National statistics

English Housing Survey 2021 to 2022: energy

Published 13 July 2023

Applies to England

Introduction and main findings

The English Housing Survey (EHS) is a national survey of people’s housing circumstances and the condition and energy efficiency of housing in England. It is one of the longest standing government surveys and was first run in 1967. This report provides the findings from the 2021-22 survey.

Impact of COVID-19 on the English Housing Survey

The 2021-22 English Housing Survey data was collected toward the end of the period of restrictions due to the COVID-19 pandemic. This was also prior to recent energy cost increases and government action to help mitigate the impact on householders. This necessitated a change in the established survey mode. Face-to-face interviews were replaced with telephone interviews and internal inspections of properties were replaced with external inspections, where the inspection was restricted to an assessment of the exterior of the dwelling and supplemented by information about the interior of the dwelling the surveyor collected (socially distanced) at the doorstep.

There were also some data we were unable to collect at all, in which case predictive modelled estimates at dwelling level were produced to supplement the ‘external plus’ inspection and indicate whether or not a dwelling: had damp problems; had any Category 1 hazards assessed through the Housing Health and Safety Rating System (HHSRS); or met the Decent Homes Standard. In these instances, we have been able to model data to provide headline figures for this report. We indicate where this has been done at the beginning of each topic area. Additionally, as interviewers were unable to identify vacant dwellings in the 2020-21 data collection year, and dwelling level data includes two survey years, all dwelling estimates for this report are based on occupied dwellings only.

More information on the impact of COVID-19 on the English Housing Survey and the modelling methodology can be found in Annex 5.5 of the Technical Report.

This report

This report predominantly focuses on energy efficiency and is split into four chapters. The first chapter gives an overview of the energy efficiency of the English housing stock between 2011 and 2021, before exploring energy efficiency in 2021 by dwelling and household characteristics. The chapter then looks at what an average dwelling with an Energy Efficiency Rating (EER) of band D might look like.

The second chapter reports on the presence of different heating systems. It then looks at heating controls and methods of keeping cool as they play a vital role in internal temperature control and comfort.

The third chapter reports on whether households had changed their fuel supplier and/or tariffs and the methods of payment households use to pay for their energy. It then explores household occupancy patterns during the winter before proceeding to analyse the characteristics of households who work from home.

Finally, the fourth chapter focuses on the cost of improving homes that have EER of band D or lower to an EER band of at least C and identifies the more common energy efficiency improvement measures within the English stock.

Main findings

Over the past decade, energy efficiency of English housing has continued to improve.

• The proportion of A to C rated dwellings increased from 16% in 2011 to 47% in 2021. While the proportion of the least energy efficient dwellings (E to G) decreased from over a third of dwellings (35%) in 2011 to one in ten dwellings (10%) in 2021.

In 2021-22, the estimated average cost to improve dwellings to at least a band C was £7,529, with social rented homes costing the least.

• Owner occupied and private rented dwellings cost on average £7,872 and £7,430, greater than the £5,345 and £5,158 required to improve local authority and housing association dwellings.

The most likely groups of households in employment to work from home four or more days a week were owner occupiers and households living in London.

• In 2021-22 there were 5.7 million households worked at home four or more days a week.
• Owner occupiers (45%) were most likely to work at home four or more days a week, followed by private renters (35%). Local authority (14%) and housing association (19%) renters were least likely to work from home for this amount of time.
• London was the only region where more than 50% of households had an adult working from home four or more days a week

As household income increased so did the likelihood of someone in the household working from home four or more days a week.

• Around 60% of households in the highest income quintile (fifth) had at least one adult working from home four or more days a week, compared with 42% in the fourth quintile, 34% in the 3rd quintile, 23% in the second quintile and 20% in the lowest income quintile.

Social renters were more likely than other tenures to pay for their electricity bills using a prepayment meter. Owner occupiers and private renters were more likely to pay for electricity using direct debit.

• Local authority and housing authority renters were more likely to pay for their electricity using a prepayment meter (39% and 37%, respectively) than owner occupiers (2%).

Acknowledgements and further queries

Each year the English Housing Survey relies on the contributions of a large number of people and organisations. The Department for Levelling Up, Housing and Communities (DLUHC) would particularly like to thank the following people and organisations without whom the 2021-22 survey and this report would not have been possible: all the households who gave up their time to take part in the survey, NatCen Social Research, the Building Research Establishment (BRE) and CADS Housing Surveys.

This report was produced by Ana Slater, Dan Windsor, and Thomas McCutcheon at BRE in collaboration with NatCen Social Research and DLUHC.

If you have any queries about this report, would like any further information or have suggestions for analyses you would like to see included in future EHS reports, please contact ehs@levellingup.gov.uk.

The responsible analyst for this report is: Claire Smith, Housing and Planning Analysis Division, DLUHC. Contact via ehs@levellingup.gov.uk.

1. Energy efficiency profile

This chapter provides an overview of the energy efficiency of the English housing stock and how this has changed between 2011 and 2021. It also explores trends in energy efficiency by dwelling and household characteristics and concludes by looking at the features of a typical EER band D dwelling.

Trends in energy efficiency

The English Housing Survey (EHS) uses the Government’s Standard Assessment Procedure (SAP 2012) to monitor the energy efficiency of homes, through the calculation of a SAP energy efficiency rating (EER).

The EER is an index based on calculated energy costs for a standard heating regime and is expressed on a scale of 1 (highly inefficient) to 100 (highly efficient with 100 representing zero energy cost). It is possible for a dwelling to have a rating of over 100 where it produces more energy than it consumes, although such dwellings will be rare within the English housing stock.

The EER is also converted into an A to G banding system, where band A represents high energy efficiency and band G represents low energy efficiency. The EER is the primary rating presented on an Energy Performance Certificate (EPC).

The mean SAP rating for all dwellings increased by 9 points (improved) over the decade, from 57 points in 2011 to 66 points in 2021. Reflecting this, the proportion of A to C rated dwellings increased from 16% of the stock in 2011 to just under half (47%) in 2021 while the proportion of the least energy efficient dwellings (E to G) decreased from over a third of dwellings (35%) in 2011 to one in ten dwellings (10%) in 2021. Similarly, there was a small decrease in the proportion of dwellings in band D from 49% to 43%, also reflecting overall energy improvements, Annex Table 1.1 and Figure 1.1.

Figure 1.1: Energy efficiency rating bands, 2011 and 2021

^{Base: all occupied dwellings
Note: underlying data are presented in Annex Table 1.3
Source: English Housing Survey, dwelling sample}

Figure 1.2 demonstrates the transition of the individual EER bands from 2011 to 2021, highlighting the overall improvement in the energy efficiency of the English housing stock.

