African countries must urgently start the process of ending aid dependency

On why African countries must seize this moment to start the journey towards self-sufficiency

Thank you for being a regular reader of An Africanist Perspective. If you haven’t done so yet, please hit subscribe to receive timely updates along with over 21,000 other subscribers.

I: We must not waste the current crisis

The projected effects of aid cuts by the United States government are sobering. According to one estimate, USAID cuts could end up costing 500,000 lives in South Africa over the next decade. People who’ve come to depend on life-saving U.S. assistance are facing the unimaginable. Beyond the dire consequences for aid recipients, the destruction of professional careers and livelihoods both in aid recipient countries and in Washington, DC will prove costly. The lost human capital will be hard to replace. The effects of job losses and program halts will reverberate across the economies of low-income countries in the coming months and years.

It’s worth reiterating that the trend in aid cuts extends beyond the United States. Furthermore, over the last four years donor countries shifted to spending aid money on refugees within their territories and repurposing aid for climate-related projects — moves that effectively reduce the amount of aid money flowing to low-income countries. These trends, and the increasing use of foreign aid to pursue narrow foreign policy objectives, look set to continue.

In the last few years donor countries have shifted significant allocations of their aid to domestic refugee settlement and other costs. In aid-recipient countries, health and humanitarian assistant dominate as the expenditure league tables. Source: ONE

This post makes two simple points. First, low-income countries can and therefore must aspire to wean themselves off of aid dependency in critical service sectors. Foreign development assistance is obviously welcome. But it must always serve national development goals and never substitute for the state. Second, the quest for independence from aid comes with a real risk of deterioration of aid-supported service delivery systems. These systems have been critical for achieving most of the gains in human welfare witnessed in low-income countries over the last three decades. It would be a shame if we wound back the clock on immunization rates, infant and maternal mortality rates, school enrollment rates, etc. Consider the example of Ghana, a country that consciously set about ending aid dependency in 2017:

The impact of aid (and grants) on the budget is much more significant than what the [nominal figures] may suggest. For the overall budget, if we exclude compensation payments (i.e. wages, salaries, allowances, and pensions), interest payments, and statutory transfers to consider expenditure on just goods and services and on capital, then the percentage of aid averaged 38.3 percent over 2016 to 2018, while grants alone averaged 21.2 percent. This feature is even more pronounced in the expenditure of some of the Ministries and Services. For example, from 2016 to 2018, for the Ministry of Food and Agriculture, and for the Ministry of Water and Sanitation, aid as a percentage of Government of Ghana financed expenditures on goods and services and on capital averaged 102.8 and 312.1 percent respectively. The percentages were 22.7 and 46.9 respectively for the Ministry Education and the Ministry of Health; and for the Ministry of Gender, Children, and Social Protection, and the Ministry of Local Government, they were 54.3 percent and 104.1 percent.

This example shows the importance of paying close attention to systems, even in countries with nominally low levels of aid dependency. Donors’ influence in “shaping priorities and determining what is actually done” can be much higher than what the headline numbers suggest. To put it mildly, filling that gap won’t be easy. In other words, even countries that have the fiscal space to absorb aid cuts will struggle to avoid a worsening of outcomes in the short to medium term. Amidst the ongoing turbulence in the aid industry, I wish more people in LIC governments and the development community spent more time and resources getting ahead of this daunting challenge. We’ve come too far. We must not let important developmental gains slip away.

In light of these facts, one hopes that the current crisis will focus minds on the urgent need to start the journey towards ending aid dependency once and for all.

Political incentives and institutional inertia are such that it was always going to require a significant shock for governments in aid-dependent countries, their development partners, and the global development profession to fully appreciate the wages of aid dependency. We are currently experiencing such a shock, which ought to force a paradigm shift. Now is the time when everyone must do their part in pushing governments in low-income countries to start weaning themselves off of aid dependency — both in terms of building resilient public sector service delivery systems in health, education, water & sanitation, etc; and enabling growth to create the fiscal space to pay for essential public goods and services.

Interestingly, my very informal survey of online and offline commentary suggests that a non-trivial share of development experts don’t believe that ending aid dependency is achievable. As I hinted to in an earlier post, the global conversations about aid cuts stand to benefit from appreciating views from low-income countries on the need to end aid dependency. It’s true that these views often ignore or understate the positive impacts of aid, while overstating the negatives. However, it’s hard to argue against the general premise that aid dependency is bad. Period. No society should have to go through the humiliating tragedies we are already witnessing at scale based on electoral outcomes in far off lands.

As always, the aim here is to think through these difficult questions from the perspective of low-income countries.

II: Not so dead aid

Most international development experts correctly don’t subscribe to the idea that foreign aid has zero benefits. There are lots of examples of how aid has helped improve human welfare at scale around the globe. Here, public health is arguably the gold standard. Diseases that killed hundreds of thousands of kids a year have been eliminated or contained. Tens of millions have been kept alive and granted a ticket to a normal life expectancy with life-saving interventions like PEPFAR. In short, aid saves lives.

