Cover Image
close this bookNational Experiences with Shelter Delivery for the Poorest Groups (HABITAT, 1994, 140 p.)
close this folderIII. SUMMARY AND RECOMMENDATIONS
View the documentA. Leaving the poor out
View the documentB. An integrated urban shelter strategy for the poor
View the documentC. Summary of recommendations
View the documentD. Directions for future research

A. Leaving the poor out

Public-sector efforts at shelter provision for the urban poor have had little impact in developing countries. Programmes have often been limited, expensive and have required large subsidies, thus making long-term replicability difficult if at all possible. They have, furthermore, been administratively intensive and slow in implementation (UNCHS, 1991c). The needs, as perceived by the poor, have seldom been met. Moreover, squatter improvement projects have neither achieved affordability for the poorest nor cost recovery. Replicability at the national level has thus been impossible. Still, the end result of many shelter and upgrading projects and programmes for the poor is undoubtedly improved accommodation and environment. Yet, the fundamental position of the poor in the city and in society at large has not changed. The poor still live in slum-like ghettoes and are underprivileged. Real and lasting improvements in basic needs in a broader sense can only be achieved through access to education, skills and employment with adequate pay. If this is not recognized, the future of the large cities in the developing world and the societies they are part of is uncertain indeed. Increasing democratization may lay the foundation required for successful enabling strategies. This is probably a necessary condition, i.e., that enabling cannot be reached in a useful way without a commitment to democracy and pluralism in society (see box 44). Another such condition is availability of resources and economic growth as well as willingness on the part of the powerful groups of some kind of redistribution. Democratization and economic growth are needed in combination (UNCHS, 1991d).

Box 44. The urban poor's voting power

In Colombo, Sri Lanka, Hettige reports from a study of a slum community that the urban poor has become a source of political power at the local municipal level. In return for political support the poor get some degree of security against eviction and gradually some kinds of basic amenities, water being the most important. At a more general level Hettige states that owing to the large number of urban poor in a democratic political system, the voting power of the poor has given positive results regarding their living conditions:

“Provision of low cost housing, enactment of rent control legislation, tenurial reforms, granting of land rights and provision of basic services are some of the steps that have been taken by local and national authorities in recent years to provide the slum and squatter population with a cover against the market forces that pervade the urban landscape”.

Source: Hettige, 1990.

1. National housing policies

Experiences during the 1970s and 1980s have convinced international donors and other development agencies that direct action by governments to provide shelter is not the answer to the enormous shelter need. In many countries governments have built houses for the poor for decades, constantly experimenting with building materials, types and technologies to reduce costs. Although the effort made has been considerable in many countries, the total outcome is very limited in view of the increasing urban population with inadequate shelter. In the 1980s, it became abundantly evident that the houses built by the public sector were unaffordable to the poor. Furthermore, the housing shortage resulted in political favouritism in allocation and inefficiency in reaching target groups. The “provider-based” solutions of the past were overtaken by events, such as rapid urban growth, rising real building costs, fiscal austerity leading to reductions in subsidies and declining real wages. It is now widely recognized that shelter provision to the poor is beyond the capacity of local and national authorities. At the same time, evidence accumulated on the ability of poor people to shape their own environment, achieve ownership at low prices and build shelter for themselves. “Aided self-help” then became more widely accepted. This includes an important, although reduced, role for governments in shelter provision for the poor. Critics of “aided self-help” maintain that this is only promoted to relieve governments and the rich of their responsibilities to provide a better life to the poor (Burgess, 1985). Aided self-help projects were small, their replicability was low and cost recovery was difficult. A reappraisal led to the notion of enabling strategies which sought to cover a much larger proportion of the poor (eventually all), to integrate shelter strategies in macro-economic planning, to abolish laws and regulations hindering self-help and community shelter construction and to involve the private business sector in shelter provision for the poorest groups. The enabling strategy, as a “support-based” strategy, seeks to improve the functioning of markets in land, capital, building materials, skills and labour inside an appropriate regulatory framework. Governments must, in this perspective, take coherent action ensuring that land, financial and housing markets do not fail to respond to the needs and demands of the poor.

In countries where the urban majority has inadequate accommodation and public funds are scarce, resources have to be distributed broadly, and people must largely be relied upon to house themselves. An appropriate national shelter strategy must take account of differences in the balance of government and private participation, the strength and characteristics of the informal sector and the operation of input markets, to ensure that innovations are not incongruent with the local context and existing conditions. Despite variations, there are principles, approaches and new perspectives which the GSS regards as applicable to most countries. Local authorities are increasingly seen as an obstacle by people who, through the informal sector, have put up illegal structures in and at the fringes of the large cities in the developing world. It is now widely realized that the main task of governments and local authorities is to enable the poor to construct their own homes themselves, in a more efficient manner. This is a major change from the public shelter-providing role, but it does not imply less responsibility and care on the part of governments. It is not a recipe for laissez-faire. Strong and cohesive government action is required to ensure responsive supply markets. Moreover, legal and regulatory reform of shelter construction and housing finance is essential.

In the three countries reviewed in this report the housing policies adopted by the Government during the last five years are all in line with the GSS. The most important change in policy has been a departure from the view of public provision of housing through direct construction of dwellings and site-and-service projects to private business and household involvement in a deregulated shelter sector. The new emphasis is on upgrading of existing slums if possible, popular participation through CBOs and targeted subsidies to the poorest only. The role of government authorities at various levels should be to facilitate and enable individual households and local communities to improve their shelter and settlements by their own efforts, based on local tradition and available resources. Technical assistance, training and financial inputs from the public sector and from NGOs are regarded as necessary external support. This reorientation in Mexico, Indonesia and India is partly a result of the work of UNCHS (Habitat), and partly of a realization of the inability of the public sector to meet the enormous and increasing demand for decent shelter in urban areas.

