Getting ahead of the game: A twin track approach to improving existing slums and reducing the need for future slums

Transcription

1 Getting ahead of the game: A twin track approach to improving existing slums and reducing the need for future slums Paper presented at the Second World Bank Urban Research Symposium: Urban Development for Economic Growth and Poverty reduction December Washington DC, USA Geoffrey Payne Geoffrey Payne and Associates 34 Inglis Road London W5 3RL UK gkpayne@gpa.org.uk Abstract Whilst many of the Millennium Development Goals represent an ambitious attempt to reduce global poverty and improve the quality of life for the world s poor, this cannot be said for the goal concerned with urban development. Goal 7 of the MDGs aims to improve the living conditions of 100 million slum dwellers by However, current estimates suggest that there are presently nearly one billion people living in slums and that this is projected to increase to 1.5 billion during the same period and to 2 billion by Even achieving this MDG is therefore to manage a retreat rather than achieve significant progress. The important policy objective for the international community and governments is therefore to dramatically exceed, not just meet, this MDG. This can best be achieved by adopting a twin-track approach towards existing and potential future slums. For the former, innovative approaches to improving tenure security in existing unauthorized 1

2 settlements can improve living conditions, whilst revising regulatory frameworks can reduce the need for future slums by significantly improving access to legal land and shelter. The paper provides evidence from recent and ongoing research which suggests that intermediate tenure options, combined with regulatory audits of planning regulations, standards and administrative procedures can significantly improve living conditions within the human, technical and financial resources available. Introduction: The first step in solving a problem is to define it correctly. After much debate, the international community and member states have declared that the main objectives driving development policy for the coming decades are to be defined by a set of goals and targets, collectively known as the Millennium Development Goals. These set global targets for different sectors to be achieved by specified dates. Whilst many would go a long way to reducing global poverty and improving the quality of life for the world s poor, this does not apply in the case of urban development, for which there are two key targets. The most significant is Goal 7, in which target 10 seeks to halve by 2015 the proportion of people without sustainable access to safe drinking water and basic sanitation and target 11 which seeks to have achieved by 2020 a significant improvement in the lives of at least 100 million slum dwellers. 2

3 These targets have to be seen in the context of the present and projected reality. UN- HABITAT estimates that there are currently 924 million slum dwellers in the world and that without significant intervention to improve access to water, sanitation, secure tenure and adequate housing this number could grow to 1.5 billion by Other estimates indicate that the total could increase to 2 billion by To achieve MDG target 11 would therefore meet only 10.8% of existing needs and only 7% of future needs by In other words, even if the MDG target is achieved, by 2020 there could be 1.4 billion people living in slums and squatter settlements compared to the 924 million at present, an increase of 162%. Whilst the target may therefore be achievable, it does not represent an appropriate or adequate definition of the challenge facing the international community, national governments, civil society groups and professionals. In fact, it suggests that a target-driven policy agenda is detracting attention from the real issues which need to be addressed. Since the MDGs are also global targets, they also fail to provide a basis for policy at national or local levels. The real challenge is two-fold. First, there is a need to improve the living conditions of far more than 100 million people living in slums and various types of unauthorised settlements. Second, there is an equally urgent need to create conditions in which all sections of urban society, especially the poorest and most vulnerable, can obtain access to legal, affordable and appropriate shelter in ways which prevent the need for future slums and unauthorised settlements. To add to the challenge still further, both these objectives need to be achieved in ways which provide adequate levels of security 3

4 and access to livelihoods, services and credit. Even if the MDG target is achieved, it will not resolve the first of these two challenges and do nothing to address the second. A twin-track approach Meeting present needs to improve living conditions for existing slum populations will in itself be a daunting task. However, it will be inadequate unless equal efforts are made to reduce the need for future slums. A twin-track approach is therefore vital if we are to turn things around and get ahead of the game. What are the available options? First, it is important to identify the existing and potential roles of the key stakeholders the poor themselves, national and local governments, civil society groups, the private sector and the international community. Second, it is important to assess ways in which the relative strengths of each stakeholder group can be combined to maximise synergies between their contributions. In terms of upgrading existing settlements, a major issue is that the poor, especially the very poor, need to live close to locations where they can earn a livelihood. As Turner (1967,1972) noted many years ago, this is more important than having permanent security of tenure or formal credit. Yet the locations where employment prospects are greatest are invariably the locations where competition, and therefore land prices, are greatest. To upgrade such settlements therefore raises two issues: If the upgrading is undertaken in ways which grant full titles, the residents will acquire, freely or at a nominal cost, an asset which can command a high price in the formal land market. 4

