The Potential of Behavioural Interventions for Optimising Energy Use at Home

Whether we are boiling water to make tea, activating a home appliance or relaxing under a hot shower, our daily behaviours and choices at home are important drivers of energy consumption. In 2018, energy consumption from the residential sector represented 20% of the total energy demand in IEA countries.

Largest end-uses of energy by sector in selected IEA countries, 2018

Open

Individuals and households can adopt a variety of measures to optimise their energy consumption. Changes in our everyday habits can help to reduce home energy bills, lower carbon emissions and ease pressure on the power grid. Modest behaviour adustments can, over time, yield substantial energy savings on their own. Residential energy demand can be further reduced through “structural” inevestments, including upgrades to more energy-efficient appliances and retrofits to existing household equipment. Appropriate policy interventions and programmes can be designed to promote sustainable changes in behaviour and encourage investments in structural improvements.

This article focuses on the potential for enhancing energy efficiency with policies and programmes designed to educate consumers and encourage them to alter their daily habits – without resorting to large-scale structural improvements. This focus is motivated by the observation that interventions aimed at promoting behavioural change are often cheaper to implement relative to policies that seek to encourage investment. Such measures also tend to be relatively quick to design and implement. 

What are behavioural interventions and how can behavioural insights enhance the impact of energy efficiency policy?

Behavioural interventions are policies and programmes designed to incorporate the insights of scientists who study human behaviour. The aim of these interventions is to trigger socially desirable behaviours – either by removing barriers to such behaviours, or by creating disincentives to socially damaging ones. The objectives and advantages of behaviourally informed policy-making in the energy realm is discussed in depth in a recent IEA and UsersTCP report.

Behavioural policy interventions differ from traditional approaches that seek to enhance energy efficiency through economic incentives (such as subsidies for efficient purchases or time-of-use energy pricing), information provision (energy efficiency labels) or regulatory requirements (setting minimum energy performance standards). Such traditional measures assume that users will easily understand the benefits of a policy or programme and react rationally. But the mechanisms that affect the way individuals interpret and act upon information are often complex, and ignoring that complexity can limit policy effectiveness.

Integrating behavioural insights into the policy-making process requires assessing the various biases that affect the way individuals perceive, process, understand or ignore information, prompts and incentives. Meanwhile, designing policies that support changes in consumer habits and that encourage investments also involves identifying possible barriers and finding appropriate solutions to lower or remove them. Sometimes those constraints can be financial, but often they are behavioural: the perceived “hassle factor,” for example, or simple inertia. By expanding their toolkit to include appropriate behavioural levers, policy makers can boost the effectiveness of energy efficiency interventions.  

Which behavioural mechanisms prevent flexible and efficient energy use, and what are the best levers to address them?

Our energy consumption at home is shaped by ingrained habits that are hard to change, due to factors like inertia or a subconscious preference for the status quo. Such biases magnify the perceived effort associated with making alternative choices. Personal preferences and social norms also play a role in shaping our routines and behaviours. Understanding how our daily usage of a basic appliance affects our monthly bills can sometimes be complicated, even for those who are environmentally motivated. This lack of awareness can hinder energy saving efforts.

Successful behavioural interventions to bolster energy efficiency have facilitated conservation by using techniques such as feedback mechanisms, which track households’ energy consumption patterns, or targeted prompts, which alert consumers to pay particular attention to their energy usage at peak times. Such interventions make use of multiple behavioural levers (for more in-depth discussions, see IEA and Users TCP (2020) and OECD (2017)):

