Decentralisation – Urban development

Over the last 25 years, more than 75 countries – developed, developing and emerging – have transferred certain powers and responsibilities to the lower levels of government. The declared objective is to improve the efficiency and coherence of government activity, particularly as regards the provision and functioning of public services. This decentralisation trend has led AFD to undertake operations in partnership not only with states, but also directly with local authorities. It is therefore essential to understand the impact of decentralisation on development processes, and particularly on a country’s ability to provide basic services to its population.

There is a good deal of theoretical work on these questions. In particular, work on the theory of financial federalism aims to take the spatial dimension of government activity into consideration. Ways and means of putting this theory into practice are also analysed using political economy models and agency theory. The AFD Research Department is currently engaged in reviewing and summarising these theoretical contributions.

Although such background studies are necessary, operationally oriented research is also essential. In practice, decentralisation processes are highly complex. They can take a variety of forms (political, administrative, fiscal etc.), and may be implemented in a variety of contexts: developing or emerging countries, with highly centralised states or with decaying central governments, containing a number of different ethnic communities, etc.

Each country is a unique case: theory, particularly economic theory, cannot provide a unique model for organising government activity, a model that governments need merely apply in order to obtain efficient public services. However, theory does have the merit of drawing attention to essential points in this process and make it possible to construct a grammar of all the determinants and possible impacts of the spatial organisation of public action.

AFD research in progress on this topic seeks to develop analytical grids that take account of the diversity of possible decentralisation configurations, based not only on theory but also on experience on the ground.