5 Experience shows that under such conditions, many households sell the newly acquired asset, realise its capital value and re-squat hoping to repeat the process. Such actions may therefore actually result in increased informal settlements rather than their decrease. They may also result in the eviction of tenants and increased litigation, especially where land records are unclear. How can poor households be able to remain in areas where they can generate a livelihood? How can this be achieved without distorting land markets and creating ghettos of the poor within urban centres? There are no universally applicable answers to these questions. However, experience suggests that an important ingredient is to provide a form of tenure to the residents of existing unauthorised settlements which is sufficient to ensure protection from eviction, together with property rights 1 and regulatory frameworks which entitle people to use their dwelling for a range of purposes and be able to obtain basic services. In some cases, it might also facilitate high density forms of multi-occupancy rather than individual houses on individual plots. The provision of secure tenure can take a variety of forms from a simple moratorium on relocations and evictions, to temporary occupation licences, communal or individual leases, community land trusts, customary tenure, etc. The duration of such tenure forms may be short in some cases, or almost permanent in others. They may provide the option of extension or upgrading to more formal tenure systems over time, or 1 Tenure and property rights are not the same. Tenure relates to the mode by which land is held or owned or the relationship among people concerning land or its product. Property rights refer to a recognised interest in land or property vested in an individual or group and can apply separately to land or development on it. 5

6 compensation for investments made. The key objectives should be to provide adequate security and maintain or increase access by the urban poor to locations where they can increase their incomes. It is also important to reduce the attraction of higher income groups to buy out low-income residents. If such measures are combined with modest planning and building regulations and standards, the subsidy burden or opportunity cost of such measures can be further reduced and a virtuous cycle established between tenure and regulatory regimes. These can be reinforced still further if administrative procedures can be revised to lower the cost of entry to legal shelter (see Payne, 2001) thus reducing the need for new slum formation. The complementary component to upgrading in a twin-track approach is to reduce the need for future slum formation. Among the constraints to accessing affordable legal shelter, the key factors are invariably cost and regulatory frameworks, especially planning and building regulations, standards and administrative procedures. Unfortunately, these often tend to combine in that high standards impose higher costs and complex bureaucratic procedures impose delays which require informal payments to facilitate progress. In his seminal publication on regulatory frameworks, de Soto noted (1989:24) that it took 159 bureaucratic steps to legalise informal settlements in Lima and that this took on average twenty years. In a similar vein, Struyk et al (1990) found that land transfers took an average of 32.5 months for a title to be issued in West Java and estimated that this added between percent to the cost of land acquisition. 6

7 Such procedural constraints were found in the other countries, though high planning standards, often based on inherited or imported norms, rather than local needs or realities, provided further barriers. For example, the minimum official plot size in many countries is significantly higher than is regularly found in informal settlements and therefore requires a higher land price than many households can afford. It also discourages the private sector from being able to meet the needs of lower income households on a financially viable basis. Another key cost factor in official standards relates to road reservations. These are often more demanding of land areas than are required in the capital cities of Europe, where car ownership levels are dramatically higher than found throughout suburban areas of developing countries. Such land areas are not only expensive in terms of capital (or opportunity) costs, but also impose high maintenance burdens on local authorities which many are unable to meet. Finally, planning and building regulations add a third tier to the barriers obstructing access to legal shelter. As McAuslan (1989) has noted in the case of Chennai, (previously Madras) India so concerned have the authorities been to close every loophole against illegal development, corruption, exploitation of scarce resources, the exercise, and therefore the possible wrongful or non-exercise, of discretion, that the principal aim of the Madras Metropolitan Development Authority to get orderly and equitable development underway in Madras and its environs has been lost site of. McAuslan also argues that planning itself has all too often been based on the needs and interests of the political elite rather than the majority, a tendency which can actively 7