  • Simplification and framing of information is crucial to making communications user-friendly and clear. This is valid for any policy intervention or programme and applies to a broad range of information, including: home energy reports, monthly bills, web portals, appliance efficiency labels and energy audit reports.
  • Feedback mechanisms show consumers the evolution of their energy consumption patterns throughout the day and across seasons. Their purpose is to raise awareness about how daily use of appliances, heating and air conditioning affects energy expenses. Feedback can be provided in real time, through in-home displays, mobile applications or web portals fed with data from smart metering systems. It can also be transmitted less frequently, via online or mailed home energy reports.
  • Leveraging social norms and comparisons can illustrate to consumers how their consumption compares to that of their peers – comparable households in the same area. This can prompt positive competition effects, motivating users to reduce their excessive energy consumption to bring it in line with the average.
  • Goal setting, commitment devices and reward schemes are often jointly exploited in the design of so-called demand response programmes, which invite users to reduce their electricity consumption during periods of high power prices. Utility companies also incorporate them into broader customer loyalty programmes. Such schemes are usually aimed at prompting energy saving efforts in short bursts (during peak usage times or extreme weather events) but can be used to encourage conservation over longer time horizons. Rewards can be monetary (prizes or discounts) or in-kind (free services or product give-aways).
  • Changes to product design and to the physical environment which leverage smart default options can simplify efficient appliance use for consumers and facilitate energy-efficient choices. For example: revising regulations that establish the default set-point temperatures of heating and cooling appliances.

Policy interventions and programmes can simultaneously exploit multiple behavioural levers. For example, energy-saving competitions and games can leverage social comparisons among consumers of the same energy utility, and engage participants through real-time feedback as well as set milestones and targets that trigger rewards. The next section summarises the demonstrated impacts that behavioural policy interventions and programmes have on energy savings.

What do we know about the impacts of behavioural interventions?

Assessing the effectiveness of any type of policy intervention or utility-run programme requires credible data on energy consumption as well as rigorous evaluation techniques. Assessments should compare usage before and after the intervention, and data from “treated” groups need to be measured against a control group.

Previous studies have aimed to synthesise the evidence that interventions affect user behaviour. These have either focused on a defined geographical area – North America in the cases of Sussman and Chikumbo (2016), and recent studies by the Boston-based Consortium for Energy Efficiency (2017) and (2019) – or on specific policy levers such as feedback mechanisms, as in the case of Karlin, Zinger, and Ford (2015).

Recent studies carried out in the United States and reviewed by Sussman and Chikumbo (2016), summarised in the chart below, demonstrate that interventions based on different behavioural levers also affect consumer behaviour differently. This can result in varying levels of energy savings. The impact of competitions and games, for example, is generally detected immediately after their implementation. Alternatively, regular feedback programmes like home energy reports tend to deliver continuous and relatively stable savings – provided the programme’s duration is sufficient to allow consumers to adjust their consumption patterns.

Energy savings across behavioural interventions implemented in the United States

Open

Several factors help to explain differing assessments of behavioural impact:

  • Evaluation approaches such as whether or not a control group is being compared to the group treated with the intervention itself, and what units of measurement (households vs buildings) are being used.
  • Characteristics of the recipients of the interventions, notably the type of home they live in, whether they are homeowners or tenants, and their energy consumption level before the intervention, which is itself driven by numerous socio-economic characteristics.
  • Characteristics of the intervention, including its duration and the point in time at which the evaluation is performed.

While being mindful of these factors, we can still draw a number of useful conclusions from these previous analyses:

  • Even relatively small levels of energy savings per participant can compound to yield high aggregate reductions in energy consumption in absolute terms. The precise source of the energy savings may vary according to programmes. Home energy reports, for example, have been shown to deliver savings as consumers reduce usage of air conditioners and heaters or switch off idle appliances and lights.
  • Opt-out interventions, which involve automatic programme enrollment, can reach a larger number of participants than opt-in programmes that require consumers to explicitly signal their interest in joining. This illustrates how a small change in programme design can make a large difference. Automatic mailings of home energy reports are an example of an opt-out intervention, while a competition or game requires users to opt in. Consumers who choose to participate in a conservation programme are generally already motivated to reduce their energy consumption. As a result, competitions and games tend to achieve higher maximum energy savings – in some cases as much as 14%.
  • Behavioural interventions appear to deliver larger cuts to electricity consumption than to natural gas consumption. Home energy reports are estimated to reduce household electricity consumption by as much as 2.2% and natural gas consumption by up to 1.6%. The savings that can be achieved through competitions and games, meanwhile, are around 14% for electricity and 10% for gas. In residential buildings, natural gas is primarily used for ambient heating and hot water, while electricity has a wider range of uses including, air conditioning, lighting and appliances. This could indicate that preferences and habits relative to heating – such as temperature settings and time of use – are more difficult to modify compared to other energy-consuming behaviours in the home.

