Energy, Scottish Economy

Decomposing Glasgow’s CO₂ Emissions: Key Insights

Targets, targets, targets

We are now a month out from the Scottish Government’s decision to drop their interim target of a 75% reduction in greenhouse gas emissions by 2030.

With so much focus on emissions at the national level and Glasgow City Council’s own 2030 net carbon zero target looming, we felt this would be a great time to share some of our latest research, as part of the Glasgow Environmental Monitoring of Indoor and Outdoor Air (GEMINOA) project.

So, if you’re interested in understanding more about Glasgow’s emissions are calculated and the data used to benchmark progress against targets, you’ve come to the right place!

So many measures, so little time

It may come as no surprise that there are many reported emissions figures available. Three widely-used measures at the UK level are “territorial”, “residence” (production) and “footprint” (consumption). Each measure draws a different circle around which emissions are in and which are out. A clear description of all three measures is available from the Office for National Statistics. Having multiple emissions measures is both a blessing and a curse. No single measure will tell us all that we need to know about the emission of greenhouse gasses. Different perspectives can help in evaluating economic and environmental policy – depending on which emissions are being impacted.

The downside is that it can be downright confusing for academics, engaged readers and policy makers alike to know which measure is being used, where they differ in terms of what elements are included in each and which are not, and how policy actions will impact on emissions under each measure.

Adding to the potential for confusion, “emissions” can be a rather vague term, frequently used to refer to one or more of the seven main greenhouse gasses which generally cover carbon dioxide (CO2), methane (CH4), nitrous oxide (N2O), hydrofluorocarbons (HFC), nitrogen trifluoride (NF3), perfluorocarbons (PFC) and sulphur hexafluoride (SF6).

The focus of this blog is to understand “territorial” CO2 emissions estimates of local authority emissions produced by the Department for Energy Security and Net Zero (DESNZ) and used by GCC when reporting on its net zero carbon target.

Understanding the DESNZ measure

UK local authority annual greenhouse gas emissions estimates are published by DESNZ, formerly the Department for Business Energy and Industrial Strategy (BEIS). These are published around 18 months after the year to which they relate ends, so that the latest data released in 2023 cover the period between 2005 and 2021 and provide estimates of three major greenhouse gasses: CO2, CH4 and N2O. Luckily for us, CO2 emissions are presented separately. Emissions figures for 2022 are likely to be published shortly.

What’s included in the DESNZ measure?

UK territorial estimates account only for emissions produced within its borders. Therefore, emissions from international aviation and shipping, as well as those from the production of imported goods -regardless of whether they were consumed by UK households, firms, or the government -are excluded.

At the local authority level, the approach mirrors that of national data: Glasgow’s estimates are intended to reflect emissions produced within the boundaries of the local authority. This is true for several sources of emissions such as agriculture, large industrial installations, as well as land use, land use change and forestry. There is, however, a need to account for emissions related to the production of electricity which is consumed within a local authority’s boundaries (irrespective of where that production occurs).

This gives rise to the “end user” allocation of energy emissions based on consumption – which appears to have a credible justification, in that it avoids penalising local authorities which are merely producing energy which is consumed elsewhere in the UK. Much of this blog is focussed on the significant consequences of this method (spoiler alert!).

Transport emissions are also estimated using “end-user” allocation – so that all emissions associated with road and rail use in a local authority is allocated to that area – while waste management emissions are allocated based on where the waste is produced rather than where it is stored.

What has been happening to Glasgow’s CO2 emissions?

Glasgow City

Figure 1 shows CO2 estimates for Glasgow City from the latest data covering the period between 2005 and2021. First, we note that Glasgow has made substantial progress in reducing its CO2 emissions. The city’s emissions dropped from 4,206 kilo-tonnes of carbon dioxide equivalent (ktCO2e) to 2,435 ktCO2e, approximately a 42% decrease. For comparison, Scotland as a whole experienced a reduction of 39.1% over the same period.

Figure 1: Grand total CO2 emissions for Glasgow, 2005-2021

Source: DESNZ

Glasgow’s decarbonisation journey follows a clear downward trend with several significant drops punctuated with minor increases. The pandemic restrictions on travel throughout 2020 led to a sharp reduction in emissions – as would be expected. Figures for 2021 show the post pandemic rebound as restrictions were partially lifted. The progress made in reducing reported emissions is clear. However, to determine whether the pre-2020 trend might continue, it is essential to understand the drivers behind these reductions. We must also consider whether these factors are unique to Glasgow or if they are observable at the regional/national level.

Glasgow City Region

Figure 2 shows the change in CO2 emissions across all Glasgow City Region (GCR) local authorities, with values indexed to 2005 as the base year (2005=100). The data reveals a decline in emissions for all local authorities over this period. Despite some localised variations around 2019, which will be discussed later, the overall downward trend across GCR is consistent with that observed in Glasgow.

The broad similarity in emission trends across all local authorities suggests that regional and national factors are likely key drivers of the observed reductions.

