Thanks God for Owen Barder. I will have to read pretty much everything
on his web site.

I have not read the http://www.cgdev.org paper but it sort of ties in with my
interest in economics.

Taking a global view we might speculate that the world financial system
has to be re-engineered. Many development economics problems would
disappear.
But how to tackle that?
The stage I am at, in my research:
1) Arno Peters ( a German professor) and his “equivalence principle”
2) Bernard Lonergan (Canada’s national treasure) and his work on “circulation analysis”
3) pretty much don’t waste your time on econometrics and the current output of economics departments

Who could do that – of course engineers.

We build the most complex systems in the world: supercomputers, space
shuttles, power generation grids, bank financial software systems etc. etc.

We have good brains for numbers, we have creativity and a knack for inventions/new designs, we have the tools to model and test models by
simulations or all kinds of numerical analysis.

We have the collaboration tools – look at the success of the Open Software
projects. Developers from all over the globe are able to create high quality artifacts – without ever meeting face-to-face etc. etc.
Engineers Without Borders – Canada/Poland

Taking a group of 13 low-income African countries (chosen based on data availability, in a limited time frame, on a bad internet connection): weighted average per capita GDP is $1,330, while weighted average per capita gov’t revenue is $183. Thus required per capita expenditures would be as follows:

Infrastructure: $153
Health: $27
Education: $37
Social Protection: $8
Watsan: $58
Total: $283

Therefore, ignoring other government expenditures (agriculture, parliamentary expenses, diplomatic corps, defense, etc.) there is already a financing gap of $100 (per capita), or 35.7%. This is in line with what donors are financing in many countries, but remember the above targets ignore many areas of government expenditure – thus I don’t think the above targets are realistic for poor African countries.

Beyond realism, the numbers generated are also quite strange. Using the above targets, the government would be expected to spend more on educatoin than health, more on infrastructure than health and education combined, and twice as much on watsan than on health. Based on my experiences in Africa, this does not seem like a good balance of expenditures.

Finally, the watsan target seems really strange to me. Taking some generally accepted numbers: a borehole costs around $10,000 to install, provides enough water for 250 people, and lasts for about 15 years. Thus the per capita cost of clean water could be something like $2.67 per year. Allowing that this is a VERY rough estimate, $58pp/year still seems like a lot, especially as compared to other expenditures.

This comment was mostly just an excuse for me to play with numbers, but I’ll try to draw a conclusion anyways: established expenditure targest are a dangerous concept. (I’m not specifically taking issue with the above targets – just using them as an example. I’m also not meaning to imply anyone has advocated for targets. Just adding to the discussion).

While I agree in principle that scarce international aid should be allocated strategically based on need and potential for impact, good ideas like that tend to breakdown in practice – especially in the world of international development. I’d hate to see a useful concept (spending money where it’s most needed, rather than where lobbies dictate it should go) turn into a uniformly applied and largely unscrutinized set of expenditure targets (the big, simple type of idea, that dev’t industry tends to favour), instead of a move towards rigorous process of evidenced based decision making (the hard to implement type of concept that dev’t industry tends to shy from). Just another hurdle to jump in the long process of making aid work I guess…

Owen – very interesting analysis. I think that does suggest that Phil is right that these targets are simply not consistent with each other.

I love the idea of being a bright young campaigner- sadly I fear I am neither bright nor young any longer!

I am not for a second seeking to say that misallocated aid costs lives- my only point is that if it wasn’t for the politics there wouldn’t be any aid in the first place. It may well then still be true that it is better to have less aid given better- and many would agree with this. Would the world be a better place without Pepfar, but with US aid to health at previous levels? Maybe. But that doesn’t seem to be your argument unless I am misunderstanding it. Instead you are saying that this aid is being given so badly it is costing lives and should be given much better. I would agree but all I am saying I think this argument is a bit academic as the aid would not be there without the politics.

I do think there is a strong argument for more clever and constructive campaigning that can tug the heartstrings of northern publics whilst delivering the kind of aid that will make a difference. We have tried to do this at oxfam, and since 2005 have been campaigning for more general and sector budget support given over a minimum of three years, often in the face of a lot of criticism from other ngos and AIDS activists in particular. We pushed really hard for MDG contracts from the EC for example, some of the best aid around in terms of supporting public services. This came out a lot from our experience with debt relief, which we feel needs far more credit as the biggest and most successful example of budget support, something I saw for myself whilst living in Malawi.

max
(a rapidly aging policy wonk)

With regards to: “(Tentative steps in this direction are, of course, being fiercely resisted by the trade union of evaluators),” what specifically are these tentative steps? Which organizations? What types of evaluations? Details please.

