal neighbourhoods and Capacity building of community-based organisa- communities in the process of building their cities tions and local NGOs to support incremental hous- through their own endeavours and ingenuity. They ing processes is next in importance to that of formal demonstrate a process that has been shown to be both local government in the league of priorities for ca- effective and efficient in terms of its responsiveness pacity building in support of incremental housing to their occupants’ fluctuating needs and fortunes. processes. The emerging role of neighbourhood and However, they are often constrained by a lack of of- community groups, as a new tier of local governance ficial or recognised supports that would extend the that comes between individual households and mu- effectiveness and efficiency of incremental housing nicipal authorities, is almost without precedent. processes for the development of the city as a whole. Although urban community organisations are right- As pointed out above, the starting point for this is ly taking on many of the traditional management the understanding of the principle of subsidiarity functions of municipal authorities, it is important and a political will to devolve authority down to the that they remain ‘non-governmental’ so that they can level of organised urban communities, coupled with maintain an independent watchdog role over munic- investment in innovative capacity building. ipal authorities, holding them to account and guard- ing the demands and interests of their constituents. In recent years such approaches have been applied to the remedial upgrading of existing informal neigh- The private sector generally takes responsibility bourhoods in many cities. Less common are strate- for building and maintaining its own capacity to gies that address the growing needs of new low-in- compete. There are situations, however, where the come urban households through the provision of ap- informal private sector and some formal sector en- propriately located, affordable, serviced land—sites terprises need assistance in the form of legislative de- and services. A recent study in Bogotá, Colombia has regulation and incentives that encourage and enable shown that the cost of developing serviced land for them to enter the market for the production of low low-income housing is almost one third the cost of cost housing and infrastructure. In many situations regularising established informal settlements118. there is also the need for easy access to management training for small and informal sector enterprises. 117 UN-Habitat, 2003 (Second Edition, 2010). This is often as much in the interests of small and 118 Fenandes , 2011–05–21. 48 References Abrams, C. 1964. Housing in the Modern World: Burra, S. 1999. “Resettlement and Rehabilitation of Man’s Struggle for Shelter in an Urbanising World. the Urban Poor: The Story of Kanjur Marg.” DPU Cambridge, MA: MIT Press. Working Paper no. 99, London. ACHR (Asian Coalition for Housing Rights). Business in the Community (2010) http://www.bitc. 2004. “Negotiating the Right to Stay in the City.” org.uk/resources/case_studies [accessed Feb 2011] Environment and Urbanization 16 (1). pp 9–25 CHF. 2004. Strategic Assessment of the Affordable ADB (Asian Development Bank). 2005. Housing Sector in Ghana. Silver Spring, MD: CHF “Providing Affordable Housing for the Urban International. Poor in Philippines.” http://www.adb.org/media/ Articles/2003/3958_Philippines_Urban_Housing/. Choguill, C. L. 1995. “The Future of Planned Urban Development in the Third World: New Directions.” Adedeji , D. S., and A. Olufemi. 2004. “Planning In Housing the Urban Poor: Policy and Practice Policies and Affordable Housing in Nigeria: An in Developing Countries, ed. B. C. Aldrich and R. Analysis of Abuja Master Plan Scheme and the Re- Sandhu. London: Zed Books. Validation of Certificate of Occupancy.” http://www. york.ac.uk/inst/chp/hsa/autumn04/daramola.pdf. Cohen, M. A. 1983. “The Challenge of Replicability: Towards a New Paradigm for Urban Shelter in Angel, S., and T. Chirathamkijkul. 1983. “There Developing Countries.” Regional Development are Realistic Mechanisms for Sharing the land.” In Dialogue 4 (1): 90–99. Land for Housing the Poor, ed. Archer Angel et al. Singapore: Select Books. –––––––. 