Commission on
Human Settlements
3 January 1995


Fifteenth session
25 April - 1 May 1995
Item 7(a) of the provisional agenda
       Activities of the United Nations Centre for
       Human Settlements (Habitat): progress report of
       the Executive Director

Report of the Executive Director


Almost twenty years have passed since Habitat: United Nations Conference on Human Settlements took place at Vancouver. In the run-up to Habitat II, scheduled for Istanbul, Turkey, in 1996, the Commission on Human Settlements has requested a review of "national action to provide housing for all since Habitat". The report comprises five chapters: chapter I provides the introduction to and background of the study, while chapter II reviews national and international action on developing and implementing shelter strategies, including a summary of their results. Main bottlenecks and successful approaches in the shelter process are analyzed in chapter III, paying particular attention to scale sustainability and poverty-reach, and to the needs of particularly vulnerable groups. Chapter IV takes a broader view, identifying a series of common themes, issues and challenges which emerge from the experience of the last two decades and assesses the extent of progress which has been made. The conclusions and recommendations of the report are presented in chapter V.

I. Introduction and background 1-9
II. Review of national action and international support to national shelter strategies 10-58
  A. The formulation of national shelter strategies 11-18
  B. Appraisal of shelter performance by sector 19-40
    1. Land 20-25
    2. Housing Finance 26-29
    3. Building materials and the construction industry 30-31
    4. Shelter production 32-34
    5. Services and infrastructure 35-37
    6. Wider Goals 38-40
  C. Assessment of capacity to make, monitor and implement national shelter strategies. 41-45
  D. International support to national shelter strategies 46-58
    1. Resource flows 47-48
    2. Support to urban management and capacity-building 49-58
III. Bottlenecks and successful approaches 59-70
  A. Bottlenecks 59-63
  B. Successful approaches 64-70
IV. Eemerging themes, issues and challenges 71-83
V. Conclusions and recommendations 79-83
VI. Selected bibliography and references


[Back to Table of Contents]

1. Between 1990 and 2030, it is estimated that the global population will have increased by approximately 3.7 billion people. Ninety per cent of this increase will be located in developing countries and 90 per cent of that will take place in urban area. At least 600 million people already live in life-threatening situations in cities, and up to one third more live in sub-standard housing. At least 250 million urban dwellers lack access to safe drinking water, and 400 million lack access to adequate sanitation. Although the proportion of people in the developing world defined as "poor" may be declining in some regions (particularly East Asia), absolute numbers are still rising - from 1,051 million people in 1985 to 1,133 million in 1990, according to the World Bank.

2. These statistics provide a stark reminder of the scale and urgency of the problems facing development in urban areas. "In the interest of better living environments for all, you will, as the community of nations, commit yourselves to nothing less than a revolution, a revolution of new solutions, new ideas, new policies, new tools and new institutions." With these words, the Secretary-General of the second United Nations Conference on Human Settlements (Habitat II) issued at call to action to the first session of the Preparatory Committee for Habitat II. But such calls to action, and to new ideas and solutions have been heard before. Almost twenty years ago, the participants in Habitat: United Nations Conference on Human Settlements made a commitment to a bold and imaginative program of action designed to address the shelter problems of the poor. As Habitat II draws near, the poor are entitled to ask what real progress there has been on those recommendations. It is, therefore, appropriate and timely that national and international action to provide housing for all since 1976 is reviewed.

3. Over 90 per cent of the Growth in world urban population between 1990, and 2020 will take place in developing countries, an extra 2.2 billion city-dwellers who will place enormous strains on the capacity of urban economies and administrations to support their need for land, shelter, services and jobs. Rising urbanization has not been universally accompanied by declining shelter conditions over the last 20 years since local responses vary greatly. Recent research has thrown up some surprising conclusions, while confirming the underlying seriousness of the situation facing poor people in most countries. While "commercialization" (the exchange of land, housing and associated services on the market) does seem to be a global trend, empirical evidence shows that the impact of commercialization on access to shelter, and on shelter standards, vanes significantly from one city to another. In some cities, land and house prices, rents and construction costs seem to have risen more rapidly than per capita incomes. However, this picture is not a universal one. Elsewhere, the housing options of the urban poor may not be declining, and may even be expanding as incomes keep pace with prices and/or people adapt, to declining conditions' in the housing market in different ways.

4. Nevertheless, there is little solid evidence to show that the shelter situation of the poorest people in the great majority of cities has improved over the last two decades, and there is some indication that the gap between these groups and the better-off is growing wider. This would certainly confirm the patterns that are emerging in most North American and European cities, where increasing social and economic polarization is giving rise to real concerns about political stability and urban integration. From this diverse picture, three conclusions suggest themselves. First, the recognition of diversity is itself an important conclusion, and a theme which reappears throughout the report. The shelter conditions of the urban poor in developing countries are not declining across the board - trends vary greatly from one city to another, and according to time, gender, social class, location and other characteristics within the sane city. Secondly (and this is an important qualification to the first conclusion), there is no evidence that the shelter situation of the poorest people in developing-country cities has improved over the last two decades, as measured by affordability, tenure, standards and access to services. In absolute numbers, more urban residents are facing a shelter situation which must he considered inadequate. Thirdly, there is evidence from many cities (both developing and industrialized) that inequality is increasing and, with it, the attendant dangers of social and political conflict.

5. The shelter options of the urban poor are closely linked to trends in urban poverty (though, as has been made clear, the links between incomes and housing vary greatly within and between one city and another). The pattern of events here seems rather clearer, with most observers confirming a significantly rising trend in urban poverty (both absolute and relative) over the last 20 years,linked to the wider macro-economic changes outlined below. The urban poor have suffered disproportionately from the economic adjustment process, which started in the 1980s. Trends in urban shelter are intimately related to trends in the wider economy. This has been one of the most important lessons that have been learned in the twenty years since 1976. However, as a recent UNCHS (1994, 1) paper puts it. "From a macro-economic perspective the road from Vancouver was a bumpy one". From the viewpoint of many developing countries, particularly in sub-Saharan Africa, this can he seen as something of an understatement. The fact that many countries have indeed had a "bumpy ride" has been of profound importance to the implementation of the Vancouver recommendations.

6. Trends in the world economy have impeded the realization of the Vancouver Declaration in two main ways, neither of which was foreseen at the time. First, the general economic outlook for developing countries appeared much rosier in 1976, prior to the oil price shock of the late 1970s and at a time of considerable optimism about the future prospects for the world economy. Secondly, the strategy chosen by the international donor community to address economic decline and indebtedness in developing countries has been "economic and structural adjustment", the liberalization of developing-country economies and their further integration into global markets via the maintenance of market exchange rates, equalizing the balance of payments, export promotion, private-sector growth, the removal of subsidies, and cuts in public expenditure. The long-term impact of adjustment policies is a subject of continued debate. In the short term, there is widespread agreement that too much was attempted too quickly, and without adequate reference to the social impact of adjustment, local variations and political realities. As a result, people in cities across the developing world (but, again, particularly in Africa) suffered a decline in their real incomes, an increase in food and energy prices, and widespread dislocation in labor markets.

7. Trends in housing policy since 1976 mirror the evolution of general ideas about development and development intervention during that period, and occasionally (as in the case of gender planning) they have also made a significant contribution to these broader changes. In determining the actions and decisions of donors and governments, and shaping the political and intellectual frameworks within which different actors play out their roles in shelter development, such shifting orthodoxies are of paramount importance, Four influences deserve mention in this respect. First, there has been a significant shift in the reluctance of international agencies (and some developing-country governments) to invest in urban areas for fear of stimulating rural-urban migration and retarding still further the development of the rural and agricultural sectors. Secondly, the Last two decades have seen a profound change in the intellectual underpinnings of development policy and practice, from the interventionist climate of the 1960s and early 1970s, through the "neo-Liberalism" of the 1980s, to the beginnings of another re-assessment in the 1990s which focuses on harmonizing the comparative advantages of markets, States and third-sector organizations. This "new policy agenda" is not, however, exclusively economic in character. It also signals official support for political reform and democratization, the process by which citizens come to play a fuller role in decision-making and the wider political process. Popular participation and decentralization, while applied very imperfectly in practice, now receive much more official support than they used to. Thirdly, development agencies of all types have recognized the limitations of traditional "blueprint" approaches to development intervention and are embracing (with varying degrees of enthusiasm) "process" approaches, which emphasize learning, experimentation, flexibility and the gradual nature of progress. Fourthly, the whole concept of "development" has changed since 1976, with the rise of gender planning over the last 15 years, the acceptance of "sustainable development", and, with the forthcoming World Summit for Social Development to be held in Copenhagen in 1995, the integration of these various elements into a new paradigm of social or human development which is holistic in character. The next stage in the development of shelter policies will be determined by Habitat II in 1996, and may well involve a sharper debate between the "human development agenda" of the United Nations agencies and the World Bank's greater emphasis on markets and economic development, albeit both under the broad umbrella of the "enabling approach" of the Global Strategy for Shelter to the Year 2000.

8. Clearly, the changes since 1976 have altered the urban, social, economic, and political landscape in dramatic fashion, Have these changes made it more, or less, difficult to implement the recommendations made at Vancouver? On the positive side, the evolution of shelter policies over the last 20 years provides a clear affirmation of the basic philosophy espoused in Vancouver. The development of enabling approaches and the rising attention given to gender concerns and environmental sustainability have provided a supportive intellectual and policy framework within which to apply the recommendations of the Vancouver Declaration' and Action Plan (with some notable exceptions, which are described below). However, the economic context within which shelter policies have been applied in practice has not been favorable to many countries over the last two decades. Rising urbanization and urban poverty, coupled with declining government resources, have rendered implementation of the Vancouver Action Plan extremely problematic. A lack of commitment to implementation among some governments and donors has also hampered progress. One of the clearest lessons of the last 20 years is that significant shelter improvements cannot be achieved in a declining macroeconomic climate. As the Asian Coalition for Housing Rights puts it, "although the Habitat I recommendations were innovative and appropriate, the story of housing in the Asian region in the seventies and eighties is a dismal one."

9. However, problems of implementation do not invalidate the achievements, nor dim the vision of the Vancouver Conference. This remains a milestone in the development of shelter policy, and, as laid out below, its influence can be seen clearly in the evolution of later policy statements. The Conference put the human settlements sector on the international agenda for the first time. There are at least three areas in which it has been especially influential:


[Back to Table of Contents]

10. The Vancouver Action Plan recommended that "all countries as a matter of urgency should establish a national policy on human settlements" (UNCHS, 1976, recommendation A.1), a goal which was given further impetus in the Global Strategy for Shelter. This chapter examines the record of national and international action in support of this recommendation, looking first at the process of strategy formulation; secondly, at the implementation of national shelter strategies, and, thirdly, at national capacity to develop and monitor results. The final section of the chapter examines the record of international support to national shelter strategy formulation and implementation, with a particular focus on resource flows and support to capacity-building. A comprehensive evaluation of strategy formulation and implementation across the world is impossible in a report of this length, so a small number of case studies are used to illustrate key themes. Examples of bottlenecks and successful approaches to particular aspects of shelter development are provided in chapter III. As a further caveat, it should be remembered that less than 20 years have passed since the Vancouver Conference. Although some progress is to be expected in this time, most national shelter strategies did not get underway until the late 1980s. A systematic evaluation of the impact of these strategies is, therefore, premature.

A. The formulation of national shelter strategies

[Back to Table of Contents]

11. The recommendations concerning national shelter policy focused on the need to integrate such policies with national economic and social development plants, promote equity of access to benefits and services, and be evaluated at regular intervals (recommendations A.1 to A.7). Detailed requirements for national strategies (as opposed to policies) were not defined until 1988. Nevertheless, some countries went ahead after Vancouver and developed comprehensive shelter policies. An early review of national action across 17 countries (carried out by the International Institute for Environment and Development (IIED) between 1978 and 1980) found evidence of progress in all of them (especially in Egypt and Singapore), though few had addressed land issues in any systematic way (Searle and Hughes, 1980, 101). By 1987, a second evaluation (commissioned by UNCHS) found encouraging signs in Botswana and Jamaica, and emphasized the importance of having a written shelter policy as a means of generating national understanding and consensus (Van Huyk, 1987).

