The UK’s hidden carbon footprint
By Anne Owen, Lena Killian and Rutger Hoekstra
When we talk about the UK’s progress towards net zero by 2050, the headline number is usually territorial emissions – the greenhouse gases released within UK borders. But that number tells only part of the story.
Crucially, the territorial emissions number does not include significant pieces of the UK’s carbon footprint, including:
The consumption-based emissions account (CBA) was developed to fill gaps in emissions reporting. It includes aviation, shipping and UK-registered businesses located overseas, adds emissions associated with imports and removes emissions associated with exports. This helps to reveal whether the UK’s apparent progress is due to real change, or simply the result of shifting our production (and emissions) to other countries. The Climate Change Committee’s Seventh Carbon Budget recommends continuing to report these consumption-based figures.
Figure 1 shows that both the UK’s Territorial and Consumption based emissions accounts are decreasing, but the pattern in the consumption account is much more closely related to fluctuations in GDP. The consumption account contains both domestic emissions and emissions emitted overseas in the supply chains of imported goods. Since 1990, the portion of the UK CBA that is imported increased from 34% to 61% in 2022. This highlights the importance of accurate, reliable and timely trade data.
CBA are modelled using large databases which detail economic transactions between global industries. Known as Global Multi-Regional Input-Output databases (GMRIO), these systems can be used to allocate the UK’s share of Global emissions that are associated with the final demand of products bought by UK consumers. The University of Leeds constructs the UKMRIO database which is used to produce the UK’s consumption-based emissions account, an Official Statistic reported by Defra[1] (Department for Environment, Food & Rural Affairs).
While data on transactions between domestic industries is well reported in UK economic accounts, data on imports and exports misses much of the detail required to fully estimate traded emissions. The economic accounts record how much industries, households and Government spend on imported goods, but not where in the world these goods are sourced from. To construct the UKMRIO, the UK economic accounts are supplemented with trade data from one of the GMRIO databases.
Figure 2 shows the results from fourteen years’ of UKMRIO databases produced by University of Leeds. The variation in emissions can be explained by changes in the source of trade data used and the methods used to construct the UKMRIO[2]. Our researchers were concerned about the 2022 estimate calculated by the initial 2025 release of the UKMRIO. It was expected that the CBA would follow GDP change but GDP in 2022 did not show the increases observed in this CBA estimate. Post and during the COVID-19 pandemic, international supply chains have been disrupted, and we are experiencing higher levels of inflation globally. It is important that the trade data used in the UKMRIO is up to date and not based on proportions and prices from pre-2020. Our team were concerned that the trade data from the EXIOBASE 3.8.3 GMRIO database used in the UKMRIO relied on old trade patterns and prices from 2019-2022.
With ESCoE funding, we set out to test how using different Global Multi-Regional Input-Output (GMRIO) databases[3] affects the UK’s consumption-based emissions accounts.
One of the databases we examined was the FIGARO GMRIO, produced by Eurostat. This uses actual rather than estimated data from 2020-2022. When we compared FIGARO’s 2022 results with earlier estimates based on the EXIOBASE database, we found a key difference: FIGARO did not show the same post-2020 spike in emissions (see figure 3). This suggested that FIGARO may provide a more accurate picture of the UK’s recent carbon footprint. However, FIGARO is only available from 2010 onwards. This meant that we had to decide when to switch from the old trade database to FIGARO. Since the EXIOBASE trade database is considered to be accurate pre-2019, we could choose any point between 2010 and 2019 to switch over to FIGARO. Figure 3 shows that EXIOBASE-based estimates and FIGARO-based estimates are very similar in 2015. Therefore, we decided to make the switch to FIGARO for the years 2015-2022.
The UKMRIO is not only used to produce the consumption-based account for emissions. It is also used to determine the UK’s material footprint. The material footprint calculates the total mass of extracted metal ore, fossil fuel material and non-metallic minerals as well as materials associated with biomass that are required to meet UK demand for goods and services. The calculation relies on detailed data from the agricultural and mining industrial sectors.
Here we found a limitation in FIGARO: while the UKMRIO database has four separate entries for the mining of coal, oil, metal ore and minerals, FIGARO has a single mining and quarrying sector. This is less of a concern for emissions accounting but critical for the material footprint.
As shown in Figure 4, using FIGARO’s raw data gave a much higher material footprint than EXIOBASE. The second part of the ESCOE funded project allowed us to produce an aggregated version of FIGARO[4] which split the mining sector into four entries. This adjustment brought FIGARO’s material footprint results in line with the EXIOBASE initial estimate.
“The consumption-based emissions data produced by Leeds is a unique resource for understanding the UK’s climate impact beyond its territorial borders. As domestic emissions decrease, emissions from imports are an increasingly important part of the UK’s overall contribution to climate change. No other dataset provides the same completeness or level of detail as CBA produced by Leeds. The improvements made to the data in recent years have given us more confidence in their accuracy and robustness, which is something the Climate Change Committee has called for in their recent Seventh Carbon Budget publication.” – Luke Maxfield, The Climate Change Committee
“While the UK’s emissions targets are set on a territorial basis, the consumption-based emissions provide helpful insight to supplement these. As they cover emissions associated with the consumption of goods and services by households in the UK, including emissions in imported good, the measure highlights the global considerations necessary when considering the UK’s contribution to climate change. It is important that the methodology and data sources used to produce these consumption- based emissions are regularly reviewed to ensure the best available data sources and methods are used.” – Defra Statisticians
The ESCOE project gave us the opportunity to fully explore the implications of switching to a new source of trade for the UKMRIO database.
The result is the 2025 release of the UKMRIO database and its associated consumption-based accounts for both emissions and materials – a clear improvement from previous releases.
The work supports climate policy and gives a clearer understanding of the UK’s contribution to climate change. It also responds directly to the Climate Change Committee’s call for more robust and reliable consumption-based emissions data.
ESCoE blogs are published to further debate. Any views expressed are solely those of the author(s) and so cannot be taken to represent those of ESCoE, its partner institutions or the Office for National Statistics.
[1] https://www.gov.uk/government/statistics/uks-carbon-footprint
[2] The reduction in CBA estimate between 2022’s release and 2023 was due to an improvement in the method used to remove taxes from imports
[3] We assessed the Gloria GMRIO (https://ielab.info/labs/ielab-gloria) produced by the University of Sydney, the FIGARO GMRIO (https://ec.europa.eu/eurostat/web/esa-supply-use-input-tables/database) produced by Eurostat and the ICIO GMRIO (https://www.oecd.org/en/data/datasets/inter-country-input-output-tables.html) produced by the OECD
[4] The consumption-based emissions results shown in figures 1 and 3 use this disaggregated version of FIGARO