Can Social Experiments Help Fight Poverty?

Award-winning French-American economist Esther Duflo would argue that yes, randomized and controlled social experiments are essential to fighting poverty. 

When it comes to solving global poverty, how can we be sure that the aid we give is the most impactful and most efficient? To eradicate poverty, do we just need to spend more money? Or rather, does money speed up the process of corruption or lead to dependency? When we look to the past, it is impossible to tell exactly, precisely what historical impact humanitarian aid has had on impoverished communities.

In her TED Talk, Duflo introduces a thought experiment to her audience: Let’s say you’ve been given a few million dollars, and you want to use it to help the poor. So what do you do? How do you go about spending your money?

Since the mid-1990s, Duflo and her colleagues have turned to field experiments to successfully, scientifically obtain reliable, robust, and actionable solutions to global poverty. The key, they’ve discovered, is to break the question of global poverty into small, manageable issues. 

In both her TED Talk and the Nobel Lecture she gave after receiving the 2019 Prize in Economic Sciences, Duflo introduces her audience to a number of these experiments and their global implications. 

How can we ensure that children are receiving immunizations that could save their lives? 

Duflo emphasizes that immunization is the cheapest way to save a child’s life, and still, at least 25 million children every year don’t receive the immunizations they should. She explains, “The technology is there, the infrastructure is there, and yet it doesn’t happen.”

Duflo and her colleagues conducted a randomized, controlled field experiment in 134 villages in Rajasthan, India. In one third of the villages, they made immunizations easy by offering a monthly camp where mothers could get their children immunized. In the next third, they made immunizations easy by offering the camp, but they also created an incentive of one kilo of lentils, which, although small, could convince parents to bring their children now rather than later. The final third of villages in the trial were left as an unchanged control group. 

They discovered that offering the camp alone increases immunization rates from 6 percent to 17 percent. Adding the incentive of lentils increases immunization rates sixfold, to 38 percent, and because the nurse is being paid anyway, the cost per immunization actually decreases when incentives are given compared to when they are not.

How can we protect millions of individuals from Malaria? 

Malaria kills nearly one million individuals every year and is the leading cause of under-five mortality. Bed nets to protect from mosquitoes are a cheap and easy fix, but some say that, if offered for free, the insecticide treated nets will be undervalued and used, perhaps, as fishing nets instead.

In Kenya, researchers distributed vouchers with 100 percent, 50 percent, and 20 percent discounts on bed nets. They discovered that first, individuals who only got partial discounts were less likely to actually purchase bed nets. Second, people will use bed nets if they have them, regardless of how they got them. Third, people who received a bed net for free were more likely to purchase another one year later compared to individuals who initially had to pay for their bed net. Duflo concludes that “People do not get used to handouts; they get used to nets.”

How can we ensure that the poorest individuals receive benefits intended for them?

Indonesia has a rice subsidy plan called Raskin which costs 2.2 billion dollars and subsidizes rice for the country’s poorest families. Unfortunately, the poor receive just 30 percent of their intended subsidy. Researchers wondered if distributing an ID card informing individuals of their eligibility for the program would increase its impact. What would be the ideal ways to distribute the card? What would make it most impactful?

After conducting randomized trials, research determined that a wide distribution of cards was more effective in increasing individuals’ knowledge of the subsidy program than a wide distribution of informative posters spread across a town. The subsidy program was most utilized when specific price information was included on the cards than when it wasn’t. An impression of accountability through a feature like a clippable coupon on the card was not influential in increasing individuals’ use of the subsidy program. These discoveries increased poor families’ take-up of the rice subsidy program by 26 percent. Leakage was also reduced, making the program more cost effective for the government, who quickly scaled up the program after the study was concluded. 

---

Field experiments like these can span a huge spectrum of social and economic issues. Duflo has played an integral role in conducting research as far-ranging as “Nudging Farmers to Use Fertilizer: Experimental Evidence from Kenya,” “Assessing the effect of conditional cash transfers on pregnancy outcomes in France,” and “Deterring Drunk Driving in India.” 

Poverty is an undeniably massive global issue, but rather than let its scope overwhelm us, we can choose to learn from Duflo’s experiments. She divides global problems into manageable pieces and approaches these with an open mind and a scientific framework. She tells her TED Talk audience, “These economics I’m proposing… it’s a slow, deliberative process of discovery.” And yet, by uncovering answers and transforming them into actionable policy, Duflo prompts change for the better. 

To learn more about randomized, controlled social experiments that study poverty alleviation and other development topics, check out The Abdul Latif Jameel Poverty Action Lab, a global research center founded by Esther Duflo, Abhijit Banerjee, and Sendhil Mullainathan, that ensures policy is influenced by scientific evidence rather than intuition. As of last year, over 400 million individuals across the globe have been positively influenced by policies stemming from discoveries made by the J-Pal network. 

You can also find more articles and information about this topic on Novel Hand’s Poverty & Development resource page.