VI. OFFICIAL DEVELOPMENT ASSISTANCE: EFFECTIVENESS
- Aid is effective and increasingly so. In countries which follow appropriate policies, a 1 percent injection of GDP can produce a 0.6 percent increase in growth, a 1.2 percent decline in poverty and a similar decline in infant mortality. In 1990, a one-time aid increase of US$1 billion allocated across countries in proportion to existing ODA would have permanently lifted an estimated 105,000 people out of poverty. By 1997-98, US$1 billion would have lifted 284,000 people out of poverty. In other words, the estimated poverty-reduction productivity of ODA nearly tripled during the 1990s.
- Raising questions about the effectiveness of aid may deflect attention from the insufficient volumes of aid which donors provide. On occasion it may suit donors for attention to be focussed on questions of aid effectiveness. That said, we believe strongly that there need not be a tension between aid volume and aid effectiveness, and that improvements in aid effectiveness are essential if increases in aid volume are to be generated. As such, we are firmly of the view that donors and recipients of aid must work together to improve the effectiveness of aid. As Clare Short told us:
"More effective use of aid could produce extra resources; a greater focus on the poorest countries and better collaboration amongst donors, including the pooling of budgets, could also maximise the efficiency of aid in diminishing poverty. Overall better allocation, co-ordination and untying, by bilateral donors and by international institutions, could make aid 50 percent more effective".
The stakes are high: aid can, and must, be made more effective. This will require concerted action from both the donor and the recipient ends of the aid relationship to improve the allocation, design and delivery of aid.
The Good Recipient: Committed to Poverty Reduction and Good Governance
- The Monterrey Consensus emphasised that developing countries have the primary responsibility for their own development, stating that:
"Good governance is essential for sustainable development. Sound economic policies, solid democratic institutions responsive to the needs of the people and improved infrastructure are the basis for sustained economic growth, poverty eradication and employment creation. Freedom, peace and security, domestic stability, respect for human rights, including the right to development, and the rule of law, gender equality, market-oriented policies, and an overall commitment to just and democratic societies are also essential and mutually reinforcing".
We share these views, and believe strongly that developing countriestheir governments, but also civil society, business and the mediahave the primary responsibility for ensuring that aid is used effectively. The ineffective use of aid or the diversion of aid into the pockets of elites pushes the poor deeper into poverty and robs them of the assistance provided to them.
- The pre-requisite for poverty reduction and the effective use of aid is the commitment of the recipient government to poverty reduction. Whilst emphasising the importance of country-specific factors, ODI suggested to us that the effective use of aid is built on the following foundations: effective governmental systems for public expenditure management; a tolerable degree of macroeconomic stability; political stability; physical security; and a framework of law which enables enterprise and encourages businesses to invest. In addition to the various dimensions of good governance, aid effectiveness depends too on the ways in which aid resources are spent. As Judith Randel of Development Initiatives told us:
"investment in human development is actually effective, it does deliver. Investment in public health, investment in education, are extremely effective ways of spending money but we spend a tiny, tiny amount of the aid budget on that. We [the DAC donors on average] spend 2 percent of aid on basic health and 1.5 percent of aid on basic education globally. We have not really invested that much in the things that work in the context we think they are likely to work".
- At the Copenhagen Social Summit in 1995 it was proposed that donor and recipient governments come to bilateral agreements whereby donors would allocate 20 percent of their ODA to basic social services including basic education, primary health care, reproductive health care and family planning, nutrition, safe drinking water and hygiene, if recipients agreed to allocate 20 percent of public expenditure to enable universal access to basic social services. This 20/20 Initiative offers aid donors and recipients the opportunity to take joint responsibility for a policy of social development. DFID does not have a target for the proportion of its aid spent on basic social services, but in 1997/8 18 percent of the UK's aid went to basic social services compared with a DAC average of 11 percent. If the UK is serious about improving the effectiveness of aid, scarce resources must be invested in the sectors which are likely to provide the best returns in terms of poverty reduction. We recommend first that DFID considers adopting a target for its aid allocation to basic social services, second that DFID encourages its fellow DAC members to increase their allocations in line with the 20/20 Initiative, and third that DFID continues to encourage the recipients of its aid to prioritise spending on universal access to basic social services.