Figure 1.2: Energy efficiency rating bands, 2011 to 2021

^{Base: all dwellings
Notes:
1) from 2018 the SAP 2012 methodology used new U values for cavity, solid and stone walls, both insulated and uninsulated
2) EER A and B bands are grouped. There are currently insufficient numbers of Band A properties existing for which meaningful estimates can be made through a sample survey
3) due to the COVID-19 pandemic, EHS surveyors did not conduct any inspection of vacant properties in 2020. Although an external inspection of vacant homes occurred in 2021, the 2021 combined survey dwelling sample is for occupied properties only
4) underlying data are presented in Annex Table 1.2
Source: English Housing Survey, dwelling sample}

Energy efficiency profile by tenure

There was an increase in mean SAP rating across all tenures over the 10-year period. The biggest improvements were to owner occupied and private rented dwellings, both increasing by 10 SAP points, from 56 and 55 in 2011 to 66 and 65 in 2021, respectively. Dwellings in the social rented sector continued to outperform the private sector with an average SAP of 63 in 2011 and 70 in 2021, Annex Table 1.1.

Reflecting the findings above, dwellings in the social rented sector were more energy efficient in 2021 than those in the private sector, with 69% of properties having an EER of A to C compared with less than half of the private sector (43%). More specifically, housing association (70%) and local authority dwellings (67%) were more likely to be in bands A to C, compared with private rented (45%) and owner occupied (43%) dwellings, Annex Table 1.3.

The proportion of the least energy efficient dwellings (EER rating of E to G) also decreased over the last 10 years. Most notably, the proportion of private rented and owner occupied dwellings in bands E to G decreased from 41% and 38% in 2011 to 14% and 10% in 2021, respectively. Despite this improvement, 11% of dwellings in the private sector had an EER of E to G (11%) in 2021 compared with 3% of the social rented sector.

Table 1.1: Profile of the most and least energy efficient dwellings, by dwelling and household characteristics, 2021-22

	A, B or C rated dwellings are most prevalent/ likely in this group		E, F or G rated dwellings are most prevalent/ likely in this group
Dwelling characteristics		%		%
Tenure	housing association	70	private rented	14
Dwelling age	post 1990	81	pre 1919	25
Dwelling type	purpose built flat, high rise	85	converted flat	22
Region	London	56	South West	15
Household characteristics		%		%
Age of HRP	16 to 24	61	75 or over	14
Income	4th quintile	49	highest income (5th quintile	11

^{Notes:
1) these are trends and may not be statistically significant
2) percentages are within each group. For example, 70% of housing association
dwellings are in EER bands A to C, while the remaining 30% are in bands D and E to G
3) underlying data are presented in Annex Tables 1.5 and 1.6
Source: English Housing Survey, dwelling and household sub sample}

Energy efficiency profile by dwelling characteristics

In 2021, the most energy efficient homes, in EER bands A to C, were newer dwellings. The majority of dwellings built after 1990 were in bands A to C (81%), whereas 18% of dwellings built before 1919 had an EER of A to C. Conversely, the oldest dwellings in the stock were more likely to have an E to G rating with around a quarter of dwellings built before 1919 (25%) in bands E to G compared with newer dwellings (2% to 9%), Figure 1.3.

Figure 1.3: Energy efficiency rating bands, by dwelling age, 2021

^{Base: all occupied dwellings
Note: underlying data are presented in Annex Table 1.5
Source: English Housing Survey, dwelling sample}

Purpose built high rise flats, were the most energy efficient with the highest proportion of A to C rated dwellings (85%), compared with all other dwelling types. Converted flats, on the other hand, made up the highest proportion of E to G rated dwellings (22%) compared with other dwelling types, Figure 1.4.

Detached houses and bungalows had the highest proportion of the E to G rated dwellings compared with other house types. Detached houses and bungalows were more likely to be in bands E to G (15% and 12%, respectively) than semi-detached (9%), small terraced (8%) and medium/large terraced houses (8%).

Figure 1.4: Energy efficiency rating bands, by dwelling type, 2021

^{Base: all occupied dwellings
Note: underlying data are presented in Annex Table 1.5
Source: English Housing Survey, dwelling sample}

Energy efficiency profile by region

Assumptions about weather conditions, including external temperatures and wind speeds, are standardised for all dwellings when modelling the EER bands under the SAP methodology. Patterns of energy efficiency by region are therefore not affected by differences in regional weather and are instead driven by differences in the physical characteristics of dwellings. The mix of tenures, dwelling types, ages of dwellings and building characteristics within each region means that it is hard to quantify the effect of region alone as a driver of energy efficiency and its improvement.

The mean SAP rating increased across all regions over the 10-year period. Reflecting these findings, the proportion of dwellings in bands A to C also increased, while the proportion of the least energy efficient homes (E to G) decreased across all regions. In 2011, around a third of dwellings across all regions were in bands E to G (31 to 39%), whereas in 2021, the proportion of dwellings in the lowest EER bands (E to G) decreased to around 10% for most regions, Annex Table 1.4.

Dwellings in the North East and London generally had the highest SAP ratings in 2021 compared with other regions. In 2021, over half of the dwellings in London were in bands A to C (56%) compared with dwellings in the East of England (50%), South West (45%), North West (44%), West Midlands (43%), Yorkshire and the Humber (41%) and East Midlands (39%). The larger proportion of A to C dwellings in London likely reflects the prevalence of newer dwellings and the higher proportion of flats in the capital compared with other regions, Figure 1.5.

Figure 1.5: Proportion of A to C rated dwellings, by region, 2021

^{Base: all occupied dwellings
Note: underlying data are presented in Annex Table 1.5
Source: English Housing Survey, dwelling sample}

Energy efficiency profile by household characteristics

Households with younger HRPs were more likely to live in more efficient homes. In 2021, households where the HRP was aged 16 to 44 were more likely to live in EER band A to C rated homes (54% to 61%) than older households, aged 45 or older (39% to 46%). Households where the HRP was aged 75 or older were more likely to live in dwellings in E to G (14%) compared with households where the HRP was aged 25 to 64 years (7% to 10%), Figure 1.6.

Figure 1.6: Energy efficiency rating bands, by age of HRP, 2021-22

^{Base: all households
Note: underlying data are presented in Annex Table 1.6
Source: English Housing Survey, household sub sample}

Households that did not receive any housing support were more likely to live in the least energy efficient homes (EER bands E to G) (10%) than those who did (7%), Annex Table 1.6.

Typical average dwelling in England: Band D profile case study

Around 10.1 million dwellings in 2021 had an EER band of D, a large proportion (43%) of the English Housing stock. This section outlines the characteristics of a typical EER band D dwelling.

The proportion of band D dwellings was higher among those that were owner occupied (71%), constructed pre 1919 (27%), and semi-detached houses (29%). They were more commonly located in the South East (15%), North West (15%), or London (13%), Annex Table 1.7.

Figure 1.7: Characteristics of a Band D dwelling, 2021

A band D dwelling typically had 150mm or more of loft insulation (55%), a boiler system with radiators as the main heating (92%), gas heating (89%), water heating with central heating (93%) and with double glazing throughout (83%). They were also generally more likely to have insulated cavity walls (37%), although over a third had uninsulated solid walls (34%).

2. Heating and feeling comfortable in the home

This chapter explores the presence of different heating systems, and how these varied by dwelling characteristics and regions. It also reports on the presence of heating controls as they can play a vital role in improving internal temperature comfort and assist with fuel cost savings. Finally, the chapter reports on the measures households undertook to keep cool in their home.

Space heating

Space and water heating account for a large proportion of domestic energy use. By 2050, buildings will need to be almost completely decarbonised, by making use of a combination of technologies to minimise their carbon emissions and maximising their energy performance.

