Group vs. individual livelihoods: Can the graduation model help mitigate information asymmetries that hinder group enterprise development?

How to cite

Karlan, Dean; Quidt, Jonathan de; Bryan, Gharad; Udry, Christopher. 2024. "Group vs. individual livelihoods: Can the graduation model help mitigate information asymmetries that hinder group enterprise development?." AEA RCT Registry. AEARCTR-0013583. https://doi.org/10.1257/rct.13583-1.1

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

Dean Karlan* , Jonathan de Quidt , Gharad Bryan , Christopher Udry

* Primary investigator

Registry ID
AEARCTR-0013583
DOI
https://doi.org/10.1257/rct.13583-1.1
First registered
Last updated
Status
On going
Trial dates
2024-03-04 – 2026-01-10
Countries
Private
Registry record
http://www.socialscienceregistry.org/trials/13583

Abstract

The graduation model has demonstrated consistent evidence of cost-effectiveness in uplifting ultra-poor households’ livelihoods, though these often remain household enterprises. Graduation models, with intensive coaching, training, and group aspects, may facilitate forming larger, jointly-owned, enterprises. We test group versus individual livelihoods while providing coaching to address hypothesized information-asymmetry barriers to forming equity-sharing group enterprises. We also compare large and small grants, to test for increasing returns to investment above certain thresholds. We will examine process changes, such as whether more participants choose to start group enterprises, short-run business outcomes, and group dynamics, as well as longer-term livelihood profits and household wellbeing.
Leveraging the 20-year Philippines Socioeconomic Panel Survey, we track individual, household, and community-level at regular intervals of 4 years. We seek partial funding for a follow-up survey 4 months post-intervention, complemented by management data from the implementing partner, to enhance our understanding of medium-term impacts of the program.

Experimental design

See the attached protocol.

Design details

Randomization method
Randomization is done by computer with R software.
Randomization unit
Barangay (Philippines' municipality)
Clusters
405 communities
Observations
N = 6,075 participants in 405 communities
Arms
1,215 participants in 81 communities in each of the 3 treatment arms, and 2,430 in 162 communities in the control group.
Primary outcomes
This RCT is primarily focused on household welfare outcomes and individual psycho-social outcomes, which will be measured through the in-person follow-up survey that is the focus of this proposal. Key outcomes at the household level include monthly income, monthly consumption per capita, savings and borrowing, productive assets, labor, and food security. At an individual level we will measure subjective well-being, life satisfaction, subjective socio-economic status, community engagement and trust, and risk preferences. These will be combined with intermediary outcomes from a business health survey and monitoring data, including proportion of participants that elect to set up a group business, business survival, revenue growth, profitability, and indicators of business group dynamics including decision-making processes, division of labor and specialization within teams, profit-sharing, and collaboration among group members.

Cited by

No citation data available for this registration yet.

Keywords