The Good Donor: Providing Sufficient, Timely and Quality Aid
- Effective aid relationships require donors to play their part too, and to be accountable for the parts they play. As Belen Vazquez of Action Aid succinctly put it: "A good donor is a donor that accepts responsibilities in the process. The responsibilities of the donor are to provide sufficient aid, timely aid and aid of good quality". ODI emphasised the importance of donors engaging seriously with the development strategy and process, the need for donors to provide flexible resources to implement the strategy, and, the importance of donors being accountable for the delivery of timely and efficient aid using simple and transparent procedures. Quality aid is focussed on the elimination of poverty and the attainment of the MDG targets. Quality aid does not require the recipient to spend the aid on goods and services provided by donor country firms. Quality aid is untied aid. In this, we are in full agreement with DFID, which, having led the way in untying all its bilateral aid in April 2001, has been pushing other donors to follow suit, or at least to make progress by untying bilateral aid between EU Member States.
- There are various reasons why tied aid is ineffective aid. Firstly, tied aid has multiple objectives, including supporting donor-country firms; therefore its focus on poverty reduction is inevitably diluted. Secondly, tied aid cannot generate the local supply networks which would multiply the impact of an injection of aid. This was exemplified by a case we saw for ourselves in Ghana and which Clare Short mentioned in her evidence, in which the enterprise of Tema Tropical Cables, in gearing itself up to produce wire for electrification projects, was thwarted by the tying of aid, and the purchase of wires for electrification, back into Europe and North America. Thirdly, tied aid can force poor countries into the ridiculous situation of having to purchase incompatible services or componentsof vehicles, of hospitals, of schoolsfrom a range of different donors. Fourthly, tied aid hinders competition between potential providers of goods and services, leading to inefficiency, and is arguably in contravention of EU internal market and free competition rules. As Belen Vazquez of Action Aid put it "Development assistance should not be used as a form of subsidising home companies". The OECD estimates that untying all aid would in effect provide an additional US$5 billion of resources for development every year. In May 2001 the OECD DAC donors did agree to untie aid to the least developed countries, but little progress has as yet been made. We applaud the Government's actions in untying UK aid, and encourage the Government to continue its efforts to persuade its more recalcitrant European partners to untie their aid too.
- As well as untying its aid, a good donor harmonises its aid procedures with those of other donors. Harmonisation and coordination reduces the bureaucratic burdens placed on recipient countries, improves the quality of aid by reducing its transactions costs, and enables poor countries to focus their resources on poverty reduction rather than complying with a range of complex and varied procedures. As Tony German of Development Initiatives told us, "what makes a good donor is not making the fragmentation even worse by everybody having their own individual little programme. ... it is better for more countries to get together to do joint programming rather than having to have their own little bit, which only makes it much more difficult for host finance ministries to track the money and then to be accountable for how the money is being spent". Ideally, and subject to an assessment of the risks involved and the requirement of accountability, aid is best provided through direct budget support. Aid delivered in this manner can be distributed nation-wide in pursuit of poverty reduction, rather than spent on isolated projects; it is also more likely to result in sustainable ongoing development. In addition, the injection of resources directly into a government's revenue stream minimises transactions costs, and helps to build the capacity of the recipients' systems for financial management and accountability. But, as we emphasise further below, the choice of aid instrumentsas well as the allocation of aid and the selection of development interventions in generalmust take account of country specifics rather than follow a preset formula.
