3 Delivering on Commitments for Aid, Debt Relief, and Trade

International Monetary Fund
Published Date:
May 2006
  • ShareShare
Show Summary Details

The “year of development” sharpened the international community’s focus on aid and trade, providing an unprecedented opportunity to accelerate achievement of the Millennium Development Goals (MDGs). Reaffirming the Monterrey Consensus (United Nations 2002), donors promised a major expansion in official development assistance (ODA) and strongly backed efforts to enhance the quality of aid. They also agreed to provide wider and deeper debt relief to the poorest countries. While trade liberalization through unilateral reforms and regional agreements moved ahead, multilateral negotiations under the Doha process made only modest progress. By contrast, considerable momentum was built on aid for trade.

Donors are delivering more assistance, and the prospects for scaling up aid have brightened. At their summit in Gleneagles, the Group of Eight (G-8) leaders pledged to increase aid to Africa by $25 billion a year by 2010—more than doubling assistance to the region—and Development Assistance Committee (DAC) members have agreed to expand aid to all developing countries by about $50 billion. These commitments call for a much faster pace of ODA growth (when public budgets could be under pressure), introducing some uncertainty for future flows. Scaling up assistance and ensuring that more of the increment is available for pursuing the MDGs make it particularly important to strengthen monitoring of donor commitments and flows. Scaling up will also require more coordination among aid delivery channels—bilateral funds, multilateral funds, global funds, and private funds.

At the country level, a strengthened poverty reduction strategy (PRS) process and enhanced consultative group and roundtable (CG/RT) coordination mechanisms will be central to implementing the scaling up and results agenda. Drawing on the PRSs, enhanced CG/RT processes can provide a practical, mutual accountability framework for linking resources to results.

The Paris Declaration in March 2005 gave a boost to the aid effectiveness agenda. Building on the principles of ownership, alignment, harmonization, managing for results, and mutual accountability, the Paris Agenda specifies monitorable actions to improve aid quality. Broad-based support for this agenda has translated into progress at the global level, such as adoption of global targets (for 2010) for the 12 indicators in the Paris Declaration, and at the country level, such as customizing several indicators and targets to the country context. Much remains to be done, however, and vigorous implementation of the agenda is needed to deliver more effective development assistance.

Major progress was made in 2005 in extending and deepening debt relief to the poorest countries. The G-8 Proposal (now called the Multilateral Debt Relief Initiative) to cancel 100 percent of the debt that heavily indebted poor countries (HIPC) owe to the African Development Fund (AfDF), International Development Association (IDA), and the International Monetary Fund (IMF) will cut the external debt of these countries by another $50 billion. Nineteen countries have already received $3.4 billion in debt relief from the IMF under the Multilateral Debt Relief Initiative (MDRI); IDA’s Board of Executive Directors has approved MDRI financing and implementation modalities for IDA, and those for AfDF are due for consideration and approval by the AfDF Board of Executive Directors in April. The MDRI initiative will greatly reduce the debt burden indicators in HIPC and create fiscal space for growth and human development objectives. To realize the potential benefits of debt relief, recipient countries need to strengthen their public expenditure management. They also need to manage postrelief borrowing so as not to undermine long-term debt sustainability.

The MDRI commits donor countries to providing additional resources to ensure that the proposed debt forgiveness does not undermine the ability of the three multilaterals to continue to provide financial support to low-income countries—or the institutions’ overall financial integrity. IDA and the AfDF have established baselines for assessing the additionality of donor financing. Monitoring against these baselines is needed to avoid substitution between regular donor contributions and debt relief compensation and to improve mutual accountability.

The Doha Development Agenda has great potential to help reduce global poverty, but progress was modest at the Hong Kong ministerial meeting in December 2005. World Trade Organization (WTO) members face a challenge in concluding the Doha Round by the end of 2006: all must raise their sights to avoid losing a good opportunity to harness trade for growth. While multilateral liberalization struggled, bilateral and regional trade agreements (RTAs) continued to mushroom during 2004–5, and unilateral reform continued at a steady pace in several regions, reflecting reductions in tariffs for agricultural products entering high-and middle-income countries and liberalization by developing countries in manufacturing and agriculture. The level of protection in least developed countries (LDCs) remained essentially unchanged. Barriers to products originating in high- and middle-income countries experienced the greatest reduction.

Following high-level political endorsement at the G-8 meeting in Gleneagles, proposals for significantly increasing aid for trade were endorsed by the governors of the World Bank Group and the IMF at their annual meetings. The Hong Kong ministerial created a task force to operationalize aid for trade and a consultation process to identify ways of increasing resources. Meanwhile, donors are substantially increasing the resources for aid for trade, continuing a trend over the last few years. The next year promises to be critical, as initiatives launched in 2005 begin to bear fruit, but challenges remain in ensuring that the increased aid is both additional and effective.

Aid Volumes and Quality

Rising Aid Levels

Assistance from DAC countries was nearly $80 billion in 2004, up from $69 billion in 2003, and it rose to an estimated $106 billion in 2005, boosted by $19 billion in debt relief to Iraq and Nigeria (OECD 2006a, 2006c). While nominal net ODA has risen by more than 50 percent from 2001 to 2004, the increase measured in real terms (at constant exchange rates and prices) is more modest at 18 percent—an average annual growth rate of 5.6 percent. But 2004 did see a strengthening in the pace of the ODA trend that continued in 2005 (ODA increased by 31.4 percent in real terms in 2005), partly in response to special factors.1 Higher aid flows mirror an increase in DAC countries’ aid effort; ODA as a share of gross national income (GNI) climbed from 0.22 percent to 0.26 percent in 2001–4 and to 0.33 percent in 2005 (figure 3.1). Only five donors have ODA/GNI ratios at or above 0.7 percent.

FIGURE 3.1DAC members’ net ODA, 1990–2005, and prospects for 2006 and 2010

Source: OECD 2006a, 2006c.

Note: The data for 2005 are preliminary.

ODA from non-DAC donors that report flows to the DAC rose by 9 percent (in nominal terms) to $3.7 billion in 2004.2 Saudi Arabia continued to account for the largest share of assistance by this group. Other donors are beginning to emerge in importance, including Korea, Kuwait, Taiwan (China), and Turkey. New European Union (EU) members that are not DAC members are also beginning to provide larger volumes of aid.3 Elsewhere, major emerging market countries, such as China and the Russian Federation, are playing an increasing role in development assistance. Data on so-called South-South assistance are incomplete, however, making it difficult to obtain comprehensive information on South-South aid volumes and prospects.4,5

Private giving is increasing. Grants from nongovernmental organizations (NGOs) grew at a brisk pace in 2004, providing over $11 billion in assistance in 2004. Private giving surged in 2005 in response to a string of catastrophic natural disasters such as the Indian Ocean tsunami of December 2004 and the South Asia earthquake of October 2005. Global private giving for tsunami-related humanitarian relief was $5.1 billion or 38 percent of the $13.4 billion of total amounts pledged (Inderfurth and others 2005).

Rising ODA volumes offset the overall decline in official nonconcessional lending. Net nonconcessional lending by multilaterals slid further in 2004 because of less borrowing from the IMF and other multilaterals and because of prepayments (see chapter 4); large prepayments pulled down multilateral net lending in 2005. Nondebt private flows, by contrast, have shown a rising trend in recent years: foreign direct investment (FDI) flows rebounded in 2004 and continued to rise in 2005; inward remittances to developing countries continued to surge. However, for poor countries, especially in Sub-Saharan Africa, ODA continues to be the largest source of external financing.

Nearly half the increase in net ODA from 2001 to 2004 has been in the form of debt relief and technical cooperation, a quarter was for emergency assistance, and a tenth for flexible bilateral forms of financing (table 3.1).6,7 Although there was a shift toward flexible forms of financing in 2004, the broad pattern observed over the past few years is likely to continue in the near term as debt forgiveness for Iraq and Nigeria are reflected in ODA flows and as humanitarian relief efforts remain high.

TABLE 3.1Composition of net ODA: less reliance by donors on special-purpose grants in 2004(all levels are in constant 2004 US$ billions)
ODA level 2001ODA level 2003ODA level 2004Increase in ODA in 2004Increase in ODA from 2001 to 2004Distribution of the increase in ODA from 2001 to 2004 (in %)
DAC ODA67.475.179.54.412.1
DAC ODA by type
Special-purpose grants29.640.638.4-2.28.873
Debt forgiveness3.59.17.1-2.03.630
Technical cooperation17.019.718.8-0.91.915
Food aid + emergency relief5.
Administrative costs3.
Flexible bilateral ODA14.813.
Contributions to multilaterals23.021.425.13.82.117
Grants by NGOs8.710.911.40.52.6
Source: OECD DAC database.Note: Flexible ODA is DAC members’ ODA less special-purpose grants and contributions to multilaterals.
Source: OECD DAC database.Note: Flexible ODA is DAC members’ ODA less special-purpose grants and contributions to multilaterals.

Much of the increment in ODA (from all donors) reflects global and regional security concerns, with Iraq and Afghanistan accounting for over half of the increase in net ODA from 2001 to 2004 (figure 3.2). Among other large recipients are the Democratic Republic of Congo and Madagascar, with additional aid mostly reflecting debt relief.

FIGURE 3.2ODA increases concentrated in a few countries

Source: OECD DAC database.

Brighter Prospects for ODA?