Instead of people participating in governments' projects, governments need to participate in people's projects (Slingsby, 1989). A distinction should be made between self-help as an instrument of government policy to reduce costs, and genuine community involvement for the needs of the poor themselves.

“In most Third World countries even the political and economic arguments for low-cost housing investment cut little ice with urban managers until the advent of aided self-help schemes which appealed because of their low cost, low commitment to social reform and high aid content.” (Drakakis-Smith, 1987).

Evidence from the three countries shows the merit of the new housing policy. There are government programmes and involvements of slum communities that have succeeded in improving the living conditions of poor groups. The magnitude of these positive experiences relative to the need is, however, still limited in all three countries. All aspects of the GSS have yet to be fully implemented in practice. Moreover, the complexities involved in shelter provision for the poor are so severe that a solution is a very long-term hope only.

2. Availability of land and housing

A major problem is the lack of urban land. This increases the price of the land occupied by squatters, forcing financially weak governments (due to international recession and structural adjustment policies, among other things) to try to resettle the poor at the urban fringes. Without due regard for and incorporation of, income-earning opportunities in such areas for the resettled people, a drift back to more central parts of the cities and their traditional sources of livelihood has occurred all too often (see box 45). Another problem is the reluctance of the large-scale private sector to invest in housing for the poor, due to low profitability and widespread defaults on payments. Extensive formal house production for the non-poor is, however, indirectly beneficial for the poor because it eases some of the demand on lower-quality dwellings. The financial difficulties of governments and the lack of improved real incomes among the population at large make for a slow growth in the formal housing supply. This has led, in the three countries, to a downward demand pressure on lower-priced housing. The effect of this has been the squeezing out of lower-income households and part of the poor. Households with middle incomes may, due to a general housing shortage in most big cities, also end up living in slum housing conditions (Pugh, 1992).

Box 45. Factors influencing shelter choices of the poor

· Access to income earning opportunities, community networks of security, health and education facilities. The locations are also influenced by the cost of transportation in relation to variations in level of rent or site costs.

· Amount of space in the dwelling and on the plot. This factor is of particular importance for households with an intension of carrying out home-based enterprises, home-based work for outside middlemen or letting a room.

· Security of tenure. A high degree of security of tenure ensures higher investments in the plot, and is therefore important for the households with intentions and means of investing in shelter.

· Level of provision of basic infrastructure such as water, electricity, roads, paths, drainage and waste disposal.

· Options available to construct an affordable shelter and obtain some degree of privacy.

In India, there has been a marked change in the policy on vacant land. In the case of private land, squatting has become virtually impossible. Public land, on the other hand, is rapidly being put to use. This has meant that squatting as a housing option will be on the decline, except at distant urban fringes where development has not taken place, and land values remain low. The poor are thus pushed out on to the pavements, and the lower-income group seems to be doomed to rely on unauthorized and unregularized areas at the urban fringes.

In general, insufficient information exists on how informal housing markets catering to the poor and the lower-income group operate (UNCHS, 1990a). This is a draw-back for an efficient enabling strategy. In for instance Seoul, Republic of Korea, even during a period of exceptionally rapid and sustained economic growth and expansion, the shelter conditions for the poor deteriorated (Hardoy and Satterthwaite, 1990). The need for better living conditions among the urban poor in the developing world is on a scale difficult to comprehend. The current level of funding from national and international public and private aid agencies is extremely small in relation to the need. Even if levels were multiplied many times, the impact would be very limited.

3. Affordability

The most fundamental problem of shelter for the poor, is the existence of a large group of people in the big cities who cannot afford to help themselves to improved shelter, even with outside support. It is not only the pavement dwellers but also many other poor people who squat on any available vacant land, and who place shelter low (or not at all) on their list of priorities.

The destitute are too poor, sick, disabled or old to look after their own welfare. A large part of the working poor, especially the food poor, are also unable to make long-term plans. The money they obtain is used for day-to-day survival. These people are not in a position to take advantage of upgrading projects. It is difficult to envisage a shelter strategy directed at these people short of total government provision. Most of the people in this category were traditionally taken care of by rural communities. In urban areas, many likewise obtain help and accommodation in slum settlements. A strategy aiming at the poor should allow for space and shelter for the above category to be organized and constructed by the small communities and their associations. The completely destitute may in this way be assisted indirectly by the authorities and outside NGOs through the help they provide to the working poor. Today, those looking for accommodation in slums in many cities in the third world are not destitutes migrating from impoverished rural areas. They are, rather, young people who are getting married, or others who, for a number of reasons, can no longer stay with their relatives or friends in the city. The lack of alternative accommodation forces many with low-paid work to look for shelter in the slums or slum-like areas.

In countries such as Botswana, Zambia and Zimbabwe, the World Bank has encouraged standards of housing and services that are low enough to be affordable by many of the poor households. Water is, for instance, provided in stand-pipes, sanitation is based on pit-latrines, and plot-size is limited. Several other suggestions, such as allowing the use of traditional building materials, have not yet been accepted by the governments (Larsson, 1991). The shelter strategies implemented have thus benefited the households just below the poverty line, and not the poorest among the poor.

The experience in the three countries is that subsidies have been reduced and targeted to the poor. In India, for instance, the NHB and the Housing Finance Development Corporation promote and regulate housing-finance institutions nationally. They have managed to mobilize resources for housing on a large scale. As is the case in the other two countries, the poor do not have access to these formal sources of loans. The interest rates required for cost recovery of housing projects effectively exclude even the working poor. Collateral restrictions should be eased, for instance by allowing the poor to pool their assets and apply for loans as cooperatives or credit associations. It is an interesting idea to allow collective community collateral for housing loans (see boxes 46, 47 and 48).

