Final Report Summary - CARBON CAP (Carbon emission mitigation by Consumption-based Accounting and Policy)
Executive Summary:
Carbon-CAP is a FP7 project that has identified promising demand-side strategies and their associated policy instruments, providing an additional suite of instruments to complement production based policies. Each of these is assessed against:
- Criteria of Effectiveness (how much carbon reduction is achieved in a given product or service if applied successfully)
- Scope (how much of the global flow of carbon is then affected)
- Economic Equity (how are the costs distributed within society), and
- Political, Legal and Institutional Acceptance.
The action ran from October 2013 to end of December 2016.
The results indicate that policy instruments which change the characteristics of products available to consumers (such as minimum standards or requirements) should have priority, while policies that affect consumer choices between products on the market could be applied at a second stage, and as a way to support the priority measures. The total greenhouse gas (GHG) reduction potential of all options combined is considerable: reducing around half of EU footprint emissions. The options with the highest potential appear to be in the food, building and transport consumption categories, so new policies may need to be put in place to tackle these emissions.
Partners are:
1. The Netherlands Organisation for Applied Scientific Research (TNO), Delft, The Netherlands
2. Wirtschaftsuniversität Wien (WU), Vienna, Austria
3. EC DG JRC Institute for Prospective Technical Studies (IPTS), Seville, Spain/Brussels Belgium
4. Institute of Environmental Sciences, Leiden University (LU-CML), Leiden, The Netherlands
5. Norges Teknisk-Naturvitenskapelige Universitet NTNU (NTNU), Trondheim, Norway
6. Cambridge University Centre for Climate Change Mitigation Research (4CMR), Cambridge, UK
7. Cambridge Econometrics (CE), Cambridge, UK
8. Climate Strategies (CS), London, UK
9. Deutsches Institut für Wirtschaftsforschung (DIW), Berlin, Germany
10. International Centre for Trade and Sustainable Development (ICTSD), Geneva, Switzerland
11. University College London, Institute for Sustainable Resources (UCL-ISR), UK
For more information contact the co-ordinator at: arnold.tukker@tno.nl
Project Context and Objectives:
Climate policies are formulated on national or regional level and differ in stringency and approach reflecting differences in economic development, political culture and will. They further mainly focus on production sectors. Yet growing consumption is a main driver behind rising greenhouse gas (GHG) emissions. Further, our economy is increasingly a single, global economy: international trade has risen threefold since 1990. Current climate policies are mainly shaped via territorial emission reduction methods with approaches from a consumption oriented perspective have added value:
1. Consumption-and trade oriented policies can explicitly address issues like carbon leakage and for instance identify situations where an apparent reduction of GHG emissions in a country is mainly the result of structural change in which carbon –intensive industries were relocated abroad.
2. Such policies are more directly addressing consumption as a driver for rising GHG emissions.
Complementing existing policies and initiatives (e.g. Energy 2020, Kyoto Protocol, Energy Roadmap 2050), consumption–based carbon emission accounts (CBCA) and consumption-based and trade related climate policy (CBTP) offer a possibility to address and partially mitigate the effect of differences in national climate policies. They can do so by taking into account GHG emissions during the entire life cycle of consumed products and services, aiming to reduce those.
The life-cycle perspective has already been adopted widely at the micro-level, and has shaped policies at that level significantly. Mitigation policies that include complementary consumption and trade components thus bear the promise of providing more balanced cost-effective and efficient solutions compared to a focus on production alone. It may also be possible to identify and stimulate low-impact consumption patterns with no lower or even higher quality life, both in developed and emerging economies.
The project has the following overall objectives:
1. To stimulate innovative European and international climate policies and services due to improved shared knowledge base on consumption emissions.
2. To realize a more effective policy mix for achieving the objectives of the EU Climate and Energy package and the Roadmap for moving to a competitive low carbon-economy in 2050.
The potential to complement current domestic GHG reduction efforts with policies that address consumption patterns is clear. However, developing a new, more balanced mix of policies will require overcoming various important gaps and hurdles.
The following gaps will be addressed:
Gap 1: Quantification of global emissions related to consumption of goods and services and understanding drivers for upward trends.
Gap 2: Understanding of levers, potential mechanisms, and feasibility of demand side tools and polices.
Gap 3: Understanding of the effectiveness and impacts of demand side tools and policies.
Gap 4: No shared view on added value, implementation challenges and acceptability of demand side tools/policies and related accounts, and no ‘roadmap’ of evolution from production towards consumption-based policies.
The project is specifically designed to overcome these gaps and convince a critical mass of the climate policy community of the added value of demand side policies and accounts. The plan is to include a well-designed process of interactive learning between the project team and key players in this policy area.
The project has the following specific scientific and technical objectives. It will test and improve available methods and tools to conduct consumption-based accounting. We further will analyse the potential, effectiveness and implementation challenges of demand side (including trade related) tools and polices. Using the improved tools and methods, a selection of simulated design of consumption based polices, and model-based assessment of their efficiency will be carried out. This will be done at the macro-level, but also for specific key sectors. In relation to each gap above we will address the following specific scientific and technical objectives:
- Gap 1: Review and assessment of consumption–based carbon accounts and upward drivers
- Gap 2: Identification and evaluation of demand side tools and policies in relation to desired technical and behavioural improvement options.
- Gap 3: Modelling and assessment of impacts of consumption-based emission reduction pathways at macro-level.
- Gap 4: Creating an implementation roadmap for consumption based accounts and policies endorsed by critical mass of stakeholders via policy-science brokerage activities.
Project Results:
Global drivers of change in carbon emissions
Current climate policies are mainly shaped via territorial emission reduction approaches. There is a number of arguments why complementing these territorial approaches with approaches from a consumption oriented perspective have added value: (1) consumption- and trade oriented policies can explicitly address issues like carbon leakage and for instance identify situations where an apparent reduction of carbon emissions in a country is mainly the result of structural change in which carbon-intensive industries were relocated abroad; and (2) such policies are more directly addressing consumption as a driver for rising greenhouse gas (GHG) emissions.
