Review and update of UK abatement costs curves for the industrial, domestic and non-domestic sectors RM 4851
Final report to the Committee on Climate Change
Restricted Commercial ED43333 Issue 3 August 2008 2
Restricted – Commercial Review and update of UK abatement costs curves for AEA/ED43333/Issue 3 the industrial, domestic and non-domestic sectors
Title Review and update of UK abatement costs curves for the industrial, domestic and non-domestic sectors
Customer Committee on Climate Change
Customer reference RM 4851
Confidentiality, This report is the Copyright of Defra and has been prepared by AEA copyright and Technology plc under contract to Defra. The contents of this report may not reproduction be reproduced in whole or in part, nor passed to any organisation or person without the specific prior written permission of Defra. AEA Technology plc accepts no liability whatsoever to any third party for any loss or damage arising from any interpretation or use of the information contained in this report, or reliance on any views expressed therein.
File reference
Reference number ED43333 – Final report (Issue 3)
Heather Haydock AEA Energy & Environment The Gemini Building Fermi Avenue Harwell International Business Centre Didcot OX11 0QR
t: 0870 190 6487 f: 0870 190 6318
AEA Energy & Environment is a business name of AEA Technology plc
AEA Energy & Environment is certificated to ISO9001 and ISO14001
Author Name Steve Pye (AEA) Ken Fletcher (Carbon Consortium) Ann Gardiner (Ecofys) Tana Angelini (Ecofys) James Greenleaf (Ecofys) Tricia Wiley (AEA) Heather Haydock (AEA)
Reviewed by Name Heather Haydock
Signature
Date 19th August 2008
AEA Energy & Environment iii Review and update of UK abatement costs curves for Restricted – Commercial the industrial, domestic and non-domestic sectors AEA/ED43333/Issue 3
iv AEA Energy & Environment Restricted – Commercial Review and update of UK abatement costs curves for AEA/ED43333/Issue 3 the industrial, domestic and non-domestic sectors
Executive summary
This report by AEA Energy & Environment, Ecofys and the Carbon Consortium is the final report under the contract to review and update cost curves for energy end use sectors. This project forms a key part of ongoing work being undertaken by the Committee on Climate Change (CCC) to help inform the requirements set out by the Climate Change Bill. These include:
The setting of 5-yearly carbon budgets The degree to which the emissions reduction effort needed to meet these budgets is shared between: domestic and overseas activity (i.e. the level of subsidiarity); the traded sectors (EU ETS and CRC) and non-traded sectors (transport, households, etc).
To inform the above, the CCC is developing an integrated marginal abatement cost curve (MACC) for the UK, covering all sectors.
The objective of this study is to review and update the MACCs associated with the industry and builidngs sectors (known here as domestic and non-domestic sectors), and integrate where appropriate other cost curve work being undertaken by Government on heat, products and microgeneration.
The three main sector-based MACCs came from the ENUSIM model (industry sector) and BRE models – BREDEM (Domestic) and N-DEEM (Non-domestic). The key updates to the industry model were revision of assumptions across three key sectors – chemicals, food and drink and engineering.1 This included some preliminary consultation with these sectors, to consider key assumptions made and industry data for incorporation. For buildings sectors, a key focus was the integration of information on heat-based technologies, microgeneration, behavioural measures and energy using products.
In addition, a key part of ths work was to provide cost curve information from sector energy models as flexible spreadsheet tools, to be used by the CCC. This included flexibility such as changing discount rates, fuel prices and emission factors, and including / excluding hidden and missing costs.
The MACCs provide an understanding of the remaining abatement potential (both cost-effective and not) at a given point in time based on a set of assumptions about future fuel prices, discount rates etc. This view of technical potential, however, does not provide a clear picture of whether this could be achieved or realised at a given point in time; it only reflects what has already been taken up – in the reference case2 – and what remains. Therefore, in this study, we have ensured that the spreadsheets have the functionality to incorporate realistic potential % factors, which when applied to the technical potential give a more realistic estimate of what could be achieved by a given year. These are very uncertain across all sectors and require additional development. For the non-domestic sector, they remain as default values. CCC plans to build on this work to develop a better understanding of the realistic potential and to explore the effects of using different data and methodological assumptions.
Domestic sector
The domestic building sector offers significant opportunities for cost-effective abatement at around 36 MtCO2, as shown in the figure below. This is technical potential (based on all possible interactions); a realistic abatement potential in 2022 of 16 MtCO2 has been estimated. In the technical potential, significant opportunities arise from the introduction of more efficient products and solid wall insulation (SWI). The realistic potential is much lower, particularly due to much lower levels of SWI uptake being assumed.
Microgeneration and district heating offer significant abatement opportunities, albeit not cost-effective. Potential of such measures is approximately 60 MtCO2, but with only 3 MtCO2 cost-effective.
1 EU ETS sectors were subject to significant revision in 2005. 2 Based on policies announced prior to the EWP 07.
AEA Energy & Environment v Review and update of UK abatement costs curves for Restricted – Commercial the industrial, domestic and non-domestic sectors AEA/ED43333/Issue 3
Domestic buildings sector (2022, 3.5% DR, measures in isolation)
1000
53 800
600
400
50 51 200
£/tCO2 48 49
47 0 41 39 40 Total 97.4 MtCO2 32 34 21 26 141720 -200 22 8 9
6 5 -400 2 3
-600 0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100
MtCO2
1 Reduced standby consumption 19 Loft insulation 0 - 270mm 37 A-rated condensing boiler 3 2 ICT products 20 Pre76 cavity wall insulation 38 Loft insulation 150 - 270mm 3 Electronic products 21 Glazing - old double to future double 39 Fossil DH to CHP 4 Integrated digital TVs 22 DIY floor insulation (susp. timber floors) 40 Glazing - new double to future double 5 A+ rated wet appliances 23 A-rated condensing boiler 2 41 Residential biomass (off gas grid) 6 A++ rated cold appliances 24 Room thermostat to control heating 42 Urban retrofit DH 7 A rated ovens 25 Loft insulation 25 - 270mm 43 A-rated condensing boiler 4 8 A-rated condensing boiler 0 26 76-83 cavity wall insulation 44 Paper type solid wall insulation 9 City centre DH 27 Loft insulation 50 - 270mm 45 A-rated condensing boiler 5 10 Efficient lighting 28 Loft insulation 75 - 270mm 46 Modestly insulated cyl to high performance 11 Induction hobs 29 Thermostatic radiator valves 47 mini wind turbines 12 A-rated condensing boiler 1 30 Post '83 cavity wall insulation 48 Photovoltaic generation 13 Glazing - old double to new double 31 Improve airtightness 49 Heat pumps (off gas grid) 14 Glazing - single to new 32 Installed floor insulation (susp.TFs) 50 Micro-CHP Fuel cell 15 Insulated doors 33 Electric high rise to CHP 51 Solar water heating 16 Insulate primary pipework 34 Solid wall insulation 52 Hot water cylinder 'stat 17 Uninsulated cylinder to high performance35 Loft insulation 100 - 270mm 53 micro wind turbines 18 Glazing - single to future double 36 Loft insulation 125 - 270mm
Non-domestic sector
The figure below shows a comparable picture of abatement opportunities for the non-domestic sector. Approximately 60% of the cost-effective potential comes from heating based measures, with the rest from more effiicient products in the main.
