Thrive on competition

Great progress has been made in reducing global poverty over the last forty years, and official development assistance (ODA) from rich countries has helped. Nonetheless, the gap between rich and poor has grown wider, and policymakers in rich countries understand that more – and more effective – aid is needed. Today, hundreds of official agencies are trying to promote development. Forty-six governments run bilateral aid programmes, each of which is administered through multiple agencies. These countries also fund some 233 multilateral development agencies. Thousands of international NGOs, tens of thousands of developing country NGOs and hundreds of thousands of community-based organisations are also part of the effort.

Many-piece orchestra

The more players there are, the more difficult coordination becomes. Harmonisation is of special concern for European Union member states. They agree in principle that they need a common foreign and security policy – and, accordingly, a common development-assistance policy. EU members directly provided $ 49 billion of bilateral ODA in 2006 and, through multilateral bodies, another $ 19 billion. Two thirds of total aid thus come from Europe. In sub-Saharan Africa, EU countries contribute three-quarters of total ODA.

Unsurprisingly, calls for harmonisation have crescendoed in Europe. But to produce harmony, the aid industry does not so much require a gran maestro in Brussels. It needs sheet music – a coordinating system that provides solid information on who should play when and what, and how loud as well as more efficient instrumentalists, the aid agencies themselves.

Aid will not have the necessary transformational impact when spread too thinly. Although total aid in 2005 was over $ 100 billion, only $ 38 billion was for investment in development projects and programmes – and of this perhaps half actually got to the intended beneficiaries. On top of this, there are two additional problems. First, projects are generally getting smaller: Johnston and Manning (2005) counted more than 60,000 ongoing projects – triple the number in 1970 – of which 85 % cost less than $ 1 million. Second, ODA flows are becoming more uncertain and volatile.

Fragmentation puts at risk the relevance, efficacy, efficiency, and sustainability of aid. Since 1970, aid has kept fragmenting (Kharas 2007). Poor countries, particularly in sub-Saharan Africa, are those that suffer from the highest degree of aid fragmentation. And though one might think that fragmentation simply stems from more donors committing more resources, fragmentation is higher in countries with low levels of aid. Yet these are precisely the countries marked by weak local capacities.

Another symptom of low aid effectiveness is the volatility of aid flows over time. Year-to-year changes in aid receipts can have high costs as planning becomes difficult for finance and line ministries, long-term projects that require recurrent expenditures are forgone, health interventions suffer from intermittent treatments, and infrastructure gains are eroded without maintenance.

Like fragmentation, volatility has been rising. Ironically, development funding is among the most volatile sources of finance. The coefficient of variation of aid – a measure of volatility – is five to six times higher than the equivalent volatility in recipient countries‘ gross national incomes or tax revenues. Aid is thus unsuitable for funding recurrent programme costs. Aid cannot be relied on to pay teachers, health staff and other relevant professionals. In our estimates, high ODA volatility reduces effectiveness by around 20 %.

Given that aid flows are both fragmented and volatile – with both trends worsening – it is encouraging that donor governments are looking at options for systematic harmonisation. The OECD’s Paris Declaration on Aid Effectiveness of 2005 called for increased aid alignment and harmonisation through, among other things:
– simplification of disbursement and monitoring systems,
– more effective division of labour and
– inter-agency cooperation.

Dimensions of harmonisation

There is preliminary evidence that donors are making modest progress (World Bank 2007). It is evident, however, that more needs to be done, and Europe is assuming a role of leadership in these matters (see box). Only time will tell if the initiatives will lead to substantial progress, but it seems obvious that – difficult though it may be in practice – greater harmonisation within and through the EU can generate more efficiency.

Harmonisation has three distinct components:
– If aid agencies reduce the number of focus countries and sectors the number of visiting missions would decline – and so would bureaucratic overheads for countries dealing with multiple donors. Well-coordinated donors could also ensure that major development gaps do not emerge and that the overall effort is coherent. These benefits, however, must be weighed against significant transaction costs on the side of the donors.
– More important efficiency gains would result from aid agencies specialising in the fields they really have comparative advantages in. The current system does not guarantee that will happen. The assessment of agency strengths is still in its infancy. It is difficult to judge where each agency’s comparative advantage lies. Donors decide for themeselves which countries and sectors they will focus on. Their decisions are based on several factors, including non-economic ones. Today, it is hard for any donor government to judge whether it is more efficient to undertake a specific investment bilaterally, cofinance a multilateral programme (and which of the many that exist), or fund an NGO. Such decisions, however, determine how efficiently millions of dollars are spent.
– To reduce administrative burdens for recipient governments, donors should agree on common procurement rules, financial management practices as well as environmental and social safeguards. They should preferably use systems already in place in the target countries.

