Addressing corruption in fragile states: what role for donors?

4. Evaluation phase

Explore a new aid architecture

As mentioned in the introduction, there are strong voices for turning around the conditionality debate: instead of providing aid to encourage reform, some argue that it should be given to countries that have already demonstrated a desire to implement key reforms. In the language of the principal-agent problem, donors should spend less time trying to write contracts that force an alignment of incentives and instead give more aid to countries that on their own demonstrate similar motivations and objectives. Some donors have begun to be more selective, including the World Bank in the allocation of its concessional IDA funds, some European donors in terms of providing budget support, and the US with its new Millennium Challenge Account.

In this section we will explore and elaborate on an option for a new aid architecture that tries to address the twin challenge of the lack of ownership by government and pressure by donors to disburse funds. The idea is to move towards ex-post conditionality, so that aid is used to finance reforms, rather than to buy them. In other words, disbursement would be based on evidence of progress in governance in the social sectors. This option is more likely to succeed in countries with some state capacity.

The challenge for development agencies now is to find an architecture that will facilitate an increase in aid flows in the medium term. The current hope rests with the concepts outlined earlier which focus on partnership, trust, harmonisation, ownership, dialogue, capacity building, and on checking for development-oriented outcomes. In Uganda, for example, it seems that it has not been possible to uphold the focus on outcomes as the donors have tried to fill the vacuum left by a regime that has moved from strong to weak ownership of the development process. Some would also argue that the process has not been genuine on the donor side either in recent years as the driving force has been to disburse with little regard for outcomes. Evidently the goal of the GoU has been to amass as much in aid monies as possible.

How then can we begin to conceptualise a situation where donor funds are used more effectively and where the government can be enticed to get back into the driving seat? The recipe we put forward emerges from the concern that there is too much focus on the grand statements, lofty plans and promises made by many governments in the difficult partnership category, stated intentions that are never implemented. We hold that it is not the amount of money allocated that is the problem in most countries but rather its illicit use. The problem is made worse by the lack of follow-up on the donor side. The toleration of corruption seems to be high as the development partners are disbursing as usual in the face of so many broken promises.

So is there some way we can reward results rather than promises and in the process minimise the leakages that we now have to endure in the many difficult partnerships? Dr Alf Morten Jerve of the Chr. Michelsen Institute sees the need to experiment with ways of financing development that are not linked to the yearly budget cycle of the agencies. Many have put forward the suggestion that funds be earmarked for regions, specific countries, sectors within countries, or even local level entities like a commune. The point is that the funds would not be removed if results took more time than stipulated to materialise. Conditionality is therefore off the table as funds would be there to fund national priorities.

Some will immediately argue that Uganda indeed has experience of such funding through the Poverty Action Fund (PAF), which was established in 1998 as a response to concerns about the accountability and reorientation of the budget, given additional resources from HIPC and other donors for pro-poor expenditure. The PAF has been instrumental in introducing mechanisms for promoting budget allocations to sectors with strong pro-poor perspectives. However, the Uganda Debt Network (UDN), which has been monitoring government expenditure through the PAF, has documented examples of corruption, major time lags between the release and receipt of PAF funds, and serious problems with the quality of the services delivered, which have compromised the potential impact of the PAF on poverty reduction.

However Dr Jerve, considers that minor modifications to the idea of a fund might make a big difference in terms of what we can expect to achieve. The recommendation is to set up a fund that releases resources to the recipient after results have been delivered. We would see funds being released based on the school building actually built in district A, that it has the agreed number of teachers teaching the agreed number of hours to a real number of students, in the first, second, third quarter etc. Likewise, funds would be released for the road leading to the village once it is actually there for all to see. This delayed payment from the fund would mean that the government would have to use its own domestic revenue as the seed money, and the government would carry the risk if the results never emerged. The National Fund money would thus be released based on performance and would only pay for parts that the government would incur.

Some will call this a form of post hoc conditionality that brings little new, and would rightly point out that this model is not equally suited to all forms of aid. It should therefore be noted that other types of fund would exist to support NGOs and other more limited commitments, and funding should be made available to improve government systems further in order to enable them to deliver the results that will be sought. Most importantly, support must be given to building capacity to receive budget support effectively if that is to be brought back in as a result of a proven record of not squandering donor and nationally generated funds.

We would be contradicting ourselves if the model did not allow for long-term democracy-building initiatives supporting drivers of change – a part of the longer-term democratisation process. Indeed, the suggested model would go some way towards restoring the damaged accountability relations between the government and the electorate as it would become very tempting for the population to vote out a government unable to access funds due to its corrupt behaviour. As the fund would grow due to the failure to deliver, the incentives to vote the regime out would increase.

Prepare for the if/when

Donors should also be “preparing the ground” for a future point at which the government is more willing and able to address corruption. One starting point may be to identify anti-corruption needs that are likely to emerge after an election, for example, and to plan follow-up action so that donors are ready with practical support for any new AC commitments that emerge.
For example, if the current regime faces allegations of grand corruption, demands are likely to arise rapidly for action on recovering stolen assets once political change takes place. Donor countries with major financial sectors may want to outline a contingency plan for how they will assist if/when these demands are made. Work on asset recovery, for example, could proceed well in advance of a regime change, since it mainly requires action between and within individual donor governments.
Under these circumstances it is important for the international community to be able to move quickly to support that newly found political will. Again, collective action allows donors to leverage their relative strengths (some may have more money available, others expertise). The Embassies and stations in fragile states must therefore be given funds and autonomy to act fast and concisely when things turn around.

[top]