Aspects in the current EU climate policy landscape that require attention in evaluating consumption-based carbon policy instruments:
- Consumption-based emission accounting requires reliable emission data and robust methods to track embodied emissions in trade and consumption.
- A transformation towards low carbon consumption produces winners and losers. The changes in consumption patterns in big consuming countries impact producing countries.
- In order to reduce emissions embodied in trade and consumption, policies can intervene at the level of the producers, intermediaries or final consumers.
- Carbon pricing policies were often guided by the principle of implementing a carbon price as much upstream as possible in the expectation that the carbon price will feed through the value chain and thus incentivise all involved actors to shift their decisions toward lower carbon options.
- The analysis of food labelling approaches showed that numerous voluntary carbon labelling initiatives have emerged, mainly driven by the private sector, although some public bodies and international organization have also been involved.
- The effectiveness of a policy instrument to achieve a reduction in embodied emissions in trade and consumption depends also on its synergy and potential for conflict with existing (non-carbon) policy instruments.
The analyses in efficiency (emissions, energy, and labour per unit output), the changes due to trade related effects (both for intermediate producers and final consumers), the changes due to technology effects (both for intermediate producers and final consumers), and the change due to affluence and population indicate that trade is an important driver for global greenhouse gas emissions growth. However, it is not as important as growth in affluence and overall industry efficiency. This is only true, however, when looking at global emissions growth. When taking into account regional shifts in greenhouse gas emissions footprints over time, the displacement of industries from developed economies in the European Union and the OECD and the increase in imports to final demand contributes to emissions growth, mainly from combustion. For non-combustion emissions, changes in trade partners seems to decrease GHG footprints. Different dynamics act on the footprint growth over time and in different regions. Greenhouse gas emissions and energy consumption are mainly driven by the increase of consumption per capita in developing economies, such as China, and in the European Union. This growth in affluence reduces (or even reverses) gains in carbon and energy efficiency. It can be seen that trade is an important driver for labour footprints change in developed economies, in a higher proportion than for energy and greenhouse gas footprints. That indicates that the displacement of industries to labour-abundant countries might not have a significant effect in the growth of emissions embodied in trade.
Consumption based carbon accounting
The underlying question of the Carbon-CAP project is, to assess whether a consumption based carbon accounting and consumption based climate policy can have an added value to the already existing production based accounting and associated policy as a means to reduce GHG emissions. A consumption based approach differs from a production based approach in the definition of the system boundaries. A production based approach, requires a geographically identified system. A consumption oriented approach requires a functional, cradle-to-grave or “footprint” approach, usually including processes in different geographical areas. So far, policy has mainly focused on production and nations, and therefore, has used a territorial approach. This is also apparent when looking at GHG emission databases: they are organized by country and by activity. A consumption based approach needs something different. Requirements to a consumption based information base:
- GHG emissions should be linked to consumption activities and consumption categories
- GHG emissions should be specified on a cradle-to-grave basis
- The information base should provide information at a relevant spatial and time scale
- The quality of the information should be sufficiently reliable
- The information base should allow for analysing the past as well as forecasting the future, or rather, imagining the future under different assumptions.
Since the advent of environmental footprint approaches in general, and consumption based carbon accounting (CBCA) approaches specifically, many policy makers have been looking at ways to derive consumption-based policies. Whilst these efforts can be lauded, it has not been clearly established in the literature that consumption-based policies are more effective or more cost-effective than traditional policies based on control of territorial emissions. Further complicating the policy arena is that many policies could be considered both traditional and consumption-based (insulation of houses, for example). Alternatively, CBCA can be seen to be policy relevant, whilst not policy prescriptive. CBCA can give a key macro-level indication about the carbon intensity of an economy relative to baselines and targets. Such reporting of emission accounts can further underline the need for multi-lateral action, and for the increased responsibility needed to be shouldered by economies with growing net-import of emissions. CBCA can further strengthen resolve around uptake of instruments around, for example, the clean development mechanism (through encouraging investment from the developed world in trade partner countries in the developing world), or for the need for additional investment in emission offsets.
Promising consumption based policy measures
More than 30 policy instruments have been assessed in Carbon-CAP, covering products and services in Transport, Manufacturing, Food, Buildings, Paper/Plastics and Textiles. To assist with choices between policy instruments, a shortlist of promising instruments were ranked in three tiers. The first tier contains instruments that are judged to be strong across the four criteria of acceptability (economic, legal, international/political, institutional). The third tier contains instruments for which there is a significant barrier to acceptance on at least one of the criteria.
To effectively reduce emissions at the global level, consumption-based climate policy instruments will have to be part of the policy mix. Introducing instruments in a portfolio has three main advantages. First, consumer-oriented policy should not have the effect of wholly ‘individualising’ responsibility solely on end-users. It should spread responsibilities across many sectors, across consumers and across producers. Second, emissions are caused by many different decisions at many different levels from primary production to consumption to disposal. Consumer-oriented policies only act on part of these, and individual consumer-based instruments further focus the scope of application. Finally, experience has shown policies are often most effective when developed in mutually reinforcing ways since weaknesses in any one instrument can be counterbalanced by strengths of another instrument. This often helps in negotiations between groups implementing and affected by an instrument.
The assessments carried out by the project provide a useful first overview of promising instruments and a starting point for identifying opportunities and challenges to focus on in future deliberations and analyses. A key lesson is that consumer choice is difficult to influence when consumers have equal access to high and low carbon goods that meet the same needs. Therefore, the rankings of effectiveness and acceptability of instruments developed in this briefing reflect a tiered approach in which instruments that alter the range of products available, their ease of access and/or the cost (due to carbon charges) are applied first. The second and third ranks of instruments might then be considered means to support the instruments in the first rank. This is consistent with the lesson that instruments are most effective when introduced as complementary portfolios.
Modelling consumption based emission reduction
The modelling of the Carbon-CAP project assesses the effects of consumption-based emission reductions options on emissions and the economy. It focuses on the three main areas for improvement options: food, the built environment and transport. It uses a suite of three different models, each one based on different assumptions, to test the robustness of outcomes in relation to different modelling approaches.
