| |
Contributions from
the Column
Focus
The EU’s performance-driven aid
There is no substitute for ownership
Reviewing conditionality
“We will not accept aid at any terms”
Relationships matter, procedures do not
07/2005
|
|
No substitute for country ownership
It is by now well known that imposing strict conditions on countries in the hope of leveraging reform has not been helpful. Like other donors, the World Bank is therefore trying to bring the use of conditions into line with the principle of country ownership.
[ By Stefan Koeberle ]
Empirical research has repeatedly shown that programmes designed by external donors are less likely to be sustained than those that result from the political process within a country. Therefore the World Bank has been stressing the issue of country ownership over the past few years. Ownership does not necessarily mean full consensus, because there are always different interests and different stakeholders at work. Therefore, it makes sense to define ownership as a reasonably good likelihood of a programme being implemented and a country staying committed to it. The idea is to support a broadly owned policy rather than just a few individual champions within a government.
The World Bank does not intend to prescribe the reform content, in the sense of listing items of what it will or will not support. If a country is truly committed, support may be appropriate even in areas that are controversial, such as privatisation, state enterprise reform, or public sector modernisation.
In spite of the importance of country ownership, donors in general cannot forsake conditions. After all, they must make sure that effective use is made of their funds and they need to report to their public and their legislators at home. In discussing conditionality, we should thus acknowledge four tensions among competing objectives up front. These are:
– Country ownership versus donor ownership. While country ownership depends on political action within the recipient country, the donor’s need for accountability also implies a sense of donor ownership. This tension between two diverging objectives cannot simply be wished away.
– Predictability versus performance. Predictability has emerged as a key objective in aid dependent countries, some of which have been adversely affected by volatile aid flows in the past. If a country is performing adequately, this does not pose a problem. But there is no easy answer to the question of how donors should react to inconclusive or even unacceptable performance.
– Accountability versus flexibility. On the one hand, we need flexibility to deal with unforeseen circumstances and difficult changes, on the other hand, we need a clear and transparent implementation framework that can be shared with different stakeholders. These objectives are not always compatible.
– Donor commitment versus recipient commitment. This issue goes beyond aid as such and tackles the broader political environment. Donor commitment should result in policy coherence. For example, aid flows should go along with development-friendly trade policies and the reduction/phasing out of agrarian subsidies. That would serve as a major incentive for developing countries to undertake mutually agreed reforms. The Monterrey Consensus was expected to bring such matters into line, but the tension still remains.
Recent trends
Aware of such dilemmas, the World Bank has broadly moved away from traditional conditionality, which focused on short-term macroeconomic adjustment and removing major economic distortions. The popular debate on conditionality tends to focus narrowly on the number of conditions per loan, which has usually been perceived as excessive. In practice, however, this issue is quite complex. The number of conditions itself says little about their relevance and usefulness. Nonetheless, over the past few years, the average number of conditions in World Bank policy-based loans has been substantially reduced. Recent management directives have reiterated the importance of limiting the number of conditions. The average number of conditions fell from above 35 in the late 1980s to about 15 in 2004.
Over the past few years, the Bank has increasingly used a programmatic approach for its policy-based lending. This approach involves a series of single-tranche operations within a medium-term framework specified at the outset. It spells out completed prior actions, monitorable progress indicators, and expected prior actions for subsequent operations. Sequentially, the Bank’s Board of Executive Directors decides upon these sequential operations. To the extent possible, disbursements are aligned to the receiving country’s budget cycle. By replacing traditional conditionality with an approach that builds on completed actions, instead of promises, programmatic lending contributes to systematic policy implementation and provides sufficient flexibility to support complex medium-term institutional reforms that are often unpredictable. Poverty Reduction Support Credits (PRSCs) are emerging as a vehicle that incorporates these principles.
