International actions

Recommendations for the international system to support country-level processes

Each of these problems is significant. But each is also solvable through committed and specific actions on the side of development partners. Here are 10 key “to do's” for the donors.

The multilateral and bilateral development agencies and other relevant international institutions need to make explicit their support for MDG-based poverty reduction strategies.

Criticisms of aid come in many forms. Some critics charge that aid is inherently flawed because it strengthens governments, often corrupt governments, at the expense of the private sector. This is the famous argument of the late British economist Peter Bauer. Some charge that aid is not needed, since private saving and investment can and should be the backbone of economic growth. Some have taken the middle ground that aid works when it is channeled to well governed countries. This is the conclusion of the highly influential study by Burnside and Dollar (2000).

Our view, explained in the text, is that aid is most useful if channeled to the countries that truly need it (mainly those stuck in a poverty trap) and channeled to the right sectors (mainly infrastructure and human capital). It works best when delivered to well governed countries. And aid used to support public investment complements private saving and investment, rather than competing with private capital.

Many negative conclusions about the link between aid and economic growth have come from cross-country regressions of economic growth on aid volumes (and other variables). The volume of aid is often found to be statistically insignificant as a determinant of economic growth, leading some authors to conclude that “aid is ineffective” in promoting economic growth. An important weakness in such studies is that they tend to examine the links of growth to overall volumes of aid without paying attention to how the aid is actually delivered. Specifically, much aid comes in the form of technical assistance (for consultants from the donor country), administrative costs of running bilateral and multilateral agencies, and emergency food aid. It is not really surprising that such aid is not correlated with economic growth in the recipient country. Food aid, especially, is given in the midst of deep crises. So a regression of economic growth on food aid would tend to prove (erroneously) that aid causes output to decline, instead of the correct conclusion that an output decline (caused by drought, for example) causes emergency aid to rise!

In an important new study, Clemens, Radelet, and Bhavnani (2004) correct for this typical shortcoming by considering only aid volumes that effectively support investments and services on the ground in the recipient country, taking out emergency aid, technical assistance, and other kinds of aid that do not translate into growth-promoting investments and services. They find that aid, when measured properly, contributes significantly to economic growth. This suggests that aid is effective, if it is well targeted and administered as direct support for country-level investments. Of course, a minimum adequacy of governance is required for a country to be able to channel aid into investments.

Aid can and must be disbursed in ways that align the incentives of donors and recipients to support positive development outcomes. As this report argues, elements of a successful disbursement strategy include aid in the form of budget support for national poverty reduction strategies based on the Millennium Development Goals. While there have been real problems with the way that aid has been distributed in the past, governments in rich and poor countries alike are learning from their mistakes to design more effective ways of delivering financial assistance to those who need it most.

In sum, foreign aid can play a hugely positive part in growth and poverty reduction when properly targeted and administered toward vital infrastructure and human capital. This finding is underlined by the recent experience of Mozambique, Tanzania, and Uganda, which all experienced substantial social sector improvements financed largely through development assistance. Mozambique is a particular success story over the past decade, having averaged real per capita economic growth rates of 5 percent while receiving aid ranging from 20 percent to 60 percent of GNP every year since 1993.

Donor governments need to distinguish among countries so that aid is focused where it will make a difference, and so that donors do not shortchange the countries that need the most help by focusing on those with greater -geopolitical attention. Different types of support will be needed for middle-income countries, well governed poverty trap countries, and poorly governed poverty trap countries (box 9). Special attention should also go to conflict countries and countries with special needs, such as landlocked or small island -economies.

To address long-term development needs systematically, countries should produce an MDG needs assessment through 2015 and a corresponding 10-year policy framework. This framework should then guide the more detailed and shorter term MDG-based poverty reduction strategy.

Table 5: Estimated official development assistance for direct MDG support and MDG capacity building, 2002

Most middle-income countries can finance the Goals largely through their own resources, nonconcessional flows (market-based loans from the World Bank and regional development banks), and private capital flows. Donor efforts should be directed at helping these countries to eliminate the remaining “pockets of poverty.” Some middle-income countries also need further debt cancellation, especially on debts owed to creditor governments (Paris Club debt). The successful conclusion of the Doha Development Agenda of multilateral trade negotiations, with increased access to rich world markets, will bring benefits middle-income countries. Many middle-income countries, such as Brazil, China, and Malaysia, already are donor countries. We recommend that they and other successful poverty-reducing countries, such as India, step up their donor efforts, including financial contributions and technical training for low-income country partners.

Well governed poverty trap countries

For well governed countries caught in a poverty trap, even a significant increase in domestic resource mobilization will not be enough to achieve the Millennium Development Goals. Substantial co-financing through official development assistance is required, especially for Least Developed Countries, to scale up the needed investments in infrastructure, human capital, and public administration. The key for well governed poverty trap countries is to base aid on a true MDG-needs assessment, and then to ensure that aid is not the binding constraint to scaling up. These countries should be fast-tracked in 2005.

Poorly governed poverty trap countries: lack of volition

For countries like Belarus, Myanmar, the Democratic People's Republic of Korea, and Zimbabwe, where the problem is the will of the political leadership, there is little case for large-scale aid. Aid should be directed to humanitarian efforts or through NGOs that can ensure delivery of services on the ground. Any aid directed through the government should be conditional on significant improvements in human rights and economic policies.

