Objective

This review of literature and experience aims to summarise the main lessons that have emerged from the use of policy based operations (PBOs) since they first emerged in the late 1970s and early 1980s, focusing on lessons from the academic and policy literature as well as from the implementation experience of different agencies undertaking PBOs including the emerging evidence on good practice.

Main Findings

• The 2010 IEG evaluation of Bank support under PRSCs concludes that “upstream harmonization of the PRS process and its integration in the policy matrix has been limited.” The evaluation also notes that in many large budget support groups the Bank had limited influence in shaping the overall agenda. It observes that “more also remains to be achieved in the harmonization of results indicators, capacity building, and in reporting arrangements (IEG, 2010, p. xvii). Explaining the difficulties of harmonization and their high transaction costs, which are not always recognized and sometimes crowd out substantive issues, the IEG evaluation report finds that the Bank has sometimes reverted to means outside the joint matrix to achieve its objectives.

• The recent retrospective report of OPCS notes that during consultations some governments voiced concerns about the rigidity of the harmonization framework. These difficulties have emerged at the time when there was a need to adjust the content of performance assessment frameworks (PAF) to reflect new structures or issues that have arisen as a result of the food and financial crisis. (Box 6, p. 35 of the 2009 OPSC report).

• Stakeholders indicated that budget support provided through the PRSCs has been fairly or very predictable (that is, over the short-term or in-year). Shortfalls were not a major issue Yet other donors have managed to achieve a higher proportion of first quarter disbursements than the Bank did. This may suggest that the Bank is rather more rigid than others in its enforcement of conditionality. Data from the Strategic Partnership with Africa Budget Support Working Group, which compares commitment and disbursement data for 11 African PRSC countries, show that close to half of total budget support aid was committed for the first quarter (49 percent). Most of this (42 percent) was disbursed. By the last quarter, disbursement proportions exceeded commitments. Disbursement delays may also be due to lags in country actions for loan effectiveness. However, good alignment has required not only matching disbursement with the expenditure cycle but, ideally, matching the annual budget preparation and review cycle with PRSC preparation and follow up. This latter matching has evidently been much more difficult to achieve.

• Nonetheless, aid flows from all sources have improved in predictability in recent years (see IEG 2010 evaluation). PRSC countries did benefit, though not dramatically, from greater stability in year-to year aid flows, perceptibly, increased stability in volume, and a more stable share of fast-disbursing funds in total Bank resources. These conclusions change little if non-PRSC policy-based lending is included with PRSCs. Quarterly regularity was good although not optimal for budget planning. Governments generally also saw room for improvement in terms of timing of disbursements, the level of co-financier resources that would be available and the ability to commit finances beyond the current year.

• The IEG evaluation observes that borrowing countries strongly concur that the PRSC is well-aligned with countries’ development strategies and overwhelmingly note that alignment had improved over time, as government teams had gained knowledge of the PRSC and work with Bank counterparts. Task team leaders also consider that alignment of broad directions has been very good (95 percent agreement) and a majority (56 percent) affirm that alignment has improved over time.25 PRSCs have been selective about the areas of the PRSPs that they emphasize, sometimes including themes (in the sense of specifying actions in the policy framework) beyond the scope of PRSPs. About half of surveyed clients (54 percent) believe that the PRSC introduced new elements, outside the national development strategy. Yet, an overwhelming number (91 percent) believe that such contributions have generally been a positive contribution.

• An independent evaluation of PRSCs in Vietnam26, which was carried out in 2006, concludes that “there are clear benefits of PRSP support to the government’s reform programme”. The benefits are in the highlighting of the policy dialogue; improving policies and locking-in government’s commitment to specific reforms; speeding up the implementation of policies; and providing additional resources to support policy actions, as well as strong national ownership of the reform agenda. On the less positive side the evaluation notes that PRSCs have been less effective in influencing “strategic breakthroughs and grassroot level reform.”

 

Main Lessons

• For the Bank Group: Other agencies have simplified and unified the instruments they are using for PBOs, as well as preparing detailed procedures to guide all aspects of decision-making that are specific to the PBO instrument. The AfDB appears to significantly lag all the other agencies reviewed in both these respects with a considerable lack of clarity about the instruments and guidance for PBOs. 

• For recipient RMCs Governements and the Bank Group: There is generally recognition that the distinction between budget support and balance of payments support is not a clear cut one (especially in the modern context where currencies are largely convertible) and that while there are distinct design issues related to PBOs directed at addressing short-term externally driven shocks, embedding a distinction between budget support and balance of payments support in policy guidance would not be particularly helpful. Specific objectives need to be defined for a specific operation. More generally, it is important to regard PBOs as supporting an agreed programme of action or policies, rather necessarily than as an instrument for policy change or reform, though this confusion appears still to be widespread in donor practice and may reinforce the tendency to continue to use conditionality, and to consider results in terms of policy changes effected, rather than in terms of the use which has been made of donor support to increase public spending, which should be what is principally used to judge the results of a PBO, with policy change usually related more directly to complementary actions such as policy dialogue and advice, rather than to the provision of funding. Ability to engage effectively in dialogue may depend on the extent to which authority is delegated to country offices and teams, and to the quality and quantity of in-country representation. 

• For the Bank Group: There is a general (though not universal) move towards single tranche operations, often based on prior actions, though with these occurring, as with the World Bank, in a programmatic framework of repeated operations in support of a medium-term government plan of action (like a nationally owned PRS). While there is evidence that single tranche operations provides reasonably predictable within-year aid flows compared to multiple tranche operations, they may risk unpredictability over the longer term without a programmatic framework. This may be difficult for the AfDB to implement specifically for ADF operations without some changes to the ADF framework. However, addressing this appears to be an important requirement for improving the effectiveness of AfDB use of budget support. 

• For the Bank Group: Other agencies do not appear to operate specific quantitative restrictions over the proportion of total loans or grants that are provided using PBOs, unlike the situation for the African Development Fund where the share of PBOs is capped, although agencies may set targets for the use of PBOs for fiduciary risk management or other strategic purposes or as part of an overall country strategy. 

• For the Bank Group: In relation to engagement in fragile situations and the role of PBOs, there are some differences in agency practice but it is generally recognised that there is an important need for the use of fast disbursing instruments but that these need to be supported to a greater degree than elsewhere by complementary actions to improve effectiveness and to manage risks. 

•  For the Bank Group: The difference between general and sector budget support relates to the scope of policy dialogue and the extent to which funds are earmarked, either notionally or through the management of funds through special accounts to ensure that they are used in particular agreed ways 

• For the RMCs Governements: A general weakness of donor practice with PBOs has been a failure adequately to set out the intervention logic by which the different elements of the PBO (finance provided, definition of conditions, policy dialogue, and complementary measures) are supposed to contribute to achieving defined results, and a consequent failure to establish effective monitoring of intermediate results. A particular weakness has been in relation to the link between a PBO and the achievement of PFM improvement results. This is likely also to be an area in which AfDB practice could be improved