Expected (based on income) vs realized levels of schooling (left) and child mortality (right). A good deal of the “excess” improvements not attributable to growth can be tied to foreign aid and/or the emergence of global norms in the quest to improve human welfare. Source: Lant Pritchett.

You can clearly see all this in data. Maternal and infant mortality rates are trending downwards. It’s hard to find any large clusters of human settlements anywhere around the globe that is yet to go through the mortality transition. Low-income countries are reaching levels of life expectancy well above what would be predicted by their income levels — in part because of health systems built and financed through aid.

Life expectancy has increased in all countries around the world regardless of income level. Source: Our World in Data

It’s hard to isolate the direct impacts of aid on development, but no impartial observer who knows the trends in health outcomes can deny that aid played an important role in these achievements.

However, the critics of aid also have a point. Aid dependency is bad. The loss of policy autonomy by aid-dependent governments is bad. The erosion of government legitimacy and accountability mechanisms that happens under conditions of severe aid dependency is bad. The fact that too often aid projects are poorly designed by people who pay little attention to contexts and don’t care to consult would-be beneficiaries is bad. The distortion of labor markets and policy focus that due to aid dependency is bad. The humiliating loss of dignity that comes with being perennially dependent on charity from strangers is bad.

Dambisa Moyo’s Dead Aid and William Easterly’s Tyranny of Experts are recommended readings for calibrating our understanding of the limits of aid.

And above all, aid creates the wrong incentives for ruling elites in aid-dependent countries. Many simply farm out whole public sectors to non-governmental organizations. The substitution of the state with NGOs stunts the development of state capacity, thereby keeping many countries forever reliant on non-governmental systems for the delivery of essential public goods and services. This reality partially contributes to the atrophy of elite ambition in these countries. Lacking the capacity to get anything done, these elites have largely been reduced to parroting common-denominator global or regional compacts as their development goals (and fluffing even those). For too long too many people have become content with all manner of aid-financed palliative fixes to gaps in the public sector.

No one should be surprised that self-respecting thinking people aren’t satisfied with this state of affairs.

In the recent past discussions about aid dependency and its negative structural impacts on low-income countries have largely receded into the background, with many development professionals touting new frontiers in evidence-based impacts and cost-effectiveness (from donors’ perspective) as reasons to stay optimistic about the current (official) aid paradigm. However, this isn’t a solution to aid dependency. Regardless of impact and/or cost effectiveness, aid dependency is still aid dependency. Furthermore, one could argue that the current arrangement reinforces low-income countries’ general lack of policy autonomy and dependence on foreign expertise. How I wish we could confidently claim that formerly aid-dependent governments are now able to embed research into their policy design and cost-effective implementation processes in line with their own developmental priorities. How I wish this was the core mission of those of us that champion evidence-based policymaking and cost-effectiveness.

Pointing out the negative effects of aid dependency shouldn’t be viewed as a dismissal of all the positive impacts of aid, or the sincerity of those who champion altruistic global solidarity through aid. The simple hard truth is that good intentions alone are not enough. To be blunt, no one should expect to be uncritically applauded when the net effect of their good intentions is a hollowing out of public sector capacity to provide essential public services; the erosion of government legitimacy; or costly exposures to erratic policy shifts elsewhere like we are currently witnessing. No amount of “do no harm” absolution rituals can make up for this.

It follows that we must urgently shift to forms of development cooperation that actually enable low-income countries to stand on their own feet. We must not pretend that we can cleverly nanodevelopment our way around glaring systemic failures at national levels.

Is now an inopportune moment to have these difficult conversations? Absolutely not. Some have argued that discussions of aid’s failings will be weaponized by those out to end all aid. Yet the same people haven’t shied away from proudly proclaiming the soft and hard power benefits of aid to donor countries. While it’s entirely within reason for pro-aid citizens of donor to fight their domestic budgetary/distributive politics the best way they can, few have paused to consider that thinking people in aid-recipient countries don’t like having their economic, political, and cultural affairs unduly directed from afar. Which is why I believe that for the sake of everyone’s political education, now is the perfect time to discuss all that is good, bad, and ugly about aid. There should be no room for moving on under the old aid paradigm.

III: The way forward

Aid-dependent countries’ quest to wean themselves off of aid won’t be easy. This is for two reasons. First, lots of these countries simply don’t have the fiscal capacity to make up for the assistance from donors. Their economies are too small and their don’t have access to credit. Second, regardless of income levels, these countries have come to rely on donor-directed systems for service delivery (see the example of Ghana above). In the best case scenario, initial teething problems will precipitate short-term deterioration in service quality and outcomes. In the worst case scenario, some countries will experience sustained system-wide deterioration or collapse of critical functions like vaccinations, disease surveillance, collection of vital statistics, pre and postnatal care, etc.

Correlation between aid received per capita and GDP per capita. Both axes are on a logarithmic scale. Source: Our World in Data

Relatedly, historical legacies of spatial developmental inequalities mean that even moderately successful countries might experience concentrations of systems failures (think the northern two fifths of Kenya, for example). This is why data on aid per capita may offer a clearer picture than aid as a proportion of budgets or national income. Like in the Ghana example, aid’s most important impact has been in making systems work (however imperfectly). And here, as shown in the figure below, most countries on the Continent remain exposed despite recent declines in aid as a share of budgets or GDP.

