Monitoring systems in Africa are increasingly reflecting PRS principles but the challenge of monitoring PRS implementation is still significant.

The proposed systems consciously build on existing systems, whilst strengthening them with selected participatory instruments and the involvement of a broader range of stakeholders. There are several specific commitments to improve the dissemination of information. Half the cases provide more scope for civil society involvement in the monitoring system through providing space for CSOs as regular interlocutors at specific events or as working group members on a more regular basis.

The ‘missing middle’ identified by Booth and Lucas (ODI, 2002) in I-PRSPs and early full PRSPs is gradually being addressed: most systems appear to have given considerable attention to intermediate indicators. In many cases indicators have been chosen in consultation with sectors or other stakeholders and generally reflect PRS priorities (with the exception of governance in some cases), although many will need to be revisited and streamlined in the near future. Most systems have yet to establish clear ‘monitoring chains’ whereby it is possible to track change through the stages of input–output–outcome–impact. Systems are also still weighted towards monitoring final poverty impacts with an emphasis on technical as opposed to strategic criteria for monitoring PRS implementation. Some confusion still exists over how to combine poverty monitoring and PRS monitoring within a common system.

Governments generally identify the sources of data to be used in tracking specific indicators, but the provision of baseline figures for these indicators is still erratic. The systems all intend to draw on a number of types of data - including surveys/censuses, routine data, and participatory data – although feeding these into policymaking in an integrated and usable format remains a challenge. Dissemination of key outputs such as annual impact reports is improving with two examples of good practice (Tanzania, Uganda). Most countries have plans to strengthen their routine data systems, although the efficacy of these plans may be open to question. There is some use of ‘light’ instruments (instruments that require less time to conduct than, for instance a full household survey and allow for quicker analysis of results) to complement more traditional surveys, although some governments are reluctant to introduce new instruments because of cost and capacity constraints.

In most cases, the monitoring system’s institutional arrangements are not yet functioning. There are a number of issues with the proposed frameworks, including duplication of functions and the capacity of key agencies/ departments. Even in the cases of Tanzania and Uganda where the system is functioning, there is still work to be done in embedding links between agencies. There is early evidence of an (imperfect) link between monitoring systems and budget processes, though much remains to be done to strengthen this.1 Cabinets and parliaments have a formal role in some systems