By Obaid Khan, Humanitas Global
In this two-part series on evidence-based solutions to poverty, we first focus on the stylized insights to randomized controlled trials (RCTs) for solving the poverty puzzle.
Economists are and have often been criticized for portraying people as rational profit-maximizing economic agents without empirical evidence. Using abstract economic theories that are incompatible with ground reality can translate into extreme worldviews that either propose increased aid to end poverty or going cold turkey on paternalistic aid distribution. A recent wave of economists and practitioners has suggested a paradigm shift in development policy and practice by using tools from behavioral economics and hands-on experimental research. With these tools, it is possible to examine and understand what interventions work in some settings and not in others. In a sea of evidence, books like, ‘Poor Economics’ and ‘More than Good Intentions’ aptly document evidence-based and context-oriented field research to avoid the pitfall of prescribing policies as universal solutions. Organizations like J-PAL and IPA, neatly highlight the evidence with focusing focus on the context-specific rigorous evaluations of interventions to ascertain the effectiveness of proposed poverty solutions. Academics and practitioners have made an honest attempt to better understand the heterogeneous behaviors and ground realities of the poor. The trend of ‘humanizing’ the target population can help move the discourse and practice from universal to context-specific policy prescriptions to tackle poverty.
The overarching idea reiterated in the development discourse is that evidence, not rhetoric, must inform development policy and practice to tackle poverty. In the health realm, innovative thinking and adaptive policies are needed to improve health outcomes. For example, authors Banerjee and Duflo of ‘Poor Economics’ found that free provision of chlorine tablets does not necessarily improve usage, but placing them in dispensers near water collection point does. Similarly, Karlan and Appel in their book ‘More than Good Intentions’ found that in Seva Mandir catchment area, monitoring attendance and tying incentives to teacher’s attendance improved attendance and student scores. However, similar strategy failed to improve the attendance of health officials in nearby clinics and centers. Evidence from free versus subsidized mosquito nets in a study conducted by J-PAL and IPA, illustrates how health service utilization behaviors can change with marginal price changes.
The complex nature of poverty makes it difficult to prescribe universal solutions without contextualizing the issue under consideration and its inter-connectedness with other factors. Most sustainable development solutions might well lie in the middle ground, between the polarizing assertions that aid is the be all and end all of development and that aid does more harm than good. The primary tools employed, in development, need to be reorient towards rigorous randomized controlled trials, evidence from behavioral economics and context-based learning approach in different settings.
Academics have come to the realization that anecdotal evidence from field experiences complement randomized control trials to ascertain the effectiveness of various poverty or development-related interventions in different settings. Lessons from behavioral economics can also be utilized to explain results and finding from the field research, some of which may be counter-intuitive in neo-classical economic frameworks or models. Varying contexts have been and should be rightly focused on across health, education and agriculture sectors. Despite the evidence generated and valuable lessons learned, widespread application of evidence-based interventions is still elusive for developmental challenges. In the next post, we will explore how and why RCTs and field evidence should be an integral part of development discourse and practice.