2007. “Aid, Density and Urban Form: Anticipating Dakar.” Built Environment 33 (2). Banks, N. 2008. “A Tale of Two Wards: Political pp145–156 Participation and the Urban Poor in Dhaka city.” In Environment and Urbanization 20 (2). pp 361–376 Cotton, A., and M. Sohail. 1997. “Community Partnered Procurement: A Socially Sensitive Option.” Buckley, R. M., and J. Kalarickal, eds. 2006. Thirty Journal of Appropriate Technologies for Water Supply Years of World Bank Shelter Lending, What Have We and Sanitation 16 (2): 24–27. Learned? Washington, DC: World Bank. 49 T H E C A S E F O R I N C R E M E N TA L H O U S I N G Cotton, A., and R. Franceys. 1991. “Services for Washington, D.C., 28 July. http://events.iadb.org/ Shelter.” In Liverpool Planning Manual 3. Liverpool: calendar/eventDetail.aspx?Lang=En&id=593. Liverpool University Press. Government of Sri Lanka. 1983. Housing Options and D’Cruz, C., and D. Satterthwaite. 2005. “Building Loans Package (HOLP): Million Houses Programme. Homes, Changing Official Approaches: The Work Colombo: National Housing Development of Urban Poor Organisations and their Federations Authority. and their Contributions to Meeting the Millennium Development Goals.” IIED Poverty Reduction in GTZ. 2009-a. “Memorandum on Sustainable Urban Urban Areas Working Paper 16, London. Development in Syria.” GTZ, mimeo, Damascus. E & U (Environment and Urbanization). 2009. –––––––. 2009-b. “Informal Settlements in Aleppo: “Getting Land for Housing: What Strategies Rapid Profile of All Settlements.” GTZ, Municipality Work for Low-Income Groups?” Environment and of Aleppo, mimeo. Urbanization Brief No.19. pp1–6 Hamdi, N., and R. Goethert. 1997. Action Planning ENDA-RUP. 2005. “Parcelles Assainies.” http://www. for Cities: A Guide to Community Practice. Chichester: hicnet.org/document.php?pid=2611. John Wiley & Sons. Ferguson, B., Navarete, J. (2003) ‘A financial frame- Hasan, A. 2000. Housing for the Poor: Failure of work for reducing slums: lessons from experience Formal Sector Strategies, 263–98. Karachi: City in Latin America’ in Environment & Urbanisation, Press. Vol.15, No.2, London. –––––––. 2009. Participatory Development: The Story Fernandes, E. (2010) “The City Statute and the le- of the Orangi Pilot Project—Research and Training gal urban order” in The City Statute: A Commentary, Institute and Urban Resource Centre, Karachi, Cities Alliance, Washington, pp.55–70 Pakistan. Oxford: Oxford University Press. –––––––. (2010) ‘The City Statute and the legal ur- Herald, The. 2009. “Co-operatives Build over 2,000 ban order’ in The City Statute: A Commentary, Cities Houses for Low Income Earners.” The Herald, May 4. Alliance, Washington, pp.55–70 IHC (International Housing Coalition). 2008. Gattoni, G. 2009. “A Case for the Incremental Multilateral and Bilateral Funding for Housing Housing in Sites-and-Services Programs.” Paper and Slum Upgrading Development in Developing presented at the IDB conference, Washington D.C., Countries. Washington, D.C.: IHC. 28 July. http://idbdocs.iadb.org/WSDocs/getDocu- ment.aspx?DOCNUM-2062811. Jere, H. 1984. “Lusaka: Local Participation in Planning and Decision-Making.” In Low Income Goethert, R. 2009. “Brief Comments on ‘A Case Housing in the Developing World: The Role of Sites for the Incremental Housing Process’ by Goerge and Services and Settlement Upgrading, ed. G. Payne. Gattoni.” Presentation at IDB conference, IDB, Chichester: Wiley. 50 REFERENCES Kearne, D. H., and S. Parriss. 1982. “Evolution of Pressures are Mounting.” Engineering News, Shelter Programs for the Urban Poor: Principal Johannesburg. http://www.engineeringnews.co.za/ Findings.” World Bank Staff Working Paper no. 547, author.php?page=49&u_id=133& (screen 50, 11 World Bank, Washington, DC. July) Lall, S. 2002. “An Evaluation of a Public Sector Low Omenya, A. 2004. “Housing Networks in Different Income Housing Project in Alwar: India.” Working Political Economies: The Cases of Ivory Park, Paper 6, Society for Development Studies, New Delhi. Johannesburg and Dandora, Nairobi.” Paper pre- sented at the Institute of Housing for South Africa, Lankatilleke, L. (1986) “Training and Information for International Housing Conference on Housing in Institutional Development for the Implementation the Next Decade: Quo Vadis? Cape Town, October of the Million Houses Programme of Sri Lanka”, 3–6. Habitat International, Vol.10, No.3, pp.109–129 Patel, S., and D. Mitlin. 