The following section illustrates these insights with case studies of how different behavioural levers have been applied to concrete policies and programmes to prompt energy efficient behaviour change. For more examples and additional details, see UsersTCP and IEA (2020).


Successful behavioural interventions: Recent examples

Regular feedback through home energy reports

Home energy reports are briefs that utilities mail out to consumers on a regular basis, comparing their energy consumption to the neighbourhood average as well as to the most efficient neighbours with a similar household size. Consumption is reported both for the most recent period and over time, outlining the overall savings achieved with respect to the average, or the potential savings that could be reaped by aligning consumption to that of most efficient neighbours – thereby leveraging comparisons with social norms. Home energy reports also compare energy consumption with the same period in the previous year, and provide energy-saving tips in a visual format that is more user-friendly and transparent than standard energy bills.

The first home energy reports were popularised in the United States by Opower (now part of Oracle) more than a decade ago. Since then, these non-price interventions have successfully prompted energy savings in different geographical, cultural and energy market contexts. It is worth noting that, in the United States, achieving a 2% average reduction in energy savings would call for an 11-20% short-term increase in the electricity price.

Impacts of home energy reports across selected countries

Country

Year

Households involved in study

Energy savings

Notes

United States

2009-2010

600 000, including both treatment and control groups

2%

Average user savings relative to control group not receiving reports. Savings ranged between 0.3% (bottom decile) and 6.3% (top decile)

Malaysia

2015

450 000 receiving reports

1-3%

Average user savings relative to control group. In absolute terms, energy savings amounted to about 50 000 MWh.

Japan

2018

300 000 receiving reports

2%

Average energy savings relative to control group

Sources: Alcott (2011), Social norms and energy conservation; Oracle (2019), “Consumers Reach Nearly 23 TWh of Energy Savings with Oracle Utilities Opower” (press release); and Sachar, S. et al. (2019), White Paper on Behavioural Energy Efficiency Potential for India.

A comparison of the net savings of 28 US-based home energy report programmes indicates that that such interventions tend to generate increasing energy savings over time. This reflects the learning patterns of consumers as they become more receptive to the reports and start to adopt energy-saving habits. The level of savings tends to stabilise roughly two years after receiving the first home energy report, indicating the importance of continued feedback to sustaining energy-saving efforts.

Net energy savings from US-based home energy reports programmes

Open

A new generation of home energy reports, such as those developed by Uplight, a Colorado-based utility, further exploit the granularity of the data provided by smart meters to gain insights into energy consumption across the day. They also make use of other communication platforms, such as mobile applications or web portals, which allow consumers to navigate the information more dynamically. These tools can also enable tailored demand-response requests.

Real-time feedback through digital devices

While feedback about previous energy consumption patterns, both in absolute terms and compared to peers, encourages energy conservation, the overall impact on behavior varies depending on a number of design features. Adusting feedback frequency, for example does not tend to induce a significant variation in energy savings, whereas a change in the feedback medium does. More specifically, a meta-analysis of feedback programmes found energy savings to be lowest when feedback was delivered through detailed bills or home energy reports. Savings were marginally better when feedback was communicated either through an in-home display, paper visuals (e.g. door hangers), and highest when transmitted via computer (e.g. through softwares or websites). The differences observed between feedback categories did not appear to be statistically significant, however.  

The large-scale deployment of smart meters is facilitating the implementation of real-time feedback mechanisms through connected digital devices. The United Kingdom, for example, has been testing the effects of pairing smart meters with feedback media such as in-home displays and mobile applications. A 2015 impact assessment by the Smart Metering Early Learning Project observed that consumers with smart meters and in-home displays used 1.5% less natural gas and 2.2% less electricity in 2011 compared with those equipped with conventional meters.

More recently, a comparison of the impact of in-home display feedback versus mobile phone applications indicated that the former resulted in marginally greater energy savings. Although both displays and apps are designed to provide essentially the same information, their unique functionalities and designs can be exploited to cater to different sets of consumers. For example, mobile apps can incorporate push notifications to communicate a consumer’s energy use. Similarly, users who are subject to time-of-use rates can also be alerted to variations in energy prices.

