Real aid - An agenda for making aid work - Actionaid International

2

CONTENTS

ACRONYMS & ABBREVIATIONS 2 EXECUTIVE SUMMARY 3 INTRODUCTION  4 CHAPTER 1 AID AND RIGHTS  7

1. INTERNATIONAL AID CAN WORK TO PROTECT BASIC RIGHTS 9
2. THE CASE FOR MORE AID 9
3. MUCH MORE AID IS NEEDED  12
4. AID AND NET RESOURCE FLOWS 13
5. CONCLUSION 13

CHAPTER 2 REAL AID  15

1. MEASURING REAL AID 17
2. CALCULATING REAL AID 17
3. REAL AND PHANTOM AID – THE DONOR RANKINGS  28
4. CONCLUSION 28

CHAPTER 3 ACCOUNTABLE AID  33

1. DONORS ARE NOT ACCOUNTABLE  35
2. THE CASE OF AID CONDITIONALITY 36
3. AN INTERNATIONAL AID AGREEMENT  39
4. CONCLUSION 48

CHAPTER 4 CONCLUSIONS AND RECOMMENDATIONS  51 REFERENCES  54 ACKNOWLEDGEMENTS AND CONTACTS 56

ACRONYMS & ABBREVIATIONS 2

ADB Asian Development Bank

BWI Bretton Wood Institution

CCF Christian Children's Fund

CG Consultative Group

CITES UN Convention on Trade in Endangered Species

CSO Civil Society Organisation

DAC Development Assistance Committee

EC European Commission

ECOSOC UN Economic and Social Council

GDP Gross Domestic Product

GNI Gross National Income

HIPC Heavily Indebted Poor Countries

IAA International Aid Agreement

IFF International Finance Facility

IFI International Financial Institutions

IMF International Monetary Fund

MDGs Millennium Development Goals

NGO Non-Governmental Organisation

NPRS National Poverty Reduction Strategy

ODA Official Development Assistance

OECD Organisation for Economic Co-operation and Development PEAP Poverty Eradication Action Plan

PRGF Poverty Reduction and Growth Facility

PRSC Poverty Reduction Support Credit

PRSP Poverty Reduction Strategy Paper

TA Technical Assistance

TAACC The Apac Anti Corruption Coalition

UN United Nations

UNDP United Nations Development Programme

UPE Universal Primary Education

WHO World Health Organisation

WTO World Trade Organisation

EXECUTIVE SUMMARY

We live in an age of unprecedented prosperity and technological progress, yet the most basic social and economic rights are routinely violated on a massive scale. Each day, 30,000 children die from easily preventable disease. 100 million children see their right to a primary education denied. 800 million people go to bed hungry each night.

Aid has a critical role to play in achieving these rights. It is not a magic bullet. But experience shows that where aid is deployed effectively as part of a wider development strategy, it makes a lasting difference in helping people to lift themselves out of poverty. It is equally clear that more aid is urgently needed. Estimates of the financing needs of the poorest countries vary, but they all point to the same conclusion – that current donor efforts are grossly inadequate.

Yet increasing aid by itself is not enough. Put simply, the aid system is not geared to achieving the poverty reduction goals that were agreed in 2000. Donors must radically improve the quality of their aid if it is going to make a fully effective contribution to the fight against poverty. At present, two thirds of donor money is phantom' aid that it is not genuinely available for poverty reduction in developing countries.

Failure to target aid at the poorest countries, runaway spending on overpriced technical assistance from international consultants, tying aid to purchases from donor country's own firms, cumbersome and ill-coordinated planning, implementation, monitoring and reporting requirements, excessive administrative costs, late and partial disbursements, double counting of debt relief, and aid spending on immigration services all deflate the value of aid.

In this report, we show the extent to which the official aid figures exaggerate rich countries' generosity. By discounting for phantom aid, we establish the amount of real aid' that is promoting basic rights in the poorest countries. The results highlight the urgent need for reform by the donor countries:

In 2003, real aid was only $27bn, or just 0.1% of the donor countries' combined national income

For the United States and France, two of the world's largest aid donors, almost 90% of their contributions are phantom aid

3

The G7 countries are the worst performers when it comes to real aid. On average, the world's seven largest economies give just 0.07% of national income in real aid. In other words, they must increase real aid tenfold to reach the UN target of 0.7%

In real aid terms, the Norwegians are 40 times more generous per person than the Americans, and 4 times more generous than the average Briton.

The problems underlying the gulf between official and real aid are not new. Donors have signed up to numerous international agreements to improve the quality of their aid. Yet this agenda has made little headway. At the heart of this failure there lies a lack of accountability on the part of donors for either the amount of aid they commit, or the quality of that aid. Meanwhile, donors continue to make excessive demands on recipients for upward' accountability, attaching rafts of intrusive policy conditions to their aid, and restricting the ability of developing countries to

plot their own development paths.

This report argues that the share of real aid in official aid flows is unlikely to increase unless this one-way' accountability is replaced by a system of genuine mutual accountability, which balances the legitimate interests of donors, recipients and, most importantly, poor people.

ActionAid is calling for a new International Aid Agreement to make aid real and accountable, with four key elements:

Clear policies from developing countries on the criteria for accepting aid

Mutual commitments in place of one-sided conditionality, that are monitored transparently at the country level

National and international forums where donors and recipients can review progress on an equal footing, overseen by a UN Commissioner on Aid

New mechanisms to substantially increase the volume and predictability of aid.

INTRODUCTION 4

International aid is in the spotlight as never before. The aid pessimism' of the 1990s has been supplanted by widespread agreement that significantly more aid is needed if poverty is to be reduced and basic rights protected. In the wake of the UN summit on Financing for Development in 2002, G7 countries are competing to push for their own proposals for financing development.

Yet aid increases will not help to reduce poverty in the absence of major improvements in the quality of that aid. At present, far too much aid is driven by geopolitical and commercial objectives rather than by efforts to protect the rights of poor people. If aid currently has a mixed record in terms of its impact on poverty reduction, that is because it is often not what it is designed to do. Poor quality aid from unaccountable donors is a blunt instrument in terms of its impact on poverty. As this report argues, far- reaching changes are needed by donors to make aid a sharp tool in the fight to realise basic rights for all.

Recipient governments also need to reform. Accountability, transparency, democracy and the protection of human rights must all be improved. But where donors promote these changes, they need to happen in the context of genuine mutual accountability between rich and poor countries. Donors must support and encourage developing country efforts to reduce poverty by meeting their international commitments to provide more and better aid. They must reach these commitments, not as they are currently measured, but in terms of real aid that is truly available to support poor countries' efforts to protect basic rights.

Aid donors fall far short of meeting the official international aid target of 0.7% of national income. ActionAid's new real 0.7% rankings', presented in this report, show that when it comes to real' aid they are falling even further behind. According to our analysis, more than 60% of aid flows are phantom'; that is they do not represent a real resource transfer to the recipient. For the worst performing G7 donors, the figure is as high as 89%. Real aid stood at only US$27 billion, or 0.1% of donor national income in 2003, with G7 donors at an average of only 0.07%.

5

In other words, despite political grandstanding on the issue, G7 donors are only one tenth of the way towards meeting the 0.7% target. And this paltry contribution pales in comparison with the value of reverse flows from South to North, in the form of ecological debts, unfair trade rules and South-North financial flows, which stood at US$710 billion in 2003.

This report argues that for aid to be fully effective, it can no longer be treated as a voluntary, charitable transfer from North to South. Instead, aid must be part and parcel of a wider redistributive agenda designed to protect basic rights. For this to happen, poor people's voices, needs and priorities must be put front and centre in the design of aid programmes.

This means that current patterns of accountability must change. At present, donor agencies hold recipients accountable, and are in turn accountable to their own taxpayers. But donors continue to use unfair, undemocratic and inappropriate policy conditionality in a way that skews recipient accountability away from the citizens of poor countries. Meanwhile, neither governments nor poor people in recipient countries are able to hold donors properly to account for the quality or quantity of aid they provide. This must change. ActionAid is proposing a new international aid agreement to make aid more accountable, and effective.

This report falls into three parts. Firstly, we show why aid must be provided as an entitlement based on rights. Secondly, we present our assessment of real' aid by donor, and show how far donors are falling short of meeting the real 0.7% target, and compare this with the extent of South-North flows. And finally, we present our proposals for a new aid architecture based on mutual accountability between donors and recipients. Our focus is primarily on government-to-government development aid – although important, we do not discuss aid from NGOs, or humanitarian aid, or aid in kind. Our findings are based on desk- based research, new analysis of donor aid flows and new country case study research in Vietnam, Cambodia, Uganda and Ethiopia.

CHAPTER 1 – AID AND RIGHTS 7

"EVERYONE, AS A MEMBER OF SOCIETY, HAS THE

RIGHT TO SOCIAL SECURITY AND IS ENTITLED TO REALISATION, THROUGH NATIONAL EFFORT AND INTERNATIONAL CO-OPERATION AND IN ACCORDANCE WITH THE ORGANISATION AND RESOURCES OF EACH STATE, OF THE ECONOMIC, SOCIAL AND CULTURAL RIGHTS INDISPENSABLE FOR HIS DIGNITY AND THE FREE DEVELOPMENT OF HIS PERSONALITY."

UN UNIVERSAL DECLARATION OF HUMAN RIGHTS, 1948

8

Sixty years ago, the Universal Declaration of Human Rights acknowledged that all humans have inalienable rights, indispensable for their dignity and the development of their personality.' The UN Declaration has since proven to be a powerful tool for promoting political and civil rights. But the violation of economic, social and cultural rights, also enshrined in the UN Declaration, until recently attracted less international attention. This is now changing. Increasingly, the international community is recognising that people who die because they lack access to medicines or clean water are suffering a rights' violation as urgent as those dying through persecution, repression or war.

In 1986, the UN declared the Right to Development: "an inalienable human right by virtue of which every human person and all peoples are entitled to participate in, contribute to, and enjoy economic, social, cultural and political development, in which all human rights and fundamental freedoms can be fully realised."2 And again, in the Millennium Declaration in the year 2000, 189 countries announced that: "We are committed to making the right to development a reality for everyone and to freeing the entire human race from want."3

But despite growing recognition of the importance of economic and social rights, the violation of these rights continues on a massive scale. Every day, 50,000 people die from poverty related and preventable

diseases. Thirty thousand of them are children. One billion adults are unable to read and write, and 100 million children are outside school. Almost one quarter of people in poor countries lack access to clean water. 4 This lack of basic rights in poor countries both stems from and reinforces highly unequal power – within and between countries – which marginalises poor people's needs and priorities.

The aid system itself reflects these power imbalances. Partly for this reason, aid alone cannot be expected to solve the widescale violation of social and economic rights. Aid is effective only as one part of a coherent development strategy.

For rich countries, this requires policies on trade and investment,

debt and security that are consistent with development goals, and which ensure a transfer of resources from North to South rather than South to North. For poor countries, social and economic rights cannot be pursued

in isolation from efforts to improve democracy, transparency and the rule of law, or from those to reduce conflict.

But aid remains part of the solution. Rich countries must ensure that they provide sufficient resources, of the right quality, to ensure the protection of basic economic and social rights. They must also ensure that aid fosters, rather than undermines, downward accountability from governments to citizens.

1. 1 INTERNATIONAL AID CAN WORK TO PROTECT BASIC RIGHTS International aid can work to protect basic rights, if provided properly. Since 1970, for example, aid has contributed to a doubling of school enrolments and a halving of infant mortality.5 Devastating diseases such as smallpox and river blindness have been rolled back or eradicated through well-targeted donor interventions. In East Asia, aid to South Korea and Taiwan was instrumental in helping lay the foundations for rapid economic growth that in turn enabled governments to extend basic economic and social rights.6

Yet in far too many countries, aid has not done an effective job of reducing poverty or protecting basic rights.

This failure is often blamed on venal governments siphoning off aid for their own purposes, the most notorious case being the Mobutu regime in the Democratic Republic of Congo, which stole an estimated US$4 billion of public funds.7 But this is at best a partial picture, not least because donors themselves have often been complicit in such corruption.8 As often as not, aid has failed to reduce poverty because it has never even reached the recipient country, but has instead been paid to donor country companies and consultants, often for overpriced and inappropriate goods and services that have few sustainable benefits. Forty per cent of global aid, and 20% of aid to Africa continues to be officially tied. The Development Assistance Committee (DAC) of the Organisation for Economic Co-operation and Development (OECD) estimates that the value of aid would be boosted by US$5-7 billion each year if aid were untied.9

When donors have provided aid in effective and accountable ways, poor people have benefited. In Uganda, for example, donors have been providing both budget support and debt relief to fund poverty reduction programmes. This aid has been untied, reasonably predictable and well co-ordinated. As a result, it has helped to fund a huge expansion in primary school enrolment, from three million children in 1997 to almost eight million by 200310. While problems of educational quality and weak accountability remain, there is no doubt that effective donor aid has helped many more children to access a basic education. In Southern Africa,

a seven-country donor-financed measles immunisation campaign reduced the number of cases from 66,000 in 1996 to 117 in 2000, almost eliminating it as a cause of child death.11 Examples such as this demonstrate the potential for aid – where the objectives and modalities are right – to make a tangible difference to poor people.

1.2 THE CASE FOR MORE AID Despite the growing consensus on the need for more aid to achieve the MDGs, set out most recently in the Millennium Project report and the UK government's Commission for Africa, a number of key arguments continue to be made against aid. These arguments include:

countries cannot absorb more aid

aid has failed to reduce poverty

aid will substitute for domestic revenue

aid increases dependency

aid undermines macroeconomic stability

aid will be wasted through corruption

But on closer inspection, none of these arguments makes the case for not increasing aid, although some of them do underscore the need for the right kind of aid to be provided:

More aid can be absorbed

Some donors are concerned that additional aid money cannot be well spent by poor countries with already over-stretched government capacity. Yet the evidence does not bear this out. Most studies find that negative rates of return on aid do not set in until it reaches at least 25-50% of GDP, above the aid levels of all but six low-income countries in 2003.12 Moreover, many of the obstacles to absorbing aid can be overcome through investment, for example in management and statistical systems and encouraging behaviour change.

Limited absorption capacity is often

a consequence of administratively cumbersome aid. Poorly coordinated multiple planning, monitoring, reporting and auditing requirements from donors directly limit recipients' ability to absorb aid quickly and effectively. Experience from countries such as Ethiopia has shown that using country systems and harmonised procedures reduces this burden, and significantly increases implementation rates for donor

programmes.13 Donors can also increase absorptive capacity by providing additional aid through channels other than central government, including local government, civil society groups and the private sector, while taking measures to ensure this doesn't undermine the long term capacity of central government.14 Donors could also help to build capacity by shifting from ineffective and inappropriate Technical Assistance towards genuine capacity building'

and skills transfer.

Aid can foster growth

In one recent review of 70 studies of the impact of aid on growth, 40 found

a significant positive relationship, and only one found a significant negative relationship.15 Another recent study found that where aid has been real', and has focused on priority investments and basic services, it has contributed significantly to economic growth.16

Where growth has failed to foster poverty reduction, inappropriate World Bank and International Monetary Fund (IMF) policy conditions have often played a role. Measures such as trade liberalisation, privatisation and fiscal austerity, undertaken as a condition of aid, have often failed to achieve the promised benefits for poor people. For example in Tanzania, regarded by the aid donors as a model performer in terms of its record on economic liberalisation, average GDP growth of almost 4% since 1990 has led to a reduction of just three percentage points in extreme poverty.17

Aid complements domestic revenues Despite large-scale aid flows to low- income countries, there was an improvement in the efficiency of revenue mobilisation across low- income countries between 1999 and 2003. Many poor countries are relatively efficient at raising revenues, and in Africa the tax effort is strong, once the size

of the tax base and the loss of revenue instruments through trade liberalisation is taken into account. Country studies have also shown that on balance aid has had a positive impact on revenue collection efforts, for example in Ghana and Uganda.18 In Malawi aid has been associated with strong resource mobilisation and reduced domestic borrowing, and has helped to protect public spending from the effects of volatile domestic financing.19

BOX 1: CAMPAIGNS AGAINST CORRUPTION  10 IN APAC DISTRICT, UGANDA

Apac District is a poor rural district in northern Uganda. Its local government administration has been rated as one of the most corrupt in the country. This has led local people to start campaigning to ensure that their government cleans up its act.

