Transition to Low-Carbon Ag-Tech Solutions: A Marketing Campaign for Adoption of Solar Irrigation Pumps in Kenya

How to cite

Mulwa, Richard; Jeuland, Marc; Czajkowski, Mikayla. 2025. "Transition to Low-Carbon Ag-Tech Solutions: A Marketing Campaign for Adoption of Solar Irrigation Pumps in Kenya." AEA RCT Registry. AEARCTR-0016298. https://doi.org/10.1257/rct.16298-1.0

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

Richard Mulwa* , Marc Jeuland , Mikayla Czajkowski

* Primary investigator

Registry ID
AEARCTR-0016298
DOI
https://doi.org/10.1257/rct.16298-1.0
First registered
Last updated
Status
In development
Trial dates
2025-07-24 – 2026-04-15
Countries
Private
Registry record
http://www.socialscienceregistry.org/trials/16298

Abstract

Climate change poses a significant threat to agricultural sustainability in Kenya, with adverse effects on the livelihoods of millions of smallholder farmers and Micro, Small, and Medium Enterprises (MSMEs). Increasingly harsh weather patterns, extreme temperatures, and unpredictable rainfall have led to reduced crop yields, loss of livestock, damaged infrastructure, and overall food insecurity. Alongside adaptation efforts to cope with the adverse impacts of climate change in the agricultural sector, mitigation interventions aimed at transitioning the country into an inclusive green economy (IGE) are being adopted in various sectors of the economy. These interventions aim at transitioning the country to a low-carbon development path that is resource efficient and socially inclusive. However, there are limited low-carbon agricultural-technology solutions that can aid agricultural MSMEs to adapt to the changing climate, enhance productivity, and reduce greenhouse gas emissions. In addition, several challenges including limited access to resources and opportunities for smallholder farmers especially women and youth, and agricultural MSMEs hamper the adoption of low-carbon agricultural-technology solutions. This study aims to generate evidence for better-informed national and county-level policies and programs aimed at promoting the adoption of low-carbon agricultural-technology solutions among smallholder farmers through: (1) targeted informational awareness campaigns on smallholder farmers including youth and women; (2) training and capacity building to improve the welfare and empowerment of youth and women using low-carbon agricultural technologies; and (3) the adoption of solar water and irrigation pumps increase the welfare and empowerment of youth and women.

Intervention

Our intervention will be an informational campaign in form of an enhanced marketing campaign targeting potential customers of Sunculture (local MSMEs selling solar water pumps). An encouragement design is suitable for evaluating the impact of this intervention because we cannot exclude potential customers from participating in the informational campaign program. We shall therefore randomly select households to receive the information campaign aimed at increasing the awareness of low-carbon farming technologies supplied by Sunculture. This should increase the probability of potential adoption relative to those who do not receive the informational campaign advertisement. The encouragement design will enable us to evaluate the local average treatment effect (LATE).

Based on a conservative power calculation, we shall use a sample size of 1000 households for each intervention, where 50 percent of the participants will be assigned to the treatment group (500 households) and 50 percent to the control group (500 households). The power calculation assumes a 5 percent level of statistical significance and a statistical power of 80 percent. There are several potential issues regarding internal validity that may arise from using the encouragement design and random assignment. First, contamination arising from the control group participating in the program may arise. One way of estimating impact in such a case would be to use random assignment as an instrument to estimate the LATE.

Experimental design

We propose to randomize non-adopting households into two groups: a) standard marketing, and b) enhanced marketing. The enhanced marketing intervention (hereafter “marketing +”) compared to standard door-to-door marketing is conducted in accordance with SunCulture sales agents’ typical approach. The “marketing +” intervention will include the following extra elements:
a. More targeted information delivery, in the form of a video shown on a smart phone or tablet that combines evaluation results with farmer testimonials
b. Invitation to attend a “rainmaker” party with a local model farmer
c. Referral incentives delivered to local model farmer based on converted customers
The informational campaign will comprise enhanced marketing campaign targeting potential customers of SunCulture.

Delivery modality: All of the door-to-door components will be delivered by a University of Nairobi team trained in collaboration with SunCulture according to its standard approach. SunCulture will decide when these teams are ready to deliver the intervention. The “rainmaker” party will be organized and supervised by SunCulture in coordination with University of Nairobi (for communication of timing to households recruited in the door-to-door component). Alongside the delivery modality, the following were collected: baseline data among non-adopters, and endline data collected at least one growing season later to assess causal impact among the adopters in the new group. The data will be collected in the counties of Makueni, Machakos, Uasin Gishu, Trans Nzoia, Bungoma, and Kakamega.
2. Baseline: This was completed in April-May 2025.
3. Intervention: This will be in July -August 2025.

Design details

Randomization method
A Randomized selection of non-adopting households using a computer application will divide the households into two groups: a) standard door-to-door marketing based on SunCulture’s sales’ agents’ approach, and b) enhanced marketing intervention, i.e., “marketing +”. The enhanced market will have more targeted information delivery (combining evaluation results with farmer testimonials on video) and referral incentives. An invitation will be sent to non-adopting households to attend a “rainmaker” party with a local model farmer.
Randomization unit
The unit of randomization will be an individual household.
Clusters
800 households
Observations
800 households
Arms
400 control farmers and 400 treated.
Primary outcomes
1. Adoption of solar water pumps 2. Energy use patterns 3. Crop yields, incomes and savings (change in welfare)
Secondary outcomes
1. Time management 2. Perceptions of climate risk and vulnerability

Cited by

No citation data available for this registration yet.

Keywords