Customer engagement and demand response with goal setting, prompts and rewards

Utilities are already exploiting competitions and games to engage with their customers, prompting users to set goals for reducing their own energy consumption in exchange for small rewards. Such programmes can be tailored for different time horizons. Longer-term programmes have been used to steer radical transformation of energy habits over time. There is also evidence that such measures can be used to achieve more targeted, short-term goals.  

For example, BC Hydro, a Canadian electricity utility, exploits multiple behavioural incentives to engage consumers in its voluntary energy saving programme, Team Power Smart, which challenges participants to reduce their energy consumption by 10% over the course of one year. The programme offers financial and in-kind rewards (including special offers in partner stores, as well as participation in special events). It also leverages social norms by encouraging participants to share their success stories as a way of maintaining motivation. Regular prompts and seasonal challenges are also used to sustain momentum. Over four and a half years, the programme has attracted more than 91 000 households and reduced their combined electricity consumption by 25.6 GWh relative to a comparable control group. This type of programme demonstrates that frequent consumer engagement that combines regular feedback with other behavioural levers results in greater energy savings.

In Australia, the Jemena Electricity Network set up demand-response challenges targeting energy demand at peak time during extreme weather events. Consumers there volunteered to participate in the scheme in exchange for small financial rewards. The challenges led to a 26% to 42% reduction in energy use over designated three-hour intervals. Jemena also experimented with community rewards – inviting consumers to donate the rewards gained by participating in the challenge to a local charity of their choice – which resulted in slightly lower energy savings.

Smart defaults in regulation and programme design

Redesigning policies, programmes or products so that the default usage option or setting is also the cleanest and most efficient one, is another way to facilitate sustainable consumption decisions and patterns. This method takes advantage of consumer inertia, while still enabling users to select a different option if they prefer.

One example of a smart default programme is India’s recently approved regulations for air conditioners. Appliance manufacturers are required to establish the default temperature settings for their devices at 24 °C. The goal is to optimise comfort and energy savings, while still allowing consumers to adjust the operating temperature according to their preferences.

Smart defaults can also increase the reach and potential of demand-response programmes by making registration automatic. Instead of inviting users to participate, Powershop, another Australia-based utility, enrolls its consumers in its demand-response challenges by default, sending them spontaneous prompts to reduce their consumption at critical times. This type of opt-out approach managed to reach eight times the number of participants, delivering three times the energy savings, compared to opt-in prompts that were tested within the same programme. Potential usage reductions were estimated at 2-3 MWh per demand-response event. Powershop’s experience shows that “traditional” tools like financial incentives for voluntary efforts can be still more effective when they are designed in a behaviourally-informed way.


What are the energy savings and emissions reductions that can be achieved through behavioural interventions?

As set out in the IEA World Energy Outlook 2020, a combination of end-use efficiency gains, electrification and behaviour change will be required in order to meet the target of net‑zero emissions by 2050. Given the urgency of climate policy objectives, the speed with which greenhouse gas emissions can be reduced is just as important as the cost of emissions abatement.

It is therefore all the more important to exploit all approaches to energy-savings to their fullest potential. As illustrated by the case studies cited here, behavioural interventions can encourage flexible and efficient energy use. Such measures can be used in tandem with other policy approaches, including financial incentives to improve structural energy efficiency, or energy price signals that internalise the social costs of greenhouse gas emissions and other pollutants from power generation.

In the United States, potential energy savings from behavioural interventions in the residential sector are estimated at between 16% and 20% of home energy demand. In this region, the greatest potential savings come from regulatory adjustments to default temperatures for heating and cooling, as well as from hot water use. These types of behaviour changes can be facilitated through feedback mechanisms and smart devices.

India estimates the potential for energy savings through behavioural interventions to be in the range of 3.4 to 10.2 TWh per year by 2030, which translates to about 1.8 to 5.3 Mt CO2 of avoided greenhouse gas emissions per year. While reaching full potential requires planning for the next ten years, deploying well-known programmes such as home energy reports to a subset of Indian states today could deliver energy savings of 720 to 2 140 GWh per year within two years.

