Back to History Current Version

Testing a Budgeting Intervention to Address Seasonal Hunger

Last registered on June 04, 2025

Pre-Trial

Trial Information

General Information

Title
Testing a Budgeting Intervention to Address Seasonal Hunger
RCT ID
AEARCTR-0015744
Initial registration date
April 04, 2025

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
June 04, 2025, 9:48 AM EDT

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

Locations

Region

Primary Investigator

Affiliation
UC Berkeley

Other Primary Investigator(s)

PI Affiliation
UC Berkeley
PI Affiliation
Stanford
PI Affiliation
UC Berkeley
PI Affiliation
UC Berkeley
PI Affiliation
UC Berkeley
PI Affiliation
One Acre Fund Malawi
PI Affiliation
One Acre Fund Malawi

Additional Trial Information

Status
Withdrawn
Start date
2025-04-01
End date
2027-08-31
Secondary IDs
Prior work
This trial does not extend or rely on any prior RCTs.
Abstract
In Malawi, 6.7 million people experience hunger between harvests with serious impacts on health, child development and economic outcomes. Our prior research (Cognitive constraints in consumption smoothing, RCT ID AEARCTR-0004409) applied techniques from the psychology literature to the complex planning problem facing smallholder farmers: how to allocate annual harvest income across the many months before the next harvest.

We will conduct a randomized evaluation of a low-cost, light-touch and portable planning exercise that draws on a large literature in psychology demonstrating how humans have difficulty making realistic plans for the future. This is a scale up of our original project, conducted in collaboration with One Acre Fund (1AF), a social enterprise that provides agricultural input loans and training services its smallholder client farmers.

Our planning intervention gives farmers a structured way of budgeting their available harvest income to cover expenses; this makes spending plans more realistic and creates urgency around the need to save early in the season. The exercise prompts farmers to 1) think through all potential income and expenditures using categories, 2) use a visual planning board to create a plan for spending and savings over the year and 3) attach labels representing the plan to existing maize stocks. The intervention is farmer-centered, low-cost (uses simple, cheap, and easily replicable materials), quick, suitable for group settings and low-literacy, and adaptable across contexts.

Farming households are randomly assigned, at the group level, to receive the budgeting intervention (treatment) or not (control). We will measure impacts on savings, consumption (including work) and farm output.

Registration Citation

Citation
Banda, Esau et al. 2025. "Testing a Budgeting Intervention to Address Seasonal Hunger." AEA RCT Registry. June 04. https://doi.org/10.1257/rct.15744-1.0
Sponsors & Partners

There is information in this trial unavailable to the public. Use the button below to request access.

Request Information
Experimental Details

Interventions

Intervention(s)
We use a cluster randomized trial to test the impact of a budgeting intervention integrated into 1AF’s operational activities at scale.
The budgeting intervention will involve 1AF farmers in the treatment arm attending a 1.5 - 2 hour planning session. A facilitator (Field Officer) describes the budgeting intervention and accompanying visual planning board to the participants, completing a demonstration planning exercise for all participants to observe and inviting all participants to complete their own, household based, planning board at the session. The intervention involves thinking through all possible projected income and expenses for the upcoming year, using categories outlined on the planning board. Participants use stickers to represent money (Malawi Kwacha) and bags of maize they have harvested or expect to harvest on their boards. The facilitator will encourage questions from the group and will walk around regularly to check in and make sure everyone understands the exercise. The facilitator will end the session by helping any participants that request support in completing their board. Participants are given the boards to take home with them. 1AF farmers in the control arm do not receive the budgeting intervention.
Intervention Start Date
2025-07-14
Intervention End Date
2026-08-31

Primary Outcomes

Primary Outcomes (end points)
1 - Savings - Household available cash on hand (survey questions) and maize stocks (direct observation by enumerators)
2 - Off-farm labor supply - Total family off-farm labor hours over the previous month, wages, earnings
3 - On farm investment - Labor: days and hours worked and hired in the past week. Other inputs: total expenditure on each input (herbicides, pesticides, fertilizer)
4 - Harvest output - Total output in value in Malawi Kwacha
5 - Market level outcomes - Maize prices and ganyu temporary wage rates
Primary Outcomes (explanation)

Secondary Outcomes

Secondary Outcomes (end points)
1- Grain sales and purchases - Quantities, prices and dates for crop purchases and sales
2 - Consumption/expenditures - Modified LSMS expenditure module
Secondary Outcomes (explanation)

Experimental Design

Experimental Design
The budgeting intervention (treatment) will be administered by agricultural Field Officers (FO) who oversee 20-25 groups of farmer clients with ~20 farmers per group.

Groups will be randomly assigned to treatment and control within each FO – specifically, we will evaluate 3 control groups and 3 treatment groups per FO.

Outcomes are measured through household surveys over the course of at least one year.
Experimental Design Details
Not available
Randomization Method
Sample selection: We first sample FOs, then sample farmer groups for selected FOs, and finally assign selected farmer groups to treatment and control. Among FOs in our 4 study districts, we first drop new FOs. From the remaining FOs, we randomly select 100 FOs. Within each selected FO, we randomly select 9 farmer groups, conditional on (a) accessibility during the rainy season and (b) group clustering.

The 9 selected groups per FO are randomly assigned to one of the following arms: Treatment, Control or Extra. The extra groups are excluded from the evaluation sample, but help achieve scale targets for our partner organization.
Randomization Unit
Randomization unit is the farmer group – entire groups are assigned either to treatment or control.
Was the treatment clustered?
Yes

Experiment Characteristics

Sample size: planned number of clusters
600 farming groups
Sample size: planned number of observations
3,600 households
Sample size (or number of clusters) by treatment arms
300 farming group treatment, 300 farming groups control
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
IRB

Institutional Review Boards (IRBs)

IRB Name
UC Berkeley Committee for the Protection of Human Subjects
IRB Approval Date
2025-06-03
IRB Approval Number
2025-01-18192