12. However, it was not until the elaboration of the Global Strategy for Shelter that clear and comprehensive guidelines were laid down for national shelter strategies, The International Year of Shelter for the Homeless (IYSH), one year earlier, had stimulated a good deal of action in preparing draft national strategies: out of 145 countries, which participated in IYSH, 55 established new shelter policies or strategies, 11 started preparations for new strategies, and the rest "had made at least one significant policy change or new program" (UNCHS, 1990d, 35). The Global Strategy for Shelter was built on these foundations, and intended to be implemented in three phases: the formulation of national shelter strategies and design of new institutional arrangements (1989-1991); their introduction in practice (1992-1994): and full-scale operation of national programs (1995-2000). These phases were intended to be evaluated according to criteria established by the twelfth session of the Commission on Human Settlements, as follows (UNCHS, 1991b, 118): strategy formulation (definition of goals, political commitment and the process of strategy formulation); strategy implementation (mechanisms for monitoring and evaluation, and legislative, financial and institutional re-organization); and strategy performance (building and infrastructure output, equitable distribution of outputs, community participation and environmental conditions). The Global Strategy for Shelter itself laid down four requirements for national strategies (UNCHS, 1990a, 14-15):

13. Agenda 21, adopted by the United Nations Conference on Environment and Development (UNCED) added one further requirement to this list, namely that national shelter strategies should include measures to promote "sustainable energy development and transport systems" (UNCED 1992, para 7,52). This is a formidable list, especially for countries in which government resources are already stretched.

14. Over the last three years, a number of studies have been carried out to review progress in developing national shelter strategies (Struyk, 1990; UNCHS, 1990e; DPU, 1991; UNCHS, 1991b; UNCHS, 1992g; Wakely and others, 1992). There have also been numerous country-specific studies, which are cited below, A good deal of information is therefore available on strategy formulation, especially from Colombia, Costa Rica, India, Indonesia, Jamaica, Kenya, Mexico, Nicaragua, Nigeria, the Philippines, Sri Lanka, Thailand, Uganda and Zimbabwe. In general, these studies conclude that a start has been made in developing national shelter strategies in most countries, but that no country has a complete strategy as yet. In addition, the output of the strategy-formulation process varies widely from one country to another, some have concentrated on modifying existing plans from earlier years, while others have focused on particular sectors of shelter development (UNCHS, 1992g, 91-100). In most cases, some elements of strategy were already in place prior to 1988 as a result of earlier attempts at policy-formulation, beginning in 1976 (UNCHS, 1991b). For example, Costa Rica and Zimbabwe have concentrated on changes at central-government level; Nicaragua and Uganda have emphasized the development of district and municipal strategies: and Indonesia and the Philippines have focused on the development of links between central and local-level planning and action (Wakely and others, 1992, 2). Some countries (such as Uganda and Zimbabwe) have kept to their original schedules, with strategies and mechanisms for implementation agreed and adopted by 1992; but most have fallen behind schedule for one reason or another. In Indonesia the Government recognized the need to adopt an "enabling approach" to shelter in its Five-Year Plan for 1989-1994, though it still has no single, explicit strategy document (Hoffman and others, 1990; Van der Hoff and Steinberg, 1993). Kenya signaled its move away from State-dominated shelter policies to the enabling approach with a Sessional Paper on Housing in 1986, followed a year later by a National Shelter Strategy and in 1990 by a new Housing Policy Paper. However, Mitullah (1994) reports that the impact of these changes has been "disappointing", in pan because housing policies have been fragmented and uncoordinated, and because of problems in the institutions intended to implement them, Similar problems have been noted in the United Republic of Tanzania (Siren, 1990).

15. India's new National Housing Policy (1992) evolved from an earlier policy adopted in 1988, and embodies a clear commitment to the principles enshrined in the Global Strategy for Shelter: "the Government will devise and implement strategies which will enable the various agencies to complement the efforts of one another and to ensure the most efficient utilization of resources" (UNCHS, 1992d, 21). However, Indian NGOs have yet to be persuaded that the new policy will be implemented effectively; "although India has continued to describe its, policy for housing as an enabling one, it has done little to create mechanisms which assure in concrete terms the enabling principle" (Patel and Burra, 1994, 1). A similar judgement comes from Bolivia, where Richmond (1994) reports that, though "bold in its outlook and wide in its application", the National Plan for Housing (1989-93) has not been implemented. Mechanisms for implementation have also been the weakest area of strategy development in the four countries reviewed by UNCHS in 1990 (Barbados, Jamaica, Kenya and Zimbabwe). A common failing is that the functions of, and linkages between implementing agencies are not spelled out in any detail (UNCHS, 1990e, 15). All four country strategies define an increased role for the private sector in shelter production and qualitative changes in the roles of government agencies, but again this means different things in each case. For example, government maintains a more significant role in Zimbabwe than in the other three countries reviewed. In common with other similar studies, UNCHS found wide variations in the details of scheduling, the groups and institutions involved in drawing up the strategies, the degree of technical assistance involved, and the precise steps that were taken. Struyk (1990, 317) makes the telling comment that three out of the four were closer to "policies" than "strategies",since they lacked a detailed timetable and procedures for implementation.

16. Key factors in differentiating successful strategies from those that remain "on paper" are high-level and continuous political support, careful research on housing market conditions prior to strategy formulation, and local control - donor involvement in initiating the process should be avoided (Struyk. 1990, 319). There is also evidence that broad-based consultation among public, private and third-sector institutions produces a stronger outcome, whether this is achieved through a committee structure (as in Kenya) or via a series of hearings (as in Indonesia). In Nicaragua, an innovative process of workshops and data collection at municipal level produced a number of "diagnoses" of shelter issues which were then discussed at meetings of all the relevant actors, resulting in the production of local action plans during 1990 and 1991 (Wakely and others. 1992. 12). A similar process was undertaken in the Philippines, which focused on developing strategies at the regional level. Uganda was also successful in involving a wider range of staff at both local and district levels in the development of a new National Shelter Strategy in the early 1990s, the focus being on building from the district level upwards rather than the central level downwards. A similar "bottom-up" process in Zimbabwe did not work so well, since the State system is more highly centralized than in Uganda (Wakely and others, 1992).

17. Taking the four requirements for national shelter strategies laid out in the Global Strategy for Shelter to the Year 2000 (summarized above), the following conclusions can be drawn:

18. In summary, substantial progress has been made in most countries in re-defining aims and objectives through the national shelter strategy process to coincide with the principles of the enabling approach, echoing the philosophy laid down in 1976. In many cases there has been wide consultation with private-sector groups and popular movements during the strategy process, though (necessarily) the details of who was involved at what stage vary widely. However, it is also true to say that most "strategies" lack a detailed plan of action, time-scale, provision for resourcing, and indicators for monitoring and evaluation, They are closer to "policies" than "strategies." This is one reason why implementation has thus far been disappointing. The lessons learned from this phase of strategy development confirm the results of an earlier evaluation of the enabling approach, which identified four preconditions for success (UNCHS, 1987, 205-206).

B. Appraisal of shelter performance by sector

[Back to Table of Contents]

19. It is one thing to develop a national shelter strategy, and another to implement it effectively. Weaknesses in implementation have already been identified as a key issue in most of the strategies developed to date. At this point in the report it is worth pausing to reflect briefly on the performance since 1976 of shelter policies in different sectors - land, finance, infrastructure and so on. Detailed examples of bottlenecks and successes can be found in chapter III; the emphasis here is on broad trends, and the time-scale used for evaluating performance in the last twenty years; national shelter strategies have not been in existence long enough to permit an evaluation of this sort, so the conclusions laid out below refer to the whole of the period since 1976. To what extent have the recommendations of the Vancouver Action Plain been implemented

1. Land

[Back to Table of Contents]

20. As mentioned in the Introduction to this report, the spirit of the Vancouver Conference was more strongly interventionist than later shelter policies, and nowhere is this more evident than in the Vancouver recommendations on land markets. The accent was on "social justice" promoted via public ownership of land to prevent the "concentration of wealth" through private accumulation in the land market (UNCHS, 1976, 28). Detailed recommendations included "complete government sovereignty" over land (D.1): zoning, legal and fiscal controls, and direct intervention (land banks etc.) to exercise public control over changes in land use (D.2): re-capturing for the State any increase in the "unearned increment resulting from the rise in land values" due to changes in land use (D.3); and improved land registration and information systems (D.7). All but this last recommendation has since been diluted considerably as public and private roles in shelter have been progressively re-defined over the last 20 years.

21. However, a common conclusion of studies of shelter development since 1976 is that land supply has been the area of greatest failure (UNCHS, 1991f, 18). As the Global Strategy for Shelter put it, "the greatest failure of governments in the housing sector has been the incapacity to stimulate a supply of sufficient, affordable and officially-recognized serviced land" (UNCHS, 1990a, 32). At the root of this problem is the fact that "often with the best of intentions, governments have adopted policies which have contributed to land shortages rather than land availability, emphasizing control and regulation over land use and supply rather than enabling and facilitating the supply and use of land" (McAuslan, 1994,5). A major reason for this is that, although very inefficient, such measures benefit those who operate them, rewarding politicians, bureaucrats and speculators in equal measure. Land has become a vehicle for political influence as well as economic gain (Baross and Van der Linden, 1990), and the crucial importance of creating alliances to act as a counterweight to this process forms a major theme of chapter IV. Whether it is right to lay so much of the blame for inefficiencies in the land market at the door of governments is debatable. Urban land markets are and have always been highly imperfect, tending to be dominated by powerful interests aiming at manipulating supply in order to secure speculative gains (Doebele, 1994). Large amounts of land are usually held vacant, even in cities where demand is very high: in Bombay, for example, it is estimated that at least 20,000 hectares of vacant land exists, sufficient to re-house the entire population of the city's illegal settlements, rental tenements and pavements (Hardoy and Satterthwaite, 1989, 101). Although reliable comparative price data acre rare (especially from illegal or semi-legal land markets), some researchers conclude that poor people have to pay higher prices in many developing-country cities in the private market than they would have to pay if government had supplied land directly to consumers (Baross and Van der Linden, 1990). Whether or not this conclusion is representative of developing-country cities in a whole, there can be no argument with the fact that formal land markets are failing to provide land for all who need it: rising tenant populations and the continued growth of housing which must be considered "sub-standard" from any point of view bear witness to this fact.

22. Government intervention is essential if supply constraints are to be removed, but misdirected intervention can make the situation worse rather than better. The often-quoted failure of the Urban Land (Ceiling and Regulation) Act (ULCRA) to unlock the huge supply of vacant land in Indian cities is an example of well-intentioned government action, which has had the opposite effect to that intended. By 1990, only 5000 out of an estimated 327,000 hectares of available land had been acquired under the Act; the freezing of vacant land had contributed to an even faster rise in land prices, and the administrative costs of administering the legislation were enormous (UNCHS, 1991f,14). This is a familiar story, and it is certainly true to say that the record of government regulation in most land markets has been depressing. Regulatory reform (with the emphasis on liberalization) forms a major component of the national shelter strategies developed since 1988. In the United Republic of Tanzania for example, the recent repeal of laws governing nationalization of property worth more than Tsh 100,000 has led to an immediate increase in the land and housing stock available for purchase (UNCHS, 1991e, 27).

23. One of the mitigating factors in this situation is the existence in many cities of a range of "informal" (illegal or quasi-legal) land markets which do provide land to the urban poor at prices many of them can afford, albeit usually unserviced and subject to indeterminate property rights. These (and markets, such as the illegal subdivisions of Santafe de Bogota, are the main source of lots for poor people (something which the Vancouver Action Plan did not really acknowledge), and if governments are unable or unwilling to address supply constraints in the formal market there is certainly no reason to block the operation of these informal mechanisms. Most governments are coming round to this position, concentrating on facilitating the operation of informal markets in such a way as to protect low-income purchasers (for example, through upgrading property rights) and maintain at least a minimum level of planning and infrastructure standards. A "secure claim on the system", as Doebele (1994, 48) puts it, is often more important than legal title per se in promoting investment in shelter by poor people. This, and provision of basic infrastructure in a coordinated way in areas deemed "informal", is the most important measure government can take to improve the functioning of urban kind markets. For example, once basic services had been installed in illegal subdivisions in Lima and Lusaka, take-up of legal titles was very slow, but upgrading continued (Doebele, 1994, 48). This less formal process of land development also allows poor people to capture more of the increase in land values for themselves, rather than seeing such profits being channeled to urban elites or to governments (though government can and should recapture part of the increase through property tax revenue and service charges).

24. Despite the gloomy overall record on the land question, there are some examples of innovatory approaches. These include:

25. Despite these innovations, the scale of unmet needs in the land markets of most third-world cities remains huge. Encouragement to informal land markets makes sense, but without action at a more fundamental level to increase access to land among the urban poor, this will only bring a temporary respite.