2004. “The Work of SPARC, Lee-Smith, D., and P. Memon. 1988. “Institution the National Slum Dwellers Federation and Mahila Development for Delivery of Low Income Housing: Milan.” In Empowering Squatter Citizen: The Roles An Evaluation of Dandora Community Development of Local Governments and Civil Society in Reducing Project in Nairobi.” Third World Planning Review, Urban Poverty, ed. D. Mitlin and D. Satterthwaite. Vol.10, No.3: 217–38. London: Earthscan. Mangin, W. 1967. “Latin American Squatter Payne, G., A. Durand-Lasserve, and C. Rakodi. 2009. Settlements: A Problem and a Solution.” Latin “The Limits to Land Titling and Home Ownership.” American Research Review Vol.2, No.3 : 65–98. Environment and Urbanization 21 (2), London, pp.443–462. Max Locke Centre. 2005. The Rough Guide to Community Asset Management. London: MLC Press. Pervaiz, A., P. Rahman, and A. Hasan. 2008. “Lessons from Karachi: The Role of Demonstration, McLeod, R., and K. Mullard. 2006. Bridging the Documentation, Mapping and Relationship Building Finance Gap in Housing and Infrastructure. London: in Advocacy for Improved, Urban Sanitation ITDG Publishing. and Water Services.” IIED Human Settlements Discussion Paper, Water-6, London. MIT (2001) Upgrading urban communities: a re- source for urban practitioners. http://web.mit.edu/ Phonphakdee, S., S. Visal, and G. Sauter. 2009. “The urbanupgrading/index.html [accessed Feb 2011] Urban Poor Development Fund in Cambodia: Supporting Local and Citywide Development.” Mitlin, D. 2008. “Urban Poor Funds: Development by Environment and Urbanization 21 (2), London, the People for the People.” IIED Poverty Reduction pp.569–586. in Urban Areas Working Paper 18, London. Rahman, M. 2005. “Role of NGOs in Urban Housing Ndaba, D. 2008. “South Africa’s Affordable- for the Poor in Dhaka, Bangladesh.” Global Built Housing Programme is Maturing but the Delivery Environment Review 15 (1): 16–29. 51 T H E C A S E F O R I N C R E M E N TA L H O U S I N G Riley, E., and P. Wakely. 2005. Communities and Srinivas, H. “Sites and Services.” Global Communication: Building Urban Partnerships, 85– Development Research Centre. http://www.gdrc. 110. Rugby: ITDG Publishing. org/uem/squatters/s-and-s.html. Rowbottom, S. 1990. “Senegal Case Study, Parcelles Stein, A., and I. Vance. 2008. “The Role of Housing Assainies: From Project to Place—After Three Finance in Addressing the Needs of the Urban Poor: Decades.” CHF Occasional Paper, International Lessons from Central America.” Environment and Development Matters: Framing a New Perspective Urbanization 20 (1) (London) pp.13–30. on Urban Development: Back to the Future, volume 3, CHF International, June. Tipple, G. 2004. “Settlement Upgrading and Home- Based Enterprises: Discussions from Empirical Schlyter, A. 1995. “Squatter and Slum Settlements Data.” Cities 21 (5): 371–79. in Zambia.” In Housing the Urban Poor: Policy and Practice in Developing Countries, ed. B. C. Aldrich Turner, J. F. C. 1967. “Barriers and Channels for and R. Sandhu. London: Zed Books. Housing Development in Modernising Countries.” Journal of the American Institute of Planners 33 (3), SIDA (Swedish International Development Agency). pp.167–181. 2007. “Increasing Access to Housing and Financial Services: SIDA Experiences in Central America.” –––––––. (1972) “Housing as a Verb” in Turner, J.F.C. 1st Asia-Pacific Housing Forum, 24–26 November, & Fichter, R. (eds), Freedom to Build, Macmillan, Singapore. New York, pp.148–175. Siddiqui, T. A. 2005. Incremental Housing Development –––––––. 1976. Housing by People. London: Marion Scheme (Pakistan): An Innovative and Successful Boyars Publishers. Scheme for Sheltering the Urban Poor. Karachi: Action Research for Shelter. UNCHS (United Nations Centre for Human Settlements). 1991. The Incremental-Development Silas, J. 1984. “The Kampong Improvement Scheme—A Case Study of Khuda-ki-Basti in Programme of Indonesia: A Comparative Case Hyderabad, Pakistan. Nairobi: UNCHS, Nairobi. Study of Jakarta and Surabaya.” In Low Income Housing in the Developing World: The Role of Sites UN-Habitat. 1993. The Urban Poor as Agents of and Services and Settlement Upgrading, ed. Payne. Development: Community Action Planning in Sri Chichester: G. Wiley. Lanka. Nairobi: UN-Habitat. Soliman, A. M. 1995. “A Tale of Informal Housing –––––––. 