In August 2004, the Bank formally recognised the significant changes that had taken place in its policy-based lending, by issuing a new operational policy (OP/BP 8.60), which replaced the term “adjustment lending” with “development policy lending” (D+C/E+Z, October 2004, p.358). Among other things, the Bank retired prescriptive passages on privatisation, financial sector reform and public sector reform. Instead, the new rules explicitly aim at supporting a country’s policy as well as its institutional actions to promote growth and achieve sustainable reductions in poverty. These programmes are expected to be based on country- and sector-wide analytical work (by the country itself, third parties or the Bank) covering fiduciary arrangements, the environment, forests and natural resources, as well as likely social impacts. OP/BP 8.60 mandates that conditions should be confined to those actions that are critical for the implementation and expected results of country policies. These policies, in turn, must have been discussed with local stakeholders and include a framework that allows for adequate monitoring and evaluation.
Good practice
On this basis, there seems to be an emerging consensus among aid practitioners defining how donors should apply conditionality. One way to structure a possible set of good practice principles is to look at (I) considerations in applying selectivity, (II) design criteria, and (III) how conditionality is actually implemented.
A key guiding principle is clearly the notion of ownership as discussed above. The second principle would be criticality. Conditions should focus on what is indispensable for successful policy implementation. Third, all countries are very different and the customisation to country circumstances needs to go much further than in the past. Where donors sense that policy performance is adequate, there is little need to rely on very specific or overly detailed policy matrices or policy conditionality. In such cases, a broad assessment of a country’s overall performance may do. Moreover, where performance is adequate, donors should no longer focus on reforms as the key objective – but rather on staying the course and on support of the overall expenditure programme. In other countries, the era of big reforms may be over, but policy-based lending may still be necessary in support of the much more protracted and complicated process of institutional change. In cases, finally, where budget support is considered for fragile states, it must be clearly tailored to the country’s capacity constraints.
Among the design criteria, perhaps the most controversial thought is that donors should not attempt to leverage policy reform through conditionality. In the past, there was a tendency of trying to use conditions to cajole countries further than they would have gone otherwise in undertaking policy reforms. Second, different programmes require different time horizons. It is not always possible to spell out the specific steps and the specific individual actions beforehand. Nonetheless, it is insufficient to just list normative indicators at the end. This is particularly true for areas such as institutional governance or public sector that are not easily measured with quantifiable indicators. In any case, these issues require a long-term perspective.
As the third design aspect, it may be advisable to go further in basing disbursements on actions that have taken place rather than on promises. Within the programmatic framework, however, it must become clear that expected prior actions (“triggers”) are truly only indicative. They are not meant as a substitute for conditionality, which needs to be done to the letter, but rather serve to assess the probable future extent of support for the reform policy.
As for implementation, any conditionality should be transparent and monitorable. Where possible, this implies involving parliaments and other stakeholders. It also requires spelling out expected results and monitoring actual results to determine whether progress is being made. This is not the same as making disbursements contingent on outcome indicators. In fact, it would make sense to use outcome indicators realistically and pragmatically. Many recipient countries are weary of being held responsible for outcomes that are not under their control. Moreover, there is often a lack of data or outcome measurements conducive to quantification. Donors should focus more on this topic, without necessarily becoming doctrinaire about it.
The notion of predictability remains an important implementation issue, especially with donors coordinating their efforts more closely. This is particularly so in the case of countries with neither entirely bad nor entirely good performacne. Such cases call for a graduated response that should not be based on a mechanistic approach, but rather use judgement. In cases of uneven or inadequate performance, donors need to be able to send signals without dramatically disrupting aid flows.
Most importantly, donors should rely on the recipient countries own procedures and systems. If a country, for instance, has an adequate system of budgetary reporting to its parliament and there is a viable medium-term expenditure framework built around its priorities, donors can – and should – rely on these structures much more heavily than they have done in the past. To the extent that government’s are accountable to their own citizens for implementing poverty reduction and improving service delivery, traditional conditionality can be replaced by a long-term partnership based on mututal accountability between recipient governments and international donors such as the World Bank.
Dr. Stefan Koeberle
is a World Bank manager of the Country Economics unit in the Operations Policy and Country Services Vice Presidency.
skoeberle@worldbank.org
|