The poor seldom save any money, and if they do, they do not save in a bank giving them the right to a loan. The community-organized arisan system in Indonesia, which is for specific shelter improvement purposes, seems to be successful in several places. The growing amount of female-headed households necessitates a particular concern for the inability of women to secure loans. Specific housing loan schemes should be established for them (UNCHS, 1986a). Yet, saving among the poor is possible. This is evident from the work of SPARC (Society for Promotion of Area Research Centres) in Bombay. SPARC encouraged pavement dwellers to save regularly for the future purchase of land and building materials. Women were furthermore assisted in the design of dwellings adapted to their needs. In Mexico, CENVI (Centro de la Vivienda y Estudios Urbana) helps residents of inner-city rental tenements to form cooperatives and then gain access to land and credit for self-construction. The essential point here is externally induced organization and the perception created among the poor that saving will eventually result in a major good.

In some instances, cross-subsidies may be applicable and of benefit to the poor. In Bombay, for instance, the authorities charge an above-the-market price for land for commercial use, thus allowing a charge below the market price for other land. From Kenya, Macoloo (1988), however, reports that the upgrading of squatter areas tends to subsidize the well-off, instead of assisting the poor for whom the projects are intended. Furthermore, the segregation of social groups in different residential areas tends to perpetuate the differences in the provision of basic infrastructure and services. More mixing of groups and charging differential fees for land and services according to ability to pay may be a method of improving the quality of life of poor urban residents.

Box 46. Group credit for the poor

Commercial banks are normally unwilling to lend to the poor. Furthermore, credit channelled through State institutions seldom reach the poor. Those that are willing to lend to the poor, however - traders and local moneylenders - charge exploitative rates of interest. The special State credit programmes directed at the poor, having low interest rates, have low recovery rates and high administrative costs. Part of the funds have also in many cases been diverted to non-poor groups. In Bangladesh, the Grameen Bank has been successful in overcoming the above-mentioned problems.

The Grameen Bank was started in 1976 by Muhammad Yunus. He developed, by experimentation, a system of lending small sums of money for small projects to poor (primarily) women. He found that loans to individuals and to “large groups of people” (around 10) were unsuccessful. The successful model was to form groups of only five members and to organize several groups from the same community into local centres. By 1987, the Bank had 247,000 members in 10,800 centres (Hulme and Turner, 1990). Essential aspects of the Bank's operations are: the members of a group should come from the same socio-economic community; only the poor are eligible as members (in rural areas those owning more than 0.4 acres of land are excluded); a member of a group can only obtain a loan if the other four members have kept their loan obligations to the Bank; the leadership of a group must rotate, and a member cannot be a leader for a second time before all have been once; the centres consist of six groups; a centre is actively meeting, discussing and socializing regularly; centres are strictly segregated according to gender. The Bank has fieldworkers who visit the centres regularly. Bank workers are rotated every year to work in different places and branches of the Bank. Interestingly, the Bank has achieved a recovery rate of 98 per cent, and it has received substantial amounts of money from international donors. The Bank's principles are now being replicated in other countries.

Source: Fuglesang and Chandler, 1986; Hossain, 1988.

A large part of the poor cannot pay a fixed and regular instalment for shelter over an extended period of time. It is thus necessary that housing agencies seek collaboration with agencies concerned with small-scale enterprise development. Integrated shelter and economic development programmes for the poor should be formulated. And shelter production should be viewed as an important and integral part of economic activities (UNCHS, 1986b; 1993a; forthcoming).

4. Flat housing

Experiences from the three countries show that flat housing for the poor is not a successful option. The reasons are both that it is too expensive to build high-rise dwellings even on slum land for the poor to be able to afford the rent, and that this form of living is unpopular. UNCHS (Habitat) also notes that the building of high-rise flats are capital-intensive, and that in most cases they require a considerable import component. They are also rather difficult to maintain with local resources (UNCHS, forthcoming). Furthermore, the poor often use their dwelling for small-scale enterprises. Women can, in this way, combine caring for children and earning a little from trade or production (UNCHS, 1989a: forthcoming). This is difficult in a flat. Experience is similar in Karachi, Pakistan. Flats on slum land became far too expensive for the original inhabitants, and this type of accommodation was not a suitable alternative for their demolished houses (Akerboom, 1992) (see box 49).

Box 47. Working Women's Forum (WWF) in Madras

The WWF is an NGO that works in 400 of the 1400 slums in Madras City. Azad (1988) undertook a case study of women in one of the poorest of the squatter settlements, Periyar Nagar, where female incomes are at the bottom of the income scale. As access to credit often is a key to development, a major impediment to petty trading for women is lack of cash credit. The sources the traders rely on are typically local moneylenders, wholesalers, shop-owners, kinship networks and chits (rotating credit associations). Most of them charge high interest rates. Annual interest rates of 40-120 per cent are not uncommon. Cash credit is needed by the poor for many other purposes than trade and production. As Azad puts it: “... the low income household in Madras City is a veritable storehouse of crisis: illness, death, ritual, marriage, male-desertion, sudden unemployment for males, monsoon, floods, and so on.” The moneylender is an easily available source of cash credit on these occasions. The WWF's credit loan programme, which is geared towards employment and income generation is based on the priorities of women and the realities of their living conditions in the slums in Madras.

In a Home Improvement Grant scheme in Periyar Nagar it was found that beneficiaries could not be given the allocated money because it was then used for other than shelter purposes. Another finding was that home-based workers lacked a worker consciousness which street vendors and headloaders, among others, acquire through constant struggle in markets and streets. The home-based workers perceived their roles as extensions of their domestic work. To combat this, WWF made it compulsory that women were mobilized into small groups. It is, however, not possible to obtain group loans through the public lending institutions. Moreover, poor women have all kinds of difficulties in getting loans from the formal system; they are typically not taken seriously when applying for credit. At WWF's office this is not the case. The women can walk in at any time, without fear or shyness, badly or well dressed.