At their summit in Gleneagles, G-8 leaders committed to increase aid to Africa by $25 billion a year by 2010, more than doubling assistance to the region. Coupled with pledges by the EU, DAC members have also agreed to expand aid to all developing countries by about $50 billion. These promises would raise the average share of ODA to GNI to 0.36 percent in 2010. In May 2005, the 15 DAC EU members set an intermediate target for their collective ODA/GNI of 0.56 percent for 2010 and revised upward their target for 2006 to 0.42 percent from 0.39 percent.8 They also reaffirmed their commitment to reach an ODA to GNI ratio of 0.7 percent by 2015. These pledges represent a $38 billion increase (in 2004 U.S. dollars) in ODA by 2010 for EU members that are DAC countries, a near doubling of assistance for this donor group. Several DAC members have also announced a timetable for achieving an ODA target of 0.7 percent of national income. If DAC countries’ commitments are delivered, ODA measured in 2004 prices will be around $100 billion in 2006 and will rise to about $128 billion by 2010.

To achieve these targets, ODA will need to grow at an accelerated pace. Overall, real ODA will need to grow 50 percent faster from 2004 to 2010 compared with the average annual growth rate from 2001 to 2004 (figure 3.3); for EU members that are DAC countries, the pace of growth will need to triple. Implementation of Paris Club debt agreements for Iraq and Nigeria boosted aid volumes in 2005, and debt forgiveness grants will remain high in 2006. Beyond the near term, more of the increase in ODA will represent a transfer of resources; perhaps raising difficulties for donors that deliver a large share of aid through debt relief. For several countries—such as Italy and Greece—the acceleration in the growth of ODA required to meet commitments will be much greater. A faster pace of growth of aid (when public budgets are under pressure) could introduce some uncertainty in medium-term aid volumes.

FIGURE 3.3Acceleration in ODA needed to meet commitments

Source: OECD DAC database.

Note: Prospects for ODA in 2006 and 2010 are based on DAC members’ announced commitments. Not all DAC members have made commitments beyond 2006.

Although much of the expansion in ODA has been delivered in the form of special purpose grants (table 3.1), beyond the near term, the proposed doubling of aid to Africa by 2010 and the completion of major debt workouts are expected to shift the pattern of aid delivery. Moreover, if donors deliver on commitments, the pattern of distribution of aid to poor countries could change as well.

Funding Development Assistance through Innovative Mechanisms

Innovative financing mechanisms could augment aid flows and development investment and improve the predictability and flexibility of aid. Several of these mechanisms are in early stages of implementation: the International Finance Facility for Immunization (IFFIm) is being established as a pilot IFF; a pilot advance market commitment (AMC) proposal is being developed; and an increasing number of countries are moving forward with airline departure tax legislation—one possible use for resulting revenues is the International Drug Purchase Facility (IDPF) proposed by France. These efforts will test key aspects of the proposed innovative mechanisms.

International Finance Facility and IFF for Immunization. The pilot IFFIm will channel funds pledged by France, Italy, Norway, Spain, Sweden, and the United Kingdom through the existing governance structure and country programs of the Global Alliance for Vaccines and Immunization (GAVI). Work to implement the IFFIm structure is ongoing, and a first bond issuance is expected in mid-2006. In addition, France and the United Kingdom plan to establish a joint working group to consider the implementation of a full IFF for health and education and partly funded by the airline departure tax.

Advance Market Commitments for vaccines. AMCs for vaccines could complement the IFFIm program to strengthen global immunization efforts. Under an AMC, donors would guarantee a set envelope of funding at a given price for a new vaccine that meets specified target requirements. G-8 finance ministers have agreed to consider a specific AMC pilot proposal in April.

Airline departure taxes. The proposed airline departure tax has gained steady support. France has passed legislation enabling collection of an airline departure levy, with revenues estimated at €200 million a year. Over a dozen countries have said they will implement the tax, and others plan to follow suit. The United Kingdom indicated its intention to use part of the revenue from its existing Air Passenger Duty to provide a long-term stream of finance to the IFFIm and the IFF. Many countries have welcomed France’s proposal to use departure tax revenues and other contributions to fund the IDPF, which would provide long-term, predictable finance to purchase drugs used to treat the big pandemics affecting the poorest countries and to lower the prices for these drugs.

Blending arrangements. The World Bank has reached out to the multilateral development banks and development partners to develop more specific proposals for the use of blending arrangements in gap and middle-income countries as part of a larger work program. In IDA countries, an ongoing pilot program provides for the use of blended grants (from trust funds) and IDA credits to achieve global or regional public goods in health projects.

Enhanced Monitoring of Donor Commitments

Prospects of significant additional amounts of aid raise the importance of better monitoring of commitments and flows. The purpose of monitoring will be to assess progress in implementing promises and to build momentum for emulating best practice. Equally importantly, monitoring can be a useful tool to better understand the scale of resources that will become available over the medium term, and how this scaling up will translate into availability of resources at the country level. By providing reliable information on resource availability at the country level over the next few years, monitoring can facilitate improvements in transparency and coordination of aid and can help improve its predictability. While recognizing the challenge of providing three-year forward projections on aid at the country level, donors have agreed that the DAC undertake an effort to collect such information.9,10

Scaling up will also require better coordination among aid delivery channels—bilateral funds, multilateral funds, global funds, and private funds. The growth of global programs and funds and the emergence of new bilateral and private donors are increasing aid delivery channels. Better coordination among donors will be essential to delivering aid effectively (box 3.1). For example, global funds need to support country-led strategies and priorities and not undermine the capacity of national authorities for coherent planning, financing, and service delivery. Likewise, bilateral donors need to shift toward delegated cooperation so as to make use of the comparative advantages of individual donors. The first step to achieving better coordination will be sharing of information on planned donor activities.

Delivering Aid More Effectively

Higher aid volumes need to be matched by improvements in the quality of aid. Several factors influence aid quality and, hence, its effectiveness. Chief among them are the harmonization and alignment of aid, the modality through which assistance is delivered, the allocation of aid, and the volatility and predictability of aid flows.

Progress on Harmonization, Alignment, and Managing for Results

Following the Paris High Level Forum last year, intensive work undertaken by the Working Party on Aid Effectiveness (WP-EFF)11 resulted in an agreement and adoption of global targets (for 2010) for 11 of the 12 indicators in the Paris Declaration.12 The preliminary baseline data suggest that the gaps to be bridged to reach the agreed global targets are quite large, indicating considerable scope for progress. For example, only 15 percent of donor missions are undertaken jointly with other donors, well below the 40 percent target set for this indicator, and only 9 percent of partner countries undertake mutual assessments of progress in implementing agreed commitments and more broadly their development partnerships, against a target of 100 percent (figure 3.4).

FIGURE 3.4Indicators of progress: gaps between baselines (preliminary) and targets

Source: OECD 2005.

Note: Comprehensive baselines will be established in 2006.

A distinct feature of the Paris Declaration is a mutual commitment undertaken by partner and donor countries to an international monitoring process. To advance this process, the WP-EFF has created a subgroup dedicated to work on developing technical guidance, survey instruments, and a methodology for collecting and reporting data. This work will draw on and be supplemented with available information from the DAC peer review mechanism, the New Partnership for Africa’s Development (NEPAD), Strategic Partnership with Africa (SPA) surveys, and WP-EFF subgroups for public financial management and procurement.13

Progress in implementing the Paris Framework at the country level has been mixed. In a few countries, such as Mozambique, Tanzania, Uganda, and Vietnam, governments and development partners have made solid progress; this progress did not happen overnight but has been evolving over a number of years. Evidence of this progress can be seen in government-led efforts to address harmonization and alignment issues in program-based approaches (table 3.2). Four factors that appear common to this group of countries are strong government leadership in setting out priorities on development assistance, an effective implementation process for poverty reduction strategy, a mature government-donor aid relationship, and a well-functioning aid coordination mechanism focused on concretely addressing difficult operational issues. In other countries, such as Bangladesh, Burkino Faso, Cambodia, Ethiopia, and Ghana, there has been accelerating moderate and concrete progress, while in countries such as the Kyrgyz Republic, Malawi, and Senegal, there is progress, but it is still limited and does not yet cover a broad front. In countries not reflected in the table, harmonization and alignment actions are generally less advanced.

TABLE 3.2Country-level progress on selected harmonization and alignment actions
Harmonization road mapJoint/collaborative assistance strategiesCoordinated country-level monitoring and evaluationCoordinated budget supportSectorwide approachesJoint analytic workCountry-specific efforts to harmonize procurement and financial managementIndependent monitoring process
Substantive progressa
Moderate progressb
Burkina Faso
Progress, but limitedc
Kyrgyz Rep.
Source: World Bank desk surveys.

Countries showing substantive progress (at least 4 black dots).

Countries where there is progress but not across a broad front (at least two black dots).

Countries where actions are being taken but progress is limited.

Note: denotes substantive action; denotes moderate action; denotes little or no action.
Source: World Bank desk surveys.

Countries showing substantive progress (at least 4 black dots).

Countries where there is progress but not across a broad front (at least two black dots).

Countries where actions are being taken but progress is limited.

Note: denotes substantive action; denotes moderate action; denotes little or no action.

BOX 3.1Africa Action Plan: an opportunity to coordinate aid flows to Africa

The implementation of the World Bank’s Africa Action Plan will create opportunities to coordinate aid flows in Africa, ensuring that donor allocations reflect country priorities. Building on the IDA 14 base, the Bank will partner with others to increase both the volume and effectiveness of resources directed at Africa’s development.