vi AEA Energy & Environment Restricted – Commercial Review and update of UK abatement costs curves for AEA/ED43333/Issue 3 the industrial, domestic and non-domestic sectors
Non-domestic buildings sector (2022, 3.5% DR, measures in isolation)
900
700 53 54
500 50
300 £/tCO2 49
46 100
-100 32 24 25 14 7 11 20 2 4 6 -300 0 5 10 15 20 25 30 35 40 45 50
MtCO2
1 Lights- Most EE Replacement Tungsten Total 31 OffEq - Most EE Monitor Total 2 Heating - More efficient air conditioning Total 32 presence detector Total 3 Compressed air Total 33 Motor - 4 Pole Motor - EFF1 replace 4 Pole Total 4 Heating - Optimising Start Times Total 34 Heating - TRVs Fully Installed Total 5 Lights - Sunrise-Sunset Timers Total 35 Heating - most EE boiler Total 6 Heating - Programmable Thermostats High Total 36 Most EE pitched roof insulation Total 7 Lights - Basic Timer Total 37 Most ee double glazing Total 8 Lights - Light Detectors Total 38 Most EE flat Roof insulation Total 9 stairwell timer Total 39 Most EE external wall insulation Total 10 Most EE fridge Total 40 Lights - HF Ballast Total 11 Heating - Reducing Room Temperature Total 41 Most ee cavity wall insulation Total 12 Most EE freezer Total 42 Biomass 13 Photocopiers - Energy Management Total 43 Lights- Most EE Replacement Tungsten Total 14 Monitors - Energy Management Total 44 Most EE fridge-freezer Total 15 Printers - Energy Management Total 45 Lights - Metal Halide Floods Total 16 OffEq Fax Machine switch off Total 46 Lights - IRC Tungsten-Halogen - Spots Total 17 Vending Machines Energy management Total 47 Variable Speed Drives Total 18 OffEq - Most EE Monitor pc only Total 48 Heating - most EE boiler Total 19 Computers - Energy Management Total 49 Most EE fridge-freezer Total 20 Lights - Turn off Lights for an extra hour Total 50 Most EE freezer Total 21 OffEq - Most EE Monitor Total 51 Most EE fridge Total 22 presence detector Total 52 Most ee cavity wall insulation Total 23 Motor - 4 Pole Motor - EFF1 replace 4 Pole Total 53 Lights- Most EE Replacement 26mm Total 24 Heating - TRVs Fully Installed Total 54 Most EE pitched roof insulation Total 25 Heating - most EE boiler Total 55 Most EE flat Roof insulation Total 26 Most EE pitched roof insulation Total 56 Lights - HF Ballast Total 27 Most ee double glazing Total 57 Lights- Most EE Replacement 26mm Total 28 Most EE flat Roof insulation Total 58 Most EE external wall insulation Total 29 Most EE external wall insulation Total 30 Lights - HF Ballast Total
There is large carbon abatement potential in the all measures in isolation case, shown above, however, less than half is cost effective. The main cost effective measures are lighting measures (for example replacement with the most energy efficient), and also specific wall and roof insulation measures. Heating measures tend to be one of the most cost effective measures, and have high abatement potential. Where all possible interactions are taken into account there is a fairly large reduction in the abatement potential.
AEA Energy & Environment vii Review and update of UK abatement costs curves for Restricted – Commercial the industrial, domestic and non-domestic sectors AEA/ED43333/Issue 3
Industry sector
In 2022, the industry sector could provide around 9 MtCO2 of cost-effective abatement potential, as shown in the figure below. This is approxmately 30% lower than what is available in 2008, due to continuing uptake of measures under a business-as-usual (reference) case. Due to the number of measures included in the modelling across a range of different sub-sectors, characterising the measures making up this abatement potential is not as straightforward as for the sectors above.
Industry MACC in 2022
2022
200 Original w ith 15% DR (+ user selected EFs) 150 User selected variables (exc. H&M costs) User selected variables + Low H&M costs 100 User selected variables + High H&M costs
50
0
£/tCO2 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 -50
-100
-150
-200 MtCO2
As with all of the sector MACC curves, these are sensitive to assumptions on fuel prices, discount rates and additional costs (as shown in the figure above). Sensitivity analysis has been undertaken in this analysis on such variables, and is presented later in this report.
CHP
New CHP modelling has been undertaken as part of this work to better understand the potential of such a technology in future years. Estimates of the technically achievable potential by year increases from 700 ktCO2 pa at 2008 to almost 10 MtCO2 by 2022, with a projected cost effective contribution of almost 7.5 MtCO2 by 2022. Realistic potential is estimated to be between 40-50% of the technical potential. For the non-doemstic sector, a technical potential of around 8 MtCO2 has been estimated, with realistic potential estimate again at around 50%.
Given the large potential offered by CHP, it is considered very important to explore what might be realistic in more detail. This will be done to some extent through the industry consultation exercise.
In addition to the development of cost curves, some preliminary analysis was done to assess the overlap of policies announced as part of the Energy White Paper 2007 (EWP 07). It is clear that much of the cost-effective potential across different sectors MACCs is likely to be taken up by policies announced as part of the EWP 07. For example, the proposed new Suppliers Obligation would lead to increasing uptake of cost-effective insulation measures in exisitng homes. Similarly, product policy aims to deliver further increases in the uptake of cost-effective energy-using appliances. However, many abatement opportunities still exist for which policy development may be required to ensure increasing take-up.
viii AEA Energy & Environment Restricted – Commercial Review and update of UK abatement costs curves for AEA/ED43333/Issue 3 the industrial, domestic and non-domestic sectors
Through review and integration of other work, the energy end use sector cost are more comprehensive than previous versions, and are sourced from the most up-to-date available information. However, improvements are still required, and need to be considered. These include:
Need for a better understanding of what is realistic potential. The sector-based cost curves have tended to have been constructed to assess technical potential. Further work is essential to improve the understanding of realistic potential across all sectors, in particular non-domestic which remains undeveloped. Data improvement. Currently a consultation is being undertaken across the three main industry sectors that were updated as part of this work. This process will help to improve the robustness of the estimates, although further detailed consultation is still recommended. Interaction of measures. Further analysis is required in the building sectors to understand the level of implementation of new measures in a given year, and the resulting interactions. This potentially requires scenario-based analysis, and further modelling.
AEA Energy & Environment ix Review and update of UK abatement costs curves for Restricted – Commercial the industrial, domestic and non-domestic sectors AEA/ED43333/Issue 3
x AEA Energy & Environment Restricted – Commercial Review and update of UK abatement costs curves for AEA/ED43333/Issue 3 the industrial, domestic and non-domestic sectors
Acknowledgements
Many different people have contributed their expertise to this study, most notably:
. On building sector cost curves, the team at BRE including Les Shorrock, Christine Pout, Jon Henderson and Fiona Mackenzie . On CHP, Mahmoud Abu-ebid and his team at AEA . On the review of building sector cost curves, Chris St John Cox at AEA . On the product policy cost curve, MTP colleagues at AEA, Fiona Brocklehurst and Sarah Winne . On the industry cost curve, Ernst Worrell and Maarten Neelis of Ecofys . On the heat cost curve, Jonathan Stern at the OCC
In addition, Mark Weiner from the CCC Secretariat has played a key role in advising on different aspects of this analysis, and co-ordinating input between different organisations inputting to this work.
AEA Energy & Environment xi Review and update of UK abatement costs curves for Restricted – Commercial the industrial, domestic and non-domestic sectors AEA/ED43333/Issue 3
Table of contents
1 Introduction 1 1.1 Key issues to address 1 1.2 Approach to study 3 2 General issues arising from review 6 3 Review and update of industry sector MACC 10 3.1 Review of industry sector MACC 10 3.2 Update of industry sector MACC 12 3.3 Realistic potential scenario 23 3.4 Emissions gaps Gaps in the coverage of sector emissions in the MAC curves 24 4 Review and update of building sector MACCs 26 4.1 Generic issues for domestic / non-domestic sectors 26 4.2 Domestic Sector MACC 28 4.3 Non-domestic sectors (commercial / public sector) 29 4.4 Update of building sector MACCs 32 5 Integration of cross-sectoral MACCs 45 5.1 Review of cross-sectoral cost curves 45 5.2 Proposal for integration: Heat technologies 52 5.3 Proposal for integration: microgeneration 59 5.4 Proposal for integration: Product policy MACC 60 5.5 Proposal for integration: Behavioural measures 62 6 MACC outputs 64 6.1 Industry 64 6.2 Buildings (domestic) 66 6.3 Buildings (non-domestic) 69 7 Comparison of abatement potential with the impact of policies 71 7.1 Supplier Obligation 71 7.2 Zero carbon homes 72 7.3 More energy efficient products 72 7.4 Carbon Reduction Commitment (CRC) 72 7.5 Climate Change Agreements (CCAs) 73 7.6 EU Emissions Trading Scheme (EU ETS) 74 7.7 CHP measures 74 8 References 75
xii AEA Energy & Environment Restricted – Commercial Review and update of UK abatement costs curves for AEA/ED43333/Issue 3 the industrial, domestic and non-domestic sectors
Appendices Appendix 1 Sectoral growth rate projections Appendix 2 Fuel price projections Appendix 3 Relating the MACC annualised costs to the cost of targets using the DCF (NPV) calculation Appendix 4 An overview of ENUSIM Appendix 5 Description of ENUSIM Appendix 6 ENUSIM file structure and process steps Appendix 7 ENUSIM based bottom-up modelling approach – methodology linking MAC curves to target or policy costs Appendix 8 Hidden and missing costs Appendix 9 Industry consultation to inform ENUSIM estimates
AEA Energy & Environment xiii
Restricted – Commercial Review and update of UK abatement costs curves for AEA/ED43333/Issue 3 the industrial, domestic and non-domestic sectors
1 Introduction
This report by AEA Energy & Environment, Ecofys and the Carbon Consortium is the final report under the contract to review and update cost curves for energy end use sectors. This work is being undertaken on behalf of the Committee on Climate Change (CCC).
This project forms a key part of ongoing work being undertaken by the CCC to help inform the requirements set out by the Climate Change Bill. These include: The setting of 5-yearly carbon budgets The degree to which the emissions reduction effort needed to meet these budgets is shared between: domestic and overseas activity (i.e. the level of subsidiarity); the traded sectors (EU ETS and CRC) and non-traded sectors (transport, households, etc).
To inform the above, the CCC is developing an integrated marginal abatement cost curve (MACC) 3 for the UK, covering all carbon emitting sectors including: Electricity generation Industry Domestic / non-domestic buildings Transport
A MACC provides an understanding of the cost of abating successive units of pollution, and provides an understanding of the potential reductions that are still possible at the different abatement cost levels.
The objective of this study is to review and update the MACCs associated with the industry and builidngs sectors (known here as energy end use sectors), and integrate where appropriate other cost curve work being undertaken by Government on heat, products and microgeneration.
This review and update exercise is important for the improvement of MACCs, and involves comprehensively identifying options for carbon reduction, properly assessing the full costs of adopting energy efficiency and other carbon reduction measures, and in doing providing robust advice to Government concerning carbon budgets.
1.1 Key issues to address
This review and update of the energy end use MACCs focuses on a range of key issues, many having been identified by a previous review undertaken by Enviros (2006) as part of the Energy Efficiency Innovation Review (EEIR). In this report, we have both reviewed each issue for the different cost curves (and associated models) and proposed further developments to the approach where necessary.
These can be broadly split into methodological issues and data issues, and are listed in Table 1.1 below.
3 The transport sector MACC is being developed by AEA (technology options) under a seprate contract and Frontier Economics (demand side measures) whilst the electricity generation MACC is being developed by the CCC. The integration of these MACCs will be done using a Framework being developed by CCC and external consultants.
AEA Energy & Environment 1 Review and update of UK abatement costs curves for Restricted – Commercial the industrial, domestic and non-domestic sectors AEA/ED43333/Issue 3
Table 1.1 Key issues for consideration in review and update of Energy End Use MACCs
Issues Description Data issues Consistency and To ensure comparability and a robust approach, in particular on fuel prices and selection of underlying discount rates. assumptions Hidden and Missing These cover ‘missing costs’ incurred by the project “host” during different stages Costs of the project’s cycle4 and not always quantified fully in the engineering costs of a project. Examples include the time taken to identify or appraise a project and disruption to production. In addition, there are also ‘hidden costs’ (such as perceived delivery risk and ongoing management/supervision time) that are not necessarily reflected in the cost of the abatement measures. Lack of model To ensure that all assumptions concerning data and methodology are available transparency for review and clearly presented. Coverage of measures, While most of the data in the cost curves is of high and recent quality, for some and quality of the data sectors the underlying data is nearly 10 years old. For some sectors, behavioural measures have not been included. Emerging measures available in future years also need to be considered. Sufficient The MACCs must be able to provide a sufficient level of detail for the CCC (e.g. disaggregation of output in terms of setting 5-yearly carbon budgets). In addition, more recent policy data developments such as the Carbon Reduction Commitment (CRC) require the ability to focus on different sections of the UK economy. Methodological issues Timing of measures A standard cost curve provides an indication of the cost of achieving various levels of potential emissions reduction, but over an infinite timescale. The MACCs need to properly account for the period over which this potential can be realised so there is a need to differentiate between replacement measures (which, for example, may need to be brought forward and hence greater cost is incurred than during end-of-life replacement) and those which can be installed at any point. Technological progress Alternative approaches for incorporating expectations on technological progress and availability of new technologies in the MACCs, either directly in the form of specific new technologies or indirectly via an assumption of continuous historic improvements in energy intensity over time. Limited account of Need to avoid double counting of emissions savings that, for example, occur due interactions/overlaps to the interaction of measures aimed at more efficient energy end-use (such as between measures insulation), and more efficient supply of energy (e.g. CHP) (and adequate accounting of the diminishing returns obtained from the addition of multiple energy efficiency measures). Penetration rates of The rate at which measures are taken up needs to account for a number of measures elements – uptake in baseline, policy impact, shape of penetration curve and structural shifts. These may not be accurately reflected by existing approaches:
Where the above issues are not properly taken into account, there is the real possibility of underestimating / overestimating cost-effective potential. Therefore, it is crucial that the above issues are modelled in as robust a way as possible.
In addition to reviewing and updating core sector-based sector MACCs, it is also important that the latest developments in MACC development are integrated into the wider UK MACC. In particular, we are considering four areas of cost curve development undertaken in the last 12 months, where these are relevant to the buildings and industrial sectors:
. Heat, including renewable heat, district heating and CHP . Energy using products . Micro-generation . Behavioural measures
4 i.e: Project identification; project appraisal; project commissioning; production disruption and additional engineering.
2 AEA Energy & Environment Restricted – Commercial Review and update of UK abatement costs curves for AEA/ED43333/Issue 3 the industrial, domestic and non-domestic sectors
1.2 Approach to study
The programme of work was split into four tasks:
1.2.1 Task 1: Review of existing MACCs
The objective of this task was to establish a common understanding of the existing work on MACCs as a starting point for informing methodological development and updating under this project. This process included review of existing studies, independent review of models / cost curve data and consultation.