In the survey of credit for shelter in Periyar Nagar, Azad found that 80 per cent of those who had received loans from five to nine times, had managed to improve their shelter substantially by strengthening walls, the roof and adding a toilet or electricity. Of those with three or four loans, all had made solid improvements. Among second-time loanees, 50 per cent had successfully applied the shelter loans, whereas the figure for the one-time loanees was only 5 per cent. The study clearly showed that, among first-time loanees, the main spending priority was not shelter, but rather the paying back of old debt, food, or investments in gold or income-generating activities. For all those involved in the shelter credit scheme improvements of the drainage in the community was seen as a major benefit.

Azad concludes from the study of the various activities of WWF activities that the organizing of work and credit in small groups is essential for success. A long-term impact of extreme importance is the small but gradual change in women's perception on issues such as dowry, inter-caste marriage and education of girls.

Source: Azad, 1988.

Box 48. Thrift societies in Sri Lanka

Thrift and credit cooperative societies are the oldest type of cooperatives in Sri Lanka, with a history dating back to the beginning of this century. A thrift society is an autonomous CBO seeking to promote thrift habits, self-help and voluntary mutual aid in order to improve the living conditions of its members. The societies obtain their funds from membership shares and deposits, which have increased during the last years. External funding is also received, although most societies are somewhat sceptical about too large external assistance and thus dependence. In 1990, there were 6871 primary thrift societies at the local level in Sri Lanka. The majority of these societies are unlimited liability societies formed during recent years, and catering to the needs of the poor in slums and other substandard settlements.

Vidanapathirana (1991) found that women constituted only 35 per cent of the membership of urban societies. Women are generally cautious regarding joining in new institutions, and most of them do not want to take any chances with their meagre savings. A further conclusion is that the societies recently established in urban areas have so far not been able to improve the lives of their members. Of 21 urban societies studied, only three were operating soundly. Still, Vidanapathirana found that urban thrift societies in many places had acted successfully as a harbinger in the process of mobilization for self-help and cooperation. Moreover, the relatively unsophisticated system adopted by the urban thrift societies minimizes the cost of lending to urban small borrowers. If properly organized, these societies can be a viable alternative to the banking problem of small borrowers who cannot satisfy the conditions imposed by the formal system. A lesson learnt is that it is necessary to organize cohesive groups that are manageable in terms of numbers, attributes and interests. The absence of cohesion makes good communication and consensus very difficult to achieve. In sum, assistance is required to uplift the organizational and management capabilities of the urban thrift societies.

In the words of Vidanapathirana:

“In a free economy where market forces determine the priorities of both the State and the private sector, it is essential to admit that these deprived communities would be further isolated in society, if organisations such as thrift and credit co-operative societies are not adequately mobilised and equipped to address their demands.”

Source: Vidanapathirana, 1991.

Due to limited available space and rapidly rising urban land prices, governments may yet find it necessary to build some multi-storeyed flat housing for the poor. Pavement dwellers and the food poor living in the inner-cities cannot solve their shelter need in either the private or household housing markets. Governments should thus restrict their housing production to particular target groups that cannot be provided for by any other sector. It should, however, be realized from the outset that experience shows that cost recovery then, at best, will be very difficult.

5. Renting

Renting is becoming more and more common in urban slums. This is due to lack of space and rapid urban migration (see box 50). Legal as well as illegal slums are becoming extremely “overcrowded”. By the early 1970s, Desai and Pillai (1990) found that in a slum area of Bombay as much as 70 per cent of the resident households were tenants. In fact, people seldom come to an area and build a shelter for themselves. Most vacant space is already taken or controlled by groups of intermediaries offering accommodation and/or space for sleeping in the streets. Often, high “deposits” of money must be paid to these people at the outset to be allowed to rent a space. In most slums today it is necessary to have or borrow some money to obtain accommodation.

Box 49. Flat housing

The observations made by Seabrook in Bombay can be made in most of the largest cities in the developing world.

“The workers live, for the most part, in the chawls - tenements. Many of these are of brick or stone. Three or four storeys high, they are constructed round a central courtyard... Some of the windows are covered with wire netting, others protected by bars, prison-like; but never glass... The yards are mostly beaten earth; but here and there are flowers or a plantain... Elsewhere, pools of stagnant water collect... rubbish accumulates where the animals scavenge... Although the chawls have communal wash-houses and lavatories, the smell of the wind from the sewage and the polluted sea makes the atmosphere sulphurous... In this apartment twelve people sleep: father and four sons in the bed, two grown-up sons on the floor; mother and three daughters sleep in a cot in the corridor, with the two small children at the foot of the bed. The lavatories that serve the whole block are at the end of the corridor: dingy tiling, a pervasive smell of urine, a constant nuisance of flies and mosquitos. The rooms are airless, even though the day is relatively cool.

... However ugly urban poverty, the proximity to food supplies means that people do not die of hunger in the way that can occur in remote country places... there is always something edible to be scavenged from the heaps of vegetable matter around the street markets.”

Source: Seabrook, 1985.

Those without are left with the streets. An interesting idea, tried by the authorities in Indonesia, is to support an extension of private dwellings in the slums with a room for letting. This may be of real benefit to the poor who cannot afford to own shelter but have a minimum of relatively permanent income. Controlling rent increases is, however, necessary if the poor are to be able to afford such accommodation. It is, on the other hand, essential to allow rents to rise periodically in order for the private sector to look upon flat housing as a profitable investment (UNCHS, 1990c). To find the right balance between these two objectives is not easy.

Box 50. Urban-rural links and urban housing

A number of urban residents in developing countries retain strong links with their rural home area. In many cases the wife and children remain to look after a small piece of land and some animals, while the husband seeks casual wage work in town. The linkages are of many types and of varying strengths, and the complex patterns of urban-rural interaction among the poor and low-income groups differ according to local context. These links are, generally speaking, much more common in Africa than in Latin America and least so in Asia.