8 discourage low-income households from living in close proximity to more fortunate groups except, of course, to proved them with services. Ongoing research in six countries 2 has provided similar evidence of the extent to which regulatory frameworks raise the barriers to accessing legal shelter to levels which most poor households, an increasing proportion of middle-income groups, are unable to overcome. As globalisation reduces the opportunities for government to directly control land and housing markets, and accelerates de-regulation in key economic sectors, the role of the regulatory framework in managing urban growth takes on a greater importance, especially as it is one of the few policy instruments determined and controlled by government. In Lesotho, for example, it has been found that more than a dozen administrative steps need to be followed if a person is to meet all the legal requirements to obtaining land. The system is highly centralised with all grant applications requiring Ministerial approval. In Tanzania, 13 steps, taking seven or more years, can elapse between the identification of an area for implementing a planning scheme to the time letters of offer are issued. Each of these steps has its own internal sub-steps. The irony is that regulatory frameworks established to achieve planned urban development have widely become a means of preventing this. 2 Regulatory guidelines for affordable shelter an ongoing research project directed by Geoffrey Payne and Associates and funded by the UK Department for International Development. The project includes case studies in Bolivia, India, Lesotho, South Africa, Tanzania and Turkey. 8

9 Applying new approaches in Cambodia A twin-track approach is presently under preparation in Cambodia in a number of urban projects, one of which is being supported by Cities Alliance, GTZ and UN-Habitat 3. In this project, it was initially proposed to improve tenure security and living conditions for all households in informal settlements by providing them with a Temporary Occupation Licence similar to that used for informal businesses in Kenya. However, it was realised that the administrative burden of identifying eligible families and issuing them all with TOLs would have been excessive. It was therefore proposed to announce a Moratorium on Relocations and Evictions (MORE), provisionally for a period of six months. This would enable people to go to work in the morning, secure in the knowledge that their home and possessions would still be there when they returned. The authorities would then be able to use the time gained to agree criteria for determining which settlements could be regularised and upgraded on a long term basis and identify sites for moving the families of settlements considered unacceptable. Finally, longer term tenure options could then be formulated for introduction at the end of the moratorium period. At the same time, a regulatory audit was undertaken of planning regulations, standards and administrative procedures to identify options for reducing the costs of new legal housing developments and thereby reduce the need for informal settlements in the 3 The local consultant to the project is Mr Din Somethearith. He is supported by a number of local advisers including Mr George Cooper and Dr B.H.S. Khemro. The international consultant to the project is Geoffrey Payne. 9

10 future. This revealed a number of areas where modest short term changes and more substantial longer term reforms could be beneficial (see Appendix 1 for details). However, as John Lennon noted, life is what happens when you are busy making plans. In the event, things moved rapidly and in May 2003, the Prime Minister announced a plan to upgrade 100 settlements a year in Phnom Penh for the next five years. Assuming a notional 100 families in each settlement (most have many more than this), this will involve improving the conditions of 10,000 households annually. If individual titles were to be allocated, this would involve issuing an average of one title every 12 minutes for five years assuming 50 weeks a year and 40 hours work a week (without lunch or toilet breaks!) 4. The administrative burden therefore remains a key consideration, especially since the status of land ownership in some areas remains unclear. In the absence of clear criteria for determining which settlements are upgraded and which are scheduled for relocation, there is a risk that pre-emptive action will be taken to remove some settlements and that others will be selected for upgrading according to political, rather than more objective, considerations. For these reasons, it was decided to revive the proposed Moratorium on Relocations and Evictions (MORE) to provide short term tenure security for all households and to introduce new forms of longer term tenure in pilot projects for both upgrading and new urban housing. 4 The COFOPRI land titling programme in Peru claimed to have allocated 1 million titles in four years, or an average of 250,000 a year. Again, assuming a 50-week working year and a 40-hour week, this would have entailed COFOPRI issuing more than two titles a second for 4 years. Although the COFOPRI programme was undoubtedly successful in many ways, many titles were actually issued earlier by local authorities. 10

