Is overseas development aid working?

These are extraordinary times for those who work in overseas development. We are living through a recession of historic proportions, and yet thus far there are few overt calls for a reduction in overseas development aid. The main questions for the 2010 election will be about public spending: where to make cuts, and how deep should they be? In the past, faced with similar pressures, there would have loud and widely voiced calls for a retrenchment of overseas development assistance (ODA), and a refocusing of government priorities on issues closer to home. But this time round there is a consensus, at least between Labour, the Conservatives and the Liberal Democrats, for continued increases in the aid budget, based on a commitment to meet the target of 0.7% of GDP by 2013. Such consensus in the UK is particularly important, since the UK is widely seen as a leader – in some respects the leader – of the global aid community. We should not be complacent. MPs report an increasing number of their constituents are questioning this commitment, and such questions will become stronger and louder, the longer the recession lasts. But the moral debate over whether or not to provide large amounts of aid to people in countries worse off than ourselves seems to have been won. So the question facing us is no longer about whether to be generous, but about how to do so. As an example of this, the Department for International Development (DFID) produced a new White Paper in July, which was followed very swiftly by a Conservative Policy Green Paper. Both quickly reiterated the commitment to 0.7%, and then went on to discuss ways to make UK aid more effective. Both papers promise bold and significant reforms. The White Paper showcases DFID’s aim of helping to make poor country governments more responsive and accountable to their citizens. The Green Paper sets out many changes which an incoming Conservative government would make within DFID, notably a greater focus on measurable results and transparency. But neither paper makes the case for the broader and deeper review of aid effectiveness which we need. Despite some welcome new policy ideas, both papers contain by implication a great deal of business as usual. The EC – whose aid budget is funded to the tune of over £1 billion by the UK – is also in the process of reviewing its approach to aid, and the UN will re-examine development progress at a major summit next year. Meanwhile, one of the elements on the table at the UN Climate Change Conference in Copenhagen this December is how to pay for the adaptation. Funds are badly needed to support adaptation in response to the effects of climate change, such as rising sea levels, changed rainfall and reduced crop production. Such changes will in many cases have serious social knock-on effects, and the need to get started on adaptation is rightly seen as urgent. Huge sums of money are being suggested – Gordon Brown recently proposed $100 billion per year – and developing countries are not surprisingly using this as a bargaining chip at Copenhagen. They want to wrest control of adaptation funds from the OECD countries which they expect to provide most of the money. In many ways climate change adaptation is akin to development, and the mechanisms and processes for funding and enabling it can legitimately be seen as part of the “development discourse”. Agreement on carbon mitigation measures at Copenhagen is almost certain to require agreement on adaptation funds. So it’s critically important that we take a step back and review how well aid is working. This article attempts to do so, by taking a look at the way aid is delivered, and recommending that we take advantage of present circumstances for a substantial rethink. How well is aid working? Despite billions of dollars being spent on development assistance over the past 60 years, it is doubtful that aid is really working as intended, i.e. providing sufficient impetus to overcome the strong forces that keep people poor. This is a nuanced issue fraught with dilemmas: not surprising, since the definition of “development” is both broad and vague; and since history tells us that sustainable change can only be measured over longer periods than those which the development institutions tend to use. It is easy to produce measures of accomplishment in countries which receive large amounts of aid, such as improved school enrolment, better health outcomes and increased GDP. These can often plausibly be ascribed to aid, and thus given as examples of “effective aid”. For example annual GDP increases averaging around 7%, and a more than doubling of primary education rates in Mozambique since 2000. But many people in both developed and developing countries, and especially those involved in development, are increasingly sceptical about aid. They question the degree to which aid is a primary cause of progress; and point out that aid distorts local institutions and thus acts as a potential obstacle to progress. Large sums of aid money have not secured a better life for Zimbabweans; nor allowed people in Sri Lanka or the DRC to live in peace; nor prevented political violence and mass displacement in Kenya or Georgia. The truth is, we will not be able to measure the results of development assistance for many years yet, because success must be measured in terms of sustained improvements. But that should not prevent us from applying intelligent historical analysis to the issue, and thus assessing the strengths and weaknesses of the ways in which aid is conceived and delivered – i.e. the institutions of aid. If asked to project ourselves forward in time to 2030 and imagine whether, looking back, we would call today’s aid policy and practice successful, many of us would find it hard to say yes. The mainstream consensus on aid has for several years been built around four identifiable core elements. First, the need to strengthen the normative role of the state, and its capacity to provide basic human services and an enabling environment for economic growth. Second, the idea of “pro-poor development”, i.e. making broad social and economic progress while protecting the vulnerable and creating opportunities for poor people to improve their lives. Third, the need to get the macro-economics right for growth. And lastly, that success should be measured in terms of progress towards the Millennium Development Goals. The MDGs represent the internationally recognised and UN-approved aspirations for human improvement, and cover key aspects of progress (poverty and