2.Housing finance

[Back to Table of Contents]

26. The Vancouver Action Plan paid comparatively little attention to financing issue, much less than the Global Strategy for Shelter, for example. The "financial arrangements" laid out in recommendation F,8 of the Action Plan stated that special attention should be paid to protecting investors from inflation (especially the poor), encouraging joint ventures between public and private capital, selective use of public funds "where private investment is unlikely", encouraging community and cooperative savings schemes, support to mortgage finance, and "ensuring that systems for financing community infrastructure result in an equitable distribution of costs within and between households" (UNCHS, 1976, 43). Little guidance was provided on how these difficult goals were to be achieved, and as with the land question, the record of the last two decades in providing affordable housing finance to poor people has been disappointing.

27. This is not surprising, since all housing-finance systems have to reconcile three partially-conflicting objectives: affordability to borrowers, viability to lenders, and resource mobilization for the sector as a whole (Renaud, 1984). In practice, one of these objectives usually ends up by dominating the others: affordable housing-finance systems turn out to be unsustainable, while viable systems tend to exclude the poor (UNCHS, 1991f, 25). Finding the right balance has proved a consistent and largely unresolved dilemma. The financing of housing and infrastructure is more difficult than that for other sectors because lending (long) and borrowing (short) usually follow different time-scales: administrative costs are therefore high, and housing finance is very sensitive to movements in inflation. Rapid inflation or interest-rate fluctuations are highly damaging to housing-finance systems, yet most developing countries lack the stable national economies which are required to prevent these problems from occurring (UNCHS, 1990c). Lack of access to affordable loans has proved a particular problem for women, due to gender biases in loan approval and repayment procedures and women's lack of collateral where property rights are registered in the name of the male head of household (Dandekar, 1993). In these circumstances, it is vital for governments to promote reforms in formal housing-finance systems, while simultaneously encouraging the development of less formal systems, which are more affordable to the poor (and on which they have always relied). Equally important is the need to link the two together, so that affordable housing-finance can be provided in secure fashion over the long term - for example by linking savings-and-credit groups of poor people to the commercial banking system, which brings with it benefits of sustainability and scale which NGOs and CBOs cannot match (Arrossi and others, 1994).

28. Examples can be cited of how housing-finance can be made more affordable to poor people. They include easing regulations on collateral (for example, Nigeria's new Housing Policy recognizes collective guarantees from cooperatives in applications for mortgages): cross-subsidizing interest rates (though there is little room for interest-rate subsidies if housing-finance systems are to be sustainable): flexible repayment schedules (as in India, where commercial banks allow for graduated repayments of loans): special funds and earmarked funds for low income groups (as used by the Bangkok Municipal Government to start a "community mortgage programme" modeled along the fines of its successful counterpart in the Philippines); catalyzing household savings (as in Hyderabad, Pakistan, where the authorities provided a small start-up subsidy in the form of cash and construction materials in order to galvanize subsequent, larger contributions from the community): and promoting NGO involvement in housing-finance, in the form of housing or savings cooperatives or using NGOs to mediate between low-income builders and the formal housing-finance system (UNCHS, 1991f, 29). Such credit guarantee schemes have been used successfully in Chile, where a local NGO (EVGL) coordinates loans for construction materials from formal-sector institutions: the scheme was successful enough to persuade local banks to risk some of their own funds in lending directly to poor households (Arrossi and others, 1994). Many similar studies show that commercial banks can be successfully involved in low-income credit provision as long as they receive adequate guarantees to reduce the risks of default. The experience of the Grameen Bank in Bangladesh shows that financial institutions, which deliberately blend together the characteristics of the commercial private sector and NGOs can be successful in lending to the poor on a very large scale.

29. Reforms in the formal housing-finance system, though usually not of direct benefit to the very poor, are important in reducing the pressures for downward filtering in the housing market. For example, Zimbabwe's new National Shelter Strategy pays considerable attention to improving the efficiency and reach of mortgage finance. With high-level backing from the Government, it has proved possible to persuade building societies to reduce the cost of housing loans in return for greater access to capital (in this case, tax-exempt bonds with 25 per cent of each issue earmarked for on-lending to "lower-income groups": UNCHS, 1990e, 20). Similar efforts have been made in Colombia and India via the Housing Finance Development Corporation (UNCHS, 1991f, 30). However, these innovations remain the exception rather than the norm. Formal housing-finance remains out of reach for most of the urban poor, and despite the availability of informal credit systems (which demonstrate that poor people can repay loans at market or even above-market interest rates), lack of access to affordable credit in sufficient quantities remains a formidable obstacle to "shelter for all."

3. Building materials and the construction industry

[Back to Table of Contents]

30. Recent research has confirmed that "self-help" housing is in reality a complex set of housing systems in which owner-builders and small-scale construction companies are both involved - and often the latter more so than the former (Mathey, 1992). For example, in Karachi, roughly 90 percent of all housing in informal settlements has been built by small, private contractors using informal credit (Hasan, 1994b, 17). Although the "self-help" spirit was very much in evidence at Vancouver in 1976, the Conference paid considerable attention to the need to support indigenous construction companies, local building materials, and appropriate standards. Recommendation C,3 of the Vancouver Action Plan states that "standards for shelter, infrastructure and services should be compatible with local resources, evolutionary, realistic, and sufficiently adaptable to local culture and conditions" (UNCHS, 1976, 18): recommendation C.7 cites the "special importance of the construction industry" and recommendation C,8 advocates strong government support to informal-sector construction. This includes "restructuring the system for marketing and distribution of building materials and tools to favor purchase in small quantities at irregular intervals and under easy credit terms", and "simplifying and adapting building and licensing codes without sacrificing recognized basic health requirements", this last recommendation coming as a timely reminder of the need to preserve some basic standards while lowering many others (UNCHS, 1976, 21). Misapplied standards have proved to be of "more benefit to developers than to the poor", as in the Colombian "normas minimas" which stimulated construction activity but at the cost of higher residential densities and lower-quality infrastructure (UNCHS, 1991f, 17). Many countries have made progress in reforming building and planning standards through a variety of variable zoning systems (as in Sudan) and "special development zones" (as in Jamaica and Kenya), though in some cases standards have simply been abandoned (Hardoy and Satterthwaite, 1989, 132). Less progress has been made in making explicit provision for the needs of women and the promotion of rental housing; in building standards or in support to the construction industry.

31. The impact of reforming building and planning standards is often dramatic. For example, World Bank research in Malaysia has shown that government regulation increased housing costs by up to 50 per cent, and has fallen particularly heavily on builders in the informal sector (World Bank, 1991, 39). Similar evidence comes from Zimbabwe, where the reduction in inappropriate building standards under the National Shelter Strategy reduced costs by 29 percent per serviced plot (though mainly via reducing the size of the plots: Schilderman, 1994). Appropriate standards should signal the "basic minimum required" (for example, that the dwelling would not collapse on its inhabitants) and allow for flexibility so that standards can evolve over time in tune with gradual upgrading, As the Minimum Building Standards now in place in Malawi show, performance standards are much more efficient than prescriptive standards, for example: "walls should stand up to local load requirements and climate", rather than "walls should be made of concrete" (Schilderman, 1994). These are important innovations, but it is important not to lose sight of the fact that building and planning standards areas often as not ignored by informal-sector builders anyway. In this situation, regulatory reform takes on less significance, though this is not an argument for abandoning it altogether: it is a matter of priorities. Similarly, the informal sector already produces most of the construction materials used by low-income builders. For example, the "unorganized sector" produces over 232 billion bricks each year in India (UNCHS, 1991f, 38). Government support to materials production and small construction companies via the lowering of standards and the promotion of access to affordable credit is important, but it is not so central to "shelter for all" as large-scale and decisive action in the land and finance markets.

4. Shelter production

[Back to Table of Contents]

32. The Vancouver Action Plan recommended that governments adopt an "integrated approach to shelter development" and special support to the "informal sector in its efforts to provide shelter for the less advantaged" (UNCHS, 1976, 21). This was to be achieved by promoting sites-and-services: support to "self-help construction," "aided self-help", and the local construction industry: and an "increased public role in renting" and rent subsidies (recommendations C.1, C.8 and C.9). The key to much of this lay in measures which are now accepted as standard practice (but which in 1976 were highly innovative): regularization of tenure in informal settlements, provision of basic services and infrastructure on empty lots, support to the efforts of the poor themselves, and so on.

33. The evolution of national shelter strategies over the last decade shows clearly that these recommendations have been listened to, and accepted. While action has been on too small a scale to keep pace with shelter needs, the basic principles laid out in the Vancouver Declaration have not been challenged since 1976: direct construction by government has given way to support for the efforts of private-sector builders and poor people themselves. Upgrading programs have been particularly popular, and some have achieved a significant scale (though often with considerable disguised subsidies). For example, the Million Houses Program implemented in Sri Lanka between 1984 and 1989, included 19 different housing options to choose from and has since been supplemented by an even more ambitious plan to support the construction of 1.5 million dwellings between 1990 and 1995 (UNCHS, 1992, 22). Indonesia's Kampung Improvement Programme is another often-quoted example of large-scale upgrading, implemented in over 500 urban areas since 1968 and responsible in part for the fact that dwelling size and residential densities in Indonesian cities are actually improving over time, despite the continued poor performance of the formal construction sector (Silas, 1994). What is needed is a continuous upgrading and/or site-and-service program (rather than projects) in which land is assembled, services installed, and plots leased or sold at prices the poor can afford on a continuous basis (UNCHS, 1990c). Other successful upgrading programs (such as Aguablanca in Call, Colombia, and Ruamjai Samakki in Bangkok, Thailand) demonstrate that the failings of earlier projects are being tackled - costs and subsidies have been reduced significantly,access to the very poor (though still no, the poorest) has been promoted, and security of tenure granted (UNCHS, 1991f, 57).

34. There has been less progress in other areas of shelter production, with the lack of attention paid to rental housing being particularly disturbing. There are some successful examples of rehabilitation of inner-city rental tenements (for example, in Bombay and Mexico City, where a combination of enlightened government support and dynamic community organizations enabled low-income tenants to remain in the city centre after upgrading had been completed), but the general record of official support to the rental market has been disappointing (UNCHS, 1993c, 89-91). Despite the obvious (and in many cities growing) importance of renting and sharing, housing policies have been slow to respond. National shelter strategies usually identify the promotion of rental housing as an objective, but have done little to promote it. There are some exceptions to this rule - for example, in Colombia and Indonesia, where credit is being provided specifically for investment by landlords in small-scale rental housing (Hoffman and others, 1990). Instead of promoting the widest possible range of housing alternatives so that people can choose what they want, shelter strategies still tend to promote one form of tenure (usually home-ownership) over others. This is despite the fact that research has shown that landlordism among the urban poor is usually a small-scale and non-exploitative affair (Edwards, 1990: Gilbert and Varley, 1992); that many low-income households actually prefer renting to ownership; and that a variety of policy measures are already in hand to stimulate rental housing production (UNCHS, 1990f). These include rent decontrol, relaxing planning standards (to promote the building of extra rooms), and fiscal incentives to landlords (Malpezzi, 1990).

5. Services and infrastructure

[Back to Table of Contents]

35. The Vancouver Conference included infrastructure in its recommendations on the "comprehensive" or integrated approach to shelter, and made explicit the need to balance equity in access with cost-recovery in financing, a theme which is just as current today. "Infrastructure policy should be geared to achieve greater equity in the provision of services and utilities, access to places of work and recreational areas, as well as to minimize adverse environmental impact" (recommendation C.11). "Active use of pricing policies as a mechanism for improving equity in access" was recommended, and priority was to be given to the provision of safe drinking water and hygienic waste disposal, with "quantitative and qualitative targets" needing to be set for both (UNCHS, 1976, 23). Recommendations on transport, communication, health and education services were also included in the Vancouver Action Plan, with particular attention being given to the importance of decentralization and the need for "greater management at the community level" (recommendation C.15). The evolution of shelter policy and practice since 1976 has broadly followed these recommendations, though their impact varies markedly from one city to another, though there are plenty of successes to report particularly in community provision and maintenance of water and sanitation (see chapter III), there has not been a generalized increase in access to adequate basic services among the poorest sections of the urban population in the developing world. Devas and Rakodi (1993,9) estimate that between 30 per cent and 50 per cent of the third world's urban population still lack access to safe water and sanitation: and that an additional US$80 to 100 billion is required to close this gap by the end of the century. As a result, a large proportion of the urban poor (varying from one city to another) continues to depend on private water vendors, scavengers and others for whose services they often pay proportionately much more than the costs of a piped water or sanitation system (UNCHS, 1993c).

