Investigating energy performance of build-to-rent buildings in england

Abstract

The domestic sector made up 16% of UK’s total greenhouse gas emissions in 2021. The build-to-rent market is experiencing a rapid expansion, but the energy performance and distinctiveness of this building sector are still generally undiscovered. Therefore, 423 build-to-rent flats from three build-to-rent developments in England were evaluated with top-down approach. Energy use intensities of flats were compared against domestic energy benchmarks and design targets. Possible determinants of energy consumption in build-to-rent flats, including EPC rating, number of bedrooms, air permeability, glazing area and glazing orientation were investigated through correlation analyses. Research discovers discrepancy between actual energy performance of the studied flats and design targets supporting UK’s 2050 net zero target. In this case study, number of bedrooms is identified as the most noticeable factor influencing energy consumption. Results of this study provide an introductory insight of the build-to-rent sector and serve as a starting point to encourage further exploration.

Practical application: This paper specifically focuses on build-to-rent buildings within the domestic sector. Energy use situations of build-to-rent flats in the UK were investigated with actual measured data provided by a build-to-rent developer. Results demonstrate the discrepancy between energy use in the build-to-rent flats being studied and existing domestic benchmarks. This paper provides insights to researchers, building developers and policymakers regarding the energy performance of this upsurging build-to-rent building sector and its associated influencing factors. It also contributes ideas on the importance of data sharing and development of energy benchmarking of the build-to-rent sector.

Keywords

Build-to-rent buildings energy benchmarking energy performance net zero occupancy level

Introduction

The UK government committed to accomplish 100% reduction of greenhouse gas emissions by 2050 referencing to 1990 levels¹ and operational carbon accounts for a higher percentage (28%) than that of embodied carbon (11%) when considering the 39% of global carbon emissions from real estates.² In view of this, attention should be paid on managing energy demand and efficiency in the building sector apart from focusing on engineering and technological advancements. As in 2022, the residential sector was responsible for 16% of UK’s CO₂ emissions.³ In recent times, the build-to-rent market has experienced a rapid growth within the domestic sector. There are over 250 thousand build-to-rent homes completed, under construction and in planning stage as in Q1 2023.⁴ The build-to-rent sector is presumed to worth £170 billion by 2032 and have the potential to become an integral part of UK’s housing delivery and economic growth.⁵ As a major build-to-rent landlord in the UK, Grainger PLC has an operational portfolio value of £3.3 billion from over 10 thousand homes and is currently moving towards the development of around 5600 homes which may help generate another £1.6 billion.⁶ The COVID-19 pandemic has promoted noticeable surge in the search of rental housing.⁷ It has also boosted the demand for build-to-rent developments as people have an increased emphasis on wellbeing, space for working and access to outdoor space when considering home environments.⁸ Therefore, understanding the characteristics of this blooming building type is crucial.

Build-to-rent developments may include different amenities owned by developers such as concierge, roof terrace and co-working space, and may include utility bills in the rents.⁹ These imply that pattern and quantity of energy usage of build-to-rent buildings may possibly differ from that of traditional residential homes. However, there is currently no statistics, energy reports or benchmarks specifically about the energy use of the build-to-rent sector. Similarities or differences between build-to-rent buildings and conventional buildings are still not yet comprehensively validated. Build-to-rent developers often directly own their developments,⁷ so they are keen on understanding the operational energy of their buildings and carrying out suitable measures to improve energy performance, and hence reduce operational costs. Accordingly, more attention should be paid on investigating the energy consumption, energy efficiency and factors affecting energy usage of this emerging building sector.

Studies found that the energy consumption of multi-unit residential buildings (MURBs) in Toronto is between 242 and 453 kWh/m².¹⁰ As for apartments in Jordan, the annual average energy use intensity (EUI) was found to be 31.6 kWh/m².¹¹ This shows that significant difference in study locations’ climate and weather conditions may be a possible cause of variations in energy consumption of MURBs, meaning that research targeting at a specific region may not reflect energy usage situations of MURBs elsewhere. Besides, research discovered that window characteristics, type of heating system, penetration rate of air conditioner, building envelope and demand on end-use appliances are important determinants of energy consumption in residential buildings.^12–15 But the findings are discovered based on general residential buildings instead of build-to-rent buildings. Energy Performance Certificate (EPC) has been used to facilitate energy performance regulation of existing residential buildings in the UK.¹⁶ However, the reliability and usefulness of adopting EPC ratings as parameter of energy performance evaluation were being questioned.¹⁷ There are also other studies providing suggestions on developing star-rating systems for residential buildings energy benchmarking.^18,19 However, asset ratings and proposed benchmarking systems are not dedicated to the build-to-rent sector, and hence the development of benchmarks for build-to-rent buildings remains unknown. There are barely studies justifying the suitability of adopting existing benchmarks regarding general residential homes and flats to evaluate this new building type.

Building energy benchmarking serves as a public yardstick to evaluate the energy use performance of buildings with regards to weather conditions, physical characteristics, device usage and management circumstances.²⁰ It has a substantial position in controlling and supervising energy consumption of buildings. The existing domestic energy benchmarks and targets available for use are listed (Table 1).

Table 1.

List of existing domestic energy benchmarks available for use.

