What actually happens when you look closely at poverty

The debate about global poverty has long been dominated by large, confident theories. Aid works, says one camp; aid corrupts, says the other. Free markets will lift all boats; structural investment in health and education is what countries need. Abhijit Banerjee and Esther Duflo — MIT economists who later won the Nobel Prize — spent years doing something quite different: they went and looked at what actually happens when specific interventions meet real lives.

Poor Economics is the account of what they found. Working across dozens of countries and using randomised controlled trials to test specific programmes, they examined why poor households make the choices they do — why they sometimes spend on items that seem frivolous to outside observers, why they borrow at high rates even when cheaper credit is available, why they sometimes do not take up free vaccinations or insecticide-treated bed nets that could prevent serious illness.

The answers are consistently more interesting than the grand abstractions suggest. Poor people often live in information environments that differ markedly from those of wealthier people — they frequently lack reliable data about the actual returns on education or the efficacy of medical treatments. They also face psychological pressures and time constraints that push strongly against long-term planning in ways that standard utility-maximisation models simply do not capture.

Banerjee and Duflo are sceptical of both aid-maximalists and aid-sceptics. Their evidence suggests that specific, well-designed interventions — a particular form of immunisation programme, a specific structure for school attendance incentives, a carefully calibrated microcredit scheme — can work consistently, while poorly designed versions of ostensibly similar interventions reliably fail. The difference lies in the detail, and the detail only becomes visible when you look closely at real lives.

The book is a sustained case for empiricism over ideology in development economics, and it is written with genuine respect for the people being studied. Banerjee and Duflo are not trying to solve poverty from a comfortable distance; they are trying to understand it from the inside, one careful study at a time. That methodological humility is itself an important and undervalued contribution to a field often dominated by confident prescription.

Banerjee and Duflo were awarded the Nobel Prize in Economic Sciences in 2019, jointly with Michael Kremer, for their experimental approach to alleviating poverty — a recognition that the field had genuinely changed because of their methodology. Poor Economics is the most accessible account of that approach and its findings, written for a general audience without sacrificing the complexity that makes the work valuable. It is also, quietly, an argument about intellectual honesty in a field that has sometimes preferred confident prescription to careful observation. The evidence, it turns out, is more interesting and more useful than the theories that were meant to replace it. The evidence, it turns out, is more interesting and more useful than the confident theories it was meant to replace, which is itself an important and undervalued lesson for the field. That is itself an important lesson for any field that prefers confident prescription to careful observation.