One example of this enhanced coordination is the Africa Catalytic Growth Fund, which will provide a new pilot mechanism to complement IDA resources while maintaining country ownership and integrated expenditure management systems supported by IDA. The UK government has pledged £ 200 million in funds. The first call for proposals under this fund was launched in March 2006.

The Bank is also collaborating with other partners, including the EU, the U.K. Department for International Development, France, the African Development Bank, and the government of Japan. In meetings with the Japan Bank for International Cooperation, a number of projects in transport, power, water, and urban sector were singled out for collaboration and potential cofinancing.

Over the past year countries and donors have made the most notable progress in five areas:

  • Customizing Paris indicators. Examples are Ghana, Nicaragua, Uganda, and Vietnam.
  • Collaborative or joint assistance strategies. In Nigeria the assistance strategy was undertaken jointly by DFID and the World Bank. In Cambodia, the Asian Development Bank, the U.K. Department for International Development (DFID), UN agencies, and the World Bank collaborated on the assistance strategy. In Uganda, a strategy was developed jointly with about 10 bilateral and multilateral partners.
  • Budget support. The 2005 budget support survey by the SPA found that 28 percent of total aid to 14 countries is being provided as budget support, compared with 26 percent in 2004; 61 percent of these programs have made a multiyear commitment, with an average of 3.1 years; and 87 percent of such aid committed in 2005 was disbursed during the fiscal year when it was scheduled, compared with 75 percent in previous years.14 Moreover, the surveyed governments were of the view that memorandums of understanding underpinning budget support operations have served to reduce transaction costs. Results from the SPA survey also suggest an improving trend in recipients’ perception of donor behavior (figure 3.5). The survey responses show, for the second year in a row, an increase in satisfaction overall and in all the areas of donor practice. The survey responses indicate that in 2004–5, donor practices with lower scores tended to improve more.
  • Sectorwide approaches. Sectorwide approaches (SWAPs), which initially focused on the health and education sectors, are now used in the areas of infrastructure, roads, water, agriculture/rural development, and justice in some countries. Recent SWAPs increasingly involve multidonor pooling, use of common procedures, and reliance on country systems.
  • Joint analytic work. Increasingly, donors are undertaking analytic work jointly; for example, 2005 calendar year data from the World Bank indicate that 22 percent of its major economic and sector work was conducted with other partners (see chapter 4).15

FIGURE 3.5Satisfaction improving with donor practices, 2003–5

Source: SPA Budget Support Survey 2005.

The encouraging developments at the country, regional, and global levels do not provide grounds for complacency as there remain many obstacles to further broadening and deepening implementation. In donor countries, political factors as well as incentive structures within donor agencies are critical to progress. Politicians and civil society—especially NGOs and private groups, who could be affected by changes in aid modalities—might not attach the same commitment to harmonization and alignment as senior managers of aid agencies (de Renzio 2006). Thus, senior managers of donor agencies need to work with politicians and civil society to build broad and effective support for moving forward on the Paris framework. At the same time, the internal incentive systems—both at the institutional and individual level—need to be compatible with a sharpened focus on harmonization, alignment, and results (ODI 2004).16 Although aid agencies are making progress on improving their internal incentive systems, more work is needed.

Enhancing Aid Quality: Aid Modality, Allocation, and Predictability

Improving aid modalities. The modality (instrument) through which donors deliver aid matters. Donors tend to use a range of modalities; the choice of modality typically is influenced by political and other constraints on the donor side and by partner country circumstances. One issue that arises is whether the effectiveness of aid can be enhanced within current aid modalities. This is particularly so for technical cooperation (TC), which continues to be a key donor tool for supporting capacity building, although there are issues surrounding its effectiveness.17 The DAC estimates “free-standing” TC (that is, cooperation aimed at capacity building and not related to investment) to be nearly $20 billion, or about a quarter of total net ODA. TC to Africa was around $4.5 billion in 2004, nearly a fifth of total assistance to the region.

To better support sustainable capacity development, technical cooperation needs to be aligned with the newer paradigm for capacity building (World Bank 2005a). This will require recipients to have more ownership and control over TC, so that assistance can be provided where it is really needed. In addition, the focus of technical assistance should be on building capacity at the level of institutions, organizations, and individuals. Mainstreaming technical assistance in programs and projects can also enhance its effectiveness. Finally, untying TC and providing it as budget support could yield savings through the competitive hiring of experts.18

Although bilateral food aid is a small component of total aid (about 5 percent), it continues to be an important tool for providing emergency assistance and for addressing hunger and malnutrition.19 Most food aid is tied, which raises the issue of whether untying can lower the cost of providing assistance for food. A recent Organisation for Economic Co-operation and Development (OECD) study finds that providing direct transfers in kind are at least 30 percent more costly for donors than unrestricted financial assistance (OECD 2005c). Nor is tied food aid cost-effective for the recipient country: transfer of food in kind was found to be about 50 percent more costly than locally procured food and 33 percent more costly than food imports from a third country.20 It appears that financial assistance or more flexible sourcing is preferable, except where local procurement might not always be an option, especially in areas with food shortages, or where well-functioning internal markets are lacking, and where weak trade linkages could hamper imports from third countries (also see box 3.2).

Improving allocations. Although the causal mechanisms through which aid has an impact on poverty reduction are not settled in the literature, it would appear that to achieve the MDGs, aid should be targeted to countries with poorer populations and governments committed to poverty reduction. Aid allocation patterns suggest that donors are paying more attention to poverty and to quality of recipients’ policies and institutions. Results from the selectivity model of Dollar and Levin (2004) indicate that in 2004 over two-thirds of bilateral donors had a significant relationship between aid and poverty and that the poverty elasticity of aid had strengthened for most of these donors. The results for policy elasticity of aid likewise show a strengthening of the relationship between aid and the quality of policies and institutions (World Bank’s Country Policy and Institutional Assessment); although some large bilateral donors are still not very selective (figure 3.6). Overall, these elasticities are much higher for multilaterals than bilaterals. Donors favor other criteria in aid allocation as well, including geostrategic considerations, vulnerability to shocks, and former colonial ties (Amprou, Guillaumont, and Guillaumont-Jeanneney 2005).

FIGURE 3.6Strengthening trend in donors’ poverty and policy focus

Source: Staff estimates.

Fragile states. Fragile states present a special challenge for the donor community: Global Monitoring Report 2005 highlighted the issue of “aid orphans” receiving much less aid and “aid darlings” receiving much more aid than can be explained by policy and poverty criteria. A set of principles for international engagement in fragile states gained broad support among donors in 2005. These principles are now being piloted in nine countries.21 Early reports suggest that the principles strongly complement the Paris Declaration, reinforcing messages on coordination—and, by extending the framework to include nonaid actors such as security and diplomatic actors—the pilots are helping to focus attention on some of the most difficult coordination challenges in fragile states. The pilots are also helping to focus attention on some of the more complex concepts identified in the principles, including the operational implications of the state-building objective. Experience from these pilots will be used to illuminate areas for further policy and operations work.22

BOX 3.2Measuring the “quality of aid”

Aid “quality” refers to the form and modalities of aid that make it more effective as a resource for advancing development objectives. One approach is that of the Center for Global Development, which aims to measure aid quality of donors in its Commitment to Development Index (CDI). The CDI adjusts for “selectivity” of aid flows to poorer and better-governed countries, for tax incentives that encourage charitable giving, and for discounts tied to aid and small average project size (Roodman 2005). The CDI methodology is under review for improvement. From the recipients’ perspective, important aspects of aid quality include fragmentation across donor programs and aid volatility. Some of these aspects are to be monitored under the Paris Declaration on Aid Effectiveness, which may contribute to strengthening the future monitoring of aid quality.

Emerging experience points to the dynamic nature of fragile state environments, implying that strategies should aim to look ahead at the direction of reform dynamics, avoiding an oversimplistic design of assistance based on past performance.23 Experience has also shown that careful attention to prioritization, harmonization, and results is critical in fragile states, particularly in situations of fast transition, which often suffer from fragmentation and a multiplicity of actors. The newly created UN Peacebuilding Commission will bring together key international actors to marshal resources and advise on postconflict peace-building and recovery strategies, focus attention on reconstruction and institution-building efforts to help lay a solid foundation for sustained development, improve coordination among actors (both within and outside the United Nations), develop best practices, and help ensure predictable financing for early recovery work.24

Improving the predictability of aid. The prospect of higher volumes and a shift from projects to general budget support could increase the volatility of aid (Bulir and Hamann 2005), even as ramping up and maintaining scaled up service delivery call for greater predictability in resource flows. A recent study (Gelb and Eifert 2005) shows that although predictability poses a special challenge for budget support, there are practical ways of addressing this issue. For example, applying performance-based allocation rules with a flexible precommitment rule, such that aid levels adjust sharply only in response to major performance changes, can allow for precommitment of aid in a multi-year framework while avoiding drawn-out periods of misallocations. In addition, donors could fund a country’s reserve holdings—that is, a buffer reserve fund of two to four months of imports—so that the reserves could cushion a shortfall in disbursements arising from exogenous factors unrelated to country performance.

