This is the final post in my series on the energy performance of buildings. The first two (here and here) dealt with issues relating to the fabric of buildings, while in this post I will look at the behaviour of occupants and how their choices impact building energy use.
Consumers influence building energy use in two ways: their passive energy use in terms of heating and lighting, and their active energy use in terms of appliances. The Government is keen to see consumers reducing their energy use – this was one of the key drivers behind the troubled smart meters programme, but its own cost benefit analyses now suggest that the programme will have little impact on consumption choices.
The role of consumers in energy markets is changing, or at least, the role of some consumers is changing. These consumers are increasingly engaged, acquiring and optimising a range of energy assets such as micro generation, domestic storage, smart home technologies and the new flexible tariffs that are emerging. The Government and many within the industry hope that these consumers are leading the way for a new model of consumer, or rather, “prosumer” (consumers that also produce energy) behaviour.
But at the same time, the Government laments the fact that half of all consumers has never switched supplier despite knowing this could result in lower energy bills. A large number of consumers is stubbornly un-engaged, so how likely is it that these consumers will morph into the prosumers of tomorrow?
Smart meters are unlikely to deliver meaningful change
There is an interesting discrepancy between the expected reductions in energy consumption due to the use of smart meters and the extent to which consumers report looking at their in-home displays (“IHDs”). Studies in 2018 by Citizens Advice and the Government both indicated that around two thirds to three quarters of consumers checked their IHDs at least once a week, but the most recent cost-benefit analysis for the smart meters programme downgraded the expected benefits again.
Indeed, a 2019 survey by Electrical Contractors’ Association the found that on 7% of households reported saving money with their smart meter and 9% reported bills had gone up!
It would be interesting to see the 2018 surveys be repeated to see whether IHD use falls over time once the novelty wears off, or indeed whether “looking at” the IHD was in any way correlated with consumption decisions. A 2020 study carried out at Keele University found that IHDs were of limited use in helping consumers to make choices about reducing consumption. Among the reasons for this were lack of context (data in kWh rather than £), lack of timeliness (data only updating after 12 hours when the previous prepayment meters displayed available credit in real time) and a lack of disaggregated data at the appliance level.
The study participants were also critical of energy saving information provided by their suppliers saying it was too generic to be useful and lacked context, although they did report increased awareness of switching off appliances when not in use. The researchers also identified a focus among participants on turning lights off, suggesting that this was one of the most visible aspects of energy consumption.
There was general cynicism as to the purpose of smart meters with the participants suggesting they were of use primarily to suppliers, and that environmentally conscious consumers would probably already minimise consumption without the need for a smart meter.
“Despite the significant investment in infrastructure and installation of devices, there has been little change in consumer energy behaviours…. The consensus from previous research and these findings is that consumers are incentive-driven and that, currently, there is a lack of adequate incentives to prompt energy reduction,” – Fredericks et al, Keele Univeristy
The participants were more receptive to phone-based apps than the IHD, but that may simply be a reflection of the study demographics, since the participants were all postgraduate students. This does however provide an insight into how more digitally-literate consumers might be engaged, and indicate that the IHDs are of limited use in helping consumers to visualise their consumption in a way that has a meaningful impact.
The issue of whether meaningful savings can be achieved is also worth exploring. While removing wasteful habits like leaving lights on and fridge doors open can be improved (with or without smart meter data), consumers may feel they knowing or being able to visualise and contextualise their consumption would not motivate them to reduce their consumption.
As one Keele study participant pointed out, consumers are unlikely to chose their evening meal based on the energy required to prepare it, and this may apply to a large proportion of what might be termed “discretionary” consumption (ie beyond what is needed to heat the property, keep food refrigerated etc). Streaming box-sets on Netflix uses more electricity at the point of use than reading a book but that does not mean that people will stop streaming and start visiting libraries instead.
This all suggests that major lifestyle changes may be required in order to significantly reduce household energy use.
Potential for energy load shifting to improve building energy performance
Load-shifting, or the act of moving energy intensive activities to cheaper times of day is becoming increasingly common among large industrial and commercial consumers, but it is much rarer in the domestic context. Although radio activated meters have historically been popular in the UK, facilitating cheaper overnight tariffs such as Economy 7 (which provided off-peak rates between midnight and 7am) these were mainly used so that households could charge storage heaters and heat water by immersion during the cheaper night-time period. They fell in popularity as gas became a cheaper alternative to electric space and water heating, and storage heaters were found to be inefficient and complicated to use.