Benchmark/target	Category of interest for study	EUI value (in kWh/m²)	Details
CIBSE TM46:2008²¹	Long-term residential	Electricity: 65	Benchmarks were proposed to complement the procedure of operational rating for DECs, and they aimed to reflect the performance of a building regarding to its building type
CIBSE TM46:2008²¹	Long-term residential	Fossil-thermal: 420
CIBSE benchmarking tool²²	Electricity-heated flats	Good practice (25^th percentile): 85 Typical (median): 123	Benchmarks are mainly expressed in EUI and are based on CIBSE guide F: Energy efficiency in buildings or cumulative frequency distribution of datapoints from datasets
	Gas-heated flats	Electricity EUI Good practice (25^th percentile): 26 Typical (median): 37
	Gas-heated flats	Fossil-fuel EUI Good practice (25^th percentile): 94 Typical (median): 138
Low energy transformation initiative (LETI) target²³	Residential buildings	35 (no more than 15 for heating)	Design targets were set up to support the building stock to achieve net zero carbon by 2050. Target EUIs were developed according to the 2030 and 2050 renewable energy generation in great Britain and the estimated proportion of that energy being applied to various building types
Royal institute of British architects (RIBA) target²⁴	Domestic/residential buildings	35	Voluntary targets were set up to help practices better visualize what should be done to support UK’s whole building stock to achieve the 2050 net zero carbon target
Net zero carbon building standard²⁵	Flats	35 (no more than 15 for heating) (note: performance levels tentative to updates)	Performance levels were created to evaluate operational energy performance of new buildings from various sectors and support them to achieve net zero

After evaluating the background information and existing research regarding the residential property market, it can be seen that there is lack of validated understanding regarding energy performance and benchmarking regarding build-to-rent buildings, especially in the UK context. Also, there is lack of studies revealing the building characteristics influencing energy consumption of build-to-rent buildings. Considering the continuous expansion of the build-to-rent market in the UK, research has to be conducted to review the energy performance of build-to-rent flats and identify aspects which require changes to support net zero. In view of this, this study intends to serve as pioneer research on this building sector in the UK by utilizing actual energy data to evaluate energy performance of build-to-rent flats and carry out energy-related analyses.

Methodology

This research made use of secondary energy consumption data of build-to-rent buildings provided by Grainger PLC, one of the major developers in the build-to-rent sector. This section illustrates methods endorsed to improve quality of data and analysis, and justifies the approaches selected to achieve research aims.

Data collection

Analysis of this research was performed based on secondary data provided by Grainger PLC. The data collection periods and the raw data processing procedures including data cleaning, consolidation, calculations, estimations and adjustments were managed by third party. The dataset includes raw metered energy consumption values, processed annual energy usage data and operational ratings of flats from Grainger’s 10 different build-to-rent developments across England, Developments A to J. The initial dataset consists of 2200 entries of energy consumption data from the developments. Table 2 displays some general information of the build-to-rent developments. Specifications of the apartment flats, for example floor area, floor level and apartment layout were provided. Additional information regarding intrinsic features of the build-to-rent buildings and their associated flats like floor plans, window sizes and air permeability rates were available for certain developments.

Table 2.

General information of the 10 build-to-rent developments from Grainger PLC.

	Location	No. of units	Building fuel type
Development A	London	100	ELEC + HEAT
Development B	London	163	ELEC + HEAT
Development C	Sheffield	237	ELEC
Development D	Manchester	510	ELEC
Development E	Southampton	132	ELEC
Development F	Hampshire	107	Apartment: ELEC Townhouse: ELEC + GAS
Development G	Bristol	194	ELEC + HEAT
Development H	Milton keynes	139	ELEC
Development I	Manchester	376	ELEC
Development J	Leeds	242	ELEC + HEAT

Processing of energy data

A set of criteria was adopted to come up with data records for energy performance evaluation of build-to-rent buildings in this research (Table 3).

Table 3.

Set of criteria used to come up with data records for energy performance evaluation of build-to-rent buildings.

	Criterion	Description
Data cleaning	1	Build-to-rent apartment flats
	2	Flats with no missing energy data points (including both estimated and actual energy consumption data)
	3	Flats with actual energy consumption data
Data processing	4	Flats powered by electricity only
	5	Energy record periods of flats determined based on check-in (move-in) and check-out (move-out) inventory values only
	6	Energy consumption data collected across time periods ≥300 days
	7	Flats from development with ≥60% of desired energy data (data recording period ≥300 days) compared with its sample size

A comprehensive summary of the data processing steps performed is shown (Table 4). The sample size reduced while the granularity increased throughout the data processing and analysis research flow.

Table 4.

Summary of steps performed for processing of energy data.

Process	Details	Sample size	Development involved
Data cleaning	Initial dataset	2200	A to J
	Build-to-rent apartment flats	2165
	Removal of data entries with missing energy data points	2139
	Removal of data entries with estimated energy consumption data	1614
Data processing	Removal of data entries violating criterion 4, 5 or 6	522	C, D, F, H and I
Data processing	Removal of data entries involving flats in developments with <60% of desired energy data (recording period ≥300 days) compared with their sample sizes	423	C, D, and F
Final data		423	C, D, and F
Data for certain in-depth analysis		138	C

At the initial stage of the data cleaning process, there were 2200 energy consumption data entries in the dataset provided by Grainger PLC regarding 10 build-to-rent developments. Considering that 98% of the build-to-rent homes are flats,⁹ this study only focused on energy data of flats to maintain uniformity and comparability of analysis results. Other data cleaning and filtering procedures were then adopted, resulting in 1614 energy data which were considered as valid and suitable for further data processing and analysis. Further criteria were set to reduce effects of different energy consumption patterns linked with tenancy changes and minimize energy data biases associated with seasonal variations of energy demand during recording periods. A uniform recording period for electricity and heat consumption could not be established for flats powered by both electricity and heat. So, only all-electric build-to-rent flats were included in analysis and energy performance evaluation. The scope of study was then further narrowed down to three build-to-rent developments, Developments C, D and F, which had higher percentages (≥60%) of desired energy data at the flat level. Accordingly, the final sample size of the energy dataset proceeded to data analysis stage was 423, approximately 20% of the raw energy data received from Grainger PLC. Certain in-depth analyses regarding factors affecting energy consumption of build-to-rent buildings were conducted with the 138 data entries from Development C only due to inadequate information for the other two developments.

