Article 05: Height, Age and the Fuel Question: What Physical Characteristics Really Do

A series mining the PhD thesis "London and UK Office Buildings: Investigating Energy Use and Landlord-Tenant Influences" (Azhari, 2025).

Key takeaway. Across thousands of Greater London offices, a small number of physical characteristics leave a measurable energy signature. Height above six storeys, post-1980 construction and electrified heating are the most useful for portfolio triage. None of them is decisive on its own.

Chart

Single-variable effects on median EUI

Each comparison shifts the median by between 13 and 38 per cent; none is decisive on its own.

A triage rule before the details

Most asset managers know the rough shape of their portfolio long before any data analysis. The largest buildings, the newest builds, the City addresses and the all-electric assets are usually the ones with the heaviest bills. The Greater London empirical data confirms much of that intuition, but with a couple of important caveats that change where the marginal pound of retrofit should go.

The triage rule that emerges from the thesis is: prioritise large, post-1980, six-storey-plus offices in central boroughs, but do not expect any one of those signals to swing the EUI by more than a fifth on average. Stack three or four of them together and you are looking at a high-intensity asset. Have only one and the picture is more ambiguous than the conventional wisdom would suggest.

Borough and tenancy: the proxies that hide more than they show

Median electricity EUI is highest in the City of London, followed by Westminster. Median gas EUI follows a different pattern: Islington has the lowest, and the City and Westminster also sit low on gas. The combined picture suggests that the central boroughs concentrate premium high-rise, air-conditioned, electrified offices, while less central boroughs lean to smaller, older, gas-heated stock.

Chart

Median electricity use intensity by borough

Electricity EUI peaks in the City of London and Westminster, the premium, high-rise, air-conditioned core.

• 1Kensington and Chelsea108
• 2Hammersmith and Fulham95
• 3Camden130
• 4Islington125
• 5Hackney100
• 6Westminster185
• 7City of London210
• 8Tower Hamlets175

Figure 4-1 (redrawn from the source chart). Median electricity EUI across London boroughs (kWh/m² per annum).

Chart

Median gas use intensity by borough

Gas EUI follows a different pattern: it peaks in outer boroughs and is lowest in Islington, the City and Westminster.

• 1Kensington and Chelsea88
• 2Hammersmith and Fulham95
• 3Camden95
• 4Islington80
• 5Hackney90
• 6Westminster85
• 7City of London86
• 8Tower Hamlets88

Figure 4-2 (redrawn from the source chart). Median gas EUI across London boroughs (kWh/m² per annum).

Borough, in other words, is a proxy. It does not cause the energy use. It is a label for the bundle of attributes (size, height, fuel, tenant mix, premium servicing standards) that tend to co-occur in a given postcode. Treat it as a triage signal rather than an analytical variable.

Tenancy follows a similar pattern. Single-tenant and multi-tenant offices show no statistically significant difference in median gas EUI. Multi-tenant offices in mixed-use buildings (those that share a footprint with shops, gyms, residential or other uses) record a median electricity EUI about 13 per cent above the other two groups. The likely explanation is that the shared-use activities (food and beverage, gyms, late-evening retail) inflate plug load and ventilation demand without the energy boundary being clean enough to separate them. It is a reminder that the Self-Contained Unit boundary discussed in Article 4 helps but does not perfectly isolate office activity from everything else.

Height: the six-storey threshold

Height is the cleanest physical signal in the dataset. The electricity EUI distribution shows a step at around six storeys: offices below the threshold sit in one regime, those at or above in another. This is consistent with Godoy-Shimizu and colleagues (2018), who reported a similar threshold at five to six storeys in an earlier study.

Several mechanisms are plausibly at work. High-rise offices have more exposed envelope above the surrounding stock and therefore higher heat-loss and solar-gain loads. They are more likely to have glazed façades, which increases both winter heat loss and summer cooling. They are more likely to carry full mechanical ventilation and air-conditioning rather than openable windows. They are more likely to be premium-fit-out commercial space with longer hours and denser plant. Disentangling these mechanisms is beyond what 3DStock can do; what the model shows is that the threshold exists and is stable across the sample.

