To assess the effectiveness of the newly-designed UPG, researchers from the BRAC Institute of Governance and Development (BIGD), Innovations for Poverty Action (IPA), Northwestern University, University of Maryland, and Yale University have teamed up to conduct an impact evaluation, using a randomized controlled trial (RCT) on the 2019 UPG cohort.
Researchers: Nusrat Jahan; Tanvir Ahmed Mozumder; Sam Hsu; Md. Raied Arman; Akmam Binte Arif
Partners: Innovations for Poverty Action (IPA); Northwestern University; University of Maryland; Yale University
Timeline: 2019-2022
Status: Completed
Contact: Tanvir Ahmed Mozumder
tanvir.mozumder@bracu.ac.bd
Publications:
Context: First implemented in 2002, the Ultra-Poor Graduation (UPG) programme is BRAC’s attempt to address and resolve ultra-poverty in Bangladesh. From the beginning, one of the key characteristics of the programme has been to make necessary adjustments in its design based on programmatic learning and research findings to make it more effective. Consequently, the UPG programme has gone through many iterations and modifications over the years. The overall macroeconomic changes in the country have also led to significant changes in the profile of the ultra-poor. In response to these changes, in 2017, BRAC modified the selection criteria as well as the intervention design. This version divides the ultra-poor into three groups.
In comparison to its earlier iterations, the revised UPG programme is expected to be more cost-effective and adept at addressing the evolving nature of poverty in Bangladesh. In order to understand and assess the impact of this new design, BRAC Institute of Governance and Development (BIGD), Innovations for Poverty Action (IPA), Northwestern University, University of Maryland, and Yale University have undertaken a rigorous impact evaluation employing a randomized controlled trial (RCT) design on the 2019 UPG cohort.
Objectives: The objective is to see the socio-economic impact of the UPG program and disseminate the research findings by presenting the impact estimates across various outcomes of interest.
Methodology:
For the overall impact assessment, in 50 randomly selected BRAC branches that implemented UPG in 2019, 279 villages were randomly assigned to receive the programme. In each of these villages, 60% of the households were randomly selected to receive the programme (treatment). In these branches, another 276 villages were randomly assigned to the control group, where the programme was not delivered. To assess the impact of the intervention, the treatment and control groups are compared based on the data collected through in-person surveys conducted between January and April 2022. Of the 3,358 surveyed households, 1,678 were treatment and 1,680 were control. The study results presented in this report are based on these households.
Findings: Using the learnings from the different UPG cohorts, BRAC identified four core pillars of the Graduation approach, namely (1) livelihood promotion, (2) social protection, (3) financial inclusion, and (4) social empowerment. When BRAC redesigns and adjusts the programme’s components, sequence, and durations according to the context, it does not veer from these core pillars. These four interrelated broad outcomes result from positive changes in a host of different factors including assets, food security, savings, financial inclusion, health, social integration, and the productive abilities of the households.