Figure 2: Indexed GCR CO2 emissions 2005-2021 (2015=100)

Source: DESNZ and FAI calculations

What is driving Glasgow’s emissions reductions?

First, some data transformation (skip to next section for analysis)

The DESNZ data provide emissions estimates for a range of “sectors” including industrial, commercial, public, domestic, transportation, agriculture, land use, land use change and forestry, and waste management. These do not map to industrial sectors of the economy as typically defined in, for instance, the System of National Accounts, on which economic and employment data is gathered, but rather correspond to those typically seen in emissions inventories.

Emissions from energy use (electricity, gas, and “other”) are reported for the sectors shown in the first four rows of the left table shown in Figure 3.

Our analysis required disaggregating the DESNZ categories to separate emissions from the consumption of energy (electricity and gas). The disaggregation is shown in the right-hand table in Figure 3.

Figure 3: Disaggregation of DESNZ categories to separate level of energy consumption from emissions

Source: FAI

In addition to the emissions data, we used DESNZ subnational electricity and gas consumption data to let us isolate whether changes in emissions from either category was due to changes in consumption alone, or other factors as we will discuss shortly.  The same process was used for road transport emissions using local authority Department for Transport (DfT) road traffic statistics.


Figure 4 shows the cumulative change in CO2 emissions in Glasgow from 2005 to 2021, broken down by the sources shown in Figure 3. Focusing on the combined green bars, it’s clear that the key driving force behind the overall decrease has been reductions in electricity emissions. The 72% drop – roughly 1295 ktCO2e – in electricity emissions over the period is responsible for 73% of the total reduction seen across all sources.

Using the local authority electricity consumption data, it was possible to go further than the DESNZ stats and attribute the proportion of the reduction that was due to lower domestic and non-domestic consumption of electricity (i.e., a reduced overall consumption of electricity). These factors can be seen in the diagonal and vertical lined bars, respectively. Accounting for these factors means that any additional changes in emissions related to electricity end-user consumption in Glasgow must have come from the decline in the emissions intensity of electricity consumption.

Using this method, we can see that 47% of the total reduction in CO2 emissions is driven by a reduction in the emissions intensity of electricity consumption i.e., the increasing share of renewables in the UK’s generation mix. For more detail on the evolution of the generation mix (see DUKES for more detail).

Figure 4: Glasgow CO2 emissions reduction by source (ktCO2e), 2005-2021

Source DESNZ and FAI calculations 

Figure 5 shows the percentage reduction in total CO2 emissions driven by the adoption of renewables for each of the local authorities within Glasgow City Region.  As would be expected for a national-level driver of emissions reductions – the impact is relatively consistent across all GCR local authorities.

Figure 5: Share of total CO2 emissions reduction in each local authority in Glasgow City Region from lower emissions intensity of electricity consumption between 2005 and 2021 (%)

Source DESNZ and FAI calculations 

The primary factor in the decarbonisation journey of all Glasgow City Region (GCR) local authorities has been the reduction in electricity emissions, largely influenced by UK-wide decarbonisation and renewable energy policies.

The data highlights the problem with setting local government targets based on CO2 estimates so heavily impacted by national factors—a direct consequence of “end-user” allocation. While local governments can undoubtedly take actions that affect emissions, many of the crucial levers needed to sustain or increase the current decarbonisation rate are beyond their control.


Figure 6 shows CO2 emissions from gas, electricity, and road transport in Glasgow from 2005 to 2021. Compared to electricity, the reduction in gas emissions has been modest at 19%. While decreased consumption in both domestic and non-domestic sectors contributed to the reduction in electricity emissions, lower consumption was the primary factor in declining gas emissions. The upcoming release of UK local authority CO2 estimates, due next month, is expected to show even larger reductions in gas emissions. This anticipated decline is likely due to the milder weather in 2022, which was about 0.8 degrees Celsius warmer than 2021 on average across the UK, combined with changes in consumer behaviour prompted by high gas prices following Russia’s invasion of Ukraine.

Figure 6: Total electricity, gas, and road transport CO2 emissions (ktCO2e) in Glasgow, 2005-2021

Source DESNZ and FAI calculations 

Road Transport

Road transport accounts for 99% of transport emissions in Glasgow. Road transport has seen relatively modest reductions in emissions compared to the significant declines observed in electricity. Despite a 19% decrease since 2005, road transport emissions still account for about one third of Glasgow’s total 2021 emissions.

Reaching the net zero carbon target in Glasgow will require a significant step change in rate of road transport emissions reductions. A crucial aspect of this challenge involves understanding both the local and national contributions to the calculation and allocation of these emissions.

Road transport emissions included for Glasgow are calculated on an “end-user” basis, meaning the city is assigned a share of national road transport emissions proportionate to the vehicle miles travelled within its boundaries across three road types: urban, rural, and motorways. Importantly, this method allocates emissions based solely on where the miles are driven, regardless of the journey’s origin or destination.

