Government electricity · Estimate: possible, not certain
Lower electricity use in the government sector
Electricity per government subscriber: 129.3 MWh in 2024
If measures that cut building energy use in other countries were applied in the government sector, its use would fall each year by about
8,630 GWh Estimate: possible, not certain
enough for 464 thousand homes for a whole year
Range in the evidence: 7,129 to 10,506 GWh
6.0 million tonnes less CO2 emissions (Range in the evidence: 4.9 million to 7.3 million)
Range in the evidence: 383 thousand to 565 thousandWhat the numbers show
Published figures show the average government subscriber used 129.3 MWh in 2024, against 34.0 for commercial . The whole government sector used 37,520 GWh in 2024 [1].
Subscriber categories differ in size and activity, so the gap does not measure efficiency on its own.
Government: 2021: 133.4 → 2022: 125.9 → 2023: 123.9 → 2024: 129.3
Commercial: 2021: 27.2 → 2022: 30.0 → 2023: 35.0 → 2024: 34.0
What was tried elsewhere: comparison
Replacing lamps with LED
LED lamps use less electricity than old lamps and give the same light.
27% to 29% of lighting energy only [5] — Three federal buildings, USA, 2014
Strength of evidence (how far to trust it): Low
Improvements by a specialist firm
Energy-service contract: a specialist firm carries out the energy upgrades, and its fee is tied to the measured drop in energy use.
Median 19% to 28% of baseline energy [2] — Local government buildings, USA, 1996–2008
Strength of evidence (how far to trust it): Medium
Swapping air-cooled for water-cooled chillers
A chiller is the unit that cools the building; the water-cooled type uses less electricity than the air-cooled type.
Electricity fell 17.5% in six months [6] — Hotel and convention centre, Indonesia, 2020–2022
Strength of evidence (how far to trust it): Low
Tuning how an existing building runs
Checking the building systems (cooling, ventilation, controls) and adjusting them to run as designed.
Median 16% of whole-building energy [4] — Existing buildings, USA, data 2009
Strength of evidence (how far to trust it): Medium
How it was applied in published programmes
- Each building's use was measured monthly, and buildings were ranked by use per square metre [7].
- Building systems (cooling, lighting, pumps) were checked and the biggest users identified [7].
- Low-cost changes were made first, larger ones in stages, and building operators were trained [7].
- Monthly measurement tracked results, compared with the year before the changes [7].
Linked target
“… reduce carbon emissions by 278M tons annually by 2030”[source]
Published target: 278 million tonnes a year (2030)
This target is not measured directly by these figures.
Arithmetically about 2.1% of the amount in the published target (an approximate sum, not a measure of contribution to the target). Estimate: possible, not certain