This paper reviews the extent of evidence for the effect of Public Financial Management (PFM) interventions on outcomes in low and middle income countries. It is based on a database of 197 studies compiled after a rigorous search process. The number of studies in the field was surprisingly low, reflecting a series of challenges in acquiring robust evidence. Studies tended to emphasise the budgeting stage of the financial management cycle, with relatively few studies about the implementation stage. Evidence on outcomes was patchy. Most studies considered multiple interventions, reflecting the need for holistic approaches to PFM systems, but this frustrated attempts to attribute results to particular interventions. The most common outcomes noted were improved transparency and improved accountability. Evidence maps summarise the findings, showing the extent of evidence across combinations of intervention and outcome. These suggest that we have scant evidence for most combinations, with the exception of work in participatory budgeting and related fields.
An accompanying database is also available. It provides a searchable list of the documents that met the inclusion criteria and enables users to generate bar charts showing the frequency of documents in each criterion.
Features of the evidence base
- The most prominent institution covered was the Ministry of Finance. This reflects the importance of the MoF in PFM practice, but other institutions seemed relatively neglected, especially the political bodies in the budget-making stage at national level.
- The most prominent categories of intervention were changes to processes and systems (especially budgeting systems) followed by relatively “concrete” interventions such as passing new laws and developing IT systems. “Softer” interventions were less represented.
- The most common specific intervention types were related to participatory budgeting or similar community level work. This helped inflate the figures for budgeting work in general. Whilst there is considerable interest in such developments, they relate to a sub-set of local government or community level spending, which represents only a small proportion of total public expenditure. Thus national PFM systems appear under-studied relative to the participatory approaches.
- Geographically the most prominent region covered was Sub-Saharan Africa, reflecting the weight of much PFM work. It may suggest PFM thinking could be excessively driven by African experience.
- The dominant methodology used was case studies. Only three studies were experimental. A significant number of meta reviews had been conducted, but these were spread thinly over a large range of different interventions and were often selective in the countries or funders examined.