Methods of analysis

After processing the empirical dataset used in this research, quantitative analysis methods were utilized to support the energy performance and benchmarking assessments of build-to-rent buildings. Top-down analysis was conducted in this research by first evaluating the overall energy performance of build-to-rent buildings and then figuring out factors affecting energy usage in build-to-rent flats. To investigate the major building intrinsic features and characteristics affecting the energy consumption of build-to-rent buildings, cross-sectional analysis was performed. It was also adopted to make comparisons between energy performance data provided by Grainger PLC and existing residential energy benchmarks to explore the need of establishing specific benchmarks for the build-to-rent sector. Other than that, research presented that energy consumption per square meter was one of the most suitable indicators when conducting energy consumption evaluation and benchmarking.²⁶ Therefore, energy use intensity (EUI) was calculated for the flats to evaluate the energy performance of flats in build-to-rent developments and compare the consumption values across existing domestic energy benchmarks. In addition, data visualisation was utilized to facilitate the interpretation of analysis results. Boxplots are viewed as a useful visualization tool when performing comparison of data of several distributions,²⁷ and hence were adopted in this study to clearly display the distribution and carry out comparison of energy consumption. The visualisation of continuous data through the generation of boxplots enables clear identification of dispersion and possible outliers in the energy datasets. Scatter plots were also utilized to determine and display the correlation between building characteristics and energy use through observing linear regression lines and their associated R-squared values.

Building characteristics for analysis

Building characteristics which may be determinants of energy performance of build-to-rent buildings were identified (Table 5) by gathering findings from existing literature related to the residential sector. Due to the limited information provided regarding the build-to-rent developments, only five building characteristics were analysed in this study: EPC rating, number of bedrooms, air permeability, glazing area and glazing orientation.

Table 5.

Building characteristics which may be factors influencing energy use in build-to-rent flats.

Characteristics	Existing findings
Air permeability	Air leakages found in residential buildings may lead to a rise in heating and cooling demand for up to 25% and 12% respectively²⁸
Building age	High level of inter-dependence exists between energy consumption, compactness, and age of buildings²⁹ Building ageing along with component degradation is discovered to have negative effects on energy performance of buildings³⁰
Building fabrics specifications	Significant decrease in primary energy consumption and energy required for heating can be achieved by having reduction in the U-value of residential buildings³¹
Building size	Building size is found to be a strong determinant of residential energy consumption.³²
EPC rating	Buildings with better EPC ratings should be more efficient than buildings of lower EPC ratings provided that they use the same heating fuel, and thus resulting in lower fuel bills³³
Geographic location	Buildings situated in different climatic zones have different energy demands³⁴
Glazing area	Considering that 20-40% of energy loss in buildings is in relation to windows, buildings’ energy performance is greatly affected by building envelope including window sizes³⁵
Glazing orientation	A substantial factor affecting energy use in buildings is glazing orientation as it relates to buildings’ capability to make use of solar radiation for heating and lighting.^36,37
Number of occupants	It is discovered that there is association between household sizes and energy consumption in residential buildings³⁸
Surface-to-volume ratio	Energy use of buildings has the most notable correlation with surface-to-volume ratio among different geometric parameters of buildings tested³⁹
Ventilation method	Energy consumption differs among buildings due to the difference in ventilation mechanisms used⁴⁰

Results

Actual energy use intensity results

Energy use of developments C, D and F

As shown in Table 6, the overall median electricity EUI of the electricity-powered flats in Developments C, D and F was found to be 65 kWh/m². It was much lower than the existing domestic energy benchmark stated in the CIBSE benchmarking tool (123 kWh/m²).²² Developments C and D shared similar median electricity EUI values with the median electricity EUI of the overall dataset. As for Development F, a relatively higher median electricity EUI of 87 kWh/m² was observed and this may indicate that the weather conditions, operation mechanisms or building characteristics of this development differ from the others, resulting in a more scattered EUI distribution and higher energy demands.

Table 6.

Overall and detailed annual energy use of electricity-powered flats in developments C, D and F.

	Type of flat	Median electricity EUI (kWh/m²)	Standard deviation	Sample size
Overall	-	65	20.58	423
Development C	Total	65	22.57	138
	Studio	74	23.73	33
	1-bedroom	66	22.92	53
	2-bedroom	62	19.07	52
Development D	Total	64	16.94	246
	1-bedroom	67	15.42	75
	2-bedroom	63	17.26	146
	3-bedroom	58	18.56	25
Development F	Total	87	22.30	39
	1-bedroom	78	23.70	20
	2-bedroom	91	20.17	19

Looking into Developments C and D, the change in median electricity EUI values of flats in relation to their apartment layouts shared a similar trend. EUI values decreased when number of bedrooms increased. This aligns with Chen et al.⁴¹ that flats in larger sizes tend to be less energy intensive. However, Development F did not experience the same trend. A possible cause is that the sample size of Development F is relatively small after categorisation of apartment layouts, so the results obtained were less representable.