The models used are E3ME, EXIOMOD and FIDELIO. E3ME is a macro-econometric model of the world’s economic and energy systems and the environment, based on post-Keynesian principles. EXIOMOD is a Computable General Equilibrium (CGE) Model. FIDELIO is a dynamic econometric input-output model that combines aspects of CGE models and linear ‘Input-Output philosophy’.
The IEA WEO 2014 current policies scenario is used as the main reference scenario for this report. In addition to the reference scenario and improvement options scenarios (food, transport and buildings), two additional sets of scenarios were included in the modelling exercise. The first additional scenario is the Nationally Determined Contributions Scenario (NDC), which is based on the pledges that were put forward at the Paris Conference of the Parties in 2015 and NDC plus improvement options. The second scenario is a Material Charge Scenario that considers a specific taxation instrument in Europe. Both the additional scenarios were modelled using E3ME only.
The improvement options in all three sectors that where selected for this study (food, buildings and transport) combined have a maximum potential to deliver household emission reductions at EU level of 47-67%, and total production-based emission reductions of 16-26% relative to the reference scenario in 2050. The impacts on production-based CO2 emissions outside the EU are small and mostly negative. However, when behavioural responses to individual policies, that could realise the improvement options, are considered, then there is a maximum 14% reduction (7.5-14%) in total CO2 production-based emissions reductions in the EU by 2050, relative to the reference scenario. These more modest reductions could be considered more realistic as many of the actions and policy measures considered were voluntary.
The most potential for reducing EU territorial CO2 emissions comes from the transport and buildings scenarios. The food options have less potential when CO2 only are considered. Therefore, because of the rebound effect a small increase in the EU households’ emissions is possible. All models show a slight decrease in consumption-based EU CO2 emissions under the combined food scenario (-0.2 to -3.6%). The range of impacts on consumption-based emissions is slightly wider range than the reductions in the territorial emissions (-0.3 to -1.2%).
The economic impacts of the combined scenario on the EU’s GDP are small, ranging from about 0.78% loss to a very small positive impact (0.06%) in 2050, depending on the model. The overall impact on employment in the EU in 2050 is very small and positive. Sectoral employment impacts vary depending on the improvement option and span from a 20% decrease in the vehicle manufacturing sector to a 61% increase in the provision of transport services.
The modelling undertaken in Carbon-CAP shows that various consumption-based policies relating to food, buildings and transport have considerable potential to reduce territorial CO2 emissions in Europe, especially with regards to household CO2 emissions. However, the reduction in consumption-based emissions is rather small and the relative gap between consumption- and production-based emissions is expected to increase. More detailed modelling will be required to estimate more comprehensively the total GHG emissions reduction potential of each of the individual policy packages and their implementation over the modelling time horizon 2020-2050. The consumption-based emission reduction measures used in this study have small impact on trade related emissions and therefore more attention should be paid to designing and assessing policies that address these emissions, including innovative measures such related to the options provided by the ITC sector. It should be also highlighted that there is high uncertainty related to modelling the policy impacts, especially on consumption-based emissions. Also using macromodels in addition to MRIO models for impacts assessments is feasible and desirable as it will help to asses feedback effects between in the economic system. Further development of model in this area is necessary to achieve more robust results, but differences in the results will remain as the models build on different theoretical backgrounds and these have to be considered while conducting impact assessments.
Further Research
Based on the feedback received through discussion with business stakeholders at the end of the Carbon-CAP project, it is also important to note the following, which could be explored in future related research projects. There is agreement amongst businesses consulted that this is an important conversation and there is interest on the part of business in discussing how to select the most effective policies for reducing consumption based emissions. Collaboration initiatives between government, business and consumers will be necessary to identify, develop and implement the most effective policies. Guidelines and criteria that governments should consider when engaging with business on new policy approaches for reducing emissions based on consumption might also be useful. More research may still be needed to iron out remaining uncertainties in the use of consumption-based accounting systems and developing whole-economy models capable of analysing and forecasting consumption-side emissions. More research could be undertaken on the full scope of the implementation of different consumption-side policy packages. In addition, Carbon-CAP established that it is possible to use consumption-based emissions from macroeconomic models as an indicator to assess future policies as well as allocating historical responsibility. This additional step could add substantial value to comprehensive policy assessments. However, there is a lot of work to be done on improving consumption-based emissions accounting in macromodels, as policy impacts on these emissions resulted in greater uncertainty than the impacts on production-based emissions. This need for further research does not preclude the possibility of the EU starting to recognise and quantify consumption-based emissions, and trade-embodied carbon, and to step up efforts to identify and implement new consumption policies with high levels of anticipated take-up and effectiveness.
Potential Impact:
The potential to complement current domestic GHG reduction efforts with policies that address consumption patterns is clear. However, developing a new, more balanced mix of policies will require overcoming various important gaps and hurdles. To overcome the gaps, the project answers the following specific scientific and technical objectives.
Gap 1: Quantification of global emissions related to consumption of goods and services and understanding drivers for upward trends.
As for the drivers, Carbon-CAP showed the following. The analyses in efficiency (emissions, energy, and labour per unit output), the changes due to trade related effects (both for intermediate producers and final consumers), the changes due to technology effects (both for intermediate producers and final consumers), and the change due to affluence and population indicate that trade is an important driver for global greenhouse gas emissions growth. However, it is not as important as growth in affluence and overall industry efficiency. This is only true, however, when looking at global emissions growth. When taking into account regional shifts in greenhouse gas emissions footprints over time, the displacement of industries from developed economies in the European Union and the OECD and the increase in imports to final demand contributes to emissions growth, mainly from combustion. This is caused by the fact that this displacement takes place to countries with in the period until now had a carbon intensive energy infrastructure, such as China. For non-combustion emissions, changes in trade partners seems to decrease GHG footprints.
Gap 2: Understanding of the levers, potential mechanisms, and feasibility of demand side tools and policies.
From previous research confirmed by Carbon-CAP, it is known that the areas of food, mobility and built environment dominate the life cycle impacts including carbon emissions from final consumption. Carbon-CAP identified a number of improvement options in these areas.
