Social Science and Public Policy at Peradeniya

I wrote last week of the conference on the social sciences and humanities I attended at Peradeniya; my account was of one set of insights I gleaned there – that helped me understand my own world better. Often, we think of intellectual pursuits of having little relevance to practical things, and I would think understanding Facebook better, as I tried to in my previous column would be an antidote to that. This week, I want to offer a taste of even more practical, useful social science research that was presented at that conference by Dr. Suresh de Mel, a senior lecturer in Economics there, and also, an old and good friend from my school days.

But first a sketch of the issues involved. How do we figure out whether particular public policies work? Especially those that affect the everyday lives of ordinary people, most of them are not rich, but poor. Often governments, or NGOs, set up nowadays as alternatives to elected governments, for better or for worse, advocate or implement micro social programs that they claim will help poor people. Let’s take an example, micro credit. Now, how do we know if it works? Under what conditions does it work or not, and how if it does work, can it be made to work better?

Superficial evaluation
Usually, and up until now, programmes have been implemented and then later, an evaluator is hired to assess the success or failure of the programmes. Unfortunately, a micro credit programmes, for let us say catalysing small entrepreneurs, will take at least two years in real on the ground time to have any impact worth assessing, but the evaluation will typically have only a six week time frame. Results are thin; many ordinary, well understood, secure social science techniques cannot be used in something like a six week time frame, and we end up with a superficial evaluation. The cycle repeats itself.
De Mel, and his collaborators, and he has several drawn from universities in the US and the UK, build on pioneering, but breathtakingly simple techniques pioneered by Esther Dufulo and Abijith Bannerjee, located at what’s called the poverty lab at MIT, whose new book, Poor Economics: A Radical Rethinking of the Way to Fight Global Poverty has won prizes and got rave reviews. The technique, borrowed from clinical trials in medicine is the double-blind field study, done over the long duration. In the paper de Mel and his colleagues presented at the Peradeniya conference, one of many, perhaps over a dozen, papers they have authored on this cluster of issues, micro credit infusions for small entrepreneurs was the object of inquiry. Rather than act as six week evaluators on something already done, in this study the researchers themselves, using careful randomised techniques, select participants in several groups, adjusting in each study for gender, previous business experience, aptitude, and training. Then, results are tabulated over many months, during and then after the micro credit infusion has ended. I will not be able to do justice to their finding here but suffice to say, they seemed far more granular and yet more robust, that any ‘project evaluation’ I have read.

Breathtaking result
Learning from one result was breathtaking – of the female entrepreneurs who were selected for a micro credit infusion, the gains of the infusion declined to almost zero after 16 months. Does that not give you pause? De Mel should be congratulated for the careful, painstaking work he is doing; all of us can learn from it.
These techniques, it is important to understand, are not simply limited to questions about micro credit or even, in the strict sense economics. Applications are wider as endless as the researchers’ imagination. For example MIT poverty lab experiments have included asking questions as useful and yet unasked as, how absenteeism among school teachers could be reduced. It turns out that a controlled study proved that having a student photograph the teacher, with a time/date stamped digital camera, at the beginning and end of the school day, with a small financial incentive to reward increased attendance, to be hugely successful; much more so than the old idea of signing the attendance register!
I hesitate of course to proclaim panaceas. No doubt there are none. But yes, we always will have public policies. It is a good idea to test them out.