36. There has been much debate over the last 10 years about the relative roles of the public and private sectors in infrastructural development (UNCHS, 1993c). The capital costs of installing piped water systems and waterborne sewerage, roads and street lighting, electricity connections, clinics and schools, are too high for low-income residents to finance themselves (at least US$800 per household according to Arrossi and others. 1994), and require city-wide coordination if they are to be developed efficiently. The role of government in ensuring adequate coordination, quality of service and access among all income groups is clear. Elsewhere, however it is much more difficult to generalize, with the most efficient and equitable mode of provision varying greatly within and between cities according to the characteristics of the service in question, the income group concerned, and the local capacities of public and private sectors. This is a useful antidote to those who (for ideological reasons) see private provision as the norm in all circumstances. Where services are "divisible" (i.e., consumers can be charged separately according to usage), and where competition between service-providers is possible, forms of private provision may be both more efficient and more equitable than public provision, as long as the State retains an overview of quality and access and the ability to intervene, if necessary. This situation applies to urban transport and refuse collection in many cities. However, where monopolies are likely, private provision will only be effective if government has the capacity (and is willing to use it) to protect consumers from excessive charges and to promote affordability. Water, sewerage and electricity most often fall into this category. The impact of contracting out service-provision to private-sector companies depends on the existence of a genuinely competitive market and a government authority willing to enforce performance standards (Davey, 1993). Community provision of services (especially water and sanitation) has also proved successful on a small scale, though it must be carefully coordinated with developments elsewhere in the city if the efficiency of the service network as a whole is to be preserved. The achievements of the Orangi Pilot Project in Karachi, Pakistan in this respect are well known: by the end of 1988 almost 38,000 dwellings had been connected to a waterborne sewerage network, built by the community themselves at half the cost of the cheapest government or private contractor (see chapter III). The experience of the Urban Basic Services Program in India are somewhat similar (and are described in more detail in the final section of this chapter): since 1972. Water, sanitation and other services have been provided under the program to over 25 million people though problems of scale and sustainability remain (UNCHS, 1991f, 21).

37. Experience since 1976 has shown that the services have to be those poor people need if they are to be sustainable (otherwise people will not pay for them), an obvious conclusion but not one which is always heeded (UNCHS,1991f), and that gender considerations must be included explicitly in the planning and provision of infrastructure - otherwise, women especially may have difficulty in securing access to them (Moser and Peake, 1987). Cost recovery tends to be easier where there is a high degree of participation by consumers in the design and operation of services (and equity can be promoted through the use of cross-subsidies of various kinds), while sustainability and equity cannot be secured where municipal government is weak, since service provision is a complex process which requires effective supervision. For all these reasons, infrastructure has proved a fertile testing ground for a wide range of public-private partnerships in developing-country cities (UNCHS, 1993c, 104).

6. Wider goals

[Back to Table of Contents]

38. Before completing this brief survey of progress by sector since 1976, it is worth considering the impact of shelter policy and practice on wider developmental goals such as equity, sustainability and environmental conservation. The Vancouver Declaration and Action Plan made explicit reference to these goals, with commitment to geographical, social and gender equity a particularly strong theme. "Human settlements policies should aim to improve the condition of human settlements by promoting a more equitable distribution of the benefits of development, and by making such benefits and public services equally accessible to all" (recommendation A.4). Although later developments in shelter policy maintained a similar commitment to equity on paper, in practice questions have been raised about "who really benefits" from "enabling approaches." As John Turner (1992a, 17) has written, "In high-income industrialized countries, as well as those still undergoing rapid urbanization, supra-local, corporate interests appear to be gaining more from enabling policies than those they are meant to support. The worry here is that the balance between public responsibilities and private freedoms in the shelter process is in danger of shifting too far towards the latter, with the result that those with less "market power" in the city (the urban poor) are denied access to adequate shelter and services. These dangers are inherent when more reliance is placed on market mechanisms in highly unequal societies with imperfect markets and a weak State apparatus. Sub-standard housing, exploitative rents, dangerous jobs and insanitary services are just as likely as the thriving land, housing and employment markets which theory predicts a theme which is taken up in chapter IV. The implications of these conclusions for women are particularly severe (see chapter IV). The potential conflict in enabling shelter strategies between the need for liberalization ("freedom to build", private-sector incentives) and the need for regulation (to correct market imperfections and curb speculation) is a particular illustration of the wider dilemma facing all economies which aim to be both equitable and efficient. The successful implementation of the recommendations made at Vancouver depends on the ability of governments and citizens to find the right balance between these two sets of principles (UNCHS, 1991f). The example of the London Docklands development shows what can go wrong even in a well-endowed economy when private capital is encouraged to invest on extremely valuable land close to the city centre by using public incentives and relaxing planning regulations - a "monumental speculative development" which is not in the public interest (and certainly not of benefit to lower-income households who used to live there), and which needed to be baled out by public funds in 1992 (Devas and Rakodi, 1993, 97).

39. Sustainability has been another recurrent problem, with many innovations limited in scale and duration to particular projects, rather than making a lasting impact on the wider systems of land provision, housing-finance and infrastructure. Even in successful slum-upgrading schemes such as Hyderabad and Visakhapatnam in India, it has proved difficult to sustain community interest in maintaining services after the initial period of investment in housing and infrastructure is completed, particularly in "softer" areas such as health, education and social development (Asthana, 1994).

40. In conclusion, the record of the last 20 years has been very mixed, though there has been a fair degree of continuity in the application of shelter policies, and there are plenty of project-level successes to report (as well as one or two city-level success stories such as Curitiba in Brazil). One commentator goes so far as to suggest that "the record of the last forty years is a melancholy one, relieved only by the enormous capacities of people to provide for themselves, and to make their own cities, despite the obstacles so often put in their way by official agencies" (Harris, 1992, 23). This is probably too pessimistic (and academic) a conclusion, but a good dose of reality is no bad thing. At the most general level, it is true to say that a reasonable start has been made in many countries in initiating the enabling approach to shelter born at Vancouver and formalized in the Global Strategy for Shelter to the Year 2000 in 1988. The process of national shelter strategy formulation which has gathered pace over the last few years is particularly important in equipping governments and their citizens to face the challenges of "shelter for all", though the challenge of turning policy into action remains as demanding as ever, particularly in the most difficult areas such as land and housing-finance.

C. Assessment of capacity to make, monitor and implement national shelter strategies

[Back to Table of Contents]

41. Having assessed the development of national shelter strategies and the record on implementation over the last two decades, it is now time to consider the issue of national capacity to make, implement and monitor strategy and policy. In many ways this is the most crucial indicator of all, since if the capacity exists "on the ground" to develop and adapt policy and practice on a continuous basis then any shelter strategy will stand a good chance of meeting its objectives. If this capacity does not exist, no strategy will be successful because it will be incapable of adapting to changing circumstances. This is the real meaning of "sustainability" -sustaining people's ability to plan, monitor and evaluate their own development programs in line with local values and aspirations.

42. This chapter has already highlighted problems in implementing national shelter strategies as a key lesson of the last 10 years. The record here among countries, which have produced strategy documents, is very variable. Some (such as Jamaica, Uganda and Zimbabwe) have progressed from approval to implementation quite quickly; others (such as Congo, Kenya and Mali) have not implemented their strategies but have used them as a foundation for further policy development (Struyk, 1990, 317; Wakely and others, 1992. 4). Detailed plans for implementation (timetables, roles and responsibilities, resource implications and so on) tend to be the weakest section in most national shelter strategies, even in Zimbabwe where the functions of and linkages between different ministries and agencies have not been adequately spelt out (UNCHS, 1990e, l5). In some cases (such as Barbados), implementation has been thrown off course by changes in political regime or donor support, showing the crucial importance of continuity in both senses. Weaknesses in implementation are, of course, not a phenomenon of the last few years alone; they go right back to the Vancouver Conference and are a common failing of all shelter policy statements in the last two decades, including the Global Strategy for Shelter (UNCHS, 1991f). The review of building bye-laws in Kenya undertaken between 1978 and 1980 was never implemented, just as the later National Shelter Strategy was used only as a basis for further debate.

43. There are a number of factors which seem to underlie poor implementation. First, and most obvious, is the simple lack of resources available to overstretched municipal administrations for research, policy development, monitoring and so on. Inadequate property-registration and land-information systems undermine the municipal revenue base and cost recovery from services is often very poor. Linked to resource issues are weaknesses in the skills base of municipal administrations to undertake the complex and demanding tasks essential to the enabling approach - achieving the right balance between the "invisible hand" of the market and the "top-down hand" of the centralized State in the way described earlier in this chapter is extremely difficult to accomplish in practice, especially since the balance - the "improving hand" as Devas and Rakodi (1993) put it - changes over time and from one activity to another. Cities are also arenas for choice, conflict and negotiation between competing interests, and the municipal authorities must be able to manage these negotiations effectively, a theme which is taken up in chapter IV.

44. In addition, over-centralization remains a problem in many countries, where there is a reluctance to devolve authority down to the municipal level, never mind to the level of the neighborhood. In Argentina, for example, centralization remains the norm despite many attempts at urban reform, and municipal governments have been unable to carry out even the one task they have been given - the collection of property rates (Herzer, 1992). Confusion over responsibility for implementation among different ministries and departments is also common, something which has not been helped by the rapid pace of institutional reform during the process of strategy formulation in many countries. For example, between 1979 and 1989, responsibility for urban shelter policy changed hands three times in Nicaragua, and twice in Indonesia and the Philippines (Wakely and others, 1992). In Zimbabwe, responsibility for policy decisions is divided among four separate ministries, and for implementation, between two levels of government; the municipalities are responsible in theory for project implementation and service-provision, though in practice they lack the capacity and resources to do this effectively. As a result, "the principles of enablement have not penetrated very deep" (Wakely and others, 1992, 38-42). Centralization has also proved a problem in Indonesia, with little autonomy being granted to local authorities and severe bureaucratic delays in the regional pilot project (Central Java) chosen for strategy formulation (Wakely and others, 1992, 57). These weaknesses and how they might be addressed are explored in more detail in chapter in.

45. The Vancouver Action Plan recognized all these issues, and called in its recommendations for improved settlement planning and monitoring within the constraints imposed by local resources. "Settlement planning should be based on realistic assessment and management of the resources actually and potentially available for development" (recommendation B.3). This spirit of pragmatism, which reappears throughout the document, is an essential counterweight to the more visionary parts of the Vancouver Declaration, and is equally valuable today. Proper implementation of a small number of key policy measures under severe resource constraints is preferable to a failure to implement a wider range of unrealistic recommendations. It was also recommended that "planning at all scales must be a continuous process requiring co-ordination, monitoring, evaluation and review, both for different levels and functions as well as from the people affected" (B.16). Indeed, the need for wide public participation in the design, implementation and monitoring of shelter policies comes through very strongly in the Vancouver Action Plan, a clear precursor of the Global Strategy for Shelter and the wider support for "democratization" highlighter: in chapter 1. "Public participation should be an indispensable element in human settlements, especially in planning strategies and in their formulation, implementation and management: it should influence all levels of government in the decision-making process to further the political, social and economic growth of human settlements" (recommendation E.1). Public participation was to be "maximized", with special mention being made of groups that "traditionally have not participated in the decision-making process", i.e. tenants, women, young people, disabled people and the elderly (UNCHS, 1976, 36). Few shelter policies in later years have matched up to these goals, though all have tried. Many national shelter strategies have made real efforts to involve a wide range of actors and interests in design and discussion, and there are encouraging signs elsewhere too. For example, in Columbia, representatives from consumers can now sit on the boards of public utility companies to participate in quality control, and the development of national shelter policies in India over the last five years has included (if not always responded to) a wide range of shelter NGOs and popular movements including the National Federation of Slumdwellers (UNCHS, 1991f, 70). Other examples of participation in shelter development are described in chapter 111, while chapter IV tries to make a more systematic evaluation of the record thus far in involving marginalized groups in decision-making over shelter, especially women.

D. International support to national shelter strategies

[Back to Table of Contents]

46. As described above, encouraging progress has been made in many countries in developing strategies to implement the recommendations of the Vancouver Conference and the Global Strategy for Shelter to the Year 2000. However, given the weak resource base of most developing countries and the demanding agenda for change, which the enabling approach represents, international support to national action also has an important role to play. Two areas of international support have been particularly influential: financial resource flows, and support to capacity-building.

