The UK Department for International Development is to be commended for encouraging some of its staff to maintain a blog to explain to the public what they do.
In Bangladesh, Adam Jackson has posted some interesting reflections on his visit to a health programme (in which DFID supports the government) and a Chars Livelihood Progamme.
Our health review team visited a District hospital where mothers who would never normally have access to safe delivery facilities had very recently given birth thanks to a voucher scheme funded by DFID and a number of other donors. Fifty miles away in the Chars I and the other workshop participants visited a village and met a number of women – some of the most vulnerable people on the planet – who had been given assets of their choice (typically a pair of cows) and had their homes raised on clay plinths above the seasonal flood level, as well as a range of other support to enable them to become self-sufficient. … Both of these programmes contribute to the Millennium Development Goals, and produce results that few people interested in the welfare of the poorest would argue with.
Adam makes the excellent point that both programmes work, albeit to achieve different kinds of objectives. Working through Government may be slower and more uncertain, but in the long run it is an investment in Government systems which, in the end, Bangladesh will need as it becomes more prosperous and no long relies on foreign aid. The Chars programme reaches people more quickly, but does not contribute to building lasting institutions. Clearly, both programmes have an important place, and donors need to be better at understanding that we are working towards multiple objectives and need many different types of instrument.
We need to understand better than we do: (a) how much immediate development benefit do we give up, if any, and how much institutional improvement do we gain, by working through governments? and (b) can providing services through parallel channels such as NGOs actually do harm to the long-run evolution of national institutions, for example by hiring away skilled staff, or by reducing the focus on and accountability of government institutions which should, in the long run, be playing those roles?
Adam’s call for rigorous, transparent evaluation is welcome. I would add that it should be independent and more focused on impact and less on process than current evaluation.
6 responses to “Aid to government, aid to NGOs – both working in different ways”
Excellent points Owen. This is something I’ve been thinking about quite a lot recently, though I had been thinking about it in terms of two separate issues: 1) the problem of working through NGOs which reduces the pressure on governments to provide services, hindering institutional capacity in the long term (as you point out, also by hiring away the limited talent) and 2) Aid needing to put more weight on capacity building in general – e.g. working with governments, hiring and training as many locals as possible, etc. But here you are combining the two in a very thought provoking way.
I think ultimately the answer will vary based on a number of factors. I wonder if the framework should be primarily a function of the current institutional state. For example in a place like Liberia or Rwanda strengthening institutions and building capacity should take priority but in a places like Zimbabwe it should be on delivery services since the governments are so corrupt.
Thoughts? Especially re: what do you think the considerations would be for one model vs. the other.
[…] Barder muses on the pros and cons of giving aid directly v via governments, while over on Global Dashboard, Alex Evans wonders if the next new thing in aid will be the kind […]
Well said.
Though I think this ignores one of the challenges to sustainability that working through governments entails. If governments are funded and guided by international donors, particularly other nations, than the stakeholders aren’t the citizens of the country. The governments aren’t liable to their own citizens.
I get building institutional frameworks just by expansion and use, but what are the challenges to having the country’s own people do this? And aren’t THOSE the challenges we want to meet when it comes to building governments that resolve conflict and provide services?
Last point, I think working only with NGOs doesn’t necessarily handicap the government’s institutions. It makes direct government aid less popular, for sure. NGOs have proven far more successful at engaging communities than governments in general, so why not work directly with NGOs instead of helping governments create schemes, which typically aren’t all that helpful in the long run? Governments can provide legal frameworks and conflict resolution well, but I’m not sure they can do sustainable development work well.
Maybe I don’t understand what “services” exactly you are referring to.
I agreee with Adam that transaprent evaluation is required. The CLP is evaluated by an independent organisation (PAEA) but is funded by DFID. The evaluations are passed to DFID before being made public. It is widely reported that several of the evaluations have critised DFID heavily and scored the program lowly, but that DFID has sent the report back saying it needs higher scores on a flagship program in Bangladesh. The report then comes back with high scores as the company has a long term contract with DFID to evaluate the CLP.
Also many of the advisors to CLP were employed by DFID in the past and have given kep posts in the project to relatives and friends.
The latest I heard about the CLP evaluation was that PAEA was fired and DFID did the evaluation themselves from their country office with their own staff. The project got some of the highest scores given to a DFID project as a result and the project was extended for another 5 years to waste another $110 mil of UK taxpayers money……
And how come Michael Scott has a contract to work with CLP as an advisor with Maxwell Stamp. On their website his ToR says 75 days from 2008 to June 2009. He used to work at DFID as the main advisor on the project and has gone straight into a well paid role with the consulting company. I thought there were rules about this, but maybe they don’t apply to him?