11 The audit confirmed that many of the planning regulations determining acceptability were based on conditions applicable when Cambodia was a French colony and there was no significant pressure on land. Thus, setbacks from railway lines, levels of ground floor plot coverage, land use restrictions, road reservations and complex administrative procedures for obtaining permissions and certificates, etc, were excessive given current realities. In fact, it appears that some standards are even higher and more onerous than those required in western Europe. The gap between regulations and realities has inevitably been filled by the people themselves and land not allocated for productive use has been extensively settled throughout Phnom Penh. As a result, only a small proportion of the population even attempt to follow official procedures, thereby adding to the national proportion of officially unplanned settlements and offsetting efforts to upgrade the existing ones. The research on tenure security and regulatory frameworks reinforced the conviction that both subjects are central to the effective implementation of a twin-track approach to upgrading existing settlements and developing new ones. These are, after all, two of the major policy instruments directly controlled by central and local governments and enforceable even where market forces are otherwise pre-eminent. Tenure and regulatory reform also offer the prospect of achieving change at the scale required. How can such an approach be implemented? The key initial consideration is to identify local champions for change. The idea for the tenure moratorium was made by a local 11

12 official and taken up by the research team. The Prime Minister and the Governor of Phnom Penh are both supportive of initiatives which can be seen to improve living conditions for the majority whilst at the same time increasing local and international investment in the local economy. Local offices of international agencies, particularly UN- Habitat and GTZ, are supportive of opportunities to transfer innovative practices from international experience. Likewise, local and international NGOs have combined to exert pressure on government but also agreed to work with government in order to effect change. The key is then to maximise synergies between all these actors. Proposals for tenure At the time of writing, it has been agreed that new forms of tenure will be tested in pilot projects for upgrading and new development. Given the reservations about titling as the main option for providing tenure security expressed above, and the limited capacity of the land administration system, it has been proposed to introduce MORE as soon as possible for all unauthorised settlements in Phnom Penh. The period proposed is for six months. At the same time, it is proposed to introduce community land leases or trusts in settlements scheduled for in-situ upgrading and tenure improvement. Evidence from previous research (see Khemro, forthcoming 2004) demonstrated that low-income households are modest in their tenure needs and do not necessarily require titles as long as they can be guaranteed reasonable access to employment locations. It is therefore proposed to introduce twelve-year community leases and ten-year sub- 12

13 leases to individual households. This is considered adequate to encourage those with funds or access to credit to invest in home improvements without raising land values to the level at which downward raiding by higher income groups would become widespread. This will also hopefully enable these areas to be accessible to future lowincome households if any existing residents leave. It is proposed that households pay a modest amount for such leases, in return for which they would receive social facilities and other improvements which they would help determine. At the termination of the lease period, the community would then be entitled to one of three options: an extension of the lease for a further ten year period on revised terms to be defined; upgrading to a more long term or individual tenure system; or termination with compensation at full market value at the date of termination for the improvements made to the land. Given that it is unlikely that local authorities would possess sufficient funds to pay compensation on more than a token number of cases where there was a genuine public interest involved, this is considered sufficient to grant security of tenure to residents of most unauthorised settlements without distorting urban land and housing markets or imposing an excessive administrative burden on the urban authorities. The proposal has received support from the Minister of Land Management, Urban Planning and Construction, the Governor of Phnom Penh and other key stakeholders and it is hoped to introduce it in pilot projects during the next few months. When promoting the option of communal land leases, it was assumed that there would be many examples in other countries from which the project team and local officials 13

