Introduction

The scale and the complexity of urban housing problems in developing countries raises major questions concerning the role of the shelter sector within social and economic development strategies. State intervention in the shelter sector exerts both direct and indirect influences on patterns of savings and investment. These in turn influence the quality of life for all. While public-sector expenditure on shelter may be somewhat modest compared with that of other sectors, total investment frequently represents a considerable part of total capital accumulation in a country. Furthermore, on the household level, it usually represents the largest single investment ever to be made. The social and economic significance of the shelter sector is thus considerable.

In their attempts to improve existing housing conditions and stimulate the supply of shelter for their increasing populations, governments in developing countries have pursued a wide range of policies, programmes and projects. Yet, the wide range of institutions and professions actively involved in the shelter sector complicate the assessment of these approaches. Further complications arise from the close links to developmental and welfare strategies. This diversity and ambiguity may account for the lack of specific shelter policies in many countries. It also partly explains the widespread adoption of projects as the main vehicle for public-sector intervention.

The obvious advantages of projects are that they are designed to produce a measurable and visible output for a pre-determined investment, are subject to established administrative procedures, and can be targeted at specific social groups. They also offer the potential for flexibility in the use of resources and in organizational learning and capacity-building. Furthermore, they fit comfortably within the grant provision and lending practices of international aid agencies (Honadle and Rosengard, 1983). In a qualified defence of the project approach, Rondinelli (1983) sums up the advantages of the project approach as outlined below:

· Projects are identifiable, bounded and organized sets of development activities;

· Projects can be effective means of translating development plans and policies into specific courses of action;

Projects are vehicles for mobilizing and allocating resources to development activities;

· Projects can be analysed and appraised before funds are committed;

· Projects are temporary activities that can lead incrementally to accomplishing larger development goals;

· Projects can be used to undertake unique, innovative or non-routine development activities;

· Projects can be used to channel development resources to specific groups of beneficiaries and to particular locations;

· Projects can be formulated as manageable units of activity guided by well-defined planning and administrative procedures;

· Projects can be organized in a variety of ways and undertaken by a wide range of organizations;

· Projects can be externally supervised and controlled and used to exert influence on broader development policies.

Conventional projects for new housing have generally involved the construction of housing units in specific locations for a number of previously selected households unable to afford acceptable housing provided by the private sector. Early approaches towards existing substandard, or unauthorized settlements regularly focused on slum clearance and relocation projects in an effort to impose pre-determined notions of officially acceptable shelter solutions, without reference to their affordability or acceptability. The failure of either of these approaches to bring housing within the reach of low-income populations in rapidly expanding urban areas has led to criticism not just of these specific projects, but of the relevance of the project approach itself.¹

^1/ For a discussion of the relevance of the project approach in general, see Honadle and Rosengard (1983), Rondinelli (1983) and Morgan (1983).

During the last two decades, however, several innovations have taken place in the project approach. This suggests that a re-evaluation is in order. Sites-and-services and settlement-upgrading projects have been adopted by countries at different levels of economic development and of different political persuasions in all parts of the world. These approaches emphasize the benefits of incremental development, flexibility and efficiency in the use of scarce resources and the need to achieve a “multiplier effect”. The changes have coincided with major innovations in shelter policy. They involve a shift away from attempting to meet housing needs through direct provision, towards enabling or facilitating approaches. Such approaches are designed to support the efforts of existing suppliers of land and housing, such as the formal and informal private sectors and community groups.

Rather than suggesting a scrapping of the project approach as such, due to shortcomings of the past, this report identifies how it can better reinforce broad policy objectives. The project approach will thus, not only because of its present pre-eminence, remain a major factor in enabling the urban poor to obtain affordable and acceptable housing as cities continue to grow. This report assesses the experience gained to date through the project approach and identifies some major lessons learned. It also offers recommendations intended to improve the role of projects within the framework of enabling shelter strategies as described in the GSS (see UNCHS, 1990). It is based upon reports commissioned by UNCHS in Colombia, Indonesia, Sri Lanka, Turkey and Zimbabwe, though reference is also made to other cases where appropriate. It would be presumptuous to claim that the report is either comprehensive or definitive, since the sample is only based on five countries. Yet, the countries selected cover a wide range of experiences, and the projects reviewed broadly reflect approaches that have been widely adopted.

In assessing the experience gained, it is important to relate examples to their context, since any approach is likely to have limitations as well as benefits and some will work better in some contexts than in others. The real question when assessing the impact of a policy, programme or project is whether its merits outweigh its defects. This report, therefore, concentrates on the major characteristics of the project approach as these have been observed in the five countries mentioned above in particular, and in other countries for which general information has been made available.

The specific projects reviewed were selected according to some general criteria. The projects are:

· Located in urban areas;

· Explicit about their focus on the shelter needs of low-income groups;

· Completed and evaluated (to enable assessment that could lead to policy changes);

· Considered as successful in meeting their objectives to address the needs of low-income groups.

Basic information on the case studies reviewed in this report are presented in table 1.

Table 1. Projects reviewed in this report

Source: Herlianto (1990), Jayaratne (1990), Mutizwa-Mangiza (1990), Tokman (1990) and Utria (1990).

* Low-(L) or middle-(M) income groups