In 2021, around 23.7 million occupied dwellings had a primary heating system present. The vast majority of these were boiler systems with radiators (89%) which includes heat pumps, followed by storage heaters (5%), room heaters (3%), communal (2%), and less than one percent had warm air systems or other heating systems, Annex Table 2.1.

Space heating by tenure

Owner occupied dwellings were more likely to have a boiler with radiator system (93%) than local authority (89%), housing association (84%) and private rented (77%) dwellings, Figure 2.1.

Among the less common heating systems, storage heaters were more prevalent among private rented (11%) and housing association dwellings (9%) than owner occupied (3%) and local authority (3%) dwellings. Similarly, there was a higher proportion of room heaters in private rented dwellings (7%) than housing association (2%) and owner occupied dwellings (2%).

Communal heating was less prevalent across the stock compared with other heating systems but was more common in local authority dwellings (8%) than housing association (5%) and private rented (4%) dwellings.

Figure 2.1: Space heating, by tenure, 2021

^{Base: all occupied dwellings
Note: underlying data are presented in Annex Table 2.1
Source: English Housing Survey, dwelling sample}

Space heating by dwelling characteristics

Older dwellings built before 1980 were more likely to have a boiler system with radiators (87% to 96%) than newer dwellings built after 1990 (82%), whereas dwellings built between 1981 and 1990 were more likely to have storage heaters (10%) than dwellings built in or before 1980 (2% to 6%), Figure 2.2.

Room heaters were more common among dwellings built before 1919 (4%) than in dwellings built between 1919 and 1980 (1% to 2%). They were also generally more common in dwellings built after 1990 (4%).

Figure 2.2: Space heating, by dwelling age, 2021

^{Base: all occupied dwellings
Note: underlying data are presented in Annex Table 2.1
Source: English Housing Survey, dwelling sample}

Detached dwellings had the highest proportion of boiler systems with radiators (99%) compared with all other dwelling types, bar semi-detached houses (98%), while purpose built, high rise dwellings had the lowest (29%), Figure 2.3.

Purpose built, high rise flats were most diverse in terms of heating systems. Around 20% of purpose built, high rise flats had storage radiators, a figure higher than bungalows (5%), small terraced (4%) and medium/large terraced houses (1%). Purpose built, high rise (15%) and converted (14%) flats were also more likely to have room heaters than purpose built, low-rise flats (6%), small terraced houses (4%), bungalows (1%) and medium/large terraced houses (1%).

Communal heating was markedly more common in purpose built, high rise flats compared with other flat types. Around a third of purpose built, high rise dwellings had a communal heating system (34%), more than purpose built, low rise (7%) and converted flats (4%).

Figure 2.3: Space heating, by dwelling type, 2021

^{Base: all occupied dwellings,br> Note: underlying data are presented in Annex Table 2.1
Source: English Housing Survey, dwelling sample}

Space heating by region

There were also regional variations with findings reflecting the different distributions of tenure, types and ages of homes within each region.

Dwellings in the East Midlands were more likely to have a boiler system with radiators (94%) than dwellings in the East of England (91%), South East (89%), North West (89%), South West (86%) and London (83%). The lower proportion of boiler systems with radiators in London likely reflects the higher proportion of flats in that region, which also explains why dwellings in London were more likely to have communal heating (8%) than dwellings in all other regions (1% to 2%), Figure 2.4.

Storage radiators were generally more common in dwellings in the South West (8%) and South East (7%) than other regions (3% to 6%). Room heaters were more likely in dwellings in the South West (4%) than dwellings in the East of England (2%), East Midlands (2%), South East (2%) and West Midlands (2%).

Figure 2.4: Space heating, by region, 2021

^{Base: all occupied dwellings
Note: underlying data are presented in Annex Table 2.1,br> Source: English Housing Survey, dwelling sample}

Heat pumps

Heat pumps make use of heat from a source (air, ground or water) and have the potential to provide heating using less energy than traditional systems. Being electrically powered, they also have the potential to be low carbon if the source of that electricity is itself low carbon.

In 2021, less than 1% of the stock or 179,000 dwellings had a heat pump for space and/or water heating, Annex Table 2.2.

Of the 179,000 dwellings that had a heat pump, around three quarters were owner occupied (74%), 15% were owned by housing associations and 8% were owned by local authorities. The figures for private rented dwellings were too small to report on.

Heating controls

This section reports on main heating systems and whether they had heating controls, which offer households flexibility and comfort around the internal temperatures of their homes, as well as contributing to potential fuel cost savings.

Heating controls can be split into two categories: non-storage and storage heating controls. Non-storage heating controls refer to heating controls present in dwellings with boiler systems with radiators, room heaters, communal, warm air, and other systems. On the other hand, storage heating controls are present in dwellings with storage heating systems only. Depending on their heating system, it is likely that most dwellings had several different types of heating controls present.

Figure 2.5: Heating controls, 2021

^{Base: all occupied dwellings
Note:
1) underlying data are presented in Annex Tables 2.3 and 2.4
2) responses are multi-coded – respondent could have chosen more than one answer
Source: English Housing Survey, dwelling sample}

Starting with non-storage heating controls, the most common heating controls in 2021 were programmers (95%), followed by boiler thermostats (89%), thermostatic radiator valves (TRVs) (89%), room thermostats (88%), time and temperature zone controllers (3%) and weather compensators (less than one percent). As for storage heating controls, manual and automatic charge controls were most popular (61% and 26%, respectively) while celect type controls were less common (one percent of dwellings with storage heaters), Figure 2.5.

Heating controls by tenure

Owner occupied and housing association dwellings were more likely to have non-storage heating controls (99% and 98%, respectively) than private rented dwellings (97%). Looking at the individual non-storage heating controls in more detail, owner occupied, local authority and housing association dwellings were more likely to have a boiler thermostat present (86% to 91%) than private rented dwellings (82%). Owner occupied dwellings were also more likely to have programmers (96%) than all other tenures, Annex Tables 2.3 and 2.4.

Housing association and local authority dwellings had a higher proportion of TRVs (93% and 92%, respectively) compared with owner occupied (89%) and private rented dwellings (85%). Owner occupied dwellings (89%) were also more likely to have TRVs than private rented dwellings (85%).

Overall, housing association dwellings were more likely to have storage heating controls present (98%) compared with owner occupied and private rented dwellings (90% and 82%, respectively). More specifically, housing association and owner occupied dwellings (68% and 66%, respectively) were more likely to have manual charge controls than private rented dwellings (52%).

Heating controls by dwelling characteristics

Non-storage heating controls were more common in dwellings built between 1919 to 1964 (99%) and 1981 to 1990 (99%) than in those built before 1919 (97%). More specifically, boiler thermostats were more prevalent in dwellings built in or before 1990 (87% to 93%) than dwellings built post 1990 (84%), Annex Table 2.3.

Programmers were more common in dwellings built between 1919 and 1964 (96%) than dwellings built before 1919 (93%). The same trend was evident for TRVs, however, TRVs were also more common in dwellings built post 1990 (89%) than those built pre 1919 (86%).

