Social Development Policies, Results and Learning: a Multi-agency Review

This report is one of the outputs from a multi-agency review, for DFID's Social Policies and Systems Project (SPSP), on social policies and systems practices in other development agencies. The aim of this paper is to inform on approaches taken by donor agencies in developing and implementing social policies and systems and to assess the effectiveness of these investments in terms of positive social policy outcomes. It discusses the relevance, for policy-makers and practitioners, of the information produced in the evaluation of the impact of aid interventions.

The paper starts by analysing the role of social policy in the strategies of the donors organisations. It concludes that social policy is mainly implicit, sporadic, inadequately described, and in some cases inconsistent. Although social policies of partner governments are often discussed in policy documents and evaluations, it is not generally clear how such assessments are used by donors to shape their country strategies, and still less clear whether (and if so how) they inform donors' overall aid, foreign and domestic policies.

Before undertaking the assessment of the outcomes of the social policy programmes, the paper discusses the relevance of the information generated in the their evaluation. It is argued that there are few examples of systematic use of impact information to shape sector or country programmes or aid policies and allocations other than at project level. Furthermore, modifications of policies, macro-level practices, priorities, and aid allocations are influenced by a variety of factors, among which information about donors' own achievements plays only a very minor role. On the other hand, it is argued that this information is not useful for policy-makers and planners. The paper shows that the main sources for information and influence, even at the head office level, are informal ways, such as short memos; personal rather than institutional memory; networks rather than rationally planned communication channels; chance encounters and lunch meetings rather than organised meetings. It is concluded that externally-driven evaluation studies are widely regarded as wasteful, since people learn best through participatory assessment of what they themselves are achieving. Searching for generic information on approaches that get good results may therefore be an unfruitful exercise. More effective would be for social development advisers to sharpen up their concepts of social development by defining categories of people, problems and processes that are of critical importance to good development.

Finally, the paper reviews the process of evaluation of the outcomes of social policy projects undertaken in these institutions. It concludes that the evaluation work is dominated by the purveying of truistic 'lessons' which are not new lessons. What is needed is more analysis of why these lessons are being ignored.