My article on OpenDemocracy today discusses whether aid works.
Some supporters of aid have made what seem to me to be extravagant claims that aid should aim to bring about economic and social transformation of developing countries, so accelerating economic growth and industrialisation. But this is a very high bar to set. Aid may well help to increase the probability of economic take-off but there are lots of other conditions that need to be in place for the transition to an industrialised market economy to happen, and aid is not a sufficient condition (nor, probably, a necessary condition) for it to occur. Even if aid does play an important contributory role, it would be statistically very hard to demonstrate a link between aid and economic growth.
Although the effect of aid on economic growth is uncertain, there can be no doubt that aid makes a huge difference to people’s lives. Aid provides food, health care, education, clean water, financial services, and modest incomes which transform the lives of the people who receive them. You can see this both in individual families – like the girl I met in northern Amhara, pictured here, who has health care and education because of aid – and in the overall statistics, which show that there has been a vast improvement in the quality of life on almost every measure other than income.
Aid may not always transform societies, but it does enable people to live much better lives while those transformations are taking place. And that represents a huge increase in the sum of human welfare.
I believe aid could and should work much better. Living in a developing country, I see all kinds of waste and inefficiency in the aid system that makes me angry. But it makes me angry because I also see how much difference aid makes when it is used well. I would like to see aid becoming much more transparent and accountable, so that it becomes subject to evolutionary pressures to improve.
This means, by the way, that I do not subscribe to the view that the aid system should be regarded as temporary. In the UK we hope that people will be on unemployment benefit temporarily before they are able to get back to work, but we don’t expect the system as a whole to come to an end. So I think that we should expect that at least for our lifetimes, it will be right and necessary that we transfer income from the richest people in the world to the poorest people in the world. I do not know which countries will be rich, on average, in fifty years time, and which will be poor; but I expect that the world will still need, and I hope it will still have, a permanent system to help those temporarily in need wherever they happen to be.
Aid would work better in future if we accept that we will need a permanent system to provide temporary help to those who need it, and set about designing a better system to do that.
Related reading:
- Phil Vernon at openDemocracy (to which my article was a reply)
- Roger Riddell at openDemocracy
- Ranil at AidThoughts
- Chris Blattman – Could Aid Slow Growth
7 responses to “Aid works even if it does not cause development”
What of those who argue that the issue is not one of whether aid is effective or ineffective, but whether aid is effective, ineffective, or harmful? Your argument seems to overlook this third possibility, which I think is an important one.
Empirical work on the link between autocratic tendencies and aid combined with other work on how dependency on local revenues creates an important de facto mechanism of accountability may not be conclusive, but I do think rather clearly shows that there is a good possibility that aid has serious negative effects on the political systems of recipient countries.
Owen replies:
Thanks for your comment, Nick. I deal with this issue a little in my longer article on openDemocracy, and at greater length in a 2006 Working Paper. The short answer is that I think that it is a theoretical possibility that aid can do more harm than good, but I do not believe that it is true in practice, nor is there any (macro or micro) evidence for it happening in the real world. I do think that some changes are needed to the way that aid is given – for example, by making it much more transparent and linking it more explicitly to results – which would make aid more effective and reduce this risk.
Hi Owen,
I have 2 questions that I would like you feedback on.
a). Am I correct in thinking that Foreign Aid falls into 3 distinct type of categories and each category has its own aim(s):
1)humanitarian relief aid – given to victims of natural disasters such as earthquakes, cyclones and floods
2) military aid
and
3)economic development assistance
b) It is not important that we specify what type of aid(s) we are talking about when discussing its effectiveness/ineffectiveness?
Owen replies –
Yes, you are right that we need to be clear what purpose(s) we have in mind for aid when we assess its effectiveness.
Of course can categorize aid in lots of different ways. Your categories are a good start. But I think a lot of aid may fall into another category: aid that may lead to economic development, but will help people to live better lives in the meantime. For example, how would you categorise paying for children to go to school, or childhood vaccination? They are not technically humanitarian aid, but can we demonstrate that they will have a significant impact on economic development. I would be in favour of paying for children to be vaccinated and to have an education whether or not it makes a difference to a country’s economic development.
One of the disadvantages of the aid effectiveness literature – and the analysis that goes along with it – is that it has a one-dimensional view of what aid is supposed to achieve. And it seems to have become almost politically incorrect to embrace the idea that aid may simply serve as a transfer payment from people who have plenty to people who have little.
Owen
Hi Owen,
Thank you so much for explaining your answer the way you did. It lays bare another layer/viewpoint on the whole aid effectiveness literature that I had not thought about.
Politics, Economics, and Ethics – very interesting.