1. Resource flows

[Back to Table of Contents]

47. Agenda 21 estimates that the provision of "adequate shelter for the urban and rural poor by the year 2000" will cost around US$75 billion annually, and that at least US$ 10 billion of this amount will have to come from external assistance (UNCED, 1992, 7.10). These sums are most unlikely to materialize, certainly if one projects past and current figures for external aid to human settlements into the future. Between 1980 and 1991, less than 1 per cent of total official development assistance from members of the Development Assistance Committee of the Organisation for Economic Cooperation and Development, and less than 3 per cent of non-concessional loans, went to human settlements development (Satterthwaite, 1994, 28). These figures vary considerably from year to year and donor to donor, and there are problems in defining exactly what "human settlements" and "shelter" refers to. The figures cited above include housing, infrastructure, services and support to urban management. Nevertheless, even accounting for these variations, the pattern of external assistance shows that resource flows have always been low (certainly since 1976). There has been some increase among some donors to some areas of human settlements during the last few years, but the overall percentages remain much lower than the comparable figures for other sectors.

48. There have, however, been significant changes overtime in the pattern of these resource flows, as shelter policy and priorities have evolved. In the 1970s, external aid went mostly to housing projects - sites-and-services schemes and upgrading programs. For example, between 1972 and 1976, the World Bank projected lending of approximately US$700 million for this purpose (World Bank, 1972, 7). The actual figure for the period 1972 to 1981 was US$2,014 million to 62 projects, projected to reach US$4 billion in the following five years (World Bank, 1983, 49). During 1983, the Bank began to move away from support to upgrading and sites-and-services projects and towards lending to housing-finance as a component of wider economic adjustment programs; between 1980 and 1991 almost 50 per cent of World Bank non-concessional loans to shelter were for this purpose (Satterthwaite, 1994, 3). By 1989 the Bank had entered a third phase of lending to shelter development, concentrating this time on support to urban management, capacity-building and social safety-nets in urban areas (Pugh, 1994). These changes came in reaction to wider shifts in housing policy (particularly the Global Strategy for Shelter with its emphasis on the enabling environment) and to the higher-than-anticipated social costs of the adjustment process in urban areas. By 1991, World Bank loans to "urban development" had reached US$1,25 billion, representing some 5,5 per cent of total lending for that year, or 10,9 per cent if "water and sanitation" loans are added to the total (Hannah, 1991). This was some way below the peak of Bank lending of 9 per cent (in 1988), but an increase over the previous year's figure. Lending for the period 1991 to 1995 is projected to reach around 9 per cent of total World Bank lending, or some US$ 15 billion over a five-year period. Grants and loans from other multilateral institutions and from bilateral donors, such as the British Overseas Development Administration (ODA) and Swedish International Development Authority (SIDA), have followed the same pattern, though in the case of the bilateral agencies funding still tends to be tied more closely to particular integrated urban development programs, such as slum upgrading in India (for ODA) and (for SIDA) in Costa Rica (via FURPROVI, a local NGO). Both these programs have achieved significant results, and are being expanded to other cities, though problems of sustainability remain (ODA, 1990). Investment in infrastructure and services as increasing across the board (especially in water and sanitation, primary health care and social safety-nets), as are loans and grants to urban management, while all donor agencies are paying more attention to gender-aware approaches to shelter and to programmes which recognize the particular needs of women (Satterthwaite, 1994, 7).

2. Support to urban management and capacity-building

[Back to Table of Contents]

49. In the late-1980s there was a shift in donor-agency priorities in the urban sector away from shelter projects and housing finance, towards support for capacity-building within the framework of the "enabling approach". The critical importance of effective urban management in developing and maintaining the "enabling" legal, regulatory and financial environment led donors to re-focus their attention on the development of the human skills and capacities required to make and evaluate policy.

50. Support to the development of national shelter strategies. After the adoption of the Global Strategy for Shelter to the Year 2000 in 1988, a number of donors (UNCHS, the United Nations Development Programme (UNDP), the Finnish International Development Agency (FINNIDA), and the Danish International Development Agency (DANIDA) agreed to support the process of national shelter strategy formulation. The Government of the Netherlands provided additional finance for a program of activities concerned with the role of women in shelter strategies (UNCHS, 1991b; 1991e). Ten sub-regional seminars were held in 1990 and 1991 to exchange ideas and experiences on strategy development and identify suitable indicators for monitoring, and UNCHS has also provided short-term advisory services to countries (including Costa Rica, Malaysia and Zimbabwe) which have requested them. A recent evaluation of that part of the support program funded by FINNIDA concluded positively: the program should continue and build on the solid start that had been made (DPU, 1991). In a separate but related move, DANIDA has been funding a training program aimed at promoting community participation in shelter development in Bolivia, Sri Lanka and Zambia since 1984. With a focus on building capacity to undertake "process" rather than "blueprint" approaches to settlement planning, the DANIDA program has provided important support to institutional development in relation to the enabling approach (UNCHS, 1992e). A similar initiative, the Community Management Program aims to re-orient local government policy and practice in Costa Rica, Ecuador, Ghana and Uganda (UNCHS, 1992d).

51. The Urban Management Program (UMP). Funded by UNCHS, UNDP and the World Bank, and with additional contributions from European bilateral donors. UMP is the most significant capacity-building initiative currently in operation. It was the logical outgrowth of rising donor interest in management issues in the late 1980s and early 1990s, exemplified by the major international policy statements on shelter issued around this time: the Global Strategy for Shelter to the Year 2000 (1988): UNDP's 1991 strategy paper; Cities, People and Poverty, and two papers from the World Bank on Urban Policy and Economic Development (1991) and Housing: Enabling Markets to Work (1992). The common thread in these policy statements was the need to strengthen urban management so that government was better able to play the role of enabler in promoting private-sector activity while protecting the interests of the poor. Indeed, so much attention has been given to management that more fundamental issues of equity and sustainability tend to be relegated to a secondary position. There is a danger that "management" will be seen as a new panacea for deep-seated structural problems in the urban economy and housing market. In the World Bank papers, a second clear theme is the heavy emphasis placed on urban productivity (and removing constraints to raising it), along with the need to link shelter policy more explicitly to macro-economic management. The potential and potential problems, of this agenda are explored in chapters III and IV.

52. "Urban management" covers the areas of governance, development policy and investment coordination, the management of assets and services, regulatory frameworks, collective functions (such as revenue collection), and public/private partnerships (Clarke, 1991). The first phase of UMP (1986-1991) concentrated on building capacity in the following areas: land management, infrastructure, and municipal finance, with a central focus on making markets work more efficiently via the loosening of government controls. Environmental management was added to this fist in 1990, and urban poverty-alleviation a year later at the start of the UMP's second phase (a surprisingly late addition, perhaps). In 1992, regional units in all four developing regions were established to increase support to capacity-building. The second phase of UMP re-focused support on facilitating private economic activity, strengthening capacity to involve NGOs and CBOs in policy dialogue, forward and strategic planning, and service-delivery (Harris, 1992, 99). From the beginning, UMP activities have been established only at the request of the host government (as a sign of commitment), followed by a "consultative process" in which public, private and third-sector representatives together identify key issues for action (UNCHS, 1992f). There have been 42 such consultations to date. The special position and potential of city mayors has also been increasingly recognized through a number of initiatives aimed at facilitating networks and discussions among mayors, and supporting the development of municipal action plans backed by strong executive leadership (for example, the meeting of over 100 mayors sponsored by UNDP and UNCHS which took place in New York in August 1994). As the latest annual report for UMP states, this is a "demand-driven, operationally-decentralized" initiative, which makes maximum use of local resources. By the end of the second phase in 1996, UMP aims to have created a "South-South technical assistance network that can address critical management issues" (UNCHS/World Bank/UNDP, 1994, 6). These are all excellent sentiments in principal, though the view of UMP from many shelter NGOs is less glowing (fearing a widening gap between rhetoric and reality), and, as the Swiss Development Agency has commented recently, there is a danger that the program will "fall into the trap of rapidly accumulating highly visible, quantitative results to the neglect of more substantive and consistent - but gradual - qualitative effects" (UNCHS/World Bank/UNDP, 1994, 37).

53. The Housing Indicators Program. This is based on a model in which housing outcomes (and beyond them, socio-economic outcomes) are determined by the interaction of supply and demand factors. The resulting indicators are desegregated by groups of countries (according to gross national product), but not by gender or neighborhood. They include the house-price (and rent)-to-income ratio, floor area per person, infrastructural expenditure per capita, water connections and "homelessness" (UNCHS/World Bank, 1992). The program builds on an earlier initiative (the City Data Program, launched in 1991) which aims to strengthen capacity to use a global urban data-collection and information system, piloted in Kenya (UNCHS, 1992e). It is too soon to judge the effectiveness of these initiatives in strengthening capacity to monitor shelter developments.

54. Urban research in the developing world. This is a program supported by the Ford Foundation, the World Bank, UNDP and UNCHS, and implemented by the University of Toronto Centre for Urban and Community Studies. This initiative has surveyed the output and pattern of urban research across the world over the last 20 years, concluding that the volume of research has declined in some regions (such as sub-Saharan Africa) but not in others (such as Asia and Latin America). The program has recommended increased international support for networks of local researchers in the urban areas of developing countries, both to improve the quality and policy relevance of research outputs and to build local capacity to undertake research (Siren, 1994).

55. The United Nations Children's Fund (UNICEF) Urban Basic Services Program (UBS). Although not restricted to capacity-building, UBS is somewhat similar to UMP in "promoting empowerment strategies via partnerships between governments, NGOs/CBOs and international donors - in this case UNICEF (1993a, 1). Since the start of the Program in the late-1970s, it has spread to over 168 towns and cities in India and to more than 40 countries worldwide. At the height of the Program, 20 per cent of total UNICEF assistance was directed to urban areas. The principles of the Program are well-known: an integrated approach to urban problems founded on community participation through networks of neighborhood committees, local volunteers and para-professionals, and emphasizing partnerships between municipal governments. CBOs and NGOs particularly in the areas of water and sanitation, primary health care and nutrition (Iskander, 1992). In concrete terms, much has been achieved by UBS over the last 20 years: for example, a reduction in infant mortality in Addis Ababa from 130 per 1.000 live births in 1983 to 50 in 1990 (Iskander, 1992, 5), expansion from four settlements to over 90,000 urban residents in Guatemala since 1987 (UNICEF, 1991): and significant improvements in immunization coverage and access to safe drinking water in many other cities in Bangladesh, India, the Philippines, Turkey and elsewhere (UNICEF, 1992a, 1992b, 1993b). At the level of capacity-building, social mobilization and advocacy, the program has helped many participating governments to "shift from sectoral, agency-based, vertically-administered programs, to city-based, multi-agency, intersectoral approaches to planning and programming" (UNICEF, 1992a, 39). This is a significant achievement: the UBS approach has shown that significant improvements in urban health and social indicators can be obtained by involving poor people fully in the planning and implementation of their own development efforts, in partnership with local government and NGOs.

56. However, UBS has (along with other similar assisted urban community development initiatives funded heavily by external donors) suffered from problems of scale and sustainability, lack of continuity in urban management and a tendency toward "mechanical planning which calls for the achievement of specific targets within too short a time frame" (Cousins and de la Soudiere, 1992). The approach is very time- and staff-intensive, and in many cities over-dependent on UNICEF for funding and technical support (UNICEF, 1992a). An evaluation undertaken in 1993 concluded that the UBS "has not been applied vigorously or consistently and has suffered from a lack of analysis regarding impact and costing, a realignment is timely and necessary" (UNICEF, 1993, 1). This realignment will take the form of "urban development with a human face:" economic policies favoring human development: partnerships for poverty-reduction and meeting the decade goals for children (community-based service provision), primary environmental care and action for "children in difficult circumstances" (UNICEF, 1993, 1).

