EXECUTIVE SUMMARY

1. This report deals with the problems encountered by the poor in their struggle to find shelter. The report specifically addresses the issue of how current shelter policies are leaving the poor out of the shelter-delivery process. Brief descriptions of successful projects of shelter improvement for the poor are included. These cases indicate the direction shelter policies should take.

2. The report is primarily based on national experiences from India, Indonesia and Mexico with examples from some of the major cities in these countries. In addition, experiences from Nigeria are included. On specific aspects in the discussion of the shelter situation of the poor and on policies and strategies of betterment, references are made to other developing countries.

1. National shelter policies for the poor

3. The national policies adopted by the case-study countries dealt with in this report are all in line with the Global Strategy for Shelter to the Year 2000 (GSS). The focus is on enabling all actors in the shelter production and improvement process to utilize their full potential and resources. The final decision on the kind of shelter to build and the organization to adopt to achieve it must be left to the people concerned and their needs and priorities. The role of the government and public authorities at various levels should be one of facilitating shelter construction by establishing more appropriate regulatory frameworks and shelter financing schemes allowing the private sector, non-governmental organizations (NGOs), community-based organizations (CBOs) and individual households to make their effort and contribution. It is realized in the case-study countries that public provision of shelter to all the urban poor is impossible to achieve. In fact, most houses are built by the people themselves with and without the assistance of private small and large-scale contractors. An efficiently functioning market for building materials, among other things, is thus of utmost importance.

4. Programmes, subsidies and initiatives for shelter for the poorest groups are, however, not abandoned by public authorities in the case-study countries. Policies have changed from being a mere provider of physical and social infrastructure and some assistance in plot and building-material provision to the poor, to an emphasis on setting the conditions right for self-help and mutual aid, as well as community participation in upgrading projects.

5. The implementation of the shelter policies adopted has not been a total success in the case-study countries. Legal restrictions and building regulations have, for instance, not been reformed to any substantial degree. The availability of affordable and relevant building materials for the poor has not been significantly improved. Popular participation in the various stages of upgrading planning and implementation is only rarely taking place. Generally speaking, much is as before in the actual shelter practices, public as well as private. A change that has taken place, is the inclusion of cost recovery in upgrading programmes. The beneficiaries are to pay according to the full cost of improvements made in water provision, path construction, drainage, street lighting and so on. Cross-subsidization is, however, applied to some extent, making the better-off people in parts of an upgrading area pay more than the poor in other parts. Cost recovery and “the user pays” principle is seen to be required in order for replicability of shelter and upgrading programmes to be possible. The poor have evidently been caught in a cost-affordability squeeze. In addition to betterment taxes, the costs of building materials have increased more than wages have risen. The real incomes of most poor people in the developing world have deteriorated during the latter half of the 1980s and now in the 1990s. This new aspect of the shelter policies of cost recovery and reduction in subsidies have been detrimental for improved shelter for the poor. An ameliorating fact is that public authorities are not everywhere and always enforcing “the user pays” principle. The recovery rates for betterment charges and plot payments are very low.

6. A negative (and unintentional) consequence of reduced public construction of housing for the non-poor, without a simultaneous increase in private-sector building, is a downward pressure on cheaper housing. The gentrification of certain parts of urban residential areas pushes the poor into areas with worse environments and accommodation. This may be a temporary phenomenon until markets function better, thus giving the right signals and prices for a higher involvement of private capital in housing schemes.

7. The case-study countries invest relatively little of public funds on housing. One of the most important aspects of the GSS is thus not yet implemented, and it may not be fully accepted at policy-decision levels. That is the fact that housing for the better-off and adequate shelter for the poor are relatively effective sectors in which to invest in order to have an impact on national economic growth. Adequate shelter, including dwellings, privacy, security, lights, basic infrastructure and services and a location making the journey to work feasible - provided by the poor themselves through an enabling strategy - creates relatively much employment, high multiplier effects and national backward linkages.