The following MACCs have been reviewed as part of this analysis:
1. Industry. Cost curves sourced from the ENUSIM model 2. Buildings sector (Domestic / non-domestic). Cost curves developed from BRE building models (BREDEM and N-DEM). BRE have supplied most of the information for these sector MACCs. 3. Cross-sectoral MACCs. There are a range of other MACCs that have been developed by other government departments including a heat cost curve (OCC), products policy cost curve (Defra), and microgeneration cost curve (Defra). As part of this project, these cost curves have been reviewed, and proposals made concerning their integration within a full UK MACC.
For all cost curves except that associated with ENUSIM, consultation has been undertaken with respective experts. We consider that the expertise within the team concerning ENUSIM is adequate for a comprehensive understanding of the issues.5
A review framework was devised against which to assess the different cost curves against the key issues described in Section 1.1. This framework provided the basis for consistent review, and enables comparison across different cost curve areas. This exercise underpinned our understanding of different approaches and assumptions, and the extent that integration is possible with current cost curves as developed. The review framework used the criteria listed in Table 1.2.
5 Note that specifc consultation on indsutry sector data issues has been undertaken as part of this study.
AEA Energy & Environment 3 Review and update of UK abatement costs curves for Restricted – Commercial the industrial, domestic and non-domestic sectors AEA/ED43333/Issue 3
Table 1.2 Review framework criteria
Issue Information required Purpose of MACC Policy appraisal or technical potential assessment? Fuel prices / emission Are future fuel prices consistent with BERR projections? What emission factors factor are used for carbon, particularly in relation to electricity Time horizon Years covered by the cost curve – do they cover those periods considered by the CCC analysis? Baseline measures What measures are included in the baseline? Is this consistent with BERR’s UEP 29 projection (excluding measures announced under the EWP 07)? Sector disaggregation At what level of disaggregation are the cost curves available? How many, and which sub-sectors are explicitly represented? Spatial disaggregation Do cost curves provide information for the UK only, or is this information disaggregated by Devolved Administration (DA)? Coverage of measures Is the coverage of measures comprehensive? Are both technical and behavioural measures included? Potential of measures What are the key sources of information informing this? Penetration rates What is the source of information for penetration rates, and does it appear to be from robust sources? Hidden and missing Are hidden and missing costs included in the costs across the cost curve costs measures? Other costs Are there any other costs included e.g. policy costs, damage costs? Measure of cost- What is the measure of cost-effectiveness – NPV divided by lifetime CO2 effectiveness abatement OR net annualised cost divided by average annual CO2 abatement? Investment costs / What investment costs are provided, and what is the assumed investment period? investment period Discount rates What are the rates used, reflecting either a private sector perspective (e.g. 15%) or a societal perspective (3.5%)? Interaction between Are the interactions between two options taken up captured in the cost curve, and measures if so, how? Timing of measures Are the additional costs of early replacement reflected in the costs of the replacement measure?
1.2.2 Task 2: Development of methodology for updating and improving the MACCs
Based on the review of the different MACCs, the objective of this task was to propose methodologies for updating and improving cost curves. The extent of methodological development of cost curves is dependent on:
Understanding of the extent to which the issues (outlined in Section 1.1) have already been addressed in the different cost curves. Insights into why the methodological approaches currently deal with the issues in the way that they do. Knowledge of the gaps in or limitation of the current approaches, in particular the impact on the robustness of the final MACCs, and the scope for further development.
In considering these it is important to realise that each of the models (ENUSIM and the BRE models) has evolved independently over the years and that the Enviros study was only partially able to bring some uniformity into the assumptions used and the modelling processes. This project had the objective of bringing the disparate models closer in both their assumptions and in their treatment of factors such as hidden costs, interactions between different measures, accelerated scrapping effects and adequate treatment of emerging technologies.
Many of the issues that need addressing in terms of methodological development of cost curves and consistency of approach were highlighted in the review of cost curves for the Energy Efficiency Innvoation Review (Enviros 2006). This work builds on that previous analysis.
4 AEA Energy & Environment Restricted – Commercial Review and update of UK abatement costs curves for AEA/ED43333/Issue 3 the industrial, domestic and non-domestic sectors
1.2.3 Task 3: Provision of updated MACCs for the domestic, commercial and industrial sectors
Revised MACC curves were developed, and cost-effective savings analysed. Sensitivity of different parameters were tested to identify the impact on cost-effectiveness. The process and approach to development of the MACCs is described in Sections 3 to 5, while the outputs are presented in Section 6.
1.2.4 Task 4: Comparison of abatement potential identified in the revised MACCs with the impact of Government policies and estimates of emissions reductions
The CCC is interested in understanding the difference between what recently announced policies will deliver and the remaining potential. This task reviews estimates for a number of different policies announced in the Energy White Paper 2007 measures.
Policies include:
. Supplier Obligation . Zero Carbon Homes . More energy efficient products . Carbon Reduction Commitment (CRC) . EU ETS . Continuation of CCA's
The analysis is presented in section 7.
AEA Energy & Environment 5 Review and update of UK abatement costs curves for Restricted – Commercial the industrial, domestic and non-domestic sectors AEA/ED43333/Issue 3
2 General issues arising from review
During the course of the review of MACCs and discussion with CCC, a range of issues pertinent to all cost curves were considered. This section describes some of these key methodological issues. The subsequent sections focus on the review and updates associated with specific MACCs.
2.1.1 Private versus Social discount rates
Discounting is used to ensure comparability in a given year of costs and benefits that will be accrued in future time periods. It is based on a principle of time preference, where an individual prefers to receive goods and services now than in future periods, whilst defering costs to future years. A societal view of this is the social time preference. Interest rates on loans reflect this increased emphasis on consumption now, whilst incurring the additional costs of borrowing through future interest payments.
Different discount rates have been applied across different cost curve analyses. The question is why you would use different rates in different analysis.
From a private perspective, an analysis is trying to reflect the real costs of investments faced by individuals and firms. Private interest rates measure the costs faced by private individuals and firms when making investment decisions. These rates are complex and varied, and may reflect the cost of borrowing, of raising equity, or simply of not investing in something else (where available capital is constrained). These can range between 8-25%, depending on the type of investment.
From a societal perspective, the use of a lower discount rate to reflect society’s view of abatement costs and investment risk, as required under Government policy, is used. Broadly, this is set at 3.5%.
Social discount rates (SDR) reflect society’s preferences over time, reflecting some elements of risk as well as changes in income and marginal utility (the ‘happiness’ an extra £1 of income brings) over time. Society prefers money (consumption) now over the future because a) society will be richer in the future so the enjoyment they receive from an extra £1 will be lower, b) there may be an element of impatience (people don’t want to have to wait) c) there is a risk £1 in the future won’t happen (or society won’t be around to enjoy it).
Which type of perspective to take generally depends on the purpose of the analysis. Much of the past work on MAC curves has tended to focus only on the MAC curve from a private perspective – what is cost effective for the individual or firm? In this case the private discount rate has been used since it reflects the costs that a private person faces when taking decisions.
The work of the Committee poses a different question. From the perspective of the society, what is the abatement potential and how much does it cost. In this case we need to know two things: . The real resource cost to the economy of investing in a particular project. . The time profile of costs and benefits – and how society values that profile.
The latter is not controversial and is reflected in the SDR. However the former presents more problems. It is clear that this cost reflects the physical resources taken to build the project (e.g. the manufacture of a new boiler). However it should also include any resource cost associated with funding/investing in that project. At its simplest level this is the opportunity cost of not investing in something else. Private interest rates reflect a constraint on the available capital, so for instance investing to reduce emissions means foregoing investment elsewhere which would have had a return higher than the SDR. This represents a real resource (opportunity) cost to the economy. By not investing in the alternative we are foregoing higher production in the future, hence the private cost of capital should be included in the appraisal of costs and benefits. 6
Other issues relating to discount rates include whether they should be sector specific, and be applied consistently across all measures, or whether discount rates should be considered on a measure-by- measure (or project basis). Within ENUSIM for example, past practice, which has aimed at simulating industrial behaviour, has used different ‘private rates’ depending upon the type of technology being
6 Much of the above arguments have been developed by the CCC Secretariat, and are further described in internal CCC papers.
