Experiments for development?

May 6, 2010

I just watched the TED talk by Emily Duflo, a French economist who believes it is possible to know which development solutions work and which don’t – by using randomised experimental trials to test different approaches to development problems. I was fascinated with her idea, and can clearly see the how policy-makers and development practitioners could find it useful.  She is right, there is a lack of information on the outcomes of development, and the more information policy makers have the more likely they are to make good decisions.

But I also felt very uncomfortable with the idea.  And I don’t think it was just my natural aversion to numbers and to economists. After some reflection I realised I had three core issues with Duflo’s idea, and these concerns were strong enough to drive me out of my unintentional thesis-writing induced blogging hiatus to post this (particularly as a quick google search didn’t turn up any other bloggers discussing it).

My three concerns are these:

1. Generalisation and over-simplification. Although Duflo does briefly state at the end of her talk that one of the limitations of the idea is that what works in one place may not work in another, I think she underestimates how much of a limitation this is, or perhaps she overestimated just how much the results of an experiment in social policy can be generalised.  Her example of the education experiment is one example.  This experiment showed that one of the most effective ways of increasing the number of years children stay in school is through worm treatment.  This is very likely the case in many areas… where there is a worm problem.  Anyone with a health background might have been able to remind her that this will of course only continue to be effective if treatment is repeated over and over… or if the source of the worms is eradicated. Just giving a worm treatment is simplistic.  Unfortunately while most development problems are more complex and multi-faceted, experiments by nature are only able to test a limited number of variables and therefore lend themselves to simplistic solutions.

2. Developing world comunities as experiment subjects. While I know Duflo and her research group probably have great motivations and want to help find development solutions, I feel very uncomfortable with the idea of using experimental methods in this context.  There are significant ethical issues related to knowledge and power, including questions who is designing and running the experiment, who owns the data, how the participant communities are represented. There’s a paternalism inherent in the idea of doing experiments like this, where white western researchers define the problem and offer the  solutions. I admit I don’t know how much input she has had from participant communities and developing world researchers, but at this point the whole idea of doing experiments on developing world communities seems rather neo-colonial to me.

3. Ignorance of structural and institutional injustice. This relates to point 1 and over-simplification. Duflo thinks that underdevelopment can be addressed by addressing the smaller problems – getting kids immunised or into school for example.  This is similar to an idea I have heard from research participants in my own research – that development is the sum of many small changes, from the bottom up.  I was not convinced when I did my research, and I am not convinced now.  While there is, and will always be some value in making changes at a grassroots or community level, I don’t believe that these changes are enough.  Solutions like those offered by Duflo do not take into account structural and institutional conditions that perpetuate systems of poverty. Racism and sexism, class and caste systems, corruption, these all contribute to poverty and until these are addressed (and the way in which they are addressed will differ between communities) there will be no significant development.

While I can appreciate the value of Duflos work for those working on specific development problems in defined areas, these issues ring big alarm bells for me.  It seems to much of a western-expert-led quick-fix-solution to me and I wouldn’t like to see it replacing long term qualitative research and development experience, and even less, usurping the knowledge of developing world communities themselves.

As an antidote to the discourse on saving the developing world, perhaps Duflo might benefit from some conversation with Design for the First World, a competition for designers, artists, scientists, makers and thinkers in developing countries to provide solutions for First World problems.

I think the world need more of this kind of thinking.

Our fellows in the first world often come to visit and give us their well intentioned but often very problematic ‘solutions’. We thought, why don’t we pay back? Dx1W is a competition for designers, artists, scientists, makers and thinkers in developing countries to provide solutions for First World problems.


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: