28 July 2011

New Aid Instruments for South Sudan

I went to brief the next crop of ODI fellows heading to South Sudan, in London on Tuesday (good luck folks!). ODI are also running what seems to be a really interesting Budget Support Initiative, providing further technical assistance from a more experienced team. One of the issues the team, and the broader donor community is grappling with, is what lessons to learn from recent experience in other countries with direct budget support, and what new funding mechanisms are appropriate for the new South Sudan as the pooled trust funds expire (disclaimer: I know very little of the detail of arrangements in other countries, or the state of arrangements over the last year in South Sudan - however - I'm going to wade in anyway).

I think you would have to be mad to recommend direct budget support for South Sudan. For two simple reasons.

1 - The Government of South Sudan already has more per capita revenues than its more developed neighbours Kenya or Uganda.

2 - Spending control is poor.

If you want to increase aid to South Sudan, I would propose testing some new ideas, like

1 - Direct unconditional cash transfers to citizens. Run a pilot, it would be relatively easy to rigorously test and monitor to see whether markets can respond to new demand in remote areas, or

2 - Cash-on-delivery aid; offer the government a contract promising payment for outcomes, incentivising the government to figure out the best way of delivering effective services to its people.

And finally, are you really sure we should be giving more aid to South Sudan anyway? Remember it has only 8 million people, less than many cities. Remember it is an enormously expensive place to do business, so you get far less bang for your buck than is possible in cheaper environments. Better to build 1 school in Sudan or 10 schools in India?

And I am now regurgitating old arguments, but not increasing aid does not have to mean not increasing international support. There are lots of other good places to start, such as creating trade and investment policies which are more effective for development. 


Anonymous said...

@bokuka: "Better to build 1 school in #Sudan or 10 schools in India?" @rovingbandit http://t.co/f2scqcZ If India, how do u change underlying issues?

Lee said...

I think the underlying issues are tough to change - a lot of it is geography and population density driving costs. South Sudan is landlocked and sparsely populated, meaning it is always going to be an expensive place to operate, even as peace, stability, and new roads will make things somewhat cheaper. Urbanisation is one way of reducing the per capita cost of delivering public services such as education and health - its cheaper when people are close and you don't have to travel so far.

Cynan Houghton said...

Those particular pooled funding mechanisms might be about to expire - count on a few new ones cropping up though (that won't be UN managed...)

Nick Travis said...

Lee - interesting post, although I generally disagree with the argument. The fact that budget execution is so bad is precisely why aid which uses government systems is needed. At present the government has oil revenue, doesn't need aid money as such and doesn't have to show accountability for budget expenditure. To improve the efficiency of budget execution the government needs to build its own systems as well as improving accountability for those funds. It seems to me that getting aid on budget and using government systems will (1) improve those systems and (2) start to have positive spill over effects onto the non-aid funded programmes particularly in service delivery.

Cash transfers have a lot a benefits, although as you say in your comments below the costs are likely to be extremely high so how efficient such a proposal is likely to be is unclear. I don't think cash on delivery aid would work - the government doesn't need money as such from donors so there is no incentive for it to deliver on its promises and would just result in inertia and an unneccessary delay in implementation.

Lee Crawfurd said...

Nick - Thanks for the comment. Do you know if there is much evidence on the impact of budget support on budget accountability? I'm not totally convinced by how much it would help. 

I don't think cash transfers are going to be more a more costly way of delivering aid than providing services, I just think that all operations are expensive in S Sudan. 

I think cash on delivery is worth a try. Not the whole aid budget, but a chunk, as an experiment. If the government doesn't deliver on a set target, they don't get paid, so there is no waste, and if you had a target such as education or health outcomes, the government would still be working towards delivering education and health services regardless, so I'm not sure where you get the inertia and delay.

Nicktravis86 said...


I think DFID have just done a whole bunch of evaluations on GBS for their partners in SSA but I haven't read them yet nor do I know if they talk about its impact on budget discipline. It's a fair point you make, although I think carrying on the way we are for much longer is just going to make things worse. The fiduciary risk is very high I agree, but unless donors take a risk
the systems won't improve (they are currently getting worse) and we'll
be having the same conversation in five years time. Even though the government doesn't "need" aid, services would suffer a lot if it was withdrawn - tying future aid to systems development gives donors some leveradge with which to drive improvements. There are some good examples of this kind of thinking having a positive impact, such as the Health Pooled Fund in Liberia.

I agree that CTs and CODA should be tried as an experiment - there's not much to lose when you consider the current efficiency of aid in South Sudan!

Post a Comment