Main Findings

• Donors have committed PFM Reform support of around USD 104 million over the decade, largely in the form of institutional capacity building activities and advisory TA. Government’s inputs were staff time, the recurrent cost of new PFM institutions like the ODPP and the common procurement service, the CIAU and the common internal audit service, and development budget spending on PFM institutions. Of significant importance for the evaluation are government’s inputs in respect of IFMIS development and roll-out after 2004, which was almost entirely funded by the GoM and costed an estimated USD 15 million in project costs.

• The establishment of formal structures to coordinate PFM reforms (the PFEM steering committee, technical committee and unit, and various working groups) has been important in creating capacity to support Malawi ownership and leadership of reform plans. The GFEM group – and CABS – provided a structure to harmonise donor PFM inputs. However, until 2010 budget support performance assessment frameworks and PFEM Action plans were an amalgamation of individual reform interests rather than a coordinated and sequenced response to PFM weaknesses. The GoM has recognised this in the drafting of the next PFEM Reform Plan.

• Even if reform inputs were directed at areas of PFM weakness, PFM reform models were not sufficiently adapted to the institutional context, the level of ownership of reforms or the  capacity for reform implementation. The procurement reform model, the internal audit model and even some of the budgeting reforms follow international best practice, but are not suited to a context in which the technical and managerial skills to run the envisaged systems are scarce in society overall, and in which the civil service context means that these skills are difficult to recruit and to retain when trained. The result is a sophisticated system on paper, which does not deliver functionality on the ground. A less sophisticated reform might not have checked all the boxes, but could have provided a greater improvement in functionality. While reforms in the early years of the period proceeded without sufficient GoM ownership, the targeting of reform inputs became more responsive to the level of ownership over the period, primarily because ownership emerged rather than because donors changed their approach. It is not clear that going forward donors will continue to be proactive, providing support which takes into account the PFM reform cycle, or will again be reactive, with their support out of sync with PFM reform windows of opportunity.

• Major reform outputs were delivered over the period, signalling significant change in the capacity for budget preparation, budget execution and audit. Most of the outputs put in place over the period had external support. However, the effect of the political change in 2004 and the degree to which it drove reforms from within is significant. In retrospect, among the reform achievements of the GoM, the procurement and roll-out of an IFMIS that functions stands out as perhaps the most significant change, supporting a series of further secondary reforms. Yet, this functionality was achieved with relatively little direct external support.

• Between 2006 and 2011 Malawi PEFA scores show a significant improvement of just over half a “notch” on average, i.e. from 2.1 (C) to 2.52 (C+), with 31 of 65 indicators improving, 29 remaining stable and only 9 deteriorating. This places Malawi above the median value for the 100 countries in the De Renzio et al sample (2010) for three of the six PFM functional clusters. Strategic budgeting and resource management improved significantly, budget preparation and accounting and monitoring showed small improvements and external accountability and internal control moderate improvements. Note that for internal control the IFMIS improvements were already factored into the 2006 report. However, a comparison of the three PEFA reports (2006, 2008 and 2011) show that most improvement occurred prior to 2008, and that for most clusters there was some deterioration between 2008 and 2011.

• The implementation of budget classification changes and the integration of these into the chart of accounts for reporting against budgeted allocations, improvements in in-year predictability of cash transfers from the Treasury, as well as the improvement of the links between policies and budgets have been key parts of the GoM reform programme since 2000s. But there is no systematic effort to provide quality information at service delivery unit level yet. Donors have been supporting these reforms since the HIPC action plan, and the GoM through the Budget Division in the MoF has been pursuing performance oriented reforms since the mid-1990s. If sustained improvements in these areas can be achieved, they ought to impact positively on service delivery, including for women and vulnerable groups.

• Areas with high improvement in PEFA scores (of 1 notch and more) had seen significant government inputs. Areas with medium improvement in PEFA scores (less than 1, more than 0.33 points) relate to three scenarios: (i) reforms were supported by both government and donors, but the level or quality of the outputs were not sufficient to generate significant  functionality improvements; (ii) reforms were supported by donors in the first period, but there has been a decline in donor funding to sustain the reforms; (iii) reforms were supported by donors throughout the period, but GoM inputs were intermittent. PFM sub-clusters with low improvement or a deterioration in PEFA scores are associated with low government inputs.

Main Lessons

Main Recommendations

Recommendation(s) to the Co-Financier:

• Donors need to be cognisant of the PFM reform / political cycle and local context in Malawi in order to maximise the effectiveness of their inputs. This has two aspects: (i) Given their high influence on reform choices, donors need to be more aware of the local context and constraints on reform choices. This may mean making much higher use of local and regional technical advisory capacity and in line with Malawi preferences, focusing on advisory rather than managerial technical assistance; (ii) Secondly, donors need to match the scale and focus of their support to make full use of PFM reform windows on the one hand, and on the other, sustain critical PFM functions when the political conditions turn adverse. This means that donor PFM programmes should be responsive to changes in conditions for reform. They should accelerate the amount of funding and other inputs and support new reforms / use policy dialogue to argue for new reforms when conditions turn positive. Equally it also means that when conditions turn adverse – e.g. when PFM slips from government’s critical priorities or when the ownership and leadership of reforms in a subsector and reform drive are not present – donors should not attempt to introduce further reforms, but rather focus their efforts on sustaining existing reforms and results.