What is particularly striking about the 19% decrease in road transport emissions is that it occurred alongside a relatively consistent increase in the number of miles driven in Glasgow, as shown in Figure 7.

Figure 7: Annual traffic by all vehicles in Glasgow City in vehicle miles (millions), 2005-2022

Source: Department for Transport

From 2005 to 2019, miles driven in Glasgow grew by around 11%. Pandemic restrictions in 2020 can be seen to have a significant impact in 2020 with a partial rebound occurring in 2021 and 2022 – though miles driven in 2022 remain below the pre-pandemic figures. We will have to wait until later this year when 2023 traffic data are released to assess whether there has been a shift in driving patterns or if the recovery towards pre-pandemic trends is merely delayed.

Despite the increase in miles driven, emissions have decreased, indicating that CO2 emissions per mile are falling (represented by the solid blue bar in Figure 8). This improvement is largely due to an increasing proportion of the vehicle fleet meeting stricter emissions regulations. In 2013, about 50% of the UK’s petrol car fleet complied with the Euro 4 emissions standard. By 2020, this figure had risen to just under 90%, according to the National Atmospheric Emissions Inventory.

Figure 8: Decomposition of cumulative change in emissions in Glasgow City by source – road transport emissions highlighted (ktCO2e), 2005 to 2021

Source DESNZ and FAI calculations 

There are two important questions here. First: how to reduce emissions from transport in the city? And second, how are those changes reflected in the currently adopted DESNZ measure?

To answer the second question, it’s critical to understand how local authority specific data is combined with devolved administration/ national data and international estimates of vehicle emission factors to produce road transport emissions estimates.

In exploring the DESNZ methodology on road transport emissions, we show in Figure 9 how local authority road traffic data is combined with national/regional data on vehicle fleet composition and detailed international emissions to arrive at emissions estimates for urban, rural, and motorways in Glasgow annually.

Figure 9: Road transport emissions methodology

Source: DESNZ and FAI 

As it stands, the process of estimating road transport emissions involves integrating diverse data sets, which complicates the understanding of where the effective levers for change lie. Currently, emission estimates from road transport are not highly sensitive to local variation in Euro standards or the proportion of electric vehicles, as this data is not provided at local authority level. Instead, these factors are determined using nationwide or devolved administration-wide data from the DfT’s Automatic Number Plate Recognition (ANPR) system and vehicle licensing statistics (for those LAs in Scotland, a Scottish vehicle mix is used, rather than one for Glasgow). This presents a challenge for local governments aiming to reduce emissions through transport policies that encourage the adoption of electric vehicles for instance.  To the extent that there are differences in the vehicle fleet in each area, or the extent to which vehicles miles driven in an area are driven by locally-owned vehicles, this could lead to differences between “true” emissions form road transport and allocated emissions.

Key Takeaways

Policies promoting a shift toward electric vehicle fleets may not immediately impact emission estimates as expected. Over time, the increased proportion of electric vehicles will influence Scotland’s average fleet statistics, but the DESNZ approach to calculating emissions from road transport mean that the emissions reduction benefits will be distributed across miles driven across all local authorities. This suggests that it would be important to understand and manage expectations around the immediate effect of such policies on emissions estimates under this approach.

The DESNZ measure may need to evolve to incorporate local authority level data on vehicle fleets as an extension of the “end-user” allocation method. The current system would appear to fail to recognise – in the emissions metric – those local authorities that work to reduce the emissions intensity of their local vehicle fleets (such as promote electrification of private car, bus or other road transport use. Enhancing the sensitivity of emission calculations to local vehicle data could be one of the most effective levers through which distinct local authorities actions would reduce their allocated emissions.

Summary and next steps

Our research has primarily focused on closely examining the DESNZ “territorial” CO2 emissions estimates for Glasgow City and Glasgow City Region. We’ve found that national policies, specifically, electricity decarbonisation, has had an outsized impact on emissions since 2005. The large impact that the “end user” allocation of energy emissions raises questions about whether other emissions metrics may be more suited to evaluating local government policy actions aimed at driving decarbonisation.

Finding the right measure of emissions at a subnational level is however a tricky task. Perhaps a suite of measures might be needed to recognise the way in which policy actions could impact on emissions measures across the territorial, production and footprint perspectives. This is one of the next stages which we are looking at in this project.

The GEMINOA team is looking forward to an eventful year ahead. Our team will continue to delve deeper into emissions metrics comparing multiple measures (“territorial”, “footprint” and “real time modelled” estimates) at the local authority level and comparing specific environmental/economic policy impacts.

Don’t forget to check out the GEMINOA website or the GEMINOA project page to catch up on our previous blogs and follow along for our next output!



Jack is an economist at the Fraser of Allander Institute.

Grant Allan is a Senior Lecturer in the Department of Economics. Grant has research interests in applied regional economic analysis and modelling, particularly in the areas of energy and tourism.