Encouraging households to enrol into a voluntary demand response program

How to cite

Akesson, Jesper. 2026. "Encouraging households to enrol into a voluntary demand response program." AEA RCT Registry. AEARCTR-0017881. https://doi.org/10.1257/rct.17881-1.0

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

Jesper Akesson*

* Primary investigator

Registry ID
AEARCTR-0017881
DOI
https://doi.org/10.1257/rct.17881-1.0
First registered
Last updated
Status
On going
Trial dates
2026-01-02 – 2026-05-31
Countries
Private
Registry record
http://www.socialscienceregistry.org/trials/17881

Abstract

Countries worldwide are accelerating the transition to renewable energy in pursuit of Net Zero targets. However, variable renewable sources such as solar and wind are inherently inflexible and cannot readily respond to short-term fluctuations in electricity demand. Residential consumers—who account for approximately one quarter of global energy consumption—therefore represent a substantial and underutilised source of demand flexibility. One approach to unlocking this potential is through demand response programmes that incentivise households to shift electricity consumption away from peak periods (e.g., 4:00–7:00 pm on weekdays) in exchange for financial rewards such as bill credits or free electricity hours. This study conducts a large-scale field trial in collaboration with a major UK energy provider. The primary objectives are to evaluate the effectiveness of behaviourally informed email interventions on programme sign-up rates and to estimate the causal impact of programme participation on household electricity consumption during peak and off-peak times.

Intervention

This study evaluates whether behaviourally informed email communications increase enrolment in a demand flexibility programme and how programme participation affects household electricity consumption. This programme incentivises residential customers to shift electricity usage away from weekday peak hours (4pm–7pm, Monday–Friday) by offering free electricity hours on Sundays to customers who opt in to a monthly challenge.

Eligible residential customers are randomly assigned to receive either no trial-specific email communication (control) or one of several behaviourally informed email invitations encouraging sign-up. Treatment emails differ in their behavioural framing, while the underlying programme offer and sign-up process remain constant. The primary outcomes are programme sign-up and subsequent electricity consumption during peak and off-peak periods.

Experimental design

The study consists of two parallel randomised controlled trials conducted among residential customers in Great Britain. One trial targets low energy users and the other targets high energy users without electric vehicles. Within each trial, eligible customers are randomly assigned to a control group or to one of three behaviourally informed email treatment groups. Outcomes include programme sign-up and electricity consumption following the intervention.

Design details

Randomization method
Computer-generated random assignment. Randomisation is conducted separately for each customer cohort.
Randomization unit
household
Clusters
0
Observations
Low energy users: 100,656 customers High energy users (non-EV): 154,652 customers Total: 255,308 customers
Arms
Low energy users: Control: 25,164 Treatment 1 (Baseline email): ~20,000 Treatment 2 (Empathetic messaging): ~20,000 Treatment 3 (Community success stories): ~20,000 High energy users: Control: ~20,000 Treatment 1 (Baseline email): ~20,000 Treatment 2 (Challenge framing): ~20,000 Treatment 3 (Community success stories): ~20,000
Primary outcomes
Programme sign-up Electricity consumption (total, peak, and off-peak)
Secondary outcomes
Email open rates Email click-through rates Bill amount and free electricity hours earned Customer attrition (churn) during the intervention period

Cited by

No citation data available for this registration yet.

Keywords