hunger, education, child and maternal health, gender equality, HIV, environmental sustainability, and the international partnerships needed for successful development). They were intended as a global commitment to measurable change by 2015 – though it is universally accepted by now that none of the goals will come close to being met by that date. Meanwhile, there’s also been a fairly strong consensus about how development aid should be delivered. Mostly it is in the form of large, more or less fungible grants and cheap loans to recipient governments. These tend to be linked to so-called “conditionalities”, under which the recipient government agrees to specific changes in its macro-economic or sectoral policies, or in government systems and procedures, as a condition for receiving the funds. These grants and loans have become ever-larger, as the UK and other donor governments have increased their aid budgets while trying to drive down transaction costs. Of increasing importance is the use of imported technical planning tools such as Poverty Reduction Strategy Papers (PRSPs) – through which the World Bank and other donors “help” the recipient government to define “its” local programming priorities, which then become central to the national budgeting process. The relationship between donors and recipient governments has been defined in recent years in terms of “partnerships”, but in reality these partnerships are very unequal. It is not difficult to see what is wrong with this picture. In many poor countries, the “normative role of the state” is practically non-existent, and the state is in fact primarily a vehicle for the enrichment and maintenance of powerful elites, whose interest is in protecting their own privileges, rather than supplying security, welfare and an enabling environment for social and economic progress to fellow citizens. This is at odds with the focus on “pro-poor”, so in this respect donors are pulling against, rather then with the grain of the local political economy. The Millennium Development Goals are a worthy expression of what a better world might look like, but in and of themselves they are far too simplistic to provide a genuinely useful way to measure progress on the intended timescale (i.e. by 2015); and they act as a perverse incentive, since they focus on rapid changes in phenomena, rather than deep-rooted changes at the underlying level. The emphasis on macro-economic growth is not necessarily a problem, provided those charged with implementation understand that short-term growth is often at the expense of other, potentially more important factors such as improvements in stability, inclusion, and the rule of law; too often however, they do not. Looking at aid mechanisms, we can see that they are mainly blunt instruments, whereas sharper, more precise actions with leverage potential are more likely to succeed in the huge task of societal transformation needed for development. As such they entail significant risks of preventing, rather than promoting the kinds of changes which are needed if poverty is genuinely to be overcome. PRSPs and other conditionalities are – despite the rhetoric of ownership and partnership – northern tools for southern problems. Ever larger subsidised loans, budget support grants and sector-wide approaches feed the actual, rather than the desired governance system, i.e. become a source of patronage and personal enrichment, and hugely strengthen the power of incumbency. They imply a level of trust that citizens of the state concerned may not yet have in their government, even if it is nominally democratic. It is also impossible to prevent the misuse of funds transferred under such instruments. The “partnerships” between donors and recipient countries tend to be conversations taking place in two mutually unintelligible tongues, in which the overwhelmingly technical goals of the donors are readily agreed to by host country representatives whose interest is overwhelmingly political. They are often not real partnerships at all, but rather a tacit agreement to work around one another’s actual (often conflicting) priorities. Put most simply, each donor’s priorities are represented by the MDGs (as modified by its own particular political priorities in the country/region concerned), while the host government partner is focused more on feeding aid into the clientelist-patronage political system to stay in power. So, despite the injection of funds, services and well-meaning advice, the net effect of aid is likely to be a strengthening of the status quo and thus, of the forces which keep poor people poor. But if this analysis is so readily made, it stands to reason that it must be available to most people involved in the aid institutions. Why, then, have policy and practice not been radically changed? So why don’t those in the aid sector address the problem? Aid tends to be “tweaked”, i.e. changed in appearance from time to time, without addressing the fundamental problems raised above, because of a set of mutually reinforcing factors which taken together prevent a rigorous review and revision from taking place. These include a lack of clarity and rigour; the unfitness-for-purpose of the aid institutions; a mutually reinforcing web of vested interests; and the genuine difficultly of getting agreement among diverse stakeholders with different interests. One problem is that, despite reams of new policies over the years, the aims and objectives of aid remain imprecise, so the sector lacks both clarity and rigour. Aid as a sector serves many different purposes: poverty eradication, poverty reduction, “development”, attaining the MDGs, improving governance, reducing conflict, reducing immigration to OECD countries, improving national security in the West, providing markets for donor-country goods and services, to name but a few. While there is overlap between them, they are not all the same thing. In any case, despite years of debate and discussion, and despite the apparent clarity provided by the MDGs, the underlying theories of change as subscribed to by different stakeholders are usually vague, and often at odds. So there is a problem of a single “sector” comprising a variety of different aims, leading to a laziness of language and preventing a rigorous approach to impact evaluation. Despite what one might expect of a sector worth billions of pounds per year, there is no clear and proven theory of change supporting aid policies and practices. Despite employing excellent staff, aid institutions taken as a whole do not apply an intellectual framework which