6 AEA Energy & Environment Restricted – Commercial Review and update of UK abatement costs curves for AEA/ED43333/Issue 3 the industrial, domestic and non-domestic sectors considered, with ‘strategic investments’ carrying a lower rate (8-10%) than smaller energy efficiency type investments (15%+).
The current approach proposed by the CCC secretariat is to apply the social discount rate and apply a mark up where capital costs can be shown to represent an additional cost.
2.1.2 Purpose of MACCs
An important question that needs to be considered in the context of cost curve development is what is the purpose of the cost curve? Cost curves are being developed to support different types of analysis:
Assessments of technical abatement potential (provided by BRE and ENUSIM models). Such cost curves provide an understanding of cost-effective potential, and total abatement potential after accounting for current policy. This type of assessment does not tend to limit penetration of measures (except on a temporal basis) as it is showing what could be achieved if the right policies were put in place (independent of timescale limitations).
Assessment of abatement potential under different policy scenarios (provided by product cost curve). This type of assessment is much more focused on understanding cost-effective potential under a set of assumptions relating to development policy (policy scenarios) as carried out for regulatory impact assessments.
Both are relevant depending on the question that is being addressed. For this current work, the objective is to build a cost curve that reflects total technical and cost-effective abatement potential. However, the CCC has asked that we also consider a realistic potential, which is reflecting more our understanding of what policy is capable of delivering given timescales, practicalities and the strength of the multiple barriers and market failures that stand in the way of a greater uptake of abatement measures. ,Reflecting this level of realism is reflected in the estimates, important for understanding the processes of carbon budget setting. In this type of MAC curve the trend is for increasing potential with time rather than the more typical reduction in potential with time as technology take up in the reference scenario tends to reduce the level of available ‘remaining’ technical potential with time; as typically seen with ENUSIM and BRE MACC outputs.7
Figure 2.1 illustrates the difference between the reference case, realistic potential and technical potential concerning uptake of a given measure. The reference case is the solid S-curve, showing what is being taken up under a BAU case. Realistic potential is the difference between the solid line and the dashed line, showing an increase and more rapid uptake over time. Technical potential is the defined by the straight line at the top of the graph, and is estimated as the difference between the BAU case and technical potential.
7 This is the case on an individual measure basis; however, a factor that could increase technical potential over time is allowing for new technologies to become available in future years.
AEA Energy & Environment 7 Review and update of UK abatement costs curves for Restricted – Commercial the industrial, domestic and non-domestic sectors AEA/ED43333/Issue 3
Figure 2.1 Uptake of measures over time in a MACC
Penetration Technical potential may be less than rate 100% take up 100% Technical Potential Technologies will be taken up over time (even without further policy interventions)
MAC Curves look at ‘Technical potential’ – the gap between business as usual take up and the technical potential
Defining realistic potential will be the key challenge for the MAC curves and budget setting Time
2.1.3 Baseline measures
For a given budget year, the CCC wishes to understand the abatement potential (and that which is cost-effective) above what is already going to happen – known as the baseline or reference case. It is therefore important to have a consistent approach concerning which measures are and are not included.
For the purposes of this study, all measures announced prior to the Energy White Paper in 2007, which are reflected in the UEP 29 energy projections, are to be included. All cost curves reviewed have developed baselines consistent with this requirement.
2.1.4 Measure of cost effectiveness
Cost-effectiveness is often measured as either:
NPV/lifetime CO2 abatement or Net annualised cost/average annual CO2 abatement
In technical terms it seems the former measure is the most appropriate; and this lifetime measure approach is recommended by the Government’s Interdepartmental Analysts Group (IAG). The logic is it is required to reduce carbon emissions in the most cost-effective way over the lifetime of a technology. Net (current) annualised cost is very closely related, but generally suggests technologies are more costly, because it is a current value calculation.
However in practice we may be severely constrained by the models available to use. Many of these can only calculate a net annualised cost. In these circumstances it will be required to extract the cost information for a NPV Discounted Cash Flow (DCF) calculation8.
Capital costs Energy savings cost benefits (pa) Other savings or cost (O&M costs, labour savings, quality savings etc pa)
8 It is possible to go directly from the marginal cost data in the MAC curve to a NPV Discounted Cash Flow (DCF) calculation. This is simple if a linear progression of delivery is assumed; however, it is more complex in those cases where non-linear progress is to be evaluated. Also, it does not reveal the various elements of the cost analysis such as total capital investments, total energy cost savings and other cost savings.
8 AEA Energy & Environment Restricted – Commercial Review and update of UK abatement costs curves for AEA/ED43333/Issue 3 the industrial, domestic and non-domestic sectors
We consider the relationship between annualised and NPV lifetime costings further in Appendix 3. We also review the methods use in the past for policy and target costings, by the Carbon Trust and for the EE Innovation Review (Defra) in Appendix 4, to help inform the current debate on budget setting.
2.1.5 Carbon emission factors for electricity
In producing an overall energy end-use MAC curve this project will retain the flexibility to convert energy savings into carbon using a range of coefficients, hence testing sensitivity of carbon savings to alternative assumptions about the long-term marginal plant. But this is only relevant when the energy end-use cost curve are used in stand-alone mode. Ultimately it is envisaged that in developing carbon budget scenarios the CCC will either:
. feed energy savings from the energy end-use and other sectoral models into the BERR energy model, which will then translate energy savings into carbon savings consistently with the marginal generation plant in the relevant scenario; . link up the energy end-use and other sectoral model into the UK MAC curve model, which has its own power sector module so that changes in the generation mix will affect the carbon intensity of electricity, and feed through to the end use sector MACCs.
Both routes allow for exploration of dynamics between power sector and downstream abatement in end user sectors. For instance plant less carbon intensive marginal generation could make electricity saving options appear less attractive and suggest that a lower amount of electricity savings is desirable from a carbon reduction perspective. In turn, this could again feed back to the power sector, and impact on generation levels and the marginal type of generation.
AEA Energy & Environment 9 Review and update of UK abatement costs curves for Restricted – Commercial the industrial, domestic and non-domestic sectors AEA/ED43333/Issue 3
3 Review and update of industry sector MACC
3.1 Review of industry sector MACC
The characteristics and assumptions about the industrial model, ENUSIM, from which cost curves are derived are listed in Table 3.1. Further description of ENUSIM can be found in Appendices 3 to 5.