Comparison between energy use and domestic energy benchmarks

The median electricity EUI of electricity-powered flats in Developments C, D and F were compared to the related domestic energy benchmarks and design targets (Figure 1). Figure 1 showed that the median electricity EUI of all the flats involved in this case study were significantly lower (by 36 to 60 kWh/m²) than the typical energy consumption value of 123 kWh/m² for electricity-powered flats, as stated in the CIBSE benchmarking tool.²² It indicates that the average energy consumption levels of build-to-rent buildings may be lower than typical performance of residential buildings. However, the median annual electricity EUI regarding the studied build-to-rent flats was still 29 to 53 kWh/m² higher than the annual electricity usage targets introduced by LETI and RIBA (35 kWh/m²).^23,24 This implies that much work may have to be done to enable the flats to meet the ambitious energy design targets which aim to advocate achievement of net zero by 2050. As domestic energy benchmarks in CIBSE TM46 (mentioned in Table 1) are intended to demonstrate typical energy usage in homes power by both thermal and electrical sources,²¹ they are not suitable to evaluate electricity-powered build-to-rent flats, and thus not analysed in this research.

Figure 1.

Median electricity EUI of electricity-powered flats in Developments C, D and F compared to existing domestic energy benchmark and design targets.^23,24

Factors/characteristics affecting energy consumption in build-to-rent flats

With the analyses regarding the annual energy consumption distributions and characteristics of the electricity-powered build-to-rent flats in Developments C, D and F, it can be found that energy use in flats from different developments vary. To gain a more comprehensive understanding of the determinants contributing to different energy consumption values in build-to-rent buildings, various building characteristics and operation practices were analysed.

EPC rating

The range and distribution of electrical EUI of the electricity-powered flats in Developments C, D and F with EPC ratings B and C were found to be similar (Figure 2). By observing the distribution of electrical EUI categorized in terms of EPC rating, build-to-rent flats with EPC rating of B and C both shared the same median EUI of 65 kWh/m². Although EUI values for flats with EPC rating C were more dispersed than the ones with EPC rating B, the two categories shared similar range of EUI values. These indicate that there is insignificant correlation between electricity EUI and EPC rating based on the analysed build-to-rent flats. EPC rating may not be effective in demonstrating the actual energy performance of flats in this case.

Figure 2.

Distribution of electricity EUI of all-electric flats in developments C, D and F categorized in terms of EPC rating.

Number of bedrooms

Relationship between electrical consumption and categories of apartment flat layouts regarding electricity-powered flats in Developments C, D and F was studied (Figure 3). Looking into Figure 3, the electricity consumption of the flats increased with the increased number of bedrooms, resulting in a clear positive trend. Considering that floor area of flats generally increases with number of bedrooms in this study (Figure 4), larger floor areas may lead to higher electricity demand. It agrees with previous research which proved that building size as an important predictor of residential energy consumption.³² In addition, number of bedrooms may serve as a proxy for number of occupants, and thus increasing number of occupants may lead to increase in electricity demand. It agrees with understanding from existing literature that higher aggregate electricity consumption is to be found in larger households.⁴² Besides, IQR of electricity consumption increased as the number of bedrooms increased, meaning that energy usage tends to be more dispersed in flats with more bedrooms. This may be due to increased uncertainties associated with occupants’ living styles and reliance on electricity intensive products in flats with more bedrooms. Above results indicate that type of apartment layout can to some extent explain the variations of electricity consumption in build-to-rent flats of this case study.

Figure 3.

Distribution of electricity consumption of all-electric flats in Developments C, D and F categorized in terms of apartment flat layouts.

Figure 4.

Distribution of floor area of all-electric flats in Developments C, D and F categorized in terms of apartment flat layouts.

Taking into account the floor area of flats, the distribution of electricity EUI of electricity-powered flats in Developments C, D and F categorized in terms of apartment flat layouts was studied (Figure 5). Considering studio, 1-bedroom and 2-bedroom flats, a slight downward trend of overall electricity EUI values could be observed when the number of bedrooms increased. The range of EUI values also decreased with increased bedroom numbers. Considering that regulated loads tend to remain reasonably similar in flats of different sizes, lower fluctuations in energy intensities would be detected when electricity consumption is distributed in larger flats. This may indicate that number of bedrooms to some extent affects electricity EUI of the build-to-rent flats involved in the case study. However, the correlation is not obvious and one possible reason is that flats with higher energy consumption levels tend to have greater floor areas. When the energy usage is adjusted with regards to floor area, the influence of bedroom number on electricity EUI values is relatively insignificant. As for 3-bedroom flats, the overall electricity EUI values did not follow the downward trend and the IQR of EUI values was relatively large as compared to other apartment flat layouts. A possible explanation is that the sample size of this apartment flat layout is limited, and thus affecting the quality of results. Another reason may be that all 3-bedroom flats were from Development D and were situated in a relatively colder geographic region. Therefore, more uncertainties regarding heating preferences of occupants and temperature variations in those flats may be detected, resulting in larger IQR of electricity EUI values. It can be observed that the median electricity EUI decreased as the floor area of flats increased for all four apartment flat layouts. It is supported by existing research which proved that EUI values in homes decreases as household size increases.⁴³

Figure 5.

Distribution of electricity EUI of electricity-powered flats in developments C, D and F categorized in terms of apartment flat layouts.

Air permeability

Considering the electricity-powered flats in Development C, air permeability (in m³/m²h) of flats which were extracted from as-built Standard Assessment Procedure (SAP) reports were plotted against electricity EUI (Figure 6). Looking into Figure 6, a clear trend or pattern regarding the distribution of data points could not be identified. It showed that air permeability may not be a significant contributing factor of variations in electricity consumption regarding the build-to-rent flats analysed in this research. The results agree with existing view that the relationship between air permeability and space heating energy use is weak in non-Passivhaus dwellings.¹⁵ It may be due to the strict air permeability standards adopted in the construction of build-to-rent developments in this study and there were minimal differences in air permeability rates among the studied flats. The differences in occupancy patterns, ventilation practices and temperature preferences associated with the studied flats may also possibly explain the weak correlation between air permeability and electricity EUI.

