The role of the international community in tackling global poverty has always been contentious. While there is an evident moral obligation for richer countries with the expertise and resources to assume a more active role, immense uncertainty remains over how this assistance should be organised. Dr Rachel Glennerster, Chief Economist for the United Kingdom’s Department for International Development (DFID), shared with me where she believes the answer lies: in evidence-based solutions. Randomised controlled trials are an integral part of her research, which includes the evaluation of community-driven development in Sierra Leone, empowerment of adolescent girls in Bangladesh, and health, education and micro-finance in India. “If you have a new approach to teaching in school, choose 200 schools, randomly pick some to adopt the new approach and see if there’s better learning in the schools with new teaching.”
Rachel’s rich diversity of experience working on the issue of international development, such as in the International Monetary Fund and the International Growth Centre, a research centre based jointly at the London School of Economics and Political Science and the University of Oxford, has made her realise the importance of evidence-based solutions for ending poverty. “There are many different approaches to anti-poverty – education, health, government and women’s empowerment – where you can test ideas before putting money into them. You want to know if they work or if they have harmful effects.” This ensures that the policy recommendations are feasible and applicable for implementation when DFID works with governments or non-governmental organisations in developing countries to roll out programmes.
Though the research “can just rely on existing test scores from children’s end-of-school examinations”, she notes the dearth of data in developing countries, which requires researchers to collect new data. Offhandedly, Rachel rattles off examples of data that could be collected, such as how much in microcredit a family consumes, children’s test scores, and even blood and hair samples. “You can test anything, from asking people to measuring physical things on the ground. It just depends on what the programme is designed to do.”
Some scepticism lingered within me as Rachel explained the effectiveness of evidence-based solutions. How receptive would DFID’s beneficiaries be to intervention? After all, DFID is making recommendations from an elevated position and expecting developing countries to adopt these supposedly effective solutions to a local problem. Rachel concedes that what people are open to varies a lot and “rigorous evidence is never the only thing people take into account when making a decision”, but stresses that it is an increasingly important component of the decision-making process. “Not everybody wants to hear that their programme is not effective. But some have said, ‘we absolutely want to know if our programme works because if it doesn’t, we don’t want to be wasting people’s money and time, and more importantly, local people’s time.” Local governments and non-governmental organisations do want anti-poverty programmes to work, and although intervention is not always guaranteed or successful, the beneficiaries often appreciate feedback from an evidence-based perspective.
From the other side of the world, how then does DFID see their role in such intervention? “Part of the role of international aid is to help build an evidence base that all developing countries can use.” Rachel refers to evidence-based solutions as a global public good that everyone has access to and is able to utilise. Yet, there will invariably be a tension between local politics and international aid. “Most people agree that we should be getting rid of fossil fuel subsidies because it is not just bad for the environment but also for the poor as most of the money is spent on subsidising rich people who use more gasoline. But it is very politically difficult for the government to suddenly increase prices. And it is not for the British government to tell them what they should do.” International aid is thus a collaborative effort, where the recipient country is actively seeking reform and the aid agency is careful in exercising sensitivity to local politics. She highlights the need to “understand the local context before you can give good recommendations on how to help – the political nuances of which ministers are pro-reform, the political space available, and whether it is a suitable time for the country to take on a challenging reform.”
Interestingly, despite the political differences in each country, Rachel is surprised by how human beings act very similarly. She laughs while pointing out that “human beings are extremely bad at spending money now for uncertain and unseen benefits that come a long time in the future” and cites the debate over vaccines in the West at the moment. “If you charge for preventative health, even a small charge like forcing people to walk for it, it really reduces the take-up of preventative health. This is something you find in different products, in different people and in different countries because fundamentally, we as human beings are very bad at making that trade-off.” This commonality in human nature makes policy interventions somewhat similar in their directions, even as local differences are taken into account.
Although it is only recent decades that the incorporation of evidence-based research, like the use of randomised controlled trials, into development policies has gained traction, data analysis has been in practice for a long time. It is sometimes difficult to identify concrete relationships in the research, and Rachel cautioned against identifying correlations instead of precise cause-and-effect relationships. She agreed this is a challenge evidence-based research faces and raised criticism that it might translate into less attention channelled towards issues that are hard to measure. She remained unfazed, however, “It just means you need to get better.”