Two food-related improvement options – a reduction in food waste, and a switch to less emission-intensive diets – were selected for modelling. The two options are also modelled together as a combined food scenario. A selection of various voluntary and mandatory policy instruments can be used for shifting consumer behaviour towards these low carbon options, for example including supply chain procurement requirements and product labelling and educational campaigns.
Two buildings-related improvement options – a shift to near-zero emissions buildings (NZEBs) and an increased installation of natural fiber insulation for the existing housing stock (NFI) – were selected for modelling. Both options were also modelled together as a combined buildings scenario. In these scenarios, a selection of voluntary and mandatory policy instruments could be used for shifting consumer behaviour towards these low carbon options. The policies include standards, approved technology lists, financial incentives, product labelling and educational campaigns.
Transport has the largest number of sub-scenarios modelled, covering eight different transport-related improvement options. These include reducing the number of cars, enhancement of carpooling and sharing, lighter and smaller cars, and reduction of air transport. All the individual policy measures are also modelled together as a combined transport scenario. In these scenarios, a selection of voluntary and mandatory policy instruments could be used to incentivise a shift in consumer behaviour towards low carbon options; the policy measures include strengthening existing standards, introduction of new standards, infrastructure improvements, subsidies and educational campaigns.
Gap 3: Understanding of the effectiveness and impacts of demand side tools and policies.
The combination of improvement options and supportive policies were modelled with three different types of models:
1. E3ME, the global macro-econometric (Energy-Environment-Economy) E3 model of Cambridge Econometrics (CE). One of the strengths of this model is the underlying econometric specification, which provides a strong empirical basis for the analysis .
2. EXIOMOD is a Computational General Equilibrium (CGE) model that was recently developed by TNO. It is based on detailed EXIOBASE multi-regional environmentally extended input output tables (MREEIO), covering 43 countries, 5 rest of the world regions, 129 economic sectors and 40 GHG and non-GHG emissions.
3. FIDELIO is a model from the European Commission’s Joint Research Centre’s Institute for Prospective Technological Studies (IPTS). FIDELIO is a dynamic econometric input-output model based on Eurostat’s supply and use tables and the WIOD database covering 27 EU countries, 7 large countries outside Europe, 59 products/ economic sectors, 3 types of GHG emissions and 5 types of non-GHG emissions.
The modelling results show that various consumption-based policies have considerable potential to reduce territorial CO2 emissions in Europe, especially with regards to buildings (4-7%) and transport (4-14%). For food, direct emission reduction of e.g. limiting food waste is significant, but unfortunately the cheaper food basked then causes a rebound effect that annihilates this reduction. A limitation of the modelling is that CH4 is not included in E3ME and FIDELIO; EXIOMOD suggests the combined food scenario results in 14% reduction in methane emissions. Overall these three areas of consumption based options and policies contribute however just in a limited to the already planned production-oriented emission reductions. The net emission transfers between the rest of the world and Europe will stay stable or may slightly reduce, but as a percentage of total (reduced) emissions will grow.
A significant proportion of embodied carbon from imports is unavoidable. This is because it arises from foreign mining operations which are not conducted in the EU, and domestic alternative materials do not exist. In relation, the consumption-based emission reduction measures used in this study have small impact on trade related emissions and therefore more attention should be paid to designing and assessing policies that address these emissions. Policies addressing rebounds are highly relevant too.
Gap 4: No shared view on the added value, implementation challenges and acceptability of demand side tools/policies and related accounts, and no “roadmap” of evolution from production towards consumption-based policies.
There are a variety of databases, tools, methods and models that can be used to assess aspects of consumption based carbon emissions. For assessing the past and the present, environmentally extended Input Output analysis (possibly hybridized with LCA) seems to be the most ideal approach. Our research showed that by far the largest uncertainty in such carbon footprint analysis comes from data already used widely in climate negotiations: sector and country level CO2 emission data. Jus harmonizing such data across IO databases reduces uncertainty in footprint analyses with 50%. In a similar way, forward looking models such as E3ME, EXIOMOD and FIDELIO could be harmonized. This will allow for a robustness of calculating past and future consumption based carbon emissions with a level of uncertainty that is not significantly worse as current information on country- or sector level carbon emissions in the past, or carbon emissions modelled for the future.
Consumption oriented carbon reduction policies have added value and can support an additional reduction of carbon emissions compared to current scenarios. In this, implementing policies addressing rebound effects (like carbon taxes) are essential. Policies that would also address trade-related emissions directly or even indirectly however do not sit well in the UNFCCC process. If introduced formally, administrative challenges and other objections for instance if compatible with WTO rules will be the result. It seems much more fruitful that given a specific situation with regard to trade related emissions, trade partners may agree on a voluntary basis to share responsibilities for such emissions differently as analysed via a purely production based approach. Overall, it is therefore clear that consumption oriented policies should be seen as a complement to the traditional production oriented approach.
Further research
Based on the feedback received through discussion with business stakeholders at the end of the Carbon-CAP project, it is also important to note the following, which could be explored in future related research projects. There is agreement amongst businesses consulted that this is an important conversation and there is interest on the part of business in discussing how to select the most effective policies for reducing consumption based emissions. Collaboration initiatives between government, business and consumers will be necessary to identify, develop and implement the most effective policies. Guidelines and criteria that governments should consider when engaging with business on new policy approaches for reducing emissions based on consumption might also be useful. More research may still be needed to iron out remaining uncertainties in the use of consumption-based accounting systems and developing whole-economy models capable of analysing and forecasting consumption-side emissions. More research could be undertaken on the full scope of the implementation of different consumption-side policy packages. In addition, Carbon-CAP established that it is possible to use consumption-based emissions from macroeconomic models as an indicator to assess future policies as well as allocating historical responsibility. This additional step could add substantial value to comprehensive policy assessments. However, there is a lot of work to be done on improving consumption-based emissions accounting in macromodels, as policy impacts on these emissions resulted in greater uncertainty than the impacts on production-based emissions. This need for further research does not preclude the possibility of the EU starting to recognise and quantify consumption-based emissions, and trade-embodied carbon, and to step up efforts to identify and implement new consumption policies with high levels of anticipated take-up and effectiveness.
