Does aid work?
Is aid an effective tool for promoting development?
Aid has its critics. For many decades the most prominent of these were proponents of a let them eat cake approach to international development: William Easterly, Dambisa Moyo, Angus Deaton and, in the case of the Pacific, Helen Hughes.1
Aid, the more nuanced of these critics argued, did little good. The more strident of them went further and claimed it actually harmed the countries it was given to.
Sadly, these critics now have some very influential company: the World’s wealthiest man and the World’s most powerful man. With aid being slashed, all of a sudden the question of whether aid helps or not has become an urgent one.
In my first article in this series I examined whether development was good or not, in my second article I interrogated the merits of economic development. In this article I’m going to ask whether aid helps promote development.
The crudest form of the argument that aid doesn’t help takes the following form:
“We have given trillions of dollars in foreign aid, and poor countries are as poor as ever!”2
There’s a big problem with this argument: it’s not true.
In a planet as populous as ours, absolute amounts of money — trillions! — always seem huge, but set against the scope of global need they’re chickenfeed. In 2007, the year William Easterly’s famous critique of aid was published, the median recipient country received US$55 in aid per person a year. Or to put it another way, just 15c per person a day. We may have given trillions, but 15 cents per person a day is next to nothing. Not nearly enough to eliminate global poverty.
The claim that developing countries still are as poor as ever is also incorrect. Poverty has been falling for decades now, even in the most troubled part of the world, Sub-Saharan Africa. Aid is unlikely to have been the sole source of this fall. However, the claim that aid doesn’t work because we’ve given so much money and everyone is as poor as ever is — empirically speaking — nonsense.3
More sophisticated claims about aid’s ineffectiveness have often been based on the findings of cross country econometric studies. There have been a number of much touted papers that have appeared to show that, on average, aid doesn’t promote development. Some of the papers are rigorous (for example), but they’re not the final word on the matter. Other studies have produced very different findings. Findings that seem to show that, on average, aid increases economic development, improves governance, and helps promote human development. My reading of the relevant literature is that, on balance, it suggests that aid tends to help.
I say “suggests” rather than “shows”, because the relevant literature has many limitations. Untangling cause from effect is very hard in statistical work of this nature. It’s also tricky to isolate the effects of one cause (aid in this instance) from other potential influences. There are ways of trying to overcome these problems, but they’re imperfect. To make matters worse, the data involved are often poor quality too.
And there’s an even bigger problem: most of the studies focus on aid’s average effects.4 Yet, as anyone who’s worked in aid will know, not all aid projects are created equal.
Some aid projects have been spectacularly successful. Aid played a catalytic role in the elimination of small-pox. Aid funded bed-nets drove rapid falls in malaria in Sub-Saharan Africa. Aid funded anti-HIV work has saved millions of lives.5 Aid helped Timor-Leste eliminate Malaria.6 Meanwhile, on the other side of the equation, the Trump administration’s gutting of US aid will likely kill tens of millions of people. (And each of these deaths is — as Tim Hirschel-Burns has very painfully documented — a tragedy in its own right.)
On a more modest scale, Australian aid has helped Solomon Islands run successful elections since the end of the civil conflict there. Similarly, the New Zealand funded reconstruction of Gordons Market in Port Moresby has improved people’s livelihoods and made it a much safer place for the predominantly female vendors who work there. (My examples come from Melanesia not because it’s a textbook case of anything, but because it’s where I work.)
Yet for all it’s successes, aid has also had notorious failures. Aid sent to President Mobutu during the Cold War, for example, was often wasted, sometimes stolen and ultimately helped the dictator cement his hold on power.
More mundane — garden gnome type — aid failures are harder to find: they occur in faraway places, and headline-averse civil servants are never keen that news of money squandered make the return journey home. Such failures are common enough though. Take, for example, aid-funded attempts to provide electricity to the people of Enga in Papua New Guinea. As Radio New Zealand recently reported, the project has collapsed amidst tribal fighting and Engans are still living in the dark. (I’ll return to this project in my next post, because it’s a very teachable example of failure.)
Aid hasn’t transformed the developing world. But what do you expect? It’s been a trickle not a flood. And, at it’s best, despite being a trickle, it has saved many many millions of lives.
Sure it’s also failed in instances, but that’s no reason to cut it. When you do that, there’s a real risk you’ll do harm — unspeakable harm if you’re an evil buffoon in charge of the World’s most powerful country.
The lesson to learn from aid’s failures is not that we should all become more callous and selfish. It’s that we should try much harder to make aid work.
How? That will be the subject of my next post.
To be fair, not all economists have made these arguments. Some economists, most notably Jeffrey Sachs, have been vocal advocates of foreign aid.
Actually, that’s not even the crudest form of argument made against aid. The arguments made by Trump and Musk have been much more dishonest and/or stupid.
The fact that poverty is falling, it should be noted, also shouldn’t be taken as vindication of any particular prescription for promoting development (the triumph of the market, for example). Given scientific and technological advantages poverty ought to be falling, and the success of any ideological prescription should be measured in terms of progress against background rates of decline.
Sometimes the studies will be average affects across a sector (for example governance aid) or in a region (for example Africa) or across a particular period of time (post Cold War, for example). Even so, they’re still averages which mask the diverse impacts of individual aid projects.
Sadly, that should probably read “had”. Listen to this depressing podcast on the negative impacts of the Trump aid cuts.
All of these examples are health related for three reasons: (1) success in combating disease is easy to measure and verify; (2) the elimination of disease tends to be sustainable — i.e. once you’ve gotten rid of it, it is less likely to come back and (3) vaccinations, ARVs and the like are easier to deliver successfully than some of the more ambitious tasks sometimes attempted though aid funding such as changing norms or the functioning of government departments.