Variation in aid dependence across the globe. Source: Our World in Data

So what should be the way forward? I hate to sound like a broken record to regular readers, but the first and most important step towards ending aid dependency must be laser-focused attention to growth. There’s simply no other way to create the fiscal space to absorb aid cuts in the long run. Rather than imagine that “good governance” and “accountability” can magically squeeze value out of African states’ paltry budgets and tiny economies, the focus should be on growth (and accompanying pro-growth improvements in tax administration). All else equal, bigger economies will mean more money available to spend on essential public goods and services and less reliance on aid for the same (higher-income countries may “graduate” into receiving expensive security assistance).

Fulfillment of basic human needs vs GDP per capita. Basically, growth is good for development. Source: Lant Pritchett.

Working to fully absorb systems and improve on aid-dependent systems will be hard. Yet to understand why this must be done anyway it’s worth looking back at how the Ghanaian government justified its quest to end aid dependency back in 2018:

…. there are certain basic services, such as basic education, health, sanitation and water resources, that after 62 years as an independent nation, we need to cater for ourselves without stretching out our hands to donors. This is important for our self-respect as a nation. Further, dependence on donors for such basics has tended to have the unfortunate effects of absorbing an inordinate amount of the time of our officials and distorting our national socio-economic priorities.

Spot on. Ending aid dependency is about both dignity and policy autonomy. Of course, seif-sufficiency must also ultimately be about achieving concrete observable results (more on this below).

Former Liberian Minister for Public Works, Gyude Moore, recently expressed similar sentiments:

Low capacity further down the [bureaucratic] hierarchy limits the extent of delegation. As head of agency, your day is inexorably cluttered with “the mundane”. You are so dependent on “donor partners” who pout if the minister himself doesn’t show up, another huge section of your time is in meetings and events to assuage the fragile egos of your partners. The smaller and poorer your country, the worse this is. The daily whack-a-mole, leaves very little time for strategic long-term thinking as local politics and external shocks conspire to suck the life out of well-intentioned plans.

Consequently, dependence on external assistance persists even when we clearly want independence in our policy making.

Moore’s words are a reminder the enormity of the problem at hand.

On paper, if the money is available it should be possible to absorb aid-dependent systems into regular mainstream ministerial operations without too much difficulty. After all, the vast majority of pivotal aid workers in these systems are usually country nationals. However, in most countries the biggest challenge to this process would be the divergence in logics of operations between ministerial bureaucratic systems and the aid sector. The former tend to disproportionately focus on inputs, political business cycles, procurement, rent seeking, and redistributive politics; while the latter hews relatively closer to global development compacts, relatively expensive delivery mechanisms, onerous reporting requirements, and attention to measurable results. Of course there’s a fair amount of overlap across these attributes (aid can be captured for redistributive purposes, examples abound of notoriously corrupt aid projects, tied aid is ultimately mostly about procurement etc).

It follows that a merger of the two approaches to service delivery would require honest assessments of sectoral distributive politics. Which isn’t necessarily a bad thing. It’s OK if service delivery is used as a currency in politics, as long as professional standards are maintained and bureaucrats retain sufficient operational autonomy. The goal should be to quickly get to doing the best results-oriented work possible under prevailing political and institutional realities, rather than engaging in delusional attempts to change the context before attempting to merge approaches. Simply put, governance gaps in low-income countries shouldn’t be cited as an excuse to perpetuate aid dependency.

IV: Ending aid dependency will require more than empty slogans

To conclude, it’s worth going back to the example of Ghana. In the end, the country failed to achieve its stated goal of substantially reducing aid dependency under President Nana Akufo-Addo. Ghana’s experience provides an important lesson. Because of the unavoidable decolonial and nationalist undertones that come with the quest for independence from aid, there will be a strong temptation to engage in empty sloganeering (similar to what we’ve seen with other endeavors at decolonization). As far as slogans go, “Ghana Beyond Aid” certainly had a nice ring to it. A number of Akufo-Addo’s hot takes even went viral. But in the end all the sloganeering meant little as his administration presided over policy failures that culminated in a humiliating sovereign default.

The point here is that tangible results must be central to the serious work of ending aid dependency. Empty slogans won’t cut it. Getting measurable results ought to be the only thing that matters. The hard-earned developmental gains of the last 40 years must be protected at all cost.

Finally, no one should be naive about potential political obstacles to ending aid dependency. As noted above, incentives and institutional inertia are on the side of the status quo. Donors want to maintain their influence. Ruling elites in low-income countries want to keep shirking their responsibilities. Lots of aid jobs would be on the line. Lots of voters would have a say about deterioration in the quality of services and outcomes. No one is going to just do the right thing.

Yet those of us who want to see more impactful modes of development assistance must insist on a paradigm shift. The future must be different, and better.