In Chawente Sub-Country, the Christian Children's Fund (CCF) is training primary school children in how to monitor the public funds coming into their school, using support from the DFID- financed Commonwealth Education Fund. CCF works with

30 schools across the district, helping children to hold their teachers to account. Although the project is new, it is already having an impact: the District Education Officer is taking up

the findings of the children and school inspections are improving. Where head teachers are not co-operating,

parents are demanding to know what has happened to the funds. Children are learning more about what accountability means in practice, and are using their knowledge to teach other children and their parents.

A group of child monitors in Chamwente Primary School said: It is important to do UPE (Universal Primary Education) monitoring so that children can learn better. Every fortnight,

we go and see the head teacher and ask him questions. We ask for the receipts for what has been spent and also check

the physical amount of things bought. We see whether the head teacher has bought the things or if he has just eaten the money. We would know if the head teacher has eaten the money because we look at the receipts to see if they are forged. We check the displays of releases of money which is required under UPE, and we count the children class by class. If something is wrong, we report this to the head teacher. If the head teacher does not accept what we say, we call the teachers and tell them. We also talk to the Christian Children's Fund.'

Adults are also working to promote local accountability in Apac. The Apac Anti Corruption Coalition (TAACC) is a network of 28 local and international NGOs, community based organisations and church groups. TAACC also trains local opinion formers to monitor the funds coming to their community for education and other projects. TAACC uses the findings of their tracking work to run radio broadcasts and debates and to organise public demonstrations against corruption.

TAACC's work is already showing results. Money diverted from education budgets has been recovered. The District Engineer has been indicted for corruption and TAACC's recommendations have also been taken up by the District Chairperson, who has written to all his staff requesting them to follow them.

Source: interviews with TAACC, CCF and Chamwente Primary School children, March 10 and 11 2005

Real aid fosters independence

Some donors are reluctant to increase aid because of concerns about poor countries becoming unduly dependent on external finance, mirroring debates in donor countries about welfare dependency. In some cases, commentators in developing countries have voiced similar concerns, arising partly from a history of intrusive donor conditionality and poor quality aid.20 If aid is provided in ways that genuinely place recipients in the driving seat' and reduce transaction costs, many of these issues can be addressed.

Longer term, perhaps the best measure of aid effectiveness is the extent to which a country reduces its reliance on external assistance. The East Asian tiger' economies and EU accession states are examples of aid being well used to foster greater economic and political independence. Although aid was only one factor in these countries' successes, the evidence shows that where governments are committed to poverty reduction, and aid is targeted and efficient, it can play a crucial

role in helping countries to grow

their economies and extend social and economic rights.21

Aid's macroeconomic impact

can be managed

Inflows of foreign aid will tend to appreciate a country's exchange rate and reduce export competitiveness, other things being equal. Governments

must either accept this appreciation, or sterilise the impact and raise interest rates. Some countries, backed by the IMF, are concerned that aid flows can undermine the performance of the private sector, especially in exports, and hit economic growth. In extreme cases such as Uganda, these concerns have even led the government to refuse any further aid.

However, most studies suggest that the impact of aid on exchange rate competitiveness is small. 22 There are also obvious trade-offs to be made between the beneficial impacts of

aid and the usually marginal costs to exporters. Any decision to reject aid

on these grounds needs to be based on a fully informed public debate. Such a debate was missing in Uganda, where the policy was decided in a closed-door discussion between the Ministry of Finance and the IMF. Line ministry officials and NGOs involved in the consultations for the national development strategy, the Poverty Eradication Action Plan (PEAP), reported that at no point were these trade-offs discussed, and in many cases were not even understood.23 For its part, the IMF has been reluctant to open up to public debate on this issue, and

has resisted using Poverty and Social Impact Assessments to run alternative macroeconomic scenarios.24

As with other critiques of aid, the key question is whether aid is genuinely

used for poverty reduction. Aid that ignores recipient needs and priorities, and which also inflates the exchange rate artificially, will result in a lose-lose situation for the recipient country. If

aid is real', it is up to the recipient government, and population, to assess the trade-offs and to decide whether the benefits of aid to the poor outweigh the costs to exporters.

Aid can help combat corruption Corruption is a very real problem across much of the developing world, and recipient country governments must take urgent action to address it. But as the UN Millennium Project has pointed out, in many countries corruption and poor governance' stems from a lack of capacity and effective systems, as well as a lack of commitment to poverty reduction. In these circumstances, donor aid can help to provide capacity and to build effective systems, and

can also support civil society in holding governments to account (see Box 1). Where government corruption is so widespread as to make fiduciary accountability impossible, donors should probably provide funding through alternative channels such

as civil society and community organisations.

Corruption is a consequence of poverty as well as a cause. Across the developing world, incentives for rent seeking' behaviour are created by public sector

FIGURE 1  ACTUAL AND REQUIRED ODA IN 2003 AND 2006

VALUE ($BN)

250

215 200

150

135

100

87

69

50

OFFICIAL ODA  PROJECTED  ODA REQ. FOR  ODA IN 2006

IN 2003 ODA IN 2006 MDGS IN 2006  IF ALL DONORS

(0.44% of GNI)  REACHED 0.7%

wages below the cost of living and

weak oversight of public servants. In Cambodia, for example, the typical civil servant only earns US$25 per month, only one fifth of the cost of living in Phnom Penh. Teachers in Uganda earn only US$50-$55 per month, far below the cost of keeping a family, even in the rural areas. The econometric evidence backs this up, showing that almost every dimension of governance is correlated with income, with higher incomes being both a consequence and a cause of better governance.25

Donors can help to put pressure on recipient governments to reduce corruption and to ensure that aid is spent on its intended beneficiaries. More importantly, local people are also starting to hold their governments to account, for example through budget tracking initiatives.26 At the same time, donors can do far more to tackle corruption within their own structures, not least by establishing fully transparent and open procurement procedures and by refusing contracts to firms and organisations implicated in corruption. While recipient governments and donors both need to ensure greater accountability and transparency, corruption should not be used as an alibi for failure to increase aid. If poverty reduction is neglected, corruption can be expected to increase and basic rights will be made more difficult to realise.

1. 3 MUCH MORE AID IS NEEDED The massive scale of unmet basic needs in poor countries gives donor claims that the aid system is operating at capacity a hollow ring. According to the UN Millennium Project, meeting the Millennium Development Goals alone will require aid of US$135 billion a year by 2006, up from only US$65 billion in 2002. But meeting the MDGs, while a critical first step, will not ensure that all basic rights have been met. Even if the MDGs are reached in 2015, more than one in five people in Sub-Saharan Africa will still be living in extreme poverty.27

Clearly, the need for increased aid is both immediate and long term. But at present this need is not being met.

As Figure 1 shows, there is a gulf between what is needed to achieve the development goals and what is currently being provided. In 2006 alone, donor commitments are likely be almost US$50 billion short of what is necessary to get developing countries on track for reaching the MDGs. This shortfall is manifested in the routine underfunding of urgent development priorities, with devastating results for poor people.

For example:

The World Health Organisation's 3 by 5' initiative, designed to get three million HIV-positive people in need of treatment onto antiretrovirals by the end of 2005, is currently under-

funded by US$2 billion. The upshot of this slow and inadequate donor response is that so far only 750,000 of the three million intended beneficiaries have been reached.28

The Education For All Fast Track Initiative, a multi-donor plan to provide low-income countries with the extra funding needed to provide universal primary education, has identified a US$2.3-$3.1 billion annual resource gap over the next two years for basic education. Donors have been slow to provide funding for the catalytic fund' designed to meet the finance gap for donor orphans' currently receiving limited donor support: a US$220m funding shortfall in 2004, and a US$260 million anticipated shortfall in 2005, will require eight of the world's poorest countries to scale back plans for reaching the education goals.29

In Ethiopia, the World Bank's internal estimates of the aid required to implement the national poverty reduction strategy – which has been approved by donors – point to a funding gap of US$1.2 billion, which threatens to jeopardise progress towards the MDGs.30

1. 4 AID AND NET RESOURCE FLOWS Substantial aid increases are needed, but these need to be dealt with in the

FIGURE 2  ROBBING PETER TO PAY PETER: NET RESOURCE FLOWS NORTH TO SOUTH

VALUE ($BN)

200

100

69

0

(100)

(100)

(200)

(210)

(300)

(400)

(400)

OFFICAL FINACIAL TRADE  ECOLOGICAL AID FLOWS FLOWS DEBT

wider context of overall resource flows to the poorest countries. At the moment, this isn't happening. When compared to reverse flows that come from countries in the South to the North, current aid levels look all the more inadequate. In addition to failing to provide anything approaching the volume of good quality aid needed, rich countries are effectively living at the expense of the very poorest people. Here, we identify flows in

three key areas – financial, trade and ecological debt. Although this is an incomplete calculation, it highlights the urgent need for aid reform to happen as part and parcel of a redistribution of resources and power between rich

and poor countries.

Financial flows

In 2003, developing countries transferred a net US$210 billion to the rich world – that is, it paid out US$210 billion more than it received in new inflows. By far

the largest component of this was the purchase of US$276 billion of foreign exchange reserves, which far outweighed net equity flows – portfolio investment and foreign direct investment – of US$149 billion. Interest payments

alone continue to take US$95 billion

of developing countries' resources, almost three times the value of what they receive in grant aid31. These figures are an understatement, since they exclude capital flight. Although difficult to measure accurately, a seminal study found that, over the period 1970-1996, the equivalent of US$285 billion left Africa.32

Trade flows

The fact that the global trading system discriminates against poor countries is well established. Tariff and non-tariff barriers, dumping and product standards cost an estimated US$100 billion per year to developing countries, 50% more than total official aid.33 Again, these figures are underestimates, because they don't include the costs of rich country protectionism in terms of reduced opportunities for employment, reduced incomes for essential goods such as food and health care, or loss of investment opportunities.

Ecological debt

Carbon emissions from rich countries are a key factor in climate change, which is impacting heavily on poor countries through increasingly unpredictable

weather and rising temperatures and sea levels. In Mozambique, for example, the devastating floods from tropical cyclones in 2000 cut annual growth rate from 8% to 2%. In Asia, climate change has been identified as a key factor in unprecedented flooding in Nepal, India, China, Vietnam, Cambodia and Bangladesh.34

It is difficult to cost exactly this North- South debt', whereby rich countries are building up a debt to the poor through their over-consumption of the global commons. However, primary estimates using International Panel on Climate Change assumptions show that stabilising atmospheric greenhouse gas concentrations at 1990 levels would have implied a global, equal carbon entitlement of about 0.43 tons per person. Yet in 2000, actual per capita emissions in the rich countries were about 3.4 tons, meaning that each person in the rich world was

over the limit by approximately 3 tons.35 According to the UK government,

the damage cost of carbon emissions is US$56 to US$223 per ton of carbon. Using a mid range estimate of US$140, each person in the rich world owes US$420 annually through excessive use of carbon36. In total, this results

in a South-North flow of around US$400 billion.

Based on these calculations, in 2003 developing countries subsidised the rich world by US$710 billion, or 10 times official aid levels. Once the real aid calculations in the next chapter are taken into account, the overall picture is even more distorted.

1.5 CONCLUSION

Almost 60 years after the Universal Declaration of Human Rights was agreed, basic social and economic rights continue to be violated daily on a massive scale. Aid has a critical role to play in righting this injustice and helping to realise these rights. Where aid has been targeted and accountable, it has been shown to make a vital contribution to advances in health and education, and to expanded economic and social opportunities. Where aid has failed, it was often not designed to achieve these objectives. More aid is needed, but it must be real aid that supports the realisation of basic rights.

Where aid is well-designed and geared to poverty reduction, concerns about its economic and political impact prove to be misplaced or exaggerated. Immediate unmet financing needs in priority areas such as HIV and AIDS and basic education demonstrate that, against

the most important test of absorptive capacity, more aid is urgently needed. But aid cannot stand alone as a strategy for reducing poverty and expanding rights. More aid has to be framed in

the context of overall resource flows

to developing countries. At present, massive reverse flows from North to South underscore the need for aid to

be part and parcel of a wider agenda

to redistribute resources and power between rich and poor countries.

CHAPTER 2 – REAL AID 15

"TODAY'S IS THE THE FIRST GENERATION WITH THE

RESOURCES AND TECHNOLOGY TO MAKE THE RIGHT TO DEVELOPMENT A REALITY FOR EVERYONE AND TO FREE THE ENTIRE HUMAN RACE FROM WANT"

KOFI ANNAN, 'IN LARGER FREEDOM: TOWARDS DEVELOPMENT, SECURITY AND HUMAN RIGHTS FOR ALL', 2005

More aid is a necessary condition of achieving the millennium development goals and realising basic rights. But it is not sufficient. The history of aid shows that all too often poor people's needs have figured as an afterthought, thereby weakening its potential impact on economic growth and poverty reduction. As this chapter shows,

real aid – the donor resources that are actually made available for poor countries – are a small proportion of total aid commitments. The case for more aid therefore urgently needs to be linked to a strong agenda for real aid, which ensures that aid increases translate into additional funding for poverty reduction in poor countries. Rather than hand ammunition to aid sceptics, a real aid agenda reinforces the political push to reach the 0.7% target – as well as ensuring that existing commitments go further, reformed aid strengthens the case

for more aid.

To date there have been few systematic attempts to estimate

how much aid reaches developing countries or poor people. Equally,

little effort has been made to rank performance across donors, or measure the extent to which transactions costs overburden recipient government capacity.37

We believe that an official ranking exercise, using more complete data than is currently available, should be undertaken as a matter of urgency.

In the absence of this kind of research, ActionAid has made a preliminary attempt to assess how much donor aid is provided in ways that can usefully contribute to achieving the international development goals – what we call real aid. We then use these estimates to compute, by donor, a set of Real ODA/GNI' ratios, showing what proportion of donor national income is actually being spent on poverty reduction38. For reasons

of data availability, we have only undertaken this exercise for bilateral donors – the scores for multilaterals, where available, have been attributed back to the bilateral agencies which fund them.

1. MEASURING REAL AID

In 2003, official ODA from all donors was US$69 billion, or 0.25% of the combined donor gross national income (GNI). But a large proportion of this aid was what we describe as phantom aid'. In effect, this aid never materialises for poor countries, but is instead diverted

for other purposes within the aid system. Our definition of phantom aid includes aid that is:

not targeted for poverty reduction, estimated to be worth US$4.9 billion

double counted as debt relief, totalling US$9.4 billion

overpriced and ineffective Technical Assistance, estimated at US$13.8 billion

tied to goods and services from the donor country, estimated at US$2.7 billion

poorly coordinated and with high transaction costs, estimated at US$9 billion

too unpredictable to be useful to the recipient – lack of data prevents an estimate

spent on immigration-related costs in the donor country; totalling US$1.5 billion

spent on excess administration costs; totalling US$0.4 billion.

In total, at least 61% of all donor assistance is phantom aid, with real aid in 2003 accounting for just US$27 billion, or only 0.1% of combined donor income39. Breaking down the findings for other donor countries reveals the scale of the gap between official and actual giving:

For G7 countries, official ODA was US$50 billion in 2003, or 0.21% of their combined GNI. Yet real aid was less than a third of this sum at US$16 billion, or 0.07% of GNI. In other words, when phantom aid is taken out, the G7 countries are only 10% of the way to the 0.7% target.

Eighty-six cents in every dollar of American aid is phantom aid, largely because it is so heavily tied to the purchase of US goods and services, and because it is so badly targeted at poor countries.

Just 11% of French aid is real aid. France spends nearly US$2 billion of its aid budget each year on Technical Assistance, and US$0.5 billion on refugee and immigration expenditures in France. Forty per cent of French aid is provided as debt relief, much of which is an accounting exercise rather than a real resource transfer.