It sounds like you are attempting to redefine Aid from being temporary development assistance to some sort of global welfare system. This is very clever but I am a little skeptical for a few reasons.
1. Welfare systems are very hard to run even within national settings, why are you optimistic that we, being the very loose and dysfunctional international community that we are (as Copenhagen in showing), can do it well on a global scale?
2. Is this moral? Welfare systems typically take care of the bottom 15%, or perhaps 20%, of people in developed nations. Most of these people have problems that are clinical, temporary, or caused by bad luck (discrimination, etc.) Without being judgemental people in developed nations recognize (to varying degrees) that some (very few) people are simply not capable of making it within the usual economic system and so we help them by means of another system, namely welfare. This is fine. But should we treat most people in the developing world (40 to 70% of the world’s population dependeing on your standard) as if they were similiarly “handicapped”? Dumping them in a global welfare system rather than working towards improving the global economic system seems too easy and potentially dangerous given the dehumanizing effects that welfare systems, especially when applied on a grand scale, can have on people.
3. Finally it seems as if you have somewhat given up on Aid as economic stimulus, or at least would like to change its focus. Given Aid’s track record over the last 40 years this is not entirely unreasonable but I don’t like the idea of giving up, especially when the alternative is such a dismal thing as any global welfare system is bound to be. If developed nations are benevolent enough to keep the developing world on life support through some welfare system, why are they so unwilling to let the people of those nations compete with them by leveling trade relations, etc.?
Owen – Your quote: “The short answer is that I think that it is a theoretical possibility that aid can do more harm than good, but I do not believe that it is true in practice, nor is there any (macro or micro) evidence for it happening in the real world.”
Here’s a bit of recent evidence for industrial-level dutch disease that seems to be relatively sound evidence for a negative relationship of aid and industrialisation:
http://aidwatchers.com/2009/12/the-effects-of-foreign-aid-dutch-disease/
There are also papers showing that aid over 10% of GFP (I think this is the figure) has significantly negative effects on the quality of institutions (through some of the mechanisms Nick describes).
I totally buy your argument on welfare. I don’t buy the argument that more is necessarily better, as opposed to a better use of what we already have.
Owen replies:
Adam – thanks. Yes, I agree that there are both theoretical arguments and some empirical studies which suggest that aid can do harm through either dutch disease or a negative impact on institutions.
But as I argued in a paper on Dutch Disease a few years ago (which you can find here) that isn’t enough to draw the conclusion that aid does net harm overall. For that to be true, the harmful effects would need to be bigger than the benefits of aid. For example, in the case of the risk of dutch disease, the additional dollar of aid (which adds a dollar to GDP, other things being equal) would have to depress net exports by more than a dollar to have an overall negative effect on GDP. That is well-nigh impossible (think of what the elasticities would need to be). The Subramanian and Rajan paper which you link to does not find negative effects on exports anything like big enough to overwhelm the direct benefits of aid on GDP (or it didn’t when I saw them present a draft paper in Berkeley a few years ago).
Remember too that a reduction in net exports represents an increase in welfare, other things – especially imports! – being equal.
So I stand by my claim that I am not aware of any empirical work which demonstrates that the overall effect of aid is harmful, though I do agree that aid can have harmful effects which we should work hard to minimize.
Dear Owen,
I very much liked this article and the way you framed the micro-macro paradox, but I am also interested in the quote that Adam commented on above. I haven’t seen any convincing macro-level studies of the harmful effects of aid, but I think the micro-level definitely needs further testing. It is often hard to examine the effect of “aid versus no aid,” because even in most randomized controlled trials the treatment questions are more subtle or focused. Where I am working in returned communities in northern Uganda, I definitely see negative consequences of aid in terms of creating dependency. At times farmers are unwilling to plant unless seeds are provided for them, and there is an assumption that WFP provisions will continue indefinitely despite repeated assurances to the contrary. This is certainly a new attitude; as a marginalized region, Northern Uganda never received much assistance prior to the creation of IDP camps throughout the area.
I don’t think cutting aid is necessarily the solution, but as assistance shifts from relief to development there is a very strong need to consider sustainability and local ownership. Giving more money and providing more services, while undoubtedly helpful, could make long-term development even more difficult. While I agree with you that some areas will need to be subsidized by aid through at least our lifetimes (particularly health care and education), I think requiring greater citizen contributions (like taxes, which have been completely eliminated for rural residents in Uganda) could help ameliorate the dependency effects of aid.
In the meantime, do you know of any papers that look at aid and dependency at the micro level? It would be fascinating to read more!
[…] development aid really make a difference? Yes, says development specialist and blogger Owen Barder in his excellent article on Open Democracy and on his blog: Although the effect of aid on […]