Behavioral and Economic Impacts of Time-of-Use Tariff in Brazil: Evidence from a Randomized Controlled Trial in the Equatorial Group

How to cite

Meloni, Luis Eduardo Negrão; Lucinda, Claudio; Moita, Rodrigo; Oliveira, Lucas Lyrio de; Pereira, Luan Michel Soares. 2025. "Behavioral and Economic Impacts of Time-of-Use Tariff in Brazil: Evidence from a Randomized Controlled Trial in the Equatorial Group." AEA RCT Registry. AEARCTR-0016313. https://doi.org/10.1257/rct.16313-1.0

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

Luis Eduardo Negrão Meloni* , Claudio Lucinda , Rodrigo Moita , Lucas Lyrio de Oliveira , Luan Michel Soares Pereira

* Primary investigator

Registry ID
AEARCTR-0016313
DOI
https://doi.org/10.1257/rct.16313-1.0
First registered
Last updated
Status
In development
Trial dates
2025-12-01 – 2026-12-01
Countries
Private
JEL codes
D03, D12, D91, L94, Q41, C93
Registry record
http://www.socialscienceregistry.org/trials/16313

Abstract

This study evaluates how different types of electricity tariff plans affect consumer behavior in low-income areas of Northern and Northeastern Brazil. In partnership with a major electricity utility (Equatorial Energia), households were randomly assigned to receive offers of alternative tariff options, including prepaid electricity, time-of-use pricing, and dynamic pricing. These options aim to increase payment regularity and reduce electricity theft by giving consumers more flexibility and control over their energy spending. The study also tests behavioral nudges like reminders and personalized consumption feedback. Outcomes will be measured using administrative billing and consumption data from the utility, and may be complemented by follow-up surveys.

Intervention

This study evaluates the impact of an innovative electricity tariff modality implemented by the Equatorial Group in two cities in Northeastern Brazil: Maragogi (AL) and Barreirinhas (MA). The initiative is a social experiment based on the application of the Tarifa Melhor Hora (Time-of-Use – ToU), which establishes fixed, differentiated prices for peak and off-peak hours, communicated to consumers in advance. The main goal of the ToU tariff is to encourage the shift of electricity consumption to periods of lower demand, thereby reducing system overload and promoting more efficient energy use.
The tariff will be applied to a sample of Consumer Units (UCs) previously selected based on specific eligibility criteria related to consumption profiles. In each municipality, eligible low-voltage consumers, including households and small commercial establishments, will be randomly assigned to treatment or control groups. Randomization will be conducted at the UC level, stratified by customer class (residential or commercial) and historical energy consumption quintiles.
Participants in the treatment group will receive the new tariff, accompanied by targeted communications sent by the utility provider. These communications explain how the tariff works and guide consumers on how to adjust their habits to achieve financial benefits. The content is personalized according to the consumer class and location and may include energy-saving tips, illustrative bill examples, and visual aids. On the other hand, consumers in the control group will continue under the conventional tariff, with no changes to prices or communications received.
Before the intervention began, smart meters were installed in all selected units. These devices enable high-frequency monitoring of electricity consumption, allowing for a more precise analysis of the effects of the new tariff. The intervention will last a minimum of 12 months, with additional monitoring after this period.

Experimental design

This study employs a Randomized Controlled Trial (RCT) to assess the behavioral and economic impacts of a Time-of-Use (ToU) electricity tariff, implemented by the Equatorial Group in two tourist municipalities in Northeastern Brazil: Maragogi (AL) and Barreirinhas (MA).

In each city, a pool of eligible low-voltage consumers—both residential and small commercial—was selected based on historical electricity consumption and administrative criteria (e.g., exclusion of units with distributed generation, recipients of the social tariff, frequent disconnections due to nonpayment, etc.).

The experimental design consists of:
1. Separate randomizations by city, generating two independent experiments (Maragogi and Barreirinhas);
2. Stratification by consumption quintiles and customer class;
3. Random assignment to treatment and control groups using a 1.3:1 treatment-to-control ratio;
4. Unit of randomization: electricity consumption unit (unidade consumidora, or UC);
5. Randomization procedure: implemented via R and Python scripts;
6. Oversampling strategy: for each sampling unit, approximately four UCs were drawn. From these, the utility selected one based on feasibility criteria (same class, municipality, and consumption quintile) to ensure a higher success rate of meter installation.

Consumers in the control group remained on the conventional tariff, while those in the treatment group were assigned the ToU tariff, accompanied by targeted communication strategies encouraging demand response. High-frequency metering was installed for all selected units, enabling precise estimation of the intervention’s effects on energy use and consumer behavior.

Design details

Randomization method
Randomization will be performed using scripts developed in R and Python.
Randomization unit
The unit of randomization is the electricity consumption unit (UC), which corresponds to an individual metering point, typically a household or small commercial establishment. Randomization will be conducted separately for ToU and for each of the two cities under concession by the Equatorial Group: Maragogi (AL) and Barreirinhas (MA). Each UC will be randomly assigned to either the treatment or control group, stratified by consumer class (residential and commercial) and historical energy consumption quintiles.
Clusters
Not applicable. The treatment is individually randomized at the electricity consumption unit (UC) level.
Observations
3,512 electricity consumption units (UCs), distributed by city according to smart-meter availability: • Maragogi (AL): 1,756 UCs • Barreirinhas (MA): 1,756 UCs
Arms
Each city follows a 1.3:1 treatment-to-control allocation. Approximate counts: • Maragogi (AL): ~993 treatment / ~763 control (total 1,756) • Barreirinhas (MA): ~993 treatment / ~763 control (total 1,756) Notes: Final counts may vary slightly due to installation success and administrative exclusions.
Primary outcomes
Peak-period consumption (kWh): Total energy consumed during the peak hours defined by the utility provider. This indicator captures the behavioral response to price signals and the effectiveness of shifting consumption.
Secondary outcomes
1) Bill amounts and payment regularity: This includes the total bill value (in Brazilian reais, R$) and the number of payments made on time. These indicators capture the financial impact of the intervention and the potential risk of default.
Sponsors
Name: Grupo Equatorial Energia Location: Brazil Url: https://www.equatorialenergia.com.br Name: Agência Nacional de Energia Elétrica (ANEEL) – Programa de Pesquisa e Desenvolvimento (P&D) Location: Brazil Url: https://www.aneel.gov.br
Partners
Name: Agência Nacional de Energia Elétrica (ANEEL) – Programa de Pesquisa e Desenvolvimento (P&D) Partner_type: government Url: https://www.aneel.gov.br Name: i4 Economic Regulation Partner_type: private_company Url: https://www.i4regulation.com

Cited by

No citation data available for this registration yet.

Keywords