In real aid terms, the Norwegians are nearly 40 times more generous per person than the Americans,

and 4 times more generous than the average Briton.

2. CALCULATING REAL AID

To arrive at a real aid figure, ActionAid has deflated the value of official aid to take account of donor practices that reduce its effectiveness. The final figures are indicative, not least because there are numerous possible methodologies and because there

is a shortage of official data that would allow a more precise calculation. Nonetheless, our results are at the more favourable end (to donors) of the range of what has been estimated by others, and arguably understate the scale of the challenge. Regardless of the exact level of real aid, the results show there is an urgent need for action to ensure donor money

makes a fully effective contribution

to poverty reduction.

2.2.1. Aid is not targeted for poverty reduction

Although the Monterrey Consensus established poverty reduction as the overriding objective of official aid, the reality is that other non-development objectives continue to drive many donors' aid allocations. As Figure 5 shows, the relationship between aid levels and need is a weak one, with resources allocated in a scattergun approach. For example:

FIGURE 3  REAL AND OFFICIAL ODA IN 2003

VALUE ($BN)

250

200 195

150

100

69 50

27

OFFICIAL AID IN 2003  OFFICIAL AID REAL AID IF ALL DONORS HAD  IN 2003 IN 2003 REACHED 0.7%

FIGURE 4  REAL AND PHANTOM AID FROM ALL DAC DONORS, 2003 18

1%

13%  ADMINISTRATIVE COSTS TRANSACTION COSTS

20%

TECHNICAL ASSISTANCE

4% TIED AID

14%

DEBT RELIEF

39% REAL AID

2%

REFUGEE SPENDING

7%

LACK OF POVERTY FOCUS

globally, only 40% of aid goes to low-income countries, despite their accounting for more three quarters of all people living in poverty40

only one third of aid goes to sub- Saharan Africa, the region where poverty is deepest and most widespread, and where the MDGs are furthest off track

there are no sub-Saharan African countries among top ten aid recipients of the EC, which gives almost three quarters of its aid to middle-income countries

three of the top five Japanese aid recipients are middle- income countries.41

Skewed aid allocations reflect the subordination of development goals to commercial and geopolitical priorities. Since the September 11 2001 attacks on the US, aid has become increasingly politicised, with aid allocations favouring strategic allies in the war on terror' (see

Box 2). In other cases, aid is strongly influenced by commercial objectives. For example, the US President's Emergency Plan for AIDS Relief (PEPFAR), which has committed $15 billion over five years, requires that funding is only provided for branded drugs. By excluding cheaper generic drugs from PEPFAR, the initiative promises US pharmaceutical companies lucrative contracts while providing far fewer people with life-saving treatment than could otherwise be the case.42

In a similar vein, Japan's Country Assistance Programme for Vietnam states that, since Vietnam is significant for Japan as a manufacturing base, a potential future export market and an energy supply base, Japan's assistance for improvements in the investment, trade and business environment in Viet Nam is expected to lead to a virtuous economic cycle not only between Japan and Viet Nam, but also between Japan and the Association of South East Asian Nations (ASEAN) as a whole.'43

Because Japan's aid to Vietnam is primarily designed to develop a market for Japanese exports and a host for its Foreign Direct Investment, it is focused on large-scale infrastructure to the exclusion of other areas. Eighty-six per cent of Japanese aid to the country goes into infrastructure aimed at promoting growth, while social development receives only 6% of grants48. Interviewees in Vietnam

also suggested that the choice of infrastructure projects in Vietnam is determined by the location of Japanese companies working in the country.49

The most accurate calculation of real' ODA/GNI ratios would take out all aid that does not go to reduce poverty, including within low-income countries. But given the lack of data, we will only be looking at aid allocations by recipient. We assume that middle- income countries should justifiably receive up to 30% of ODA, which is roughly proportionate to the number of poor people living in these countries.

BOX 2: AID AND FOREIGN POLICY

– THE RE-POLITICISATION OF AID

When the cold war ended, there was widespread optimism that aid would be de-politicised, and poverty reduction placed at the fore of international development efforts. Yet since September 11 2001, this optimism has been badly shaken as aid has increasingly been linked to the importance of countries in the war on terror'.

As the OECD-DAC has noted, recent aid increases have been mostly gone towards debt write off and reconstruction in Iraq, Afghanistan and Pakistan. Between 2000 and 2002, net ODA received by the strategically important countries surrounding Afghanistan44 more than doubled (from US$1.2 billion to over US$2.7 billion).45 Pakistan's ODA alone tripled between 2000 and 2002, with US bilateral contributions to Pakistan increasing from an average of US$40 million (1998-2000) to US$770 million in 200246. In Africa, strategic considerations have seen aid increase in the horn of Africa – Djibouti, Eritrea, Ethiopia and Somalia – from just over US$1 billion in 2000 to US$2 billion in 2003. The Gulf of Guinea oil producers, which now account for about 15% of US oil imports, have been another beneficiary of US aid as it has worked to secure its interests in the region.

Ten years ago the US gave just US$3.4 million in economic assistance to Gulf of Guinea countries. In 2003 it stood at US$93.4 million.47

Any aid going to middle income countries beyond this 30% share is counted as phantom' aid. On this basis almost US$5 billion or 7% of global aid flows

in 2003 have been discounted:50

Greek aid, 70% of which goes to middle-income countries, was most heavily discounted on this score. Spain, which gives 62% of its aid to middle-income countries, mostly in Latin America, also fared poorly.

Both Austria and the US allocated just 53% of aid to low-income countries.51 Three of the four largest recipients of US funding are Egypt, Iraq and Pakistan, which between them received 14% of US aid in 2002-03.

2.2.2 Debt relief is double counted as aid

Debt relief in 2003 amounted to US$9.4 billion, equivalent to 14% of the global aid budget. According to OECD rules, all money used to fund debt relief is counted as part of official aid. This applies both to debt relief under the Heavily Indebted Poor Countries (HIPC) initiative and to debts cancelled on export credits, as well as other bilateral debts. As Box 4 shows, all debt relief provided since 2002 has been counted as part of ODA, despite the fact that the Monterrey Consensus agreed that year explicitly stated that aid increases should be additional to debt relief.

We have discounted debt relief from real aid for three key reasons. First, cancelled debt stock – the principal and interest on the loan – are counted as ODA in the year in which the relief is agreed, even though any benefits are felt over several years. This explains recent dramatic year-on-year increases in Belgian aid, after debt relief to the Democratic Republic of Congo, and

in Portuguese aid after debt relief

for lusophone African countries.

Second, these figures exaggerate the actual transfer being made to poor countries because debt relief is valued at its full nominal value. Much of the debt relief provided to poor countries simply closes the gap between what countries were scheduled to repay and what they actually were able to repay, and has often done little to relieve budgetary pressure on poor countries. For example, most of the US$5.1 billion increase in aid to the Democratic Republic of Congo between 2001

and 2003 was debt relief, even though Congo's debt service has actually increased sharply since the country entered the HIPC process, from US$2.7m in 1999 to US$126.7m in 2003.52 Debt service for all developing countries between 2002 and 2003 fell by just US$0.3 billion. Among the 27 countries that have so far received relief from the HIPC initiative, debt service increased from US$2.6 billion to US$2.8

billion over the same period. 53 This suggests that the vast majority of the US$9.4 billion cancelled in 2003 is not directly available for reducing poverty.

Third, counting debt relief as ODA creates a public perception that more money is being spent on development than is actually the case. In the UK, for example, debt cancellation has been presented as additional to aid spending. This is double counting. If debt relief is taken out of the UK's ODA calculations, then the ODA/GNI ratio fell between 2000/01 and 2002/03, although it rose again in 2003/04.54 Funding debt relief from aid budgets is not only misleading. It also risks penalising countries that are not indebted, as aid resources are diverted towards heavily indebted countries. It also violates the principle that creditors should carry some of the cost of debt relief, given the role that reckless lending has played in the debt crisis, and the fact that much of the initial lending was not supporting development-related expenditures.

Discounting debt relief from our real aid calculations does not mean that debt relief is unimportant. Cancellation of unpayable debts is urgently needed alongside more and better aid in order to reach the MDGs and help address the negative net resource flows discussed in Chapter 1. Debt relief can also be a particularly effective form of resource

FIGURE 5  THE SCATTERGUN APPROACH: AID ALLOCATIONS AND NEED ARE WEAKLY LINKED

PER CAPITA AID ALLOCATION ($)

500 450 400 350 300

250 200 150 100 50

50 70 90 110 130 150 170 190 COUNTRY HUMAN DEVELOPMENT INDEX RANKING

FIGURE 6  PERCENTAGE OF AID GOING TO LOW INCOME COUNTRIES 2002 – 2003 21

GREECE 30%

SPAIN 38%

AUSTRIA 53%

UNITED STATES 53%

NEW ZEALAND 60%

GERMANY 62%

LUXEMBOURG 63%

CANADA 64%

FRANCE 66%

FINLAND 67%

JAPAN 67%

SWITZERLAND 68%

NORWAY 68%

UNITED KINGDOM 70% SWEDEN 73%

NETHERLANDS 73% ITALY 74%

DENMARK 75% AUSTRALIA 78% IRELAND 79% BELGUIM 79%

PORTUGAL 82%

transfer, as it is untied, stable, predictable and flexible. But funding for debt relief should be genuinely additional, rather than be paid out of current aid budgets.

3. Aid is spent on overpriced and ineffective Technical Assistance (TA)

In 2003, US$18 billion of donor money, or more than a quarter of total aid, was allocated to Technical Assistance. The official figures probably understate the amount being spent in this area, since most project and programme allocations hide significant spending on TA. Typically, TA pays for consultants, either long

or short term, to support and advise recipient governments on policy issues and to build capacity' – in Africa alone, donors employ an estimated 100,000 technical experts.55

Technical and knowledge gaps doubtless exist in developing countries, and in principle TA has the potential to improve both government commitment and capacity, and the impact of aid. But past experience has shown that TA is

rarely well used. There are three key problems. First, although TA, along

with food aid, is excluded from the OECD agreement on untying, most

of it is in practice heavily tied to donor country firms. Even where donors have fully untied, tender and contract arrangements continue to limit competition. For example, 25 of the 34 largest recipients of the UK TA contracts listed on the DFID website are British. The other nine recipients are mostly American and Canadian, and none is from a developing country.56

Second, the upshot of TA being tied

and the market being at best semi- competitive is that it is heavily overpriced. For example, in Cambodia the aid

spent by donors on 700 international consultants in 2002 was estimated to be between US$50 and US$70m – roughly equivalent to the wage bill for 160,000 Cambodian civil servants. In other words, donor-financed consultants working in the Cambodian government are paid upwards of 200 times what

their Cambodian counterparts receive.57 In India, DFID spent US$40m on TA from Credit Suisse First Boston over just six months, in the course of advising the state government of Orissa on energy privatisation. The total bill for foreign consultants on this programme eventually rose to US$110m, with most of the TA provided by Price Waterhouse Coopers.58 In Vietnam, one DFID official estimated that they typically pay foreign experts between US$18,000 and US$27,000 per month, compared to US$1,500- $3,000 for local experts.59

Third, TA does a poor job of responding to local demand, and often leads to inappropriate or irrelevant support.

In particular, TA is widely used in ways that foreclose policy options and steer countries towards donors' preferred reforms. The World Bank and IMF's joint initiative with the World Trade Organisation to provide TA through

an integrated framework' on trade

is a case in point. Diagnostic studies routinely push for the kind of

BOX 3: TECHNICAL ASSISTANCE AND OWNERSHIP

One of the key criticisms of current arrangements for Technical Assistance is that it is strongly supply-driven, thereby undermining country ownership. At present, TA is provided as a series of uncoordinated analytical, design, implementation, monitoring and capacity building activities, each managed separately through donor agencies' own systems. The fact that contracts are typically issued by the donor makes it difficult for countries to have a coherent strategy for filling technical and knowledge gaps, and raises recipient transaction costs.

Some donors have started to address these problems by putting TA money directly into the hands of recipient governments. For example, in South Africa the DFID Support for Economic Reform Programme enabled the Ministries of Finance and Trade to select their own consultants, and manage the contracts themselves. There was a strong preference for local expertise wherever possible. In Ethiopia, donors have pooled TA funding for the PRSP through UNDP – although Japan and Italy have continued to earmark their assistance. More recently, the World Bank Public Sector Capacity Programme has pooled support from donors including the UK and USA, with contracts managed directly by the government.

FIGURE 7  TA AS A PERCENTAGE OF ODA 24

LUXEMBOURG 2% IRELAND 2%

SWEDEN 4% ITALY 6%

DENMARK 6%

NORWAY 12%

SWITZERLAND 14%

UNITED KINGDOM 16%

SPAIN 16%

CANADA 17%

BELGIUM 17%

NETHERLANDS 18%

JAPAN 21%

AUSTRIA 23%

FINLAND 23%

NEW ZEALAND 24% FRANCE 27%

GREECE 32%

GERMANY 34%

PORTUGAL 45% AUSTRALIA 46%

UNITED STATES 47%

DAC AVERAGE 27%

FIGURE 8  TYING STATUS BY DONOR, EXCLUDING TA AND ADMINISTRATION 25

UNITED KINGDOM 0%

SWEDEN 0%

IRELAND 0%

NORWAY 0%

BELGIUM 1%

FRANCE 3%

JAPAN 3%

SWITZERLAND 4%

GREECE 5%

GERMANY 5%

PORTUGAL 6%

FINLAND 14%

NEW ZEALAND 19%

DENMARK 29%

AUSTRALIA 33%

SPAIN 44% CANADA 47%

AUSTRIA 49%

UNITED STATES 70%

ITALY 92%

privatisation blueprints preferred

by the IMF and World Bank – for example of Senegal's groundnut sector, urban water in Tanzania,

and cotton marketing in Burundi

and Madagascar. TA is also inefficient. One recent study found widespread duplication of TA, partly because donors do a poor job of coordinating analytical, diagnostic and capacity building work, and of using it once it is produced.60

For example, in Vietnam there are 60 different TA projects covering various aspects of the country's accession to the World Trade Organisation, funded by 23 different donors, with little apparent coordination.61

Despite these problems we do not want to count all TA as phantom aid. There are some important examples of technical assistance being provided

in a more cost-effective and ownership- friendly way (see Box 3). In the absence of any systematic evaluation of the effectiveness of TA across countries

or by donor, we have approximated

the contribution of TA to real aid in

the following three ways:62

Most TA remains tied, either officially or unofficially. Estimates of the mark- up from tying vary from around 15% to 40%, and we would expect TA to be at the upper end of this range, based on current evaluations. We therefore take off 33% of the value of TA.

The near total reliance on international TA involves extra cost items, such as generous living expenses and travel. Using local expertise in country, or third country residents, would not incur such costs. These costs will not be incurred in all cases, but are likely in at least half. Non-salary, non-profit overheads are around one third of resident TA costs. We therefore subtract a further 17% from the value of TA.

Based on the available evidence, we assume that around half of the outstanding amount spent on TA does not enhance institutional capacity or quality, or improve management and absorption of resources.

The total subtraction for tying, additional costs for expatriate advisors, and the weak contribution of TA to enhancing institutional capacity means that an estimated 75% of TA is phantom' aid. In reality the impact of TA will vary by

donor, although there is not enough data available to make such comparisons. However, based on the findings from Vietnam and Cambodia, counting 25% of TA spending as real aid is probably

a generous estimate. In total, therefore, we estimate that US$13 billion goes into phantom' TA spending. The largest providers of TA, as a proportion of their total aid budget, are the US (47%), Australia (46%) and Portugal (45%).

4. Large sums of aid remain tied An estimated 40% of all aid, excluding food aid and TA, is tied to the purchase of goods and services from the donor country. Italy and the USA are among the biggest culprits of tying, spending upwards of 70% of aid on domestic firms and organisations. More recently, donors have committed to untying all their aid to least-developed countries, although in reality procurement practices are often unchanged. For example, even though Japan has officially untied its aid, 96% of the 64 billion Yen of Japanese aid spent on large projects in Vietnam in 2003 involved projects solely or partially involving Japanese companies.64 Only four countries – the UK, Norway, Ireland and Sweden – have so far fully untied.

