Speakers:

Anjali Kumar - Lead Economist, Corporate and Global Evaluation & Methods, Independent Evaluation Group

Marcus Manuel - Head of CAPE, ODI

Mick Foster - Independent Consultant

Chair:

Alison Evans - Director, ODI

Alison Evans, Director of ODI opened the meeting by introducing Anjali Kumar to present the World Bank Independent Evaluation Group's recently published report evaluating the success of (Poverty Reduction Support Credits) PRSCs. Mick Foster, Independent Consultant, Marcus Manuel, Head of Programme for CAPE and Robert Okudi, Principle Economist in Uganda, all acted as discussants.

Anjali Kumar began by presenting the background of PRSCs. The World Bank introduced PRSCs in 2001 as an instrument for better budget support. PRSCs were seen as a departure from the modalities of budget support lending in the form of structural adjustment lending. PRSCs were designed to support country ownership, ease conditionality and increase the predictability of donor funding.

PRCSs have been a tremendously important World Bank instrument; between 2001-2009 there were 99 PRSCs approved countries, covering more than $8 billion of lending. Despite this, the share of PRSCs in recipient government budgets has been declining more recently and this matches recent evidence on budget support as a whole.

PRSCs have six core objectives and the primary objective is to support strong performing IDA countries to implement their poverty reduction strategies. The evaluation assessed the extent to which PRSCs were able to meet their core objectives, and examined their relevance and effectiveness to support development. All PRSCs from 2001 till June 2008 were covered in the evaluation.

The evaluation was based on a large number of sources including desk based reviews, field evaluations, tailored made surveys for the evaluation, relevant databases and literature from other case studies. The methods of evaluation consisted of a traditional process of triangulation to establish plausible causality. To try to understand and measure the effects of PRSC support, comparisons of the’ before and after’ changes in the performance of PRSCs beneficiaries was carried out.  In addition, comparisons between changes in PRSC countries and changes in IDA non PRSC beneficiary countries were made. Two immediate methodological challenges were evident; i) attributing development improvements to PRSCs; ii) accounting for changes made to the framework for World Bank Developing Policy Lending in 2004.

In terms of the results of the evaluation, Anjali first discussed aid design and process. In respect to conditionalities, the evaluation found that PRSCs were associated with lower levels of conditionalities than previous World Bank lending instruments (particularly in relation to legal conditions). Yet, the evaluation also discovered that over the same time period other policy lending instruments changed in parallel.  Second, over the evaluation period ‘program bench marching’, as with conditionalities, reduced in both PRSCs and other policy based lending instruments. So the influence of PRSCs was significant but not unique. Third, flexibility improved in PRSCs, measured by the fact that a number of triggers were adjusted, downgraded or postponed in PRSCs results frameworks. The evaluation also examined the predictability of funding and the regularity of disbursements, both of which improved in PRSC countries.

An assessment of alignment, ownership and operationalisation of national plans found that overall PRSC lending in countries was well aligned with national development strategies and PRSC countries had greater ownership of the process than with previous adjustment lending. The evaluation also showed an increase in sector spending in PRSC countries on education, health and public financial management as compared to previous lending. Problems with results frameworks were mentioned, such as challenges with the selection of indicators, baseline data and synchronising of PRSC and government results frameworks. In terms of donor harmonisation, the results were quite mixed. In some countries such as Vietnam the PRSCs served as a focal point for Donor Harmonisation, but in others its role was not so significant.

Anjali then went on to discuss the evaluation of PRSCs outcomes. First, in public financial management, a core objective, the accomplishments were more evident in ‘easier areas’, such as budget classification reform, whilst achievements in areas such as ‘bringing donor aid on budget’ was more difficult to achieve. Second, it was difficult to establish the effect of PRSCs on growth and poverty reduction. Growth grew faster in PRSC countries in the PRSC period compared to the preceding period, however the story was similar in non-PRSC countries. Poverty reduction fell in PRSC countries, but it also declined in non-PRSCs countries and in some cases the decline in poverty was faster in the period prior to the PRSC intervention. In conclusion, outcomes have improved in PRSC countries, but attributing this to PRSCs is difficult. .

Anjali summarised the key messages of the evaluation: aid processes under PRSCs clearly did improve, but the effect of PRSCs on growth and poverty outcomes is less clear.

Mick Foster discussed the origins of PRSCs. He explained how donors, in the 1990s, started to recognise that programme aid was supporting the budget, not just the balance of payments. Alongside this donors called for governments to set out their expenditure policies and plans to tackle poverty. Once reviewed, donors began committing funds over a medium term period to support the implementation of government policies. However, ‘poverty reduction strategies’, originally meant to be fully government owned, became a product pushed by donors with content prescribed by donors. In addition, poor deployment and coordination of donor support led to distortions in implementation.  Mick then went on to say how recent developments in Accra and Paris Declaration, as well as the evaluation recommendations have brought us back towards the original vision for PRSCs. But he asked whether the World Bank is still too remote, too obsessed with its procedures and too little committed to supporting Government driven processes and time schedules, to be fully implement the original objectives?

Marcus Manuel reflected on the difficulties of establishing a sound methodology for measuring before and after effects of aid interventions. However, he felt that the evaluation could have drawn out more effectively some of the positive findings, particularly those related to poverty outcomes. He noted that when the data was disaggregated, it was clear that in PRSC countries the rate of poverty reduction had doubled in PRSC periods vs non-PRSC periods, and such a decline was much greater than that achieved in non-PRSC countries. A similar story could be told for child mortality rates. He also suggested that the report could have focussed on understanding the institutional impact of the PRSCs, given that they were considerably different from their structural adjustment loan predecessors.

Robert Okudi discussed the report in the context of Uganda. First, he touched on the ‘branding’ of PRSCs. Robert said that the ‘poverty’ brand helps facilitate the approval of loans through Parliament, because poverty related issues are appealing to politicians. Second, Robert agreed with the benefits of simplifying the language related to conditionalities, but suggested that conditionalities, if tied to the disbursement of funds, should be tied to future not in-year disbursement changes.  Finally, he suggested the PRSC dialogue should focus on the macro level and be supported by additional sector level working group dialogue.