Issues in Managing a Scaling Up of Aid

With donors planning to boost aid to poor countries, the economic impact on recipients of substantially higher flows needs to be addressed. Some of the issues that need to be considered are the macroeconomic impact of higher aid in terms of appreciation of the real exchange rate and consequences for external competitiveness; management of monetary and fiscal policy; budgetary management issues of higher and possibly more volatile aid flows; the implications of scaled-up aid for medium-term expenditure frameworks and the composition of public investment; and identification and management of the institutional, organizational, and skills constraints to scaling up aid (Heller 2005). Several of these issues, particularly absorptive capacity constraints and the need for appropriate sequencing of investment, were dealt with in detail in Global Monitoring Report 2005.25 A discussion of macroeconomic management issues is presented in box 3.3.

At the country level, scaling up must be anchored in a strengthened Poverty Reduction Strategy process: ambitious country-owned and -led development plans linked to medium-term expenditure frameworks and focused on results. Drawing on the PRSs, the consultative group and roundtable processes can be used to identify the scope of scaling up on a country-by-country basis. The DAC and the World Bank are collaborating to enhance the CG-RT processes to provide a practical, mutual accountability-based framework for linking resources to results (OECD 2006b). The Bank’s Africa Action Plan, as well as the recently approved Catalytic Fund, will promote the formation of results and resources consultative group meetings so as sharpen the results focus of alignment and harmonization efforts at the country level, and to improve the predictability of aid (World Bank 2005c). The framework for enhancing the CG-RT processes would entail various tasks, including specifying MDG-based results that are to be targeted and the actions by partner governments to achieve these results, identifying capacity constraints and the resources needed to build capacity, obtaining donor commitments for predictable and flexible financing consistent with the Paris Declaration targets, and monitoring progress on results and commitments of aid donors and partners.

Progress on Debt Relief for the Poorest Countries

The past year saw major progress in extending and deepening debt relief to the poorest countries. In June 2005, the G-8 proposed that three multilateral institutions—the AfDF, IDA, and the IMF—cancel 100 percent of their debt claims on countries that have reached, or will eventually reach, the completion point under the HIPC Initiative, thereby freeing up additional resources to help these countries reach the MDGs. The G-8 Proposal, now called the Multilateral Debt Relief Initiative, will deepen the debt relief provided under the HIPC Initiative by further cutting the debt of poor countries by about $50 billion (in nominal terms).

Progress on the HIPC Initiative

Overall, substantial progress has been made in the implementation of the HIPC Initiative. As of March 2006, 29 HIPC have reached the decision point and are receiving debt relief; these countries account for about two-thirds of the initiative’s total expected debt relief in net present value terms. Progress toward reaching the completion point—when creditors provide the full amount of debt relief committed at the decision point on an irrevocable basis—continued in 2005; three additional countries reached the completion point, bringing the total number of countries to have done so to 18. Most of the 11 countries in the interim period between their decision point and completion point are on track with respect to their macroeconomic programs; others that experienced difficulties in program implementation are pursuing the necessary policy measures to bring their economic programs back on track (IMF-World Bank 2005).

In addition, a list of countries meeting the enhanced HIPC Initiative’s income and indebtedness criteria at end-2004 was approved in April 2006. The Boards of the IDA and the IMF had decided, in September 2004, to extend the sunset clause of the HIPC Initiative to end-2006 and to ring fence its application to countries satisfying the enhanced HIPC Initiative income and indebtedness criteria using end-2004 debt data. The list of countries that meet these two criteria and might wish to be considered for debt relief under the Initiative includes seven countries previously identified as HIPC plus four new countries. Three additional countries that meet the required criteria indicated that they do not wish to avail themselves of the Initiative.

BOX 3.3Macroeconomic management of surges in aid

Meeting the MDGs would require a substantial increase in resource inflows to low-income countries, implying the need for careful attention to macroeconomic management, particularly coordination of fiscal, monetary, and exchange rate policy in response to an aid surge.

Two related but distinct responses to aid increases—absorption and spending—help to analyze macro management options (IMF 2005). Absorption refers to the increase in net imports associated with an increase in aid, whereas spending refers to the widening of the fiscal deficit (net of aid) associated with that increase.

To absorb and spend is the textbook response to aid; the government increases investment, and aid finances the resulting rise in net imports. Even if the government spending is on domestic goods, the aid allows the resulting higher aggregate demand and spending to increase net imports without creating a balance-of-payments problem. Some real exchange rate appreciation may be necessary to enable this reallocation of resources.

In general, spending and absorbing aid is the only sensible long-run alternative to forgoing aid altogether. However, other responses to incremental aid may be justified under some circumstances and for a limited period.

To save incremental aid—that is, to neither absorb nor spend—may be a good way to build up international reserves from a low level or smooth volatile aid flows.

To absorb, but not spend, substitutes aid for domestic financing of the government deficit. Where the initial level of domestically financed deficit spending is too high, this can help stabilize the economy. Alternatively, it can reduce the level of public debt outstanding, crowding in the private sector. To spend and not absorb is a common but problematic response and usually reflects inadequate coordination of monetary and fiscal policies. This response is similar to a fiscal stimulus in the absence of aid. The aid goes to reserves, so the increase in government spending must be financed by the printing of money (which generates inflationary pressures) or by government borrowing from the private sector (which crowds out private investment). There is no real resource transfer from abroad, given the absence of an increase in net imports.

Dutch disease—the crowding out of exports through real exchange rate appreciation—is often seen as an unfortunate byproduct of aid. A permanent increase in the level of aid to a country may lead to some degree of real appreciation of the exchange rate. This appreciation draws domestic resources from the production of traded goods to the production of schools, hospitals, and other infrastructure deemed crucial to development. However, if aid-financed spending generates a strong supply response—for example, eases critical infrastructure bottlenecks to agriculture and manufacturing—and stimulates dynamic externalities, Dutch disease may be avoided. In both cases, aid finances the rise in net imports that comprise the counterpart to the reallocation of resources. In essence, this is the spend-and-absorb case discussed above.

In the long run, therefore, aid surges may well lead to some degree of real appreciation. Recent work suggests that this may have a significant adverse impact on labor-intensive and exportable manufacturing industries (Rajan and Subramanian 2005a and 2005b). It is therefore crucial that aid-financed investments be targeted so that the resulting gains in productivity more than outweigh any loss of export competitiveness. Moreover, the pace of aid absorption and spending could be adjusted if Dutch disease concerns are important and aid-generated increases in productivity are slow to materialize. In the short run, a country’s circumstances should dictate the macroeconomic responses to a surge in aid inflows: weak investment opportunities or very low international reserves could indicate that aid should be temporarily saved, but over the long term aid should be both spent and absorbed.

Debt relief under the HIPC Initiative is projected to substantially lower debt and debt service ratios for most HIPC that have reached the decision point. Net present value (NPV) of debt stocks in the 29 HIPC that reached the decision point by March 2006 are projected to decline by about two-thirds once they reach their respective completion points. The ratio of debt service to exports and fiscal revenues for these countries is estimated to have declined from an average of about 16 and 24 percent in 1998–9 to 7 and 12 percent in 2005, respectively. These ratios are projected to decline further to less than half of the 1998–9 average by 2006.

For these 29 countries, poverty-reducing expenditures on average have risen from about 6 percent of GDP in 1999 to 9 percent of GDP in 2005, a level more than four times that spent on debt service (figure 3.7).26 In absolute terms, poverty-reducing spending is estimated to have increased from about $6 billion in 1999 to $13 billion in 2005, and is projected to increase to $15 billion in 2006.27

FIGURE 3.7Lower debt service, higher poverty-reducing expenditures, 1999–2006

Source: IMF-World Bank 2005 and staff estimates.

Note: Data for 2005 are preliminary; data for 2006 are projections.

Creditor participation under the HIPC Initiative remains an issue, in particular with respect to non–Paris Club bilateral and commercial creditors. Although commercial creditors account for less than 4 percent of the total debt relief due under the initiative, most have not provided their share. Moreover, the share of relief by non–Paris Club creditors has declined, as a number of them have withdrawn their participation from the initiative because of restrictive legislation, weak debt and asset management, limited understanding of the HIPC methodology, or lack of communication with debtors. In addition, the number of lawsuits initiated by commercial creditors against HIPC has increased. Moral suasion remains the principal measure for encouraging participation and discouraging litigation by remaining commercial creditors. The Debt Reduction Facility (DRF) for IDA-only countries has been an important instrument in reducing commercial debt owed by HIPC.28

Augmenting Debt Relief: The MDRI

The MDRI deepens the debt relief provided by the HIPC Initiative by canceling the debts of HIPC to the AfDF, IDA, and the IMF, and it provides dollar-for-dollar compensation for the costs of the MDRI debt relief for the AfDF and IDA. Through the MDRI, these countries will receive a further reduction in debt of around $50 billion—an average annual debt flow relief of over $1 billion over the next decade and close to $2 billion per year in the following decade.

Key features of the MDRI. The MDRI cancels 100 percent of the debt claims of the AfDF, IDA, and the IMF on countries that have reached, or will eventually reach, the completion point under the HIPC Initiative. Unlike the HIPC Initiative, the MDRI does not propose any parallel debt relief on the part of official bilateral or private creditors, or of multilateral institutions beyond the AfDF, IDA, and the IMF. Although the MDRI is an initiative common to three international financial institutions, the decision to grant debt relief is ultimately the separate responsibility of each institution, and the approach to coverage and implementation varies. There are three major variables when determining eligibility for debt relief under the MDRI. These variables include the cutoff date of eligible debt stock, the credit coverage of the debt to be cancelled, and the group of countries to be covered under the MDRI (table 3.3).