Table 3.1 Review framework – industry
Issue Key assumption Comment Source model Enusim Purpose of MACC Assessment of cost-effective potential of measures relative to business as usual uptake of measures Fuel prices Based on BERR assumptions. Note for this work core fuel prices are set For this MACC, based on March centrally by CCC. However, further 2008 fuel prices. disaggregation is needed to correspond to the 23 fuel types (+ waste with zero cost) needed in Enusim Emission factors - Based on BERR/Defra Note for this work set centrally by CCC electricity assumptions9 (Long-term marginal factor 0.43kgCO2/kWh) One limitation of the model is that emissions factors remain constant over time. The user may input new emissions factor for each year of the cost curve in the output spreadsheet. Emission factors – Based on Defra assumptions As above. fossil fuels from the National Greenhouse Gas Emissions Inventory Time horizon Up to 31 years after base year Cost curve output can be specified for any period within this. Base year Varies by sector due to nature of Iron and steel (2003) previous updates to model and Paper, Glass, Bricks, Ceramics, Chemicals data availability. (2002) Cement and Lime (2005) All others 2006 Baseline measures Baseline measures broadly The baseline penetration rate for the include technologies is a user input. There is no direct EU-ETS for relevant sectors link between technology penetration and the (via carbon price within policy context. However, it is assumed that energy prices) these penetration rates are based on the current CCAs (for non-ETS sectors) policy framework. A more explicit link could be made using information from the OEF forecasts, but would require significant resources beyond the current project. Sector 22 sectors (excluding energy). Iron and Steel, Paper, Non Ferrous Metals, disaggregation Glass, Other Industries, Non Metallic Minerals, Bricks, Energy, Cement, Ceramics, Chemicals, Mechanical Engineering, Electrical Engineering, Plastics, Food and Drink, Textiles, Vehicle Engineering, Lime, Aerospace, Construction, Water, Other vehicles Spatial UK only Disaggregation for DAs based on sector output disaggregation (GVA) Coverage of EU ETS sector is comprehensive Model only covers final energy consumption so
9 http://www.defra.gov.uk/environment/business/envrp/pdf/conversion-factors.pdf
10 AEA Energy & Environment Restricted – Commercial Review and update of UK abatement costs curves for AEA/ED43333/Issue 3 the industrial, domestic and non-domestic sectors measures as recently updated (2006) supply measures (e.g. CHP) must be considered Energy use in all other sectors off-model. updated to a base year of 2006 Some updating of technology data for chemicals, food and engineering Potential of Based on combination of EU-ETS sectors recently updated (2006). measures Literature review Expert judgement The technology data, used to determine potential was reviewed for chemicals, food and engineering. The review relied on expert judgement in the project team. Ideally, these data should be reviewed by industry representatives but this has not been possible within the scope of this project. Penetration rates S-curves assumed. Shape of s-curve and initial penetration of measure on the curve can be varied within the model. Based on combination of:
Literature review Expert judgement Consultation with industry
N.B. the exact uptake of measures in the model depends on the interaction of the cost- effectiveness of measures (e.g. in light of energy prices and discount rates) and the interaction with the s-curve (which acts to limit the rate of penetration). This is an integral choice by the model and means that penetration rates over time are not set explicitly.
EU-ETS sectors recently updated (2006). Penetration rates for chemicals, food and engineering reviewed as part of this work. N.B. no industry consultation for this updating Hidden and missing Based on the analysis carried out NB the Enviros analysis changes both costs and costs by Enviros (2006). Technologies delivery were classified into different types and the hidden and missing costs given in the Enviros report applied to the technologies in the results spreadsheet i.e. post-ENUSIM. The option to look at either high or low hidden costs was included Other costs Policy costs not included Measure of cost- Annualised cost divided by effectiveness average annual CO2 abatement Investment costs Central estimate only For capital fixed and variable costs Investment period Specified by technology in model – not based on lifetime Discount rates Input of commercial discount rate N.B. See discussion in Section 2.1.1 on the use in ENUSIM of private and social discount rates. In the results spreadsheet, the user can input a social discount rate. The spreadsheet then calculates an annualised capital cost based on the user specified discount rate and the investment period defined in ENUSIM. This annualised capital cost is used to calculate a new specific cost for the technology. The carbon savings are not recalculated
AEA Energy & Environment 11 Review and update of UK abatement costs curves for Restricted – Commercial the industrial, domestic and non-domestic sectors AEA/ED43333/Issue 3
Interaction between Accounted for explicitly in model. Multiplicative savings where combinations of measures technologies are actually allowed (this can be varied in the model). Timing of measures Additional costs for early replacement are not accounted for in the model. Taxes on energy Energy prices include the climate The output spreadsheet includes the option to prices change levy at two rates: full and remove taxes from the energy prices. Changes 20% for companies with climate to the specific cost are calculated by comparing change agreement new energy prices (i.e. excluding taxes) relative to the prices in the original model run, hence it is possible to alter energy prices beyond taxes only. The carbon savings are not recalculated as part of this, only specific costs. Carbon allowance prices are excluded from energy prices.
3.2 Update of industry sector MACC
Figure 3.1 below provides an illustrative overview of where the various data and methodological issues interact with the ENUSIM modelling process, used to generate the industry MACCs. These can occur: Prior to the modelling runs and require ex-ante analysis As part of the modelling process (i.e. directly influence it or are a result of it) After the modelling runs where ex-post analysis is required on the output to help generate the final MACCs.
Figure 3.1 Interaction of data and methodological issues in modelling work for industry
Ex-ante analysis Modelling Ex-post analysis
g) New/updated ii) Alternative approaches measures to technological progress - Emerging techs c) Hidden and missing costs -Incorporation of - Payback criterion e) Coverage of measures/ quality of data iii) Interactions/ overlaps between measures - Sector updates: d) Lack of model - energy supply measures -Chemicals transparency (e.g. CHP) - Food & Drink - Operation of - Engineering - Contents of
Measure data
i) Timing of measures iv) Penetration rates of - Cost of potential in Enusim runs Output measures given year (i.e. impact of - Adjustment of via link to early replacement) OE modelling work Scenario data f) Disaggregation of a) Consistency of output data underlying -Time (model adjustment) b) Selection of assumptions - Fuel (model adjustment) underlying -Fuel prices - By sub-sector (SIC 2 assumptions - industry growth level), CRC & EUETS - discount rates - Devolved admin
KEY DATA ISSUES METHODOLOGICAL ISSUES MODEL COMPONENTS
12 AEA Energy & Environment Restricted – Commercial Review and update of UK abatement costs curves for AEA/ED43333/Issue 3 the industrial, domestic and non-domestic sectors
In this project, we have implemented the following data and methodological changes:
o Ensured consistency of input assumptions and provided more flexibility to change these assumptions o Updated data for most sectors o Added hidden and missing costs o Provided factors to dissagregate for Devolved Administrations and for policy coverage (ETS, CCA and CRC) o Implemented methodological changes to account for social costing (see Section 2.3.1)
To provide a tool that can be used by the CCC, a spreadsheet has been developed that aggregates the output from ENUSIM and allows the user to perform additional analysis such as adding hidden and missing costs or recalculating specific costs from a social perspective. Appendix 6 provides a process map for the updating of the industrial MACCs.
3.2.1 Consistency and selection of underlying assumptions
Discount rates The user may now input a choice of discount rates in ENUSIM and also a social discount rate in the output spreadsheet.
Projected sectoral growth rates Industry output (either physical production or gross value added) by sub-sector/product is used. These are based on the most recent modelling work by Oxford Economics (OE)10, which are consistent with BERR’s latest projections. The industrial sector projections by BERR for the EWP 2007 operates at the SIC 2-digit level; ENUSIM requires a significantly higher level of disaggregation. BERR have supplied us with their model which alignes the total industrial GVA projection used in EWP 2007, with the SIC 2-digit projections used in UEP. We have now further disaggregated this to produce pro-rata changes at a more detailed sector level as show in the tables in Appendix 1. The data in these tables link automatically to the MACC control panel total industrial GVA projections via the BERR model and our more disaggregated sub-sector projection model.