Comment: In both this post and in your working paper, you pay attention the political economy of aid, but not that of countries. I agree that IGOs are inefficiently allocating money, which implies that its possible for a different allocation mechanism to do better. I don’t quite see why country governments are assumed to be better, however. On average, they have been quite bad. I think its analogous to the rural area country diner versus the Applebee’s. Now Applebee’s isn’t great, it’s not even very good, and the diner might have some romantic mystique, but that diner food is much more inconsistent and much more likely to be terrible in multiple ways: country governments are also much more unpredictable and much more likely to be terrible than the generic IGO.

I don’t recall seeing any mechanism in your white paper for filtering the good from bad local governments, nor do I recall an explanation of why the political economy of a developing country is likely to have better allocation than IGOs.

Owen replies: Your diner analogy doesn’t work. In your example, both diners are in some way accountable to their customers. A more accurate metaphor would be two restaurants, side by side, one of which depended for repeat business on the quality of its food and the other of which was paid by someone very rich thousands of miles away to do whatever it wanted. Which one would be more likely to serve its customers well?

On evaluation, I think 3IE is a good initiative. I also think the Poverty Action Lab does good work. The ALINE approach to gathering data from beneficiaries also looks promising.

To take your example from the UK, 0.7% is far from guaranteed as we all well know, and we do fight on the overall numbers and on 0.7% despite how uncampaignable that is. However, the majority of supporters have to see the human face of what aid can deliver.

Another important political economy point is that whilst aid from the Global Fund for HIV/AIDS, TB and Malaria may have some serious flaws in terms of aid effectiveness, it is still far better in terms of aid quality than the vast majority of bilateral aid given by those who also support the fund. DFID is always keen to point out that the global fund is not as good as their bilateral aid (which is not a given)- but even if this is true then it is still effectively laundering far worse quality bilateral aid from many others (japan, italy, spain, france, the US….)

So when you take political economy into account I think the jury is out as to whether global funds actually have a net negative effect on overall aid quality. Those wanting to tackle HIV only have to deal with two behemoths, GFATM and PEPFAR. imagine the nightmare spaghetti of bilateral funds that would be there without them.

I think blanket dismissal of any new global funds is a mistake too. There are good global funds and there are bad global funds.

In Education if we had a global fund that could mobilise billions from countries in the next few years, and that combines the great governance arrangements of the global fund with the aid effectiveness of the FTI, giving budget support in a predictable way to support education plans, then that is not necessarily a bad thing.

In health i think the challenge is to get the global fund a. using country systems and giving budget support as much as possible, over long time frames, and to gradually turn it into a global fund for health, without losing the amazing gains we have seen since its inception (compare with the World Bank, whose track record on health is by their own estimation deeply ineffective).

I also see what you mean about priorities being set in advance, but I think this is a bit of a red herring. you could reasonably argue that with a few possible exceptions, say for example immediate post conflict countries, almost every government on earth should be spending around a quarter of its budget on education and health.

Max – you seem less concerned than I am about the human consequences of this donor setting of resource priorities. The Open Budget Project calculation is that about half a million children will die in Ethiopia and Nigeria (alone) as a result of misallocation of aid. This is not an argument among bright young things in a campaign planning meeting: this is somebody’s child dying because we thought it was politically attractive to earmark funds for AIDS instead of something else.

I agree with you about the need to be politically savvy about building support for more and better aid – indeed, I’ve recently written a whole Working Paper on the subject – but you can’t use politics as a trump card to to say that anything goes because it is politically expedient. There must be a serious risk that the grotesque behaviour of donors, driven by sectoral advocacy, is gradually bringing the whole aid system into disrepute.

Best wishes from Dar –

Lindsay

The good news though is that Development Aid is not the solution to development – it is at its best a small contribution to a small group of individuals. Development is basically about open and fair trade, and a global respect of human rights.