Figure 6.

Scatter plot of electricity EUI of all-electric flats in development C against air permeability (at 50 Pa).

Glazing area

Looking into the electricity EUI of the electricity-powered flats with single glazing orientation in Development C against window glazing area (Figure 7), it could be observed that there was a slight decline trend of electricity EUI when glazing area increased. A possible explanation is that the reduction in annual heating energy consumption slightly outweighs the rise in cooling load due to higher solar radiation coming through bigger windows. However, the linear trendline only had a R-squared value of approximately 0.075, indicating that variation in electricity EUI of build-to-rent flats explained by window area is relatively limited in this case study.

Figure 7.

Scatter plot of electricity EUI of electricity-powered flats in development C with single glazing orientation against window glazing area.

Another analysis was performed regarding the electricity-powered flats in Development C with single glazing orientation to determine whether energy usage would be influenced by glazing area-to-floor area ratio in the build-to-rent flats (Figure 8). It could be observed that the data points were scattered randomly in general, and an obvious trend could not be determined. A possible reason is that glazing area may not be allocated to apartment flats based on their floor area. Flats with similar glazing area-to-floor area ratios may have different apartment layouts and thus varied energy use patterns. Referring to Figures 7 and 8, factors associated with glazing area may not be notable determinants of electricity consumption of build-to-rent flats involved in this research.

Figure 8.

Scatter plot of electricity EUI of electricity-powered flats in development C with single glazing orientation against glazing area-to-floor area ratio.

Glazing orientation

The scatter plot of electricity EUI against window orientation regarding electricity-powered flats with single glazing orientation in Development C was explored (Figure 9). Flats with windows facing different orientations had different ranges of EUIs and no obvious trend could be deduced. It shows that glazing orientation may not possess a strong correlation with energy consumption in build-to-rent flats which were examined in this research. One probable reason is that the sample size adopted in this case study may not be adequate to present a clear difference in energy consumption of flats with distinct window orientations. Another possible explanation is that the windows constructed in the flats were up-to-standard. The influences of solar gain and heat loss were effectively minimised, making window orientation a less critical determinant of energy demand in the studied flats. Also, energy-intensive appliances, such as air conditioning units, which may considerably influence energy demand based on building orientation were not installed in the studied flats. This may lead to smaller variations in energy demand associated with window orientation.

Figure 9.

Scatter plot of electricity EUI of electricity-powered flats in development C with single glazing orientation against window orientation.

Discussion

Discrepancy between actual energy use and energy benchmarks

The average energy consumption of electricity-powered build-to-rent flats involved in this case study is approximately 50% lower than that of typical electricity-powered residential flats reflected in the CIBSE benchmarking tool (123 kWh/m²).²² One potential cause is that they are mostly newly built, and they adopt improved materials or construction methodologies to adhere to more stringent regulations regarding building fabrics and building services, for example revised versions of Part L under The Building Regulations in UK. This is supported by Liddiard et al.⁴⁴ that energy consumption in homes decrease from older to newer buildings due to advancements in construction technologies and strategies. In addition, build-to-rent developments are usually directly owned by their developers. Developers may design and construct their buildings according to high standards to reduce potential operation costs and acquire foreseeable benefits, resulting in better energy performance. Build-to-rent buildings may have considerable differences in construction mechanism, equipment efficiencies and operation practices as compared with conventional domestic buildings. According to this case study, existing benchmarks may not be able to fully interpret the situation of the build-to-rent flats examined and specific energy benchmarks may have to be set up to better energy performance evaluation. However, these findings are deduced based on a limited sample size so additional investigation is required to verify their validity across a wider scope.

The calculated annual energy consumption regarding the examined build-to-rent flats were much higher than the design EUI targets introduced by LETI and RIBA which are designated to advocate UK’s net zero target (35 kWh/m²).^23,24 It indicates that the build-to-rent flats in this research are not likely to attain net zero by 2050 if radical changes are not initiated. The discrepancy of values may be caused by the high plug loads associated with occupant behaviours and living habits, building defects and deterioration over time, high demand for heating, etc. Therefore, emphasis should also be put on investigating means to execute energy decarbonisation and change energy supply profiles in addition to generating ideas about reducing electricity usage from regulated and electrical plug loads in the flats.

Factors influencing energy use in build-to-rent buildings

EPC rating

Referring to the build-to-rent flats analysed in this case study, it is found that energy consumption of the flats does not differ according to their different EPC ratings, and the electricity EUI values of flats with same EPC rating are quite widely distributed. It agrees with Liddiard et al.⁴⁴ that no obvious pattern can be observed when evaluating relationship between increase of EUI and degradation of EPC rating, and thus energy efficiency expressed in EPC should not be regarded as an indicator which robustly illustrates the actual electricity use for houses in the UK stock. This research further verifies results in Coyne and Denny¹⁷ that although EPC is intended to improve building energy efficiency and inform individuals of the energy performance of a building, disparity often exists between actual energy use and EPC. Statistics from a report written by Passivehaus Trust also revealed that homes in lower EPC bands may out-perform homes labelled with better EPC ratings.⁴⁵ One possible reason is that electrical plug loads are not included in EPC energy calculations, but they are significant contributors (15-20%) to the total amount of energy consumed in buildings.⁴⁶ Besides, build-to-rent flats from only two consecutive EPC rating categories were analysed in this case study. The variations in energy performance of the flats may not be distinct enough to show a clear correlation between electricity EUI and EPC rating.

