Preliminary evidence on the effects of a behaviourally informed intervention integrated in an initial subset of 12 apartments seems to highlight the role that household composition plays in ensuring the effectiveness of behaviour-change interventions. The intervention in questions is informed by behavioural economics and involves providing social information to tenants regarding the average energy consumption of their neighbours. This approach, known in behavioural economics as a “social comparison intervention”, has been integrated in several energy conservation programmes in the private residential sector, particularly in the United States. However, the evidence of their effectiveness in social housing and in a European context remains limited. Furthermore, when these behaviour-change interventions are applied in the context of retrofitting activities, such as is the case with the Sinfonia project, they have the potential to contribute to the avoidance of the well-known “rebound effect”. This is where tenants adjust their behaviour as a response to increased efficiency in the building stock and can in some cases lead to the “take-back” of some of the expected energy savings. Integrating behaviour change interventions with technical innovations is therefore crucial to align technical and behavioural energy efficiency, ensuring best outcomes for end-users in terms of energy savings.
The social information was shared to tenants through smart meters displays installed in the context of the project to consenting households. These displays provided information of household´s own energy consumption with high granularity (every 5 minutes) and helped them access this information aggregated at different levels, daily, weekly, and hourly. The displays also provided information on household comfort levels, including temperature, humidity, and air quality. Besides this, the displays also provided tenants with tips to save energy and increase household comfort-levels.
To test the impact of providing social information, an experimental approach was taken. That is, households were randomly assigned to either a control or treatment group, at the district-level. The control group had access to information on their own energy consumption, while the treatment group was also provided with information on the consumption of a comparable group of neighbours. This reference group was chosen to be comparable based on household composition. Due to the random assignment, average behaviour between the two groups should be representative of any impacts that the provision of social information has on consumption behaviour.
The research is ongoing but results on a limited sample of 12 consenting apartments whose energy consumption was monitored for three months seems to show considerable variability in response to the provision of social information. Interestingly, household composition factors, such as gender and age of members, seems to be responsible for moderating some of the effects of the intervention. As more data from the Sinfonia project continues to emerge, it will be interesting to see if any significant differences emerge in energy behaviours between the two groups, to evaluate the effect of the social comparison intervention in this post-retrofitting, social housing context. For the meantime, the preliminary message for policymakers and practitioners working on retrofitting projects in social housing is to be mindful of contextual influence when designing behaviour-change interventions to integrate into their projects.
For more information, check out the publication.