is clear and coherent enough to meet the challenges implied by their mission. The implicit endeavour of international aid institutions is to support and promote societal transformation in other people’s sovereign countries, and in so doing to promote changes some of which are bound to go against the interests of those currently in power, both locally and nationally. But there is no successful historical example from which to learn how best to do this, excepting cases in which people have been conquered or subdued in war. This conceptual and empirical vacuum has to some extent been filled by development economists, whose language and culture permits a degree of technical prescription and certainty, while the more nuanced voices of political scientists, historians and anthropologists have been harder to hear. We need to be much clearer about the theories and purpose of aid. Given the vagueness of the purpose, it is difficult to argue conclusively that aid institutions are either fit or unfit for purpose. But if the purpose is – as I would argue – to help transform the contexts and conditions which define poor people’s opportunities and choices, so that they stop being poor, then it follows that aid institutions should be designed for and evaluated against an ability to promote transformation. This means, for example, that they need to be set up in a way that supports locally tailored approaches and promotes the fundamentally societal and political changes that are needed. They must take account of and address these issues both in the rich and poor worlds. They must reach beyond merely “delivering aid”, e.g. to include trade and other aspects of international relations. Currently they remain far too focused on delivering technical changes, and are not set up to deal with the politics of aid. We need to review and fundamentally rethink the institutions of aid, and make them more fit for purpose, so they can accomplish the kinds of tasks most critical to the purpose being met; rather than – as currently – accomplish the tasks which happen to be most feasible. Taken as a whole, the institutions of the aid sector are in some ways redolent of the old “military-industrial complex” of the Cold War, with vested interests resisting changes in the analysis which might lead to their exclusion from opportunities. From the donor side this includes the overseas development ministries, the vast UN system, the World Bank and other intergovernmental bureaucracies, some political leaders (and pop stars), some academics, NGOs and the businesses that benefit from ODA funds. In developing countries this includes politicians and bureaucrats and their clients, as well as NGOs and all others benefiting directly or indirectly from the current system. This web of vested interests creates a massive inertia, preventing those within it from seeing things clearly and devising more appropriate approaches. Incentives within the sector are often poorly aligned with the purpose of aid, and are often perverse. This web of vested interests needs to be unpicked, and some of the institutions within it persuaded to accept the need for a fundamental review of the purpose itself, and of their fitness for the purpose. Some of the “smaller” actors for change – small NGOs, innovative individuals, those outside the development mainstream, thus less significantly “vested” in the status quo – can provide some of the leadership ideas. Meanwhile the political compromises needed to come up with a development aid discourse that is acceptable to all the stakeholders are many. Development is not the only foreign policy aim of donor countries, so compromises have to be achieved among competing ministries, and in democracies, between government and the political opposition. Compromises are also necessary to reconcile the diverse interests and approaches of different countries within the “donor community”, and among the member states of the UN. Nor do all recipient countries necessarily agree on how aid should be delivered, necessitating further compromises when they come together to sign international agreements and declarations concerning aid. All these compromises are further complicated by the shadows cast by history and geopolitics, because the donor countries were and are responsible by their actions for some of the structural causes which keep people in underdeveloped countries poor. The result of all this compromise is that negotiations and agreements tend to focus on the “technical” aspects of aid, on which agreement is most likely to be reached, and shy away from the politics of what is, nevertheless, a highly political sector. The 2005 Paris Declaration and the 2008 Accra Declaration – sets of very broad and overly technical “principles” about aid, agreed on between the OECD and developing country governments – are two examples of this. On the face of it they address important aspects of aid – but they fail to deal with the kinds of issues raised in this article. There is a need for far more plain-speaking and honesty about aid than is currently the case. This should be underpinned by a strong political will and insistence on rigour, while not abandoning the important value-driven impetus in the OECD countries to assist people who are poorer than them. As a result of these factors, development aid as currently implemented, while no doubt contributing to remarkable short-term successes in specific locations such as improved primary school enrolment, rapid economic growth and reduced child and maternal mortality, too often focuses on a one-dimensional, economics-led theory of change. It also tends to reinforce and entrench the culture and systems of power (globally and locally) that keep poor people poor, potentially storing up significant political resentment that ends up being repressed and/or being expressed through instability and violence. And it shies away from promoting societal changes with the potential to bring about real progress and development. So what is needed? What all this means is that a complete rethink of aid is needed: a re-examination of its fundamental premises, incentives and operating models, to align them better with the deeply political nature of the task. This does not mean throwing out the baby with the bathwater, but it must nevertheless be done with maximum honesty and rigour, so as to avoid the kind of “tweaking” that has so often characterised changes in aid policies and practices hitherto. The various policy reviews referred to at the start of this article are welcome. But history tells us that, for the