As well as working as a form of round- tripping, tied aid is hugely wasteful, inflating procurement costs by the equivalent of US$5 billion a year – money that could be better spent reducing poverty. For example, in Cambodia USAID-funded NGOs

must award contracts over a minimum threshold to US companies. In one case, this would have meant a healthcare NGO being required to buy oral rehydration salts at four to five times the price of locally available sachets. After a search for alternative funding, the organisation was eventually able to procure locally through the WHO.65

Tying also has a track record of distorting the content of aid programmes, for example by encouraging donors to make large capital expenditures that ignore the recurrent cost implications

for the recipient country.66 Tying can

also slow aid down, at enormous cost

to recipients. Food aid to Ethiopia is a case in point: in 2003, USAID's vegetable oil stocks were still being shipped out

of the country, while urgently needed in response to the worst food emergency in a decade. Local USAID staff requested permission from headquarters

in Washington, DC to make local food purchases but were refused, apparently after pressure from the US farm lobby.67

To discount tied aid from real aid, we made a lower range estimate that tying increases costs by 25%, meaning that 20% of tied aid is phantom aid. This implies that US$2.7 billion – or roughly 4% of global aid flows – are lost through tying.68

5. Aid is poorly coordinated

Aid carries high administrative and financial transaction costs for recipients. There are 35,000 aid transactions a year, 85% of them worth less than US$1m.69 Overstretched civil servants in aid dependent countries are required

to meet a raft of disbursement, procurement, reporting, monitoring

and auditing requirements from multiple agencies, diverting scarce time

and resources from identifying and implementing local policy priorities.

Despite numerous commitments

to reduce this burden through closer coordination between donor agencies, harmonisation of procurement, reporting and monitoring procedures, and greater use of a country's own budget and administrative systems, scant progress has been made. For example, the DAC estimates that a typical African country submits 10,000 quarterly donor reports each year, and hosts more than 1,000 donor missions.70

Lack of coordination is driven partly

by donor concerns about the visibility

of their own efforts and the ability to attribute quick results. In Cambodia,

this has manifested itself in three separate strategic plans, all supposedly country owned': the 2nd Socio-Economic Development Plan (SEDPII), funded

by the Asian Development Bank and written by one of their own consultants; the National Poverty Reduction Strategy (NPRS), funded by the World Bank

and largely written by a World Bank consultant; and the UN's MDG strategy for the country. While the government

of Cambodia is now making efforts to develop its own plan, amalgamating all three, the result has been a diversion of time and resources in a country that can scarcely afford either, and widespread confusion as to which document is the Cambodian government's guiding strategy.

27

In Ethiopia, 216 missions were reported to the OECD-DAC for 2003. Since only one quarter of donors reported their missions, the true figure is not known. Senior Finance Ministry officials reported meeting with several missions on a weekly basis. The Japanese were identified as imposing an especially heavy burden, with four appraisal missions from Tokyo being typical before a project can be approved.71 More recently, eight Direct Budget Support donors have made progress

in harmonising some of their missions, and identifying lead interlocutors with government on key issues.

While lack of donor coordination and harmonisation clearly impose heavy transaction costs, these are difficult to measure. Indeed, one attempt to do so in Vietnam foundered when the UNDP concluded that the transaction costs involved in measuring transaction costs were themselves too high.72 To discount transaction costs from real aid, we have broken down countries into strong' and weak' performers, based on an assessment of donor behaviour by

the UK based non-profit organisation, Development Finance International (DFI).73 Using DFI's results, we assume that 10% of aid will be lost in transaction costs for strong performers, and 20% for weak performers. The scores for

multilateral agencies have been attributed back to the bilateral donors that fund them.

On this basis, US$9 billion – or 13% of the global aid budget – is lost through transactions costs, with Japan, Italy, France, Spain and Switzerland among the weakest performers.

6. Aid is unpredictable

Aid is highly unpredictable, with much of it arriving late or not at all, and is far less reliable than government revenues.74 For Africa, actual disbursements of programme aid fall short of projections by 14%, and by 26% for project aid. One recent survey of aid recipients found that in 25% of cases, aid disbursements arrived 6 to 12 months late, while for the EC, only 14% of aid actually arrived on time.76 Unreliable aid both undermines long term planning for the Millennium Development Goals, and creates financial uncertainty for governments.

Administratively cumbersome procedures contribute to the problem. In Ethiopia, slow procurement arrangements mean that Italy's 1999-2001 aid programme is still being implemented. Similarly, in Zambia procurement has been blamed for late and incomplete releases to the education sector, which has left large parts of the country's education strategy unimplemented (see Figure 9).77 Donor conditions also lead to aid being delayed

or suspended during the financial year. In Ethiopia, one outstanding African Development Bank condition has led to a US$90m loan remaining undisbursed in the final quarter of the financial year 2004-05.

As a result, Ministries of Finance routinely discount donor commitments on the basis of past disbursements.

In Ethiopia, the Ministry of Finance discounts African Development Bank loans by as much as 80%, and EC aid by up to 75% at the start of the financial year.78 In Uganda, the Finance Ministry reported discounts of up to 50% on donor commitments.

The lack of predictable aid is exacerbated by donors' failure to make indicative commitments more than one year in advance. Countries such as Uganda have been praised by donors for their rolling three year budget plans, yet most donors have not responded with changes to their own budgeting. According to a Ministry of Finance official donors really need to start making us commitments over at least a three year time period.'79

The lack of donor-by-donor data on disbursement delays makes us unable to calculate a discount rate for unpredictable aid, although the available evidence suggests that budgeted commitments are a poor indicator of the resources that are usefully available to recipients.

FIGURE 9  BUDGETED AND ACTUAL AID TO EDUCATION

IN ZAMBIA 2000 – 2002

ZAMBIAN KWACHA BN

350

300 BUDGET 250

200

ACTUAL 150

100

50

2000 2001 2002

7. Aid is spent on

immigration services

In 2003, US$1.5 billion of official aid

– or 2% of global flows – was spent on refugee-related expenditures in donor countries. The extent to which donors count such costs under the aid budget varies. Australia allocated Aus$47m to spending on asylum seekers, more than 5% of the total aid budget.80 France is the biggest spender under this category

– US$455m of French aid, or 6% of its budget is spent on refugee-related services. Because this money never leaves the donor country, and is not available for poverty reduction in poor countries, we have discounted it from our real aid' calculations.

8. Aid is spent on administration About 5% of donor aid is spent on administrative costs for bilateral aid agencies. Some administrative costs are inevitable and necessary for the effective day-to-day running of aid programmes. However, some costs

are more questionable. Bilateral agencies typically look after their staff well, including generous living allowances for expatriates, high hotel expenses and business class flights. DFID officials posted overseas, for example, receive allowances to fund business class flights back to the UK. Staff can opt instead to use the flight fund' for other journeys,

in effect allowing use of the aid budget

to subsidise foreign holidays. DFID administrative costs, at 11.5%, are among the highest of any donor, and well above the 8% ceiling allowed by DFID in its funding agreements with NGOs.81 Other donors with high levels

of administrative expenditure include Denmark (11.7%) and Canada (10.8%).

In total, around US$3.9 billion is spent as administrative costs by bilateral donors. We have allowed for administrative spending of up to 8%, in line with the standard established by donors for their funding partners. Discounted administrative spending above this level totals US$342m.82

3. REAL AND PHANTOM AID

– THE DONOR RANKINGS Globally, donors give only 0.1%

of their national income in real aid, compared to the UN target of 0.7%. For the G7, the figure is even lower, at 0.07% – only one tenth of what it should be under the UN target.

The donors that have progressed furthest towards the 0.7% target are the best performers in terms of the ratio of real to phantom aid, as Figure 10 shows.

In particular:

Luxembourg comes out on top, with a 0.65% real aid/GNI ratio. The other 0.7% donors – Denmark, Sweden, Norway and the Netherlands – also rank highly. Only 19% of Luxembourg's aid is counted as phantom aid.

Outside the group of 0.7% donors, Ireland, Switzerland, Finland and the UK also do relatively well in terms of aid quality, with almost 90% of Irish aid counting as real aid, 65% of Finnish aid and 71% of UK aid. However, their real aid/GNI ratios are brought down by their low official aid commitments.

The United States and France lead the way among the poor performers, with upwards of 80% of budgets going into phantom aid. In particular,

The US spends just 0.02% of national income on real' aid – or US$8 for every US citizen. Heavily tied aid, a lack of poverty targeting and a large Technical Assistance budget count against the US aid effort.

French aid is heavily discounted for its US$2 billion TA spend, debt relief – which accounted for 42% of ODA in 2003 – and its immigration-related spending, which accounted for 6% of total aid.

Among other G7 countries, Italy and Japan score moderately well in terms of their real/phantom aid ratio, but their low official aid commitments put them towards the bottom of the real aid generosity table (see figures 10 – 12).

Among the smaller donors, Greece scores poorly on real aid, largely because it does a poor job of targeting aid at low-income countries, while there is a similar story in Spain. Belgium's large share of phantom aid results from a one-off hike in the budget due to debt relief.

4. CONCLUSION

More aid is urgently needed to reach the international development goals. But equally urgently, donors must undertake far-reaching reforms to ensure that aid quality is improved so

as to make a fully effective contribution to the fight against poverty.

While meeting the official 0.7% target is important, it is far more important that donors reach real 0.7%', a target that no OECD donor meets today. ActionAid believes that all donors must commit now to reaching the real 0.7% target

as a matter of urgency – and by 2010 at the latest.

FIGURE 10  REAL AND PHANTOM ODA/GNI RATIOS BY DONORS, 2003 29

UNITED STATES 0.02 0.13 REAL

ODA GREECE 0.04 0.17 PHANTOM

ODA

FRANCE  0.04 0.37

SPAIN 0.06 0.17 AUSTRIA  0.07 0.14

ITALY 0.08 0.09

GERMANY 0.10 0.18 PORTUGAL 0.10 0.12

JAPAN 0.11 0.10

CANADA 0.11 0.12 AUSTRALIA 0.11 0.13 NEW ZEALAND 0.11 0.11

BELGIUM 0.19 0.41

FINLAND 0.23 0.12

UNITED KINGDOM 0.24 0.10

SWITZERLAND 0.24 0.15 IRELAND 0.34 0.05

NERTHERLANDS 0.52 0.28 SWEDEN 0.57 0.23

DENMARK 0.60 0.24 NORWAY 0.62 0.30

LUXEMBOURG 0.65 0.15

TOTAL DAC 0.10 0.15 TOTAL G7 0.07 0.14

FIGURE 11  SHARE OF PHANTOM AID BY DONOR, 2003 30

IRELAND 13%

LUXEMBOURG 19%

DENMARK 28%

SWEDEN 29%

UNITED KINGDOM 29%

NORWAY 33%

NETHERLANDS 35%

FINLAND 35%

SWITZERLAND 38%

JAPAN 48%

NEW ZEALAND 49% CANADA 52% PORTUGAL 53%

AUSTRALIA 54% ITALY 55%

GERMANY 65% AUSTRIA 67%

BELGIUM 69% SPAIN 72%

GREECE 81%

UNITED STATES 86%

FRANCE 89%

TOTAL DAC 61%

TOTAL G7 68%

FIGURE 12  REAL AID PER PERSON IN OECD COUNTRIES, 2003 ($) 31

GREECE $7

UNITED STATES $8

FRANCE $13

SPAIN $13

PORTUGAL $15

ITALY $19

AUSTRIA $21

NEW ZEALAND $22

AUSTRALIA $29

GERMANY $29

CANADA $31

JAPAN $36

BELGIUM $56

FINLAND $70

UNITED KINGDOM $75

SWITZERLAND $111 IRELAND $112

NETHERLANDS $162

SWEDEN $193

DENMARK $232

NORWAY $304

LUXEMBOURG $357

CHAPTER 3 – ACCOUNTABLE AID 33

"ALTHOUGH THE CONCEPT OF COUNTRY OWNERSHIP

IS GENERALLY ACCEPTED, THE VARIOUS CONDITIONALITIES INTRODUCED HAVE NOT ONLY SLOWED DOWN THE PROCESS, BUT HAVE FURTHER UNDERMINED THE CAPACITIES AND FUNCTIONS OF STATE INSTITUTIONS. THE CONCEPT OF DEVELOPMENT. HAS BEEN SUBJECTED TO CONTINUOUSLY CHANGING 'DEVELOPMENT FADS' WHICH, AT THE END OF THE DAY, MAY RESULT IN THWARTING OWNERSHIP."

MELES ZENAWI, ETHIOPIAN PRESIDENT, IN ADDRESS TO AFRICAN FINANCE MINISTERS, 2000

As Chapter 2 shows, the international aid system is failing to play its part

in securing the basic rights of poor people. Aid commitments are far below the levels promised by the international community. More

than 60% of donor commitments are phantom aid that does not represent a real resource transfer to poor countries. Meanwhile, aid flows from North to South pale in comparison to the reverse flows' from South to North, in the form of ecological debts, unfair trade rules and South-North financial flows.

Donors have made numerous pledges to improve both aid quality and quantity over the past three decades. Yet these commitments have rarely been fully implemented, and in too many cases remain shamefully neglected. In this chapter, we argue that this is caused by a fundamental lack of accountability on the part of donors. Donors are not accountable to recipient governments, and neither donors nor recipients are really accountable to poor people. This lack of downward' accountability contrasts

sharply with donors' excessive demands for upward' accountability. This is reflected in intrusive policy conditions designed to modify recipient behaviour, in micro- management of reforms, and in the failure to use recipients' own systems and procedures for channelling aid.

Significant progress towards making more aid real requires that donors are held accountable for their actions. This implies a radically new approach to aid, one that replaces the prevailing top down, donor-dominated model with a system of genuine mutual accountability that balances the legitimate interests of donors, recipients and, most importantly, poor people. This chapter sets out the key elements necessary for a system of accountable aid' and outlines an agenda for reform.

35

3.1 DONORS ARE NOT ACCOUNTABLE

1. Broken promises – aid volumes The litany of broken promises on aid provides strong evidence that the main obstacle to changing the aid system is political, not analytical. There has been no shortage of international pledges to increase the quantity and quality of aid over the past three decades: the 0.7% aid target, recommended by the Pearson report' and adopted by the UN in 1970, was re-affirmed both in 1992 and 2002 by almost all donor countries. Yet as of 2005 only five donors – Denmark, Norway, Sweden, the Netherlands and Luxemburg

– had reached the target. Since the UN target was announced, no G7 country has ever joined the G0.7, although France and Britain have pledged to do so by 2012 and 2013 respectively. As Box 4 shows, however, donor promises are easily broken. And Ireland's recent decision

to backtrack on its own commitment to reach the 0.7% target by 2010 highlights a lack of real sanctions on donors who break such promises.

A similar picture emerges on aid quality. Donors have made commitment after commitment to improving aid quality. Yet year after year these promises have been broken. Most recently, donors have made very little progress in fulfilling promises made two years ago in Rome, to harmonise and align their aid, as Box 5 shows.

