Energy and Climate Change CommitteeMemorandum submitted by Karen Turner, Peter McGregor, Max Munday and J. Kim Swales


* The central distinction between emissions accounted under production and consumption based measures is the treatment of emissions embodied in trade flows.

* Consumption based measures exclude emissions embodied in exports but include emissions embodied in imports. However, the latter are very difficult to identify. The work reported here uses preliminary data provided by the OECD. The OECD is working on an appropriate database for consumption based accounting, but this is still in the process of development. Most existing consumption accounting studies rely on data that are not in the public domain.

* For Scotland and the UK as a whole, our research shows that consumption-based, as against production-based, accounting methods identify a significantly higher level of CO2 emissions. However, this is not the case for all UK regions: we calculate that Welsh emissions are lower under consumption based accounting due to the high emissions content of exports from the region.

* It is important that any policy focus does not rely solely on production or consumption based measures, instead focussing on region-specific aspects of emissions generation. For example, we find that Scotland is “punished” in carbon trade balance terms by having adopted cleaner technologies in its export-intensive electricity generation sector.

* There are important issues concerning the incentives implied by adopting either production or consumption based accounting measures as the basis for emissions targets. For example, while production-based measures may encourage importing of “dirty” goods and services, consumption-based accounting may reduce incentives to ‘clean up’ domestic technology where production primarily serves export demands.

1. Introduction

1.1 The Fraser of Allander Institute at the University of Strathclyde, University of Stirling (under the ESRC Climate Change Leadership Fellowship, ESRC ref: RES- RES-066-27-0029), and Cardiff University (ESRC BRASS Centre) wish to submit evidence to the enquiry examining Consumption Based Emissions reporting. In this evidence we address the following questions:

How do assessments of the UK’s greenhouse gas emissions differ when measured on a consumption rather than a production basis?

Is it possible to develop a robust methodology for measuring emissions on a consumption rather than production basis?

What are the benefits and disadvantages associated with taking a consumption-based rather than production-based approach to greenhouse gas emissions accounting?

In summary we show that in the UK and Scottish cases there are important differences between findings depending on the reporting principles adopted. We explain why this is the case, and then discuss policy issues resulting from the analysis. In this submission we do not expand on the methods that we develop and apply to produce the consumption-based measures of emissions: details of these may be found in the references (below).

2. Accounting for Carbon Generation

2.1 A crucial issue affecting unilateral attempts to fulfil national emissions reductions targets under the Kyoto Protocol is the impact of international trade on domestic emissions generation. The basic problem is that emissions generated in producing goods and services to meet export demand are attributed to the producing nation’s emissions account. That is, pollution generation in any one country is partly driven by consumption decisions in others. In summary it is necessary to distinguish a “production accounting principle” (PAP) and a “consumption accounting principle” (CAP) in considering the structure of pollution problems.

2.2 Accounts constructed on the PAP focus on emissions produced within the geographical boundaries of the national economy. This is what is accounted for, and what individual national governments are responsible for reducing, under the Kyoto Protocol. In contrast, the latter, CAP, focuses on emissions produced globally to meet consumption demand within the national economy. This is what increasingly popular measures such as carbon footprints attempt to measure, and what many people regard as more appropriate, given that human consumption decisions are commonly considered to lie at the heart of the climate change problem. In an economy with no trade, total emissions calculated under both production and consumption accounting principles would be equal (by definition). Further, if accurate production and consumption accounts were calculated for all countries they should sum to the same world figure. Where there is trade and pollution is embodied in that trade through emissions generated in one region or nation to meet consumption demand in another, the CAP and PAP measures for individual regions or countries need not, and in general will not, be equal. A foreign “trade balance” in pollution will exist in terms of the difference between total emissions estimated on the basis of the production and consumption accounting principles, or more simply, the difference between the pollution embodied in exports and the pollution embodied in imports.

2.3 Thus, the question arises as to whether PAP and/or CAP measures should be used to monitor and track pollution generation at the individual country level and what, if anything, can be done about foreign “trade balances” in pollution.

3. Summary Carbon (CO2 equivalent) Accounting Results for the UK and Scotland

Table 1


3.1 Our findings, presented in Table 1, suggest that in 2004 the “carbon footprint” of UK consumption at 813.5 million tonnes of CO2 equivalent was significantly higher than the level of CO2 emissions generated within UK borders, which measured 643.8 million tonnes of CO2. That is, under CAP accounting, UK CO2 generation is 26% higher than under the PAP accounting measure that is the metric for the CO2 reduction targets under Kyoto. This implies that the UK “imports sustainability”from its trade partners. This result might be expected in the case of an advanced economy whose domestic production has shifted from manufactured goods towards more service orientated activities, whilst increasing imports of manufactured goods. However, it raises questions as to whether PAP measures (as under the Kyoto Protocol targets) give an appropriate account of the sustainability of the UK economy.