Among dwellings with storage heating systems, manual charge controls were more common in dwellings built before 1919 (62%) and dwellings built from 1945 onwards (59% to 68%) than dwellings built between 1919 and 1944 (32%), Annex Table 2.4.

Non-storage heating controls were more prevalent in detached houses (99%) than in small terraced houses (98%), purpose built, low rise flats (96%), converted (94%) flats and purpose built, high rise (92%), Annex Table 2.3.

Just over a third of purpose built, high rise dwellings had a boiler thermostat present (35%) which was considerably lower than all other dwelling types (78% to 93%). Similarly, programmers were less common in purpose built, low rise (90%), purpose built, high rise (72%) and converted flats (86%) than in other dwelling types (95% to 97%). Again, there was a lower proportion of purpose built, high rise and converted flats (both 78%) with TRVs than other dwelling types (88% to 91%).

Manual charge controls were more common in semi-detached houses (80%) with storage heaters than purpose built, low rise flats (60%), converted flats (55%), medium/large terraced houses (45%) and detached houses (40%) with the same heating system. Conversely, converted flats were more likely to have automatic charge controls (41%) than semi-detached houses (12%), Annex Table 2.4.

Heating controls by region

Regional findings reflected different tenures, dwelling types, dwelling ages and other building characteristics. It is difficult to quantify the influence of region alone on the presence of heating controls.

Dwellings in the North East (97%), South East (96%), East of England (96%) and Yorkshire and the Humber (95%) were more likely to have programmers present compared with dwellings in London (93%). Dwellings in London were also generally less likely to have a boiler thermostat present (85%) than other regions (87% to 94%), while TRVs were more common in the East of England (91%) than the North West (88%), Yorkshire and the Humber (88%), West Midlands (88%) and London (87%), Annex Table 2.3.

For dwellings with storage heating systems, there was a higher proportion of manual charge controls in the West Midlands (86%) than in the South West (70%), Yorkshire and the Humber (58%), South East (54%), London (51%) and North West (45%), Annex Table 2.4.

Keeping cool in summer

Being able to cool down in the summer months is particularly important for older households who are more at risk of ill-health due to excessive heat. This section reports on the methods householders used to control warm temperatures in their home during the summer months and explores the household characteristics of those who reported keeping cool at night by opening the windows.

Households were asked what methods they used to cool down their home. Households were able to select all methods that were applicable to them, however it cannot be determined whether they were successful at cooling their home with that method alone.

Figure 2.6: Presence of household cooling devices, 2021-22

^{Base: all households who mentioned having at least one of these cooling devices present in their home
Notes:
1) percentages are within each group. For example, 93% of households mentioned opening their windows while the remaining 7% did not mention it
2) analysis excludes ‘no answer’ responses
3) underlying data are presented in Annex Table 2.5
4) responses are multi-coded – respondent could have chosen more than one answer
Source: English Housing Survey, household sample}

The most common methods of keeping cool were opening the windows (93%), switching on the fan (52%), closing the curtains (50%), and closing the blinds (38%). The least common methods were closing the shutters (9%), switching on the air conditioner (3%) and unrolling the awning or canopy (2%). Overall, therefore, the majority of households (97%) mentioned using non-mechanical methods of keeping cool and over half (53%) mentioned using a mechanical appliance to keep their homes cooler, Figure 2.6, and Annex Table 2.5.

keeping cool in summer by tenure

There were few differences by tenure however, owner occupiers (97%) were more likely to use non-mechanical methods to keep cool than private renters (96%). More specifically, owner occupiers were more likely to mention closing the shutters (11%) than private (8%), local authority (6%) and housing association renters (5%). They were also more likely to mention that they closed their curtains to keep cool (52%) than private (45%), local authority (44%) and housing association renters (42%), Annex Table 2.5.

As for mechanical methods of keeping cool, local authority and housing association renters (55% and 54%, respectively) were more likely to switch on their fans compared with private renters (49%) while using the air con was more likely to be reported by owner occupiers (3%) than housing association renters (2%).

Keeping cool in summer by household characteristics

Households where the HRP was aged 35 to 64 or 75 or over (all 97%) were more likely to use non-mechanical methods to keep cool than households whose HRP was aged between 25 and 34 years (95%), Annex Table 2.5.

Households where the HRP was aged between 35 and 44 years were generally more likely to report closing the shutters (11%) compared with other households (8% to 10%). HRPs in the 35 to 44 age group were also more likely to mention closing their curtains to keep cool (54%) than HRPs in all other age groups (41% to 50%), Figure 2.7.

There was a higher prevalence of households where the HRP aged 25 or over who reported that they close their blinds to keep cool (34% to 42%) than households where the HRP was aged 16 to 24 years old (26%).

Figure 2.7: Common household cooling devices, by age of HRP, 2021-22

^{Base: all households who mentioned having at least one of these cooling devices present in their home
Notes:
1) percentages are within each group. For example, 93% of households mentioned opening their windows while the remaining 7% did not mention it
2) analysis excludes ‘no answer’ responses
3) underlying data are presented in Annex Table 2.5
Source: English Housing Survey, household sample}

Households where someone had a long-term illness or disability were more likely to use mechanical appliances to keep cool (57%) than other households (51%). Whereas households where no one had a long-term illness or disability were more likely to open their windows and close their shutters as a method to keep cool (94% and 10%, respectively) than other households (92% and 8%, respectively), Annex Table 2.5.

There was a higher proportion of white HRP households reporting that they close their curtains to keep cool (51%) than Asian (42%), black (37%) and other households (36%). White HRP households (39%) were also more likely to close their blinds than Asian (29%), black (27%) and other (27%) households. On the other hand, Asian and black HRP households were generally more likely to use mechanical methods to keep cool than white and other households. More specifically, Asian and black HRP households were more likely to report using a fan (64% and 59%, respectively) than white and other households (51% and 45%, respectively), Figure 2.8.

Figure 2.8: Household cooling devices, by ethnicity, 2021-22

^{Base: all households who mentioned having at least one of these cooling devices present in their home
Notes:
1) percentages are within each group. For example, 93% of households mentioned opening their windows while the remaining 7% did not mention it
2) analysis excludes ‘no answer’ responses
3) underlying data are presented in Annex Table 2.5
Source: English Housing Survey, household sample}

Keeping cool by opening windows

Households who mentioned opening their windows to keep cool were then asked whether they were able to keep cool at night during the summer by opening their windows. Just under half of households reported that they were ‘always’ able to keep cool in the summer months by opening their windows at night (49%), under a quarter (22%) mentioned that they were ‘often’ able to keep cool, 21% mentioned ‘sometimes’ and 9% mentioned that they were never able to keep cool in the summer months by opening the windows, Figure 2.9.

Figure 2.9: Households’ ability to keep cool at night by opening a window, 2021-22

^{Base: all households who mentioned opening their window to keep cool during the summer months
Notes:
1) analysis excludes ‘don’t know’ responses
2) underlying data are presented in Annex Table 2.6
Source: English Housing Survey, household sample}

Around half of owner occupiers reported ‘always’ being able to keep cool by opening their windows at night (50%) compared with 46% of private renters and 45% of housing association renters. Conversely, housing association and local authority renters (15% and 13%, respectively) were more likely to report ‘never’ being able to keep cool at night by opening windows than private renters and owner occupiers (8% and 7%, respectively), Annex Table 2.6.