57. Other relevant programs. There are many other current initiatives from within the international donor community timed at one or other aspect of urban management and/or capacity, many of which overlap with each other (and especially with UMP proper). They include: UNDP's Training Programme and Capacity-Building Strategy for Municipal Management, which has four sub-programs dealing with gender-aware planning, the training of city managers, training-of-trainers for urban finance and management for the Arab States, and a network of urban training institutions in Latin America. The Government of the Netherlands also supports (through UNDP and UNCHS) a Training Program on Municipal Finance and Management for South Asia (UNCHS, 1992e). Added to these are the Settlement Infrastructure and Environment Programs of UNCHS which provides support to the development of integrated approaches to service-provision: the Sustainable Cities Program launched in 1990 to improve environmental planning and management capacity (and formally the operational arm of UMP): and the Urban Poverty Partnership which links together the poverty-alleviation strand of UMP with similar interests in the International Labor Organization and United Nations Volunteer Program (UNCHS/World Bank/UNDP, 1994).

58. The point of citing all these programs is not to assess their impact (since few evaluations are available), but to stress the huge amount of capacity-building that is currently being supported by international donor agencies. Indeed, there are so many programs, that their real impact on local capacity must be questionable. A common complaint among under-resourced and overstretched governments in developing countries is that the sheer weight and variety of donor-supported initiatives (often with different time-scales, reporting schedules and monitoring guidelines) threatens to overwhelm the capacity of the system and the personnel within it whose capacity is supposed to be being built (La Fond, 1994). A more coordinated approach (as in UMP) is bound to pay dividends. As the formal body set up after the Vancouver Conference to coordinate shelter-related activities in the United Nations system, UNCHS obviously plays a central role in this process of increased cooperation.


A. Bottlenecks

[Back to Table of Contents]

59. The supply side. The supply of housing and related services is influenced by many factors, including land availability and prices, infrastructure, access to credit for construction and land assembly, the regulatory framework, the organization of the building industry, and the degree of speculation and politicization which exists in the housing market (which is often much greater than anticipated). Real markets are permeated by power relations of various kinds - class, gender, culture and politics, which are exacerbated when, as in the case of land and housing, commodities take on great significance for investment, speculation and patronage. The weakness of property-registration and information systems and the historical rigidities of housing supply combine to make markets even more imperfect. As the World Bank points out, it is the poor who are most disadvantaged by poorly-functioning housing markets. Removing constraints on supply in the rental housing market is particularly important, since that is where most poor people in developing-country cities live. Progress over the last 20 years in this respect has been extremely disappointing. Innovatory measures to stimulate the supply of affordable land tend to be very small in scale - land sharing in Bangkok, for example, has benefited only a few thousand families, and land readjustment less than 10,000 hectares in Indian cities. Experiments in land-market regulation, such as the ULCRA legislation in India, have usually rendered supply constraints even worse. It has proved extremely difficult to strike the right balance between liberalization and regulation in the land and housing markets through appropriate legislation and other measures.

60. The demand side. Housing demand is influenced by the level and distribution of income among the urban population (i.e. urban poverty), the availability and costs of housing-finance, the nature of property rights (or secure claims), and in the use of subsidies. According to the Head of the Urban Development Division of the World Bank, "By 1990 urban poverty had not really been alleviated very much". As a result, the World Bank, followed quickly by most other multilateral and bilateral donor agencies, refocused its policy toward poverty-reduction (including urban poverty) on a three-pronged strategy aimed at raising urban productivity: labour-intensive growth (for example, construction activity and private provision of some services using labor-intensive methods): increasing investment in human capital and the social sectors: and social policy measures (including modified social safety-nets) to mitigate the impact of poverty and economic adjustment on the most vulnerable groups. The impact of these strategies on urban poverty is debatable, though it is already clear that women often bear the brunt of economic changes in the contemporary city, with increasing workloads and responsibilities, but declining family and government support.

61. At root, poverty reduction is a political problem and as such it is much less amenable to external policy prescriptions, Economic and political interest groups which block attempts to remove constraints in the land and housing markets are also likely to frustrate the implementation of the urban productivity agenda, since they have little to gain from the market empowerment of the poor.

62. Deficiencies in urban management. Significant problems have been encountered in national action on shelter on both the supply and demand sides of the housing market. Equally important is the ability of government (especially at the municipal level) to manage markets in such a way as to promote both efficiency and equity - the essence of the "enabling approach", but something which is exceptionally difficult to accomplish in practice, especially where resources are scarce. Despite considerable support from external agencies and host governments, deficiencies in urban management retrain very widespread. Absolute shortages of resources acre obviously important in this respect, but urban managers often find it difficult to spend resources efficiently even when they have them. Further problems arise as a result of the fragmentation of urban government between functions, levels and departments, and the lack of coordination which results, In addition, the necessary reform of urban governance called for in the "enabling approach" to shelter inevitably runs up against the problems of inertia and resistance to change which exist in all bureaucracies. "The old agenda has amazing, persistence" as one critic puts it, with urban managers still caught up in marry cities in outmoded patterns of working - over-centralized, unresponsive to consumers and citizens, and inefficient. The politicization of urban administrations is common, especially since land housing and contracts are extremely valuable commodities in both economic and political terms.

63. Problems with scale, sustainability and reach. A consistent theme in evaluations of national action in the field of human settlements since 1976 has been that, whereas problems are widespread, successes have usually been small in scale, limited in reach, and transient, often relying on heavy injections of external support for their survival. The choice between large-scale coverage in programs and reforms (which risks the exclusion of the most vulnerable groups) and in-depth action (which is difficult to sustain) has been a constant dilemma for policy-makers. It has proved universally difficult to achieve scale, sustainability and poverty-reach all at the same time, In poorly-resourced economies, this is inevitable. In part, the difficulties encountered with going-to-scale and achieving sustainable change are the result of an over-concentration on special projects or programs, which have failed to alter systems, structures and resource flows at a more fundamental level. Attempts to scale up have often failed because they have not succeeded in changing (and may even have ignored) the institutions, linkages and practices which are mainly responsible for determining the outcome of the shelter process. A failure to confront vested interests and power relations has also been important. Too little attention has been paid to capacity-building among local institutions, the development of linkages between different aspects of the program (so that they support each-other), and the careful measurement of costs and benefits, which is essential to test the sustainability of each intervention. If shelter programs are sustainable, they may exclude the poorest people in developing-country cities who cannot afford to pay user charges, find time to participate in community or political activities, and/or simply lack the "market power" to take advantage of new incentives and opportunities.

B. Successful approaches

[Back to Table of Contents]

64. The supply side. A number of important lessons emerge from the innovations reviewed in the report. First, sustained government intervention is required in the land market if significant amounts of land are to be made available at prices the poor can afford: reliance on market mechanisms alone (i.e., the formal market) will not be enough. Secondly, support to entrepreneurs and builders in illegal or quasi-legal land markets will pay dividends - this shifts the focus of government action to the provision of infrastructure and services, lowering planning and building standards (though not abolishing them completely), simplifying land-registration and development procedures, and granting security of tenure (though not necessarily individual freeholds). Some form of leasehold arrangement may be the best way of promoting household investment while curbing speculation. Thirdly, all parties in partnership arrangements intended to increase the supply of land must receive some material benefit if the partnership is to work. Fourthly, strong community organizations, intermediary NGOs, and municipal governments are required to oversee these processes and provide adequate support and protection to the urban poor, especially when markets are liberalized.

65. The demand side. The central importance of securing poor people's claim to property tights (especially among women as individuals rather than as dependants) has already been mentioned, along with the use of various forms of leasehold to reduce the risks of upward filtering and speculation in low-income land and housing markets. The demand for shelter among the urban poor can also be stimulated by encouraging the use of labor-intensive methods of construction and service provision, and by strengthening links between shelter development and economic development in a variety of ways. On a wider level, demand can be increased by the kind of poverty-reduction strategies described above - the three-pronged approach recommended by the World Bank and others which focuses on support to labor-intensive growth, investment in human development and social infrastructure, and social policy/social safety-nets.

66. Successes in urban management. Urban management in "cities in crisis" is a monumentally demanding task. When laid out in policy documents, the elements of effective management seem deceptively straightforward. However, in reality such lists underestimate the scale, nature and complexity of the demands which are placed on municipal governments in developing countries, for, in addition to the obvious technical functions, governments must also play the leading role in mediating between the interests of people and capital, public and private sectors, and use and exchange value in land and housing. Governments must endeavor to find the right balance between intervention and liberalization in the shelter process in order to promote private-sector activity while protecting the interests of those with less market power and ensuring the efficient and equitable development of the city as a whole. They must plan and implement policy in the context of continual uncertainty, resource shortages, and discontinuities in personnel and political leadership, and manage land and housing markets, which are highly imperfect and liable to manipulation and politicization. Finally, there will be pressures from central government to capture the valuable revenue base of cities and maintain centralized control over decision-making. What successes can be identified in the field of urban management over the last 20 years, when faced by such a daunting agenda?

67. First, decentralization is essential so that municipal governments are genuinely empowered to raise and allocate resources, brought into closer contact with realities on the ground, and held accountable by local citizens for their performance. A second lesson of experience concerns coordination and integration of functions. Effective urban management requires the close integration of planning and development responsibilities with service provision, urban transport, environmental health and other functions. The next area concerns the reform of municipal bureaucracies, rendering them more responsive and more efficient and less vulnerable to corruption. This is always a difficult task to accomplish, given the historical inertia and inefficiency which tends to characterize bureaucracies in developing countries. Nevertheless, experience shows that it can be done. A key factor underlying efficient urban government is the close involvement of citizens and their organizations in decision-making, policy debates, program implementation and monitoring. Broadening the base of decision-making in the urban arena is essential if lasting reforms are to be achieved in urban governance. Strong, representative municipal government is the only effective counterweight to the vested interests which will otherwise block and frustrate attempts at reform. A move away from centralized bureaucratic systems and towards more open government, increasing consultation with the users of services and systems (and their involvement in managing these systems), and the formal recognition of the role of community organizations and NGOs in urban development, are common features of successful municipal governance. However, these approaches can only flourish if the political environment is conducive to CBO/NGO-government collaboration, and this takes a conscious effort on all sides. It is neither a simple nor a straightforward task. Experience shows that such partnerships work best when government has a positive social agenda and NGOs are strong and independent actors: mutual respect (rather than simple sub-contracting by government), an acceptance of NGO autonomy, and the deliberate fostering of a pluralistic political culture are all important. The legal framework governing NGO activity must be positive, as should the fiscal regime (taxation etc). If the political climate is unfavorable to social mobilization, and/or if CBOs and NGOs are seen only as implementers of projects financed by government and external agencies, then positive government-NGO partnerships will be difficult to develop.

68. In summary, experience over the last two decades has shown that the basic principles of successful urban management are as follows: an-organizational structure with well-defined and transparent functions: strong accountability mechanisms and linkages between authority and performance: continuous monitoring of service-delivery and program implementation within a clear framework of incentives, rewards and penalties; continuity and stability of personnel; qualified staff and good-quality training: a political climate which encourages and institutionalizes citizen participation, pluralism and the activities of NGOs and CBOs; strong commitment to genuine public-private partnerships; and real decentralization. Rather than copying the management models of the industrialized world, it is essential to adopt the fundamental features of good management-efficiency, transparency and accountability -and apply them in a specific urban context. Autonomous management of sufficient resources at the municipal level is critical, as is the existence of mechanisms and systems to broaden the base of citizen involvement in deciding how these resources are to be spent and accounted for.

69. Successes in scaling-up and sustainability "Multiplicative and diffusive" approaches to scaling-up are often more effective than attempts at program growth or simple replication, since they can achieve a wider coverage without a decline in quality or reach or an increase in bureaucracy and costs. "Multiplicative" approaches focus on the deliberate spread of successful approaches via interaction between NGOs and government, training, awareness-raising, networking, legal and institutional reform, and policy advocacy. "Diffusive" approaches achieve the same goals, but as a result of the informal and non-deliberate spread of new ideas and successful experiences. Even if the scale of resources devoted to successful shelter programs was increased several-fold, their impact would be limited without changes in the underlying approach adopted.

70. The special needs of vulnerable groups. There has been considerable progress over the last 20 years in incorporating into planning and decision-making an awareness of the particular needs of women in shelter development, though the impact of this process is debatable. Commentators have concluded that increasing gender-awareness training and urban research with a gender component "has had little influence on either mainstream urban researchers or on policy-makers" (Moser and Peake, 1987; Dandekar, 1993). Most of the more successful examples of shelter programs and approaches which have met the particular needs of women in cities come from CBOs and NGOs, and the difficulty of institutionalizing these successes on a larger scale (in government programs) has proved a major obstacle. Much less progress has been made in meeting the needs of other vulnerable groups, such as children and the elderly.


