You have to talk to the poor if you want to learn about poverty
Everyone basically agrees that we should be fighting poverty and helping the billions of extremely poor where we can, but the question of how to do this quickly becomes a heated debate. Some think we’re just lacking enough money, others that we need a totally new approach to delivering aid, or even that aid is damaging and should be stopped.
These discussions are almost always awash with sweeping statements, and simplistic stereotypes which are unhelpful and counterproductive – or so think Abhijit V. Banerjee and Esther Duflo, co-founders of MIT’s Poverty Action Lab (J-PAL), and authors of the excellent new book Poor Economics: A radical rethinking of the way to fight global poverty. Banerjee and Duflo argue we should move away from impossibly big questions – such as ‘does aid work?’ – to more specific answerable ones, with a focus on hard evidence rather than just theorizing.
Look at the data!
Merely making the questions more specific doesn’t achieve anything unless we can give a good answer. Here Banerjee and Duflo, along with their colleagues at J-PAL, are pioneers in the new movement of Randomized Control Trials (RCTs), to try and produce concrete data.
Take childhood vaccination. We know this is one of the most effective way of saving lives, and yet despite many programmes working in this area, the immunization rate among children in some areas remains very low – sometimes as little as 1%. To work out why this is, the researchers carried out an RCT in 134 villages in the Indian district of Udaipur, randomly assigning them to different test groups. They found that by making immunization easier by providing monthly camps in the village, immunization rates were three times higher than in the villages in the control group. But when as well as the camps a small incentive was added – a free kilo of lentils per child immunized – the rate rose to more than six times the level of the control group. Priceless information for planning cost-effective programmes.
In the book we hear colourful stories of Mexican schoolchildren, Nigerian businesswomen, and many more from the 865 million people worldwide who live on less than 99¢ per day; and we learn from RCTs investigating the most effective way to increase the use of mosquito nets to prevent malaria, to get children to stay in school for longer, how best to assist farmers in poor countries, and more.
Accepting the findings
What emerges is often unexpected – for instance giving out deworming tablets was a far more cost-effective way of improving education levels than investing in new teachers. And some of J-PAL’s findings have been the focus of some controversy for being perceived to be sceptical about microfinance. In fact, the study in Hyderabad gave good evidence that people were using microloans to make sound investments in their businesses – but they also found no improvement in empowerment of women or spending on health or education. The authors do not claim microfinance doesn’t work, but rather that it is not the cure-all solution that some have made it out to be in what the authors describe as “decades of overpromising”.
In fact what repeatedly emerges is a surprisingly complex picture. Mothers are discouraged by having to walk kilometres to health centres for immunizations, but also a small incentive can stop them putting off what they had intended to do anyway. Banerjee and Duflo repeatedly emphasise the conclusion that there is no ‘magic bullet’ solution.
A cause for frustration?
So will the many people working hard to fight poverty be dispirited by this book? Not only does it repeatedly emphasise the lack of any ‘magic bullet’ solution, but time and again we see instances of poor people apparently resisting our efforts to help. They frequently have to pay large amounts they can barely afford on curative medicine, having previously failed to take very cheap preventative measures, a farmer might decline highly subsidized fertilizer which would pay for itself many times over in increased yield, and mothers need bribing before they will have their children immunized or educated. We even hear of people in Africa with barely enough to eat spending the food grants they are given on satellite TV. Surely a humanitarian reading this will despair – why won’t these people act in their own interest?
But again, the mistake is in viewing poor people as some 2-D characters, rather than the complex individuals we all actually are. There are reasons they act how they do, which even if not objectively ‘rational’ are almost always understandable. Maybe they’re missing some important information, or their existence is so precarious they have no option but to be highly risk-averse, even if that means passing up a good deal. Like us, they will sometimes opt for immediate pleasures, such as TV or tobacco, rather than long-term investments.
The book makes the insightful observation that, with our pension plans, enforced schooling, and many other “nudges”, it is made much easier for us in the developed world to make the right choices (and yet we still all know the feeling of putting off what’s best for us – all those failed New Year’s resolutions and unused gym memberships). The poor, on the other hand, bear “responsibility for too many aspects of their lives”, and are constantly having to balance choices and risks, acting as “barefoot hedge-fund managers”.
In fact Poor Economics sounds a note of hope. We do need to move away from lazy stereotypes, unfounded speculation, and the misguided policies they produce. But if we take the time to understand the complexity of people living in poverty and their problems, real progress can be made – the authors give countless examples of successful programmes and encouraging new directions. And even if this approach at first makes the problems seem even more daunting and intractable, as they write, “success can sometimes be closer than it looks”.
You can watch Esther Duflo explaining more about RCTs here.
Watch the Banerjee and Duflo discuss Poor Economics on India’s NDTV.
Blogpost written by Ben Mason