List of Websites:
carboncap.eu
Carbon-CAP is a FP7 project that has identified promising demand-side strategies and their associated policy instruments, providing an additional suite of instruments to complement production based policies. Each of these is assessed against:
- Criteria of Effectiveness (how much carbon reduction is achieved in a given product or service if applied successfully)
- Scope (how much of the global flow of carbon is then affected)
- Economic Equity (how are the costs distributed within society), and
- Political, Legal and Institutional Acceptance.
The action ran from October 2013 to end of December 2016.
The results indicate that policy instruments which change the characteristics of products available to consumers (such as minimum standards or requirements) should have priority, while policies that affect consumer choices between products on the market could be applied at a second stage, and as a way to support the priority measures. The total greenhouse gas (GHG) reduction potential of all options combined is considerable: reducing around half of EU footprint emissions. The options with the highest potential appear to be in the food, building and transport consumption categories, so new policies may need to be put in place to tackle these emissions.
Partners are:
1. The Netherlands Organisation for Applied Scientific Research (TNO), Delft, The Netherlands
2. Wirtschaftsuniversität Wien (WU), Vienna, Austria
3. EC DG JRC Institute for Prospective Technical Studies (IPTS), Seville, Spain/Brussels Belgium
4. Institute of Environmental Sciences, Leiden University (LU-CML), Leiden, The Netherlands
5. Norges Teknisk-Naturvitenskapelige Universitet NTNU (NTNU), Trondheim, Norway
6. Cambridge University Centre for Climate Change Mitigation Research (4CMR), Cambridge, UK
7. Cambridge Econometrics (CE), Cambridge, UK
8. Climate Strategies (CS), London, UK
9. Deutsches Institut für Wirtschaftsforschung (DIW), Berlin, Germany
10. International Centre for Trade and Sustainable Development (ICTSD), Geneva, Switzerland
11. University College London, Institute for Sustainable Resources (UCL-ISR), UK
For more information contact the co-ordinator at: arnold.tukker@tno.nl
Project Context and Objectives:
Climate policies are formulated on national or regional level and differ in stringency and approach reflecting differences in economic development, political culture and will. They further mainly focus on production sectors. Yet growing consumption is a main driver behind rising greenhouse gas (GHG) emissions. Further, our economy is increasingly a single, global economy: international trade has risen threefold since 1990. Current climate policies are mainly shaped via territorial emission reduction methods with approaches from a consumption oriented perspective have added value:
1. Consumption-and trade oriented policies can explicitly address issues like carbon leakage and for instance identify situations where an apparent reduction of GHG emissions in a country is mainly the result of structural change in which carbon –intensive industries were relocated abroad.
2. Such policies are more directly addressing consumption as a driver for rising GHG emissions.
Complementing existing policies and initiatives (e.g. Energy 2020, Kyoto Protocol, Energy Roadmap 2050), consumption–based carbon emission accounts (CBCA) and consumption-based and trade related climate policy (CBTP) offer a possibility to address and partially mitigate the effect of differences in national climate policies. They can do so by taking into account GHG emissions during the entire life cycle of consumed products and services, aiming to reduce those.
The life-cycle perspective has already been adopted widely at the micro-level, and has shaped policies at that level significantly. Mitigation policies that include complementary consumption and trade components thus bear the promise of providing more balanced cost-effective and efficient solutions compared to a focus on production alone. It may also be possible to identify and stimulate low-impact consumption patterns with no lower or even higher quality life, both in developed and emerging economies.
The project has the following overall objectives:
1. To stimulate innovative European and international climate policies and services due to improved shared knowledge base on consumption emissions.
2. To realize a more effective policy mix for achieving the objectives of the EU Climate and Energy package and the Roadmap for moving to a competitive low carbon-economy in 2050.
The potential to complement current domestic GHG reduction efforts with policies that address consumption patterns is clear. However, developing a new, more balanced mix of policies will require overcoming various important gaps and hurdles.
The following gaps will be addressed:
Gap 1: Quantification of global emissions related to consumption of goods and services and understanding drivers for upward trends.
Gap 2: Understanding of levers, potential mechanisms, and feasibility of demand side tools and polices.
Gap 3: Understanding of the effectiveness and impacts of demand side tools and policies.
Gap 4: No shared view on added value, implementation challenges and acceptability of demand side tools/policies and related accounts, and no ‘roadmap’ of evolution from production towards consumption-based policies.
The project is specifically designed to overcome these gaps and convince a critical mass of the climate policy community of the added value of demand side policies and accounts. The plan is to include a well-designed process of interactive learning between the project team and key players in this policy area.
The project has the following specific scientific and technical objectives. It will test and improve available methods and tools to conduct consumption-based accounting. We further will analyse the potential, effectiveness and implementation challenges of demand side (including trade related) tools and polices. Using the improved tools and methods, a selection of simulated design of consumption based polices, and model-based assessment of their efficiency will be carried out. This will be done at the macro-level, but also for specific key sectors. In relation to each gap above we will address the following specific scientific and technical objectives:
- Gap 1: Review and assessment of consumption–based carbon accounts and upward drivers
- Gap 2: Identification and evaluation of demand side tools and policies in relation to desired technical and behavioural improvement options.
- Gap 3: Modelling and assessment of impacts of consumption-based emission reduction pathways at macro-level.
- Gap 4: Creating an implementation roadmap for consumption based accounts and policies endorsed by critical mass of stakeholders via policy-science brokerage activities.
Project Results:
Global drivers of change in carbon emissions
Current climate policies are mainly shaped via territorial emission reduction approaches. There is a number of arguments why complementing these territorial approaches with approaches from a consumption oriented perspective have added value: (1) consumption- and trade oriented policies can explicitly address issues like carbon leakage and for instance identify situations where an apparent reduction of carbon emissions in a country is mainly the result of structural change in which carbon-intensive industries were relocated abroad; and (2) such policies are more directly addressing consumption as a driver for rising greenhouse gas (GHG) emissions.
Aspects in the current EU climate policy landscape that require attention in evaluating consumption-based carbon policy instruments:
- Consumption-based emission accounting requires reliable emission data and robust methods to track embodied emissions in trade and consumption.
