Consumption-Based Emissions Reporting
Memorandum submitted by University of Surrey (CON 15)
· Consumption-based GHG emissions increased by 7% between 1990-2004, whereas on a production basis they fell.
· A robust methodology for consumption accounting is possible but robust datasets are not yet available. The UK should take a proactive role in developing and standardising robust datasets.
· Consumption and production based approaches should be adopted in parallel – taking either approach on its own would be a mistake.
· The main benefits of the consumption basis are that it provides a more complete picture of the UK's carbon exposure, it is arguably a more equitable form of sharing responsibility for carbon emissions, and it acknowledges in a more complete way the UK's historical contribution to climate change
· The main disadvantage of consumption based targets is the limited control that the UK has on emissions that arise abroad.
· There is a danger that pursuing one or two indicators at the expense of other issues (such as water) can have unexpected and undesirable consequences.
How do assessments of the UK’s GHG emissions differ when measured on a consumption rather than a production basis?
1. The difference between assessment of the UK’s greenhouse gas (GHG) emissions when measured on a consumption basis rather than production basis is illustrated in Figure 1  . UK emissions from the production perspective as reported to the UNFCCC  for the purposes of the Kyoto Treaty are shown on the lower (pink) line. Accounting in this way shows that UK GHG emissions were 15% lower in 2004 than they were in 1990. UNFCCC reporting excludes emissions due to international aviation and shipping. When these emissions are added to give us a more complete estimation of emissions from the production perspective, emissions are just 9% lower in 2004 than they were in 1990, as shown by the green line marked "Production perspective (according to Environmental Accounts)".
Figure 1: UK GHG emissions according to different accounting perspectives.
Source: Druckman and Jackson (2009b)
2. The graph shows that a marked difference is visible when we compare production perspective emissions with emissions from the consumption perspective (shown in the black line in Figure 1): consumption perspective emissions were 854mtCO2e in 1990, they fell to around 759mtCO2e in 1995, but since 1995 the trend has been rising, so that by 2004 consumption perspective emissions had reached around 914mtCO2e - an increase of about 7% over 1990 levels. Moreover, between 2000 and 2004 the rate of increase was around 3% per annum.
3. The difference between the production perspective and the consumption perspective is known as the ‘GHG trade balance’. A positive GHG trade balance confirms the hypothesis that a country’s consumption patterns are creating a ‘pollution haven’ effect abroad: in other words that the country is ‘exporting’ energy-intensive industries and importing the semi-finished or finished consumer products demanded by UK lifestyles. The graph shows that the GHG trade balance increased from 6% (45mtCO2e) in 1990 to 24% (179mtCO2e) in 2004 and therefore that the UK is effectively creating a pollution haven abroad.
Is it possible to develop a robust methodology for measuring emissions on a consumption rather than production basis and what are the challenges that need to be overcome to deliver this?
4. Estimating emissions on a consumption basis relies on having both robust methodology and robust data. The methodology used in this analysis (and other similar analyses) is known as Environmentally-Extended Input-Output (EEIO) and is well established (see for example Miller and Blair (2009)). Within EEIO methodology variations are possible with respect to how imports are accounted for. The simplest method is to assume that imported goods and services are produced using the same technology as the UK technologies, this is known as the Domestic Technology Assumption (Druckman et al. 2008). The next simplest method is known as Quasi-Multi-Regional Input-Output modelling: in this methodology the environmental intensity of goods is specific to each region in the model, but the ‘manufacturing recipe’ (as given by the Analytical Input-Output tables) is assumed to be as in the UK (Druckman and Jackson 2009a). In full multiregional modelling the manufacturing recipe is specific to each country/region in the model. It requires more data than QMRIO modelling. There are basically two types of multiregional Input Output model: those that model trade flows between all countries/regions in the model, and those that only model trade flows between the country of focus and its trading partners. The methodologies are standard, but obtaining robust data is challenging, and the number of countries/regions modelled varies - often due to data constraints.
5. Hence a key issue to consider is the robustness of datasets available for EEIO modelling. In essence three sets of data are required:
· Analytical Input-Output tables for the UK;
· Environmental accounts for the UK;
· Information on the GHG intensity of imports.
