The Social Observatory: A Living Lab for Developing a Science of Delivery

Currently home to one-third of the world’s poor, India is at the center of many of the world’s most innovative efforts to end poverty and boost shared prosperity. It is no surprise then that the Social Observatory—a joint initiative of World Bank research and operations teams to transform how development is delivered—chose India as the place to set up operations four years ago. The initiative is embedded in the $2 billion portfolio of World Bank livelihoods projects currently operating in the country.

“As researchers we are very interested in having significant policy impact,” said Research Director Asli Demirgüç-Kunt. “One of the big advantages of being a research department housed in the World Bank is working closely with our operational colleagues on initiatives like the Social Observatory, which is precisely how we researchers learn from experience.”

Livelihoods projects work with communities to lift the poorest of the poor out of poverty. The core of these projects are self-help groups, self-selected groups of women in villages throughout India which meet regularly to provide each other support and work together on a range of anti-poverty programs such as business training, microfinance, and health services.

But like many aid programs that use a community-driven approach, livelihoods projects face a challenge in adapting to diverse cultural contexts and constantly changing circumstances on the ground. Thus entered the Social Observatory in 2011, which now works with livelihoods projects in the Indian states of Bihar, Tamil Nadu, Odisha, and Maharashtra that together serve nearly 20 million people.

In a recent Policy Research Talk, Vijayendra Rao, a Lead Economist in the Research Department and the dynamo behind the Social Observatory, explained how the Social Observatory sprang out of the findings of a landmark World Bank research report on participatory development, ‘Localizing Development: Does Participation Work’. The report found that participatory, community-driven projects often fail to be sensitive to complex contexts—including social, political, historical and geographical realities—and fall short in terms of monitoring and evaluation systems, which hampers learning.

The Social Observatory assembled an interdisciplinary team of economists, sociologists, behavioral scientists, information management specialists, and others to see if they could overcome the failures of the past. The team wanted to create projects that are nimble, learned by doing, and are able to make mid-course corrections in response to local conditions—in short, projects with adaptive capacity.    

To illustrate how this approach works in practice, Rao focused on the example of Jeevika, a livelihoods project in the Indian state of Bihar. Between 2006 and 2011, Jeevika worked with nearly 400,000 clients. A retrospective evaluation carried out by the Social Observatory team found impressive impacts from the program, with a near tripling of savings, a large reduction in households with high-interest debt, and a much greater empowerment of women.

Jeevika’s impact was in part due to the concrete resources participants received such as group money or a passbook. But the embedded approach of the Social Observatory allowed them to uncover less obvious but equally important reasons for the impact. Self-help groups forged an alternative identity for poor women that cut across caste. This allowed the women to participate in the social and political life of villages in a way that was previously unthinkable. Deeply entrenched social norms were transformed in a relatively short amount of time. “Culture is not an immutable constraint for development,” said Rao.

In another project, Pudhu Vaazhvu in Tamil Nadu, the Social Observatory has invested heavily in generating project feedback loops through a new approach to data called ‘participatory tracking’. Cutting-edge technologies like tablets are paired with participatory processes such as community-based generation of questionnaires to generate a real-time picture of project performance that is relevant to all stakeholders. To make the data truly democratic in settings where literacy is limited, the Social Observatory has also developed entirely new systems of data visualization. Communities are empowered to be true stakeholders rather than passive recipients of aid.    

Shobha Shetty, Practice Manager, South Asia Region, Food and Agriculture Global Practice, also shared her perspective on the value of the multi-year partnership. “One of the biggest advantages of this really exciting collaboration has been the rigor and independence that the Social Observatory has brought to our livelihoods projects, which are big and complex,” Shetty said. Compared to other experiences, where evaluators parachute in, the program was able to “test, learn, and adapt to achieve measurable results specific to the states that they operate in.”