When/how does aid help Africa’s public services work better?

I seem to be spending most of my life at the ODI at the moment, largely because it is producing an apparently endless stream of really useful researchODI factors papers and seminars. Yesterday saw a combo of the two, as it launched Unblocking Results: using aid to address governance constraints in public service delivery (OK, maybe it still has a thing or two to learn about snappy titles…..).

The starting point for the work is that while there is a vast amount of research on the role of institutions in delivering (or failing to deliver) health, education, water etc, there is very little on the role of aid agencies when things go well. So ODI carried out a positive deviance exercise, identifying 4 success stories out of 60 initial candidates, and then delving into the reasons behind the success.

They were a rural water programme in Tanzania, a pay and attendance monitoring programme in Sierra Leone, support for the government Strategy and Policy Unit, also in Sierra Leone, and a local government programme in Uganda.

According to the report, ‘Six factors seem critical in this regard and provide clear implications for the design and implementation of aid packages that seek to address service delivery blockages. Apart from the first (windows of opportunity), they are all within the control of external partners to pursue. However, in most cases, they would also require considerable deviation from current practice.’

The six, complete with natty graphic (right) are:

Identifying and seizing windows of opportunity: plenty of overlap with my own work on shocks as drivers of change. The authors described what I now call WoOs (sorry) as the most significant common factor behind the success stories. But it’s not just a question of waiting around for a shock – prior relationships, trust and knowledge are crucial to being able to seize WoOs, as is a ready source of at least limited funding.

Focusing on reforms with tangible political pay-offs: governments listen to aid agencies when they help deliver ‘tangible goods and services that politicians could capitalise on in their campaigns.’ i.e. unless you align political self-interest with governance reform, you can forget it.

Building on what exists to implement legal mandates: concentrate on the implementation gaps that already exist, rather than rewriting current rules and laws, even if that means accepting the reviled ‘second best solutions.’

Moving beyond reliance on policy dialogue: Building on the previous point, ‘[successful] aid packages seem to focus on ‘getting things working’ rather than perfecting the framework (through the development of laws, procedures, regulations, policy processes).’ Nuts and bolts, not endless seminars.

Facilitating problem solving and local collective action solutions: yep, it’s our old friend, convening and brokering. One speaker worried about a ‘scramble to convene and broker’, which could make the per diem culture of East Africa and elsewhere look like a garden party.

Adaptation by learning: ‘Aid packages benefit from in-built flexibility that allows for regular programme adjustment based on learning and changes in the local context.’

If this sounds familiar, it’s because it is – it echoes work by Matt Andrews, the Africa Power and Politics Programme, and a lot of the stuff on ‘how change happens’ on this blog.  Sue Unsworth reckons we are reaching some kind of ‘critical mass’ of research findings. Rebecca Simson of ODI helpfully summarized the so what’s for donors in this table (it’s not in the report, but Matt Andrews wisely advised them that unless they can offer a table of so whats, the donors won’t listen).

ODI Summary

But does that mean donors are going to start adopting the lessons of such success? The obstacles in terms of ideas, institutions and incentives, are high. One useful suggestion from the seminar was to apply the same positive deviance approach to the donors – where do donors depart from standard practice and pursue enlightened approaches to governance reform, and why? As Ros Eyben has shown, an awful lot of these success stories seem to boil down to stubborn individuals in the aid sector, willing to ride the two horses of satisfying the demands of logframe/results-based aid, while operating in the real world of messy, make-it-up-as-you-go-along innovation, (however painful the result). Oh good, needs more research……

Subscribe to our Newsletter

You can unsubscribe at any time by clicking the link in the footer of our emails. For information about our privacy practices, please see our .

We use MailChimp as our marketing platform. By subscribing, you acknowledge that your information will be transferred to MailChimp for processing. Learn more about MailChimp's privacy practices here.


7 Responses to “When/how does aid help Africa’s public services work better?”
  1. Interesting to compare this post that sounds rather optimistic with the recent post by Owen Barder, where he gives a quick overview on the findings of the book ‘Why Nations Fail’ (which you also reviewed here). Basically, what comes out of Owen’s blog is that “If a country’s elite wants to improve public financial management, they can easily and cheaply obtain the expertise they need without any help from donors; and if they don’t want sound public finances, then donors pushing it down their throat won’t make it happen.”

    But Owen concludes that “If we think of politics as an endogenous characteristic of a complex system, then perhaps we have more hope of accelerating development by trying to tweak the internal feedback loops, and so shaping future system dynamics, than by offering exogenous solutions from the outside,” which resonates with the ODI report presented here.

    The question really is how often do these WoOs actually open? Is it often enough that we can make a difference? Or how can we facilitate the opening of these windows?

  2. David Jacobstein

    Hi Duncan,

    Curious about your thoughts on the second line in the box – the two pieces seem to talk past each other. I’m not sure how the donor guise of the political neutrality of development goals contradicts an acceptance that aid is inherently political and will work with the political incentive structure. It seems to me that on a strategic level, one can claim “children’s health is not political” while on a tactical level, using political understandings to cement compacts and secure support for reforms.