6. Analytical Input-Output tables for the UK are produced by the Office for National Statistics (ONS). Until recently accurate consumption accounts for the UK could not be produced because the most recent tables released by the ONS had been for 1995. However, in the summer of 2011 the ONS released Analytical Input Output tables for the year 2005. It is imperative that the ONS commits to producing analytical input output tables on a regular basis (one dataset at least every five years).
7. The UK produces annual Environmental Accounts. In order to be used with the Analytical Input Output tables it is necessary to match the sector categories: the sector definitions used in the Environmental Accounts are different from those used in the Analytical Input Output tables and accuracy is lost in the cross matching process. The Environmental Accounts categorisation is more appropriate for GHG reporting and so, ideally, the categories used for the Analytical Input Output tables would match those used in the Environmental Accounts.
8. In order to estimate emissions embedded in imports to the UK it is necessary to have environmental and economic data for each of our importing partners. This relies on having Input-Output tables and environmental accounts for each of our trading partners as well as information on trade flows. The standard database that has been used in multiregional Input-Output modelling to date has been, generally, the GTAP database produced by Purdue University (Dimaranan 2006). There are several problems with using this database: firstly it is not designed for use in input output studies, and secondly the robustness of the data must be questionable because the dataset is collated from donated data for each country. As a reward for donating data, free access to the GTAP dataset is given. However, other organisations are now appreciating the importance of input output data and the OECD, for example, are working on producing Input Output tables that promise a more robust approach.
9. The selection of which database to use is important as it influences the results. So if international consumption-based reporting were to be used in a formal sense then it would be important that all countries use an internationally agreed dataset (as well as, of course, internationally agreed methodology). That being said, however, evidence shows that use of different datasets with variations in methodology do not cause appreciable differences in findings, as shown in figure 2, which compares the results from various UK studies.
Figure 2. Comparison of UK CO2 consumption perspective emissions from different studies
Source: Druckman and Jackson (2008)
10. Figure 2 compares our results obtained using the QMRIO model by Druckman and Jackson with those from multi-regional studies of the UK by Wiedmann et al (2008) (for 1990-2004), Ahmad and Wyckoff (2003) (for 1995) and Peters and Hertwich (2008) (for 2001). Wiedmann et al’s (2008) estimates are higher than those from other studies, whereas Ahmad and Wyckoff (2003) give the lowest estimate. Druckman and Jackson’s QMRIO model is predicted to under-estimate CO2 emissions due to use of the UK Leontief Inverse for all world regions, and comparison with the other studies generally confirms this. The trends shown by Wiedmann et al (2008) and our QMRIO study agree well. Both studies show the effects of the "dash for gas"  and economic downturn in the early 1990s. Following this period emissions are estimated by both studies to have risen fairly steadily. It is important to note that all of these studies confirm that, from a consumption perspective, UK GHG emissions have risen since 1990 rather than fallen as indicated in statistics reported according to the Kyoto protocol.
11. In conclusion, in our opinion it is possible to develop a robust methodology for measuring emissions on a consumption basis but at the moment robust datasets are not yet available. The UK should be ambitious in helping to develop international data standards for use in consumption accounting.
What are the benefits and disadvantages associated with taking a consumption-based rather than production-based approach to GHG emissions accounting?
12. A key advantage of adopting consumption accounting is that this gives a more complete picture of the UK's carbon exposure: as the price of carbon rises globally, so we can expect the price of goods and services to rise in line with their carbon content. By being aware of carbon emissions from the consumption perspective and attempting to minimise them, the UK will be able to mitigate this risk. It will also of course be taking a greater responsibility for the carbon emissions attributable to UK consumption patterns.
13. One disadvantage associated with taking a consumption-based approach to GHG emissions accounting is that the UK Government and regulators do not have direct control over emissions that occur in other countries. Emissions embedded in imported goods can be classified into two types:
· Emissions embedded in semi-finished goods and services imported into UK industry, which are then used in products purchased by UK final consumers. These emissions are companies’ ‘Scope 3’ supply chain emissions . Currently many companies do not even estimate their Scope 3 emissions, let alone try to minimise them. It is therefore imperative that progress is made towards companies accounting for, taking responsibility for, and minimising their Scope 3 emissions;
· Emissions embedded in imported finished goods and services purchased by UK final consumers (households, government, and capital investment).