But now there are hopes that time-of-use tariffs will encourage consumers to shift domestic load patterns allowing network operators to make better use of their assets, something that will be important as electricity demand increases in response to electrification of transport and heating. The questions are whether consumers will be willing to do this and if so whether it will make very much difference.
On the first question, there are some loads that can obviously be shifted outside of peak demand hours such as electric car charging. Other loads such as cooking dinner are less likely to be moveable, and some may be undesirable from a safety perspective such as washing machines and tumble driers whose use should be supervised due to the risk of fire.
Studies of consumer attitudes have found that a strong tendency to classify tasks as daily or not daily, with the not daily tasks being more moveable. But there was resistance to moving tasks that were seen as short, such as showering and making hot drinks (although other studies have illustrated that showers are typically not short, consumers tend to view them as such), and there were social concerns such as the prospect of disturbing neighbours or young children by engaging in noisy tasks overnight or early in the mornings, particularly at weekends.
Emphasis was also placed on the comparative behaviour of others with participants tending to believe their own consumption was lower than average and less in need of change.
“Washing machine: again see this is the tricky one because I live in a terrace so if I knew that it was going to cost me more during the day I wouldn’t have any choice because the walls are paper thin,” – study participant
Interestingly the same study found that participants resisted moving activities to the 9pm time slot when the tariff option was a more restrictive two-level option. When a more flexible three-level choice was offered, participants were willing to move the same activities to the same slot, suggesting the greater the flexibility on offer, the more likely the participants were to behave flexibly.
Impact of lifestyle on energy consumption
Social and technological changes in the 20th century saw large increases in heating demand…
Studies have shown that despite an increase in household energy since the 1970s, heating demand has actually increased. Part of the reasons for this originate earlier in the 20th century when a significant amount of new house-building took place to replace slums and war-damaged housing. These new houses were designed differently from the traditional 2-up, 2-down houses of the past, and this fascinating study outlines how different social expectations and desires informed planning decisions.
Traditional 2-up, 2-down houses had two bedrooms upstairs and a kitchen/living room and scullery downstairs (or a kitchen and a parlour). The scullery was the “wet room” where activities such as the “wet” parts of food preparation, washing up, laundry, and personal washing took place. One of the first changes was to move cooking activities from the living room into the scullery, and separate indoor bathrooms were created, since “the house-wife does not want members of the family washing in the scullery when she is busy preparing breakfast”.
A need for more rooms was identified. Initially it was considered undesirable for eating and cooking to take place in the same room, so kitchens were deliberately made too small to accommodate a dining area. More bedrooms were added to ensure adequate separation of the sexes, and more room was required for “activities demanding privacy and quiet”, such as listening to records, watching television, do-it-yourself and homework. The idea was that bedrooms would no longer be used only for sleeping as had traditionally been the case.
This meant that bedrooms would need to be adequately heated – although all rooms had the capacity to be heated, fires tended not to be lit in bedrooms unless someone was sick in bed. These requirements resulted in major changes to building design, with “pipes, rather than people carrying water, warmth and fuel, activities requiring heat and (warm) water no longer needed to be performed near the home’s main coal fire”(which was historically in the combined kitchen/living room).
“…fires are not lighted in bedrooms except in case of sickness, nor are fires lighted in spare ground floor rooms except on rare occasions,” – Sheffield housing committee report, 1945
Gas-fired heating began to become common, and initially this led to a reduction in heating use: as gas fires provided instant heat rooms could be heated for short periods of time and there was no longer any need to bank fires overnight, but as people became used to this convenience, they began to use heating more often, for example in the mornings before going to work, which had not previously been the case (previously men went to work and women stayed in the room in which there was a cooking fire, but as women increasingly went out to work as well, morning heating became common). The use of thermostats and the removal of the need for manual intervention (adding fuel) meant that heating was on for longer than had been the case with coal fires.
But despite the increase in women going out to work, there tended to be more people in homes for more hours than had been the case when people typically went out to work at the age of 14. Now teenagers would come home from school and need a warm place to do homework, and engage in leisure and social activities. It also became more common for children to play indoors when in the past they had typically been outside all day.