14 could learn. Surprisingly, this proved not to be the case and despite efforts to identify other examples only Community Land Trusts in Kenya and the Community Mortgage Program in the Philippines, only one example has been found (see Mohit 2002). It has therefore proved difficult to reassure officials and communities that the approach is sound. It seems that most stakeholders, whether officials or residents, are reluctant to put themselves in a position of risk, which may partly explain the popularity of titling as a tenure option. To help overcome this fear of the unknown and the potential costs to all involved if new approaches fail to deliver what is expected of them, it would be helpful if UN-Habitat could strengthen its role in disseminating information on innovative tenure options as part of the Global Campaign on Secure Tenure. It is understood that this is being actively considered. If all goes well, the introduction of communal land leases or trusts will go some way to providing security of tenure for the low-income households living in unauthorised settlements. When combined with participatory approaches to the provision of essential utility networks, it is anticipated that the combination of legal and physical measures will increase local confidence that the authorities are genuinely committed to their welfare. This will help stimulate local investment in home and local improvements, especially if the payments made for the lease are seen to generate additional provision of social facilities and amenities. The first step in this process is therefore to encourage the government to introduce MORE in order to stabilise the present situation and buy time for developing longer term 14

15 options. These include the need to establish criteria for selecting those settlements which are to be upgraded and those which will need to be relocated. Draft criteria to be submitted for approval are attached in Appendix 2. It will be important to establish sound criteria for selecting the settlements to be improved, and the sequence by which such improvements are implemented, to avoid charges that the selection is based on political considerations. Proposals for regulatory reform Since not all settlements are likely to be acceptable for long term upgrading, it will be equally important to inform affected communities as soon as possible. This will enable them to appeal against such decisions to the relevant authorities. This is where the regulatory audit becomes relevant as a basis for revising urban planning regulations and standards. For example, the current requirement that a reservation of 25 metres be provided on both sides from the middle of a railway line in urban areas effectively sterilises one hectare of land for every 150 metres of track which could otherwise be perfectly acceptable for urban housing. Given that there are several hundred residential plots already developed along the main railway line serving Phnom Penh (see Fig 1), relaxing this arbitrary requirement could at a stroke make these settlements eligible for upgrading rather than relocation. 15

16 Fig 1: Housing built along the railway line into Phnom Penh. 16

17 Fig 2: Relocated settlement outside the urban area of Phnom Penh 5 In cases where appeals against relocation are rejected, it is important to provide both the affected communities and the authorities with sufficient time to make alternative arrangements. In the past, relocation has involved moving families to sites well outside the urban area (see Fig 2), where many were unable to settle due to the lack of incomegenerating opportunities and even basic services. A vital component of future relocation projects is therefore that sites be found which are within reasonable distance from existing employment areas and public utilities. Identifying such unused land within the urban area is therefore a priority task for the Municipality. 5 All photos by Dr Beng Socheat Khemro 17

18 Given the limited availability and increasing cost of urban land in Phnom Penh, it is clearly important to put all undeveloped sites to efficient use and work is currently underway by the Bureau of Urban Affairs in Phnom Penh Municipality to identify and survey undeveloped sites suitable for infill development. A UN-Habitat study is also being undertaken 6 to assess the potential application of land sharing approaches to the development of sites in private or state private ownership 7. It is intended to develop some pilot land sharing projects in the coming months and to combine these with innovations in tenure and regulatory frameworks. Lessons learned as a result will then provide a basis for changes at the urban scale and in other cities. The regulatory audit can be of use in this respect by indicating which aspects of the regulatory regime are presently impeding access to affordable legal shelter without the need for external subsidies. Appendix 1 indicates the relevant planning standards, regulations and administrative procedures which are presently constraining such access and which therefore need to be relaxed or removed. Understandably, many officials trained to assume that the existing standards and regulations exist to protect the public interest may resist any change, but since the public interest can best be satisfied by maximising access to legal shelter, it is proposed to introduce selected changes within a limited number of pilot relocation projects. These can then be monitored over a five year 6 These studies are being undertaken by Paul Rabé, a consultant appointed by UN- Habitat to advise the Municipality of Phnom Penh. 7 In Cambodia, land in public ownership is defined as either state public or state private land. State public lands are government lands put to special government use, for example for schools or parks, or lands that in their natural condition have some specific public benefit, for example forests. State private lands are all state lands that are not state public lands. 18