2. Donors are not held

responsible for results

The lack of progress towards meeting donor commitments on aid quality is matched by donors' reluctance to be held accountable for results. Very little information on aid quality is made available to the DAC, despite its role

as the main monitor of donor behaviour. In Ethiopia, for example, although 23 bilateral donors provide aid, only 11 submitted information to the recent

DAC survey. Often the information that is provided is so incomplete as to be virtually meaningless: in Senegal, France is the only donor to have reported joint missions, raising the question of who they harmonised with.83 Data on aid

quality also suffers from the fact that donors are self-assessing, meaning that reports of progress are often based on donor perceptions that are not shared by recipients. For example, whereas donors in Zambia reported that over half of all aid used local procurement systems, the government's figure was just 10%.84

Similarly, many donors have been extremely unwilling to adopt targets

that would allow them to be identified individually for what they have and have not delivered. At the Paris High Level Forum in March 2005, Japan and the US objected to targets on issues such as tied aid, in contravention of the DAC's own guidelines.85 Some donors have also rejected accountability mechanisms at the country level. In Tanzania, for example, key bilateral agencies prevented the naming of individual donors by an independent monitoring team reporting to the World Bank Consultative Group meeting.86

Donors are also notoriously bad at disclosing their planned and actual

BOX 4: MIND THE GAP: DONOR PROMISES AND PRACTICE

The Monterrey summit on Financing for Development, held in March 2002, was notable for the last minute rush of donor pledges to scale-up aid dramatically. After a decade of declining aid to the poorest countries, the European Union promised that all member states would reach the target of 0.33% ODA/GNI by 2006, while President Bush committed to increase US aid levels by US$5 billion a year, to be administered through a new Millennium Challenge Account. The summit document, the Monterrey Consensus, was explicit that these aid flows should be additional to debt relief. Three years on, the pledges look less impressive. Six months before EU member states are due to meet the 0.33% target, Italy, Portugal, Spain, Austria and Greece are giving between 0.1 and 0.25% of national income as aid; less still once debt relief is counted out. For its part, the US has taken three years to make its first Challenge Account disbursement, to Madagascar, of less than US$30m a year. Disbursements to just three other countries – Georgia, Honduras and Nicaragua – are expected during the remainder of 2005. Despite the US promise of an additional US$5 billion a year, Congress appropriated just US$1.5 billion for the Challenge Account in 2005, following a US$1 billion appropriation in 2004.

disbursements. In Zambia, more than three quarters of donor agencies fail

to notify the government about actual aid disbursements, making effective financial planning extremely difficult for the government. In Tanzania, 20 out of 39 donor agencies submitted no information about project or programme spending when asked to do so by government.87

A similar picture exists with donors' analytical work which, according to the DAC, a majority of donors' fail to share openly, and with disclosure of aid conditions, as discussed below.88

Since transparency is a prerequisite of accountability, donor behaviour suggests that for all the talk of partnership' and development compacts', there are serious limits on the extent to which donors wish to be answerable for their actions.

In sum, the public accountability for results that has pushed donors to deliver debt relief has been missing from the aid effectiveness debate. The aid system currently lacks – and urgently needs – strong mechanisms to replace one-way accountability (directed upwards to donors) with mutual accountability.

The recent history of donor efforts

to improve aid incrementally suggests that genuine mutual accountability cannot happen through a technical fix' approach. Fundamental changes are needed to the power imbalance between donor and recipient, which has so

far been left unchallenged by official processes. As the new aid agreement set

out in this chapter proposes, this requires far-reaching changes to the institutions, objectives and incentives in the aid system.

As a starting point, a more equal donor- recipient relationship will require a scaling down of donors' own excessive demands for accountability from developing countries. Pressure for quick, demonstrable results from aid spending, coupled with heavy reporting requirements, have driven the creation of parallel donor systems, complicated procedures and intrusive conditions. In turn, these have weakened capacity, distorted incentives and skewed accountability.

2. THE CASE OF AID CONDITIONALITY

Donor reluctance to be held downwardly accountable to poor countries contrasts with donor enthusiasm for holding recipient governments to account through aid conditions. Donors use conditionality in a number of ways: as a financial accountability device, a commitment device, and as a way of inducing policy change. But the underlying concern that leads to conditionality is always the same: donors lack confidence in either the commitment or the capacity of the recipient.

Donors are right to want to demonstrate to their parliaments and publics that aid is well spent, and this requires that

developing countries properly report on how aid has been used. Yet conditionality has now moved far beyond what is necessary for basic fiduciary accountability. Since the 1980s, the International Financial Institutions (IFIs) – backed by key

G7 shareholders – have become increasingly preoccupied with the structural obstacles to growth and poverty reduction, and have sought to use loans to leverage the kinds of reforms that their Washington-based economists have deemed desirable. As a result, the average number of World Bank conditions per programme tripled between the early 1980s and mid-1990s, and by the 1990s IMF mission creep' led to it bolstering

the Bank's efforts with its own structural conditions.89

As conditionality has escalated and structural reforms have become more complex, donors have also sought to micromanage many of the day-to-day functions of government by specifying the detailed steps countries must take to improve policy, and by using Technical Assistance to place donor-funded staff, many of them expatriates, in key government positions where they

can implement and monitor change.90 For example, in 2001 when Uganda was negotiating its first Poverty Reduction Support Credit (PRSC) with the World Bank, a dozen expatriate consultants were working on the staff of the Finance

FIGURE 13  LONG TERM TRENDS IN ODA/GNI RATIOS

FOR DAC MEMBERS SINCE 1965

ODA/GNI

0.7 1970 UN TARGET  1992 EARTH SUMMIT 2002 MONTERREY OF 0.7 ODA/NGI  RIO DE JANEIRO FINANCING FOR ANNOUNCED DEVELOPMENT

0.6 CONFERENCE 0.5 0.4

0.3 0.2 0.1 0.0

BOX 5: FROM ROME TO PARIS  37

– DONOR INACTION ON HARMONISATION AND ALIGNMENT

In 2003, donors meeting in Rome under the auspices of the OECD made a series of commitments to harmonise their practices and to align to recipients' own policies and procedures. Yet two years on, OECD development ministers and developing country finance ministers meeting in Paris to review progress found that minimal action had been taken on most of the key indicators.

On conditionality, rather than reducing the number of conditions, donor harmonisation has led to more donor money being linked to a largely unreformed set of World Bank and IMF conditions. Bilateral donors have also added their own policy demands to IFI programmes.

Very limited progress has been made towards alignment with recipient planning, reporting and budgeting cycles, and still less towards using national systems. The DAC survey also revealed a gulf between donor and recipient perceptions of the extent to which country systems are being used, reflecting in part the fact that many country systems are identified, designed and implemented with donor funds in order to satisfy donor demands.

No significant progress has been made in reducing transaction costs for recipients. In particular, there has been little progress towards genuinely harmonised missions or silent partnerships, both approaches that are likely to benefit recipients substantially.

There was no evidence of progress on untying aid. Despite an OECD commitment to end tied aid to the least-developed countries, one fifth of aid to Heavily Indebted Poor Countries in Africa remains formally tied.

Ministry. In some cases, this led to

the bizarre situation of donor-funded European economists negotiating PRSC conditions on behalf of Uganda's government with the donor agencies.91

The failure of this conditionality-heavy approach to achieve the changes donors hoped for – one recent IMF study found 44% of programmes suffered major interruptions – has prompted a rethink.92 Since the mid 1990s, donors have increasingly stressed the importance of local commitment to sustainable policy reform and argued for more focused conditions that respect countries' need to plot their own development strategies and place them in the driving seat'.93 Yet the evidence on the ground suggests that old habits die hard (see Box 6). Both the World Bank and IMF are continuing to attach large numbers of conditions to their programmes; streamlining' of IMF conditionality, agreed in 2001, shows no clear evidence of progress, with one recent study finding that structural conditionality in IMF programmes is increasing, rather than falling.94 The IMF's own internal evaluations show at best uneven progress towards streamlining.95 So-called strong performers' such as Tanzania – required to meet 58 formal IMF conditions between 2000 and 2003 – demonstrate the progress that still has to be made.

For its part, the World Bank's evaluations suggest that conditions per programme have fallen slightly, but because this has coincided with a move from multiple

to single tranche PRSC adjustment programmes, there's no evidence that the burden on poor countries year-on- year has diminished.96 In other cases, conditions have become less explicit as the number of upfront prior action' conditions have been reduced and implementation details' have been shifted into increasingly complex matrixes. For example, in Ethiopia donors make disbursement decisions largely on the basis of a matrix of

over 200 indicators.

The impact of conditionality has been magnified as several of the major bilateral donors have shifted towards programme aid, often harmonising with World Bank and IMF conditions. In Vietnam, for example, the World Bank's PRSC is now being co-financed by at least seven other bilateral and multilateral donors. Despite relatively strong government ownership in

Vietnam, the PRSC matrix still contains many of the traditional structural adjustment' policy conditions, including state-owned enterprise reform, financial sector reform and trade liberalisation, not all areas in which there is genuine government commitment to reform.102

As well as increasing the burden of conditionality, this trend also has potentially serious implications for aid volatility.

3.2.1 The impact of conditionality

There is a growing consensus that the impact of conditionality on poor countries has often been negative.103 This impact is felt in at least four key ways:

distortion of democratic processes

imposition of inappropriate policy choices

generation of transaction costs

stop/start financing.

Distortion of democratic processes

One of the strongest criticisms of conditionality is that it has given significant policy influence to

donor agencies, which are outside the domestic political process and therefore not answerable to the electorate. This

is most problematic when, as has often happened, conditionality leads to poor policy outcomes. Arguably, the lack of proper accountability checks on donors can breed recklessness, leading them to promote risky and unproven policy experiments. The export by the World Bank of a largely untested power privatisation model from the UK, Chile and the USA to countries such as India and Indonesia in the early 1990s (having been heavily sold' to the Bank by major consultancy firms involved in the initial privatisation programmes) is a case

in point. 104

Conditions have also often overridden parliamentary processes. Despite growing democratisation across the developing world, IFIs have continued to bypass parliaments, a trend that sits oddly with donor insistence on good governance.' In Ghana in 2003, for example, the parliament approved a budget that included plans to raise tariffs on rice and poultry, within WTO limits. These tariff increases were subsequently rescinded after the IMF made it clear to the government that this would risk the status of the IMF programme.105 The IFIs' use of one-size-fits-all' conditions, centered on a standard package of

liberalisation, privatisation and decentralisation, also forecloses the policy choices that are the lifeblood

of a democratic system. Recently, parliamentarians from across the world have started to assert their democratic right to a greater say in development policy, through initiatives such as the International Parliamentarians' Petition (see Box 7).

The lack of accountability is compounded by the un-transparent way in which conditions are set. Typically, discussions take place between a small group of senior officials in the IFIs and the Finance Ministry, with conditions often drawn from mission objectives' drawn up by Bank and IMF staff before leaving Washington. There is usually little opportunity for civil society or parliament to engage. In Uganda, for example, there is no transparency around how conditions

in the PRSC are set, and no consultation with civil society organisations. In Vietnam, NGOs complained that limited consultations gave them insufficient time to fully scrutinise or input into the draft PRSC matrix.107 Limiting discussions in this way can strengthen the power of

the Finance Ministry, and give undue precedence to macroeconomic and monetary objectives at the expense of other policy priorities. Line ministries can find themselves being forced to implement policy choices that they

have not been party to, distorting lines

of accountability.107 In some cases, conditions are not made publicly available even after they have been agreed, allowing donors and government to engage in mutual finger pointing if things going wrong, thereby blurring lines of accountability. Even when conditions are made public, they are included in a bewildering array of technical documents, making them difficult for the public

to interpret.

Imposition of inappropriate

policy choices

Policy conditions take the initiative away from countries and often lead to donor preferences being implemented at the expense of more locally appropriate policy. This happens both because local people are widely excluded from the policy making process, and because of the lack of flexibility in the model pushed by donors, in particular the IFIs. For example, one recent overview of World Bank support to the water sector

39

found that, in 12 of 14 low-income countries where the World Bank was funding the water sector, there was some form of privatisation condition.109

As a result, many of the policy changes imposed by donors have not benefited poor people in the way that was anticipated, as a number of evaluations of structural adjustment have revealed.110 Rapid economic liberalisation by model performers' has instead often led to massive social dislocation and widening inequality, with economic growth failing to translate effectively into poverty reduction. For example, in Uganda – which experienced average rates of economic growth of 6% between 1999 and 2003 – the proportion of people living below the poverty line increased from 34% to 38%.111

Generation of transaction costs Governments of poor countries with scarce capacity are diverted from their core responsibilities by the identification, agreement, monitoring and reporting of numerous conditions to different donors. As a general rule, these accountability demands from donors are heaviest where capacity is weakest – countries with low ratings from the World Bank's Country Policy and Institutional Assessment (CPIA), which is used

by the Bank to guide lending levels – have on average twice the number

of conditions than countries with high ratings.112 Huge numbers of missions are carried out by donor agencies to monitor compliance with conditions. For example, Senegal alone hosted an average of one World Bank mission a week in 2003.113

More recently, transaction costs

have been added to by new process conditions' that involve consultation and feedback, producing additional policy documents and monitoring and reporting on results. For example, the Poverty Reduction Strategy Papers (PRSPs) that low-income countries must now produce in order to borrow from the World Bank require public meetings, the writing of a new Strategy Paper, annual reporting, revision every three years, and monitoring and reporting on results in relation to the Millennium Development Goals. These requirements may be more justifiable than policy conditions, and could in principle promote a measure

of downward accountability by

governments and donors to poor people. However, the failure of donors to harmonise their operations, the lack of alignment between PRSPs and other instruments (see Box 8) and the failure to build on existing policy making processes has raised their costs relative to any benefits.

Stop/start financing

As discussed in Chapter 2, aid is one of the least predictable sources of financing, and in itself can undermine long term budget planning and be a major cause of macroeconomic instability – thereby undermining progress towards long term development goals.116 Although donor disbursement procedures are partly to blame, aid volatility also arises from non-compliance with formal conditions, where the IMF presses the off switch' on its programme and sends a signal to other donors that a country is no longer trustworthy. This happened in Rwanda in late 2003 when the IMF suspended its Poverty Reduction and Growth Facility over a dispute about reducing the budget deficit by the equivalent of 2% of GDP, and several bilateral donors followed suit. According to the Rwandan Ministry of Finance, US$66m of development assistance was forfeited over the six months that the country went off track' with the Fund – a sum equal to half the national budget.117

In summary, current donor practice

on conditionality directly limits poor countries' ability to own – still less lead – policy priorities, and skews government accountability away from citizens and upwards to donors. The IFIs remain the arbiters of appropriate policy, and compliance with IFI conditions probably unlocks bilateral aid money to a greater extent than it did a decade ago. There is no clear evidence that the IFIs are supporting a broader range of policy options as a result of the rhetorical shift towards ownership' and partnership.' Despite streamlining of structural conditions, there are few reasons

to believe that the aggregate weight

of conditionality has reduced.

3. AN INTERNATIONAL

AID AGREEMENT

At present, sanctions in the aid relationship are one sided, and therefore so is accountability. If recipients fail to comply with donor conditions, aid can be withheld.

Even where formal conditions are not breached, donors may decide that the government has not met the spirit of its commitments and refuse to disburse aid. Donor sanctions carry real bite, especially where groups of donors have chosen to harmonise behind a single set of conditions.

In contrast, recipients have very limited scope to hold donors to account when they fail to meet their side of the bargain, beyond naming and shaming' or exhorting donors to change their ways. Moreover, there are few international forums where this can happen, and it is a risky strategy for an aid-dependent country to publicly embarrass donors into action, given that many donors have strongly resisted being held accountable for their development assistance.

For aid to become truly effective in the fight against poverty, there needs to be far-reaching reform of how it is planned and delivered. The donors' self-regulating approach has failed to make significant progress and has run its course in the absence of changes to the objectives, institutions and incentives of the aid system.

ActionAid is calling for a new International Aid Agreement (IAA) through which donors, recipients and civil society organisations can truly be held mutually accountable. Such an agreement would have four key elements:

1. clear policies from developing countries on the criteria for accepting aid
2. mutual commitments in place of one-sided conditionality, monitored transparently at the country level
3. national and international forums where donors and recipients can review progress on an equal footing, overseen by a UN Commissioner on Aid
4. new mechanisms to increase the volume and predictability of aid.