[Back to Table of Contents]

71. The policy-implementation gap. As this summary has outlined, there has been a great deal of progress over the last 20 years in the evolution of shelter policy. Despite differences of emphasis between those who advocate more liberalization and those who believe in continued intervention, in land and housing markets, there is a clear line of development from the recommendations of the Vancouver Conference in 1976 to the "enabling approach" of today. In most cases it seems as though it is known what should and should not be done to improve low-income shelter, the problem is that this advice is not followed. There is a serious gap between policy and implementation, and between rhetoric and reality, particularly in key areas such as land and housing finance. What are the factors, which lie behind this problem, and what can de done to address them? Unless answers can he found to these questions, no policy, training, technical support or even increase in financial resources will have much lasting impact. All bureaucracies suffer from inertia and interest-group manipulation. Strong and continuous pressure for reform and results, exerted from the bottom upwards, acts as an essential counterweight to these characteristics. This is why strong accountability mechanisms (to users, voters and consumers) and "good urban governance" are so important to shelter development: unless there is a strong and effective constituency in the city which can pressurize the municipal authorities to implement difficult measures which are often unpopular (especially with elites), key bottlenecks in the land and housing markets will remain. Equally important is command at the municipal level over the resources required to govern fast-growing cities effectively. Resource shortages are inevitable in low-income countries, particularly when government expenditure is being retrenched under economic adjustment and/or transition. A third key area for action in trying to close the "policy-implementation gap" is to improve the functioning of the municipal authorities through new training and incentive structures. Ensuring good relations between central government and the city authorities is also vital. Whatever measures are adopted, closing the "policy-implementation gap" also depends on contextual factors, and the ability to identify and take advantage of "moments of transition" - periods or circumstances during which bureaucratic reform is more likely to be effective.

72. Building constituencies for good urban governance. Technical measures such as improved incentives- to performance and increased municipal control over revenue will help in closing the "policy-implementation gap", but by themselves they will not be enough, for the simple reason that urban shelter is not primarily a technical issue, but a political one. The reality of shelter-provision in most developing-country cities is one of large-scale inequality: land and housing markets are heavily constrained and highly imperfect: powerful landholding and financial interests derive significant material land speculative benefits from maintaining these constraints (such as land held vacant or capital invested for short-term gain), and are unlikely to give up such practices unless they are forced to. Without strong and continuous pressure exerted on local government by urban citizens there will never be a sufficient "critical mass" to induce change, oppose vested interests, and maintain the pace of reform once instituted. Since urban citizens will themselves have different interests, this process usually operates through coalitions and alliances for reform in particular areas and at particular times. If the urban poor are to benefit from reform, it is particularly important that they form, and are encouraged to form, powerful coalitions of their own. Achieving a balance between different needs and goals requires a representative political and administrative system through which the views and priorities of citizens can influence policies and actions both within the neighbourhood where they live and at city level. This is what is meant by the term "good urban governance": systems based on transparency, accountability and the existence and enforcement of clear and agreed rules and regulations.

73. The public-private mix. The most controversial and ideologically-charged subject in current debates over shelter policy and practice concerns the balance to be struck between "public" and "private" roles and responsibilities in the shelter process: the balance between "intervention" and "liberalization" in land, housing and financial markets: the "mix" between State, citizen, and business. While markets are more effective in organizing production and exchange, governments are more effective at defining the framework and "entitlements" within which markets must operate if "imperfections" (such as monopoly control over the provision of a service) and "externalities" (such as environmental pollution) are to be regulated satisfactorily. This is more-or-less agreed in theory. However, in practice, there continues to be enormous debate over the roles of State, market and people in shelter development. For one thing, service provision may not be the prime goal of urban governance: "local government is not just one possible provider which outsiders may rate against other possible providers: it is itself the institution which should be making these decisions, even if it makes the 'wrong' ones" (Batley, 1992, 3). Secondly, the reality of public-private partnership is much more complex and less predictable than the theory suggests. The balance to be struck between public and private provision will vary from one city to another, and from one service to another. The potential trade-offs between cost, quality and access (coverage) are also complex, and this makes generalization very dangerous. Empirical evidence shows that "both shifts from public to private provision and shifts from private to public provision have been shown to lead to efficiency improvements" (Batley, 1992, 4, 58), and that "the presumption that to privatize is to achieve higher levels of efficiency is given only selective support" (ibid.). Thirdly, the management capacity and resource base required to manage private-sector provision effectively is often lacking in developing-country cities. Strong public intervention is essential if these dimensions of marketization are to be tackled effectively so that poor people are able to participate in land and housing markets, and have access to non-market solutions where appropriate. "The strongest and socially most-beneficial interventions are likely to involve land and finance" (Strassman, 1994, 2). Markets work best with strong public support and guidance.

74. Dealing with diversity. The discussion of the changing intellectual orthodoxies described in chapter I emphasized a move over the last 20 years away from rigid models and standardization in development theory, policy and action, towards a "learning process approach" which recognizes the fundamental diversity and uncertainty of the contemporary world. This is a major advance, though it remains to be consolidated in many areas of development policy, especially among economists. The reality of shelter development in developing-country cities since 1976 certainly bears out the value of such an approach, with levels of diversity and unpredictability much higher than was commonly believed to be the case 20 yeas ago. Research and experience have shown that the housing preferences and needs of poor people in cities vary greatly according to family characteristics, incomes and occupations, gender, location, tenure, form and structure and other characteristics, and diversity is a major theme running right throughout the body of the report - in land and housing markets, public-private roles and responsibilities, the impact of different policy interventions, and trends in shelter quality and affordability. Generalizing about trends over time and space in real land prices has proved to be a dangerous activity, Each new intervention in land and housing markets is likely to create new (and often unforeseen) problems and opportunities, an increasingly complex web of interests and conflicts emerges in the city which makes it even more difficult to predict what is going to happen as a result of a particular policy measure or action.

75. If it is true that diversity is the rule, the next step cannot be to recommend particular models, modalities and arrangements in the public-private mix, land and housing markets and other areas of shelter provision, since these prescriptions are bound to be inappropriate in some, if not most, cases. This means that structures of urban governance, decision-making, policy-formulation, and monitoring must be capable of adapting to and dealing with diversity and uncertainty. Not only must structures and systems be adapted but also (and perhaps more importantly), the attitudes of urban planners and managers have to change, so that those involved become more adept and confident in handling the complexities and uncertainties of real and ever-changing land and housing markets, competing interests, shifting alliances, and new solutions. There is a need to shift attitudes among policy-makers from "absolute control or prediction" to "arriving at better decisions", or as J, M, Keynes once said, "It is better to be roughly right rather than precisely wrong". In practice, this means that urban managers need to focus on increasing people's choices in shelter development and expanding their opportunities, while leaving the details of "micro-management" to people themselves. "It is not specific practices which should be transferred, but a capacity to identify and analyze problems and practices in any particular case and to build solutions on what exists and what works." Experience shows that there are a number of steps, which can be taken to promote these kinds of attitudes and structures. An emphasis on the basic principles of effective urban management is much more important than "copying the management models of the industrialized world": transparency, accountability and efficiency applied differently in practice according to local context. Also crucial are the capacity to promote conflict-resolution between competing interests: develop a common civic identity and consciousness which can override these differences, in order to build widespread public support for urban reform in the city as a whole: facilitate linkages and connections between different groups: and strengthen the institutions of "civil society" to act as a counterweight to State or commercial power. All in all, urban management must move from "planning the city" to a "city that plans". The performance of current urban management and capacity-building programs in relation to these goals is a matter of some debate.

76. Holistic approaches to shelter development. Over the last 20 years, there has been an increasing recognition of the linkages which exist between successful shelter development and wider economic, political, environmental, social and cultural factors, Shelter planning in isolation from these factors is unlikely to be effective. This concern for a more "holistic" or integrated approach to shelter development was present in the Vancouver Declaration and Action Plan, but in a fairly primitive firm. A great deal more is now known about how to put the holistic approach into practice, though (as in all areas of shelter development), practice continues to be deficient. There are two levels at which such integration is essential. First, at the macro-policy level, research and experience have shown that shelter development is intimately linked to national and local economic growth and development. Secondly, at the level of urban management, integration is equally important. Experience shows that the strongest partnerships between public and private sectors (including NGOs and CBOs) are those in which the different sectors coordinate with each other at every stage and every level, and also in which a range of related functions are integrated under a single policy framework, and into the wider financial and other systems which can promote the sustainability of the partnership and its benefits. If such holistic approaches to urban development are to succeed, they require indicators which can measure progress and impact in new ways, looking at the whole urban environment rather than one aspect in isolation from the others.

77. Issues of difference. The incorporation of a gender perspective in shelter policy and planning, and the recognition of the particular needs, priorities and contributions of women in shelter development, have been one of the most positive developments in this field since 1976. In general, the conclusion must be that, while on the surface important progress seems to have been made, progress at a more fundamental level has been disappointing. Two aspects of women's participation in shelter policy are crucial: their direct participation in formulating such policies, and the inclusion of gender-specific considerations in policy-formulation. "With few exceptions, both are poorly recognized in all regions of the world" (UNCHS, 1991c, 9). Policy-markers have generally ignored the special issues surrounding women's needs for shelter, especially the key issue of their entitlements and claims to land and housing, and the legal status of these entitlements. If there has been disappointing progress in terms of recognizing and incorporating the needs and rights of women in shelter policy and practice, there bas been even less progress in making urban, planning sensitive to other "issues of difference". Gender is not the only important dimension of social differentiation in the shelter process: age, ethnicity and disability are also important, though few would deny that gender inequalities are by far the most pervasive and damaging of these factors. It is, however, important that urban planners begin to incorporate an awareness of these other dimensions of difference in their work, if shelter development is going to be responsive and relevant to the particular needs of (for example) the elderly, children and people with disabilities.

78. Going to scale, A consistent and persistent theme of shelter policy and practice over the last 20 years has been the widespread failure of innovations and successful approaches to achieve results on a significant scale. Many examples of small -scale success have been described in earlier chapters of this report, but in most cases they have remained small in scale. Alternatively, if they have been scaled up to cover larger populations, their sustainability and poverty-reach has been called into question. The goal of combining scale, sustainability and coverage has proved to be extremely elusive, For the most part, the problem has been seen in terms of program or organizational growth and replication: find a successful project or agency, help it to serve more people, and try the same approach elsewhere. This approach has a number of obvious limitations. It assumes similarities in context and conditions between different cities which simply do not exist in most parts of the developing world. It ignores the very real costs of program and organizational growth (such as increasing bureaucracy and distance between agency and consumer) which may undermine the original comparative advantages of successful initiatives as they become larger. And it ignores the special factors which usually underlie success (such as charismatic leadership or a particularly propitious set of external circumstances). "Experiments must be conducted in an environment where the institutional capacity for replication is already in place: or the links to centres of power must be strong enough so that the experiment has a path along which it can later spread to broader institutional networks" (such as government) (Tendler, 1989, 1043). The theme of linkages (to government, to commercial banks, and to other NGOs and CBOs) is repeated time and time again in the literature on scaling-up in urban areas, and provides a clue as to the factors which underpin success in this crucial area. In conclusion, the most effective way to scale up successful approaches in the field of urban shelter is to concentrate on developing the attitudes and capacities which underpin success throughout the institutions and structures which control the flow of resources, ideas and policies at municipal and national levels, be they public or private. The costs associated with organizational or program growth can then be avoided, and the danger of trading-off scale against sustainability or poverty reach can be reduced. If the policy and institutional framework is conducive, and if the capacities exist at different levels to take advantage of these opportunities, scaling up will occur. But if these conditions are not present in developing-country cities, an amount of external pressure will result in success. This confirms the central importance of capacity-building and the enabling environment in shelter development.


[Back to Table of Contents]

79. This report reviews 20 years of national action in support of shelter for all since the Vancouver Conference. There has been continued progress throughout this period in the development of more pragmatic and potentially effective shelter policies, culminating in the Global Strategy for Shelter to the Year 2000 and the "enabling approach" to shelter development it enshrines. At the national and municipal levels, this approach is gaining ground all the time. A number of significant project, program and city-level innovations and successes have also been achieved, and are described in the body of the text, These include NGO credit schemes, large-scale upgrading programs and successful urban planning in cities such as Curitiba in Brazil and Hyderabad in Pakistan. Such examples serve to demonstrate that significant progress in shelter is possible, even within the severe constraints imposed by low incomes and under-researching. However, these successes must be set against a wider background which is less impressive. There has not been a significant improvement in aggregate in the shelter conditions and shelter options of the urban poor in developing countries since 1976, and in many cases there has been a serious decline. In most areas of shelter practice, large-scale, sustainable solutions still seem very distant. Resource constraints clearly play a key role in acting as a barrier to progress, they do not tell the whole story: misplaced policies and actions, and a lack of imagination, vision, courage and commitment, are also major factors.

