This paper synthesises the key findings from case studies in five countries (Ghana, Malawi, Mozambique, Tanzania and Uganda), each of which examined how public expenditure management has been linked to poverty reduction policy goals. Each of our case study countries entered the 1990s with a pattern of public expenditure in which the efficiency and effectiveness of public expenditure was very low, and its benefits went mainly to the non-poor.

The hypotheses we set out to examine in this research can be summarised as follows. In order for public expenditure to better serve the interests of the poor, political will to confront difficult choices is necessary, but not sufficient. It needs to be allied to more effective public expenditure management; macro-economic and budget stability, and budget systems that turn policy analysis into actual cash releases to implement the intended policies. This in turn must be allied to reforms that bring the incentives facing those required to implement expenditure programmes more into line with the objectives of policy.

In achieving all of this, we hypothesise that transparent flows of information will be important in keeping Government honest, and that wider publicity on the nature and extent of the problems faced by the poor will help to secure increased focus on improving their lot.