Entries in randomized control trials (2)


Hope, hybrid value chains, and graduating from poverty

Nicholas Kristof writes about how the realization that life can be better -- or the power of hope -- can transform a person, family and community. His recent New York Times column focuses on the findings of a large-scale randomized control trial involving 21,000 people in six countries; this trial looked at the impact of a type of aid package called a "graduation program."

Graduation programs (designed to help people graduate from poverty) give very poor families a significant boost that continues after the program ends. These programs can help the extreme poor establish sustainable self-employment activities and generate lasting improvements in their well-being. This program targets the poorest members in a village and provides a productive asset grant, training and support, life skills coaching, temporary cash consumption support, and typically access to savings accounts and health information or services.

Kristof quotes Sir Fazel Abed, who started the Bangladeshi aid group that developed the graduation program:

“Poverty is not just poverty of money or income. We also see a poverty of self-esteem, hope, opportunity and freedom. People trapped in a cycle of destitution often don’t realize their lives can be changed for the better through their own activities. Once they understand that, it’s like a light gets turned on.”

We like this kind of initiative; it's another form of the hybrid value chain partnerships that we develop with Maua and Bloom.

-- Clara Shen



Measuring social impact

Traditional measurement of the social impacts of investment focuses narrowly on counting the total number of units distributed or the number of beneficiaries supported, says Pablo Anton Diaz, a development effectiveness consultant at IDB's Opportunities for the Majority Initiative. While not discounting that data, he states it fails to determine the true benefits of intervention and contains no analysis into how to improve the impact's effectiveness as well as operating efficiency. That's because the standardized metrics of management information systems (MIS) are not only limited but are reliant on self-report data, which contains an inherent bias. Diaz points out the randomized control trials (RCT) and impact evaluations, though more expensive both in money and time, offer more insights into what outcomes are attributable to the investment.

He believes the best information comes from striking a middle ground between MIS and RCTs, and cites his own example of Ecuador as an example. Diaz explains that local financial institutions work directly with beneficiaries, either on small economic development or larger infrastructure projects. Leveraging that relationship, he was able to mine valuable client-level data, such as income level and savings amounts. He also discovered efficiency disparities between projects when he compared the different costs among projects that were for very similar uses and scales.

We have seen great work done in the area of RCT by some of our fellows and partners, and we are interested to hear your thoughts about the best ways to measure social impact.