Fuel prices The ENUSIM bottom-up simulation model requires projections of delivered fuel prices in three plant size tranches (small, medium, large) for all relevant fuel streams11. Details of the prices used are provided in Appendix 2 for a 20% and 100% Climate Change Levy (CCL) level (appropriate to CCA and non-CCA sectors). Prices are based on wholesale price projections used by BERR for the EWP 07 projections, translated into the delivered size tranche information using BERR published industrial delivered fuel prices 12. The standard outlook of ENUSIM is therefore to look at cost-effectiveness from a private perspective (the perspective of businesses investing in the abatement measure), although a spreadsheet developed as part of this project will allow for appropriate corrections to be made to derive cost effectiveness indicators based on a social perspective.13
For electricity wholesale prices we have obtained projections directly from the BERR modelling team, as these figures have not been published and are developed through their electricity model. Figures in Appendix 2 show the individual delivered fuel price projections by fuel type. They show significant increases over prices used in previous MACC studies.
Fuel prices include average CCL levels across users as shown in Table 3.2, taken from the BERR Quarterly Energy Prices publication.
10 See: Oxford Economics (2007) Report on modelling the macroeconomic impacts of achieving the UK’s carbon emission reduction goal. Report for BERR. 11 Domestic and commercial sector delivered fuel price projections can be taken directly from the MACC control panel. 12 BERR Quarterly Energy Prices (Dec 2007), Table 3.1.3. 13 CCC is looking at the fuel prices to use, which may differ from the private fuel price – they are currently looking at using long run marginal savings (as the social resource measure of the cost of a unit of electricity)
AEA Energy & Environment 13 Review and update of UK abatement costs curves for Restricted – Commercial the industrial, domestic and non-domestic sectors AEA/ED43333/Issue 3
Table 3.2. Average amount of Climate Change Levy paid by fuel type
Fuel Full rate of levy Average amount paid Q4 06 Q1 07 Q2 07 Q3 07 Coal £12.01/tonne £8.1/tonne £6.0/tonne £6.4/tonne £6.0/tonne Electricity 0.441 p/kWh 0.32 p/kWh 0.32 p/kWh 0.24 p/kWh 0.23 p/kWh Gas 0.154 p/kWh 0.08 p/kWh 0.09 p/kWh 0.08 p/kWh 0.07 p/kWh LPG £9.85/tonne - - - -
In producing the Appendix 2 tables of delivered fuel prices with 20% and 100% CCL levels, we have firstly removed the average CCL levels shown above, and then added back actual CCL values, as shown in Table 3.2.
Table 3.3. Revised CCL levels from HMRC notice, 2007
Fuel p/kWh £/GJ At real 2006 Electricity 0.441 1.225 1.187 Gas 0.154 0.428 0.415 Coal 0.161 0.447 0.433 LPG 0.072 0.199 0.193
3.2.2 Coverage of measures, and quality of the data
Recent updates of ENUSIM involved those industrial sectors in the EU ETS shown in Table 3.4. These sectors (generally energy intensive sectors) accounted for around 38% of total industrial sector emissions. In addition, combustion plant emissions covering chemicals, paper and vehicle manufacturing (to assess the role of boilers and CHP technologies) covered a further 9% of industrial emissions (excluding the paper sector, which is already accounted for). In the previous update there was no specific updating of the end-use technologies in chemicals or vehicle manufacturing, with the emphasis more on combustion plant measures.
Table 3.4 Comparisons EU ETS sector emissions (2003)
Direct Indirect Total Percentage of EU ETS sectors emissions, emissions, emissions, total industrial ktCO2 ktCO2 ktCO2 emissions
Cement 5,026 885 5,911 4.9% Lime 488 51 539 0.4% Paper 3,660 1,921 5,581 4.6% Glass 1,028 506 1,534 1.3% Brick making 969 213 1,182 1.0% Ceramics 140 71 211 0.2% Steel 14 27,221 3,487 30,708 25.2% Chemicals heat 10,224 8.4% Paper heat 3,083 2.5% Vehicles manufacturing heat 550 0.5%
Totals excluding heat based emissions 37.5%
(and with heat) (48.9%)
14 The steel sector was analysed after the main AEA/CC report and was reported separately. This report is not publicly available due to confidentiality issues.
14 AEA Energy & Environment Restricted – Commercial Review and update of UK abatement costs curves for AEA/ED43333/Issue 3 the industrial, domestic and non-domestic sectors
Those sectors not reviewed in the updating work are listed in Table 3.5. These evidently represent a significant percentage of UK industrial emissions and will form a substantial part of the coverage of policies such as the CRC. Chemicals, food and drink and engineering together represent 38% of total industry emissions, equivalent to the combined emissions level of all of the EU ETS sectors studied in the previous updating work.
However, there is a further complication in that most of the non-EU ETS sectors are made up of very diverse activities as shown in Table 3.5, and each of which is in need of updating.
Table 3.5 Comparisons of non-EU ETS sector emissions based on BERR energy statistics15
Direct Indirect Total Percentage of Non EU ETS sectors SIC emissions emissions emissions total industrial ktCO2 ktCO2 ktCO2 emissions
Food and drink All 15 6437.6 5694.0 12131.6 10.0% Operation of dairies 15.51 432.7 439.3 872.0 0.7% Manufacture of beer 15.96 577.4 124.1 701.5 0.6% Maltings 15.97 264.9 74.4 339.3 0.3% Alcoholic beverages 15.91 238.0 292.7 530.7 0.4% Baking 15.81+2 727.3 762.5 1489.8 1.2% Sugar 15.83 1012.8 32.6 1045.4 0.9% Meat and fish 15.1,15.2 692.1 1381.4 2073.5 1.7% Grain milling and starches 15.6 426.3 391.8 818.1 0.7% Fruit and vegetables 15.3 242.4 458.4 700.8 0.6% Animal feeds 15.7 402.5 284.2 686.7 0.6%
Chemicals All 24 8916.8 10270.6 19187.4 15.7% Organic basic chemicals 24.14 2243.0 2243.0 4486.0 3.7% Fertilizers 24.15 1103.8 678.0 1781.7 1.5% Inorganic basic chemicals 24.1,2,3 2264.1 3648.3 5912.4 4.9% Pharmaceuticals All 24.4 1091.2 852.9 1944.1 1.6% Soap,detergents, toiletries All 24.5 334.6 302.9 637.4 0.5%
Engineering All 28 to 5367.3 9770.6 15137.9 12.4% 35 Mechanical engineering 28,29 1999.6 3957.4 5957.1 4.9% Electrical engineering and All 30 to 897.2 3171.1 4068.3 3.3% electronics 33 Vehicle engineering All 34 to 2470.5 2642.1 5112.5 4.2% 35
Textiles All 1777.6 1583.7 3361.3 2.8% 17,18,19
Mining and quarrying All 13,14 791.4 769.0 1560.3 1.3%
Rubber and plastics All 25 4236.6 5198.4 9435.0 7.7% Tyres 25.11,12 1404.4 299.7 1704.0 1.4% Plastics 25.2 2469.4 4549.4 7018.8 5.8%
Total non-EU ETS emissions 49.9%
To improve the data used for non-EU ETS sectors that have not recently been updated, the following steps were undertaken:
15 http://stats.berr.gov.uk/energystats/ecuk.xls
AEA Energy & Environment 15 Review and update of UK abatement costs curves for Restricted – Commercial the industrial, domestic and non-domestic sectors AEA/ED43333/Issue 3
Step 1: Update base year energy and emissions values by detailed sub-sector (to 2006 from the original 1995 base year in the model); Step 2: Estimate the current (2006) device energy breakdown (e.g. by cement kiln type) in each sub-sector; Step 3: Re-evaluate existing technologies/measures16 in terms of their costs/energy saving values and their current penetration levels in sectors (i.e. % penetration in 2006 - normally the model starts from 0% penetration in 1995); Step 4: New technologies/measures and their timescales for introduction.