reasons given above, they are likely in practice to lead to tweaking rather than significant reform. So what are the kinds of improvements we ought to be arguing for? There are no easy answers to this – which is why a broader, deeper and more honest debate is so badly needed. Whoever wins the next general election will need to muster a great deal of political energy – domestically and across the world – to make significant changes to the UK’s development policies. The political party conferences this autumn, where the next crop of MPs will mingle with those staying on, are an important opportunity to engage in this discussion. Here are a few suggestions. Clarity of purpose is critical. History tells us that poverty reduction comes about not as a result of people being “lifted out of poverty”, which is a curious and misleading phrase, but because the context and conditions in which they live are transformed – by changes within and outside their own sphere of control – thus widening their opportunities and choices. Thus an appropriate and simple goal for development aid would be to help transform the contexts and conditions which define poor people’s opportunities and choices, so they stop being poor. But whatever the purpose chosen, it needs to be clear and simple, so that fitness for purpose can be assessed and improved, progress can be evaluated, and appropriate policies and strategies can be devised. Each development context is different, so once the overall purpose is clear, aid agencies must craft country by country strategies to establish in explicit terms how they plan to influence and help transform the context and conditions which define poor people’s opportunities and choices, so they stop being poor. This is not the same as “poverty reduction” as normally understood. Aid mechanisms must comply as far as possible with the principle of subsidiarity. What this means is that strategy, decisions, implementation and monitoring take place at the lowest appropriate level. This simple principle maximises the possibility of coherence, accountability, relevance, appropriateness, feasibility, flexibility and participation – all of which are critical to aid effectiveness. The current drive towards ever-larger aid institutions, linked by increasingly rigid agreements about coordination, makes it hard to comply with this principle. The posture of aid agencies in developing countries must be political, not technocratic. This means, for example that donor’s local offices need a political, rather than a functionary style of leadership. The leaders of country aid programmes must be expected to engage politically, play an accompaniment role, negotiate cannily, and hold the recipient government to account. The current title of the senior DFID person in-country is Head of Office, which surely sends the wrong message entirely. Donors also need to engage with other like-minded internationals in political coalitions; this is not the same as contributing to bureaucratic coalitions such as multi-donor trust funds. Meanwhile aid agency staff should create relationships within the local society allowing them to understand the context better, and influence it by their interactions; they should see their role explicitly as promoting the interests and voices of marginalised constituencies within the country, and thus taking care to understand their interests and views. Finally, they need to get out of the capital cities more, working and supporting others’ work at the local level, recognising that peace and state are built there, as much as in the capital. Aid agencies must find ways to sharpen the way they use financial aid, and make it more explicitly part of the political aid contract. They can do this through a more effective use of conditionality, by negotiating context-specific conditions about improvements in governance, citizenship, peace and development, and taking them at least as seriously as the macro-economic and technical conditions most often used today. Having reached agreement on the right conditions, they should require recipient governments to show significant progress (rather than just intent) on politically difficult issues, in order to unlock funds. And they should then follow this up rigorously by implementing real sanctions in the event of non-fulfilment (e.g. cancelling rather than suspending funds). To ensure objectivity and fairness, they need to establish new systems, appropriate to the context, for reviewing progress against funding conditions, e.g. setting up panels including local MPs, donors, civil society and external evaluators. An active civil society is critical to a responsive state and an inclusive political settlement. Aid agencies must therefore promote and energise civil society as part of their country strategies. In contexts where the state is hostile to civil society, this can be done by supporting large numbers of NGO service delivery projects in sectors which are not seen as inherently “political”, e.g. health care and agriculture. By doing this, agencies can promote and foster civil society interactions from which new ideas about policy and politics, and change leaders will emerge. Changes in policy and approach need to be backed up by changes in the organisational culture and systems of the aid agencies. For example aid agencies must give equal prominence to societal and political outcomes, as to economic growth. They urgently need to review and revise incentives, so that staff are rewarded for progress towards long-term achievements, and not for perverse incentives such as funds disbursement. This means focussing accountability on achieving specific outcomes in specific contexts, rather than overarching outcomes like the MDGs. It also means basing the staffing and structure of country offices on the country strategy, rather than vice versa; and posting more staff and more senior staff to complex assignments, and keeping them there for longer. And finally it means creating an institutional environment more conducive to questioning, and a less defensive dialogue with parliaments, civil society, the media, etc. about the limits of ODA’s impact. These are simply suggestions, and encouragingly some of them are reflected in DFID’s White Paper and the Conservatives’ Green Paper. But past experience tell us that changing only some aspects of development policy and practice is not enough. What is needed right now is a sustained and – crucially – a more honest dialogue about how to make our overseas aid more effective. It’s what poor people in developing countries deserve.