Table 2


3.2 The Scottish results, on the other hand, are more complex. Scotland also “imports sustainability”, but with a much larger difference between CO2 allocated under PAP accounting methods (48.9 million tonnes of CO2 in 2004) and under CAP (71.5m tonnes of CO2). There is a 46% increase as we move from measuring CO2 generated within Scotland, to measuring the full amount of CO2 embodied in Scottish consumption in the form of Scotland’s ‘carbon’ (CO2 equivalent) footprint. However, a key underlying determinant of Scotland’s CO2 trade deficit, particularly with the rest of the UK, is the fact that Scotland is a net exporter of electricity. In this respect, the results in Table 2 reflect the (increasing) prevalence of renewable generation technologies in Scotland, which means that the export side of Scotland’s CO2 trade balance with the rest of the UK has a lower CO2 intensity and content than it would if these technologies had not been implemented. While of course there should be concern over the CO2 content of Scottish imports, the key implication that emerges from our results is that Scotland, by fostering renewable technologies in electricity generation, is helping to lower the carbon footprints of its trade partners, particularly the rest of the UK. However, this is not reflected in the CAP accounting measure. Moreover, such changes make Scotland’s balance of trade in emissions worse.

This is illustrated by the fact that if we were to assume that UK average polluting technologies apply throughout the nation in the CO2 accounting exercise, not only is Scottish domestic CO2 generation (estimated under PAP) higher (at 66.7 million tonnes), we find that it would actually run a CO2 trade surplus with the rest of the UK, and its CAP-based carbon footprint would only be 17% higher than its domestic CO2 generation, compared with the 26% difference at the UK level. Thus, it would seem that Scotland is ‘punished’ in CO2 trade balance terms by having adopted cleaner technologies in producing what are usually quite CO2 intensive exports (see McGregor et al, 2008).

3.3 The central question to ask is whether either of the PAP or CAP measures are appropriate on their own for CO2 accounting. Both PAP and CAP methods are clearly dependent on consumption and technology decisions taken both within and outwith the reporting region or nation. Moreover, in the case of regional economies, it raises questions as to the role of indicators and targets across different regions that may play different (complementary) roles in delivering on economic and environmental sustainability. For example, in work for Wales (reported in Turner et al, 2011b), we find that CO2 emissions are lower under CAP than under PAP. This largely due to the fact that a significant share of Welsh PAP emissions are generated in producing steel for export to the rest of the UK to serve as inputs to the wider, but on average less CO2-intensive, manufacturing sector.

3.4 It also should be noted that calculation of CAP measures is extremely data-intensive and subject to error. The crucial problem is that information is required on production and polluting technologies in all of the countries from which the reporting economy imports. As explained in Turner et al (2011a), our estimates of the pollution content of imports are based on data provided by colleagues at the OECD. However, as we discuss in our previous research that dataset is imperfect and incomplete, although OECD is currently working to improve it.

4. Conclusions

4.1 Three broad issues should be considered in considering the use of CAP and/or PAP measures. The first concerns the incentives that each implies. If we are to focus on CAP-based targets at the national and/or regional level, this implies that we are completely unconcerned with any reduction in the emissions embodied in what we export, since under CAP this is the responsibility of consumers in another jurisdiction. Rather, we are solely concerned with reducing the emissions embodied in what we consume. On the other hand, focusing solely on a PAP-based target may lead to replacing domestic production with imports of CO2-intensive goods and services. Generally, the strategy employed to meet any target set must take account of what the target itself does and does not cover. For example, pursuing a CAP target may lead to the potentially paradoxical situation where domestic governments are not incentivised to constrain domestic pollution where this is generated to serve foreign demands. On the other hand, a CAP-based measure gives the reporting authorities an incentive to reduce the emission intensity of technologies used to produce imported goods, over which the reporting policymakers have little say (and no jurisdiction).

4.2 Moreover, it is also necessary to consider issues of cost and accuracy in deciding on which pollution measures to adopt. Measurement under PAP is costly but in the UK it is already in place at the national, though not at the regional, level (certainly not based on consistent official statistics). However, as noted above, appropriate databases to measure emissions under CAP for any country or region are not yet in place. Most studies rely on the subscription-based GTAP database produced at Purdue University in Indiana. The producers of this data-set emphasise that it is not intended for formal pollution accounting, relying as it does on subscriber contributions of input-output data for different countries. Formal and internationally comparable CAP measurement would require internationally agreed database construction (including coverage of international trade flows in input-output format) and accounting methods.

4.3 It is important to note that we are not suggesting that there is no merit in consumption-based targets. Rather, given the interest in the policy community in CAP based measures, we are taking this opportunity to pose questions such as: why CAP rather than PAP targets are being promoted; what impacts are policymakers really concerned about, and what are the tradeoffs involved in the variety of measures that are available.


McGregor, P, Swales, J K and Turner, K (2008) The CO2 “trade balance” between Scotland and the rest of the UK: Performing a multi-region environmental input–output analysis with limited data, Ecological Economics, Vol. 66, Issue 4, pp 662–673.

Turner, K, Yamano, N, Druckman, A, Ha, S J, De Fence, J, McIntyre, S and Munday M (2011a) An input-output carbon accounting tool with carbon footprints estimates for the UK and Scotland, Fraser Economic Commentary, special issue (January 2011) titled “Energy and Pollution”, pp 6–20. Fraser of Allander Institute, University of Strathclyde. 2011.

Turner, K, Munday, M, McIntyre, S and Jensen, C D (2011b) Incorporating jurisdiction issues into regional carbon accounts under production and consumption accounting principles, Environment and Planning A, 24, pp 722–741. 2011.

October 2011

Prepared 17th April 2012