2003. Global Report on Human Settlements: in Egypt.” In Housing the Urban Poor: Policy and The Challenge of Slums, Chapter 5, 79–95. London: Practice in Developing Countries, ed. B. C. Aldrich Earthscan. and R. Sandhu. London: Zed Books. 52 REFERENCES –––––––. 2007. Enhancing Urban Safety and Security: Poverty and Environment Report Series 6, Ottawa, Global Report on Human Settlements, pp.96–107. Canada. van der Linden, J. 1986. The Sites and Services Wakely, P., and A. Aliani. 1996. “Where We Have Approach Reviewed. Aldershot: Gower. Come From: Historical Perspective and Major Trends.” In UN-ESCAP “Living in Asian Cities.” New –––––––. 1992. “Back to the Roots: Keys to Successful York: UNDP. Implementation of Sites-and-Services.” In Beyond Self-Help Housing, ed. K. Mathéy, 341–53. London: Wakely, P., H. Schmetzer, and B. Mumtaz. 1974. “A Profil Verlag. Building Clinic in Baghdad.” In Architectural Design 6 (74) (London), pp.627–629. Wakely, P. 2008. “Land Tenure in Under-Served Settlements in Colombo Sri Lanka.” IDRC Urban 53 54  |  Section Bibliography Housing policy A review of the sites and services approach within the context of a discussion of 20 years  Buckley, R. M., and J. Kalarickal, eds. 2006. of housing policy. The article reveals the pow- Thirty Years of World Bank Shelter Lending, erful political, administrative, and economic What Have We Learned? Washington, D.C.: interests that acted against the large-scale im- World Bank. http://siteresources.world- plementation of such project and favoured the bank.org/INTHOUSINGLAND/Resourc- status quo of informal self-help housing devel- es/339552–1153163100518/Thirty_Years_ opment. Shelter_Lending.pdf.  UN-Habitat. 2003. The Challenge of Slums, An extensive review of the World Banks’ Global Report on Human Settlements 2003. record in the shelter sector, based on the out- London: Earthscan. comes of over 278 projects in 90 countries, Broad review of the phenomena of slums, this examines the evolving objectives and out- their types, causes, context, and workings, as comes of Bank projects. Policy instruments well as policy responses to them. Illustrated are also reviewed in the areas of land market with city case studies from around the world. issues, housing finance, housing subsidies, and the Banks’ direct involvement in low-income Self-help housing processes housing.  IHC (International Housing Coalition). 2008.  Pugh, C. 2000. “Squatter Settlements: Their Multilateral and Bilateral Funding for Housing Sustainability, Architectural Contributions and and Slum Upgrading Development in Develop- Socio-Economic Roles.” Cities 17 (5): 325–37. ing Countries. Washington, D.C.: IHC. Examines the reasons for urban squatter An overview of funding trends for slum settlements and the objectives of policy re- upgrading and housing finance by bilateral sponses in terms of improving sanitary services donors and multilateral organisations and the and legality. The article looks at the dilemmas results of interviews with representatives of of institutional and organisational approaches foreign assistance agencies. to the improvement of squatter settlements.  Nientied, P., and J. van der Linden. 1988. “The  Payne, G. (ed) (1984) ‘Low income housing ‘New’ Policy Approach to Housing: A Review in the developing world: The role of sites and of the Literature.” Public Administration and services and settlement upgrading’, Wiley, Development 8 (2): 233–40. Chichester. 55 T H E C A S E F O R I N C R E M E N TA L H O U S I N G Though nearly 30 years old, this book the Urban Poor in Developing Countries. Lon- provides a range of still relevant strategic ap- don: Earthscan. proaches to incremental housing by govern- Presents and analyses the main conclusions ments working with low-income communities, of a comparative research programme on land pointing to transferable successes and drawing tenure issues at a global level, looking at how cautionary lessons from practice. innovative solutions can be found and imple-  Tipple, A. G. 2000. Extending Themselves: User mented to respond to the demands and needs Initiated Transformations of Government Built of the majority of urban households living in Housing in Developing Countries. Liverpool: informal settlements. Liverpool University Press.  