14. In some respects households can be considered as the ultimate final consumers as, in the final analysis, all economic activity ultimately serves households  . Therefore, as ultimate final consumers, households also arguably responsible for all emissions embedded in goods and services, regardless of whether they are embedded in semi-finished or finished goods and services. However whereas households are generally aware of their emissions due to direct energy use (e.g. for space and water heating, lighting and appliances), they are much less where of the embedded emissions. As illustrated in Figure 3, embedded emissions account for over half of household emissions (around 55% in terms of CO2, and 62% in terms of GHGs (Druckman and Jackson 2009a; Druckman and Jackson 2009b). Of these emissions, the proportion that occurs outside the UK has risen from 33% in 1990 to 41% in 2004 (see Figure 4).
Figure 3. Trends in CO2 attributable to UK households 1990-2004.
Source: Druckman and Jackson (2009a)
15. It is, of course, hard for consumers to directly influence the quantity of carbon emissions embedded in their purchases. But it is, however, important that they become increasingly aware of these embedded emissions, and where possible (a) reduce their purchase of carbon intensive goods, and (b) exert pressure on government and companies to reduce embedded emissions.
Figure 4: Percentage of embedded GHG emissions from imported goods
Source: Druckman and Jackson (2010)
Would it be (a) desirable and (b) practicable for the UK to adopt emissions reduction targets on a consumption rather than production basis?
16. This question is phrased as an "either/or" question regarding consumption versus production accounting, but we do not think that should be a straight choice: both approaches are valid and they should be used in combination. Taking either production or consumption accounting targets alone would be a mistake. Figure 1 illustrates the inadequacy of adopting production-based target only. However, adoption of a consumption-based target could also produce sub-optimal results. This is illustrated by the case of Scotland, which, from the consumption perspective has CO2 emissions 46% higher than on the production basis (Turner et al. 2011). One of the reasons behind this is that Scotland exports electricity to the rest of the UK and, due to the renewable content of this electricity, the exported electricity is relatively low carbon. This illustrates that in some cases higher consumption-based emissions compared to production emissions is not necessarily a "bad" thing as, in this case the local conditions in Scotland are more favourable for producing electricity from renewables than the local conditions in the rest of the UK. If renewable electricity was not exported from Scotland to the rest of the UK, the overall greenhouse intensity of the U.K.'s electricity supply would be higher, thus raising total GHG emissions. So rather than simply trying to optimise actions according to one indicator for another, it is best if decisions are made in each case according to local conditions with the overarching goal to minimise global GHG emissions.
17. In consideration of the practicality of adopting a consumption-based reduction target, as explained above, consumption accounting is, arguably, a more appropriate choice if we want to understand emissions attributable to UK lifestyles. However, by their very nature, emissions that occur abroad in the production and distribution of goods and services for final consumption in the UK are at least partly under the control of foreign regimes, so we in the UK have less ability to control them. Although there are some strategies that could mitigate such impacts, in a globalised market economy targets for reducing emissions over which we have limited control is problematic.
18. Finally, as discussed earlier, the datasets to carry out robust consumption accounting are still lacking and therefore it is not practicable to adopt a formal consumption based target as yet. We advocate that the UK works towards adopting emissions reduction targets on a consumption basis in parallel with the production basis in future. Initially consumption accounts should be used as a more informal indicator.
What are the potential implications at the international level of the UK adopting a consumption- rather than production-based approach to GHG emissions accounting?
19. An important aspect of international negotiations is that of trust. Although we do not advocate adoption of consumption accounting as an alternative to production based GHG emissions accounting, we do advocate production and consumption approaches being taken together. By including consumption-based accounts, the UK would be seen to be appreciating a more equitable form of sharing responsibility for GHG emissions, and be acknowledging in a more complete way the UK's historical contribution to climate change. The UK has already shown world leadership by passing the Climate Change Act 2008, and it could again gain leadership by moving towards adoption of consumption accounting alongside production accounting.