European countries have also found the potential savings to be sizeable. Ireland, for example, estimates that modest changes in behavior, such as adjustments to indoor temperature settings, would lead to significant reductions in energy use: 2.4 TWh per year in the case of residential buildings alone, and 6.5 TWh overall when commercial and public buildings are included. This would enable the country to lower its total energy consumption by about 5%.

The ease and speed with which behavioural interventions can be implemented mean that countries are able to meet their greenhouse gas emissions targets sooner, and at lower cost, thereby accelerating their energy transition and reducing the overall impact on the climate. Drawing on data from programmes that have been implemented in the United States, it is possible to compare the efficacy of behavioural versus “structural” energy-efficiency programmes according to their cost per dollar of climate damage avoided, as well as the average annual number of participants involved in each programme. (Structural programmes typically involve measures such as comprehensive assessments of home energy performance as well as incentives for retrofits or the installation of more efficient heating and cooling systems. Behavioural programmes use levers such as home energy reports, high bill alerts or home energy audits.)

Such comparisons show that behavioural interventions tend to reach a considerably larger consumer base, for about one-fifth of the cost of structural efficiency interventions. Because of their differing costs, outreach profiles and time horizons for emissions reductions, these approaches are highly complementary and contribute significantly toward a clean-energy transition.

Structural and behavioural energy efficiency programmes in the United States

 

Structural Programmes

Behavioural Programmes

 

Cost per USD of avoided climate-change damage

Illinois

2.04

0.43

Maryland

9.75

2.03

Massachusetts

22.82

5.42

 

Average annual programme participation

Illinois

30 509

1 700 000

Maryland

6 100

902 900

Massachusetts

25 652

705 000

Source: Hibbard et al. (2020), Utility energy efficiency program performance from a climate change perspective. A comparison of structural and behavioral programs; Analysis Group report for Oracle.

Normalising behavioural interventions can optimise energy consumption at home and beyond

Insights from the behavioural sciences have alreasy led to a number of successful and innovative policy interventions and programmes, including home energy reports and sustainable default features for appliances. At the same time, they can enhance the effects of more traditional approaches to promoting energy efficiency by informing the design and implementation of measures ranging from product efficiency labels to dynamic energy-pricing schemes.

Normalising behavior-changing interventions and fully integrating them in the efficiency policy toolkit requires a supportive regulatory environment. Long-term energy efficiency objectives, accompanied by energy efficiency obligations for utilities, can provide a framework for unlocking the potential of behavioural interventions. Rigorous assessments of the impacts of behavioural policies and programmes, facilitated by smart metering and digital devices, can enable adjustments to these efforts when and where they are needed. They can also make it possible tailor programmes to suit specific consumer groups or building sectors.

While this article has discussed the potential of behavioural interventions to reduce household energy consumption without resorting to building or appliance upgrades, behavioural insights can also be applied to policies that seek to increase investments in structural measures such as home retrofits or energy-efficient appliances. The two are mutually reinforcing: structural adaptations, through more efficient building construction, transportation infrastructure and urban design investments, can facilitate the adoption of sustainable behaviours.

Further, behavioural insights can also inform the design of policy incentives to support sustainable behaviour change in other areas critical for the energy transition, such as mobility and transport: behavioural interventions can encourage eco-driving practices and a modal shift towards zero-carbon options such as walking, cycling and public transport, as well as support the electrification of private vehicles.

Integrating behavioural interventions in the policy toolkit, alongside market-based instruments such as energy taxes and regulatory interventions such as product efficiency standards, can support the achievement of ambitious energy efficiency objectives necessary for the clean energy transition. While supply-side changes in the energy market – from electrification to decarbonisation of power generation – are fundamental to achieve net‑zero emissions by 2050, behaviour change can contribute to achieving demand-side objectives, together with structural efficiency improvements and with large-scale changes needed across all areas of energy use, from industrial operations to transport demand. Ultimately, a solid understanding of behavioural drivers of energy consumption and barriers to sustainable energy use is necessary to design people-centred energy policies for the transition