- A transformation towards low carbon consumption produces winners and losers. The changes in consumption patterns in big consuming countries impact producing countries.
- In order to reduce emissions embodied in trade and consumption, policies can intervene at the level of the producers, intermediaries or final consumers.
- Carbon pricing policies were often guided by the principle of implementing a carbon price as much upstream as possible in the expectation that the carbon price will feed through the value chain and thus incentivise all involved actors to shift their decisions toward lower carbon options.
- The analysis of food labelling approaches showed that numerous voluntary carbon labelling initiatives have emerged, mainly driven by the private sector, although some public bodies and international organization have also been involved.
- The effectiveness of a policy instrument to achieve a reduction in embodied emissions in trade and consumption depends also on its synergy and potential for conflict with existing (non-carbon) policy instruments.
The analyses in efficiency (emissions, energy, and labour per unit output), the changes due to trade related effects (both for intermediate producers and final consumers), the changes due to technology effects (both for intermediate producers and final consumers), and the change due to affluence and population indicate that trade is an important driver for global greenhouse gas emissions growth. However, it is not as important as growth in affluence and overall industry efficiency. This is only true, however, when looking at global emissions growth. When taking into account regional shifts in greenhouse gas emissions footprints over time, the displacement of industries from developed economies in the European Union and the OECD and the increase in imports to final demand contributes to emissions growth, mainly from combustion. For non-combustion emissions, changes in trade partners seems to decrease GHG footprints. Different dynamics act on the footprint growth over time and in different regions. Greenhouse gas emissions and energy consumption are mainly driven by the increase of consumption per capita in developing economies, such as China, and in the European Union. This growth in affluence reduces (or even reverses) gains in carbon and energy efficiency. It can be seen that trade is an important driver for labour footprints change in developed economies, in a higher proportion than for energy and greenhouse gas footprints. That indicates that the displacement of industries to labour-abundant countries might not have a significant effect in the growth of emissions embodied in trade.
Consumption based carbon accounting
The underlying question of the Carbon-CAP project is, to assess whether a consumption based carbon accounting and consumption based climate policy can have an added value to the already existing production based accounting and associated policy as a means to reduce GHG emissions. A consumption based approach differs from a production based approach in the definition of the system boundaries. A production based approach, requires a geographically identified system. A consumption oriented approach requires a functional, cradle-to-grave or “footprint” approach, usually including processes in different geographical areas. So far, policy has mainly focused on production and nations, and therefore, has used a territorial approach. This is also apparent when looking at GHG emission databases: they are organized by country and by activity. A consumption based approach needs something different. Requirements to a consumption based information base:
- GHG emissions should be linked to consumption activities and consumption categories
- GHG emissions should be specified on a cradle-to-grave basis
- The information base should provide information at a relevant spatial and time scale
- The quality of the information should be sufficiently reliable
- The information base should allow for analysing the past as well as forecasting the future, or rather, imagining the future under different assumptions.
Since the advent of environmental footprint approaches in general, and consumption based carbon accounting (CBCA) approaches specifically, many policy makers have been looking at ways to derive consumption-based policies. Whilst these efforts can be lauded, it has not been clearly established in the literature that consumption-based policies are more effective or more cost-effective than traditional policies based on control of territorial emissions. Further complicating the policy arena is that many policies could be considered both traditional and consumption-based (insulation of houses, for example). Alternatively, CBCA can be seen to be policy relevant, whilst not policy prescriptive. CBCA can give a key macro-level indication about the carbon intensity of an economy relative to baselines and targets. Such reporting of emission accounts can further underline the need for multi-lateral action, and for the increased responsibility needed to be shouldered by economies with growing net-import of emissions. CBCA can further strengthen resolve around uptake of instruments around, for example, the clean development mechanism (through encouraging investment from the developed world in trade partner countries in the developing world), or for the need for additional investment in emission offsets.
Promising consumption based policy measures
More than 30 policy instruments have been assessed in Carbon-CAP, covering products and services in Transport, Manufacturing, Food, Buildings, Paper/Plastics and Textiles. To assist with choices between policy instruments, a shortlist of promising instruments were ranked in three tiers. The first tier contains instruments that are judged to be strong across the four criteria of acceptability (economic, legal, international/political, institutional). The third tier contains instruments for which there is a significant barrier to acceptance on at least one of the criteria.
To effectively reduce emissions at the global level, consumption-based climate policy instruments will have to be part of the policy mix. Introducing instruments in a portfolio has three main advantages. First, consumer-oriented policy should not have the effect of wholly ‘individualising’ responsibility solely on end-users. It should spread responsibilities across many sectors, across consumers and across producers. Second, emissions are caused by many different decisions at many different levels from primary production to consumption to disposal. Consumer-oriented policies only act on part of these, and individual consumer-based instruments further focus the scope of application. Finally, experience has shown policies are often most effective when developed in mutually reinforcing ways since weaknesses in any one instrument can be counterbalanced by strengths of another instrument. This often helps in negotiations between groups implementing and affected by an instrument.
The assessments carried out by the project provide a useful first overview of promising instruments and a starting point for identifying opportunities and challenges to focus on in future deliberations and analyses. A key lesson is that consumer choice is difficult to influence when consumers have equal access to high and low carbon goods that meet the same needs. Therefore, the rankings of effectiveness and acceptability of instruments developed in this briefing reflect a tiered approach in which instruments that alter the range of products available, their ease of access and/or the cost (due to carbon charges) are applied first. The second and third ranks of instruments might then be considered means to support the instruments in the first rank. This is consistent with the lesson that instruments are most effective when introduced as complementary portfolios.
Modelling consumption based emission reduction
The modelling of the Carbon-CAP project assesses the effects of consumption-based emission reductions options on emissions and the economy. It focuses on the three main areas for improvement options: food, the built environment and transport. It uses a suite of three different models, each one based on different assumptions, to test the robustness of outcomes in relation to different modelling approaches.