TABLE 3.3Key features of the MDRI by institution
Implementation dateJanuary 1, 2006July 1, 2006January 1, 2006
Cutoff date for debt reliefEnd-2004End-2003End-2004
Country coverageAll HIPC plus all non-HIPC with per-capita income of US$380 a year or less—Cambodia and TajikistanAll HIPCAll HIPC in Africa
Timing of delivery of debt reliefFor postcompletion point HIPC (plus eligible non-HIPC), once they meet the qualification criteria For precompletion point HIPC, once they reach the completion pointFor postcompletion point HIPC, once they meet the qualification criteria For precompletion point HIPC, once they reach completion point, at the beginning of a quarter following confirmation of eligibility by the Board of Executive DirectorsFor postcompletion point HIPC, once they meet the qualification criteria For precompletion point HIPC, once they reach the completion point
Modality of debt reliefStock cancellationStock cancellation and adjustment of gross assistance flows by amount forgivenStock cancellation and adjustment of gross assistance flows by amount forgiven
Source: AfDF, IMF, and World Bank staff.
Source: AfDF, IMF, and World Bank staff.

The MDRI implementation date is the beginning of 2006 for the AfDF and the IMF and mid-2006 for IDA; the actual delivery of debt relief occurs only when countries have been confirmed to qualify for such relief. Qualification for MDRI relief by HIPC that have already reached the completion point is contingent on satisfactory performance since the completion point in three key areas: macro-economic performance, implementation of a poverty reduction strategy detailed in a Poverty Reduction Strategy Paper (PRSP) or a similar framework, and public expenditure management systems.29 HIPC that have not reached the completion point will automatically qualify for MDRI relief once they reach the completion point.30

Twenty countries were assessed in 2005 and 19 were found to satisfy all the criteria and hence qualify for debt relief under the MDRI from the IMF.31 Debt relief amounting to special drawing right (SDR) 2.3 billion (about US$3.4 billion) was delivered to these 19 countries in early January 2006.32 Debt relief from IDA is expected beginning this July for 17 completion point countries.33 Debt relief from AfDF is expected once donors finalize the MDRI implementation modalities for the AfDF.

Additionality of donor financing—establishing baselines for donor contributions. The MDRI also commits to providing additional resources to ensure that the proposed debt forgiveness does not undermine the institutions’ overall financial integrity or ability to continue to provide financial support to low-income countries. The AfDF’s estimated cost from debt cancellation is $9.06 billion (UA 5.84 billion) in nominal terms; the costs to IDA from the MDRI are about $37 billion (SDR 24.8 billion) in nominal terms;34 and the estimated cost of full debt relief for the Fund is around $5 billion (SDR 3.5 billion) in end-2005 NPV terms, of which about SDR 1.3 billion is already being financed through the HIPC Initiative.35,36

Actual additionality of donor financing is required to ensure that IDA countries will benefit from the MDRI. There will be no additionality if donors’ replacement resources to cover forgone reflows of IDA and AfDF were merely deducted from the regular financial support of these institutions’ donors. Thus, IDA and AfDF have established baselines on which the additionality of donor financing can be assessed:

Baseline for IDA contributions

  • The contribution baseline has been set with reference to IDA14 regular contributions and assuming an inflation rate of 2 percent per year for the SDR basket of currencies, regular contributions in IDA15 would increase by 6.12 percent over each donor’s regular contribution to IDA14 in SDR terms.37 This would lead to an aggregate contribution baseline for regular contributions of around $16.3 billion (SDR 10.817 billion) in IDA15 (table 3.4).38
  • Compensatory financing of IDA’s forgone credit reflows due to the MDRI will be additional to this contribution baseline. In addition, the financing framework of future replenishments will also include special financing items, namely, compensation for IDA’s HIPC-related costs and financing of forgone principal reflows due to IDA grants.
  • The agreed contribution baseline would be indicative in nature and intended to demonstrate transparently the additionality of debt relief financing. The level of contributions, and therefore the size of successive IDA replenishments, will, as now, remain the sovereign decision of each government in the IDA donor community.39 The agreed baseline for additionality will provide a basis for mutual accountability among donors, and an important public signal of the commitment to avoid substitution of regular donor contributions for debt relief compensation.
TABLE 3.4Indicative donor commitments to IDA and AfDF over the next decade(Baseline: constant regular donor contributions in real terms—US$ billions)
IDA 14

IDA 15

IDA 16




Regular donor contributions
In current replenishment (net of supplemental, incentive, accelerated funds)14.93.59
In future replenishments (constant in real terms)a16.317.83.814.04
Special financing commitments of donors2.64.45.6
of which: Financing of MDRI costsb0.
Total indicative financing commitments of donors17.520.723.43.714.124.57
Source: AfDF and IDA staff estimates.

For IDA, based on regular donor contributions in IDA14, at the IDA14 foreign exchange reference rates, using a 3% per year USD deflator (equivalent to a 2% per year deflator in SDR terms). For AfDF, based on regular donor contributions in AfDF-X, at the AfDF-X foreign exchange reference rates, using a 3% per year USD deflator (equivalent to a 2% per year deflator in UA=SDR terms).

For IDA, based on disbursed and outstanding credits; cut-off date of 12/31/2003 and implementation date of 07/01/2006; includes 38 HIPC and 4 countries potentially eligible under the HIPC Sunset Clause. For AfDF, based on disbursed and outstanding credits at end-December 2004 as cut-off date, and implementation date of 01/01/2006; includes 32 HIPC and one country potentially eligible under the HIPC Sunset Clause.

Source: AfDF and IDA staff estimates.

For IDA, based on regular donor contributions in IDA14, at the IDA14 foreign exchange reference rates, using a 3% per year USD deflator (equivalent to a 2% per year deflator in SDR terms). For AfDF, based on regular donor contributions in AfDF-X, at the AfDF-X foreign exchange reference rates, using a 3% per year USD deflator (equivalent to a 2% per year deflator in UA=SDR terms).

For IDA, based on disbursed and outstanding credits; cut-off date of 12/31/2003 and implementation date of 07/01/2006; includes 38 HIPC and 4 countries potentially eligible under the HIPC Sunset Clause. For AfDF, based on disbursed and outstanding credits at end-December 2004 as cut-off date, and implementation date of 01/01/2006; includes 32 HIPC and one country potentially eligible under the HIPC Sunset Clause.

Baseline for AfDF contributions. To establish the additionality of donor replacement funds beyond AfDF-X, an inflation rate of 2 percent per year is assumed such that the aggregate nominal donor contributions of UA 2.454 billion in AfDF-X would increase at the compound rate of 6.12 percent to UA 2.604 billion in AfDF-XI.40

Impact on IDA/AfDF assistance. The MDRI will affect gross assistance flows from IDA and AfDF through a two-step process. First, 100 percent stock cancellation will be delivered by relieving eligible countries of repayment obligations and by adjusting their gross assistance flows by an equivalent amount. This feature helps allay moral hazard and equity concerns associated with debt cancellation. Second, additional resources provided to these institutions by the donors (to compensate for the forgone debt service from the country) will be reallocated to IDA-only and AfDF-only countries through the existing performance-based allocation systems of IDA and AfDF. This feature helps strengthen the link between resource transfers and country performance levels.

As a result of debt relief, total IDA assistance flows—the sum of new IDA commitments and forgone reflows—would be expected to remain at least intact for all beneficiary countries. Therefore, in terms of net resource transfers, no MDRI-recipient country would be expected to be worse off following debt relief. However, new IDA commitments may decline over time for some countries receiving debt relief, depending on the difference between future annual performance-based allocations they receive from IDA and forgone debt service at each period. Longer-term projections of future performance-based allocations are, however, subject to a considerable degree of uncertainty. A country’s allocation in any given year will depend on a number of factors, including its performance relative to other countries; whether it is eligible for IDA grants and, consequently, subject to a volume discount under IDA’s grant allocation system; the size of IDA’s overall available resource envelope; and the extent of country graduations from IDA and reverse graduations back into IDA. Many alternative scenarios could therefore be devised, leading to diverse outcomes.

The challenge for countries receiving debt relief is to ensure that financial resources freed up by debt reduction are used for reaching the MDGs. Here, sound public financial management is key to achieving results; chapter 6 discusses the quality of public financial management in HIPC and suggests ways of improving performance. It is also critical that debt reduction does not undermine recipient countries’ capacity to mobilize domestic resources for development.

Lower debt burdens and debt sustainability. The MDRI will significantly reduce debt burden indicators in HIPC. Debt cancellation, once implemented, would cause debt ratios in HIPC to be significantly lower than those for grant-eligible non-HIPC. For the 18 postcompletion point HIPC, about 80 percent of the debt outstanding after HIPC relief is owed to multilateral creditors; in these countries the average NPV debt/exports ratio would fall from over approximately 140 percent after HIPC relief to a projected 59 percent after implementation of the initiative (figure 3.8). This would put debt ratios in these countries at less than half the average in non-HIPC that are currently eligible for grant assistance from IDA due to elevated debt ratios.

FIGURE 3.8Impact of MDRI on debt ratios in HIPC

Source: Staff estimates.