Aid as Investment: Targeting, Conditions and Low-Return Countries
- As both the Secretary of State for International Development and the Chancellor of the Exchequer have been at pains to emphasise, aid ought to be conceptualised as investment in the developmental capacity of poor countries rather than as charity. Clare Short argued that, "we have got to get international development out of the charity box into the investment and economic box", investing aid to enable currently poor countries to develop and ultimately to engage with the increasingly global economy. If aid is to be regarded as an investment, it should be targeted to countries where it will secure the best returns in terms of eliminating poverty, and with conditions attached to ensure that it is invested wisely. If invested aid produces good returnsin part through sensible targeting and the attachment of conditionsthen it is likely to encourage further aid. We fully agree with DFID and HM Treasury; aid should be conceptualised as investment. However, careful attention does need to be paid: first, to the allocation criteria which drive aid/investment decisions; second, to the implications for countries which are seen as offering low rates of return; and, third, to the nature of the conditions which investors might attach to aid flows.
Diagram Source: Committee's own.
- The current allocation of aid is shaped by a range of factors, including historical and cultural ties, and the realities of geo-politics. Efforts to improve the effectiveness of aid by improving its allocation are built upon conceptual models of the development process and, more specifically, of how aid contributes to poverty reduction. The international community's efforts to improve the effectiveness of aid depend upon the quality of these models, and their sensible interpretation and application.
- Particularly since the World Bank's report on "Assessing aid" was published in 1998, the widely-accepted wisdom amongst donors has become that aid should be targeted to poor countries which follow what the World Bank and the donor community deem to be sensible policies in terms of good governance, budgetary and macro-economic stability, and economic liberalisation. The simple model is that aid works by improving institutions, contributing to economic growth and raising income levels, and hencemaking certain assumptions about the distribution of income gainsreducing poverty. Such mechanisms are said to work best when policies supportive of economic growth, and ideally of pro-poor economic growth, are pursued. Indeed, the World Bank's model suggests that in countries with poor policies, aid is ineffective or even harmful. In short, the received wisdom has been that aid works best in, and therefore should be targeted at, poor countries which pursue World Bank-approved policies. Similarly, the model suggests that conditions should be attached to aid to encourage policy reform.
- Over the last three or four years, however, there has been some debate about this model of aid effectiveness, specifically about the relative importance of the right policy environment in determining the effectiveness of aid. Critics of the World Bank's approach to aid allocation have suggested that aid can be effective across a range of policy environments. Consequently, there are questions about the appropriate criteria for targeting aid and selecting countries, and also, about the sorts of conditions which ought to be attached to aid. DFID's position seems to be that the World Bank's adviceto target aid to poor countries with good policyis broadly right, but that aid can also work in poor policy environments by helping recipients who are committed to poverty reduction to pursue policy reform. As we were told by Suma Chakrabarti, the DFID Permanent Secretary, in an evidence session about DFID's Annual Report, his Department's application of allocation criteria may therefore be less "mechanistic" than that of the World Bank.
- This is not the place to rehearse complex and technical debates about aid effectiveness, but the very fact that there are such debates suggests to us that donors ought continuously to re-examine their models of aid and aid effectiveness, to ensure that their aid allocations and their use of conditionalities contribute to maximising the impact of aid. It would be helpful if DFID made explicit its underlying model of the ways in which aid is effective and its aid allocation criteria, and encouraged other donors to do the same. Such transparency is fundamental to any notion of a shared and jointly-owned development partnership, and would provide developing countries with the information necessary toif they so wishimprove their performance in terms of the criteria.
- Our view, as we have stated emphatically in relation to EU aid, is that aid must be targeted firstly towards poor countries and countries with large numbers of poor people. A second important criterion is the nature of a country's policy environment, or more precisely the prospects of a country's policy environment changing for the better through the establishment of an aid-related policy dialogue. Aid resources must not be squandered in countries which lack the commitment to development and poverty reduction. In addition, aid must not be used to prop up governments that are not furthering the interests of their people. This must not mean that we forget such countries, or, more specifically, the millions of poor people who live in such countries. Rather, the nature of donor engagement with such countriescountries which the World Bank labels Low-Income Countries Under Stress or LICUSought to be different. As Clare Short told us: "Even where there is a lack of reform but where there are large numbers of poor people, if you can get it [aid] in effectively, it still has high effectiveness in poverty reduction".
