|Assessment of Experience with the Project Approach to Shelter Delivery for the Poor (HABITAT, 1991, 52 p.)|
|V. Shelter projects and national policies|
A precondition for assessing the impact of projects on shelter policies is that such policies exist. In many countries, government interventions in the sector reflect a wide range of separate objectives and measures adopted by different ministries and development authorities that are accountable to different constituencies and operate under different criteria. This is partly a reflection of the diffused nature of the shelter sector, with its relevance to economic, social and environmental policies and its relative lack of importance as a specific activity in the eyes of many governments. The outcome of such a situation, however, is that it is difficult to assess the impact of discrete interventions such as projects.
Where national shelter policies exist, the impact of projects appears, on balance, to be modest. In both Turkey and Sri Lanka, a range of strategies has been adopted in the shelter sector. In Turkey, it appears that projects have failed to generate any change in national shelter policy. However, the acceptance that they have failed to address the demand for housing has led to the acceptance of non-project approaches that provide security of tenure and services in informally developed settlements. In Sri Lanka, the upgrading of informal settlements was undertaken as a complementary activity to the supply of new plots through the MHP (Jayaratne, 1990: 37). This was due to a recognition that informal settlements were making a positive contribution to total housing supply, especially for the low- income groups. The innovations achieved through the MHP have made a major impact on the shelter options open to low-income households. Yet, it would not be correct to conclude that this was due to the impact of projects upon policy, since in practice the policy was formulated first and projects were then initiated to implement the policy objectives.
In Indonesia, experience with the project approach shows that it has exerted a considerable influence on national housing policies since independence in 1945. Large projects, such as that at Klender in east Jakarta, have enabled a range of income groups to be served on a relatively self-financing basis, through extensive use of cross-subsidies (Herlianto, 1990:32). Whilst this does not amount to an impact on policy as such, it is significant that a private-sector project in Bandung was awarded a Best Achievement Award by the Government. The project was later emulated by other private developers. By the same token, the NGO projects in Semarang were officially endorsed in 1989, when the State Minister of Housing established a task force to encourage and enhance the role of NGOs and community participation in housing development (Herlianto, 1990: 56).
Among all the cases where government has used the shelter sector as a central element of policy, the case of Colombia deserves mention. The 1983-1986 National Development Plan identified housing as a driving sector of the economy, with the intention of both providing housing at the scale required and stimulating economic activity and employment (Utria, 1990: 17). This resulted in a major expansion of housing supply that initially benefited all social groups, including the poor. Later in the decade, however, the populist element weakened as projects concentrated on housing for middle-income groups. The shelter sector policies of the new Government of Colombia place the emphasis on establishing a national housing system based on subsidized social housing for low-income groups (Utria, 1990: 34). The scale of such subsidies, however, will place considerable strains on the Colombian economy. Only the future can tell whether this policy can be sustained.
Evidence from other countries tends to confirm the modest impact of projects on policy. This can best be appreciated in the case of sites-and-services projects, which have been undertaken throughout the developing world. Despite their widespread application, Hardoy and Satterthwaite (1981: 254) found that in only two out of 17 countries surveyed had such projects become the central part of urban housing policies. This may be for several reasons, though three appear common. First, such an outcome is not intended in the first place; secondly, it may be that governments lack the confidence or technical resources to initiate new approaches; or thirdly, that strong vested interests oppose any change to existing programmes. Whatever the reason, or reasons, it is generally true that the impact of projects on policies has been limited. As Hardoy and Satterthwaite have also noted (1981: 254), projects of any type divert attention away from the real priority, reforming the urban land market so that it does not automatically exclude lower-income groups from legal housing.
Settlement-upgrading or -improvement projects do not appear to be as vulnerable to the above criticisms. The impressive Kampung Improvement Programme in Indonesia is by no means the only example of government agencies undertaking supportive, realistic and replicable improvements to the living conditions of a significant proportion of low-income households in large cities. They have also been more successful in reaching the very poor than new development projects.
The lack of independent evaluations of projects, in terms of their internal objectives as well as their wider policy impact, makes it difficult to identify future roles with confidence. More effort is needed to evaluate the experience gained so far, and greater willingness is equally required to accept and act upon such evaluations. At present, there are too few incentives for public-sector personnel to learn from previous experience and to rectify the limitations of completed projects in preparing new ones. Correcting this will require instilling a new institutional culture in many housing and planning agencies. This can only be achieved in full through the education of the next generation of professional staff and the reorientation of existing personnel.
In the final analysis, the impact of projects on policy also depends upon external factors that are beyond the control of project planners. The Ismailia Demonstration Projects in Egypt for example, achieved a far greater impact upon housing policy than the project team had even contemplated. Ironically, this was not because all the project objectives had been achieved. In fact, changes to the project made during the process of implementation made it more difficult for low-income households to participate. Yet, the standard of construction achieved by residents of what was intended as a self-financing project for low-income groups, was so high that senior officials and politicians, including the President, were sufficiently impressed to announce that future government housing policy would concentrate on the provision of infrastructure and finance, rather than conventional housing supply (Davidson, 1984: 146).