BOX 3.4The MDRI and “free-riding” risks

The MDRI provides for post-HIPC irrevocable debt stock cancellation and hence will significantly lower debt stock burdens for the debt relief recipient countries. This relief comes on top of HIPC debt relief already committed. After this debt relief, the debt stock ratios in most of the recipient countries will be significantly below that of many middle-income countries, which primarily borrow on nonconcessional terms. As a result there is an increased risk of “free-riding”—situations in which nonconcessional lenders may indirectly obtain financial gain from debt forgiveness, grants, and concessional financing activities of IFIs.a This situation could lead to an excessive buildup of debt if nonconcessional borrowing is not carefully managed. The figure below displays the new debt burden indicators for the 18 completion point HIPC after MDRI.

Post-MDRI debt relief: HIPC versus selected lower-middle-income countries

Source: Global Development Finance database and staff estimates.

Note: Debt data are public and publicly guaranteed debt and IMF credits at end-2003. The debt burden indicators for middle-income countries would be even higher than those for HIPC if the indicators were based on total external debt, because private debt is much larger in middle-income countries than in HIPC.

Lower debt ratios alone would not necessarily lead to changes in commercial risk ratings for these countries as other factors such as political risk would also be considered. However, credit rating agencies have been paying attention to this new reality, and Standard and Poors (S&P) announced its plans to assign sovereign debt ratings for many post-MDRI countries. Following a 2002 initiative by the United States to encourage African countries to apply for sovereign debt credit ratings so as to boost African participation in capital markets, several Sub-Saharan African countries obtained ratings: Burkina Faso, Cameroon, Madagascar, Mali, and Mozambique were issued ratings by S&P in 2004 (other than Botswana, which has an S&P rating of A, the ratings for other sovereigns are less than BB and hence speculative). Ratings represent a move toward transparency, help markets evaluate risk, and can help attract private investors.

Although increased FDI would be welcome in MDRI countries, lower debt stock ratios may result in a higher risk in the next few years that the fiscal space freed up by debt relief could be filled with new nonconcessional sovereign borrowing. This raises concern for the long-term development of the MDRI recipient countries, as excessive amounts of nonconcessional borrowing could potentially erode the benefits of the debt relief provided by multilateral creditors, namely sustainable debt burdens and additional fiscal space to support achievement of the MDGs. Most MDRI recipients, especially the postcompletion point HIPC, have had low and infrequent nonconcessional borrowing, most notably since the original HIPC Initiative was announced. In addition to the discipline imposed by the IMF arrangement itself, a condition of HIPC assistance, continued high debt burdens, and the likelihood of countries benefiting from HIPC and future debt relief initiatives may have been strong deterrents for commercial creditors to extend nonconcessional loans.b This may no longer be the case after MDRI.

a. While the similarity to the classical “free rider problem” in public economics is not perfect, the “free riding” debate in the present context reflects a concern that nonconcessional lenders may be willing to finance even unproductive investments knowing that, as a result of MDRI and the potential for future grants, the country will be able to make its debt service payments.b. The universal creditor coverage under the HIPC Initiative and the common reduction factor that applied to all creditors were meant to address some of the potential free-rider problems. Without such features, there is inherently a higher risk of free-rider problems from the MDRI.

The MDRI would leave African and Latin American and Caribbean (LAC) HIPC with different debt ratios. Among African completion point HIPC, the average debt/exports ratio would be about 45 percent, whereas among the LAC completion point HIPC it would be 92 percent. This is largely due to the fact that LAC HIPC would receive debt relief only from IDA and the IMF (currently there is no agreement for LAC HIPC to receive debt cancellation from other regional development banks), while African HIPC would benefit from debt relief from IDA, IMF, and the AfDF, which together hold, on average, 80 percent of debt outstanding in these countries.

Continued measures are needed by HIPC and by creditors to ensure that debt sustainability is maintained after HIPC and MDRI relief, just as similar measures are needed for other low-income countries (box 3.4). A review of the framework for assessing debt sustainability in low-income countries is currently under way to address the issue of how to apply the framework to MDRI recipients to ensure that they do not rebuild unsustainable debt burdens while not unnecessarily constraining access to resources for development.


World trade, though strong in 2005, slowed slightly from 2004. Exports of merchandise reached $10.2 trillion in 2005, growing 14 percent. While this rate is lower than the 2004 growth rate of 21 percent, it is still above the average of 8 percent over 1995–2004. Fuel exporters experienced a 37 percent increase in exports in 2005, but global nonoil exports also performed well, with 121 percent growth.

Developing-country export growth, at 24 percent, continued to outpace the global average. Middle Eastern and North African countries reported the highest level of export growth at 37 percent, up markedly from 28 percent in 2004, reflecting higher energy prices. China continued to accelerate its exports, with a 28 percent increase in 2005. Merchandise exports from Sub-Saharan Africa slowed slightly from 2004, but still experienced a 27 percent increase overall. Least developed countries experienced a remarkable 32 percent growth, while industrialized countries expanded their exports at a more modest 9 percent.

BOX 3.5Estimating the impacts of global trade liberalization

The global welfare gains from full liberalization of merchandise trade are estimated at $280 billion a year by 2015, of which $86 billion would accrue to developing countries. This is an underestimate as it does not take into account the gains from services liberalization, trade facilitation (each day that goods are delayed in ports and customs imposes a cost on consumers equivalent to 1 percent of the value of the shipment), or possible dynamic productivity gains. The size of global gains also depends on actions by governments to both liberalize trade and implement complementary measures to facilitate adjustment and job creation. Although developing countries and LDCs as a group stand to benefit, much depends on the depth and content of the associated reforms.

Recent analysis (see Hertel and Winters 2006, as well as Global Monitoring Report 2005) demonstrates that some countries, such as Brazil and China, would make immediate gains from an ambitious Doha Round, while others could experience losses in the short run. Contrary to popular belief, global trade reforms would improve the income distribution in countries such as Brazil and reduce poverty, because households in many of the poorest areas in such countries rely heavily on low-skilled labor employed in agricultural activities or related sectors. But poorer households in richer parts of the country could lose because of higher food prices.

That some poor countries may experience short-term poverty increases points to the need for assistance to address adjustment costs and to implement measures to stimulate growth and reduce poverty. The answer is not to preclude trade reforms from occurring, but to complement them by actions to reduce transactions costs and improve the investment climate. Without improved transport and market infrastructure, along with increased investments in training, extension services, new seeds, and the like, the gains from trade reform for the poorest countries will be reduced. Aid for trade is critical to enhance the overall gains from trade and manage the costs and downside risks in vulnerable countries.

Sources: Hertel and Winters 2006; Anderson and Martin 2006; Djankov, Freund, and Pham 2006.

In addition to cyclical factors, the trade performance reflects continuing unilateral trade reforms. Average tariffs in developing countries have fallen from 16.3 percent in 1997 to 12.2 percent in 2005. Additionally, China’s WTO accession-related reforms, along with the impact of the elimination of quotas on world trade in textiles and clothing on January 1, 2005, imply that some of the potential gains from global trade reforms identified at the launch of the Doha Round in 2001 have already been realized.41 But multilateral trade liberalization can still bring significant gains to the world economy (box 3.5).

Multilateral Negotiations

After the failure of Cancun in 2003, the July 2004 framework put the Doha Round negotiations back on track, but progress remained difficult. The 6th WTO ministerial meeting in Hong Kong in December 2005 was originally expected to agree on negotiating modalities for agriculture and nonagricultural products, but it became clear that agreement would not be reached. Expectations for the meeting were recalibrated, avoiding an acrimonious collapse, but leaving little time to achieve an ambitious outcome.

The road map agreed at Hong Kong requires agreement on divisive issues in the first half of 2006 to enable the round to be finalized before the expiration of U.S. Trade Promotion Authority in July 2007. Negotiating modalities for agriculture and industrial products are to be agreed by April 30, 2006, and comprehensive draft schedules for liberalization, by July 31. A new round of revised services offers is to be submitted by the end of July and final draft schedules of commitments, by October 31, 2006.

Aside from the road map, the ministerial produced fairly modest outcomes.42 First, agricultural export subsidies (modest in dollar terms, but highly distorting) will be phased out by 2013.43 Parallel disciplines are to be developed on equivalent programs, including food aid.44

Second, developed-country members will (and those developing countries who can should) provide duty-free and quota-free market access for products originating from LDCs by 2008 or no later than the start of the implementation period of the Doha Round. But because of sensitivities regarding clothing, leather, and rice, commitments may be limited to 97 percent of tariff lines, significantly undermining the value for LDCs. For example, more than 70 percent of Bangladesh’s exports to the United States are covered by only 70 tariff lines, or less than 1 percent of all U.S. tariff lines.45 Only 39 tariff lines account for 76 percent of Cambodia’s exports to the United States. Moreover, the value of any preferential access will depend on the restrictiveness of the rules of origin.

WTO members now face a considerable challenge in concluding the negotiations this year. All members need to galvanize the political will to undertake necessary reforms—the EU on expanding agricultural market access, the United States on reducing agricultural domestic support, and key developing countries on further market opening in manufactures and services—to conclude an ambitious Doha Round. Failure to do so risks either a modest Doha Round outcome that results in little new liberalization, or that the Round enters a period of drift. As the first trade negotiation with development explicitly at its core, it is vital that the Doha Round deliver reforms with concrete benefits for developing countries.