Number of bedrooms

Analysis results of this case study show that increase in number of bedrooms may lead to rise in energy usage of build-to-rent flats and variations of energy consumption. It may be due to the increase in electricity-related human activities in flats with more bedrooms and probably more occupants. Besides, considering the build-to-rent flats viewed in this research, greater number of bedrooms results in higher variations of energy consumption. This may be associated with the rise in uncertainties regarding occupant behaviours like occupants’ habits, level of reliance on energy intensive electrical appliances, purchasing decisions of appliances, etc. Previous study suggested that occupant preferences and behaviours are important contributors to variations in energy performance of buildings,⁴⁷ so with more inhabitants in a flat, energy usage patterns may differ more. In addition, Office for National Statistics⁴⁸ disclosed that around 33% of flats, maisonettes or apartments was under-occupied by occupancy rating (bedroom) in England in 2021. Therefore, there is a notable probability of households having more bedrooms than required, leading to higher variability of living situation in larger built-to-let flats. This may serve as a valid explanation of why electricity EUI values decreases when number of bedrooms increases in this study. However, the results were generated based on number of bedrooms so the deduced interpretation regarding the correlation between number of occupants and energy usage remains provisional. Additional work has to be performed to generate understanding of the overall situation of the build-to-rent sector and validate the findings of this case study.

Air permeability

This study discovered that there is probably no perceptible correlation between air permeability the studied build-to-rent flats and their energy consumption values. The results from this case study disagree with findings from existing research regarding residential buildings. Xing et al.⁴⁹ and Hashemi & Khatami⁵⁰ proved that improvement in air tightness can help reduce heat loss and energy consumption in buildings. It was further validated in Poza-Casado et al.²⁸ that air leakage in residential buildings may induce increase in heating and cooling demand for up to 25% and 12% respectively. One potential explanation of the inconsistency in outcomes is that the variations of air permeability rates among the analysed flats were relatively minimal, and thus no trend could be identified. The limited sample size of build-to-rent flats examined in this research may also contribute to the difference in outcomes. Besides, building fabrics and construction details of the flats under investigation were generally very up-to-standard with air permeability rates much lower than the maximum air permeability stated in building regulations for new dwellings in UK (8.0 m³/m²h).⁵¹ Therefore, the electricity EUI of the analysed flats may seem less dependent on air permeability rates.

Glazing area

In this case study, glazing area and glazing area-to-floor area ratio both tend to have weak correlation with energy consumption in build-to-rent flats, supplementing to previous finding that moderate increase in window-to-wall ratio does not significantly influence energy consumption and carbon footprint.⁵² A possible cause of the results obtained is that glazing area may not necessarily be allocated to build-to-rent flats according to flat sizes. A relatively larger window may be constructed in flats with a smaller wall or floor area, resulting in a lower demand for heating. Another point is that glazing area mainly affects energy consumption associated with heating and cooling demands but not plug loads. As plug loads account for a substantial proportion of energy consumption in a flat,⁴⁴ glazing area may not be dominant enough to explain the difference in energy consumption of build-to-rent flats. Considering that 20-40% of energy loss of a building envelope is associated with windows,³⁵ additional research is required to study the influence of window-to-wall ratio on energy consumption in build-to-rent buildings.

Glazing orientation

Previous research proved that building orientation potentially has substantial impact on energy use as it determines a building’s ability to utilize solar radiation for heating and lighting to increase energy efficiency.^36,37 However, the result of this case study contradicts with existing literature and the correlation between glazing orientation and energy consumption of the examined build-to-rent flats is found to be weak. One probable reason is that the window dimensions in the flats under investigation differ. As glazing area may also affect demand for energy,⁵³ clear association between glazing orientation and energy consumption cannot be easily established. Besides, flats analysed in this case study may experience different degree of overshadowing due to their surrounding environment, resulting in variations in energy demand.⁵⁴ Overall, factors which may affect energy consumption in relation to glazing orientation should be studied to identify potential connections between glazing and energy consumption of build-to-rent flats with improved accuracy.

Recommendations for build-to-rent sector to move towards net zero

Below are suggested measures for the government, build-to-rent developers and occupants to support the build-to-rent sector to move towards UK’s 2050 net-zero target:

• Cultivate the eagerness of energy data sharing and improve quality and transparency of data collection within the build-to-rent sector to facilitate energy performance evaluation.

• Develop specific energy benchmarks for the build-to-rent sector to enhance energy performance evaluation.

• Ensure quality of building specifications and fabrics and carry out periodic checks or maintenance to prevent deterioration over time.

• Educate occupants regarding appropriate operational behaviours and energy-saving living habits.

• Support the revision of rating system used to display the operational energy performance of build-to-rent buildings.

Conclusion

This study serves as pioneer research of the build-to-rent sector by gathering evidence of energy use characteristics and accomplishing energy analyses regarding build-to-rent buildings in England. The intrinsic features of build-to-rent buildings which may affect their energy consumption are also investigated. Parameters including EPC rating, number of bedrooms, air permeability, glazing area and glazing orientation were analysed in this study. This case study highlights the possible discrepancy of actual energy consumption in build-to-rent flats and existing domestic energy benchmarks as well as design targets. The study also suggests possible measures to support the build-to-rent sector to move towards the 2050 net-zero target.

Key findings picked up after conducting a case study which involved investigations and analyses of energy consumption of build-to-rent flats in England include:

• Existing domestic energy benchmarks may not be suitable to assess the performance of build-to-rent flats.

• EPC ratings may not be appropriate for representing the actual energy performance of build-to-rent flats.

• Number of bedrooms which may serve as a proxy of number of occupants or floor area is the most noticeable factor affecting energy consumption of the studied build-to-rent flats.

• Window orientation and area as well as air permeability of a flat may not be key factors affecting energy consumption of build-to-rent buildings.