2. Poverty

8. It is difficult to estimate how many among the 1.6 billion urban dwellers in developing countries live in inadequate housing with little or no provision of water, sanitation and other services. It is typical for 30 to 60 per cent of the urban population in developing countries to live in illegal settlements, and in overcrowded and deteriorating tenements. There is normally a mismatch between estimates made on the basis of case studies in the large cities and official figures which tend to show a much better provision of basic amenities and less slums and squatter settlements. In the case-study countries, urban poverty, slums and pavement dwellers are increasing. Lack of land for shelter construction has become an acute problem everywhere, and vast slums are found at the outskirts of the cities. The poor urban communities in the big cities rank among the most life-threatening and unhealthy living environments that exist. The poor's living conditions are far beyond the individual's capability to alleviate, and rest on lack of employment and resources. Shelter policies cannot solve the issue of mass poverty in urban areas in the developing countries. It can, however, have an important positive impact if rightly conceived and implemented together with the poor themselves. The impact would be greatly enhanced if shelter provision became an integral part of an overall development strategy for the urban poor.

9. There is no common agreement on the content and operationalization of the term “poverty”. It is useful to distinguish between the following degrees of material and psychological hardship: food poverty (absolute or starvation poverty); subsistence poverty (divided into a narrow and a broad basic needs concept); and relative deprivation (of either a social coping or a social participation kind). The last category of limited access to social participation can easily be confused with the concept of inequality and with different life-styles in a society, By making the concept “poor” completely relative to the typical living conditions in a society, the people with for instance inadequate shelter (e.g., a leaking roof, no heating etc.) cannot be termed poor if that is normal in their “society.” Hence, it is necessary to retain an absolute meaning in the concept of poverty, referring to material inadequacies and a life characterized by hardship (this will also, to some extent, be relative through time and space).

10. This report limits the category “poor” to those in food poverty, plus those in subsistence poverty in a narrow sense. It happens that case studies, which establish a poverty line for a country, also include some expenditures for travel and recreation. This kind of broad subsistence poverty is making the term “poor” too broad to be applicable in this report. The term “lower-income group” will be used for this category. This group of people has a sufficient material consumption to allow them an active and fairly secure life concerning food, water, clothing and shelter. They do, however, have an income below what is typically necessary for complete social coping and participation in a society's normal activities and events. The term “low-income groups” is used when referring to both the poor and the lower-income group.

11. In Indonesia and Mexico, the bottom 20 per cent on the urban income scale constitutes the urban poor. This is not meant to be an exact figure but indicative of the magnitude only. The urban population between 20 and 40 per cent of income distribution makes up the lower-income group. In India, the urban poor cannot meaningfully be limited to the 20 per cent with lowest income. The relevant figure is 35 per cent. The lower-income group is made up by those with incomes from 35 to 45 per cent on the income ladder. The above categories of people in the three countries have fairly comparable levels of needs satisfaction on average. The internal variation in the respective categories of levels of poverty may be somewhat different. This will have little bearing on the discussion in this report. The size of the categories “food poor” and “narrow subsistence poor” in the countries cannot be established, due to lack of reliable and comparable data. Again, such a quantification is not required for the purpose of this report. It suffices here to underline that urban poverty in the narrow needs satisfaction sense applied here, is extensive and growing in absolute numbers.

3. Shelter needs and delivery

12. The urban shelter shortage in India is estimated at between 6 and 13 million units. Some studies arrive at higher figures. One third of all urban households are without access to a latrine, and 45 per cent have only one room. One estimate holds that 30 per cent of the urban poor live in dilapidated shack-like structures and mud dwellings. The total production of houses during the last 30 years is a tiny proportion of the real need for shelter for the poor.

13. In Indonesia, urban shelter needs are enormous and increasing. An estimated total of 2 million units are required annually to cover the national need during the 1990s. The Government is supporting the dwelling needs of the poor to only a very limited degree. It is, however, supporting the shelter needs indirectly, by extensive slum (or kampung) upgrading programmes of basic infrastructure such as roads, water and drainage. This improves the value of plots somewhat and induces people to invest in their shelter.

14. The housing need in Mexico is vast, estimated at 6 million units in 1990. During the 1990s, there is an estimated annual need for 277,000 units due to population growth and 308,000 units due to deterioration of the housing stock. In Mexico City's metropolitan area alone, as many as 10 million people are living in slums. During the 1947-1990 period, public housing agencies were involved in the construction of only about 804,000 housing units for the low-income groups. A total of only 26,378 units were constructed for the poor in 1991. The slum settlements are often found in swampy, inhospitable salt-flats, on steep hillsides and on garbage dumps. The land and street layouts are chaotic and services are often lacking.