Increasingly, the international community is reaching agreement on a range of public good' issues where the costs and benefits spill across countries. The Kyoto Protocol on Climate Change; the International Criminal Court; the Heavily Indebted Poor Countries initiative; the World Trade Organisation, and the UN Convention on Trade in Endangered Species are all examples of such

BOX 6: THE LIMITS OF OWNERSHIP 41

Donors' reluctance to reduce their reliance on conditionality points to a wider failure to act on the implications of a partnership approach to development and respect the need for national ownership and leadership of policy. There are at least two reasons for this failure. First, donors' own working definitions of ownership are a key part of the problem: for many, ownership is taken to mean recipients do what we want them to do but do so voluntarily'.97 As one recent IMF paper argued, "ownership does not require that an IMF-supported programme be a government's first choice, nor that it be the programme that officials would have preferred in the absence of IMF involvementwhat is essential is that the responsible and controlling officials be committed and that opposition can be overcome".98 Such weak definitions of ownership mean that, even if countries are placed in the driver's seat of reform, donors are often still sitting in the back seat with the map, giving directions.

Second, incentives in the aid system encourage the continuing use of conditionality, especially the underlying principle that aid should either induce or respond to good policy'. Programmatic aid thereby encourages donor agencies to define themselves as a policy vanguard, seeking to buy as much reform as possible from governments that, at the margin, must remain reluctant reformers.99 This approach – reflected in the World Bank's reinvention as a Knowledge Bank'– implies that donors possess superior policy knowledge to recipients, or at a minimum pursue a more constant policy course and are insulated from vested interests'.100 This is hardly conducive to the partnership and ownership principles promoted by donors. As Ethiopian president Meles Zenawi has said, "we are not yet sure that it is fully recognised that national ownership must mean that from time to time national decisions will be made that are at variance with donor priorities or established practice. We should always be willing to listen to advice, and draw on the knowledge and experience at the disposal of donors, but we should also take responsibility for our national policiesachieve our own successes and make our own mistakes." 101

agreements. While these initiatives vary in the degree to which they have been implemented, or universally supported, they have made some important progress in ensuring that countries work together towards a common interest.

Aid, in contrast, is managed in an

ad hoc way that fails to balance the interests of donors and recipients. Many commentators have pointed to the cartel- like behaviour among agencies, including a tendency to carve up markets' that are usually former colonies or geopolitical allies, a mutual interest in avoiding inter-agency comparisons or public accountability, and a tendency to respond to shortcomings by adding new delivery models to the existing architecture.118

In many ways, architecture' is a misnomer for a construction as ramshackle as the current aid system:

the OECD-DAC plays a key role in co-ordinating and monitoring aid effectiveness efforts, but is inevitably constrained by its responsibility for representing the views of its own members – the donor countries. The IFIs act as a linchpin of the aid system through their signalling and analytical role, but their legitimacy, and therefore their ability to fairly represent donor

and recipient interests, is severely compromised by an undemocratic and un-transparent governance structure. The UN, while nominally tasked with responsibility for donor co-ordination,

is limited in its ability to impose logic and coherence on the aid system by its declining share of development aid and its own fragmentation. New mechanisms are needed as a matter of urgency, that are fully representative of both donor and recipient interests, and that have sufficient teeth to truly hold both rich and poor to account.

3.3.1 Clear financing policies from aid recipients

As a first step, recipient governments should set out clear demands on their external financing needs and the conditions under which they will and will not accept aid. Such a strategy has already been used by some recipients to strengthen their hand in the aid relationship. India, for example, recently announced that it would only accept aid from the six largest donor agencies in the country, forcing smaller donors such as the Netherlands to close down

their programmes.119 In Vietnam, the government has also taken a pro-active role in forcing donors to harmonise and align their aid to government strategies.

Once clear criteria for aid quality have been set out, recipients should use these to prioritise some donors over others. This is already occurring in some heavily indebted African countries, which are working with

BOX 7: THE INTERNATIONAL PARLIAMENTARIANS' PETITION

To mark the 60th anniversary of the creation of the World Bank and IMF, more than 1,000 Parliamentarians from 46 countries have signed a petition calling for greater democratic oversight of IMF and World Bank policies:

recognising that the IMF and World Bank have voiced a commitment to ensuring that individual countries determine their own economic policies, and noting that key economic policies continue to be imposed by both the World Bank and IMF as conditions for receiving debt relief and new loans, with the Boards of the IFIs retaining the power of veto over all measures including those in Poverty Reduction Strategy Papers. We therefore call on the Bretton Woods Institutions and their principal shareholders to ensure that the democratically elected representatives of recipient nations are the final arbiters of all economic policies in their countries. It is vital that national parliaments in recipient nations have the right and obligation to be fully involved in the development and scrutiny of all measures associated with IFI activities within their borders.'

43

UK-based organisation Development Finance International to rank and prioritise their donors according to 23 criteria that they have developed themselves.120 Criteria applied by recipients could include the extent to which the aid is concessional, flexible, predictable and tied.

Ideally, this kind of exercise could empower recipients to reject aid that does not meet their minimum criteria. However, this is often not realistic, at least in the short run – resource constrained governments may be reluctant to turn down badly needed funds, and geopolitical realities may make it difficult for recipients to refuse aid from large and politically powerful donors. But if global aid continues to increase, and if more predictable sources of development finance are secured, some of these concerns should be addressed.

Donors whose aid scores poorly or is rejected by recipients could improve aid quality directly, for example by untying, reducing the burden of monitoring and reporting, or increasing concessionality. Other options would include redirecting aid through multilateral organisations, leaving

the country or sector, or increasing aid to other recipients:

1. Redirect aid through multilateral organisations

Channelling aid through multilateral organisations in place of providing it through several smaller bilateral agencies has clear potential to reduce transaction costs for recipients. Multilateral aid has other advantages, including that it is untied and more poverty-focused and predictable than most bilateral aid.121 Such a shift would reverse the trend of recent years, in which the share of aid spent through multilateral organisations has fallen slightly. However, any shift towards multilateral agencies would have to be accompanied by comprehensive reform. Many multilateral institutions would score poorly on the real aid measure in Chapter 2, and the World Bank and IMF in particular need to radically reform their approach to conditionality. There also needs to be greater voice and accountability of poor countries within the UN system, in particular the World Bank and IMF – see Box 9 – and greater

efficiency and transparency within UN funds and specialised agencies.

2. Make use of silent

partnership agreements'

Donors who score poorly on aid quality could also engage in silent partnership' agreements, where they piggy back on another donor's capacity and disburse money through them. This can help limit planning, reporting and accounting requirements, and reduce the demands on government for policy dialogue.

There is already some experience

with silent partnerships, mostly in the education sector, and between donors with similar approaches. In Mali, for example, Sweden has channelled its money for education through the Netherlands' aid programme, while in Rwanda the Swedish and UK govern- ments have recently agreed an education silent partnership.122 However, there is substantial scope to increase the number of such arrangements, and limit

the number of donors with whom governments must directly engage.

3. Leave the country or sector and increase aid to other recipients

One of the primary reasons for poor donor co-ordination and high transaction costs is the large number of donors operating in each country. Worldwide, there are now approximately 90 official donors dispensing aid – most recently joined by the aid agencies of 10 new members of the European Union – each with their own systems, procedures and priorities. The growing number of donor agencies has been out of proportion to any increase in the value of aid, with fragmentation of aid delivery increasing most in the poorest region, sub-Saharan Africa, as figure 14 shows.123 In Ethiopia, there are over 40 active bilateral and multilateral agencies. In Zambia, for example, the education sector alone involves 20 agencies.

Such donor fragmentation' can also actively undermine the quality of public administration, locking poor countries in a vicious cycle where the demands placed on government by numerous donors has a long term effect on their ability to manage aid effectively.124

Donor fragmentation could be reduced by ensuring that fewer donors focus on each country. Donors that score badly in one particular country could be forced to leave and focus their attention elsewhere.

Even donors that score well could help to improve the quality of their aid by focusing larger volumes of resources on fewer countries. Such a shift would need to be accompanied by discussions at the international level, however, to ensure that aid is allocated fairly across countries, on the basis of poverty needs.

2. From conditionality to

mutual commitments

At present, accountability is heavily skewed towards donors and away

from recipients and poor people. A genuine partnership between donors and recipients requires a shift towards a more reciprocal and transparent relations (see Box 10). Conditionality needs to

be replaced by a limited set of mutual commitments, which could be regularly and transparently monitored as part of ongoing reviews of progress towards the international development goals.

These commitments would be most legitimate, and therefore most difficult to evade, if they were based on a minimum set of international and national standards that donors and recipients are already signed up to. Examples of such commitments could include:

The UN Declaration on Human Rights

The UN Convention Against Corruption

Environmental commitments included within UN Conventions, such as CITES, Convention on Bio- diversity and Montreal Protocol.

The Convention on the Elimination of all forms of Discrimination Against Women (CEDAW)

Such an approach could also de- politicise and simplify conditionality, and give countries more space to identify their own policy priorities.

On the recipient side, examples of such mutual commitments could include governments committing to spend money on intended beneficiaries in

a transparent and accountable way, thereby marrying donor fiduciary concerns with strong downward accountability to citizens. Effective budget processes must include full consultation with all stakeholders,

in particular poor men and women, that genuinely feeds back into policy decisions and implementation.

On the donor side, accountability could centre on existing donor commitments to reach 0.7%, to making that funding high quality and predictable, and to reforming other areas of policy that are currently discriminating against poor countries in the global economy. Through the 2002 Monterrey Consensus, donors have already committed to fully funding the Millennium Development Goals in countries that have a coherent strategy for meeting them. An agreement on how practically to do this is overdue.

Making sure that all aid is real aid –

that responds to recipient needs and

is predictable, coordinated, untied and carries minimal conditions – is a key challenge. But it is also vital that other aspects of donor policy are consistent with the development needs of the poorest countries. In particular, donors need to stop the reverse flows of the kind outlined in Chapter 1. This means that donor countries should cut carbon emissions, stop export dumping, cancel debts and ensure that poor countries have access to northern markets. Donors could be asked to produce their own Poverty Reduction Support Papers that would detail how they will

contribute to meeting the Millennium Development Goals across the full range of public policy. Precedents for such papers already exist. Sweden's Global Development Act of 2004 requires all

government ministries to develop policies that will positively contribute to the achievement of the MDGs. These PRSPs' could be independently assessed – perhaps by mixed government-civil society review teams drawn from developing and developed countries. As with aid recipients, commitments within donor PRSPs should, where possible, be drawn from already existing international commitments.

Monitoring compliance

Compliance with these mutual commitments could be reviewed

at national Consultative Group (CG) meetings. The format of CG meetings would need to be revised, with the agenda no longer controlled by the World Bank and equal space given to all stakeholders. This would include not only Ministries of Finance, but also line ministries, parliamentarians and civil society groups. All reviews of progress would be discussed publicly, with full transparency of all conditions, including progress made against each of the various commitments.

Recipients could use the CG meetings to publicly present their rankings of donor agencies, making it clear which donors are being prioritised and which donors are failing to meet their commitments. However, in cases where recipients are concerned about publicly criticising a

politically powerful donor, they could instead report the donor to a UN Commissioner on Aid, which is discussed below.

Holding donors and

recipients to account

In the case of recipient failure to

meet the mutual commitments on transparency and accountability, donors may in some cases have to withhold aid. However, in order to minimise aid volatility, a number of changes should be made to the current system of conditionality and aid disbursements:

all low-income countries should be guaranteed a minimum annual resource transfer, perhaps linked to per capita GDP or income poverty levels, to act in a similar way to a safety net'126

individual commitments should be linked to specific tranches of aid, so that failure to comply with one key condition would not cut off all aid to a country

donors should commit to not suspending any aid commitments in a given financial year, except in the most extreme circumstances.

Where donors and local civil society organisations have no confidence in the ability of the government to meet these minimum commitments, aid should be

FIGURE 14  AID DELIVERY HAS FRAGMENTED OVER TIME

DONOR FRAGMENTATION

0.85 AFRICA 0.80

0.75

0.70

ALL DEVELOPING

0.65 COUNTRIES 0.60 0.55

0.50

1975 2000

BOX 8: POVERTY REDUCTION STRATEGIES  45

– OWNERSHIP THROUGH PARTICIPATION?

Since 1999 low-income countries borrowing from the World Bank have been required to write a Poverty Reduction Strategy Paper (PRSP), which sets out a country's broad development priorities. An increasing number of bilateral donors have emphasised the importance of PRSPs in guiding their own aid allocations. The PRSP was intended both to pull together disparate policy priorities and activities into a single, strategic overview, and to change the way in which policy was formulated, by opening up the process to broad participation of government and non-government actors, thereby building local ownership and downward accountability.

Six years on from their inception, PRSPs have at best a mixed record in terms of fostering ownership and accountability. Some countries have developed their own versions of the PRSP through extensive consultation that mesh reasonably well with priorities in the budget, such as Uganda's Poverty Eradication Action Plan (PEAP). But even in Uganda, there is widespread complaint that, despite a good quality process for developing the PEAP, the IMF has continued to impose its own policy prescriptions through the conditions attached to its lending. Moreover, conditions attached to the World Bank-led Poverty Reduction Support Credit (PRSC) are only indirectly drawn from the PEAP, as shown by the disparity between the PEAP implementation matrix drawn up by the Government of Uganda, and the Bank's PRSC matrix.

Tellingly, the recent OECD-DAC Survey on Harmonisation and Alignment found no evidence that PRSPs had forced donors to adjust their own aid response. For some donors, alignment' means changing the PRSP to suit donor priorities rather than the other way round. Japan's Country Assistance Plan for Vietnam, for example, notes that as the PRSP did not make any reference to the contribution oflarge-scale infrastructure development to poverty reduction, Japan took an initiative

in reviewing the PRSP so that it would address this point.'115

In some cases such as Cambodia, PRSPs have even been written by aid-financed foreign consultants, making ownership' little more than a donor fiction. The World Bank and IMF's own recent evaluations are also critical, highlighting the reluctance of donors to discuss genuine policy alternatives and the failure of Bank and Fund loan programmes to read across to the PRSP. Many civil society groups have been critical of the use of set piece consultations to educate reform-sceptics, in place of real participation that is designed to evolve a shared reform model, and have increasingly questioned the value of their involvement. In short, donors' own reluctance to relinquish ownership of their programmes has restricted the space in which countries can pursue their own priorities, and governments can become more downwardly accountable to citizens.116

channelled through civil society. Donors should also invest in capacity building programmes with government, for example on public expenditure management.127 Aid to civil society should focus not only on basic

service delivery, but also on increasing the capacity of CSOs to hold their governments to account, for example on anti-corruption and human rights, as has happened in Cambodia and Uganda.

Aid that cannot be disbursed in this way, for example due to absorption capacity constraints, should be held in trust until such time as governance in the country improves. There is already a precedent for this kind of arrangement with UK debt relief for countries that have not yet reached Decision Point' under

the Heavily Indebted Poor Countries (HIPCs) initiative. Any debt service received from these countries is held

in trust, to be returned to once Decision Point is reached. This not only ensures that all countries eventually receive their

fair share of global aid allocations, but would also provide positive incentives for the country to reform.

Given the inevitable asymmetries in any recipient-funder relationship, it is more difficult for recipients to hold donors to account for failure to meet their commitments. Recipients would have three main options:

Publicly naming and shaming'

the donor at the country level

and exhorting them to meet commitments. This approach is more likely to succeed in less aid-dependent countries, or with smaller donors. Recipients may be wary of speaking out in public against large or politically powerful donors. Public naming and shaming could also help donors put peer pressure on each other, however. In Uganda, for example, donors are apparently using the government's partnership principles' to exhort recalcitrant donors to improve

aid quality.128

Refusing aid. As noted above, recipients could refuse aid from poor quality donors, particularly if the aid comes in the form of loans. From the point of view of donor staff, seeing their aid refused is likely to lead to embarrassment, damaged career prospects and immediate loss of employment. At a personal level, aid agency staff will therefore have strong incentives to improve the quality of their aid.