E&U. 2009. “Securing Land for Housing and Examines the scale and nature of housing Urban Development.” Environment and Ur- shortages and poor people’s coping strategies. banization 21 (2). The book looks at the meaning of housing to Various articles that examine, using case occupants, its financial importance, its impact studies, issues surrounding poor people’s ac- on sustainability, how poor people transform cess to urban land for housing, including con- their housing, and how they can be supported flict over land, effective strategies for acquiring to do so in a way that improves the productiv- land, and the roles that community-based or- ity of houses and their occupants. ganisations can play.  Turner, J. F. C. 1976. Housing by People. Lon-  Payne, G. 2002. Land, Rights and Innovation. don: Marion Boyars. London: ITDG. A classic book that examines the self-help Looks at 15 cases from around the world housing strategies pursued by poor people, where authorities have recognised the com- showing that such strategies can be highly or- plexity of land problems and evolved practical, ganised, flexible, and effective. innovative approaches to providing tenure for the urban poor. Also includes a chapter on the Housing finance legal issues of security of tenure.  E&U (Environment and Urbanization). 2008. Urban management “Finance for Housing, Livelihoods and Basic Services.” Environment and Urbanization 20 (1).  Cooke, B., and U. Kothari, eds. 2001. Partici- Numerous articles on finance for shelter, pation: The New Tyranny? London: Zed Books. including microfinance for housing improve- Examines the rhetoric of participation and ment, community development, and service the reality that it can be exploitative of poor provision and financial exclusion, with case people, showing the complexities of social and studies from many countries and regions. political relations dominated by inequality while advocating a deeper form of empower- Land ment.  Chambers, R. 1997. Whose Reality Counts? Put-  Durand-Lasserve, A., and L. Royston, eds. 2002. ting the First Last. London: Intermediate Tech- Holding Their Ground: Secure Land Tenure for nology. 56 BIBLIOGRAP H Y Argues that the central issues in develop-  Jenks, M., and R. Burgess, eds. 2000. Compact ment have been overlooked and that profes- Cities: Sustainable Urban Forms for Developing sionals need new approaches and methods for Countries. London: Spon Press. interacting, learning, and knowing in order Examines sustainable urban forms in devel- for the skills and knowledge systems of poor oping countries, looking at the implications of people to receive greater recognition. the compact city debate on various countries  Development Planning Unit. 2002. Sustainable and regions, and the need for compact devel- Urbanisation. London: Development Planning opment—including from an environmental Unit. perspective. Contains a section on infrastruc- Advocates sustainable urban development ture and transport. through linking the green and brown agendas; examines more than 70 case studies. Infrastructure and services  E&U. 2008. “City Governance and Citizen.” En- vironment and Urbanization 20 (2).  Cotton, A., and K. Tayler. 2000. Services for the Articles that illustrate the social and po- Urban Poor: Guidelines for Policymakers, Plan- litical basis for citizen action to reduce urban ners and Engineers. Loughborough: WEDC/ poverty, looking at new approaches for people- GHK. centred development and for building bridges A set of manuals that offers detailed guid- between citizens and the state. ance on the planning, design, implementation,  ———–. 2009. “City Governance and Citizen operation, and maintenance of basic services Action II.” Environment and Urbanization 21 for the urban poor—based largely on experi- (2). ence from South Asia. It emphasises the need Examines the role that local authorities can to integrate participatory approaches at the lo- play in strengthening local democracy and the cal level with strategic improvements to city- relationships between government and citi- level infrastructure. zens, with case studies from various countries and regions. 57 CITIES WITHOUT SLUMS 1818 H Street, N.W. | Washington, D.C. 20433 | USA Tel: 202.473.9233 | Fax: 202.522.3224 Iinfo@citiesalliance.org | www.citiesalliance.org