80. At this stage of the evaluation made on the topic, there would be little point in presenting yet another long list of recommendations for shelter policy and practice. The recommendations presented in other, existing policy documents already provide the framework that is needed for action. The problem has been, and remains today, the fact that policy is not implemented effectively in the field. Given the sheer diversity and uncertainty of shelter conditions and outcomes described in this report, detailed prescriptive policy recommendations would, in any case, be inappropriate. What matters most is the development of attitudes, skills and capacities in the public, private and third sectors to make, implement, monitor and evaluate policy and practice: and the creation of frameworks within which these capacities can be used to full effect.

81. National action. On the supply side, there is a need to legitimize (or re-confirm, since this was a basic principle of the Vancouver Declaration in 1976) government intervention in land and housing markets with the objective of removing supply constraints: the large-scale provision of basic infrastructure in public/private partnerships managed by the State: and the simplification of planning and building standards. On the demand side, there is a need to legitimize (or re-confirm, since this was also present in those recommendations) the central importance of promoting greater choice in the shelter process, so that people can adjust their diverse and ever-changing shelter needs to different housing submarkets, tenures, costs and standards. If on the supply side the key challenge is to legitimize liberalization oriented intervention in markets, and on the demand side it is to protect and enhance choice and "freedom to build then the central challenge of municipal management is to find and maintain the right balance between the two. There can be no hard-and-fast generalizations here, because the "right balance" will vary from one city to another, from one area of shelter development to another, and over time. Therefore, the key is to develop the capacity within municipal government to monitor the impact of different policy measures on a continuous basis, so that their impact in practice can be ascertained. The focus must be on the process of urban management, not the product.

82. International action. Obviously, the political and institutional dilemmas of shelter development in developing countries have to be addressed at national and municipal levels. However, international action is important in helping to create a more supportive environment in which this can take place. In part, this is a matter of resources, both increased aid flows from international donors for shelter and related services, and higher resource transfers from the industrialized countries to the developing world overall (better terms-of-trade, debt relief and improved economic adjustment programs which promote productive investment in the urban environment and protect the urban poor from unnecessary harm). If macroeconomic conditions in developing countries do not improve significantly, it will be even more difficult to address supply and demand constraints in the land and housing markets, and to improve urban management. The international community must shoulder the responsibility for addressing macroeconomic constraints at the international level. In addition to increased aid flows to shelter development, it is also important that international agencies re-examine their investment practices in order to ensure that they are as effective as possible in supporting urban development and capacity-building, and foster progress towards sustainable change. At present, there are too many uncoordinated initiatives from different donor agencies, each with slightly different conditionalities, criteria, timescales and reporting requirements.

83. Towards Habitat 11. Efficiency has never been the sole criterion for government action in shelter development; equity and social justice, conservation and empowerment are also, and perhaps equally, important. Over the last two decades, the world has moved towards a new synthesis of State, market and people which holds out the promise of adequate shelter for all. However, this process needs to go much further, to reject the simplistic faith shown by some analysts in the power of the market, just as earlier reliance on centralized planning was abandoned. Neither States nor markets are sufficient on their own to promote the efficient and the equitable development of cities; what is required is a new balance between the "complementary powers of the three systems". The result may be the kind of "community-based systems enabled by government and supported by industry" recommended by some commentators as an "alternative to the unsustainable capitalist financial market system and the inefficient State command system", and in which personal creativity and local initiative is given full rein. The experience of upgrading in low-income settlements is regarded by some as confirmation that collective action can coexist with individualistic desires based on private property rights. Such a vision tries to avoid some of the constraints and adverse consequences of both State and market-based systems by shifting control away from government bureaucracies and private firms, towards consumers and their interests. This vision is similar to a "social housing market" in which housing is transformed from an expensive commodity into an affordable social entitlement. In practical terms, this requires new forms of management and new forms of tenure, that confer the benefits of ownership (a "secure claim") while reducing prospects of speculative gains. Collective self-help and "limited equity ownership" with clear controls over resale are other possibilities. Whatever the precise form of these arrangements, the point to be made is that the old protagonism between States and markets must give way to bold and imaginative new thinking. The second Habitat Conference, which will take place in Istanbul, Turkey, in 1996, provides an ideal opportunity to breathe new life into this vision. This would be a fitting successor indeed to the vision expressed by the Vancouver Declaration almost 20 years ago.


[Back to Table of Contents]

Amis, P. (1986), "Urban proletarianisation and poverty in East Africa", paper presented to the African Studies Association, Canterbury.

Amis, P. (1989), "Structural adjustment. proletarianisation and survival strategies: the urban working class in East Africa" (mimeo) (Project Planning Centre, University of Bradford).

Amis, P. (1994), "Long-run trends in Nairobi's informal housing market", paper presented to the 2nd Symposium on Housing for the Urban Poor, Birmingham.

Amis, P., and Lloyd, P. (eds.) (1990), Housing Africa's Urban Poor (Manchester, Manchester University Press/International African Institute).

Amis, P., and Rakodi, C. (1994), "Urban poverty: issues for research and policy", Journal of International Development, vol. 6, No. 5, pp. 627-634.

Angelides, C., and Caiden, G. (1994), "Adjusting policy thinking to global pragmatics and future problematics". Public Administration and Development. vol. 14, No. 3, pp. 223-240.

Argyris, C., and Schon, D. (1978), Organisational Learning: a Theory Of Action Perspective (Menlo Park. Addison-Wesley).

Gilbert, A. (1992), "Third world cities: housing. infrastructure and servicing", Urban Studies. vol. 29, Nos. 3/4, pp. 435-460.

Gilbert, A., and Varley, A. (1992), Landlord and Tenant: Housing the Poor in Urban Mexico (London, Routledge).

Hannah, L. (1991), Urban Development: FY 91 Sector Review (Washington, D.C., The World Bank).

Hardoy, J., and Satterthwaite, D. (1989), Squatter Citizen: Life in the Urban Third World (London, Earthscan).

Hardoy, J., and Satterthwaite, D. (1991), "Housing policies: a review of changing attitudes and responses to city housing problems in the Third World". in Cheema, G. (ed.), Urban Management Policies and Innovations in Developing Countries (Westport, Praeger).

Hardoy, J., Cairncross, S., and Satterthwaite, D. (1990), The Poor Die Young: Housing and Health in Third-world Cities (London, Earthscan).

Hardoy, J., Satterthwaite, D., and Mitlin. D. (1992), Environmental Problems in Third-world Cities (London, Earthscan).

Harloe, M., and Martens, M. (1990), New Ideas for Housing: the Experience of Three Countries (London, Shelter).

Harpham, T., and Pepperall, J. (1994), "Decentralizing urban health activities in developing countries". Development in Practice, vol. 4, No. 2, pp. 92-99.

Harpham, T., Lusty, T., and Vaughan, P. (eds.) (1988), In the Shadow of the City: Community Health and the Urban Poor (Oxford, Oxford University Press).

Harris, N. (ed.) (1992), Cities in the 1990s: the Challenge for Developing Countries (London, UCL Press).

Harriss, J. (1989), "Urban poverty and urban poverty alleviation". Cities. vol. 6. No. 3. pp. 186-194.

Harriss, J. (1993), Urban Poverty (London, Save the Children Fund).

Hasan, A. (1993), Scaling-up the OPP's Low-cost Sanitation Programme (Karachi, Orangi Pilot Project).

Hasan, A. (1994a), Urban Housing Policies and Approaches in a Changing Asian Context (Bangkok. Asian Coalition for Housing Rights).

Hasan, A. (1994b), ,Profiles of three Pakistani cities". background paper for the Global report on Human Settlements.

Herzer, H. (1992), Local Governments in Argentina. INURD Working Paper 14 (Washington, D.C., The World Bank).

Hoffman, M., Walker, M., Struyk, R., and Nelson, K. (1990), Rental Housing in Urban Indonesia (Washington. D.C., Urban Institute).

Iskander, M. (1992), UNICEF Co-operation with Municipal Governments (New York, UNICEF).

Jaglin, S. (1994), "Why mobilise town dwellers? Joint management in Ouagadougou 1983-90". background paper for the Global Report on Human Settlements.

Jamal, V., and Weeks, J. (1993), Africa Misunderstood or Whatever Happened to the Rural-urban Gap? (London, Macmillan).

Jones, G. (1994), "Urban land management: panacea or myth?". paper presented to the 2nd Symposium on Housing for the Urban Poor, Birmingham.

Jones, G., and Ward, P. (1994), "Tilting at windmills: paradigm shifts in World Bank orthodoxy". in Jones. G., and Ward. P. (eds.), op. cit.

Jones, G., and Ward, P. (eds.) (1994), Methodology for Land and Housing Market Analysis (London. UCL Press).

Kahnert, F. (1987), Improving Urban Employment and Labour Productivity. Discussion Paper 10 (Washington, D.C., The World Bank).

Keare, D., and Parris. S. (1982), Evaluation of World Bank-Sponsored Urban Shelter Programs. Staff Working Paper (Washington, D.C., The World Bank).

Korten, D. (1980), "Community organisation and rural development: a learning process approach". Public Administration Review. vol. 40, No. 5, pp. 480-511.

Korten, F., and Siy, R. (1988), Transforming a Bureaucracy: the Experience of the Philippine National Irrigation Administration (West Hartford, Kumarian Press).

La Fond, A. (1994), Linking Investment and Sustainability. Working Paper 1(1 (London, Save the Children).

UNDP (1994), Human Development Report 1994 (New York, UNDP).

UNICEF (1991), Guatemala: Urban Basic Services (New York, UNICEF).

UNICEF (1992a), Case study of the UNICEF-assisted Urban Basic Services Programme in the Philippines (Manila, UNICEF).

UNICEF (1992b), Antalya Gecekondu Project (Ankara, UNICEF).

UNICEF (1993a), "UNICEF programs for the urban poor", paper to the Executive Board, New York.

UNICEF (1993b), Convergence in UBSP: an E.rploratory Study of Nasik and Aligarh (New Delhi, UNICEF).

UNICEF (1994), Juanita: Putting Children in the Electoral Arena (New York, UNICEF).

UNNGLS (1994), "Habitat 11 gets underway". Go-Between, No. 45, p. 11.

Van Der Hoff, R., and Steinberg, F. (1993), The Integrated Urban Infrastructure Development Program and Urban Management Innovations in Indonesia. Working Paper 7 (Rotterdam, Institute for Housing Studies).

Van Der Linden, J. (1993), "Back to the roots". in Mathey, K. (ed.), op. cit.

Van Huyk, A. (1987), "New directions for national shelter policies". in Rodwin, L. (ed.), op. cit.

Wakely, P., Ramirez, R., and Mumtaz, B. (1992), "The formulation of national shelter strategies: six case studies". Working Paper 59 (London, DPU).

White, G. (ed.) (1993), The Political Analysis of Markets (Brighton, Institute of Development Studies). Bulletin. vol. 24. No. 3.

Wierdsma, A., and Swieringha, J. (1992), Becoming a Learning Organisation (London, Addison-Wesley).

Woo-Jin, K. (1994), "Housing policy for low-income households in South Korea". paper presented to the 2nd Symposium on Housing for the Urban Poor, Birmingham.

World Bank (1972), Urbanization: Sector Working Paper (Washington, D.C., The World Bank).

World Bank (1975), Housing: Sector Policy Paper (Washington, D.C., The World Bank).

World Bank (1980), Shelter: Poverty and Basic Needs (Washington, D.C., The World Bank).

World Bank (1983), Learning by Doing: World Bank Lending for Urban Development 1972-82 (Washington, D.C., The World Bank).

World Bank (1991), Urban Policy and Economic Development: an Agenda for the 1990s (Washington, D.C., The World Bank).

World Bank (1992), Implementing the World Bank's Strategy to Reduce Poverty: Progress and Challenges (Washington, D.C., The World Bank).

World Bank (1993), Housing: Enabling Markets to Work (Washington, D.C., The World Bank).

World Bank (1994a), World Debt Tables 1993-94 (Washington, D.C., The World Bank).

World Bank (1994b), Working with NGOs (Washington, D.C., The World Bank).

Zearley, T. (1994), "Housing reforms in Mexico", Finance and Development, March. p. 47.