• The energy performance of examined build-to-rent flats are better than the benchmarks set for domestic flats.

• The current energy performance of the investigated build-to-rent flats are still far from the design targets which aim to support the achievement of UK’s 2050 net-zero target.

Although appropriate methodologies were adopted to undertake this certain case study, uncertainties and limitations still exist in the data analysis process. Due to the limited sample size of the study, the research remains as a case study and the findings may not be representable and qualified to justify the overall energy consumption conditions of the build-to-rent sector and their associated determinants. Besides, inadequate amount of detailed information regarding the buildings and flats of the selected build-to-rent developments hindered the exploration energy consumption of the amenity areas and certain factors which may affect energy use patterns. Therefore, the results obtained from this study may only reflect the energy performance of a certain portion of build-to-rent buildings within the sector. Future research is required to verify findings of this case study and provide a more complete insight of the energy use situations of the build-to-rent sector.

Research suggestions regarding ways to better understand the energy consumption of build-to-rent buildings and explore possibilities of achieving the net-zero target in this building sector are proposed:

• Conduct research on a larger scale by utilizing datasets with larger sample sizes and more build-to-rent developments.

• Perform evaluation of energy consumption in developments using electricity and fossil fuel, and electricity-powered developments to develop specific energy benchmarks for different types of build-to-rent flats.

• Study the energy performance of different amenity areas in build-to-rent buildings to acquire a more holistic understanding of the sector.

• Figure out other characteristics affecting energy use of build-to-rent buildings.

• Develop multi-regression models to understand correlations between characteristics of build-to-rent buildings and energy consumption.

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iDs

Ying Yan Li

Sung Min Hong

References

UK Government . The climate change act 2008 (2050 target amendment) order 2019. UK Government, 2019.

Wainwright

Demirel

. Multiple logics in financialisation? Moving to carbon sustainability in build-to-rent development. Environ Plann 2023; 55: 22–45.

BEIS . 2021 UK greenhouse gas emissions, final figures. BEIS, 2023.

BPF . Build-to-rent Q1 2023. BPF, 2023.

Property Wire . UK build-to-rent homes to increase fivefold by 2032. https://www.propertywire.com/news/uk-build-to-rent-homes-to-increase-fivefold-by-2032/

Grainger

PLC

. An outstanding year of record delivery with an excellent outlook- full year financial results 2023. https://www.graingerplc.co.uk

Carvalho

Liu

Zhang

, et al. Key themes of build-to-rent: developing a conceptual framework for achieving successful developments through a systematic literature review. Buildings. 2023; 13(8): 1926. DOI: 10.3390/buildings13081926.

RPS Group . What build to rent is thriving, accessed 30 November 2023.https://www.rpsgroup.com/insights/consulting-uki/why-build-to-rent-is-thriving/

Stephanie

Pollitt

. Who lives in build-to-rent? - an analysis of build-to-rent occupancy in London. London First, 2022.

10.

Huang

. Energy benchmarking and energy saving assessment in high-rise multi-unit residential buildings 2012. https://digitalcommons.ryerson.ca/dissertations

11.

Ayadi

Abdalla

Hallaq

, et al. Developing an energy benchmark for residential buildings in Jordan. 2021 12th international renewable engineering conference (IREC). 2021; 1–5. DOI: 10.1109/IREC51415.2021.9427826.

12.

Clarissa

Energy consumption trends of multi-unit residential buildings in the city of Toronto. University of Toronto, 2013.

13.

Yathreb

. Analysis of a residential building energy consumption as ‘base model’ in Tripoli, Lebanon. International Journal of Energy Production and Management 2016; 1: 359–370.

14.

Tereci

Ozkan

STE

Eicker

. Energy benchmarking for residential buildings. Energy Build 2013; 60: 92–99.

15.

Gupta

Kotopouleas

. Magnitude and extent of building fabric thermal performance gap in UK low energy housing. Appl Energy 2018; 222: 673–686.

16.

Zuhaib

Schmatzberger

Volt

, et al. Next-generation energy performance certificates: end-user needs and expectations. Energy Pol. 2022; 161: 112723. DOI: 10.1016/j.enpol.2021.112723.

17.

Coyne

Denny

. Mind the energy performance gap: testing the accuracy of building energy performance certificates in Ireland. Energy Efficiency. 2021; 14: 57. DOI: 10.1007/s12053-021-09960-1.

18.

Wong

Mui

Law

. An energy consumption benchmarking system for residential buildings in Hong Kong. Build Serv Eng Res Tecnol 2009; 30: 135–142.

19.

Gupta

Mathur

, et al. Comparison of energy-efficiency benchmarking methodologies for residential buildings. Energy Build 2023 285: 112920. DOI: 10.1016/j.enbuild.2023.112920.

20.

Ding

Liu

. A comparative analysis of data-driven methods in building energy benchmarking. Energy Build 2020; 209: 109711. DOI: 10.1016/j.enbuild.2019.109711.

21.

CIBSE . Energy benchmarks TM46:2008. CIBSE, 2008.

22.

CIBSE . CIBSE energy benchmarking tool. https://www.cibse.org/knowledge-research/knowledge-resources/knowledge-toolbox/benchmarking-registration, 2021, accessed 6 June 2023.

23.

LETI . LETI climate emergency design guide. LETI, 2020.

24.

RIBA . RIBA 2030 Climate Challenge Version 2. RIBA, 2021. https://www.architecture.com/2030challenge.

25.

UKNZCBS . UK net zero carbon building standard - technical update and consultation. UKNZCBS, 2023.

26.

Ghajarkhosravi

Huang

Fung

, et al. Energy benchmarking analysis of multi-unit residential buildings (MURBs) in Toronto, Canada. J Build Eng 2020; 27: 100981. DOI: 10.1016/j.jobe.2019.100981.