    Now, it’s certainly fair to say that donors do not do the latter very well, use of “Drivers of Change” notwithstanding. But it seems to me that the stylized current practice should be something like “donors are unaware of how political methods to further reforms” or “donors resist analyzing their work in terms of local politics” rather than being about the claim of neutrality of ends – the flaw is with the means.

    It does seem that the latter two categories would be hardest to change, given the political economy within donor agencies.

  3. Masood

    The challenge for practitioners is always to balance upward and downward accountability. The former is to donors who want log frames, result based management and so on while the latter is to the people who matter for aid, the ones whose life has to be changed. In an ideal world the two should be the same. But in aid business you are working across two cultures often very different to one another. This is the reality of aid. One of our evaluators often said that it was ironical that the organisation was most productive when it looked disorganised and messy and least productive when it was smooth, slick and perfect. Some years ago I was handling a DFID project. The provinicial minister was an expatriate UK based Pakistani who had made well in UK but returned to win an election and have the provincial local government ministry in which the Water and Sanitation Project lay which was funded by DFID. The Ministers entire agenda and views of the project was determined by the extent the project was willing to facilitate him with his many demands on the project vehicles. A few months down the road DFID arranged a visit to UK with the Minister for the project team. At london the project team was shocked when the Minister with whom the worked on a daily basis dealing with his demand expressed his ignorance about the Project team at DFID briefing when the introductions were being done. DFID was not too amused because the project had obviously not been able to influence policy, one of the big results expected of the project since they did not even know the Minister after two year of the project. The one who probably had a big laugh was the minister who understood the two cultures well and understood what policy and accountability meant in the two cultures

  4. Heidi Tavakoli

    Hi Marcus, I think your point touches on the binding constraint for effective aid. In line with Owen’s sentiment (and probably most other development practitioners) I think WoOs, where country led imperatives for change emerge, are always only a factor of country dynamics . However, once materialised, external actors may be able to play a more effective role in facilitating their translation into action. One of the biggest challenges in development is appropriately diagnosing the windows in the first place. Discerning a real imperative for change, rather than a priority that receives lip-service to appease certain constituencies (whether it be donors e.g. Matt Andrew’s signalling arguments, or parts of the populace) is certainly not easy for external actors who often have very limited understanding of the nature of the political networks in a country. It is also is an iterative process, as there is a continuous process of new doors opening while others close.
    Our Unblocking Results research draws out successful strategies used by the aid programmes to discern windows. It also suggests that employing the other enabling factors (in the natty diagram) supports the implementation of this imperative for change. By limiting the advocacy of external policy objectives and facilitating local problem solving, as well as supporting the implementation of legal mandates and politically popular goods and services, external actors are closer to supporting the grain, rather than trying to divert efforts to their priorities.

  5. Kieran

    Agree with much of above and does seem like a lot of current thinking beginning to realise this is “way foward”. Only thing I would add is a lot of good practice/development appears to be initiated/driven by “charismatic leaders” who can sell ideas, inspire people and manage all of the above. Development can be full of very well intentioned but somewhat uninspiring bureaucrats(is that why we love logframes).I sense these qualities are often not really sought or developed within development circles.

  6. I just saw Matt ANdrews’ presentation at ODI online yesterday and has thought a bit about the comment made by Per Molander: *Leave it to the politicians to define what’s politically achievable* (or something to that sort).
    It may be worthwhile turning the table upside/down and get politicians -or ‘people holding public legitimacy’ is probably better in context- to the centre of it. The table is still quite outside-in or instrumental at the moment.

    And then finally, the table should come with a Caution – Hot! Institutions are interconnected. So, there needs to be attention to the institutional spillover effects (unintended consequences). Working with what’s there might solve your immediate problem while reinforcing structures you’d rather avoid reinforcing. Of course, taking this too far will put a stop to doing anything at all, but, let’s be conscious about it.

  7. Rebecca Simson

    Hi David – Thanks for the suggestions about how to improve the table. I take your point that the two statements about political pay-offs are speaking at cross purposes. We’ll take your suggestions on board in future work.

    Hi Søren – while I fully agree with your point that politicians/policy-makers are at the centre of these processes and should be the ones to determine what is politically feasible, this research is primarily aimed at a technocratic audience. All other things being equal, how can external actors (donor agencies, practitioners) work more effectively in a political context? In my experience at least, donor officials often face the challenge that government officials/politicians in recipient countries (particularly heavily aid-dependent ones) will not outright reject aid proposals. They therefore need to find other means to gauge whether proposals have buy-in/are political feasible.

    Thanks for the warning label suggestion. I agree that there are often trade-offs between riding the political momentum and improving policy and institutional coherence. We discuss this in more depth in two of the case studies (‘addressing pay and attendance of health workers’, and ‘the Africa Governance Initiative in Sierra Leone’). Ultimately it’s a judgement call, but hopefully a better understanding of the trade-off leads to better decision-making.