19 period and, if considered acceptable by all key stakeholders, including the residents, can be gradually incorporated into the city-wide regulatory framework. The key point of regulatory reform should be to reduce entry costs to new urban housing in ways which provide sufficient security and options for long term incremental improvements. This involves permitting the most efficient use of available land and relaxing constraints on the forms of development and uses to which people can put their plots. Based on previous research at neighbourhood level (Davidson and Payne 2000) project planners should aim to achieve 65 percent of developable land within private use, ie as residential, commercial or industrial plots. Another 15 percent can then be allocated for public or communal facilities, such as schools, health clinics and religious centres, etc, leaving 20 percent for local roads and public open spaces. Since private land areas will be self-financing, and public amenities should also be paid for by the agencies providing them, only the 20 percent of public land will need to be funded by private land users. This yields a ratio of about 30 percent of unproductive to productive private land. In cases where the total area of private land is 55 percent and roads and public open space amounts to 30 percent, a situation common in many planned urban developments, the ratio of unproductive to productive land increases to about 54 percent, a substantial financial burden on residents, developers or the public sector. How this objective is achieved will vary according to cultural, climatic and other local considerations and the skill of developers or project planners. If sites selected for new infill development can be planned according to these objectives, it is hoped that the 19

20 majority of all households which may need to be relocated can be re-housed in areas near places of employment and with good access to services. Conclusions All of these proposals are the result of efforts to address the need to improve living conditions for all residents in existing unauthorised settlements and to reduce the need for future slum formation. They have little to do with global concerns as defined and represented by Target 11 of the MDGs. In fact, for professionals operating at the national or local level, which is the vast majority, an excessive focus on this target is in danger of distorting our collective assessment of the true scale and nature of the urban challenge facing us and of lowering our horizons in terms of what needs to be done. We need to get back on track. 20

21 Bibilography Davidson, F and Payne G (eds) (second edition 2000) Urban Projects Manual Liverpool University Press Liverpool B.S.H. Khemro (forthcoming 2004) Improving tenure security for the urban poor in Phnom Penh, Cambodia: An analytical case study in Habitat International McAuslan, P (1989:30) Land law, tenure and registration issues and options paper presented at the Urban Land Development Seminar, World Bank, Washington DC Mohit, Radhika Savant (2002) A level playing field: Security of tenure and the urban poor in Bangkok, Thailand in Payne, G (editor) Land, Rights and Innovation: Improving tenure security for the urban poor ITDG Publishing, London Payne, G (2001) Lowering the ladder: regulatory frameworks for sustainable development Development in Practice, Vol 11 Nos 2 and 3, May pp de Soto, Hernando (1989) The Other Path: The invisible revolution in the Third World I.B.Tauris London Struyk, R, Hoffman, M. and Katsura. H (1990) The market for housing in Indonesian cities The Urban Institute Press, Washington DC Turner, J F C (1967) 'Barriers and Channels for Housing Development in Modernizing Countries', Journal of the American Institute of Planners, Vol xxxiii, No 3; Turner, J F C (co-edited, with Robert Fichter), (1972) Freedom to Build, dweller control of the housing process Macmillan, New York, and translated into Italian and Spanish 21

22 Appendix 1: Regulatory Audit Table, Phnom Penh, Responsible authority Formal/Statutory Informal/Customary Planning Standards 1. Plot: Minimum area n/a No standard Around 3m x 3m 2. Plot: Minimum frontage n/a No standard 3m 3. Plot: Minimum depth n/a No standard 3m 4. Road width (RoW): Access MDLMUCC/BAU 3.50 m < 3.50m (some areas) 5. Road width (RoW): Local MDLMUCC/BAU 3.50m < 3.50m (some areas) 6. Type of road surface MDPT Bitumen/laterite laterite 7. Utilities: Water supply Government/private Pipe/well Pipe/well 8. Utilities: Garbage collection and sanitation Private (Cintri) Formally Informally Municipal dep of public Existing and informal 9. Utilities: Drainage work and transportation Existing and new provided connection Government (EDC)/ 10. Utilities: Electricity supply private Formal Informal 11. Utilities: Telecoms Government/Private Formal Formal 12. Public open space per ha: Total BAU No standard No standard 13. POS: Primary Schools n/a No standard No standard 14. Secondary Schools n/a No standard No standard 15. POS: Religious centres n/a No standard No standard 16. POS: Health clinics n/a No standard No standard 17. POS: Public admin n/a No standard No standard 18. POS: Community centres n/a No standard No standard 19. Ground building area Over use of ground area (shop house) MDLMUCC/MPP 75% of total area 20. Ground building area (villa house) MDLMUCC/MPP 50% of total area of building Over use of ground area of building Degree of Constraint Comments Mostly not respected Mostly not respected 22