So far, measures to coordinate efforts are most advanced in the EU as well as among other donors. Unfortunately, this is the area of harmonisation where the net benefits are likely to be the lowest, given the considerable transaction costs of coordination. Progress towards specialisation based on agency efficiency and towards common systems, however, remains too slow.

That Europe needs harmonised outcomes does not mean Europe needs a single, unified and administratively managed coordination mechanism. Instead, the aid industry should provide donors and recipients with information to catalyzse efficiency and innovation in a way similar to how markets discipline firms.

Market-disciplined aid

Markets coordinate multiple firms and consumers efficiently, matching producers with purchasers. In allocating scarce development assistance, however, aid priorities are distorted and the supply of aid is inefficient.

Even governments cannot always claim to fully represent their countries’ needs, yet developing countries need ways to signal their preferences. Otherwise, distortions can be dramatic. A notorious example is Rwanda’s health sector. Thanks to support from the Global Fund for AIDS, TB and Malaria and other donors, it disposes of $ 48 million per year for HIV/AIDS matters, which affect only three percent of the population. On the other hand, Rwanda only spends $ 1 million per year on maternal and child health programmes. In terms of dollars spent per life saved, this is grossly inefficient, and it results from a lack of donor responsiveness rather than a lack of harmonisation.

At the same time, it is unclear who recipient countries can turn to for the most efficient help to overcome key problems. Donor programmes typically differ in many details, making it hard to actually compare their efficiency or effectiveness – so the supply of aid is inefficient. By contrast, constant competitive pressure makes firms in market systems dynamic and efficient.

The aid industry, however, lacks competition. No donor agency ever exits the system because of inefficiency or ineffectiveness, nor do the best donor agencies get additional resources to expand. Though it is widely believed that multilateral agencies have many advantages, they have been steadily losing market share to bilateral agencies and a few specialised new funds. Donor typically do not specialise on specific tasks, in which they might then excel. That is only slowly starting to change thanks to so-called vertical funds, which do focus on specific sectors. But outside of health and environment, there are few such funds. There is no fund for infrastructure, education or governance, to name only three crucially important areas.

Unlike in a market economy, where the firms which expand the fastest are typically the most profitable, there is no gauge of aid-agency effectiveness to guide donors. But this is precisely what would do most to improve aid effectiveness.

Conclusion

At the heart of the debate is how best to improve aid efficiency. Harmonisation, as currently practised, is based on like-mindedness among donors and a set of administrative rules that governs each agency’s behaviour. However, that approach may also dampen competition among aid agencies and reduce accountability. After all, there is safety in numbers. If all donors act in harmony, yet fail to achieve the goal of lasting poverty reduction, it becomes all too easy to point the finger of blame at the recipient country.

Today, donor agencies themselves feel no pain from their failures. But harmonisation designed to achieve agency specialisation based on comparative advantages could have a dramatic impact on aid efficiency. The aid industry today has more need for a common score sheet and effective instrumentalists than for a gran maestro.

In our view, three things are essential:
– More and reliable information is needed to assess agencies’ performance in particular sectors. There must be a way for policymakers in donor countries to judge how to allocate dollars sensibly to aid agencies. In order to compute costs and benefits of various agencies, more rigorous evaluation methods are required. Moreover, agencies should bear accountability for sectoral programmes, not just individual projects, just as ministries are not only responsible for their own actions, but for entire societal sectors. It certainly makes sense to let development agencies compete for funds.
– Aid recipients deserve mechanisms with which they can express their preferences. So far, only few development agencies systematically undertake client surveys; and where such work is done, it is not carried out in a systematic comparative fashion and the results have little bearing on actual programmes. Recipient countries cannot substitute aid from one partner for a programme offered by another donor without losing overall aid resources.
– Systems must be designed to bring interventions to scale. Many projects are successful experiments which do not have the expected impact because of a lack of scaling up. Without an express focus on scaling up, development interventions will remain patchy. One of the most robust development conclusions is that impact is maximised through sustained engagement, not through one-off projects (Hartmann and Linn 2007).

The EU has assumed a leading role in harmonisation. To become a more effective leader, however, the EU should now take the next step, translating ambitious policy documents into concrete actions to improve aid effectiveness.