Developing countries’ own reform could account for half of their potential gains from global liberalization. Such reforms are particularly important for countries confronting possible preference erosion or higher food import prices from OECD agricultural reform (box 3.6). Countries should be supported in addressing the short-term adjustment costs of reforms with long-term benefits, and in building capacity to expand their trade. More attention also needs to be paid to the distributional consequences of liberalization.

BOX 3.6Africa and trade reform

Sub-Saharan Africa would see an increase in real incomes of $4.8 billion from deep global reform of merchandise trade, or 1.1 percent of GDP. Although small in absolute terms, this increase is significant relative to initial incomes and trade flows. Two-thirds of the gains come from reforms by other countries (including other countries in Sub-Saharan Africa) and one-third from countries’ own liberalization.

Agriculture accounts for 78 percent of the total gain for Sub-Saharan Africa, of which 12 percent (if southern Africa is excluded) comes from cotton. Cotton contributes a negligible amount (0.5 percent) to the global gains of trade reform but is very important to Africa. Global cotton trade reform would boost Sub-Saharan Africa’s cotton output by $2.2 billion per year by 2015, and cotton exports by $1.9 billion. Sub-Saharan Africa (especially non-LDCs without preferences) gains from increased agricultural market access and from higher prices for crop exports.

But while Sub-Saharan Africa as a whole gains, some countries lose from preference erosion or, in the case of net food importers, from higher food prices. Losses from preferences, as with benefits, are less than expected due to strict rules of origin and are highly concentrated among relatively few countries and products (mostly those with quota rents, such as sugar). But even relatively small losses can pose significant adjustment problems for some countries, underlining the need for increased aid for trade to support and complement trade reform.

There is a large domestic agenda to address the competitiveness problems that underlie Sub-Saha-ran Africa’s poor trade performance. Aid for trade can help by supporting policy reform and infrastructure investments. A typical import transaction in Africa takes 58 days (versus 14 days in the OECD), and each day of delay reduces export volumes by 1 percent on average. Poor roads and ports, and poorly performing customs, mean that trade facilitation measures in Sub-Saharan Africa will have significant payoffs. Services liberalization, coupled with aid for regulatory capacity, could help to improve access to new technologies, finance, and other services and could strengthen trade performance.

Sources: Anderson, Martin, and van der Mensbrugghe 2006; Djankov, Freund, and Pham 2006.

Regional Trade Agreements

Bilateral and regional trade agreements continued to mushroom during 2004–5 (annex table 3.1). As of January 2005, approximately 170 still active RTAs, up from 24 RTAs in 1990, had been notified to the General Agreement on Tariffs and Trade/WTO. Many more have not been notified: an estimated 250-plus RTAs cover one-third of world trade; 20 more await ratification, and another 70 are under negotiation. Although RTAs can include opportunities or issues not available multilaterally, from a development perspective, they are collectively inferior to nondiscriminatory liberalization. The benefits are often less than expected because of restrictive rules of origin (commonly a problem in North-South RTAs) or wide product exclusion (particularly prevalent in South-South agreements). As many as half of all RTAs may divert more trade than they create, according to Global Economic Prospects 2005, and bilateral “hub and spoke” RTAs benefit the hub (the rich country) disproportionately more than the spokes (developing countries). An ambitious Doha agreement would help to reduce discriminatory preference margins in all markets.

Unilateral Policy Reforms

Tariff liberalization by OECD and developing countries caused a two-percentage point reduction in the overall trade restrictiveness index (OTRI) for the world as a whole, to 14 percent on average.46 The OTRI of high-income countries is 11 percent, down from 12 percent for 2002; much of this decrease is due to a reduction in agricultural tariffs underpinning market price supports (annex figure 3.1). These reductions mostly affected middle- and high-income country exporters; the trade restrictiveness confronting exports from low-income countries continues to be above the world average (figure 3.9), preferences notwithstanding. In part this trend reflects the importance of nontariff measures (NTMs) as a source of trade restrictiveness (annex figures 3.1 and 3.2). For low-income countries as a group, agricultural NTMs are as important as tariffs.

FIGURE 3.9High-income countries’ OTRI, overall and toward low-income countries, 2005

Source: Staff estimates.

Agriculture is still much more protected than manufacturing, particularly in middle-and low-income countries. While LDC and Sub-Saharan exports face low barriers in manufacturing, their agricultural exports face a level of restrictiveness similar to, or higher than, that of products originating in high-income countries.

The changes in the OTRI between 2002 and 2005 (table 3.5) indicate that while world markets are now less restricted, the OTRI for LDCs has changed very little. The OTRI is still highest in the Middle East and North Africa, followed by Sub-Saharan Africa, South Asia, Latin America, and East Asia and the Pacific. Most of the decrease in the overall OTRI for developing countries is due to liberalization in middle-income countries. High-income countries are more open for Sub-Saharan Africa relative to low-income countries. But, this relative preference is lower now than in 2002, especially for Japan, where the OTRI against low-income countries has decreased substantially more than the OTRI against Sub-Saharan Africa.

TABLE 3.5Developing-country OTRI by geographic region and changes, 2002–5
OTRI for all countriesaOTRI for low-income countriesOTRI for least developed countries
Importing income group/country2005change2005change2005change
East Asia and Pacific16-5.323-3.024-1.3
Europe and Central Asia11-1.213-3.112-3.0
Latin America and the Caribbean17-1.120-0.820-0.1
Middle East and North Africa27-4.338-9.634-13.0
South Asia19-4.027-2.224-2.6
Sub-Saharan Africa230.024-1.024-0.8
Developing countries18-2.523-2.622-2.5
Least developed countries20-0.122-0.1220.0
Low-income countries20-1.624-1.224-1.1
Middle-income countries17-2.922-3.521-3.3
Source: Staff estimates.

High-income countries account for most of the OTRI for the “all countries” group.

Source: Staff estimates.

High-income countries account for most of the OTRI for the “all countries” group.

Aid for Trade

The G-8 Summit in Gleneagles in July endorsed additional aid for trade47 and requested detailed proposals from the international financial institutions, which were agreed at the September Annual Meetings. The main recommendation,48 to enhance the Integrated Framework for Trade-related Technical Assistance for LDCs,49 was subsequently taken forward by a task force of donors and LDCs in the WTO. The task force will report by April 2006; the enhanced Integrated Framework is to become operational by end-2006. The WTO Hong Kong ministerial also created an additional task force to provide recommendations by July 2006 on operationalizing aid for trade and asked the WTO Director-General to consult on appropriate mechanisms to secure additional resources for aid for trade.

Aid for trade has been steadily increasing. Between 2002 and 2003 trade-related assistance increased from 3.6 percent to 4.4 percent of total aid commitments; infrastructure accounted for a further 25 percent.50 Assistance consolidated in 2004: commitments in trade policy and regulations declined from $934 million to $811 million, but assistance to trade facilitation and trade development rose sharply. Aid for infrastructure remained at $9.3 billion; Asia was the largest recipient, but Africa’s share increased by 60 percent over 2002 to $3 billion in 2003. Assistance to Africa for cotton increased fivefold over 2002-3 to $63.5 million in 2004.

In July 2005 the European Commission announced an increase in trade-related assistance of €300 million a year to a total of €1 billion a year, which EU member states subsequently undertook to match by 2010. The United Kingdom alone will treble its aid for trade to £100m a year by 2010. Japan has promised $10 billion on aid for trade over three years, and the United States has promised a doubling of aid for trade to $2.7 billion a year by 2010.

The Bank’s trade-related lending has almost trebled over the last three years, rebounding to $1.2 billion in 2004 from about $300 million in 2002, to account for 6 percent of portfolio, thanks largely to the resumption of lending for trade-related infrastructure. Africa is the priority target; it has received some 60 projects and 30 percent of trade component lending.

Financial support from the IMF for potential adjustment costs of trade reform is generally met in the context of regular Fund arrangements. In 2004 the Trade Integration Mechanism was specifically designed to help members cope with the possible balance-of-payments impact of liberalization by trading partners. Up to $185 million (including potential augmentations under existing arrangements) has been made available thus far. Discussions are under way with selected members on incorporating floating tranches into Fund arrangements to provide flexibility on the timing of trade reforms associated with potential transitory balance-of-payments shortfalls.

There is a critical need to ensure that aid for trade is effective and that it is not viewed as a substitute for liberalization under the Doha Round, or as a bargaining chip as this is unlikely to result in decisions based on sound assessment of need. To ensure its effectiveness, aid for trade should be an integral part of national development programs. Developing countries have a better understanding of their needs and internal constraints, so local initiative and control over implementation is vital to the success of aid for trade projects. Aid for trade must be accompanied by necessary policy reforms to create a supportive investment climate and be additional, rather than a reallocation or redefinition of existing funds.