In the early post-war years, activities had moved from the kitchen/living room to the kitchen and new dining rooms and living rooms fell into disuse, but with the advent of the television, living rooms began to be used more extensively, and eventually dining rooms fell into disuse. This led to more open-plan living, which in turn boosts the need for heating since larger spaces require more energy to heat.
Today’s norms (notwithstanding the current lockdown) are for more rooms and a larger portion of houses to be heated, for longer periods as occupants spend more time indoors, spread across more rooms.
At the same time, people have a higher expectation of warmth – the days of people dressing under the covers in their unheated bedrooms are typically gone, although the research does indicate scope for heating demand to rise still further particularly with bedrooms. It is still not typical for bedrooms to be heated throughout the night, but some groups recommend constant heating for older people and small children, and there is a trend for people to wear less nightwear, meaning bedrooms need to be warmer to compensate.
…but alternative off-grid lifestyles provide some insights into how these trends can be reversed
Hope et al in their paper: Consumer engagement in low-carbon home energy in the United Kingdom: Implications for future energy system decentralisation looked at the findings of two separate studies into off-grid living: one focusing on communities on the North Norfolk coast who had electricity but not mains gas and who used at least one non-grid technology for generating heat or power. The second study examined people living full or part-time on canal boats, disconnected from any centralised energy networks.
The most significant finding from this research was that the subjects adapted to limited energy supplies by diversifying their energy sources, using traditional polluting fuels such as wood, coal and oil alongside cleaner options such as solar and air source heat pumps. However, use of solid fuels required both time and physical labour (finding and transporting fuel, and clearing out ashes etc), and some elderly and vulnerable participants found this difficult so switched to using bottled gas, but despite this most participants reported having recently installed wood-burning stoves. In fact, such stoves are growing in popularity across the developing world, creating renewed air quality concerns in places like London.
There were some interesting differences between the two groups. Those living on narrowboats found reliance on battery storage and renewables meant that shifting the time of energy use was more of a necessity than a choice since they had to use energy when it was available. However, the Norfolk participants shifted energy use to reduce the cost of energy consumption, since they generally had the choice to use grid electricity if necessary (the rural location meant that electricity outages were more common than in most other parts of the country).
The participants on narrowboats also made changes to the types of appliances they used in order to reduce energy demand. For example, a number of participants had switched from electrical kitchen appliances to manual ones and some had even replaced fridges with cool boxes, and/or were only buying what they needed to eat on a daily basis.
Interestingly, participants from both groups found modern washing machines which use cold water which is heated in the machine to be a “significant annoyance” since it meant they were unable to use hot water that was produced from other sources such as running the boat engines or from solar thermal systems.
“I don’t use that much hot water really because in fact when it comes down to it […] washing machines and dishwashers are cold-feed. And I find that really, really annoying because when I had that [solar thermal] installed I had a washing machine that wasn’t [cold feed] and now the only way you can use hot water is a shower or a bath,” – Norfolk study participant
One of the key findings of these studies is that where energy supplies are constrained, and in particular when they were visibly limited (diminishing piles of wood, falling levels of oil in the tank etc), consumers spend a great deal of time thinking about and organising the securing of those supplies, monitoring the rate of consumption, and deciding how they should be used.
The study subjects had chosen their off-grid lifestyles, but, as the authors note, it is unlikely to be considered fair at the societal level to force the population at large into such energy constraints. It is also unlikely to be beneficial in terms of economic productivity if large parts of the working population must devote significant time and personal energy into organising and managing their energy supplies and consumption in such detail.
“…future demand-side response initiatives may have skewed influence if more affluent households are able to prioritise convenience over cost but less affluent households are not. Similarly, whereas more affluent houses may be able to invest in smart appliances and microgeneration, less affluent houses may again be at a disadvantage.
Furthermore, as found by the present research, some households may also respond in ways that potentially negatively impact on their welfare – for example curtailing energy use. Again, this is more likely to impact on lower income houses, even though these already tend to use less energy than more affluent households, as people try to manage costs” – Hope et al
The challenge for policy-makers is how can some of the off-grid mindsets be replicated across the population in a way that achieves “better” use of energy without creating societal unfairness and causing economic harm to dis-advantaged and vulnerable consumers.
Impact of cognitive biases on consumer decision-making
In their paper, Household energy use: Applying behavioural economics to understand consumer decision-making and behaviour, Frederiks et al explore consumer decision-making in the context of energy use. Economic models are often predicated on the notion that people are “rational”, but in fact they routinely fail to act in ways which objectively reflect the optimal balance of costs and benefits.