Households where the age of the HRP was 45 or over (51% to 52%) were more likely to report being able to ‘always’ keep cool by opening their windows at night during summer months than households where the HRP was aged between 16 and 34 years (43% to 44%).

Households where someone had a long-term illness or disability were more likely to report ‘never’ being able to cool down at night by opening windows (10%) than other households (8%).

There was a higher prevalence of white HRP households who reported ‘always’ being able to keep cool at night by opening the windows (50%) than Asian (44%), black (41%) and other (39%) households.

3. Energy suppliers and tariffs, methods of payment and occupancy rates

This chapter reports on changes in household energy suppliers and tariffs, then examines the method of payment households use to pay for their energy bills. The second part of the chapter explores household occupancy patterns during the winter before proceeding to analyse the prevalence of home working by various household characteristics.

Changing electricity and/or gas supplier

For the whole of the survey period, prices were capped under the Ofgem default tariff cap. The default tariff cap limited the maximum standing charge and price per unit (kWh) a customer on a default tariff can be charged every six months but does not cap a total bill, which will vary depending on how much energy is used. The cap could, in theory, offer less incentive for households to change their energy suppliers.

Householders were asked whether they had changed their electricity and/or gas supplier in the 12 months prior to the survey. The survey period began in July 2020 and concluded in April 2022, therefore householders could have reported changing suppliers as far back as July 2019.

Three quarters of households (74%) did not change either of their gas or electricity supplier. Of the 26% of households who did change suppliers, 22% changed both their electricity and gas supplier, 4% changed their electricity supplier only and less than one percent changed their gas supplier only, Figure 3.1.

Figures may differ from those published by the Department for Energy Security and Net Zero (DESNZ). The DESNZ statistics are updated quarterly and are sourced from OFGEM. The DESNZ figures refer to Great Britain (not England), and report on domestic customers who switch from one energy supplier to another. For more information see the Quarterly Energy Prices Statistical publication.

Figure 3.1: Reported changes to gas and/or electricity suppliers, 2021-22

^{Base: all households
Notes:
1) analysis excludes ‘don’t know’ responses
2) underlying data are presented in Annex Table 3.1
Source: English Housing Survey, household sub sample}

Owner occupiers (25%) were more likely to report changing both their electricity and gas suppliers compared with private renters (17%), housing association renters (15%) and local authority renters (13%), Annex Table 3.1.

Among all dwelling types, households in high rise, purpose built flats were the least likely to change their electricity and gas suppliers (7%).

Changing electricity and/or gas tariffs

Households who had not changed their electricity and/or gas suppliers were then asked whether they had changed their tariffs in the previous 12 months. Around 2.8 million households mentioned changing their tariffs (16%). Around 13% changed both their electricity and gas tariff, 2% changed their electricity tariff only and less than 1% changed their gas tariff only. Most households (84%) however, did not report changing their tariff during this period, Figure 3.2.

The number of households changing tariffs may be under-reported as respondents may not have considered auto-switching (for example at the end of a fixed contract) as a change of tariffs. The Ofgem price cap may have discouraged households from changing energy tariff or if their energy supplier collapsed, they may have been switched onto a new supplier’s tariff. Furthermore, householders may have forgotten switching or be unaware of someone else within the household switching tariffs.

Figure 3.2: Reported changes to gas and/or electricity tariffs, 2021-22

^{Base: all households who didn’t change their electricity/gas suppliers
Notes:
1) analysis excludes ‘don’t know’ responses
2) underlying data are presented in Annex Table 3.2
Source: English Housing Survey, household sub sample}

Current electricity and/or gas tariffs

Households who had not changed their electricity and/or gas suppliers or tariffs, were asked what tariffs they were on at the time of the survey. In some cases, households chose more than one tariff type (i.e., both a fixed and a variable electricity tariff).

Over two thirds of households reported having a fixed electricity tariff (67%) while around a third mentioned a variable electricity tariff (33%). A similar pattern was seen for gas, Annex Table 3.3.

A similar distribution in the type of tariffs was evident for all tenures. Within each tenure, it was more common to pay a fixed electricity tariff than a variable electricity tariff. The same trend was seen for gas tariffs.

A higher proportion of owner occupiers (68%) reported paying a fixed electricity tariff than private renters (64%). Conversely, a higher proportion of private renters (36%) reported paying a variable electricity tariff than owner occupiers (32%), Figure 3.3.

Local authority renters (59%), owner occupiers (59%) and housing association renters (53%) were more likely to pay a fixed gas tariff compared with private renters (47%).

Figure 3.3: Current electricity/gas tariff, by tenure, 2021-22

^{Base: all households who didn’t change their electricity or gas tariffs/suppliers
Notes:
1) percentages are within each group. For example, 68% of owner occupied households mentioned having fixed electricity tariff while the remaining 32% did not mention it
2) analysis excludes ‘don’t know’ responses
3) please note that households may have chosen more than one tariff (i.e. both variable and fixed electricity tariffs)
4) underlying data are presented in Annex Table 3.3
Source: English Housing Survey, household sub sample}

Energy payment method

Households were asked how they paid for their electricity and gas bills. For electricity and gas, most households used direct debit to pay their bills (81% and 82%, respectively), around one in ten used a pre-payment meter (11% and 10%, respectively), 8% used standard credit for both electricity and gas and less than 1% used other methods of payment for either utility, Annex Tables 3.4 and 3.5.

Method of energy payment by tenure

Owner occupiers buying their homes with a mortgage were more likely to pay their electricity (92%) and gas (93%) bills by direct debit than any other tenure (50% to 91% and 51% to 91% respectively), including outright owners (91% for both). Conversely, social renters were less likely to pay by direct debit for electricity and gas (51% and 52%, respectively) than mortgagors (92% and 93%, respectively), outright owners (91% for both) and private renters (72% and 73%, respectively), Figures 3.4 and 3.5.

Figure 3.4: Current electricity method of payment, by tenure, 2021-22

^{Base: all households
Notes:
1) analysis excludes ‘don’t know’ responses
2) underlying data are presented in Annex Table 3.4
Source: English Housing Survey, household sub sample}

Local authority and housing association renters were more likely to pay for their electricity using a prepayment meter (39% and 37%, respectively) than owner occupiers (2%). Similarly, there was a higher proportion of housing association and local authority renters who paid for gas using pre-payment meter (37% and 35%, respectively) than owner occupiers (2%).

Housing association renters (11%), private renters (11%) and local authority renters (10%) were all more likely to use standard credit as a method of paying for electricity bills than owner occupiers (6%). The same was seen for gas payments as local authority, housing association and private renters (all 10%) were more likely to pay for their gas bills with standard credit than owner occupiers (6%).

Figure 3.5: Current gas method of payment, by tenure, 2021-22

^{Base: all households
Notes:
1) analysis excludes ‘don’t know’ and ‘no gas’ responses
2) underlying data are presented in Annex Table 3.5
Source: English Housing Survey, household sub sample}

Smart meters and pre-payment methods

The rollout of smart meters is an essential national infrastructure upgrade that will make the country’s energy system more efficient and flexible, helping to deliver net zero emissions by 2050. Smart meters are the next generation of gas and electricity meters and offer a range of new functions. For example, they can tell residents how much energy they are using in pounds and pence via an In-Home Display. Smart meters communicate directly with the energy supplier, which avoids manual meter reads and provides customers with accurate bills.