The models used are E3ME, EXIOMOD and FIDELIO. E3ME is a macro-econometric model of the world’s economic and energy systems and the environment, based on post-Keynesian principles. EXIOMOD is a Computable General Equilibrium (CGE) Model. FIDELIO is a dynamic econometric input-output model that combines aspects of CGE models and linear ‘Input-Output philosophy’.
The IEA WEO 2014 current policies scenario is used as the main reference scenario for this report. In addition to the reference scenario and improvement options scenarios (food, transport and buildings), two additional sets of scenarios were included in the modelling exercise. The first additional scenario is the Nationally Determined Contributions Scenario (NDC), which is based on the pledges that were put forward at the Paris Conference of the Parties in 2015 and NDC plus improvement options. The second scenario is a Material Charge Scenario that considers a specific taxation instrument in Europe. Both the additional scenarios were modelled using E3ME only.
The improvement options in all three sectors that where selected for this study (food, buildings and transport) combined have a maximum potential to deliver household emission reductions at EU level of 47-67%, and total production-based emission reductions of 16-26% relative to the reference scenario in 2050. The impacts on production-based CO2 emissions outside the EU are small and mostly negative. However, when behavioural responses to individual policies, that could realise the improvement options, are considered, then there is a maximum 14% reduction (7.5-14%) in total CO2 production-based emissions reductions in the EU by 2050, relative to the reference scenario. These more modest reductions could be considered more realistic as many of the actions and policy measures considered were voluntary.
The most potential for reducing EU territorial CO2 emissions comes from the transport and buildings scenarios. The food options have less potential when CO2 only are considered. Therefore, because of the rebound effect a small increase in the EU households’ emissions is possible. All models show a slight decrease in consumption-based EU CO2 emissions under the combined food scenario (-0.2 to -3.6%). The range of impacts on consumption-based emissions is slightly wider range than the reductions in the territorial emissions (-0.3 to -1.2%).
The economic impacts of the combined scenario on the EU’s GDP are small, ranging from about 0.78% loss to a very small positive impact (0.06%) in 2050, depending on the model. The overall impact on employment in the EU in 2050 is very small and positive. Sectoral employment impacts vary depending on the improvement option and span from a 20% decrease in the vehicle manufacturing sector to a 61% increase in the provision of transport services.
The modelling undertaken in Carbon-CAP shows that various consumption-based policies relating to food, buildings and transport have considerable potential to reduce territorial CO2 emissions in Europe, especially with regards to household CO2 emissions. However, the reduction in consumption-based emissions is rather small and the relative gap between consumption- and production-based emissions is expected to increase. More detailed modelling will be required to estimate more comprehensively the total GHG emissions reduction potential of each of the individual policy packages and their implementation over the modelling time horizon 2020-2050. The consumption-based emission reduction measures used in this study have small impact on trade related emissions and therefore more attention should be paid to designing and assessing policies that address these emissions, including innovative measures such related to the options provided by the ITC sector. It should be also highlighted that there is high uncertainty related to modelling the policy impacts, especially on consumption-based emissions. Also using macromodels in addition to MRIO models for impacts assessments is feasible and desirable as it will help to asses feedback effects between in the economic system. Further development of model in this area is necessary to achieve more robust results, but differences in the results will remain as the models build on different theoretical backgrounds and these have to be considered while conducting impact assessments.
Further Research
Based on the feedback received through discussion with business stakeholders at the end of the Carbon-CAP project, it is also important to note the following, which could be explored in future related research projects. There is agreement amongst businesses consulted that this is an important conversation and there is interest on the part of business in discussing how to select the most effective policies for reducing consumption based emissions. Collaboration initiatives between government, business and consumers will be necessary to identify, develop and implement the most effective policies. Guidelines and criteria that governments should consider when engaging with business on new policy approaches for reducing emissions based on consumption might also be useful. More research may still be needed to iron out remaining uncertainties in the use of consumption-based accounting systems and developing whole-economy models capable of analysing and forecasting consumption-side emissions. More research could be undertaken on the full scope of the implementation of different consumption-side policy packages. In addition, Carbon-CAP established that it is possible to use consumption-based emissions from macroeconomic models as an indicator to assess future policies as well as allocating historical responsibility. This additional step could add substantial value to comprehensive policy assessments. However, there is a lot of work to be done on improving consumption-based emissions accounting in macromodels, as policy impacts on these emissions resulted in greater uncertainty than the impacts on production-based emissions. This need for further research does not preclude the possibility of the EU starting to recognise and quantify consumption-based emissions, and trade-embodied carbon, and to step up efforts to identify and implement new consumption policies with high levels of anticipated take-up and effectiveness.
Potential Impact:
The potential to complement current domestic GHG reduction efforts with policies that address consumption patterns is clear. However, developing a new, more balanced mix of policies will require overcoming various important gaps and hurdles. To overcome the gaps, the project answers the following specific scientific and technical objectives.
Gap 1: Quantification of global emissions related to consumption of goods and services and understanding drivers for upward trends.
As for the drivers, Carbon-CAP showed the following. The analyses in efficiency (emissions, energy, and labour per unit output), the changes due to trade related effects (both for intermediate producers and final consumers), the changes due to technology effects (both for intermediate producers and final consumers), and the change due to affluence and population indicate that trade is an important driver for global greenhouse gas emissions growth. However, it is not as important as growth in affluence and overall industry efficiency. This is only true, however, when looking at global emissions growth. When taking into account regional shifts in greenhouse gas emissions footprints over time, the displacement of industries from developed economies in the European Union and the OECD and the increase in imports to final demand contributes to emissions growth, mainly from combustion. This is caused by the fact that this displacement takes place to countries with in the period until now had a carbon intensive energy infrastructure, such as China. For non-combustion emissions, changes in trade partners seems to decrease GHG footprints.
Gap 2: Understanding of the levers, potential mechanisms, and feasibility of demand side tools and policies.
From previous research confirmed by Carbon-CAP, it is known that the areas of food, mobility and built environment dominate the life cycle impacts including carbon emissions from final consumption. Carbon-CAP identified a number of improvement options in these areas.
