I’ve never been more excited about the Nobel Memorial Prize in Economics.
On October 14 it was awarded to 3 amazing economists – Abhijit Banerjee, Esther Duflo, and Michael Kremer (henceforth Laureates) – for their pioneering work in the use of experimental methods in development economics.
So far in my own career, I’ve worked the most in the field of development economics, which hinges on a fundamental question: Why are some countries poor and others rich?
The Laureates expanded the toolkit of development economists by borrowing an experimental method in medicine called randomized controlled trial or RCT and mainstreaming it in the study of poverty alleviation.
In this piece let’s talk about what RCTs are all about, how effective they are, how they’ve been used in the Philippines, and why Filipino policymakers ought to support such studies.
RCTs are nothing new. They’ve long been used in clinical studies on the effectiveness of medical treatments such as drugs.
In these studies, a new drug is tested on people who are randomly assigned to two groups: a “control” group (who consumes a placebo) and a “treatment” group (who consumes the drug).
As long as these groups are statistically identical, any difference in the two groups’ health outcomes must be the effect of the new drug.
The Laureates mainstreamed RCTs in development economics. People, too, are assigned at random to control and treatment groups, but economic interventions can range from cash transfers, the use of bed nets (kulambo), or even deworming. (READ: Poverty reduction: What works, what doesn’t)
As you can imagine, applying such experiments across communities – vis-à-vis a lab – can be daunting. Critics are also quick to point out the issue of “external validity”: cash transfers, bed nets, or deworming might work in a small Indian town but not necessarily in a Philippine town.
Nonetheless, the Laureates have demonstrated that RCTs can be used widely to assess many types of development programs.
The Laureates also showed us that RCTs are the closest we have to a scientific, evidence-based approach in assessing anti-poverty programs. This is in stark contrast to previous policies by governments and NGOs which were characterized by iffy correlations, excessive hand-waving, or sheer guesswork.
RCTs have been applied to a multitude of areas in development economics, from education, health, credit, and even gender and politics.
In education, for example, the Laureates sought to investigate how textbooks, flip charts, computer-assisted learning, deworming, and school meals could improve children’s attendance in class and performance in tests. They also investigated the impact of bonuses as a way to minimize teachers’ absenteeism.
In health, the Laureates explored how bed nets treated with insecticides could prevent the spread of malaria, how safer water supply containers could abate the incidence of diarrhea, and how mobile vaccination clinics could encourage parents to immunize their children.
Other experiments have looked at the use of small, time-bound discounts in farmers’ purchase of fertilizers, the impact of reserving seats for women in village councils, and the effects of microcredit access on households’ incomes and women’s empowerment.
A lot of these studies were described in Banerjee and Duflo’s 2011 book titled Poor Economics: A Radical Rethinking of the Way to Fight Global Poverty. See Duflo’s TED Talk on economic RCTs as well.
Across all these studies, impact is often intricately tied to the context in which an intervention is being applied. But over the years, some policies and programs turned out to be more effective than others.
In education, for example, a 2015 meta-analysis found that, “interventions that focus on improved pedagogy…are particularly effective, and so are interventions that improve school governance and teacher accountability.”
More than the effectiveness of any one intervention, RCTs have proven a valuable way to winnow bad anti-poverty programs from good.
Most importantly, the Laureates have shown us that even the simplest and cheapest programs can work wonders to improve the lives of the poor.
RCTs in the Philippines
A lot of RCTs have in fact been conducted in the Philippines.
The Philippine government’s flagship anti-poverty program, called Pantawid Pamilyang Pilipino Program (4Ps), is a conditional cash transfer program that has been the favorite of many RCTs.
Last April President Rodrigo Duterte signed the 4Ps Act, which institutionalizes the program and provides for annual funds. Unfortunately, 4Ps’ first wave of impact evaluation was conducted with a flawed RCT study. Future waves of evaluation had to resort to other empirical techniques.
Other RCTs in the Philippines have been more successful.
For instance, a commitment savings device helped households to save more in Caraga back in 2000 to 2003. A month-long reading marathon in Tarlac improved the reading scores of Grade 4 students. And telling voters in Sorsogon they can accept money from politicians (but ought to vote with their conscience) only worsened vote-buying.
One of the more recent RCTs sought to uncover the impact of religiosity on poor Filipino households’ incomes and life satisfaction.
‘Baliw na baliw sa research’
In the Philippines today, evidence-based policy-making is under serious threat.
Senators express open disdain for pure research. Senator Cynthia Villar recently lambasted the Department of Agriculture’s budget request for corn research, saying, “Baliw na baliw kayo sa research” (You’re so crazy about research).
Duterte himself cares little for research-based policies. His war on drugs continues despite overwhelming evidence of its glaring failures abroad. (READ: War on drugs? Other countries focus on demand, not supply)
But this year’s Nobel Prize in Economics is a testament to the power of research, scientific thinking, and empirical evidence in the crafting and implementation of economic policies and programs.
Is it too much to ask for leaders and policymakers to embrace rather than reject evidence-based policy-making? – Rappler.com