In a study from Kenya it was concluded

“... that urban-rural linkages are reflected in a low commitment to investment of resources in the urban setting. This does not in itself result in the development of commodity relations in the supply of housing, but it provides an ideal market for rented housing. Once having consolidated their position as promoters of housing for the rurally-tied residents who find rental accommodation appropriate, the promoters may extend the supply also to the landless. Promoters were, until the late 1980s, small and middle businessmen, politicians, traders and coffee farmers. Larger scale developers appeared in the late 1980s. They do this, for instance, by various mechanisms excluding poor people from unauthorized 'subsistence-production' of their own houses (for example, squatting) through political control of land, of planning, and through the enforcement of building by-laws. In Nairobi more than in Thika, this has resulted in extensive demolitions as the ultimate solution to force low-income families into privately rented housing.

Therefore, although strong urban-rural linkages discourage residents from investing resources in urban housing and hence contribute to the development of a rental housing market, this market is, first and foremost, shaped by the interests of the middle class and the strong political intervention in planning.”

Source: Andreasen, 1990.

In the present situation of freer markets and less public house building, it is essential to allow rents on flats to increase according to inflation in order to stimulate the private sector to construct tenements. At the same time, the wages for the workers are seldom raised in step with rising prices. The ability of the poor and the lower-income group to pay higher rents is thus limited. Hence, authorities must try to find a level of rents affordable to the least poor of the poor. This may induce the private sector to increase their involvement in rental housing and present owners to maintain the existing rental housing stock. This level of rents may however not exist. If the gap between what the private sector can accept and people can afford is relatively small, subsidies to families may be the only answer. Since more and more poor people cannot afford their own housing and must rent accommodation, the issues of flexible rent regulation, subsidies and the use of the market mechanism to increase rental housing are of growing importance for the poor (see box 51).

Most countries have some form of price control on part or all of their rental housing. Rent control is among the most visible and contentious housing-policy issues. Unlike many housing programmes, rent control has a small budget cost. This makes it attractive to policy-makers. Research has, however, shown that rent control distort markets and reduce private-sector rental housing construction (Malpezzi, 1990). Rent control is in fact, an implicit tax on housing capital from landlords to tenants. Some tenants are, however, worse off under rent control, since it constrains housing consumption. In rental-housing markets with significant uncontrolled parts, rent control can drive up the price. On the other hand - according to Tenga (1990) - rent control makes sense in situations where the operation of a free market in rental housing displaces the majority of the poor. Tenga thus holds that there are important reasons for having rent control in certain areas, while in other areas it “does not make sense at all”. An example given is low-density areas of Dar-es-Salaam where deregulation may enhance property development.

Experiences show that rent control imposed by governments restricts the private sector from making investments in flat housing, owing to low rates of return on such investments. Moreover, rent-control acts in India, for instance, have resulted in a “black market” in rental housing. The acts are evaded through the charging of market rents, initial deposits, leasing and non-issues of rent receipts (Wadhva, 1990). In large cities, such as Bombay and Delhi, landlords can evade the act by offering guest-house accommodation. The poor may thus end up paying more than if the market was competitive. Due to the fragmented and informal nature of the market, it is not possible for government either to regulate the rents or to protect the tenants against exploitation. It is the poorest households that are most prone to exploitation. This is, however, not a major problem for the poor in India due to the limited supply of and thus relatively high rents of houses in desired locations. There is an obvious need to develop a distinct rental-housing sector also catering to the poor in many urban areas.

Box 51. Rental housing

In the context of an enabling role for governments in national shelter strategies, governments should facilitate rental housing through a wide variety of measures. Such measures might include:

· Reorienting public-sector agencies, including local bodies, to play a facilitating role in rental-housing supply;

· Creating an enabling legal and regulatory environment for housing activity in general and rental housing in particular in a manner that avoids tenure biases;

· Introducing fiscal and property tax concessions to promote rental construction and encourage maintenance of rental stock, especially for poor families;

· Streamlining building codes and planning standards, including making provision for renting of dwelling units in the design of low-income settlements;

· Developing city-level strategies for decentralizing activities and employment, through rationalization of land use in inner-city areas and provision of adequate transport facilities, so as to allow for residential mobility of poor households in a changing tenure environment;

· Installing a widely accessible information network on the availability and costs of ownership and rental housing;

· Facilitating initiatives of tenant groups to organize and manage housing and services, to secure finance on easy terms and to negotiate land-sharing in informal settlements;

· Exploring ways of regularizing informal settlements, wherever possible, with safeguards for avoiding displacement of existing occupants;

· Identifying the scope for targeting subsidies for low-income households;

· Encouraging the corporate and institutional sectors to provide rental housing in existing and new urban areas;

· Streamlining management of public rental housing, with efforts to transfer management of structures and services to tenant groups on agreed terms; and

· Undertaking direct provision of rental housing only to especially needy groups or in emergency situations.

Source: UNCHS. 1990c.

In a study of four unauthorized settlements in Nairobi, Amis (1990) found - contrary to expectation - that the rent levels had a high degree of stability during the 1980s. This was due to government restrictions; i.e., a rare case of successful rent control. The objective of the rent control, the author suggests, was the Government's want of low housing cost so as to reduce the pressure on wages.

The programmes of infrastructure and service upgrading in Calcutta's bustees (where renting is the typical form of housing) have been successful in improving the living conditions of more than 3 million people (about 30 per cent of the city's population live in bustees). The improvement has expectedly led to increased rents and occupancy rates. According to Pugh (1990), “... Calcutta no longer has an air of desperation about its bustees and its basic utilities. What the programmes have not done is to change the substantial dimensions of poverty and inequality.” Renting thus provides important housing for many chronically poor and it acts as a crucial first step for households that can save and later move on to home-ownership.

