Fighting worms to fight poverty

19 Luglio Lug 2011 1326 19 luglio 2011

Esther Duflo, director of the MIT's Action Poverty Lab on numbers, development and worms

  • ...
  • ...

Esther Duflo, director of the MIT's Action Poverty Lab on numbers, development and worms

By Marco Dotti

According to Esther Duflo what we need is “a pragmatic change of direction”. Duflo, 39, is the director of the Massachussests Institute of Technology's (MIT) Action Poverty Lab and is considered by the Economist to be one of the most influential “economic thinkers” of today's world. She is also the author of Poor Economics.

Her research has focused on “a new way of thinking and acting, based on numbers” because it is through numbers – that is to say a well pondered evaluation of the effectiveness of policies – that Duflo makes her case for a “new strategy to defeat poverty”.

Your research forces us to take a reality check. Poverty is not to be defeated in one go but by solving specific, quantifiable problems. How can we improve education? How can we fight malaria?

We have to remember that healthcare and education are what the economist Amartya Sen calls “capabilities”, which is to say things which are essential to the development of our lives. In the absence of these capabilities, words like “wellbeing” and “freedom” loose their meaning. Health and education are valuable in their own right but they also represent a growth factor. This is something that all economists, including the more conservative ones, agree on. Take the Chicago School, for example. People like Theodore Schultz, who coined the term “human capital”, or Robert Lucas, who turned human capital into a factor of growth and insisted on the power of contagion. According to Lucas a healthy individual is not only more “productive” and “efficient” but also makes those around him more productive and efficient. This helps us to see why health and education are so important and why it is important to have effective policies which can be monitored and tested. Big words are of little use, or rather, big words are big words but need to be turned into something real, something with a measurable impact.

The evaluation method you used to measure the impact of aid in developing countries, “randomized controlled trials”, is a rigorous one. You also used it to test social policies, particularly in terms of healthcare, education and microcredit. Did you encounter difficulties or resistance in applying this method?

When I first started my research, about 15 years ago, it was hard to get individual or institutional interlocutors to accept these kinds of ideas. Now, thankfully, things have changed. On the one hand it isn't as difficult to get them to see the advantages of testing the effectiveness of development policies on the field. On the other, the evaluation method is far more refined and enables us to achieve surprising results.

For example?

Impact evaluation helps to dispel many development myths. The results, if accurately evaluated, can be surprising, especially if we start from assumptions that have never been verified before. For example, if you were to ask almost anyone whether they thought that distributing free school books in developing countries encouraged children to go to school, it is likely that they will say that it does.

Does it?

No. In Kenya, for example, the distribution of free school books has not had any impact at all on education levels. We are not saying that school books aren't useful, only that they tend to affect those children who already have a fairly good level of education and who speake English. So this is one myth dispelled. Could fighting intestinal worms affect education? Strangely enough it does. It is the most effective and least costly means of increasing the number of children who go to school. Every one knows it is important to fight these kinds of diseases but the direct correlation between eliminating these illnesses, going to school and being educated can only be measured with specific tools.

How does your method work?

It is based on the same method used in clinical trials to test new drugs. The distinguishing feature is that the groups tested are selected randomly, so some people are administered a placebo and others the drug to be tested. I do exactly the same thing, so in the case of having to test the welfare system and in particular the effectiveness of school support mechanisms in India, 70 schools were chosen and different classes out of these different schools were placed under observation. Randomized trials have been around for a while, it was just a case of refining the data and results, which in the past didn't always happen.

Microcredit is under attack at the moment. What are your thoughts on the matter?

It is dangerous to have opinions about microcredit. Especially when your opinion is either extremely favourable or extremely unfavourable and when it is not backed up by facts and numbers. Did we really think that a fridge, a bicycle or a small investment could help people emerge from poverty? Sure, a bike and a fridge can help poor people live better but its not a solution. Now microcredit is being attacked but it is being attacked more out of revenge than scientific fact.

Where is the problem?

The problem is the big words. The problem is “Democracy”, “Fight against Poverty”, the problem are all these capital letters, these empty ideas which fall as soon as they encounter a little bit of realism. We need to go back to real things and not set impossible objectives. We need to look not just at “how much” but “how” relationships are forged, interventions are carried out, policies are made and expectations are created. We need to start from the field and oblige ourselves to keep our word.