Table of Content
2.1 Randomized Controlled Trials
2.2 Key Examples
2.4 Political Detachment
3. Placement in Literature
“After all, we have spent billions of dollars on aid. […] Has it done any good? And sadly, we don’t know. And worst of all, we will never know.” (Duflo 2010, 01:37)
Poverty alleviation is a huge humanitarian challenge and also the supreme discipline of some economists. The effort to fight poverty had a mixed success so far and scholars claim different reasons for that outcome (cf. Besley 2012). In “Poor Economics”, Abhijit Banerjee and Esther Duflo avoid the often polemic ‘Sachs vs. Easterly’ aid debate and promise a “radical rethinking of the way to fight global poverty”: They make a convincing case about learning on the lives of the poor and the unique choices they have to face in their lives. Banerjee/Duflo understand how the poor perceive their conditions and come to the decision they make and are therefore able to craft better strategies and interventions that actually help the poor and do not produce unintended consequences. By doing that, they deliberately target the micro-level and leave out the level of politics or institutions, which is somewhat problematic, to search effective approaches in answers of the smaller questions.
This essay is structured in the following way: After an introduction to the analysis (2.) and an overview of Randomized Controlled Trials (RCTs), the main method of Banerjee/Duflo (2.1), the analysis will provide an assessment of key examples of the two scholars’ research (2.2). Subsequently, an evaluation of limitations (2.3) and shortcomings (2.4) of the book will be conducted. By evaluating the approach and findings of “Poor Economics”, the essay will be able to compare it with other scholarly works (3.). A conclusion (4.) will summarize the review and end with a personal take on what has been learned through studying the book.
Banerjee/Duflo claim that most aid today is based on untested assumptions or even misperceptions of the lives and decisions of the poor. According to them, this can be drastically changed and the conditions of the poor understood through effective interventions that were tested on the ground. Their approach can be summarized in that they want to a) try to stay out of the mainstream aid debate between Sachs and Easterly, b) understand the life, situation, behavior and constraints of the poor and c) create and test innovative interventions.
2.1 Randomized Controlled Trials
The main method Banerjee/Duflo use is the scientific approach of field experiments with Randomized Controlled Trials (RCTs), in which interventions are tested through treatment and control groups that are chosen at random. So Banerjee/Duflo, therefore sometimes called ‘randomistas’, use experiments to find better, respectively more efficient ways to fight global poverty. Besley called RCTs “one of the most important innovations in economists' tool kits” (Besley 2012) and Banerjee/Duflo are the scholars who pushed this tool in their Poverty Action Lab (J-PAL) forward and increased its credibility in the scientific community.
Some readers may feel that RCTs are morally wrong to execute because first, the other half of a group loses potentially out and second, one shouldn’t experiment with the lives of the poor (cf. Easterly 2009). But most of the time, there is only enough money to apply the intervention for a portion of the population anyway. Regarding the experimental concern, Banerjee/Duflo would argue that human rights review boards grant clearance for them. Other criticism focuses on the hype of RCTs and their increasingly preferred use over other methods of gaining knowledge of the poor (see Easterly 2009, Deaton 2010). Despite the critics, Banerjee/Duflo argue that “push[ing] on the right lever can make a huge difference” (Banerjee/Duflo 2012: x) and it is their goal to find these levers with RCTs.
2.2 Key examples
One strong side of “Poor Economics” is the rich fundus of interventions that were tested successful with RCTs but also examples of aid policies that clearly did not work. Banerjee/Duflo devote the first part of the book to the private lives of the poor and the second part to markets and institutional approaches. What overall stands out is the insight the reader can gain into the lives of the poor, about the challenges and decisions they face that are so extremely different to those that people in developed countries have to deal with.
A lot of the data and knowledge that was gathered in the book is based on careful observations and interviews of the lives of the poor. A series of views into their research will follow to show what their findings is about: Banerjee/Duflo expose how the poor spend their money wrongly in terms of health – on cure instead prevention, on private providers or traditional healers that are expensive and by demanding wrong cures (cf. ibid., 51ff.). Their explanations are sound and convincing as they single out a lack of information and trust (cf. ibid., 58f.), as well as time inconsistency problems with prevention (short-sightedness). Banerjee/Duflo call for making the right choices easier to take for the poor, in one example by providing chlorine dispensers at water wells. Their solution depicts their ideology: They are, in cases where they tested it successfully, pro incentives and try to find subsidies/nudges (cf. ibid., 63ff.) to help the poor in making the right choices. Examples for this include their bed net study, in which they found with RCTs that handing them out for free or highly subsidized makes sense as the total use actually decreased when charging for them or the immunization study, in which they found that it is after all cheaper to give out lentils to people as an incentive for immunizing their children than not incentivizing the immunization at all.
These kinds of findings can be very helpful for policymakers who are interested in crafting the most efficient interventions for their respective countries. For example, the cognition of deworming children to increase their school years proved to be very effective and was taken up likewise by governments, policymakers and NGOs. But some cases are more convincing than others, at one point Banerjee/Duflo state “that there is no smoking gun to prove that larger families are bad for children” (ibid., 110) and don’t find quality-quantity relationships/trade-offs, although they argued earlier that parents indeed pick winners and losers (cf. ibid., 88f.) out of their children to advantage one over others.