23 Planning Regulations 1. Floor Area Ratio (max) MLMUPC MPP > 3,000m2 < 3,000m2 > 3,000m2 < 3,000m2 2. Building setbacks: Front MDLMUCC/MPP 4m Sometimes not followed No set back or 2m set- 3. Building setbacks: Side MDLMUCC/MPP back Mostly < 2m Mostly not respect 4. Building setbacks: Rear MDLMUCC/MPP 2m Mostly < 2m Mostly not respect 5. Height (max permitted) MDLMUCC/MPP Variable Varied 6. Land use controls/zoning BAU Not clear Not clear 7. Density Levels: Minimum n/a No regulation No regulation 8. Density Levels: Maximum n/a No regulation No regulation 9. Boundary definition (by type) n/a n/a n/a 10. Public health requirements (ventilation, pollution) MDLMUCC/MPP Window with area at least 1.5m2 Varied Road less than 3.5m or sometimes no road for fire rescue 11. Fire regulations MDLMUCC/MPP Road: 3.5 m at least 12. Environmental regulations MDLMUCC/MPP Noise pollution Not respected 13. Septic tank for building floor area less than 80 m2 MDLMUCC 3 m3 (height at least 1.5m) Smaller or not provided 14. Reserved land along both sides of railway in urban area 25m from the middle of railway Around 2m 15. Construction along dikes to protect city from flood MPP Not allowed Already built Administrative procedures by relative costs 1. Housing construction permission: Legal payment tax for land Cadastral office Around R900-R1,800/m2 Did not pay Inform Sangkat office Sangkat office Informal payment Informal payment Inform Khan office Khan office Informal payment Informal payment Did not apply for Ask permission DLMUCC Informal payment permission 23

24 Land measuring and reporting Phumbal Informal payment Construction expert MDLMUCC Informal payment Checking construction site Phumabal Informal payment Checking drawing MDLMUCC Informal payment All offices in MPP and Approval Deputy Governor Informal payment Did not apply for permission Did not apply for permission Did not apply for permission Did not apply for permission Did not apply for permission 2. Obtaining permission for commencement of construction MPP 45 days Build without permission 3. Obtaining connections to services Not exist Not exist Not exist 5. Building which is not require for construction permission (minimum building floor area) Sangkat/ Khan 50 m2 More than 50 m2 Sources: 1. Sub-Decree Number 85, December Government Prakas (Circulation) Number 06 on alleviation of anarchic land grabbing 3. Other interviews with various peoples and institutions 24

25 Appendix 2: Draft criteria for determining settlements to be upgraded and those to be relocated. Acceptable for For relocation Criteria upgrading (land sharing and in-situ upgrading) Land ownership without legal right residing on: 1. Public park 2. Public open space 3. Small piece of public land 4. Government Ministries lands 5. State private land 6. Private land (individual/company) 7. Temple ground 8. Prime commercial land Government land use planning priority 1. The area is planned for public park 2. The area is planned for road infrastructure 3. The area is planned for sewerage system 4. The area is planned for commercial development 5. The area is planned for industrial development 6. The area is planned for residential development Geographical/physical vulnerability of the site 1. The area is located on the sewerage system 2. The area is located too close to or on the dumping site. 3. The area locates on public pavement 4. The area locates too close to water supply sources (rivers, lakes) 5. The area locates on the bank of floodprotected dike. 6. The area is located along river 7. The area locates around lakes 8. The area is located along railway lines 9. The area is located on roof-top 10. Rooms in old buildings