In 2021, 44% of households reported having a smart meter in their home. More than a third of households (34%) had both an electricity and gas smart meter, 10% had only an electricity smart meter and less than one percent reported having a gas smart meter only. The remaining households either did not have a smart meter (55%) or did not know their meter type (1%), Annex Table 3.6.

Households paying their electricity via standard credit were less likely to have a smart meter present (7%) and more likely to have a standard, non-smart electricity meter present instead (9%). As for gas, households with a pre-payment meter (12%) were more likely than not to have a smart gas meter present while households paying standard credit were more likely to have a non-smart meter present (8%), Figure 3.6.

Figure 3.6: Reported presence of smart meter, by method of payment, 2021-22

^{Base: all households
Notes:
1) analysis excludes ‘don’t know’ and ‘no gas’ responses
2) self reported presence of electricity and gas smart meters
3) underlying data reported in annex tables 3.4 and 3.5
Source: English Housing Survey, household sub sample}

Occupancy rates: home during the winter

All households were asked about when someone in the household was regularly at home during winter, when heating needs are greatest. Details on the time spent at home during the winter is critical when determining a household’s energy requirements, a major component when calculating Fuel Poverty statistics. Understanding the occupancy and heating patterns of households is also important for assessing the role of energy demand technologies, such as smart heating controls which are designed to manage energy consumption more efficiently.

This analysis looks at occupancy patterns using three years of EHS interview survey data which spanned over five calendar years.

Overall, there was considerable variation in how much time households spent at home during the winter from 2019 to 2021. Much of this variation could be attributed to the COVID-19 pandemic and the tightening and loosening of lockdown measures and social distancing policies between March 2020 and July 2021. Household responses may have been influenced by the timeframe in which the EHS data collection occurred due to overlaps with national lockdowns.

For reference, the 2019 survey year spanned from April 2018 to March 2020, demonstrating occupancy patterns before the COVID-19 pandemic. Both the 2020 and 2021 survey years were impacted by lockdowns as 2020 data was collected from April 2019 to March 2021 while 2021 data comprises of data collected from July 2020 to April 2022. For more information on the EHS fieldwork dates and COVID-19 timeline, see EHS Technical reports and coronavirus timeline.

From 2019 to 2020 there was a large increase in the proportion of people home all day during the winter (from 41% to 52%, respectively). This can be attributed to surveys from the final quarter of the 2020 survey year taking place after the second and third COVID-19 lockdowns which bookended the 2020/21 winter. The same trend was seen from 2020 to 2021 (from 52% of households to 61%, respectively), Annex Table 3.7.

Consequently, there was a decrease in those who report being home only part of the day. The proportion of households at home during winter weekday mornings decreased from 5% in 2019 to 4% in 2020 and then to 3% in 2021. For winter weekday afternoons, the proportion of households at home reduced from 6% in 2019 to 5% in 2020 and 4% in 2021. As for weekday evenings, the proportion decreased from 34% in 2019 to 27% in 2020 and to 20% in 2021.

For weekends, the proportion of households at home during the day decreased from 23% in 2019 to 19% in 2020 and 15% in 2021. In the evening, this decreased from 30% in 2019, to 24% for 2020, to 19% for 2021, Figure 3.7.

Figure 3.7: At home during the winter, 2019 to 2021

^{Base: all households
Notes:
1) analysis excludes ‘don’t know’ responses
2) underlying data are presented in Annex Table 3.7
Source: English Housing Survey, household sample}

The proportion of people who were at home at various times throughout the day during the winter decreased from 2019 to 2020 (22% to 19%, respectively) and decreased again from 2020 to 2021 (19% to 17%, respectively).

Occupancy rates: working from home

To assess the impact of COVID-19 on home working patterns, all employed HRP households in the 2021-22 survey year were asked whether anyone in the household regularly worked from home during the week. Due to the COVID-19 restrictions, which occurred during part of the 2021-22 EHS data collection, it is likely that the number of households working from home is higher than without these restrictions and may not be representative of normal occupancy rates.

In 2021-22, of the households that reported some level of work from home, 5.7 million households worked four or more days a week from home (40%), 1.6 million (12%) worked two or three days a week from home, 435,000 (3%) worked one day a week from home and 159,000 (1%) worked less than once a week from home. Around 6.3 million (44%) of households reported no time was spent working from home, Figure 3.8.

Figure 3.8: Working from home, 2021-22

^{Base: all households with a HRP in full or part-time work
Notes:
1) analysis excludes ‘don’t know’ responses
2) underlying data are presented in Annex Table 3.8
Source: English Housing Survey, household sub sample}

Occupancy rates: working from home by tenure

Owner occupiers (45%) were most likely to work at home four or more days a week, followed by private renters (35%). Households in the social sector were least likely to work from home for this amount of time (17%), Annex Table 3.8.

Owner occupiers (13%) were also more likely to work two to three days from home than all other tenures. Private renters (10%) and housing association renters (9%) were more likely to work two to three days from home than local authority renters (3%). As a result, local authority (81%) renters were least likely to have an adult working from home during the day. Housing association renters (70%) were also more likely to have no adults working from home than private renters (51%) and owner occupiers (37%), Figure 3.9.

Figure 3.9: Working from home, by tenure, 2021-22

^{Base: all households with a HRP in full or part-time work
Notes:
1) analysis excludes ‘don’t know’ responses
2) to safeguard against data disclosure, findings derived from unweighted cell counts of less than 5 and more than 0 are replaced with a “u”
3) underlying data are presented in Annex Table 3.8
Source: English Housing Survey, household sub sample}

Occupancy rates: working from home by dwelling characteristics

Households living in purpose built, high rise flats (55%) and detached houses (50%) were more likely to have an adult working from home four or more days a week when compared with all other dwelling types (29% to 40%), Annex Table 3.8.

Households living in small terraced houses (58%) were generally more likely to have no adults working from home during the day than other dwelling types (28% to 50%).

Occupancy rates: working from home by region

Those living in London were the most likely to have one adult in the household working from home four or more times a week (54%) and were the least likely to spend no days working from home (29%), Figure 3.10.

Households in the West Midlands (58%) were more likely than most other regions (29% to 50%) to have no adults working from home during the week.

Figure 3.10: Working from home, by region, 2021-22

^{Base: all households with a HRP in full or part-time work
Notes:
(1) analysis excludes ‘don’t know’ responses
(2) to safeguard against data disclosure, findings derived from unweighted cell counts of less than 5 and more than 0 are replaced with a “u”
(3) underlying data are presented in Annex Table 3.8
Source: English Housing Survey, household sub sample}

Occupancy rates: working from home by household characteristics

Households with a HRP in full time employment (43%) were more likely to have an adult working from home four or more days a week than households with a HRP in part-time employment (24%). Conversely, it was more likely for a household with a HRP in part-time employment (59%) to have no adults working from home during the week, than a household with a HRP in full-time employment (42%) Annex Table 3.9.

