Energy consumption of ICT

DOI: https://doi.org/10.58248/PN677 

Information and Communication Technology (ICT) encompasses data centres, communication networks (wired and mobile) and user-devices (such as smartphones and laptops). Energy is used in the production and powering of ICT equipment and in the provision of digital services.

There is limited evidence on the energy use of ICT, and significant uncertainty in existing estimates. There are no peer-reviewed studies of energy use of the ICT sector in the UK specifically. Estimates for global ICT energy use are based on industry energy use reports and market sales of ICT equipment. These estimates suggest that ICT (excluding TVs) accounted for 4-6% of all electricity used globally in 2020 and that this is likely to increase over the next 5-10 years.

Future ICT energy use will be affected by trends in user demand and energy efficiency, as well as by growing applications such as crypto-mining and machine learning. The effects of these trends are highly uncertain.

Key points:

• Energy efficiency improvements allow the same ICT tasks to be performed using less energy. Some stakeholders note that this has resulted in energy use of ICT remaining relatively flat in the last decade despite the growth in demand. Others suggest that improvements in energy efficiency can themselves cause increased demand and hence a focus on future improvements in energy efficiency without additional regulation may not reduce energy use.
• Because the ICT sector is highly electrified, the carbon footprint associated with ICT energy use depends strongly on the global electricity mix. Some estimates suggest that ICT (including TVs) constitutes 2-3% of greenhouse gas (GHG) emissions globally.
• The ICT sector is increasingly powered by renewable electricity sources, but experts have said that more work is needed to decarbonise the electrical grid and ICT supply chains to ensure the sector reaches climate targets. One study by the International Telecommunications Union estimated that ICT emissions could fall by 50% from 2020 to 2030, but they note that this relies on a very challenging rate of grid decarbonisation.
• Estimates suggest that user devices consume more energy than networks and data centres combined. One study attributed 60% of use-phase electricity consumption of the ICT sector to devices (excluding TVs), with networks and data centres accounting for 20% each.
• Data centres use a significant proportion of energy for cooling the ICT equipment. Cooling costs can be reduced by consolidating smaller data centres into larger ones and by building data centres in cooler climates.
• Fibre networks and 5G are expected to be more energy efficient than their older counterparts (copper networks and 3G and 4G). However, the effect of 5G on overall energy use of communication networks is contested and uncertain.
• There have been significant improvements in power management for user devices, with energy saving techniques now built into laptops and PCs.
• Many stakeholders suggest that more data should be collected and analysed on energy use of ICT. Barriers to collecting this information include proprietary data and the difficulty of disaggregating the energy used by ICT from a company’s overall energy use.
• Energy efficiency standards and regulations are in place across the ICT sector. However, stakeholders have pointed to a lack of enforcement mechanisms for standards. ICT companies have also put forward their own climate targets, but some of these have been criticised for not being sufficiently detailed.
• Some stakeholders note that greater understanding of the energy and environmental impacts of ICT could lead to changes in user behaviour. The Energy Savings Trust reported that consumer electronics account for around 6% of household energy bills. It provides advice on how individuals can reduce their energy use of ICT.
• The cost associated with running appliances in the home (including electronic devices) is likely to be of increasing concern to consumers due to the rising costs of living in the UK (see CBP-9428, Rising cost of living in the UK).

Acknowledgements

POSTnotes are based on literature reviews and interviews with a range of stakeholders and are externally peer reviewed. POST would like to thank interviewees and peer reviewers for kindly giving up their time during the preparation of this briefing, including:

• Emma Fryer, techUK
• Adam Young, techUK*
• George Kamiya, International Energy Agency*
• Jens Malmodin, Ericsson*
• Anders Andrae, Huawei*
• Professor Chris Preist, University of Bristol*
• Dr Daniel Schien, University of Bristol*
• Professor Eric Masanet, Northwestern University
• Dr Jonathan Koomey, Koomey Analytics*
• Dr Arman Shehabi, Lawrence Berkeley National Laboratory*
• John Booth, Carbon3IT*
• Dr Reinhard Gruenwald, Office of Technology Assessment at the German Bundestag
• Dr Claudio Caviezel, Office of Technology Assessment at the German Bundestag*
• Simon Hinterholzer, Borderstep Institute*
• Professor Gordon Blair, University of Lancaster*
• Dr Gabrielle Samuel, Kings College London*
• Dr Lucas Somavilla, University of Oxford*
• Professor Emma Strubell, Carnegie Mellon University*
• Professor Rabih Bashroush, University of East London
• Dr Umaima Haider, University of East London*
• Adam Turner, Department for Environment, Food and Rural Affairs*
• Dan Craddock, Department for Digital, Culture, Media and Sport*
• Frankie Evans, HM Treasury*
• Dan Bailey, Cabinet Office*
• Michael Salter-Church, Openreach*
• Abby Chicken, Openreach*
• Jonathan Chown, Openreach*
• Trevor Linney, Openreach*
• Andrew Hatton, Greenpeace*
• Rashik Parmar, IBM*
• John Midgley, Amazon Web Services
• James Johns, Amazon Web Services
• Gareth Elliott, Mobile UK*
• Adam Higgitt, Broadband Stakeholders Group
• Ofcom*
• Mark Acton, Ekkosense*
• Conor Griffin, DeepMind*
• Simon Hansford, UKCloud*
• Members of the POST Board*

*denotes people and organisations who acted as external reviewers of the briefing.