- Donor countries need to have a range of strategies for using aid (not to mention other development interventions)from providing direct budget support, to supporting sector-wide approaches, to financing projects, to building the capacity of civil societyso that countries can be considered on a case-by-case basis and aid can be used in the most effective way. In addition, countries should not be doubly-punished when their reforms are knocked off track by shocks such as commodity price changes or natural disasters. Just as there is a spectrum of types of countries, or of regimes, so donors should have an array of instruments which they can use (see figure 12). Recent events in Kenya, Malawi, and Zimbabwecountries whose governments' commitments to poverty reduction are at the very least questionableclearly illustrate the importance of working out how to help poor people in poorly-governed countries. We are pleased that DFID is alive to such issues and is working hard on pragmatic answers to what Clare Short described as "the toughest question in international development".
- In addition to targeting their aid at countries where it is likely to be most effective, all donors attach conditions to aid to encourage the adoption of policies which are thought likely to improve the chances of aid being effective. For critics of conditionality this amounts to the intrusive exercise of power by donors; for others it is a legitimate effort to encourage the adoption of policies which are likely to maximise aid effectiveness and poverty reduction, and a necessary safeguard to ensure that money provided by the tax-paying electorate in donor countries is used effectively. Donor-imposed conditionalities have been criticised for undermining the local ownership of development and poverty reduction strategies, and for weakening local accountability. As Tony German of Development Initiatives told us "a plethora of conditions is ridiculous because it simply does not work and it puts an immense burden on countries in trying to comply with all the conditions." There are doubts too about the extent to which conditionality leads to reform, with DFID's view being that it is unlikely to persuade non-reformers to reform, but that aid with conditions attached can give governments which wish to reform their policy environment more room to manoeuvre, and add momentum to the reform process.
- Despite the criticisms of conditionality, and doubts about its effectiveness in leveraging policy change, donors will no doubt continue to attach conditions to their aid. As Tony German put it in evidence, "Nobody gives aid without having a view about where it should go and how it should be spent. It is a question of striking a balance between ownership and conditionality. You have to have a view and as a donor you have a right, even an obligation to have a view about what the money is spent on". Donors have a right and a responsibility to ensure that aid is used effectively, that recipient countries are committed to poverty reduction, and that appropriate policies are in place. However, we recognise that the external imposition of conditions can undermine local ownership and accountability. Conditionality, in some form, is sure to continue, but it needs to be practised flexibly so that countries are treated on a case-by-case basis, rather than in a one-size-fits-all manner, and so that local ownership and accountability are not undermined.
- The ways in which conditions are arrived at is perhaps as important as what the conditions stipulate: are conditions imposed by donors, or negotiated between donors and a range of stakeholders in the developing country? In the jargon, the path to such negotiated conditionalities is one of "policy dialogue". John Roberts of ODI explained to us what is novel about "policy dialogue". He said:
"Uganda is a perfect example, but to a lesser extent also Tanzania, Mozambique. Uganda particularly in the late eighties was not developing in the free market [mould], it had pegged exchange rates, it had all kinds of discouragements. Uncharacteristically, the donors did not come in with a battle axe. They very gently suggested possibilities for moving to a free market. To begin with nothing happened. By about 1992/93, by a process of gentle persuasion, it had reached the point where, when advice was needed, where Uganda had sometimes turned internally to the ministry of finance, the ministry of finance would then turn to the aid donors. The aid donors fed in a message about the need for a poverty related pattern of public expenditure that led the Ugandan Government to decide what, in the light of its priorities, would be the right way to do that. They came up with their own list and in this way a pattern of public expenditure was arrived at which has now borne fruit and the rate of poverty in Uganda has gone down from over 50 percent to under 30 percent. The important thing was that the donors were willing to play it long, they did not push Uganda too hard, they condoned things like an export tax on coffee in 1994. They showed that they were willing to trust the Ugandans and in the end they were rewarded."
