Written by Ryan McGuine //
In October, economists Abhijit Banerjee, Esther Duflo, and Michael Kremer were awarded the Nobel Prize in Economic Sciences “for their experimental approach to alleviating global poverty.” All three recipients are Americans, though Mr Banerjee and Miss Duflo are immigrants. Miss Duflo is just the second woman to have won the prize, and the youngest ever recipient, at 46 years old.
Until the 1980s, the field of development economics, which seeks to determine why some countries grow rich while others remain poor, was mostly concerned with big questions. Economists studied the transition from rural and agricultural to urban and industrialized by using regressions to compare how variables affected growth across multiple countries. Unfortunately, such studies generated data that were often low-quality and difficult to interpret.
As a result of this poor data, debates about foreign aid were heavily ideological — for instance, between those think “poverty traps” abound, like Jeffrey Sachs, and those who do not, like William Easterly — and billions of dollars were spent on untested projects based on untested assumptions. Enter the Randomistas. In their own words, the recipients have sought to determine not “whether aid is good or bad, but whether particular instances of aid did some good or not,” using a technique called randomized control trials (RCTs) in economic and social studies.
RCTs have been utilized by the medical community for decades to determine the effectiveness of drugs, but more recently have been adopted by global development practitioners to quantify the impact of particular poverty reduction interventions. The idea is simple: divide a population into two groups using a randomized action like the flip of a coin or cast of a die. One group receives the intervention while the other does not, and at the end of the prescribed time period, the condition of both groups is evaluated. Compared to other econometric methods, RCTs rely on fewer assumptions, and produce results that are clearer to interpret.
Mr Kremer kicked off the field of testing anti-poverty interventions using RCTs during the early 2000s with educational policies in Kenya, and Mr Banerjee and Miss Duflo broadened the tool’s application. The pair started the Abdul Latif Jameel Poverty Action Lab (J-PAL) to promote the use of RCTs in development research, often working in partnership with Innovations for Poverty Action (IPA). With J-PAL and IPA serving as the nexus, the number of RCTs conducted over the last two decades has exploded, as shown below. RCTs are now used to evaluate interventions by researchers around the world, including those at the World Bank, NGOs, and universities.
Source: Martin Ravallion (2018), Should the Randomistas (Continue to) Rule?
In 2012, Mr Banerjee and Miss Duflo published the book Poor Economics for a popular audience. Poor Economics aggregates the lessons from RCTs conducted by the authors and others to present a coherent theory of how the extremely poor make decisions, and what kinds of interventions make them less poor. The end result is a series of policy prescriptions that might help incrementally chip away at the “three I’s of of development policy” — ideology, ignorance, and inertia — with the end goal of igniting virtuous cycles that lead to emergent institutional change.
Not everyone welcomes RCTs and the “aid effectiveness” movement they have heralded. Some argue that it is ethically dubious to prevent some group from receiving a treatment, and that RCT results are too little applicable to the real world. Another argument says RCTs over-emphasize measurable results, fostering programmatic interventions which might lift people above some arbitrary poverty line, but which fail to address the underlying poverty-creating systems (economic growth in Asia accounts for over three-quarters of global poverty reduction since 1980, causing some to suggest that Chinese government officials deserve the prize). Finally, it is possible RCTs have emboldened economists to overstate the policy implications of evidence gleaned from them, taking an engineering approach to policy-making in poor countries.
The critics have valid points, and the proper place for RCTs is probably one of many tools available to a development economist, rather than the top of a hierarchy of methods. However, rather than spending too much time criticizing the Randomistas, detractors would do better to emulate their example. Economists love innovation and entrepreneurship, and the three recipients have exemplified just such a spirit in their scholarship. No single econometric method or policy has the ability to “solve” global poverty; more experimentation, risk taking, and coalition building in academic research might though.