Targeting the ultra-poor to improve livelihoods

A multifaceted livelihood program has reached and improved the standard of living for more than three million households across 15 countries following randomized evaluations by J-PAL affiliates.
Smiling woman in a sari with two cows and a goat
Beneficiary of BRAC's program in Bangladesh. Photo: BRAC

The Bangladeshi organization BRAC designed a multifaceted livelihood program—known as the “Graduation approach”—which provides ultra-poor households with a productive asset, training, coaching, access to savings, and consumption support to help them graduate from extreme poverty. The program has since been replicated by several implementing organizations across the world. Research by J-PAL affiliates found that the model led to large and lasting improvements in consumption, food security, asset holdings, savings, and in some contexts, improved psychosocial well-being. The approach has been scaled up by BRAC to reach over 2 million households in Bangladesh, as well as an additional 1.1 million households in fourteen other countries. 

The Problem

Ultra-poor households often depend on insecure livelihoods and face a variety of impediments to sustainably transition out of poverty.

According to the most recent estimates from 2015, 10 percent of the world’s population lives on less than US$1.90 per day, the World Bank international poverty line.1 Many of these families depend on insecure and fragile livelihoods, including casual farming and domestic labor. Their income is often irregular or seasonal, putting laborers and their families at risk of hunger.

The first target of the Sustainable Development Goals is to eradicate extreme poverty for all people everywhere by 2030. Achieving this target will require the poorest of the poor to shift to more secure and sustainable livelihoods. Self-employment is often the only viable alternative to menial labor for the ultra-poor. Yet many lack the necessary cash or skills to start a viable business. To alleviate these constraints, many international and local NGOs have begun supporting multifaceted programs that foster a sustainable transition to more secure livelihoods.

The Research

J-PAL affiliates evaluated the Graduation approach in seven countries across a diverse set of contexts and implementing partners.

Originally developed in 2002 by Bangladeshi nonprofit organization BRAC, the Graduation approach combines six complementary interventions into a comprehensive livelihoods program for poor women. Implemented together over two years, the program provides a big push to help those living in ultra-poverty transition to more secure livelihoods. The six components of the program are:

  1. Productive asset: One-time transfer of a productive asset such as a cow, goat, or supplies for petty trade
  2. Technical skills training: Training to manage the productive asset
  3. Consumption support: Regular cash or food support for a few months to a year
  4. Savings: Savings account access or encouragement to save
  5. Home visits: Frequent home visits by implementing partner staff to provide accountability, coaching, and encouragement
  6. Health: Health education, health-care access, and/or life skills training

BRAC was interested in working with researchers to evaluate the model’s impact at scale and compare the results to an earlier impact evaluation led by an internal team of BRAC researchers. To do so, they partnered with J-PAL affiliates Oriana Bandiera (London School of Economics), Robin Burgess (London School of Economics), and Imran Rasul (University College London), together with Narayan Das, Selim Gulesci, Munshi Sulaiman, and Raniya Shams, to conduct a randomized evaluation beginning in 2007, when the program was scheduled to expand to 270,000 new recipients. Researchers found the program enabled the poorest women to shift out of farm labor and into running small businesses, increasing their earnings by an average 38 percent four years after the productive asset transfer. Evidence from this rigorous evaluation helped BRAC generate support to scale up the Graduation approach in Bangladesh and other parts of the world.

Bandhan Konnagar, the nonprofit arm of financial institution Bandhan, adapted the model in West Bengal, India. This adaptation was evaluated by J-PAL affiliates Abhijit Banerjee (MIT), Raghabendra Chattopadhyay (Indian Institute of Management), and Esther Duflo (MIT), with Jeremy Shapiro. They found that, in India, Graduation households experienced a five-fold increase in livestock revenue relative to comparison group households and were more likely to report having enough food to eat.

Very few programs are designed for this segment of the poor. For a total cost of Rs. 24,000 (US$360), this program provides a sustainable path out of poverty over two years.

- Mr. Chandra Shekhar Ghosh, Chairman and Managing Director, Bandhan Financial Services Limited

Across a diverse set of contexts and implementing partners, this cost-effective program has led to large and lasting impacts on ultra-poor households’ standard of living. For more information, see the Policy Bulletin summarizing different evaluations of the Graduation approach in Bangladesh, Ethiopia, Ghana, Honduras, India, Pakistan, and Peru.

From Research to Action

Rigorous evidence fostered support for replicating and scaling the Graduation approach around the world.

The randomized evaluation led by Bandiera and co-authors generated rigorous evidence of the model’s effectiveness and spurred BRAC to consider how the model could be further adapted. From 2012–2015, BRAC scaled the Graduation approach to reach over 580,000 women—more than double the number they had planned to reach before the evaluation—helping to improve the lives of more than 2 million people, when including the benefits to their immediate families.2 BRAC also began pilots in Afghanistan, Pakistan, South Sudan, and Uganda to contextualize the model to these settings.

In India, evidence generated from the randomized evaluation and subsequent policy outreach by the J-PAL South Asia office assisted Bandhan Konnagar in raising resources to expand the model. They have formalized partnerships with state governments of Rajasthan, Jharkhand, and Bihar to scale up the program. From 2012 to 2018, Bandhan Konnagar has reached nearly 60,000 ultra-poor households across the country, benefiting over 210,000 people in total across eight states in the country including Assam, Bihar, Jharkhand, Madhya Pradesh, Odisha, Rajasthan, Telangana, and Tripura.

In 2017, the state government of Bihar collaborated with J-PAL affiliates Banerjee and Duflo, with Shagun Sabarwal and Jyoti Mukhopadhyay, to evaluate the impact of a pilot program to adapt the Graduation approach in two districts, to be implemented by the state agency Jeevika with technical assistance from Bandhan Konnagar. Drawing on the pilot program, the Bihar government subsequently announced plans to adopt the Graduation approach in its Satat Jeevikoparjan Yojana (Sustainable Livelihoods Program), to reach 100,000 households across the state between 2018–2020.

As of 2020, BRAC had scaled this program to reach more than 2 million households (approximately 9 million people) in Bangladesh, as well as more than 1.1 million households (approximately 5 million people) in 14 other countries. Additionally, BRAC was selected as a 2020 grantee of The Audacious Project and will receive more than US$60 million to help an estimated 21 million people lift themselves from extreme poverty by 2026. The results from high-quality research are spurring international interest in further innovation. J-PAL affiliates and other researchers are beginning evaluations of the next generation of the program to better understand how to most effectively optimize the model’s components in different settings— including a model with a less-intensive coaching component and a version for urban areas.


Bandiera, Oriana, Robin Burgess, Narayan Das, Selim Gulesci, Imran Rasul, and Munshi Sulaiman. 2017. “Labor Markets and Poverty in Village Economies.” The Quarterly Journal of Economics 132 (2): 811–870.

Banerjee, Abhijit, Esther Duflo, Nathanael Goldberg, Dean Karlan, Robert Osei, William Parienté, Jeremy Shapiro, Bram Thuysbaert, and Christopher Udry. 2015. “A Multi-Faceted Program Causes Lasting Progress for the Very Poor: Evidence from Six Countries.” Science 348 (6236): 1260799.

Suggested citation:

Abdul Latif Jameel Poverty Action Lab (J-PAL). 2018. "Targeting the ultra-poor to improve livelihoods." J-PAL Evidence to Policy Case Study. Last modified June 2021. 

The World Bank. “Poverty.” Accessed October 5, 2018.
Based on an estimate of 3.5 people per household.