This is often seen in relation to energy use – consumers report concerns over climate change and a desire to act sustainably, but this is not reflected in their consumption levels. For example, it is not unusual for people to heat their homes to a degree that allows them to go barefoot in t-shirts all year round, when setting the thermostat lower and wearing socks and a jumper would result in equivalent comfort (while also being healthier).
Energy is a complex arena, and bills are difficult to understand and compare with different construction (standing charges versus variable charges) and units (kWh) that most people cannot relate to.
“Yet even with adequate knowledge of how to save energy and a professed desire to do so, many consumers still fail to take noticeable steps towards energy efficiency and conservation. There is often a sizeable discrepancy between peoples’ self-reported knowledge, values, attitudes and intentions, and their observable behaviour,” – Frederiks et al
These trends are also observed in work settings. Research has shown that 56% of the total energy used in office buildings is consumed during non-working hours simply because occupants leave lighting and equipment on at the end of the day.
The Frederiks study identified a set of behaviours that inhibit the making of rational choices (see box).
Research indicates that consumer choices and behaviour are often driven to a large extent by cognitive biases, and that various mental shortcuts are used to deal with complex decision-making. Yet these factors are often over-looked by policy-makers when developing energy efficiency and use initiatives.
van den Broek and Walker looked at the way in which consumers approach energy decision-making in practical terms:
“Findings show that 1) participants used as many as twenty-four different heuristics in an energy judgement task – an order of magnitude more than identified in existing literature; 2) participants are aware they use the heuristics, but awareness varies per heuristic; 3) the use of these heuristics can be changed and this in turn can improve energy literacy,” – van den Broek and Walker
The paper explores consumers’ “device literacy”, that is the ability of consumers to accurately estimate the energy consumption of household appliances. Various studies have found that participants used heuristics when judging the energy consumption of household appliances – heuristics are simple rules used to reduce the cognitive load of decision-making and prevent information overload.
Previous studies have identified three types of heuristics used to judge the energy consumption of household devices:
a size heuristic where large devices are thought to use more energy than small devices – this implies that the energy consumption of small devices can be underestimated while the consumption of large devices is overestimated;
a usage pattern heuristic where devices that are used frequently or for long periods are thought to use a lot of energy per year while those that are rarely or briefly used are thought to use less energy – this suggests that the energy use of devices that consume high levels of energy per unit of time but are rarely or shortly used may be underestimated and vice versa;
an availability heuristic where the visibility of the appliance may be a factor in decision-making, although the evidence for this is weaker.
van den Broek and Walker conducted three studies that investigated the heuristics used in energy judgements further. The first study identified many more heuristics used to arrive at an energy estimate, thereby suggesting that this decision-making process is far more complex than has previously been shown. Of the three heuristics identified in previous studies, only the size heuristic was confirmed in this study.
Interestingly, the study also showed that participants only recalled the use of half of the identified heuristics, suggesting lack of awareness of their use, however, in the second study, participants self-reported to have used all of these heuristics in an online survey. Participants tended to report using the heat and time switched on heuristics most frequently.
The third study demonstrated that the use of these heuristics can be changed in a way that improves energy literacy, although only a small proportion of the changes in energy literacy could be explained through the use of the heuristic targeted by the intervention.
There were some limitations to the studies (selection of participants being one) which could affect the results, but the wide range of heuristics is likely to be found more broadly across the population. Where this matters is the extent to which energy consumption choices would be affected. For example, knowing the relative energy use of a freezer and a hair-dryer may not affect their use as freezers tend to be on all the time and hair-dryers used only when needed. Consumers may have little discretion in the timing of the use of their hair-dryer (needing to dry hair before going to work), although there is a degree of discretion over the duration/amount of the drying (partial drying).
However, the study raises interesting questions that should be explored further if the right policy interventions are to be made to change energy consumption behaviours.
Combining energy as a service with energy visualisation tools
If we consider that the energy used in buildings relating to consumer choice comprises two elements, a basic element for the minimum heating, lighting, food refrigeration and so on, and a more discretionary element which includes entertainment, EV charging, and other “non-essential” consumption, then we might also consider that different approaches should be taken to each type of consumption.