Two food-related improvement options – a reduction in food waste, and a switch to less emission-intensive diets – were selected for modelling. The two options are also modelled together as a combined food scenario. A selection of various voluntary and mandatory policy instruments can be used for shifting consumer behaviour towards these low carbon options, for example including supply chain procurement requirements and product labelling and educational campaigns.
Two buildings-related improvement options – a shift to near-zero emissions buildings (NZEBs) and an increased installation of natural fiber insulation for the existing housing stock (NFI) – were selected for modelling. Both options were also modelled together as a combined buildings scenario. In these scenarios, a selection of voluntary and mandatory policy instruments could be used for shifting consumer behaviour towards these low carbon options. The policies include standards, approved technology lists, financial incentives, product labelling and educational campaigns.
Transport has the largest number of sub-scenarios modelled, covering eight different transport-related improvement options. These include reducing the number of cars, enhancement of carpooling and sharing, lighter and smaller cars, and reduction of air transport. All the individual policy measures are also modelled together as a combined transport scenario. In these scenarios, a selection of voluntary and mandatory policy instruments could be used to incentivise a shift in consumer behaviour towards low carbon options; the policy measures include strengthening existing standards, introduction of new standards, infrastructure improvements, subsidies and educational campaigns.
Gap 3: Understanding of the effectiveness and impacts of demand side tools and policies.
The combination of improvement options and supportive policies were modelled with three different types of models:
1. E3ME, the global macro-econometric (Energy-Environment-Economy) E3 model of Cambridge Econometrics (CE). One of the strengths of this model is the underlying econometric specification, which provides a strong empirical basis for the analysis .
2. EXIOMOD is a Computational General Equilibrium (CGE) model that was recently developed by TNO. It is based on detailed EXIOBASE multi-regional environmentally extended input output tables (MREEIO), covering 43 countries, 5 rest of the world regions, 129 economic sectors and 40 GHG and non-GHG emissions.
3. FIDELIO is a model from the European Commission’s Joint Research Centre’s Institute for Prospective Technological Studies (IPTS). FIDELIO is a dynamic econometric input-output model based on Eurostat’s supply and use tables and the WIOD database covering 27 EU countries, 7 large countries outside Europe, 59 products/ economic sectors, 3 types of GHG emissions and 5 types of non-GHG emissions.
The modelling results show that various consumption-based policies have considerable potential to reduce territorial CO2 emissions in Europe, especially with regards to buildings (4-7%) and transport (4-14%). For food, direct emission reduction of e.g. limiting food waste is significant, but unfortunately the cheaper food basked then causes a rebound effect that annihilates this reduction. A limitation of the modelling is that CH4 is not included in E3ME and FIDELIO; EXIOMOD suggests the combined food scenario results in 14% reduction in methane emissions. Overall these three areas of consumption based options and policies contribute however just in a limited to the already planned production-oriented emission reductions. The net emission transfers between the rest of the world and Europe will stay stable or may slightly reduce, but as a percentage of total (reduced) emissions will grow.
A significant proportion of embodied carbon from imports is unavoidable. This is because it arises from foreign mining operations which are not conducted in the EU, and domestic alternative materials do not exist. In relation, the consumption-based emission reduction measures used in this study have small impact on trade related emissions and therefore more attention should be paid to designing and assessing policies that address these emissions. Policies addressing rebounds are highly relevant too.
Gap 4: No shared view on the added value, implementation challenges and acceptability of demand side tools/policies and related accounts, and no “roadmap” of evolution from production towards consumption-based policies.
There are a variety of databases, tools, methods and models that can be used to assess aspects of consumption based carbon emissions. For assessing the past and the present, environmentally extended Input Output analysis (possibly hybridized with LCA) seems to be the most ideal approach. Our research showed that by far the largest uncertainty in such carbon footprint analysis comes from data already used widely in climate negotiations: sector and country level CO2 emission data. Jus harmonizing such data across IO databases reduces uncertainty in footprint analyses with 50%. In a similar way, forward looking models such as E3ME, EXIOMOD and FIDELIO could be harmonized. This will allow for a robustness of calculating past and future consumption based carbon emissions with a level of uncertainty that is not significantly worse as current information on country- or sector level carbon emissions in the past, or carbon emissions modelled for the future.
Consumption oriented carbon reduction policies have added value and can support an additional reduction of carbon emissions compared to current scenarios. In this, implementing policies addressing rebound effects (like carbon taxes) are essential. Policies that would also address trade-related emissions directly or even indirectly however do not sit well in the UNFCCC process. If introduced formally, administrative challenges and other objections for instance if compatible with WTO rules will be the result. It seems much more fruitful that given a specific situation with regard to trade related emissions, trade partners may agree on a voluntary basis to share responsibilities for such emissions differently as analysed via a purely production based approach. Overall, it is therefore clear that consumption oriented policies should be seen as a complement to the traditional production oriented approach.
Further research
Based on the feedback received through discussion with business stakeholders at the end of the Carbon-CAP project, it is also important to note the following, which could be explored in future related research projects. There is agreement amongst businesses consulted that this is an important conversation and there is interest on the part of business in discussing how to select the most effective policies for reducing consumption based emissions. Collaboration initiatives between government, business and consumers will be necessary to identify, develop and implement the most effective policies. Guidelines and criteria that governments should consider when engaging with business on new policy approaches for reducing emissions based on consumption might also be useful. More research may still be needed to iron out remaining uncertainties in the use of consumption-based accounting systems and developing whole-economy models capable of analysing and forecasting consumption-side emissions. More research could be undertaken on the full scope of the implementation of different consumption-side policy packages. In addition, Carbon-CAP established that it is possible to use consumption-based emissions from macroeconomic models as an indicator to assess future policies as well as allocating historical responsibility. This additional step could add substantial value to comprehensive policy assessments. However, there is a lot of work to be done on improving consumption-based emissions accounting in macromodels, as policy impacts on these emissions resulted in greater uncertainty than the impacts on production-based emissions. This need for further research does not preclude the possibility of the EU starting to recognise and quantify consumption-based emissions, and trade-embodied carbon, and to step up efforts to identify and implement new consumption policies with high levels of anticipated take-up and effectiveness.
List of Websites:
carboncap.eu