Poorly governed poverty trap countries: weak public administration

When the volition exists in government leadership but public administration is poor because of a lack of sound public management, one key step is to invest in public-sector capacity. This will also raise the “absorptive capacity” for aid in later years. Donors should view the poor public administration as an investment opportunity, not a barrier to achieving the Goals. Early efforts should be directed at building the government's analytical and administrative capacity at national, regional, and local levels—and building the technical expertise at the grassroots level in health, education, agriculture, and infrastructure. We expect that these countries will significantly outperform current expectations. In many countries international expectations are low but the country's potential is very high if timely donor support and debt cancellation are brought to bear, and phased in over time.

Conflict countries

Countries in conflict, just out of conflict, or falling into conflict present urgent special cases for the international community. Rapid responses are essential. A delay in well targeted aid can mean the difference between a consolidated peace process and a resurgence of conflict. Aid should be targeted at ending the violence and restoring basic services, directed in a manner to ease tensions among competing groups. Carrots (offers of an expanding aid effort) generally are much more powerful than sticks (international sanctions) in such crisis countries as Haiti and Sudan. Yet sticks are more typically applied, with few lasting results.

Geopolitical priorities

Countries with geopolitical priority (such as Afghanistan and Iraq) have urgent needs, to be sure, but may take up a disproportionate share of donor funding and public attention. If the major donors are to devote substantial efforts to these countries, they must ensure that the efforts do not divert attention and financial resources from other worthy countries. Debt cancellation for Iraq, for example, without similar debt cancellation for Nigeria would be unjustified on grounds of equity, merit, and relative need.

Countries with special needs

Developing states with special needs include:

• Small island states (isolation, small markets, natural hazards).
• Landlocked states (isolation and high transport costs).
• Mountain states (isolation and high transport costs).
• States vulnerable to natural disasters.

The geographically isolated states require special investments in transport and communications—and geopolitical help to support regional cooperation and regional integration. Hazards are rising in frequency, intensity, and impact, and traditional ad hoc responses are too slow and underfinanced. Donors should establish special emergency funds for natural hazards (droughts, floods, pests, disease) and steady funds for long-term improvements to cope with disasters.

The multilateral and bilateral agencies should organize their technical efforts around supporting countries to develop and implement MDG-based poverty reduction strategies. The UN Development Group (UNDG) should guide country-based UN Country Teams in their MDG support, and the UN Development Assistance Framework should identify the specific ways in which the Country Team will support the government to achieve the Goals. We recommend that agency specialists be trained to complement their existing sector-specific knowledge with basic skills to support country-level budgeting processes. We also recommend the establishment of multi-agency, cross-sectoral- regional technical centers to support governments and UN country teams in developing, financing, and implementing MDG-based PRSs.

As the senior UN representative on the ground, the UN Resident Coordinator's office needs dramatic strengthening, both to coordinate among UN organizations through the UN Country Team and to manage a core technical staff to support the host government in developing and implementing the MDG-based poverty reduction strategy. The local representatives of the international financial institutions should work closely with the UN Country Team in support of the host-country poverty reduction programs. The UNDG at headquarters level should support the strengthening of the Resident Coordinator position.

ODA levels should be guided by the MDG needs assessment, rather than being picked for political reasons or on the basis of incremental budgeting, as is now the case. By partnering with local research institutes, the UN Millennium Project has undertaken the first ever bottom-up needs assessments of the -country-level investments required to achieve the Goals. Although these first estimates need to be refined through the real country-level processes we are recommending in this report, the results show that the total cost of investments in low-income countries is on the order of $70–$80 per capita per year in 2006, increasing to $120–$160 per capita per year in 2015 (see, for example, the results for Ghana in table 6 on page 56). Middle-income countries will generally be able to afford these investments on their own. But the low-income countries, even after they initiate a major increase in their resource mobilization, will require $40–$50 per capita in external finance in 2006, rising to $70–$100 in 2015. To ensure the sustainability of programs, development assistance should also cover recurrent costs (such as public sector salaries, operations and maintenance) in addition to capital costs.

“Debt sustainability” should be redefined as “the level of debt consistent with achieving the Millennium Development Goals,” arriving in 2015 without a new debt overhang. For many heavily indebted poor countries, this will require 100 percent debt cancellation. For many heavily indebted middle-income countries, this will require more debt relief than has been on offer. For some poor countries left off the heavily indebted poor countries (HIPC) list, such as Nigeria, meeting the Goals will require significant debt cancellation. A corollary for low-income countries is that current and future ODA should be grants rather than loans.

To empower national ownership of MDG-based strategies and to limit the transaction costs of providing financial support, bilateral donors should use simplifying coordination mechanisms—such as sector-wide approaches, direct budget support, and multilateral financing such as that through the European Development Fund and the International Development Association. They should also follow through urgently on the actions they set for themselves in the Rome harmonization agenda.

Development partners should support developing countries in promoting overlooked priorities such as maternal health, gender equality, and reproductive health, and addressing neglected public goods, including long-term scientific capacities, environmental management, regional integration, and cross-border infrastructure.

Donors should evaluate their development, foreign, and financial policies with specific reference to the Goals. Donors should subject themselves to at least the same standards of transparency and coherence as they expect of developing country governments. Some countries have made progress by initiating their self-evaluating Goal 8 reports, but independent technical groups should publish independent evaluations of donor policy impacts and donor coherence, with data made publicly available to permit re-analysis.