Telangana Graduation with Inclusion and Livelihoods Program (TGILP) Evaluation

Last registered on July 27, 2026

Pre-Trial

Trial Information

General Information

Title
Telangana Graduation with Inclusion and Livelihoods Program (TGILP) Evaluation
RCT ID
AEARCTR-0019120
Initial registration date
July 27, 2026

Initial registration date is when the trial was registered.

It corresponds to when the registration was submitted to the Registry to be reviewed for publication.

First published
July 27, 2026, 7:23 AM EDT

First published corresponds to when the trial was first made public on the Registry after being reviewed.

Locations

Region

Primary Investigator

Affiliation
Trinity College Dublin

Other Primary Investigator(s)

PI Affiliation
Trinity College Dublin
PI Affiliation
Trinity College Dublin

Additional Trial Information

Status
On going
Start date
2026-02-17
End date
2028-08-15
Secondary IDs
Prior work
This trial does not extend or rely on any prior RCTs.
Abstract
This study evaluates the Telangana Inclusive Livelihoods Program (TGILP), a government pilot inspired by BRAC's Graduation Approach, an international model that combines financial support, assets, training, and coaching to help ultra-poor households build sustainable livelihoods. TGILP builds on this model with a view towards strengthening households’ resilience to climate change and targets groups often excluded from government programs, including Particularly Vulnerable Tribal Groups (PVTGs) and women from marginalized communities. The targeted households are likely to face overlapping challenges of extreme poverty, social exclusion, limited access to services, and rising exposure to droughts and heatwaves.

The evaluation addresses the following core research questions:

- Overall effectiveness: What is the impact of TGILP — a government-led Graduation-style program combining livelihood grants, consumption support, coaching, gender intentionality, and climate resilience modules — on poverty, income, food security, and wellbeing among ultra-poor households in rural Telangana?

- Gendered and social inclusion pathways: Do the most socially excluded and vulnerable households — in particular survivors of the Jogini system and those belonging to particularly vulnerable tribal groups— derive additional benefits from the program that allow them to overcome local social stigma and exclusion?

To address these questions, we will conduct a cluster-randomized controlled trial in 100 villages across five districts of Telangana. Villages will be randomly assigned to receive TGILP or serve as controls. Within each village, eligible “Poorest of the Poor” households will be identified, with a focus on female representatives, and about 4,000 households will be randomly selected for the study. Baseline data will be collected before rollout in 2026, with follow-up surveys 24 months later. Results will inform TGILP’s potential expansion in Telangana and contribute to the Ministry of Rural Development’s national livelihoods strategy.
External Link(s)

Registration Citation

Citation
Casabianca Gonzalez, Maria Sofia, Selim Gulesci and Doireann O'Brien. 2026. "Telangana Graduation with Inclusion and Livelihoods Program (TGILP) Evaluation." AEA RCT Registry. July 27. https://doi.org/10.1257/rct.19120-1.0
Experimental Details

Interventions

Intervention(s)
Implemented by the Society for Elimination of Rural Poverty (SERP)—the rural development agency of the Government of Telangana—TGILP aims to enable sustainable economic advancement for marginalized communities, including PVTGs and socially excluded women, such as survivors of the Jogini system (former temple sex workers). TGILP builds on BRAC’s globally recognized Graduation model, which combines asset transfers, financial support, coaching, and service linkages. Key program components include:

- Livelihood Promotion Grants of up to INR 32,000 per household to support enterprise development.
- Short-Term Stipends of INR 1,000/month for six months to stabilize consumption.
- Individual Coaching and Peer Support to build financial literacy, confidence, connections to social protection schemes and market access support.
- Gender Intentionality Measures, such as support for socially acceptable livelihood choices for women, gender-based violence (GBV) and menstrual health training, and agency building efforts.
- Climate Resilience Measures, including climate-risk mapping, climate adaptation training, promotion of drought-resilient livelihoods, access to climate insurance, and community climate action planning.
Intervention Start Date
2026-08-15
Intervention End Date
2028-02-15

Primary Outcomes

Primary Outcomes (end points)
• Poverty and economic welfare: per-capita consumption, income, asset ownership, food security
• Gender and social empowerment: intimate partner violence, social and political inclusion, participation in community and government schemes (e.g. Self-Help Groups (SHGs), MGNREGS, pensions), agency and decision-making, marriage market outcomes, gender attitudes and awareness
• Climate resilience: livelihood diversification away from climate-sensitive activities, adoption of climate-adaptive practices


Primary Outcomes (explanation)
The study's primary outcomes assess whether TGILP alleviates poverty, builds climate resilience, and reduces gender-based inequality

Secondary Outcomes

Secondary Outcomes (end points)
• Financial inclusion
• Market access
• Access to basic services: water, health, education
• Physical health, including menstrual hygiene
• Mental health

Secondary Outcomes (explanation)

Experimental Design

Experimental Design
As a first step, SERP identified program-eligible households from 100 villages in 5 selected districts (Kumuram Bheem Asifabad, Narayanpet, Mulugu, Vikarabad, Nagarkurnool). Program eligibility is defined by SERP in consultation with community-based organisations, based on marginalisation, vulnerability, and exclusion. All eligible households must be:
• Rural, located in one of 8 mandals across 5 identified districts (Kumuram Bheem Asifabad, Narayanpet, Mulugu, Vikarabad, Nagarkurnool)
• Headed or represented by a woman
• Ultra-poor, belonging to one or more of the following socially excluded categories:
• Scheduled Tribe (ST) households, including Particularly Vulnerable Tribal Groups (PVTGs): Kolams, Thotis, Gonds, Koyas, Chenchus
• Joginis — women subjected to ritualised exploitation under the now-banned Devadasi system, facing social ostracisation, economic marginalisation, and denial of entitlements
• Widows, deserted or divorced women
In addition to the above inclusion criteria, the program has the following exclusion criteria:
• Households without a female representative
• Non-rural households
• Tribal households with a government job (excluded from SERP eligibility criteria)
• Households in villages outside the 8 designated study mandals

After receiving the list of program-eligible households from SERP, we randomly sampled 3236 households to be included in a baseline survey. Once the baseline survey was completed in each district, we randomly assigned all villages within that district into treatment and control groups. As such, randomisation was stratified at the district level. In addition, we stratified the randomisation by the number of eligible households and number of PVTG households within a village (above vs below median along both dimensions).

In control villages, eligible households will not receive any TGLIP intervention for at least 24 months after the start of the evaluation period. Control villages are not informed about the programme. After the endline survey, control villages will receive the TGILP programme.
Experimental Design Details
Not available
Randomization Method
Randomization done in office by a computer
Randomization Unit
Randomisation unit is a village. Randomisation was stratified by district, number of eligible households and number of PVTG households within a village (above vs below median along the latter two dimensions).
Was the treatment clustered?
Yes

Experiment Characteristics

Sample size: planned number of clusters
100 villages
Sample size: planned number of observations
A total of 3236 households were surveyed at baseline
Sample size (or number of clusters) by treatment arms
50 villages are assigned to the treatment group and the other 50 villages are assigned to the control group
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
IRB

Institutional Review Boards (IRBs)

IRB Name
Trinity College Dublin Research Ethics Committee
IRB Approval Date
2026-03-19
IRB Approval Number
4908
IRB Name
IFMR Human Subjects Committee
IRB Approval Date
2025-09-19
IRB Approval Number
IRB0000710