Nigeria's housing problems, like those of other developing countries, are essentially of an urban character. There is an acute shortage of rental accommodation, especially for the poor in the big cities. The average urban worker has to pay as much as 40 per cent of his/her monthly income in rent. This is a major factor in the distortion of income distribution in favour of property-owning people. This constitutes an obstacle in the realization of a just and egalitarian society. There is no area of social service where the urban worker in Nigeria now needs relief more desperately than in housing (Achunine, 1993).

6. Owner-occupiers: ownership of plot and house

It is difficult to overestimate the positive impact ownership or secure tenure of a plot and even a simple shelter has on a family's life. Among a number of valuable economic and social aspects, home-ownership is a most important hedge against inflation, on which a family can protect what little wealth it has.

Slum regulation and upgrading are successful in many urban areas and in the cities reviewed in this report. The projects and programmes undertaken contain valuable lessons which have been collected, systematized and published by UNCHS (Habitat). In order to encourage the poor invest part of their meagre income in improved shelter, it is important to legalize the occupation of the land and to give them ownership of a plot. From an upgraded area (Anna Nagar) in Madras, the Wit (1992) reports that plot prices rose steadily when improvements were carried out. And there was a peak of plot selling when the beneficiaries received “lease-cum-sale” agreements to the land. For those who could afford to stay on in the area, the quality of life changed in a positive way. Housing investments soared in a number of households, there were fewer quarrels and conflicts and the position of women improved. In order to reduce the resale of plots, due to poor people's more pressing needs, it may be necessary that the community gets the legal title to the land, so that it can control who can use the land. This may avoid gentrification of upgraded settlements. Special attention should be given to the problems women have in getting a legal title to a plot. Unfortunately, experience shows that when an area is allocated to poor people, leaders intervene to secure a profit or benefit for themselves. The number of people seeking to make a quick profit out of illegal land transactions is fairly large. The best way to counter this is to have well organized communities and to put emphasis on training of CBO leaders (see box 52).

Evidence shows that most squatter settlements have some form of organization. Squatter settlements are typically social settings where relatives, friends and people with specific group identities live together. There are close social interaction, mutual aid and economic exchange. To organize and improve a settlement when external sources of funds and expertise are made available, is much easier in such communities than when “strangers” are moved to a resettlement site. It takes many years to create a common community feeling and identity in new settlements when those settling are unknown to each other and belong to different groups. To assist in the formalization of CBOs may, in such cases, be important. When resettlement of squatters is undertaken, whole communities should be allowed to move together. To reduce the drift back to the inner city, it is however essential to give individual households a choice of where to resettle. This drift back is a major problem in most cities. In, for instance, La Paz many chose to remain in centrally located rental accommodation instead of moving into owner-built housing on the city's periphery although ownership is the generally preferred form of shelter (Lindert and Westen, 1991).

Box 52. Conditions for success in community-based projects

· People and the communities must be involved in all stages of a project.

· The respective roles, responsibilities and inputs of the community and public agencies must be clearly defined at the outset.

· Public agencies must act as supporters and facilitators, and not as managers of the projects.

· The staff of the public agencies involved in the administration and coordination of the projects should have skills in organizing and motivating poor people rather than in technical matters.

· Public agencies must fulfil their part of the projects on time.

· The projects must be on a relative small scale and apply appropriate technology.

Settlement upgrading or establishment are more successful when based on community participation from the very beginning, and when a household is allowed to finish its dwelling gradually, and according to its individual financial ability. In India, for instance, there are no specific programmes directed at the needs of poor women regarding shelter. It is, among other things, essential that female-headed households also can hold a legal title to a plot and shelter. The removal of institutional forms of discrimination to give women equal rights to land-ownership and access to credit are strategic gender needs. Moreover, studies show that willingness to pay is important for successful cost recovery (Moser, 1992). Women are generally more willing to pay and are more responsible for their debts. In a study of an Indian credit programme, it was found that the participation of women in planning and implementation increased the repayment of loans.

Are well-housed people able to work harder because they live more comfortably and have a higher degree of privacy than people in slums? UNCHS (1993a; forthcoming) found that literature on this issue provides inconclusive empirical evidence. The argument is still widely adopted (see box 53). It is important to recognize that slums and squatter settlements contain a variety of people. The incomes and shelter quality in a slum are often very different, and many people are enterprising and hard working. However, empirical evidence does not support the idea that slums are gradually developing out of their defining characteristics. The reality is the opposite. Overcrowding is increasing in most places and more and more people have to rent a tiny room in ramshackle structures owned by squatter landlords.

Box 53. The possible impact of improved shelter

It has been said that improved shelter may raise productivity of work, lower absenteeism from work, improve the level of health, increase the quality of education, and decrease the incidence of violence, crime and other social problems.

Burns and Grebler (1977), however, were not able to conclude that improved shelter had these effects. This does not mean that there is no positive impact on these and other aspects of poor people's work and welfare. Such impacts are generally very difficult to isolate and relate to one causal factor since they depend on several other variables as well. Still, it is highly likely that improvements in the worst end of the shelter quality scale will have more positive impact in people's life than at the high quality end. This is an argument in favour of “second best” solutions in housing. Moreover, incremental and relatively minor improvements in services, such as drainage, sanitation, solid-waste removal, power supply and potable water, may have an immediate positive impact.

Source: UNCHS, 1993a; forthcoming.

B. An integrated urban shelter strategy for the poor

Evidence from a variety of urban shelter projects and programmes in the major cities of Mexico, Indonesia and India underlines the extreme difficulty inherent in reaching the goal of environmentally sound settlements and of facilitating a process of improved shelter for the poor. The following are reflections based on experiences from the three case study countries. A large group of the poor cannot achieve improved shelter without new economic activities increasing their income (UNCHS, 1993a; forthcoming). Upgrading of slum areas' infrastructure and basic services and development of new land for low-income housing cannot to a sufficient degree be reached without the participation of the residents themselves in organization, work and financing. It is thus essential to view shelter as a productive investment in a long-term perspective. The income multiplier of low-cost house construction is relatively good. Furthermore, this kind of shelter production uses a greater proportion of locally-made materials, higher labour to capital ratios and smaller amounts of imported machinery than formal house-building. For governments to support and make the right environment for investments in shelter for the poor is economically rational indeed. Since the provision of shelter is an economic activity with a very high labour input, it should be possible to promote both housing and employment goals simultaneously, through a common strategy (UNCHS, forthcoming).