Gas EUI does not show the same threshold effect. The intuitive explanation is that the high-rise inflection is largely an electric one: cooling, ventilation, lifts, façade lighting, plug load. Gas demand is dominated by heating, and heating efficiency depends more on plant type and control than on building height.

Age: the post-1980 inflection

Construction date sits squarely in the conventional wisdom about energy use, but the empirical pattern is more counter-intuitive than the wisdom suggests. Offices built after 1980 record a median electricity EUI 38 per cent higher than the 1945 to 1980 cohort and 25 per cent higher than pre-1914 stock. Gas EUI shows no significant variation across age bands.

Why are newer offices using more electricity? The likely answer combines several effects. Post-1980 office construction shifts toward deep-plan, fully serviced, fully air-conditioned typologies. Building regulations from the 1980s onwards drove up fabric thermal performance, which reduced heating demand (the gas side) but did not constrain electrical demand for ventilation, cooling and IT loads. Lighting, equipment and ICT loads grew over the same period. And many post-1980 offices were designed and operated for trading floors, financial services and other intensive activities that older stock often does not host.

The implication for retrofit is uncomfortable. Improving the EPC band of a 1990s curtain-walled high-rise via more insulation and better glazing does not necessarily attack the dominant energy-use driver in that asset. Controls, operating hours, IT load management, lighting and HVAC optimisation are more likely to move the bill. Article 3 and the qualitative evidence in Article 8 both reinforce that conclusion from different directions.

The 2026 National Buildings Database (DESNZ), which the author contributed to, gives the national age picture. 34 per cent of office premises in Great Britain (20 per cent of office floorspace) sit in the pre-1919 band. Another NBD finding is that 41 per cent of office premises are in conservation areas, listed buildings, or both. Heritage constraint is therefore not a minority concern. For roughly two in five office premises, retrofit options are bounded by planning and heritage rules before any technical or economic assessment begins.

Heating fuel and air-conditioning

EPC records provide an indication of primary heating fuel for the rated subset of offices. Grouping by primary fuel produces an expected pattern with one interesting wrinkle.

Electrically heated offices use 17 per cent more electricity than gas-heated offices and 35 per cent less gas. So far so intuitive. The wrinkle is that electrically heated offices still record a median gas EUI of 61 kWh per square metre per annum. That is not a rounding error. It is real on-site gas demand, presumably for catering, secondary heating, swimming pools in mixed-use buildings, or in some cases small CHP and fuel-cell installations that consume gas to make electricity. The EPC label of primary fuel is therefore a coarse signal, and any retrofit plan based purely on the headline fuel could easily miss meaningful pockets of gas demand.

Air-conditioning, also recorded on the EPC, behaves more predictably. Air-conditioned offices show a 17 per cent higher median electricity EUI than non-air-conditioned offices. Gas EUI shows no significant difference. The growth in the air-conditioning market in the UK over the past two decades (AMA Research, 2021) suggests this effect will become more pronounced rather than less, particularly as climate change pushes summer cooling needs up.

Putting it all together

Each of these characteristics on its own moves the EUI dial by between roughly 10 and 40 per cent, depending on the comparison being made. Stack three or four together (a high-rise, post-1980, air-conditioned, electrically heated office in a central borough) and you have a typical premium asset that uses noticeably more electricity than the rest of the stock. Take any one of them away and the picture loses sharpness.

This is exactly the pattern that Article 3 captures statistically: even taken together in a linear regression, these characteristics explain only 18 per cent of the variation in electricity EUI. The single-variable effects are real and useful for triage. They are not enough to predict energy use for any given building, and they are not enough to design a policy that reliably bites on the highest-intensity assets without also dragging in low-intensity ones in the same category.

For portfolio managers the takeaway is practical. Sort your stock by the stack of attributes above. Focus first on the assets that pile multiple high-intensity signals together. Resist the temptation to assume that age, EPC band or fuel alone is enough to set the order of work. And remember, from the previous article, that the marginal kilowatt-hour is more often saved in operations and controls than in fabric.

