Impact of Ultra-Poor Graduation Program on wage rate and health

How to cite

Tamim, Sheikh Arman; Sulaiman, Munshi; Rahman, Atiya. 2025. "Impact of Ultra-Poor Graduation Program on wage rate and health." AEA RCT Registry. AEARCTR-0017131. https://doi.org/10.1257/rct.17131-2.0

Download: BibTeX · RIS — for reference managers without a browser translator. Zotero and most others pick up this page's citation metadata directly.

Sheikh Arman Tamim* , Munshi Sulaiman , Atiya Rahman

* Primary investigator

Registry ID
AEARCTR-0017131
DOI
https://doi.org/10.1257/rct.17131-2.0
First registered
Last updated
Status
On going
Trial dates
2019-04-01 – 2026-12-31
Countries
Private
Registry record
http://www.socialscienceregistry.org/trials/17131

Abstract

The Ultra-Poor Graduation (UPG) programme has been rigorously evaluated over the years in different contexts, with strong evidence documenting its significant impacts on ultra-poor households’ economic outcomes (Bandiera et al., 2017; Banerjee et al., 2015, 2021). Several studies have shown that participation in the programme leads to increased income, productive asset ownership, and consumption, helping households escape extreme poverty through a multifaceted package of support. However, most of these existing studies focus on household- and individual-level effects. There remains a substantial evidence gap regarding the community-level impacts of such large-scale poverty reduction interventions, despite their potential to shape local labour markets, social structures, and norms.

Bandiera et al. (2017), using data from a large-scale randomised controlled trial in Bangladesh, find that the UPG programme successfully shifted ultra-poor women toward more productive income-generating activities. Ultra-poor women devoted more hours to livestock rearing, and fewer hours to agricultural labour, and maid services relative to their counterparts in the control villages. Near ultra-poor and non-eligible poor women from treatment villages also experienced higher wage rates in agricultural labour and maid services. These findings suggest that the UPG programme might affect community-level wage rates through shifting labour supply.

The proposed study aims to assess the community-level impacts of the UPG programme, with a primary focus on wage effects. It will also examine the relationship between community-level land inequality and wage rates. First, we will examine the impact of the UPG programme on wage rates at the community level. Second, we will investigate whether the relationship between land inequality within communities and local wage rates differs between communities with and without UPG programme exposure. Third, we will explore whether participation in the programme enhances women’s labour market agency, particularly their ability to negotiate and engage in wage employment. We will also investigate whether these effects vary in response to household-level shocks or natural shocks.

Experimental design

For implementing the research, the operational duration of each UPG program branch was used for stratified random sampling to select the research branches. These 100 branches are spread across 37 districts in Bangladesh. Among these, 50 branches have been randomly assigned to the treatment arms. From the rest of the 50 branches, half of the villages were selected as pure control, while in the remaining half of the villages, 60% of households were assigned to treatment and 40% to control. The interventions consist of a partial asset grant (partial repayment for an interest-free loan), savings support, enterprise coaching, and community resource mobilization.

Design details

Randomization method
Randomization done in office by a computer
Randomization unit
Branch. Half of the branches are selected as pure treatment branches. For the remaining 50 branches, half of the villages in each branch are selected as pure control, and in the other half of the villages, 60% of households are assigned to treatment and 40% to control.
Clusters
100 Branches
Observations
60000 households.
Arms
50 Pure treatment branches, 50 branches with pure control households, treated households, and spillover control households
Primary outcomes
Wage rate, mortality rate

Cited by

No citation data available for this registration yet.

Keywords