Urban poverty should be reduced through employment generation, particularly for women, in the provision and improvement of urban infrastructure and services and the support of economic activities in the informal sector such as repairs, recycling and small commerce. Small-scale entrepreneurs can - given less stringent rules and regulations and some technical and financial assistance - contribute to economic growth and the creation of employment. Governments have an important role in providing land and the overall framework in which the private small- and large-scale enterprises may operate efficiently. Furthermore, labour-intensive public works tend to employ the poorest workers. Even relatively simple drainage and earth works can generate a reasonable amount of employment for the unskilled. Assistance in the form of small loans to home-based enterprises will also be of value for the poor and for women (UNCHS, 1993a; forthcoming). Rules that inhibit economic activities within the home should be removed. Shelter may thus prove to become a major component in a community-based strategy to mitigate urban poverty. This has up to now not been adopted as a strategy. Shelter has been, and to a large extent still is, regarded rather as a consumption good, a cost to governments. Investments in upgrading and shelter for the poor are supporting informal economic activities. Such investments have a very low import component, thus demanding little, if any, foreign exchange. The widely held view that such investments should be regarded as welfare expenditures absorbing rather than creating resources, has no basis in fact (UNCHS, 1987; forthcoming).

The aim must be for governments to find the right balance between market liberalization and deregulation on the one hand and intervention on the other. There should be a division of labour between CBOs and local authorities. Public works on basic amenities in the slums should be offered to small contractors and temporary employment given to the poor at low pay. This must, however, be limited in scope and duration so as not to increase exploitation of non-unionized and weak workers. In addition to public works and small communities' own efforts in settlement upgrading, various economic activities should be stimulated in a planned way. This will provide local multiplier effects enabling poor people to invest more in their shelter, thus increasing further the economic circulation locally. An integrated urban shelter approach is necessary if the indecent and unhealthy living conditions are to be alleviated. The strategy has to be applied on a very large scale to have any visible impact on the enormous existing need.

Poverty is on the increase in the three countries, as in other developing nations, in absolute numbers. Urban poverty is rising even faster, and the urban population is becoming a larger and larger proportion of the total population. The urban shelter problem is so complex that only an extraordinary and combined effort by government authorities at different levels, the private sector, national and international NGOs and people themselves will have any chance of success. This does not imply that individual programmes and projects should be large-scale (UNCHS, 1991c). Rather, smaller, locally-adapted projects should be developed on the basis of real experiences from various parts of the developing world. The experience is that large-scale relocation programmes are difficult to manage and have negative social effects. Smaller-scale projects are more flexible allowing collaboration between residents, NGOs, the private sector and public agencies towards an adaptation to different local conditions and shelter and settlement goals (UNCHS, 1991f).

“Specific programmes aimed at assisting low-income groups, particularly those residing in slums and squatter settlements, can only be successful if framed within a comprehensive shelter strategy that lists priorities, identifies affordable approaches, makes provision for the fair allocation of resources and eliminates the contradictions between its various sector components and programmes.” (UNCHS, 1991a).

Community participation in shelter and settlement programmes is vital because, among other things, poor people's priorities are often misrepresented by well-intentioned planners and administrators due to the complex social and economic reality of the urban slums (Azad, 1988). Furthermore, the input even very poor people can provide in terms of planning, implementation and monitoring should not be ignored. Community participation can go a long way in reducing the following common problems: lack of proper identification of beneficiaries; lack of information and communication between the implementing agency and beneficiaries, leading to petty corruption by intermediaries; lack of knowledge of details of building standards etc., among the residents; lack of acceptance by the local people of, for instance, house demolishing to make space for roads and water supply; and lack of control of contractors resulting in delays of implementation. Community participation should thus be incorporated into the very design of urban management programmes (UNCHS, 1991e).

About 1500 million people live in urban areas in the developing world (the figure was 1360 million in 1990 and will reach 2000 million by the year 2000). The economic forces driving urbanization will continue to operate in the years to come. Governments have so far failed to ensure that the rapid urban growth is accompanied by sufficient investments in physical and social infrastructure, especially in the poorer areas. The result is an escalation of over-crowded, illegal and unhealthy settlements. Individual and collective efforts in the urban slums, although in many cases and respects remarkably successful, cannot solve many of the major and necessary tasks, such as water supply, sewerage and drainage, health services and land regularization. Water contaminated by toxic and other hazardous waste will be an increasing danger to the health of the urban poor in particular. It must also be expected that, for instance, the further growth of deteriorated and overcrowded rented accommodation will worsen an already violent social situation. It is estimated that at least 600 million people live in life- and health-threatening communities and dwellings in urban areas of the developing world (Hardoy and Satterthwaite, 1989; Cairncross 1990). The sheer size of the urban slum and poverty problem in most developing countries makes a solution depend on international assistance of an exceptionally high magnitude.

C. Summary of recommendations

Some of the key issues highlighted in this report of how to address the shelter problem of the poor within the context of the enabling shelter strategy are summarized below.

1. The funds made available to the housing sector, must in future, be at a completely different level than at present. International donor agencies and the United Nations system must make shelter for the poor a priority issue. At the national level, democratization and economic growth are required, in combination, as a basis on which to develop sound shelter strategies applicable to different local contexts. Information must be diffused in such a way that planners and politicians understand and accept that housing and shelter provision in favour of the poor are really beneficial for overall economic growth. Investments in the housing sector generally must be seen as productive investments with a very limited import leakage. It should furthermore be realized that increased allocation of funds to housebuilding for the lower-income and middle-income groups, is not only creating employment, local multiplier effects and backward linkages, it also eases the downward pressure on housing. This will improve the access of the not-so-poor among the poor to adequate housing.

