Can ‘Doing Development Differently’ only succeed if aid donors stay away from it?

May 22, 2018 8 By Duncan Green

Another day, another seminar on Adaptive Management/Doing Development Differently/Thinking and Working



Politically (let’s save words by just calling the whole thing DDD). This one was held under the Chatham House Rule, so no names or institutions. There was an interesting mix of academics and contractors – private companies who increasingly run the big contracts for DFID and other donors, and a few lightbulb moments in the discussion:

Firstly, how come DDD has become such a Thing when it seems to swim against the institutional tides of the aid business? It emphasizes systems thinking, uncertainty, learning by doing/failing and giving priority to local knowledge – that is, to put it mildly, a poor fit with the aid biz’ preference for control, logframes, risk minimization, and predictable, tangible results. At first it seemed like some kind of quixotic backlash against the forces of evil, but it seems to be getting traction all over the place – any theories as to why/how that kind of mainstreaming has happened?

One aid trend which on the face of it seems to run counter to DDD is the shift to outsourcing – DFID and others run multi-million pound contracts through a plethora of largely faceless management consultants and contractors, like Adam Smith International, who periodically get taken to task for profit gouging, poor performance or dodgy dealings. Not all management consultants are the same, but at the gold digger end of the spectrum are some companies who only want to maximise profits and minimise costs – they seem ill suited to the world of DDD, with its focus on uncertainty, deep curiosity, suspicion of blueprints, preference for experimentation, accepting and learning from failure etc. To be fair, at the other end of the spectrum are some contractors who do not seem solely motivated by $, and who are doing some serious work in the DDD area – Palladium, OPM and Abt are examples.

On the Fence

On the Fence

But a more fundamental challenge to the role of aid in DDD came from some of the Harvard crew (oops, there goes Chatham House. Sorry), who work directly with a range of governments to help pursue ‘Problem Driven Iterative Adaptation’. One speaker described its core principle as ‘always give the work back to government – it’s a coaching role, not us doing the work’.

In that model, academics may fly in to help, but they do so at the invitation of the government, and ideally, the government funds and owns the work. Riffing on my much-overused analogy of projects as linear cake recipes, the speaker said ‘PDIA accepts limits to how much we can understand about the system, which is why the cake has to be in hands of the people’. Move over Marie Antoinette.

If you introduce donors into that arrangement, ownership is diluted, and donor priorities and processes take over – the donors provide the recipe, and at best, the government gets to decorate the cake. A nice line from a TWP person: ‘Thinking politically comes naturally to many (in government, civil society etc); working politically less so. People breathe politics, then put it to one side when they talk to funders.’ See this great paper by Graham Teskey and Lavina Tyrrel for more on that.

One example: the vast majority of donor activity assumes the problem is largely known in advance – aid is siloed into education, health, infrastructure, governance etc, with calls from donors for different aid agencies/consultancies to bid for contracts. But PDIA works on the basis that you initially spend a lot of time helping local players clarify and understand the roots of a given problem, which may not look at all like what you start out with (you think the problem is poor quality education, but it turns out to be dodgy electricity or roads, or poor health among kids). Yet Aid doesn’t do agnosticism and finds it hard to hop between siloes. Tricky.

An even more fundamental challenge was raised in terms of systems thinking. The underlying model is of an external intervention into the status quo – the intervention is seen as external to the system that is being intervened in. That is questionable – donors may be only a minor perturbation of the system, in which case seeing them as separate may be a good approximation, but as soon as they start to have impact, they become part of it. Quite apart from the politics of meddling, there seems to be a fundamental contradiction in external interveners trying to understand/influence endogenous change as a separate thing from themselves.

So a purist might say ‘keep external roles to a minimum, aid donors stay away, and let local actors lead’. That has two

Not Good

Not Good

massive flaws:

  1. Yes, some governments can fund some of the PDIA style work, but many can’t – if you buy the argument that aid is still needed in many countries, then is there a way to politically sterilize it, taking it away from all the dictates of donors? One option might be to channel it through third parties, like the UN or World Bank, but they are hardly politics-free either.
  2. PDIA-style approaches start with being invited in by someone senior in government, who wants to sort out a problem. But what about governments that don’t want to reform? Can donors tiptoe around the accusations of meddling in internal affairs and support critics, opponents and future leaders with different views? Is that where DDD meets civil society support?

In other words, a good seminar, that left me with lots of questions to ponder.