Reporting the donor to the UN Commissioner on Aid. Where neither of the first two options prove sufficient, recipients should report the donor to a UN Commissioner

on Aid. This could either be done publicly or anonymously. The UN Commissioner could then take

the issue up with the donor at the international level. If done anonymously, this would help to ease recipient fears about losing access to aid. If a major donor is reported by a number

BOX 9: THE IMF AND WORLD BANK

– THE LONG ROAD TO DEMOCRACY

Developing countries that rely on the World Bank and IMF for loans and their seal of approval often find their interests subordinated to a small group of G7 shareholders that dominate the boards. The seven largest OECD economies hold 40% of the votes, while the US wields a veto and leadership positions are carved up between Europe and North America. Africa, meanwhile, holds just two out of 26 board seats in both institutions. Since 2003, board discussions – led by South African Finance Minister Trevor Manuel – have attempted to address voice and vote' issues in the IFIs and propose an alternative, more accountable structure. Yet so far, no meaningful progress has been made. The only commitment has been some additional administrative support to the two African Executive Directors' offices, which between themselves must manage the interests of 46 countries.

of recipients for failing to meet its commitments, public pressure from within the donor country should be brought to bear in order to ensure that donor behaviour improves.

3. International forums

As well as country level monitoring mechanisms, it is also important that donors are held accountable at the international level. This is because some of the problems of unaccountable aid are systemic, and don't only exist at the country level. It is also because genuine cooperation can only happen through a multilateral framework that maintains the principle of equity and safeguards the interests of the weaker party.129

In order to ensure appropriate international level representation for aid recipients, two major reforms to the current aid architecture are needed. These are:

annual international meetings to assess donor progress against commitments made

a UN Commissioner on Aid.

International meetings

on donor progress

In place of the current donor-dominated arrangements for co-ordinating aid, there should be annual international level meetings that donors and recipients attend as equals, to highlight and discuss issues of donor and recipient performance. These could

be held under the auspices of the

UN's Economic and Social Council (ECOSOC). Regional organisations such as the Economic Commission for Africa, sub-regional bodies and individual

UN member countries could also participate. There should be full involvement of both northern and southern civil society organisations

and full transparency. Discussions

at these meetings should focus on:

Reviews of progress on aid quality donor by donor, perhaps undertaken through anonymous surveys of

the views of recipients on donor

performance. Such reviews could be undertaken by amalgamating country level assessments of each donor, or by sending out a more standard questionnaire to recipients. This could then be completed by each country, with only the overall findings by donor presented. The forum could also agree and monitor transparent targets for donor performance.

Reviews of donor progress on policy coherence, for example

on environment, trade and financial policy, to ensure that donors do not take away with one hand what they give with the other.

Reviews of how much real aid donors are providing, with a view to naming and shaming donors whose real aid ratios are low. The analysis of real aid should go beyond the preliminary allocations presented in this report, and should include detailed country-by- country analysis of the proportion ofaid that focused on poverty reduction.

BOX 10: WHAT DOES PARTNERSHIP MEAN?

Partnership is one of development's buzzwords, used by official donors and International NGOs (INGOs) alike to describe their relationship with funding recipients. Partnership' involves mutual trust and accountability and shared goals. In turn, this implies a shift of control and power, something that has been difficult to achieve in the aid system.

Several official donors have recently made efforts to apply partnership principles on the ground. This has often proven to be easier outside of government-to-government relations. For example, DFID's partnership agreements with civil society groups provide general funding to NGOs on the basis of shared principles and a high level of trust. Reporting requirements are kept to a minimum. In effect, selectivity and programme support has replaced conditionality and micromanagement of projects.

INGOs, whose own relationships with Southern civil society partners often involve similar asymmetries to donor-recipient relations, have also attempted to behave more as partners. ActionAid's own planning and learning system involves two way reporting, annual participatory reviews, transparent budgets and space for local diversity' in programme planning. These examples hold some potentially important lessons for how to move donor-recipient relations towards a genuine partnership model. 125

Discussion of aid allocations across countries, ensuring that these reflect genuine poverty priorities and that there is a fair allocation of aid across countries. This would ensure that aid is allocated in accordance with poverty reduction needs and that there are no longer donor orphans' and donor darlings'.

Discussion of best practice in donor- recipient relations, where successful experiments can be presented and disseminated.

In depth discussion of particularly problematic areas, as reported to the UN Commissioner on Aid in their role as ombudsman (see below.)

UN Commissioner on Aid

A United Nations Commissioner on Aid, reporting directly to the Secretary General, should be empowered to play a refereeing role' by overseeing the preparation of periodic public reports on donor and recipient behaviour, handling complaints about breaches

of commitments and arbitrating where there is a serious breakdown in donor- recipient relations and aid is suspended.

Sanctions against donors that fail to meet commitments will inevitably be weaker than those sanctions against recipients. However, options for sanctions against donors that repeatedly fail to honour commitments or continue to provide poor quality aid could include a reduction in decision-making powers in the UN system, such as loss of Security Council seats and loss of voting rights on the boards of the IMF and World Bank, or public naming and shaming and pressure from CSOs within the donor country.

3.3.4 Guaranteed sources of development finance

Mutual accountability will not work when donors are free to provide aid at will and when aid recipients have development needs that are severely under-funded. A key element of any new International Aid Agreement, therefore, should be the development of guaranteed sources of financing for development.

Binding long term commitments

from donors at global level

Donors should ensure that they make binding commitments to increasing their own aid budgets to 0.7% of gross national income by 2010 at the latest, and

should commit to maintaining that level for the foreseeable future – at least until extreme poverty has been eradicated. Between 2005 and 2010, donors should similarly commit to ensuring that aid flows increase towards the 0.7% target, and should ensure that such flows

are insulated from domestic budgetary pressures. Many donor countries already make long term budget commitments that effectively tie future governments and parliaments to a spending floor in areas such as defence and health; there is no reason why this principle should not be extended to development spending, which currently amounts to a very small proportion of government spending in donor countries. Donors should commit to reaching 0.7% in real' aid, a target that no OECD donor has met to date.

Long term commitments at country level In order to enable recipient countries

to properly plan and budget, aid commitments to individual countries also need to be made over a longer

time period, as Chapter 2 showed. Donors should provide recipients with projected aid envelopes for at least a three year period, with aid envelopes only changed if recipients fail to meet their mutually agreed commitments, as outlined above. Such agreements are not without precedents: the UK, for example, has recently signed an indicative long term agreement with Ethiopia for a 10- year programme of aid.

Innovative sources of

development finance

As well as increases in aid funded out

of donor government budgets, innovative sources of finance could also be used to provide a predictable income stream to support progress towards the MDGs.

A number of proposals for raising innovative sources of finance have been put forward in recent years.

These include the French government's Landau report, which proposes a tax on global currency markets and a voluntary airline tax, and the quadripartite report put forward by France, Brazil, Spain and Chile.130Over the long term, using global taxation to finance a growing share of development assistance could help to facilitate the kind of shift we are calling for in this report – away from the current model, in which aid is essentially treated as a discretionary charitable contribution to the deserving poor', towards a redistributive welfare model, where aid

is treated an entitlement for countries unable to meet basic needs from their own resources.131

The UK's proposal for an International Finance Facility (IFF), whereby future

aid flows are frontloaded to provide an increase in aid up to 2015, could also have the advantage of securing more predictable financing, at least over the next ten years, and could help to realise immediate resources for development. However, it is important that the IFF, if it

is implemented, does not result in a sharp reduction in aid flows beyond 2015.132

The IFF is also a costly form of resource mobilisation, with interest payments on IFF bonds, and the administrative costs of setting up and running the scheme, leading to a net loss in aid flows of US$108 billion over its 27-year lifespan.133 The IFF is therefore a second best option to either immediate increases in ODA/GNI ratios towards the 0.7% target, or the use of global taxation to finance development.

3.4 CONCLUSIONS

Much more aid is needed if poor

people are to be able to secure their

basic economic, social and cultural

rights, and if the MDGs are to be met by 2015. Yet as Chapter 2 has demonstrated, the aid system at present is geared more towards meeting the desires of the rich than the rights of the poor. Since its post- war inception, the aid system has mutated into a confusion of official agencies pursuing competing geo-political, commercial and development objectives, in which poor people's needs often figure as an afterthought at best. Current initiatives and processes by the OECD- DAC, UN, World Bank and other donors have failed to resolve these problems. In the absence of progress, the incremental reform agenda pursued by these agencies faces a growing credibility gap.

Phantom aid will only become real aid if donors are held accountable for providing enough good quality aid to enable poor countries to secure their basic rights.

At present, this is far from being the case. Donors are self-assessing and un- transparent, and there are few institutional incentives to support poor countries' own priorities and systems. There is

little public information on aid quality by donor, and few forums in which donors and recipients can be held mutually accountable – by each other, by civil society organisations and most importantly by poor people. Meanwhile, donors

continue to impose unfair, undemocratic and inappropriate policy conditions, which directly limit the scope for downward accountability from governments to citizens.

ActionAid therefore believes that a new International Aid Agreement is needed that: holds both donors and recipients to account for what they do and don't deliver; that replaces conditionality with mutual obligations, and that ensures guaranteed sources of development financing.

   50

CHAPTER 4 – CONCLUSIONS AND 51 RECOMMENDATIONS

"OVERCOMING POVERTY IS NOT A GESTURE OF

CHARITY. IT IS AN ACT OF JUSTICE. IT IS THE PROTECTION OF A FUNDAMENTAL HUMAN RIGHT, THE RIGHT TO DIGNITY AND A DECENT LIFE. WHILE POVERTY PERSISTS, THERE IS NO TRUE FREEDOM."

NELSON MANDELA, 2005

52

Rich countries are starting to recognise that much more aid is needed if poor people are to secure their basic rights. Yet if aid quality remains unchanged, increasing aid alone will do little to help poor people. At present, far too much aid is provided in ways that subordinate the needs of the poor to rich countries' political and commercial priorities.

ActionAid believes that all donors have the obligation to provide 0.7% of their national income in real' aid – money that is genuinely available to poor governments to help them secure basic rights. At present, this is far from being the case. As we show in this report, almost all donors are falling far short of meeting even the official 0.7% target, and more than 60% aid flows are phantom.' Not one donor has met the real 0.7% target, with DAC donors giving an average of only 0.1%. Meanwhile, the G7 – the seven richest and most powerful countries in the world – give only 0.07% of their national income, or one tenth of what they should provide. And these paltry amounts pale in comparison with South-North flows, which stood at US$710 billion in 2003.

If all aid is to become real, the international aid system must radically change. Accountability relations, heavily skewed towards donors at present, must be reformed. There must be a new International Aid Agreement that holds donors and recipients mutually accountable for securing the basic rights of the poor. Donor imposed policy conditionality must be replaced by mutually agreed obligations; secrecy replaced by transparency; upward accountability by mutual accountability; and consultation' by true participation'.

ActionAid believes that the following reforms must be put in place as a matter of urgency:

1. All donors must commit to providing at least 0.7% of their national income in real aid, by 2010 at the latest. This means that:

1. Aid must be allocated according to poverty needs. At least 70% of all aid must go to least-developed countries and other low-income countries, with the remaining 30% allocated to poverty reduction needs in middle-income countries. Within countries, all aid should explicitly be provided solely for poverty reduction, rather than meeting commercial, political or strategic objectives of the donor.
2. While debt relief is important, all funding for debt relief should be additional to, rather than drawn from, aid budgets, and debt relief funding should not count towards ODA/GNI ratios.
3. Technical Assistance must be untied and purchased locally to the greatest extent possible, using pooled donor funding if necessary. There should

be comprehensive reviews of the effectiveness of TA in each country.

TA should be genuinely technical', and share skills rather than steer the country in a pre-determined reform direction. There should be a greater emphasis on capacity building rather than gap filling, with the extent of any capacity building that takes place monitored independently. The share of global aid budgets allocated to TA should also be sharply reduced.

4. All aid must be fully untied, and deliberate policies put in place to encourage local procurement. There should be regular monitoring of the proportion of donor procurement that is going to the recipient country or to its neighbours.
5. Donors must fully implement the Rome Declaration on Harmonisation and Alignment, and there should be binding targets with regular monitoring processes set up for reviewing progress.

6. All aid should be committed for at least a three year period, and only withheld under extreme circumstances (such as conflict, widespread human rights abuses or endemic corruption).
7. Funding for refugees in the donor country should not be counted as part of aid budgets.
8. Donors must reduce administration costs, and must be more transparent about the breakdown of expenditures within their administration budget.

2. There must be a new International Aid Agreement, in which donors and recipients are held mutually accountable. This will require:

1. Recipient governments being more proactive in holding donors to account. Recipients should develop clear financing policies that set out minimum standards for donor aid. Aid that fails to meet these minimum standards should be rejected. All donors, and in particular those with poor quality aid, should be encouraged to channel greater proportions of aid through a reformed set of multilateral organisations, as well as making use of silent partnership agreements and limiting the number of countries in which they provide aid.
2. A shift from donor imposed conditionality to mutual accountability. Intrusive donor imposed policy conditions should be abandoned and replaced by an agreed set of mutual obligations between donors and recipients. Recipients should commit to spending money on its intended beneficiaries in a transparent and accountable manner. Donors should agree to fully finance the MDGs, to provide good quality aid, and to ensure that other policies serve to help, rather than hurt, developing countries. Commitments should, where possible, be based on already existing international commitments. Compliance with these commitments should be monitored publicly during annual Consultative Group meetings.

3. Sanctions on both recipients and donors that fail to meet mutually agreed conditions. For recipients, this could include withholding aid, while for donors it could include refusing aid, public naming and shaming, and being reported to the UN Commissioner on Aid.
4. Annual International Aid Forums should be convened, by the UN's ECOSOC or a similar body, to monitor donor and recipient performance. Donors and recipients should attend as equals and there should be full participation of northern and southern CSOs, as well as full transparency.
5. A new UN Commissioner on Aid.

A new UN Commissioner on Aid should be created to provide impartial oversight of whether donors and recipients are meeting their commitments. This post could report directly to the UN Secretary General.

6. Guaranteed sources of development finance. Mutual accountability will never work unless aid recipients are sure that they will receive all the financing they need to meet the MDGs. Donors should make binding commitments to reach the real 0.7% target by 2010 at the latest. Donors should also work to complement aid flows with innovative sources of finance such as global taxation. Aid should be made fully predictable and reliable at the recipient country level.

REFERENCES 54

1 UN, Article 22,

UN Universal Declaration of Human Rights, 1948 2 UN, Article 1,

UN Declaration on the Right to Development, 1986 3 UN, paragraph 11,

UN Millennium Declaration, 18 September 2000

4 Calculated from 2004 World Bank and UN data.

World Bank World Development Indicators 2004, Washington, DC, 2004

5 Foster, M and Keith, A,

The Case for Increased Aid, 2003

6 World Bank, Assessing Aid: What Works,

What Doesn't and Why, Washington DC, 1998

7 For example, see The Telegraph. This is where our

money goes' by Christina Lamb, October 29 2000 8 In the case of Mobutu, donor agencies were

aware of the misappropriation of donor funds, but

no action was taken on the IMF and World Bank

boards, partly due to US concerns about pushing

Zaire into the Soviet sphere of influence. For

example, see Leslie, Winsome J, Zaire: Continuity

and Political Change in an Oppressive State,

Boulder: Westview, 1993

9 OECD-DAC estimates that tying inflates aid costs

by 15-30% – e.g. see policy brief, Untying Aid to the Least Developed Countries, OECD Observer 2001

10 Government of Uganda Poverty Eradication

Action Plan 2004/05-2008, 2004

11 Levine, R. et al, Millions Saved: Proven Successes

in Global Health, Centre for Global Development, Washington, DC

12 UN Millennium Project, Investing in Development:

A Practical Plan to Achieve the Millennium Development Goals, New York, 2005

13 Mokoro Consulting, Aid Modalities in Ethiopia:

a report to Development Cooperation Ireland, August 2004

14 De Renzio, P, The Challenge of Absorptive

Capacity, ODI report for DFID, September 2004

15 Hansen and Tarp (2000), cited in Foster, M and

Keith, A, The Case for Increased Aid, 2003

16 Clemens, Radelet and Bhavani, Counting

Chickens When they Hatch: The Short Term Effect of Aid on Growth, Working Paper 44, Center for Global Development, Washington DC, 2004

17 IMF, Tanzania's growth process and success in reducing poverty, Working Paper 05/35, 2005

18 Commission for Africa, Our Common Interest: Report of the Commission for Africa, 2005

19 Fagernas, S. and J. Roberts, Fiscal Impact of Aid:

A survey of issues and synthesis of country studies of Malawi, Uganda and Zambia, ESAY Working Paper no.11, ODI, London, 2004

20 AFRODAD, Reality of Aid: Does Africa Need Aid?

Harare, 2002

21 World Bank, Assessing Aid, 1998; World Bank,

The Role and Effectiveness of Development Assistance: Lessons from World Bank Experience, Research Paper, 2002

22 Commission for Africa, Our Common Interest: Report of the Commission for Africa, 2005

23 Interview with NGO Forum,

Uganda, 9 March 2005

24 ActionAid USA and ActionAid Uganda, Blocking

Progress: how the fight against HIV/AIDS is being undermined by the World Bank and International Monetary Fund, Briefing Paper, September 2004;

Oxfam et al, Where is the Impact?, Briefing paper, April 2003

25 UN Millennium Project, Investing in Development:

A Practical Plan to Achieve the Millennium Development Goals, 2005

26 For example, see

http://www.internationalbudget.org

27 World Bank, World Development Indicators 2004,

Figure 1c, 2004.