Couples with no dependent children (45%) were more likely to work four or more days a week from home than all other household types (21% to 38%), apart from households made up of couples with dependent children (43%).

In terms of differences by age, households with a working HRP aged between 35 and 44 years old (44%) were more likely to have an adult working from home four or more days a week than households with a HRP aged 45 to 64 and 65 to 74 years old (40% and 25%, respectively).

Households with a working HRP aged 16 to 24 (56%) or 65 to 74 (51%) were generally more likely to have no adults working from home during the week than most other age ranges (43% to 44%).

As income increased, so too did the likelihood of working from home for at least four days per week. Around 60% of households in the highest income quintile (fifth) had at least one adult working from home four or more days a week, compared with 42% in the fourth quintile. Both of these were more likely to work from home for four or more days than households in the lower income quintiles (20% to 34%).

4. Costs of improving to EER band C

This chapter looks at the cost of improving dwellings with an energy efficiency rating (EER) band of D or lower to an EER band of at least C, by dwelling characteristics and household characteristics. The section then identifies the most common energy efficiency improvement measures used within the English dwelling stock.

The Government has set an aspiration for as many homes as possible to be EER band C by 2035 where practical, cost-effective and affordable. In addition, to reduce carbon emissions and reach net zero by 2050. The average dwelling creates around 2.8 tonnes per year of carbon dioxide. The energy use of homes is influenced by multiple factors: the building’s location, orientation, design, construction and engineering services, but also the way it is used, managed and maintained. For each dwelling identified as having an EER band of D or lower, improvement measures were simulated cumulatively using SAP as the underlying methodology. After each improvement, the SAP rating was recalculated until the dwelling reached the threshold for EER band C (SAP rating of 68.5 or higher). The following analysis covers those dwellings that were able to reach an EER band C after modelling (50%). The remaining dwellings in the stock were categorised as already having an EER band of C or higher (47%), receiving at least one improvement measure but were unable to reach EER band C (2%), or not eligible to receive any improvement measures (less than one percent), Annex Table 4.1

In certain cases, a dwelling’s energy efficiency rating may be improved beyond the target band, where a dwelling’s energy efficiency rating is already close to the band C threshold. If a measure with a high SAP improvement yield is installed, for example cavity wall insulation, then the dwelling may be improved beyond a band C into the band B range.

The costs of the measures installed assumes prices from 2005, which have been inflated to reflect current costs. Some measures may have increased in cost since 2005, whereas others may have decreased.

If all applicable energy improvement measures were applied to all dwellings rated below an EER band C, then 96% of those dwellings would shift into an EER band of A to C, while just 4% of dwellings would have an EER band of D or lower, Annex Table 4.2.

Average annual costs of improving to EER band C

The average cost to improve dwellings to a EER band C was £7,529, with an estimated total cost of between £89 and £91 billion. The median cost was slightly higher, at £8,323, to improve dwellings to band C or higher, Annex Table 4.3.

As expected, dwellings with EER bands E to G had a higher average cost to bring up to band C than D rated dwellings; £13,931 compared with £6,221, Annex Table 4.4.

Average annual costs of improving to EER band by tenure

Where it was possible for energy efficiency improvement measures to lift dwellings into an EER band of C or higher, owner occupied and private rented dwellings cost on average £7,872 and £7,430, greater than the £5,345 and £5,158 required to improve local authority and housing association dwellings, respectively, Annex Table 4.4, Figure 4.1.

Figure 4.1: Average cost to improve to an energy efficiency rating band C, by tenure, 2021

^{Base: all occupied dwellings able to be improved to an EER band C
Note: underlying data are presented in Annex Table 4.4 and 4.11
Source: English Housing Survey, dwelling sample}

Average annual costs of improving to EER band by dwelling characteristics

Given the influence of dwelling age and type of dwelling on energy efficiency, it is not surprising that it was more expensive to bring older dwellings up to band C. Dwellings built before 1919 had the highest average cost to improve to band C, £10,861, followed by dwellings built from 1919 to 1944 (£7,226) and dwellings built after 1945 (£5,137 to £5,759), Annex Table 4.4.

Detached houses cost the most to bring up to band C, at £9,003. Conversely, purpose built, low rise and purpose built, high rise flats had the lowest average costs, at £4,032 and £2,450, respectively. All other dwelling types had average costs of between £7,020 and £8,218.

Average annual costs of improving to EER band by region

The North East had the lowest average cost of £4,973 to achieve an EER rating of band C, compared with all other regions which ranged from £6,676 to £8,482, Annex Table 4.4, Figure 4.2.

Figure 4.2: Average cost to improve to band C, by region, 2021

^{Base: all occupied dwellings able to be improved to an EER band C
Note: underlying data are presented in Annex Table 4.4
Source: English Housing Survey, dwelling sample}

Average annual costs of improving to EER band by household characteristics

Households where the age of the HRP was between 25 and 34 generally had a lower average cost to improve their dwelling to an EER band rating of C, at £6,558. This may reflect the type of dwelling they live in, such as flats which are more energy efficient to begin with, Annex Table 4.5.

Households in receipt of housing support had a lower average cost to bring their home up to band C at £6,240, compared with those not in receipt of housing support at £7,067, Annex Table 4.6.

Banded cost to improve to EER band C

It would cost between £5,000 and £9,999 to improve just under half (45%) of dwellings to a band C, and just under a quarter (24%) of dwellings would be improved to band C if £1,000 to £4,999 was spent improving them. Only 5% of dwellings would reach band C with improvements costing less than £1,000, and it would cost more than £15,000 to improve 8% dwellings to band C. Finally, it would cost between £10,000 and £14,999 to improve 18% of dwellings, Figure 4.3.

Figure 4.3: Banded cost to improve to band C, 2021

^{Base: all occupied dwellings able to be improved to an EER band C
Note: underlying data are presented in Annex Table 4.7
Source: English Housing Survey, dwelling sample}

Overall, it would cost less than £10,000 per dwelling to improve around three quarters (74%) of dwellings to band C, Annex Table 4.7.

Dwellings rated E to G (36%) were more likely to require £15,000 or more to reach an EER band of C than dwellings with a rating of D (2%). Conversely, D rated dwellings were more likely to require £1,000 to £4,999 to improve to band C (28%) than E to G rated dwellings (10%), Annex Table 4.8.

Improving dwellings with an EER of D was more likely to require less than £15,000 (6% to 51%) than £15,000 or more (2%).

Banded cost to improve to band C by tenure

Overall, it would be cheaper to improve dwellings in the social rented sector and more expensive to bring private sector dwellings to band C. Housing association dwellings (11%) were more likely to cost less than £1,000 to bring up to band C than private rented (6%) and owner occupied (4%) dwellings. Conversely, owner occupied dwellings (10%) were more likely to cost £15,000 or more to bring up to band C than private rented (5%), local authority (3%) and housing association (1%) dwellings, Annex Table 4.8.

Among all private sector homes, 4% required less than £1,000 to improve to an energy efficiency rating band of C while 9% would require £15,000 or more.

Banded cost to improve to band C by dwelling characteristics

In general, improving older dwellings (built before 1919) to band C was more expensive, with 17% likely to cost £15,000 or more compared with all other aged dwellings (2% to 5%). Similarly, dwellings built after 1990 (17%) were more likely to cost less than £1,000 to bring up to band C than older dwellings (3% to 10%).