Limitations

Each characteristic is analysed in isolation in the thesis. Real offices combine height, age, fuel and tenancy in correlated ways, so single-variable effects carry confounding. EPC primary heating fuel is a proxy for HVAC type. The data does not specify whether a building runs boilers, heat pumps or electric resistance. Age bands rely on GIM International UK Buildings data, which estimates rather than measures construction date for many buildings. The sample sizes for some sub-groups (seven-plus storeys, recent EPC band A) are small enough to limit the strength of group-level claims. The 2017 cross-section pre-dates significant electrification trends, so the fuel-mix findings are a baseline rather than a forecast.

References

• Azhari, R. (2025) London and UK Office Buildings Investigating energy use and landlord/tenant influences. Doctoral thesis (Ph.D), UCL (University College London). URL: https://discovery.ucl.ac.uk/id/eprint/10204821/

• Godoy-Shimizu, D., Steadman, P., Hamilton, I., Donn, M., Evans, S., Moreno, G., and Shayesteh, H. (2018). Energy use and height in office buildings. Building Research and Information. Available at: https://doi.org/10.1080/09613218.2018.1479927

• Liddiard, R., Godoy-Shimizu, D., Ruyssevelt, P., Steadman, P., Evans, S., Humphrey, D., and Azhari, R. (2021). Energy use intensities in London houses. Buildings and Cities, 2(1), 336. Available at: https://doi.org/10.5334/bc.79

• Thomas, L. E. (2006). Energy use in office buildings and HVAC system efficiency.

• AMA Research (2021). Ventilation and Air Conditioning Market Report UK 2021-2025. Available at: https://www.amaresearch.co.uk

• Better Buildings Partnership (2020). Real Estate Environmental Benchmark. Available at: https://www.betterbuildingspartnership.co.uk/our-priorities/measuring-reporting/real-estate-environmental-benchmark

• Evans, S., Fennell, P., Humphrey, D., Liddiard, R., Oraiopoulos, A., Palmer, J., Ruyssevelt, P., Shamsi, H., Amrith, S., and Steadman, P. (2026). National Buildings Database Phase 2. Department for Energy Security and Net Zero. Available at: https://www.gov.uk/government/publications/national-buildings-database-phase-2-understanding-great-britains-buildings

Read next

• Article 3: Eighteen Per Cent. What Building Characteristics Can and Cannot Explain.

• Article 10: Time to Retire ECG-19? Updating Office Benchmarks.

About this series

This article is part of a fifteen-piece series adapting the 2025 PhD thesis "London and UK Office Buildings: Investigating Energy Use and Landlord-Tenant Influences" (Azhari, 2025) for a mixed academic and industry readership. The empirical findings draw on the 3DStock model of 6,038 office Self-Contained Units in Greater London with metered energy data for 2017, supplied by BEIS under a data-sharing agreement, alongside the Better Buildings Partnership Real Estate Environmental Benchmark. The qualitative findings draw on semi-structured interviews with seven major UK property organisations, conducted during the 2021 lockdown. Interviewees and their organisations are anonymised by role and organisation type. Please cite the original thesis for academic use.

Author. Rayan Azhari completed his PhD at the UCL Bartlett School of Environment, Energy and Resources in 2025, supervised by Paul Ruyssevelt and Kathryn Janda. The research was supported by the EPSRC Centre for Doctoral Training in Energy Demand (LoLo) and UK Research and Innovation through the Centre for Research into Energy Demand Solutions.

Other articles in the series. Article 1 The 30/85/89 Problem; Article 2 Why EPCs Do Not Tell You How Much Energy a Building Uses; Article 3 Eighteen Per Cent; Article 4 Mapping the Stock; Article 5 Height, Age and the Fuel Question; Article 6 The Split-Incentive Problem; Article 7 Green Leases and Service Charges; Article 8 From 38 to 73 Per Cent Energy Savings; Article 9 NABERS for Britain; Article 10 Time to Retire ECG-19; Article 11 Can London Speak for England and Wales; Article 12 The Hybrid-Work Footprint; Article 13 Why I Used Linear Regression Over Random Forest; Article 14 Vertical Postcodes; Article 15 What Is a Building?