Preferences, Norms, or Cost in Polluting Energy Usage: Experimental Evidence from Ethiopia

How to cite

Efobi, Uchenna; Delessa, Kitessa; Alemu, Tekie; Boka, Jonse; Kahsay, Goytom. 2025. "Preferences, Norms, or Cost in Polluting Energy Usage: Experimental Evidence from Ethiopia." AEA RCT Registry. AEARCTR-0015277. https://doi.org/10.1257/rct.15277-1.0

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

Uchenna Efobi* , Kitessa Delessa , Tekie Alemu , Jonse Boka , Goytom Kahsay

* Primary investigator

Registry ID
AEARCTR-0015277
DOI
https://doi.org/10.1257/rct.15277-1.0
First registered
Last updated
Status
On going
Trial dates
2025-02-03 – 2026-06-30
Countries
Ethiopia (Africa)
JEL codes
Q5, Q54, Q56, Q57
Registry record
http://www.socialscienceregistry.org/trials/15277

Abstract

Environmental pollution is one of the key policy issues in many developing countries, and Ethiopia faces particularly significant challenges due to its heavy reliance on solid fuels, a primary source of GHG emissions. Around 94% of households still use solid fuels for cooking and other daily activities, leading to significant health and environmental issues, among others. This widespread reliance on traditional biomass contributes to deforestation, indoor air pollution, and respiratory illnesses, especially among the most vulnerable—women and children. Limited access to modern energy sources, along with economic constraints and cultural practices, makes Ethiopia a key case for studying household energy choices and developing sustainable solutions. This research aims to document household preferences for energy sources in Ethiopia, by comparing preferences for polluting fuels like biomass over cleaner alternatives such as biogas. We will evaluate how targeted information campaigns, delivered through community meetings and emphasizing social norms and economic benefits, influence these preferences and the actual adoption of cleaner energy. Using a Randomized Controlled Trial (RCT) design, we will assess the impact of exposure to normative and economic information on shifting household energy preferences. Our study will provide policymakers and stakeholders with actionable insights to design effective interventions for sustainable energy use. Households receiving norm-based information will learn about successful biogas adoption by community members and leaders, with an emphasis on social benefits and collective impact. The cost-focused group will receive detailed information on the financial benefits of biogas, including potential long-term savings compared to biomass. By integrating social norms and economic incentives, we aim to offer a holistic approach to promoting the adoption of cleaner household energy sources, particularly in developing countries.

Intervention

In collaboration with the College of Business and Economics at the University of Addis Ababa and the IGC Ethiopia office, the team will develop educational interventions about cleaner energy alternatives to biomass fuel. These interventions will include:

Two ten-minute videos will be produced and shown at community meetings and town halls. One video will focus on the economic advantages of biogas, including cost savings compared to traditional biomass. The other video will emphasize the social shift towards biogas, showcasing peers who have adopted the technology. The videos will be available in English, Amharic, and Afaan Oromo to ensure inclusivity. Both male and female actors will be featured to highlight the role of both genders in environmental sustainability.

The intervention will also leverage text messages to improve comprehension. A series of informational text messages will be sent to reinforce the content of the videos. These messages will be designed for maximum comprehension and will leverage the widespread mobile network coverage in Ethiopia (over 98%).

The overall aim of the intervention is to educate households about the economic benefits of biogas and the use of digestate as fertilizer, while also normalizing the use of biogas through real-life testimonies and engaging content.

Experimental design

Randomization:
The intervention will be conducted in 155 communities or villages from four Woredas (Lume, Adea, Ejere, and Bacho) in the Oromia region of Ethiopia. These areas are characterized by heavy reliance on biomass fuels, dense populations, and poor socioeconomic outcomes typical of places with high biomass usage (Gabisa & Gheewala, 2018; Getie, 2020; Endalew et al., 2022).

The team will work with local government and community leaders to gain buy-in before the randomization process begins. Communities will be randomly assigned into three groups: a control group (one-third of the communities) and two treatment groups (the remaining two-thirds). One treatment group will watch the video focused on cost-savings, while the other will watch the video focused on changing norms. The randomization will be stratified by community size and economic characteristics.

Within each community, 10 households will be selected for participation, resulting in an estimated sample of 1,550 households, similar to other studies (Afridi et al., 2021). However, due to the potential for intra-cluster correlation, where households in the same community might make similar energy choices, the sample size may need to be adjusted to account for this and ensure sufficient statistical power.

Households will be randomly selected using satellite imagery or community-provided lists, depending on logistical ease.

Design details

Randomization method
The randomization will be conducted in an office using a computer. Communities will be randomly assigned into one of three groups: a control group (one-third of the communities) and two treatment groups (the remaining two-thirds). One treatment group will watch a video focused on cost-savings, while the other will watch a video focused on changing social norms. The randomization will be stratified by community size and economic characteristics to ensure balanced groups.
Randomization unit
The randomization will occur at the community level. Communities will be randomly assigned to either receive the interventions (informational videos and text messages) or not. Within these communities, individual households will then be randomly selected to participate in the research.
Clusters
155 - 200 communities or villages
Observations
1,550 - 2,000 households within the communities/villages
Arms
517-667 households control, 517-667 Treatment 1, 517-667 Treatment 2,
Primary outcomes
a). Behavioural game to gauge participants responsiveness to adopting cleaner energy sources – biogas. b) Behaviour towards interest in cleaner energy sources.

Cited by

No citation data available for this registration yet.

Keywords