28 World Health Organization, 3 by 5 progress report:

investing in a comprehensive health sector response to HIV/AIDS - scaling up treatment and accelerating prevention, 2004

29 FTI secretariat, Education for All Fast Track Initiative Status Report, November 2004

30 World Bank, Ethiopia: Accelerating the Progress

Towards the MDGs: Impact on Aid Requirements, Unpublished study, 2004

31 World Bank, Global Development Finance 2004,

Volumes 1 and 2, Washington, DC, 2004

32 Boyce and Ndikumana, Is Africa a net Creditor?

New Estimates of Capital Flight from Severely Indebted Low Income Sub-Saharan African Countries, 1970 –1996, 2000

33 Oxfam International, Rigged Rules and

Double Standards

34 Simms, A, Magrath, J and Reid, H, Up in Smoke: Threats from, and Responses to, the Impact of Global Warming on Human Development, 2004

35 World Development Indicators, 2004

36 Simms, A, Real World Environmental Outlook,

cited in Pettifor, ed Real World Economic Outlook, 2003

37 One notable attempt to rank donor performance

has been the work of the Center for Global Development in Washington DC, whose Index of Donor Performance 2004 uses a similar methodology to that used in this paper

38 Unless otherwise stated, all data is from the

OECD DAC Development Cooperation Report 2004, Statistical Annex

39 This is roughly comparable to estimates by the

UN Millennium Project, which found that 24% of bilateral aid and 54% of multilateral aid is available to finance investments on the ground

40 This figure differs from the donor-by-donor

breakdown shown in Figure 5, which excludes aid money that is not specified by region

41 OECD-DAC; EU aid website

42 Kaiser Daily Report. November 1st 2004.

US Government Accounting Office Report to International AIDS Conference'.

43 Government of Japan, Country Assistance Plan

for Vietnam, page 1, 2004

44 OECD DAC Development Cooperation Report

2004 (Pakistan, Kyrgyztan, Uzbekistan, Tajikistan and Turkmenistan)

45 OECD DAC Development Cooperation Report

2004

46 USAID Greenbook

http://qesdb.cdie.org/gbk/index.html

47 USAID Greenbook

http://qesdb.cdie.org/gbk/index.html

48 Interview with Embassy of Japan in Vietnam,

18 February 2005

49 Interview with independent consultant in Vietnam,

18 February 2005

5 Donor-by-donor data excludes aid that is not

specified by region

51 Note that these figures include funding from

multilateral donors that has been imputed back to bilaterals based on their contributions to the multilaterals. The source for all data is the OECD DAC Development Co-operation Report 2004, Statistical Annex, Table 26.

52 IDA, Global Monitoring Report, HIPC Status of

Implementation Report, September 2005

53 Of course, just because debt service rose it does not mean that debt relief prevented debt service from rising even further. We do not have figures

on this. Data source is GDF for all developing countries, World Bank HIPC/debt website for HIPCs.

54 Note: 2000/01 ODA figures including debt relief

refer to 2000 statistics, 2001/02 to 2001 and so on. 55 World Bank, Can Africa Claim the 21st Century?

Oxford University Press, New York, 2000

56 Defined as either receiving three or more

contracts over the time period, or receiving contracts with a value of more than £1 million. All data taken from DFID website, www.dfid.gov.uk/procurement/constractslet.asp

57 Siddiqui, F, Strickler, C and Vinde, P, Capacity

Building Practices of Cambodia's Development Partners: Results of a Survey, 2004

58 ActionAid UK, Money Talks: how aid conditions

continue to drive utility privatisation in poor countries, 2004

59 Interview with DFID in Vietnam, 16 February 2005 60 Oxford Policy Management, A Vision for the

Future of Technical Assistance in the International Development System, Report, July 2003

61 Informal Report prepared for the Consultative

Group Meeting for Vietnam, Hanoi, Moving Towards 2010: Vietnam Partnership Report, 1-2 December 2004

62 We are grateful to Andrew Rogerson at the ODI

for suggesting this methodology.

63 Johnson , A, Martin, M and Bargawi, H,

The Effectiveness of Aid to Africa Since the HIPC Inititiative: Issues, Evidence and Possible Areas for Action, Report for the Commission for Africa, 2004

64 Calculated from JBIC website. Only includes

contracts of more than 100 million Yen

65 Interview with US funded NGO working on health

in Cambodia, 22 February 2005

66 Reality of Aid, 2002

67 Oxfam GB, Ethiopia Financing for the MDGs,

Unpublished paper, 2004

68 Data for US and Italy taken from Oxfam 2004,

Paying the Price : Why rich countries must invest now in a war on poverty. All other data from OECD-DAC

69 Manning, R, How can the Development

Community help to achieve greater progress towards the Millennium Development Goals?, WIDER Angle', 2004

70 DAC 2003, page 56, based on calculations by

Van de Walle and Johnston, cited in Trocaire, Aid Effectiveness, 2005

71 Interview with Finance Ministry official,

in Ethiopia April 4th 2005

72 UNDP/DFID, Aid Transaction Costs in Vietnam,

2000

73 Johnson , A, Martin, M and Bargawi, H,

The Effectiveness of Aid to Africa since the HIPC Initiative, 2004

74 Bulir, A and Hamann, J, Volatility of Development

aid: From the Frying Pan into the Fire?, 2005

76 Oxfam International, Paying the Price : Why rich

countries must invest now in a way on poverty, 2004 77 ActionAid, Transaction costs in aid: Case studies

of sector wide approaches in Zambia and Senegal, Paper for UNDP Human Development Report, 2004

78 Interivew with Ministry of Finance official in

Ethiopia, 4 April 2005

79 Interview with Ministry of Finance official in

Uganda, 7 March 2005

80 www.aidwatch.org.au

81 DFID Civil Society Challenge Fund Guidelines 82 We do not include multilateral agencies here

because data is not available for either the World Bank or the regional development banks. However, observation of World Bank spending suggests that administrative costs for the multi- laterals are probably even higher than for bilaterals

83 OECD-DAC, Survey on Progress in Harmonisation

and Alignment, 2004

84 Ibid.

85 www.oecd.org; Trocaire, The International Aid

System: what's next?, unpublished paper, 2005

86 Helleiner, G, Local Ownership and Donor

Performance Monitoring: New Aid Relationships in Tanzania?, 2005

87 Oxfam, Paying the Price : Why rich countries must

invest now in a war on poverty, 2005

88 OECD-DAC High Level Forum, Harmonisation,

Alignment, Results: Report on Progress, Challenges and Opportunities

89 Easterly, W, The Elusive Quest for Growth:

Economists' Adventures and Misadventures in the Tropics, MIT Press, Cambridge, 2001

90 For example, see White, H, and Dijkstria,

Programme Aid and Development: Beyond Conditionality, Routledge, London, 2003

91 Interview with former ODI fellow, March 2005

92 Ivanova, A, Mayer, W, Mourmouras, A, and

Anayiotos, G, What Determines the Implementation of IMF-Supported Programmes?, IMF Working Paper WP/03/8, International Monetary Fund, Washington, DC, 2003

93 World Bank. 2004. Development Policy Lending

Guidelines. Washington, D.C: World Bank; HM Government. 2005. Partnerships for poverty reduction: rethinking conditionality. Policy paper; World Bank. 2005. Conditionality Revisited: Concepts, Experiences and Lessons. Washington, D.C: World Bank.

94 Debt Relief International, The Role of the IMF in

Low Income Countries: Study for Swedish Ministries of Finance and Foreign Affairs, 2004

95 IMF, Lessons from Real-Time Assessments of

Structural Conditionality, March 2004; IMF-IEO, Draft Issues Paper For An Evaluation Of Structural Conditionality In IMF-Supported Programs, 2005

96 World Bank, Conditionality in World Bank-

supported Policy-Based Lending, 1980-2003, OPCS presentation, 2004

97 Helleiner, G, External Conditionality, Local

Ownership and Development, in Freedman, J (ed), Transforming Development: Foreign Aid for a Changing World, University of Toronto Press, Canada

98 Boughton, JM, Who's in Charge? Ownership and

Conditionality in IMF-supported programmes, IMF Working Paper EP/03/191, IMF, Washington, DC, 2003

99 Collier, P, 2000, Consensus Building, Knowledge

and Conditionality: http://orion.forumone.com/ABCDE/files.fcgi/139_ collier.pdf

100Wood, A, Conditionality: Past, Present and Future,

Issues Paper for ActionAid UK, 2004 101Presidential address to 2002 World Bank

Consultative Group meeting, cited in Oxfam,

Ethiopia Financing for the MDGs, internal

working paper

102Interviews with Oxfam Vietnam and ActionAid

International Vietnam, 16 and 18 February 2005 103For example, see DFID/FCO/HM Treasury,

Partnerships for Poverty Reduction: rethinking

conditionality, 2005

104Dubash, NK, et al, Power Politics: Equity and

environment in electricity reform, World Resources Institute, Washington, DC, 2002

105Christian Aid, Struggling to be Heard:

Democratising the World Bank and IMF, 2003 106Interview with Uganda Debt Network, 7 March

2005; interview with NGO Forum Uganda, 9

March 2005

107Interview with Oxfam GB in Vietnam,17 February 2005; interview with ActionAid International Vietnam

108A.Wood, Conditionality Past, Present and Future 109Wood, A, International Financial Institutions,

Conditionality and Privatisation of Water and Sanitation Systems: Report for WaterAid, WaterAid, London, 2003

110IMF, External Evaluation of the ESAF, IMF, Washington, DC, SAPRIN, 1999

111Uganda PEAP

112World Bank, Conditionality in World Bank- supported Policy-Based Lending, 1980-2003, OPCS presentation, 2004

113OECD-DAC, Survey on Progress in Harmonisation and Alignment, 2004

114Japan Country Assistance Plan for Vietnam, page 17, April 2004

115For example, see World Bank, The Poverty Reduction Strategy Initiative: An Independent Evaluation of the World Bank's Support Through 2003. World Bank/OED, Washington, DC, 2005; ActionAid USA, Rethinking Participation, Briefing Paper, 2004; Oxfam International, From Donorship' to Ownership? Moving Towards PRSP Round Two, Oxfam International, Oxford, 2004

116Bulir, A and Hamann, J, How volatile and unpredictable are Aid Flows, Implications, IMF Working Paper 01/167, Washington, 2001

117Oxfam International, The IMF and the Millennium Development Goals, Oxfam: Oxford, 2003

118World Bank, Aid Agency Competition, Public Policy for the Private Sector, Note no. 277, 2004;

Rogerson, A and Maxwell, S, Aid: What's Next? ODI Opinions no. 27, 2004; Easterly, W, The Cartel of Good Intentions, Foreign Policy, July –August: 40:49, 2002

119Financial Times, India Opts to Decline Aid from all

but Six Countries, July 8 2003

120 Johnson , A, Martin, M and Bargawi, H, The

Effectiveness of Aid to Africa since the HIPC Initiative: Issues, Evidence and Possible Areas for Action, Background Paper for the Commission for Africa, 2004

121Ibid.

122www.minbuza.nl; the UK committed £7 million

over three years in February 2005 to a silent partnership with France for basic education in Niger

123Rogerson, A and Maxwell, S, Aid Industry Needs

Slimming Down in Order to Tackle World Poverty, The Guardian, September 27 2004. Donor fragmentation as measured by Knack and Rahman is the number of donors relative to the average share of the project market'

124Knack and Rahman, Donor Fragmentation and

Bureaucratic Quality in Aid Recipients, Policy Research Working Paper 3186, World Bank: Washington, DC, 2004

125ActionAid, Accountability, Learning and Planning

System, ActionAid: IDS. 2001; The New Dynamics of Aid: Power, Procedures and Relationships, Policy Briefing no.15

126For example, see Lockwood, M, The State They're

In: Poverty and politics in Africa and the agenda for international action, Monograph, 2005

127Rogerson, A. and de Renzio, P. 2005. The Seven

Habits of Effective Aid: Best Practices, Challenges and Open Questions'. ODI Opinions, February 2005.

128Interview with DFID official in Uganda, 9 March

2005

129AFRODAD, Governance, Transparency and Mutual

Accountability Within the Aid Regime, Note, 2005 130Working Group on new international financial

contributions, Landau report, December 2004;

Atkinson, AB, (ed), New Sources of Development

Finance, UNU-WIDER November 2004

131Fitzgerald, V, Rethinking Development Assistance:

The Implications of Social Citizenship in a Global Economy, Paper for UNRISD Globalisation and Citizenship Conference, 1996

132Development Initiatives, The International Finance

Facility: Briefing Note for CONCORD Working Group on Financing for Development, 2003

133World Development Movement, The International

Finance Facility: Boon or Burden for the Poor?, 2005

ACKNOWLEDGEMENTS AND CONTACTS 56

ActionAid International is a unique partnership of people who are fighting for a better world – a world without poverty.

ActionAid International PostNet Suite #248 Private Bag X31 Saxonwold 2132 Johannesburg

South Africa

Telephone

+27 (0) 11 880 0008

Fax

+27 (0) 11 880 8082

Email mail.jhb@actionaid.org

Website www.actionaid.org

International Head Office Johannesburg

Asia Region Office Bangkok

Africa Region Office Nairobi

Americas Region Office Rio de Janeiro

ActionAid International is registered under Section 21A of the Companies Act 1973. Registration number 2004/007117/10

Design: Ranch

This report was researched and written by Romilly Greenhill and Patrick Watt of the Aid and Accountability team in ActionAid International (UK).

For more information, please contact:

romilly.greenhill@actionaid.org or patrick.watt@actionaid.org

The report received the support and input of a large number of colleagues and partners. Particular acknowledgements are due to Fern Leathers, Chris Jordan, Daniel Bekele, Fikre Zewdie, Amanda Serumaga, Florence Apuri, Keshav Gautam, Sopheareak Meas, Nguyen Hoai Chau, Shona Sarkar, Ramesh Khadka, Nguyet Tran Que, Steve Tibbett, Tom Sharman, Louise Hilditch, Felicity Daly, Simon Wright, Taaka Awori, Tony Durham and Stephanie Ross in ActionAid, Arry Fraser at Oxfam, and to Matthew Martin at Development Finance International, Elena Sisti at the New Economics Foundation, Charles Mutasa at AFRODAD, Judith Randell at Development Initiatives and Andrew Rogerson at ODI

Photographic Credits Cover

Liba Taylor /ActionAid

Page 6

Jenny Matthews/ActionAid

Page 14 Gideon Mendel/Corbis/ActionAid

Page 23 Sophia Evans/NB Pictures/ActionAid

Page 32

Chris Stowers/Panos/ActionAid

Page 40

Gideon Mendel/Corbis/ActionAid

Page 50

Gideon Mendel/Corbis/ActionAid

June 2005