While consumers may have the ability to influence their basic consumption by purchasing more energy efficient appliances (and undertaking home improvements), these are typically infrequent choices requiring significant capital outlay which is out of the reach of many consumers. Optimising the use of these devices is also challenging for many, who may lack the knowledge, time or intellectual capacity to engage with these choices.
The delivery of energy as a service reduces the need for consumers to become educated about their basic energy use and how it can be optimised day-to-day. Service providers could bundle energy supply with a degree of building management, for example remotely controlling heating to deliver the consumer’s desired comfort levels optimally.
These services could also include the provision of new appliances such as hydrogen boilers, heat pumps, solar panels, domestic batteries, and white goods, with the capital costs being spread over time in the same way that mobile phone contracts bundle the cost of the handset with minutes of phone-calls, numbers of messages and amounts of data. Subsidies could be made available to low-income households to enable them to afford more efficient appliances.
Energy as a service could also be deployed to reduce wasteful consumption since lights and some appliances could be powered off remotely or automatically.
Energy as a service would be less appropriate for some of the discretionary consumption. EV charging could be included, but it would be difficult to optimise the energy used in entertainment through time-shifting, and most people already charge devices overnight for convenience rather than cost reasons. Here the provision of improved visualisation tools that support decision-making would be valuable, allowing consumers to make informed yet straightforward choices about which devices to buy, when they should be replaced (from an energy consumption perspective) and whether there are ways in which their use could be improved.
For example, the off-grid study found that many of the subjects had replaced electric appliances with manual ones, such as whisks in cooking. Quite a few household devices are electric when manual ones can work just as well, with a little more time and physical exertion (such as toothbrushes, and a wide range of kitchen devices including can openers, carving knives, blenders and so on) – in other words devices which save minimal amounts of physical labour.
But it is important that low-income and vulnerable consumers are not forced back into the 19th century – washing clothes by hand is both more time consuming and less effective than using a washing machine, and while gadgets such as electric carving knives are un-necessary luxury items, care must be taken to avoid making energy inequalities worse.
Reducing the energy use from buildings is a challenge that requires bold thinking
In this series of posts I have explored the drivers of the energy use of buildings, and it is clear that if net-zero targets are to be met, there needs to be serious attention to the question of how building energy use can be made more sustainable.
The fabric of buildings needs to be improved to ensure that energy consumption is minimised, but in order to do this, the EPC approach needs to be reformed to make it fit for purpose. A serious effort must be made to close the performance gap, with mandatory post construction and post renovation/retrofit testing using actual data to measure energy consumption to identify weaknesses in design, failures of the construction and installation processes, or lack of user education.
Lifecycle analysis needs to be undertaken to ensure that embodied emissions are taken into account. Buildings should be built to last, and when their end-of-life is reached, their materials should be re-cycled.
Consumers should be helped to manage their consumption through new energy management services bundling energy assets and appliances (which should all be subject to mandatory testing and energy consumption certification) and better energy visualisation tools.
At each step, low income and vulnerable consumers should be supported with schemes that deliver the energy savings they promise, and subsidies should be available for low-energy appliances. And all consumers, must be kept physically safe, so appliances that pose a fire risk should not be run at night by energy service providers unless suitable monitoring is in place, and consumers on low incomes and/or with vulnerabilities should not feel pushed into operating such appliances themselves at night-time in order to make use of cheaper prices, particularly since these groups are more likely to have older and therefore less safe appliances.
I have deliberately excluded the means of heating from this analysis, other than it’s role in the EPC ratings, because different solutions are likely to be preferred in different parts of the country, and so the choice of hydrogen or heat pumps etc might be outside the control of the owners or developers of buildings.
I have also excluded other aspects of sustainability such as water management, as being outside the scope of a discussion on energy, but these areas will still be important for policy-makers.
What is clear is that improving building energy performance presents a significant challenge, and that some radical changes need to be made in order to meet the Government’s targets.
Part I: Improving the EPC system
The Government wants to see the EPC rating for all homes improved, but the system is flawed: electric heating which generates lower emissions than gas heating is heavily penalised while the condition of buildings is assumed to be perfect.
This makes it difficult for consumers to make effective improvements to buildings that both reduce costs and emissions.
The is currently no obligation for building energy performance to be measured at any stage in its life. This means that poor design, faulty installation or consumer errors are not identified or corrected, and there is no mechanism for the deign of buildings to be improved. Embodied emissions are also often overlooked: if buildings are to become genuinely sustainable, lifecyle emissions analyses are vital.