1.Not all ODA represents resources that can be used for long-term development.
2.In real terms non-DAC ODA was unchanged.
3.As part of the EU, these donors have also committed to meeting country-level targets of ODA/GNI collectively agreed to by the EU (see note 7).
4.The Forum on Partnerships for More Effective Development Co-operation (Feb 1-2, 2005) brought together DAC members and a number of non-OECD countries involved in furthering development in developing countries. The aim of the forum was to improve coordination and cooperation among the entire donor community—particularly through better information and knowledge sharing.
5.The World Bank is partnering with OECD-DAC and the United Nations Development Programme to gather information on South-South aid.
6.Both debt relief and technical cooperation declined in 2004. Contributions to multilaterals rose sharply, in part due to the increasing amounts of resources provided to the European Commission.
7.Debt relief represents additional financial resources if debt is being serviced, but amounts to an accounting exercise if it is not.
8.According to the Council of the European Union (2005), the individual objective for 2010 for the member states that are part of DAC is 0.51 percent and that for the 10 newest members of the EU is 0.17 percent.
9.Similar information on resources from non-DAC donors will not be collected.
10.DAC-World Bank-sponsored meeting on scaling up, December 6, 2005.
11.An international partnership hosted by the OECD DAC.
12.For the indicator on untying of aid, it was agreed in Paris that the target should be “continued progress.”
13.The survey instruments were field tested in five partner countries in February and March 2006. Data collection is expected to begin in May to determine initial baselines. Early progress against the baseline will be measured, aggregated, and reported. A second survey to solidify these results will be undertaken in 2008, ahead of the next High Level Forum in Ghana later that year.
14.Budget support provided to survey countries increased 30 percent in 2005 to $3.3 billion. Bilaterals accounted for 29 percent of this increase. The United Kingdom disbursed 40.9 percent of its assistance in the survey countries as general budget support, and the Netherlands, 31.9 percent.
15.To promote collaboration, the World Bank, on behalf of an international partnership, hosts the Country Analytic Work Web site at, which carries major reports of more than 25 multilateral and bilateral donor agencies.
16.The OECD-DAC’s Working Party on Aid Effectiveness is examining this issue with a view to identifying and disseminating best practice.
17.Technical cooperation is often criticized as being excessively costly because of the high cost of international experts, as exacerbating the problem of brain drain by training the best and brightest but not being able to retain them, and as being too fragmented and uncoordinated (Berg 1993 and OECD 2006a).
18.An issue here is whether budget support is the appropriate way to provide technical assistance. In providing technical assistance as budget support, designated funds might need to be ring fenced so as to ensure their availability to support technical assistance activities.
19.Food aid is provided during emergency situations in which food supply has been disrupted or the local food market has been destroyed; for humanitarian purposes to prevent hunger in poor households; and sold in local markets with the proceeds providing budget support.
20.Timmer (2006) argues that during extreme emergency situations, such as immediately following an earthquake, tsunami, or hurricane, using the nearest supply source is usually faster and cheaper than procuring food supplies from far away and that it is likely to result in more timely relief.
21.The nine pilot countries and the convening donors are the Democratic Republic of Congo (Belgium), Guinea Bissau (Portugal), Haiti (Canada), Nepal (United Kingdom), Somalia (United Kingdom and World Bank), Solomon Islands (Australia and New Zealand), Sudan (Norway), Yemen (United Nations and United Kingdom), and Zimbabwe (European Commission).
22.Heads of agency will consider a synthesis report and refine the Principles at the end of 2006.
23.For instance, the LICUS Implementation Trust Fund provides small grants to support early reform, development service delivery, and harmonize approaches.
24.This will compliment the common results framework for governments and donors alike—the transitional results matrix—that brings together the political, security, economic, and social spheres into one simple planning tool. The transitional results matrix, which was developed by the Bank and other partners (notably the United Nations), has been used to guide transitions and as a precursor to a full PRSP in the Central African Republic, Liberia, and Sudan, among other countries.
26.The definition of poverty-reducing expenditures varies across countries. Commonly included are primary education, basic health, and rural development.
27.Country authorities are implementing public expenditure management systems that would ensure the efficiency of poverty-reducing expenditures. See IMF-World Bank (2005).
28.Since 1989 the DRF has assisted in retiring $4.1 billion in principal and $3.4 billion of associated interest owed to commercial creditors.
29.To receive MDRI relief from a participating international financial institution, countries must also be current on their obligations to the institution and must be compliant with existing reporting requirements on external borrowing to qualify for relief from IDA. If the boards find that these conditions have not been met, remedial measures would need to be implemented and qualification for MDRI relief would be reassessed.
30.In addition, to fulfill the principle of uniformity of treatment in the use of IMF resources, the Fund board decided that all members with per-capita income of $380 a year or less (HIPC and non-HIPC) will receive MDRI debt relief financed by the institution’s own resources. Accordingly, Cambodia and Tajikistan are also eligible for MDRI relief from the IMF.
31.In the case of Mauritania, the board determined that it could qualify for MDRI relief only after certain remedial actions were taken.
32.Benin, Bolivia, Burkina Faso, Cambodia, Ethiopia, Ghana, Guyana, Honduras, Nicaragua, Madagascar, Mali, Mozambique, Niger, Rwanda, Senegal, Tajikistan, Tanzania, Uganda, and Zambia.
33.Mauritania will qualify for relief after implementing key public expenditure reforms. Cambodia and Tajikistan will only get debt relief from the IMF—see previous footnote.
34.The above cost estimate will be updated annually to take into account the expected time profile when HIPC would reach their completion points and become eligible for debt cancellation under the MDRI, the volume of debt relief to be provided under the current HIPC Initiative, and the foreign exchange rates to be used for valuing the USD-denominated HIPC relief in SDR terms (IDA’s base currency).
35.This estimate excludes the potential sunset clause HIPC, and the protracted arrears cases (Liberia, Somalia, and Sudan). Additional contributions will be needed to cover the cost of debt relief for these countries.
36.Staff estimates that additional subsidy contributions of over SDR 200 million (over US$285 million) would be needed to allow the IMF to lend all remaining Poverty Reduction and Growth Facility (PRGF) loan resources to low-income countries at the PRGF concessional interest rate.
37.In their letter of September 23, 2005, G-8 governments proposed that donor compensation for debt relief be additional to a contribution baseline for future IDA replenishments, set at the level of IDA14 contributions in real terms.
38.That amount would continue to increase by the SDR inflation rate for subsequent replenishments. The actual SDR inflation rate over the preceding three years would be used to determine the baseline volume of regular contributions in each future replenishment.
39.In the same vein, in their letter of September 23, 2005, G-8 governments state “that funding for IDA will continue to depend on donors’ conviction of IDA’s effectiveness in delivering development assistance; IDA reflows; and the performance, financing needs and absorptive capacity of poor countries.”
40.To maintain donor contributions in real terms, that amount would continue to increase by the UA inflation rate for subsequent replenishments, using the actual average UA inflation rate per year over the preceding three years to determine the baseline volume of regular contributions in each future replenishment.
41.This is the major explanation why more recent estimates of the potential global gains from merchandise trade liberalization are lower than earlier analyses; see van der Mensbrugghe (2006).
42.Immediately before the ministerial, agreement had been reached on an amendment of the Agreement on Trade-Related Aspects of Intellectual Property Rights regarding the use of compulsory licensing of pharmaceuticals for public health reasons (December 6) and an extension until July 2013 for LDCs to implement the agreement (November 29).
43.Although 2010 was proposed, the EU sought 2013 to coincide with the end of its forthcoming budget cycle.
44.The Ministerial Declaration introduces the concept of a “safe box” to ensure that bona fide food aid for emergency situations will not be impeded. Beyond that, commercial displacement is to be eliminated.
45.The total number of tariff lines at the eight-digit level is 10,500. The calculation was done at the eight-digit level.
46.While the tariff database has been updated to 2005, data on nontariff measures is essentially the same as for Global Monitoring Report 2005, reflecting the limited resources for tracking such policies. The OTRI measures overall restrictive-ness, not just protectionism (for a detailed discussion see Global Monitoring Report 2005).
47.Aid for trade refers to technical and financial assistance to help countries address supply-side constraints to trade and transitional adjustment costs from liberalization (for a detailed discussion, see Global Monitoring Report 2005).
48.Others were that the Bank and the IMF will report by September 2006 on the adequacy of existing mechanisms to address regional or crosscountry aid for trade and that they will strengthen the framework for assessing adjustment needs. The Bank will also better integrate trade-related needs into its country programs.
49.Information on the Integrated Framework is available at
50.Definitions and figures are taken from the WTO/OECD DAC Trade Capacity Building Database.
ANNEX TABLE 3.1New regional and multilateral trade agreements, 2004–5
Central American Free Trade Agreement (CAFTA)2005
East African Community (EAC)a2005
European Union Enlargementb2004
Greater Arab Free Trade Area (GAFTA)c2005
Australia/United States2004
Chile/Korea, Rep. of2005
Chile/United States2004
China/Hong Kong (China)2004
China/Macao SAR2004
European Union/Chile2005
European Union/Egypt2004
Romania/Serbia and Montenegro2004
Singapore/United States2004
Turkey/European Free Trade Association (EFTA)2005
Source: IMF Trade Policy Information Database (TPID) and the World Trade Organization (WTO).

Introduced a common external tariff on January 1, 2005.

The European Union expanded by 10 Eastern European countries on May 1, 2004.

The free trade area came into effect January 1, 2005 among 17 Arab states.

Source: IMF Trade Policy Information Database (TPID) and the World Trade Organization (WTO).

Introduced a common external tariff on January 1, 2005.

The European Union expanded by 10 Eastern European countries on May 1, 2004.

The free trade area came into effect January 1, 2005 among 17 Arab states.

ANNEX FIGURE 3.1OECD restrictiveness remains high for low-income countries, 2005

Source: Staff estimates.

ANNEX FIGURE 3.2Changes in OECD OTRI between 2002 and 2005: as tariffs fall, non-tariff policies become more important

Source: Staff estimates.

    Other Resources Citing This Publication