Owen replies: Par – I agree that there are many determinants of development other than aid, and that trade and human rights are important. But aid can help to accelerate development, and it can also help people to live better lives while that development is taking place. It is too precious to waste.

A good post on an important topic. I’m afraid, however your recommendations are not sufficiently grounded in reality. As you say, aid budgets come from voters and legislative bodies. This constituency is generally more persuaded by human appeals than great evidence. This mix also brings in foreign policy and diplomacy which is another driver for aid budgets. One of the reasons that Ethiopia got such a big HIV allocation was it was a frontline state in the fight against terrorism and was willing to send its soldiers to Somalia. As much as I would like to wish this factor away, we can’t and we better figure out practical ways of dealing with this influence.

As for people advocating for their cause, I am sympathetic here as well. But you can’t simply ask people to stop it. In some cases, it is driven by a sincere belief that their cause is more important than others, in other cases it is just people responding to incentives. Appeals to solve specific diseases that can be given a face and a name are a hell of a lot easier to raise funds for than for evidence best approaches. I don’t think we will get people to sign up for health system sponsorships the way hundreds of charities get donors to sponsor children.

Couldn’t agree more about not creating more global funds. I think I was the only person who blogged on the CDG website that we should wind up UNAIDS instead of inventing a new mission for it. We can thank UNAIDS for the overfunding of HIV around the world.

Heaping scorn on people for advocating for a cause is misguided and will never happen. One idea that I think merits some consideration is a professional certification of charities that establishes rules about the use of evidence and provides guidelines on appropriate fundraising methods. To counteract one set of incentives, you have to create new ones. Major donors would be constrained to fund organizations that only adhere to guidelines on what claims of success can be made from their interventions and how they use data to make those claims. Similarly, those misleading appeals to saving poor africans and sponsorships (which are really a scam) ought to be banned.

In any case, keep raising this issue..it demands more serious consideration and we are being drowned out by the malaria, AIDS, clean water, family planning, guinea worm, oncho, vaccine, polio, etc. etc. champions.

Owen replies: thanks Jeff. I agree with your point about the political economy of reform – indeed, my recent CGD Working Paper is all about how to align the political economy with reform of the aid system.

while it is clear that we tie the hands of local decision-makers to do sensible, rational allocation, is there evidence that if we untied them it would be more rational? or perhaps more importantly, what steps can we take in untying the aid to promote more evidence-based decisions locally?

Orin – Nice to hear from you. It is hard to generalise, of course, given the very diverse governments around the world. But let’s not forget that the option we have is not “rational global allocation” but an uncoordinated set of decisions among tens of donors and literally hundreds of global funds and international organisations with nothing in the system to force these independent decisions towards a good overall outcome. There is no question in my mind that the Ethiopian Health Ministry would do a better job of resource allocation than the aid they experience today.

Local decisions by a government with good local knowledge and – most importantly – accountable to local citizens seems, in general, more likely to make good allocation decisions than unaccountable decisions by international agencies. That cycle of allocation, impact and accountability is essential for any system to operate effectively, and the accountability for global decisions is completely missing. National accountability is far from perfect in many countries, developing and industrialised, but earmarking of aid from outside makes that problem worse, not better.

• Health: 15% government expenditure
• Education: 20% government expenditure
• Infrastructure: 11.5% GDP
• Social protection: 4.4% GDP
• Water and sanitation: $58 per capita
• Agriculture: 10% government expenditure

Owen replies – interesting question, Phil. On the face of it, these percentages don’t look internally inconsistent, though it is strange to think that anyone thinks that there should be international norms of this kind.

I think that one aspect of the structure of the World Bank helps moderate this effect, and this is the role and strong influence of Country Directors. They act in a way like a minister of finance (or treasury secretary) in deciding the Bank’s support program. And often they are based in the country, so they have more ability that agency staff back in HQ to understand and take into account evolving country priorities. And the Bank’s funds (and allocations) have to be agreed by the country, including first and foremost their minister of finance. I think all these structural factors are a big help in containing the more harmful tendencies to “follow fashion” of aid spending. I have seen these folks block time and again efforts by sector and issue advocates to shift far too many resources towards the latest issue that has risen in popularity and attention at the global level.
I’m not sure what the take-away is, but it has something to do with focusing on strengthening this process for allocation of spending at the country level.