Incentivizing tenants to reduce heating energy consumption: A field study

How to cite

Kandul, Serhiy; Lanz, Bruno. 2019. "Incentivizing tenants to reduce heating energy consumption: A field study." AEA RCT Registry. AEARCTR-0003856. https://doi.org/10.1257/rct.3856-2.0

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

Serhiy Kandul* , Bruno Lanz

* Primary investigator

Registry ID
AEARCTR-0003856
DOI
https://doi.org/10.1257/rct.3856-2.0
First registered
Last updated
Status
Completed
Trial dates
2018-12-01 – 2019-04-01
Countries
Switzerland ()
JEL codes
Q41, C93, D91
Registry record
http://www.socialscienceregistry.org/trials/3856

Abstract

This field experiment quantifies the impact of a social comparison feedback intervention to promote responsible heating energy consumption among a group of tenants. Using data for indoor temperature collected by a set of smart thermostats, participating households are informed about how their average indoor temperature in December 2018 compares to average temperature of their peers (i.e. observationally similar households observed during the same period of time).

Aside from a control group that does not receive the information, we randomly allocate households across three conditions. First, in the baseline treatment, the social comparison information is provided without context, which is the usual approach in alternative field studies (e.g. Allcott and Rogers, AER 2014). In the second, third and fourth conditions, social comparison information is accompanied by additional information about the landlord behavior.

While the baseline condition provides evidence about the impact of social comparison in this specific context, the two additional treatments will enable us to document the role of information about supply-side behavior in the context of social information feedback interventions.

Experimental design

Participants for this experiment are the tenants living in the buildings managed by our partner, the real-estate agency (the landlord thereafter). Participating tenants are contacted via regular mail, one letter at the end of January 2019 with information about their indoor temperature levels in the previous month (December 2018). Irrespective of the experimental condition, all the letters contain information about the indoor temperature levels and the energy saving tips. The treatment-specific information in the letters vary according to the experimental condition (see section "Interventions").

Importantly, participants are not offered any financial compensation. The intervention is purely informational, and is done as a collaboration with the landlord who wishes to find ways to communicate with their tenants.

Based on the previous work on social comparison feedback, we expect that tenants who learn that their temperature is above the average would tend to adjust their behavior towards the "norm." In order to prevent good-performing tenants (the ones with the below-the-average indoor temperatures) to increase their indoor temperature, we include social approval in forms of smiley faces.

All the letters are accompanied with a small questionnaire where participants can react to the letter, list heating appliances that are not provided by the landlord, as well as an opt-out form where participants could explicitly deny the usage of their temperature data for the purpose of the study.

Design details

Randomization method
Random number generator randomized individual households into each of the four treatment conditions. Within each household, one randomly selected tenants is chosen as a recipient of the letter. The control condition was selected beforehand based on the building level data.
Randomization unit
Individual household (between the treatments), group level randomization (treated vs. control)
Clusters
30 buildings in the treatment conditions, and 15 buildings in the control.
Observations
We contacted 623 households in the treated condition (156 in three of the treatments and 155 in one of the treatments), and 232 households in the control.
Arms
156 households in each of the three treatment arms, 155 in one of the treatment arm (30 buildings in total) 232 households in the control (15 buildings in total).
Primary outcomes
The key outcome variable is the change in average indoor temperature/humidity levels over time, comparing tenants in alternative conditions.
Secondary outcomes
We are interested in whether tenants contact us to complain / withdraw from the study, and see if it differs across treatments, and also if the choice of the person contacted (e.g. male vs. female) matters to induce energy saving efforts.

Cited by

No citation data available for this registration yet.

Keywords