20. As indicated above, datasets capable of providing a fully robust basis for consumption accounting at an international level are not yet available. In the light of this, there is a risk that introducing consumption-based accounts into international climate negotiations at this stage might further complicate and stall progress. To mitigate this risk, the UK should take a proactive role in developing and standardising such datasets at the international level.
Are there any other issues relating to consumption-based emissions reporting that you think the Committee should be aware of?
21. The danger with simple indicators such as carbon emissions (from either production or consumption perspective, or both) lies in their inability to take full account of complexity. There is a danger that pursuing one or two indicators can have unexpected and undesirable consequences. For example, GHGs are not the only issue we should be taking into account: arguably, for example, water and waste are also extremely important issues.
Dimaranan, B. V., Ed(s). (2006). Global Trade, Assistance, and Production: The GTAP 6 Data Base Center for Global Trade Analysis, Purdue University, USA.
Druckman, A., P. Bradley, E. Papathanasopoulou and T. Jackson (2008). "Measuring progress towards carbon reduction in the UK " Ecological Economics 66(4): 594-604.
Druckman, A. and T. Jackson (2008). The Surrey Environmental Lifestyle MApping (SELMA) framework: development and key results to date. RESOLVE Working Paper 08-08, University of Surrey, Guildford, UK. Available from http://www.surrey.ac.uk/resolve/Docs/WorkingPapers/RESOLVE_WP_08-08.pdf.
Druckman, A. and T. Jackson (2009a). "The carbon footprint of UK households 1990-2004: a socio-economically disaggregated, quasi-multiregional input-output model." Ecological Economics 68 (7): 2066–2077.
Druckman, A. and T. Jackson (2009b). Mapping our carbon responsibilities: more key results from the Surrey Environmental Lifestyle MApping (SELMA) framework. RESOLVE Working Paper 02-09, University of Surrey, Guildford, UK. Available from http://resolve.sustainablelifestyles.ac.uk/sites/default/files/RESOLVE_WP_02-09_0.pdf.
Druckman, A. and T. Jackson (2010). An Exploration into the Carbon Footprint of UK Households. RESOLVE Working Paper Series 02-10, University of Surrey, Guildford, UK. November 2010. Available from http://resolve.sustainablelifestyles.ac.uk/sites/default/files/RESOLVE_WP_02-10.pdf.
Miller, R. E. and P. D. Blair (2009). Input-output Analysis: Foundations and Extensions. Cambridge, UK. 2nd revised edition, Cambridge University Press
Peters, G. and E. Hertwich, G. (2008). "CO2 Embodied in International Trade with Implications for Global Climate Policy." Environmental Science & Technology 42(5): 1401-1407.
Turner, K., N. Yamano, A. Druckman, S. Jung Ha, J. De Fence, S. McIntyre and M. Munday (2011). "An input-output carbon accounting tool: with carbon footprint estimates for the UK and Scotland." Fraser Economic Commentary Special Issue: Energy and Pollution. January 2011: 6-20.
Wiedmann, T., R. Wood, M. Lenzen, J. Minx, D. Guan and J. Barrett (2008). Development of an Embedded Carbon Emissions Indicator – Producing a Time Series of Input-Output Tables and Embedded Carbon Dioxide Emissions for the UK by Using a MRIO Data Optimisation System, Report to the UK Department for Environment, Food and Rural Affairs by Stockholm Environment Institute at the University of York and Centre for Integrated Sustainability Analysis at the University of Sydney, June 2008. Defra, London, UK.
 This is estimated using the Surrey Environmental Lifestyle MApping (SELMA) framework (Druckman and Jackson 2009a; Druckman and Jackson 2009b) .
 United Nations Framework Convention on Climate Change.
 See Druckman and Jackson (2009a) .
 Industries ultimately produce goods and services for household consumers, and all government and investment expenditure (such as expenditure on hospitals, education, roads and railways) is ultimately to serve households.