Detached houses (15%) were more likely to cost £15,000 or more to reach band C than all other house types (6% to 8%), as well as converted flats (3%).

Banded cost to improve to band C by region

Overall, the cost to improve to band C would be more expensive in the South West compared to all other regions, with 15% of dwellings requiring improvements costing £15,000 or more. This is likely to reflect the type, size and age of dwellings in the area. On the other hand, the North East (42%) was more likely to have dwellings that cost £1,000 to £4,999 to bring up to band C than all other regions (15% to 29%).

Banded cost to improve to band C by household characteristics

Households where the age of the HRP was between 25 and 44 years (6% to 7%) were generally more likely to require less than £1,000 to improve their home to band C than HRP households aged 75 or over (3%). Furthermore, households with a HRP aged 25 to 34 were less likely to require £15,000 or more to bring their home up to an EER of band C (2%), compared with households with other aged HRPs (7% to 10%), bar HRPs aged 16 to 24 years, Annex Table 4.9.

Households not in receipt of housing support were more likely to require £15,000 or more to reach band C (5%) than households in receipt of support (2%), Annex Table 4.10.

Average annual energy cost to improve and savings from improving to EER band C

The average annual energy cost savings of bringing a dwelling up to band C were £285 per year, Annex Table 4.11.

Owner occupied and private rented dwellings had higher average energy cost savings per year (£303 and £285, respectively) than housing association and local authority dwellings (£161 and £138, respectively), Figure 4.3.

Average annual energy cost savings by dwelling characteristics

Dwellings built before 1919 had the highest average energy cost savings of all dwellings, at £428 per year, with dwellings built after this date saving between £192 and £240. Moreover, detached houses had higher average savings per year at £453, compared with other types of dwellings, Annex Table 4.11.

Average annual energy cost savings by region

Dwellings in the North East had the lowest average energy cost savings of all regions at £170 per year, followed by London (£245), North West (£272), Yorkshire and the Humber (£273), South East (£281), East Midlands (£285), East of England (£294), West Midlands (£314) and the South West (£374), Figure 4.5.

Carbon dioxide (CO2) savings

The average CO2 savings for a dwelling to be improved to at least a band C were 1.6 tonnes per year, over half of what an average house produces. Dwellings with a pre-improvement EER band of E to G had the largest average CO2 saving at 4.3 tonnes per year, this would equate to almost 2 years’ worth of energy, Annex Table 4.14.

Dwellings with an EER band of D had lower average CO2 savings, at 1.1 tonnes per year in comparison to E to G rated dwellings at 4.3 tonnes per year.

Carbon dioxide (CO2) savings by dwelling characteristics

Owner occupied homes had the highest average CO2 savings at 1.8 tonnes per year, followed by private rented dwellings at 1.5 tonnes per year. Social rented dwellings had marginally lower CO2 savings, at 0.9 tonnes per year for housing association dwellings and 0.8 tonnes for local authority dwellings.

Dwellings built before 1919 (2.5 tonnes) and dwellings built from 1919 to 1944 (1.5 tonnes) had the highest average CO2 savings per year, compared with dwellings built from 1945 (0.9 tonnes to 1.2 tonnes).

Detached dwellings had the highest average CO2 savings per year at 2.6 tonnes, compared with all other dwelling types which ranged from 1.0 tonnes to 1.6 tonnes.

Carbon dioxide (CO2) savings by region

The North East had the lowest average CO2 savings per year at 1.0 tonnes, in comparison to all other regions which ranged from 1.4 tonnes to 2.1 tonnes, Figure 4.4.

Figure 4.4: Average annual energy cost savings with average annual CO2 savings, by region, 2021

^{Base: all occupied dwellings able to be improved to an EER band C
Notes:
1) SAP 2012 carbon emissions factors do not reflect the decarbonisation of the grid, it is not a lifetime carbon saving, nor does it include carbon savings related to primary electricity generation
2) in certain cases, a dwelling’s energy efficiency rating may be improved beyond the target band, which may overestimate carbon savings
3) SAP 2012 uses a standardised heating pattern which may overestimate actual consumption, particularly in E-G rated homes
4) underlying data are presented in Annex Table 4.11 and 4.14
Source: English Housing Survey, dwelling sample}

Carbon dioxide (CO2) savings by household characteristics

Households where the age of the HRP was 65 to 74 would realise the highest average CO2 savings (1.9 tonnes per year) from the energy improvements to their home, in comparison to HRPs aged 16 to 44 (1.2 to 1.4 tonnes), Annex Table 4.15.

Those in receipt of housing support would have lower average CO2 savings per year at 1.1 tonnes, compared with those who were not in receipt of housing support at 1.5 tonnes, Annex Table 4.16.

Common energy efficiency measures

Improving energy efficiency in homes could reduce heating bills and improve comfort. This section provides a look at the most common energy efficiency measures in each energy efficiency rating band.

Dwellings with 150mm or more of loft insulation (52%), solid walls with insulation (71%) or cavity walls with insulation (65%), 245-litre hot water cylinders (54%), floor insulation (65%), 100% coverage of low energy lighting (55%), and non storage heating controls (49%) were more likely to have an EER band of A to C, Annex Table 4.17.

Dwellings with less than 100mm of loft insulation (61%), solid uninsulated (58%) or cavity uninsulated (55%), and a lack of floor insulation (45%) were most commonly rated as band D.

Dwellings with no loft insulation (63%), solid uninsulated walls (23%), a lack of floor insulation (10%), and no heating controls (55%) were most commonly rated E to G.

Technical notes

The main parts of this report refer to the physical dwelling, are presented for ‘2021’ and are based on fieldwork carried out between July 2020 and March 2022 The sample comprises 10,572 occupied dwellings only where a physical inspection was carried out. Due to COVID-19 restrictions, the sample does not include vacant dwellings, where in previous years’ it did. Throughout the report, this is referred to as the ‘dwelling sample’.

Results for the section on occupancy refers to household data, which are presented for ‘2021-22’ and are based on fieldwork carried out between May 2021 and March 2022 on a sample of 9,752 households. This is referred to as the ‘full household sample’.

The reliability of the results of sample surveys, including the English Housing Survey, is positively related to the unweighted sample size. Results based on small sample sizes should therefore be treated as indicative only because inference about the national picture cannot be drawn. To alert readers to those results, percentages based on a row or column total with unweighted total sample size of less than 30 are italicised. To safeguard against data disclosure, the cell contents of cells where the cell count is less than 5 are replaced with a “u”.

Where comparative statements have been made in the text, these have been significance tested to a 95% confidence level. This means we are 95% confident that the statements we are making are true.

Additional annex tables, including the data underlying the figures and charts in this report are published on the English Housing Survey page alongside many supplementary live tables, which are updated each year but are too numerous to include in our reports.

A more thorough description of the English Housing Survey methodology is provided in the Technical Report which is published annually. The 2021-22 Technical Report includes further details of the impact the COVID-19 on the 2021-22 survey. A full account of data quality procedures followed to collect and analyse English Housing Survey data can be found in the Quality Report, which is also updated and published annually.