15. In all the case-study countries, renting is becoming more common among the urban poor. This poses an additional problem, since the incentive to invest in accommodation is clearly less when the shelter is not owned by the inhabitants. This also leads to a lack of interest by the residents in betterment of their local community environment.

16. Currently, between 60 and 90 per cent of all new housing units are constructed illegally, either on land without permission, or at standards not in accordance with official regulations. This is not surprising since developing countries invest only between 2 and 8 per cent of their GDP in housing. The total external assistance to housing in these countries is a mere 4 per cent of total investments.

4. Major problems of shelter delivery for the poor

17. The experience of the case-study countries is that even the shelter programmes and credit arrangements specifically designed for them are unaffordable to most of the poor. Most public resettlement projects have high rates of reselling and return of the poorest from the relocation sites at urban fringes back to the city centres where a survival income is easier to obtain. Land for shelter construction purposes (for all income groups) is in extremely short supply in most cities in the developing world, mainly as a result of high opportunity costs.

18. The poor's reliance on intermediaries and mafia-like gangs to find shelter, is reported to be a problem in the case-study countries. Furthermore, the high interest rates on credit and loans, and the need to pay substantial initial “down-payments” even for renting a dwelling or space on a pavement, make even the tiniest and most inadequate shelter rather expensive. The need for support to the establishment or strengthening of CBOs is recognized.

19. Women are racing particular problems regarding the question of plot rights and collateral for credit. Moreover, shelter location is often essential for women's economic activities, which are often conducted at home simultaneously with care of children and old parents. Women, thus find it difficult to take part in resettlement programmes, also owing to the increased distance to markets.:

20. The availability and cost of simple and/or traditional building materials are restricting minor and gradual improvements of poor people's dwellings. This is a particular problem for female-headed households because of their much more limited spatial mobility.

21. Information about and knowledge of ongoing upgrading projects, subsidy programmes and building regulations are low among the poor. This hampers the lower-income group's, but especially the subsistance poor's, ability to provide improved shelter for themselves.

22. The major problem in rental housing for the poor is how to identify the most appropriate rate of rental increase, which ensures both that the poor can afford to pay and that owners maintain the buildings.

5. Agenda for further research

23. The problems experienced in the case-study countries and in other developing countries of improving the shelter conditions of the poor underline the need for a strategy that combines the resources and motivation of the poor themselves (organized in CBOs), various kinds of NGOs and public authorities at different levels. The poor cannot afford to rely on the market and private contractors to cover their need for adequate shelter. Public agencies do not have enough funds to provide shelter and basic amenities to all the urban poor. The case for the involvement of national and international NGOs (and other donors) is thus very strong. The challenge is thus to create a model which would integrate the different capabilities of the various actors in the shelter sector and to coordinate their efforts in specific areas and programmes. The responsibility for initiating and coordinating such a shelter strategy for the poor must primarily rest with governments. By involving CBOs from the very beginning of upgrading, resettlement and other types of shelter programmes for the poor, the potentially substantial contribution of poorer groups can be effectively activated. The focus must then be on identifying the barriers and bottlenecks to the full participation of local-level organizations, in collaboration with NGOs, in shelter construction and community betterment.

24. A more detailed summary of the recommendations is included at the end of the report, together with an outline of future research priorities. The following must be underlined: first, official statistics on the magnitude of and conditions in slums and squatter settlements have low reliability. Case studies, although of a fairly large number and widely distributed, do not provide sufficient data to make reliable estimates of the relative proportions of groups living in different kinds of shelter and environmental conditions. Secondly, very few data are available and few empirical studies have been undertaken on poverty that splits this broad category into various levels and types of poverty. The heterogeneity even inside one slum area is great. To avoid programmes and projects having negative consequences for major groups among the poor, extensive research efforts should be made on this heterogeneity of income levels, occupations and life situations in general, together with variations in poverty and livelihood in other urban housing forms such as high-rise flats, tenements etc.