Doing Development Differently: Report back from two mind-blowing days at Harvard

October 28, 2014

     By Duncan Green     

Spent an intense two days at Harvard last week, taking part in a ‘Doing Development Differently’ (DDD) seminar, hosted by Matt Andrews, who runs Harvard’s ‘Building DDD logosState Capability’ programme and ODI. About 40 participants, a mixture of multilaterals and donors (big World Bank contingent), consultants and project design and implementation people, and a couple of (more or less) tame NGO people like me (here’s the participants list).

The purpose was to learn from success, based on 15 short (7m) filmed presentations, which are all online, and ensuing discussions. The premise of the meeting was that there is something like an incipient movement around DDD. As you would expect at such an early stage, it is fragmented and messy (people using the same words to mean different things, lack of clarity on what is/is not included, overlap with other initiatives like the Thinking and Working Politically crew etc), and clearer on what it is against (linear thinking, tyranny of the logframe etc) than what it is for. So this meeting aimed to try and clarify terms and ways of thinking, and build something like a community of practice and consensus among adherents.

The main focus of the discussions was how to bring about institutional reform, in particular of the state – the theme covered by Matt’s great book. So lot’s about civil service reform and that kind of thing. But they had some outliers – I spoke about our work supporting grassroots Women’s Leadership Groups in Pakistan. The common elements that emerged from the case studies include:

Iteration: Place lots of small bets, ‘fail fast and cheap’ (a research assistant running off with $20k to have a sex change operation definitely takes the biscuit), make sure you have realtime monitoring and feedback, and mechanisms in place to let you learn and adjust the programme as you go. This is the antithesis of ‘spend 2 years designing the perfect project, and then roll implacably out’.

Success is based on relationships. Your key people have to be deeply immersed in the culture (either because they are local, or at least have lived there for decades rather than years). As an outsider, it’s only when you start getting invited to the weddings that you know you have a chance of understanding what is really going on.

Deep study of the system, based on continuous observation and listening. In Nicaragua, UNICEF sent public officials out to try and access the public services they were administering, and even made the men carry 30lb backpacks to experience what it’s like being pregnant! This is all about immersion, rather than the traditional ‘fly in, fly out’ consultant culture.

networks-200_r3Quick wins are vital to build momentum. Low morale is often a major obstacle – state officials resigned to the impossibility of change. So showing quick results both builds confidence, and gets the big bosses excited and supportive.

Brokering is often more important than delivery. A lot of the work involves brokering discussions between different state bodies who seldom talk to each other, or between donors and governments who are at loggerheads. Facilitation skills are crucial.

Incrementalism: This approach favours lots of small steps, going with the grain of existing institutions, rather than big bangs or shock therapy

So what are the problems with such an approach? In terms of making it function in the current aid environment:

First, it is heavy on staff and salaries – building relationships, mentoring and coaching etc takes time, tricky if the donor is determined to shift lots of money fast.

It is often necessary to try and ‘get the cash off the table’ so that officials’ eyes are on finding solutions to problems, rather than dazzled by dollar signs. Options include pushing it into ‘payment by results’ (I was struck by how far the World Bank has moved on this), having no money anyway (an NGO speciality) or making it clear from the beginning that big contracts are not on offer.

complexity signResults are a ‘perennial problem’. ‘We’re best when we are invisible’ said one speaker, but how do you demonstrate results while being invisible? Interesting discussions on whether we need to develop new metrics to measure trust – an equivalent to the private sector’s obsessive monitoring of brand loyalties. In any case ‘unless we can demonstrate value for money, this wave may pass’.

Who in the aid business stays anywhere for 10 years? Not if you want to develop your career, anyway. Success seems to involve handing over power and control to local staff or partners – something the big donors seem to find rather difficult.

But I have some broader concerns, which echo some of my worries about elements of the ODIs Africa Power and Politics Programme. Namely, what are the politics of all this? For all its talk of understanding context, politics etc, the overall discussion felt extremely technocratic. Understanding politics appears to be mainly of instrumental value in that it helps you implement a reform programme more effectively, but there was little discussion on who decides the content of the programme. So are we talking about tweaks or transformation? Is everything win-win, or are there sometimes conflicts because someone loses out? Does it matter whether we apply these approaches in an autocracy or a democracy? Apparently not. A few times I raised the issue of power and social justice, and how it distorts relationships and policy choices, but should be at the centre of what we do, but this was largely shrugged off – I’m starting to realize how gender advisers must feel.

And then there was a fascinating conversation about the tyranny of ‘The Project’, but that will have to wait for tomorrow.