27.

Kendrick

. The box plot: a simple visual method to interpret data. Article in Annals of Internal Medicine. 1989; 110(11): 916–21. DOI: 10.1059/0003-4819-110-11-916.

28.

Poza-Casado

Meiss

Ángel Padilla-Marcos

, et al. Airtightness and energy impact of air infiltration in residential buildings in Spain. International Journal of Ventilation, 2020, pp. 258–264.

29.

Aksoezen

Daniel

Hassler

, et al. Building age as an indicator for energy consumption. Energy Build 2015; 87: 74–86.

30.

Waddicor

Fuentes

Sisó

, et al. Climate change and building ageing impact on building energy performance and mitigation measures application: a case study in Turin, northern Italy. Build Environ 2016; 102: 13–25.

31.

Bikas

Chastas

. The effect of the U value in the energy performance of residential buildings in Greece. J Sustain Architect Civ Eng. 2014; 6(1): 58–65. DOI: 10.5755/j01.sace.6.1.5950.

32.

Huebner

Shipworth

. All about size? – The potential of downsizing in reducing energy demand. Appl Energy 2017; 186: 226–233.

33.

DCLG . A guide to energy performance certificates for the marketing, sale and let of dwellings- Improving the energy efficiency of our buildings. HM Government. 2017.

34.

Pokorska-Silva

Nowoświat

Gać

. Impact of building location on its energy demand. IOP Conf Ser Mater Sci Eng 2021; 1203; 032109.

35.

Alghoul

Rijabo

Mashena

. Energy consumption in buildings: a correlation for the influence of window to wall ratio and window orientation in Tripoli, Libya. J Build Eng 2017; 11: 82–86.

36.

Abanda

Byers

. An investigation of the impact of building orientation on energy consumption in a domestic building using emerging BIM (Building Information Modelling). Energy 2016; 97: 517–527.

37.

Yarramsetty

Rohullah

Sivakumar

MVN

, et al. An investigation on energy consumption in residential building with different orientation: a BIM approach. Asian Journal of Civil Engineering 2020; 21: 253–266.

38.

Xiao

. Critical factors of electricity consumption in residential buildings: an analysis from the point of occupant characteristics view. J Clean Prod 2020; 256: 120423. DOI: 10.1016/j.jclepro.2020.120423.

39.

Jang

Kim

. Empirical analysis of building energy consumption and urban form in a large city: a case of Seoul, South Korea. Energy Build 2021; 245: 111046. DOI: 10.1016/j.enbuild.2021.111046.

40.

Giama

. Review on ventilation systems for building applications in terms of energy efficiency and environmental impact assessment. Energies 2022; 15: 98. DOI: 10.3390/en15010098.

41.

Chen

C-F

Adua

, et al. Exploring the factors that influence energy use intensity across low-, middle-, and high-income households in the United States. Energy Policy, 2022.

42.

Kavousian

Rajagopal

Fischer

. Determinants of residential electricity consumption: using smart meter data to examine the effect of climate, building characteristics, appliance stock, and occupants’ behavior. Energy 2013; 55: 184–194.

43.

Bogin

Kissinger

Erell

. Metrics matter: an integrated approach for analyzing household electricity consumption and the potential for conservation. Energy Build 2024; 307: 113851. DOI: 10.1016/j.enbuild.2023.113851.

44.

Liddiard

Godoy-Shimizu

Ruyssevelt

, et al. Energy use intensities in London houses. Buildings and Cities 2021; 2: 336–353.

45.

Passivhaus Trust . EPCs as efficiency targets- lowering emissions, raising standards. Passivhaus Trust, 2020. https://www.passivhaustrust.org.uk.

46.

Kwong

Lim

Hasim

. Miscellaneous electric loads in Malaysian buildings - energy management opportunities and regulatory requirements. Energy Strategy Rev 2018; 21: 35–49.

47.

Delzendeh

Lee

, et al. The impact of occupants’ behaviours on building energy analysis: a research review. Renew Sustain Energy Rev 2017; 80: 1061–1071.

48.

Office for National Statistics (ONS) . Overcrowding and under-occupancy by household characteristics, England and Wales 2021. https://www.ons.gov.uk/peoplepopulationandcommunity/housing/articles/overcrowdingandunderoccupancybyhouseholdcharacteristicsenglandandwales/census2021 (2023 accessed 2nd December).

49.

Xing

Hewitt

Griffiths

. Zero carbon buildings refurbishment - a Hierarchical pathway. Renew Sustain Energy Rev 2011; 15: 3229–3236.

50.

Hashemi

Khatami

. The effects of air permeability, background ventilation and lifestyle on energy performance, indoor air quality and risk of condensation in domestic buildings. Sustainability 2015; 7: 4022–4034.

51.

HM Government . Conservation of fuel and power- approved document L- Volume 1: dwellings- 2021 edition incorporating 2023 amendments. The building regulations 2010. HM Government, 2021. https://www.gov.uk/guidance/building-regulations.

52.

Tahmasebi

Banihashemi

Hassanabadi

. Assessment of the variation impacts of window on energy consumption and carbon footprint. In: Procedia engineering. Amsterdam, The Netherlands: Elsevier Ltd, 2011, pp. 820–828.

53.

Hee

Alghoul

Bakhtyar

, et al. The role of window glazing on daylighting and energy saving in buildings. Renew Sustain Energy Rev 2015; 42: 323–343.

54.

Rahimian

Duarte

Iulo

. A machine learning approach for mining the multidimensional impact of urban form on community scale energy consumption in cities. Design Computing and Cognition. Springer, 2020, pp. 607–624.