2. Investments in rural development is no solution to urban shelter problems, neither today nor tomorrow. Although urban migration will continue under the most favourable circumstances for rural areas, and thus add to the number of people in urban areas, the natural population growth in the urban slums and in the rest of the large cities is very high in absolute figures and will continue to be so in future. It is thus essential that governments implement fully the policies of the GSS, which many governments have already adopted in principle. Reforms are necessary regarding rules and regulation to make it easier for the poor to make their own shelter with traditional materials and techniques. Although the majority of the poor must house themselves, participatory programmes and projects should gradually be expanded to reach all the working poor without adequate shelter.

3. The model for such programmes should typically include cooperation between national and local authorities, NGOs and CBOs. It is important that public agencies see their role to be one of initiating this cooperation and of combining the positive forces for shelter delivery for the poor. Furthermore, it is essential to involve the people through their CBOs (or assist them in forming CBOs) from the very beginning of a programme. The communities should also be part of the planning process and not only participate in implementing decisions already taken.

4. Experience has shown that affordable shelter for the poor, is close to impossible. Credit programmes through special banks with NGO financial support have proved to be successful in some cases. The lessons learned through the growth of, for instance, the Grameen Bank in Bangladesh must be made widely available in other developing countries.

5. Acceptance for and opportunities of group collateral and collective credit should be expanded. Likewise, local communities should be able to retain control over plots when a household decides to sell its plot. This will contribute to an avoidance of rapid reselling of land in resettlement programmes. Solutions must moreover be found to ease the access of women to plot rights.

6. Cross-subsidization must be more widely applied in shelter programmes. By mixing various groups in housing programme areas and charging differential rates according to dwelling standard, the poor may receive cheaper shelter and better physical and social infrastructure.

7. The supply of adequate and cheap building materials must be improved, through reform of regulations and organizational assistance.

8. Small-scale projects are normally more successful than large-scale ones. The former may not necessarily have a limited impact on the overall need. Restrictions should be removed to allow many small projects to be established rapidly but based on a general model modified by the particular circumstances of individual countries.

9. The important work of training CBO leaders as part of ongoing shelter programmes and projects must continue and be expanded. Training in various Melds should become an integral part of all shelter, upgrading and resettlement programmes everywhere. The importance of training in situ of the beneficiaries and participants of mobilized self-help projects must not be underestimated. Training is an essential element of empowerment.

10. While planning new settlements, it is of utmost significance to select locations and/or provide for easy access through public transport systems that enable the poor to reach places of work and employment opportunities. Experiences abound of the drift back to the inner city by poor people who have been resettled far and relatively isolated at the outskirts of cities without a simultaneous planned and implemented transport system. Since available land for shelter purposes is extremely limited in most large cities, high-rise flat accommodation is often seen as the solution. For the poor this is neither affordable nor favoured. Land at the urban fringe is then the only alternative.

11. In analyses, debates and planning on shelter delivery for the poor, the poor group should be split and categorized according to levels of poverty. The destitutes must primarily be supported through various kinds of social benefits and charity, while the working poor should be the target group for “aided participatory shelter programmes.” Furthermore, the latter category must be split into the “food poor” and the “subsistence poor” (in a narrow basic-needs sense). Pavement dwellers may be found in both these categories. The subsistence poor can be relied upon to be able to save a little, to give priority to shelter improvement and to pay part of the cost of community upgrading (e.g., pay a betterment tax). Moreover, the subsistence poor are more easy to mobilize for self-help shelter projects. The food poor must have a higher level of subsidy, assistance in building techniques and in obtaining permissions to improve their shelter conditions.

D. Directions for future research

The need for research on how to enable the various low-income groups to help themselves to obtain better shelter and living conditions is extensive. Some key issues are outlined below.

1. More empirical documentation is required to convince planners and politicians that investment in housing in general, and in shelter for the poor in particular, constitutes a boost to economic growth. It is necessary to show that in different contexts and under various circumstances the multiplier effects, backward linkages and employment impacts are significant and relatively high compared with investments in other sectors.

2. The functioning of markets for plots and shelter in slums, squatter settlements and on pavements needs to be further clarified. Furthermore, the role of intermediaries and of the mode of operation of local mafia-like groups must be analysed in specific empirical settings.

3. The question of the saving capability of different poor groups is still undecided. In the literature, support is given to the view that there is quite a substantial potential for saving for shelter even among subsistence poor, but as this is refuted by other studies.

4. Analyses must be made on how to arrive at the right balance between the ability of the poor to pay and the need to let house rents increase according to inflation, to encourage building maintenance and investments in new accommodation for rent. It is not only a question of level of rent and rate of increases but of how subsidies may be used to close the gap between what the poor can pay and what business will accept. The mechanisms and practices whereby house-owners and landlords avoid existing legislation regarding rents should be identified.

5. Research must address the question of how to establish the right balance between market liberalization and deregulation and government interventions for different subsectors of the housing sector and for different places.

6. Studies should be carried out on the mechanisms by which adopted government shelter policies fail to be implemented. What are the countervailing forces and how can they in turn be countered?

7. The poor do not constitute a homogeneous group. The variation in levels of poverty and in living conditions must be taken into consideration when planning shelter and settlement projects. For certain groups among the poor relocation may imply less of their sources of livelihood, and thus entail a worsening situation. Not only must shelter type and settlement form vary according to the degree of poverty and the kinds of sources of livelihood different groups of poor have, the type and extent of organization and external support must also be adapted to the specific needs of poor people living in special circumstances. Existing information on these issues is sporadic. Systematic research, with clearly defined and operationalized categories of poor people is highly needed.