- If the MDG targets are to be met, and more aid is to be provided by donor countries, aid effectiveness will have to be improved. There is much scope for improvement, and substantial international agreement on the measures needed to produce such improvements: good governance and a commitment to poverty reduction on the part of aid recipients; the timely and harmonised delivery of sufficient quality untied aid on the part of aid donors; and the flexible and sensitive use of targeting and conditionality to govern aid allocations and donor-recipient relationships. Most importantly, attaining the MDG targets will require a change in the relationship between donors and recipients, and the construction of development partnerships which allow for sufficient ownership at the local level, including in the production and implementation of Poverty Reduction Strategy Papers (PRSPs). (See also paragraphs 88 and 89). As Clare Short put it:
"we are trying to move it from 'We're the donors, we'll tell you what to do; jump over all these hurdles or you can't get your money', to 'This is a partnership, we have agreed on this Poverty Reduction Strategy together; let's check that it's on track'. And, often, with goodwill, countries go off the track a bit, they might have a drought, commodity prices might fall, and then it is a question of helping them to get it right, not always just saying 'You're off track, we'll cut off the money.' So we are trying to change the culture of how we interpret those conditionalities, and that work is going on and is a constant endeavour".
114 The case for aid for the poorest countries, DFID/HM Treasury, March 2002-copy placed in library and see http://www.hm-treasury.gov.uk/mediastore/otherfiles/case_for_aid02.pdf Back
115 The role and effectiveness of development assistance, World Bank, April 2002, p.xviii. See-http://econ.worldbank.org/files/13080_Development_Effectiveness.pdf Back
116 Ev 111, para 9 Back
117 Q99 Back
118 Monterrey Consensus, paragraph 11. See footnote 12 for web-site. Back
119 Q57 Back
120 Ibid. Back
121 Q16. The equivalent figures for the UK-given in table 19 of the OECD/DAC Development Cooperation Report for 2001-are 2.6 percent and 4.8 percent. Back
122 Peer review of UK aid, OECD/DAC Journal, 2001,Volume 2, Number 4, p. I-30, footnote 9. Back
123 Q17 Back
124 Q62 and Q65. See Ev 61 for ODI table concerning best-practice for donors, recipients and NGOs. Back
125 Q112 Back
126 Ev 3, para 4 Back
127 Q31 Back
128 Ev 3, para 4 Back
129 Q21 Back
130 Q104 Back
131 See, for instance, OECD/DAC Development Cooperation Report, 1999, p.129. Back
132 Assessing aid: What works, what doesn't, and why. World Bank, 1998. Back
133 Ev 67. See also: OECD/DAC Development Cooperation Report, 1999, pp.130-131; Policy implications for aid allocations of recent research on aid effectiveness and selectivity: A summary, Jonathan Beynon, DFID, 2001-copy placed in library and see http://www.dfid.gov.uk/Pubs/files/beynon_paper.pdf; and, Will the new aid agenda help promote poverty reduction? Howard White, Institute of Development Studies-copy placed in library and see http://www.ids.ac.uk/ids/pvty/pdf%20files/aidagenda.pdf Back
134 Q18 of evidence given to the Committee on 18 June 2002 by DFID's Permanent Secretary in relation to our inquiry into DFID's Annual Report. Back
135 Conditionality or contract: Perspectives on partnership for development, published in Journal of International Development, 1998, Volume 10, pp.257-268-reference to p.265. Back
136 Second Report from the International Development Committee, Session 2001-2002, The effectiveness of the reforms of European development assistance, HC417, paragraph 20. Back
137 Q112 Back
138 Q117 Back
139 Q21 Back
140 The case for aid for the poorest countries, DFID/HM Treasury, March 2002-copy placed in library. See footnote 114 for web-site. Back
141 Q40 Back
142 Ibid. Back
143 The assumption made here, and perhaps borne out by subsequent results, is that Uganda needed to develop in a free market direction. Back
144 Q59 Back
145 Q118 Back