The following updates include:
Undertake Steps 1 and 2 for each of the non-EU ETS sectors: chemicals, food and drink, engineering, textiles, mining and quarrying, rubber and plastics, construction, water (utilities), other industries. This is essential to ensure sub-sectors begin from a more recent starting point. Undertake Steps 3 and 4 (based on literature review and consultation with relevant sector experts within our consortium) for chemicals, food and drink, and engineering only – as these are the largest emitting non-EU ETS sectors (covering 38% of industrial emissions – and broadly equivalent to that covered under the EU ETS update – excluding the off-model iron and steel sector work).
For food and drink and to some extent for non-ferrous metals some restructuring of the grouping was necessary. Previous updating was carried out using information from the Best Practice Programme (the pre-cursor to the Carbon Trust), with a higher level of dissagregation of data than is now available. The energy use for a subsector is taken from BERR statistics17. The latest year for which this detailed breakdown is available is 2005 but 2006 figures have been estimated by multiplying 2005 by ratio of the sector energy in 2006 to that in 2005. It has been assumed that the split of the subsector energy by device is the same as previously as no new data are available. The output for the updated sectors is taken as gross value added from the UK Manufacturing Sector Data/Forecasts done for the 2007 Energy White Paper.
3.2.3 Geographic coverage
To allow MAC curves to be produced for the different Devolved Administrations, we have used the Annual Business Inquiry regional breakdown statistics (release date 31.07 2007) to estimate the industrial breakdown of sector UK industrial activity by region.
The breakdown has been carried out on the basis of GVA activity (which better reflects energy activity); the coverage of the main sector areas is shown in following table.
Data is not completely available for all different Administrative Regions and therefore for some sectors it has been necessary to make estimates, particularly for Wales and Northern Ireland (Food and drink and Vehicle engineering). For the water sector, data is only available for England and we have therefore split coverage of the other Devolved Administrations on the basis of the average activity split figures.
The table also shows the individual sector GVA values at the total national UK level. It is evident that in terms of GVA value, construction dominates activities, followed by mechanical engineering, paper and printing and the food sectors.
16 That are applied to the sub-sector devices to reduce energy consumption – e.g. improved controls. 17 UK Energy Consumption 2007 Table 4.6 Detailed Industrial Energy Consumption by Fuel, 2000-2005 http://www.berr.gov.uk/energy/statistics/publications/ecuk/industrial/page18171.html
16 AEA Energy & Environment Restricted – Commercial Review and update of UK abatement costs curves for AEA/ED43333/Issue 3 the industrial, domestic and non-domestic sectors
Table 3.6: Summary of Devolved Administration coverage of UK industry
Northern GVA weighting SIC England Wales Scotland Comments Ireland £m 15 Food and drink 82.8% 4.0% 11.2% 2.1% 18,738 Estimates made for Wales and NI (2002 data) 17,18,19 Textiles, clothing 87.1% 2.6% 8.1% 2.3% 4,433 Based on 2005 data 21,22 Paper, printing 90.5% 2.6% 5.5% 1.4% 18,850 Based on 2005 data 24 Chemical 89.0% 4.0% 5.5% 1.5% 15,283 Based on 2005 data 25 Rubber and plastics 84.0% 5.2% 6.1% 4.7% 8,004 Based on 2005 data 26 Non-metallic mineral products 84.2% 4.5% 5.9% 5.5% 5,356 Based on 2005 data 27 Basic metals 84.1% 10.7% 4.6% 0.6% 3,472 Based on 2005 data 28,29 Mechanical engineering 86.9% 3.5% 7.2% 2.4% 24,592 Based on 2005 data 30,31,32 Electrical engineering 76.0% 6.8% 14.4% 2.8% 9,565 Based on 2004 data 34,35 Vehicle engineering 87.6% 4.9% 4.9% 2.6% 17,554 Estimates made for Wales and NI (2005 data) 20,33,36 Other 83.6% 6.2% 8.2% 2.1% 13,649 Wood, furniture, other NEC 41 Water 82.0% 5.7% 9.3% 3.0% 3,861 Estimates made for Wales, Scotland and NI (2005 data) 45 Construction 83.7% 3.8% 8.8% 3.8% 60,198 Based on 2005 data Total 85.0% 4.3% 7.9% 2.8% 203,555
AEA Energy & Environment 17 Review and update of UK abatement costs curves for Restricted – Commercial the industrial, domestic and non-domestic sectors AEA/ED43333/Issue 3
3.2.4 Trading and non-trading sectors
Different parts of manufacturing industry are subject to different UK GHG policies and are therefore also subject to different delivered fuel prices (i.e. with or without the CCL discounts) and emissions reduction obligations. The response of industry to emissions abatement investment is therefore affected (via the ENUSIM model) by whether they are subject to Climate Change Agreements and/or EU ETS or not.
Furthermore, modelling the emissions reduction potential in the various sectors of the economy and understanding this potential in trading sectors (EU ETS and CRC sectors) versus the non-traded sectors is an important assessment for setting the carbon budget. It will enable recommendations on how the reduction pathways can be achieved cost effectively, since effort should best come from the non-traded sector(s) where the related cost is less than the envisaged market price of carbon, and from the traded sectors thereafter –ensuring no marginal costs above the wider market price of carbon.
It is therefore necessary to ensure that the MACC models are sufficiently flexible to offer disaggregated potentials for organisations not covered by the CRC, and those covered by the CRC, CCAs and the EU ETS.
A later section of this report will also look at the relationship between existing policy delivery and the marginal cost of delivery, to assess whether existing policies are equitable regarding costs (and/or delivery). To make this assessment it is necessary to adequately disaggregate the coverage of existing policies among sectors.
In this section we consider the coverage of Climate Change Agreements, the EU ETS and the forthcoming CRC amongst industrial sectors so that the appropriate MAC curves can be developed for policy evaluations; and to also identify trading versus non-trading sectors for the budget allocation process.
The coverage calculation method we use is based on real company ‘returns’ for Climate Change Agreements and for the EU ETS; we then make comparisons with the base year emissions figures calculated from ENUSIM, cross-checked against national statistics to evaluate the percentage coverage and the policy overlaps. For the evaluation of the coverage of the CRC we have used the delivery data in it the NERA/Enviros CRC study, as reported in Defra ‘Updated Partial Regulatory Impact Assessment on the Carbon Reduction Commitment (June 2007).
For many sectors there are large overlaps between CCA and EUETS coverage. This has been estimated by comparing the direct emissions sources within CCAs with those in EU ETS. We have identified the element of the CCA coverage, which is associated with direct emissions i.e. direct fossil- fuel use, and compare this with the EU ETS direct emissions (fuel related) including that component resulting from CHP plants. For most sectors the end-use emissions covered by the CCA is similar to that covered by the EU ETS (see the table below) and we have estimated the overlap level as the smaller of the two emissions figures (either from CCAs or EU ETS). For chemicals, we have a specific estimate, which indicates that 84% of the direct element of CCA emissions is covered by EU ETS and we have therefore use this figure to estimate the overlap. For vehicle engineering and aerospace, we have also assumed that there is no overlap between CCAs and EU ETS as the policies cover different elements of plant. In general however, plants covered by EU ETS also come under CCA coverage and there are relatively few sectors where direct emissions coverage is EU ETS only. Therefore taking the smaller of the two emissions figures (either from CCAs or EU ETS) is likely to be a reasonable approximation in the absence of more detailed data18.
The following table summarises the coverage of trading instruments by manufacturing sector at 2003 - 2004; these are the base years for EU ETS actual site emissions data and CCA returns data, which are then able to be compared with available yearly national statistics.
18 It might slightly overstate the overlap where there are some CCA, some EU ETS and some overlapped emissions in any given sector, but this is unlikely to be a large effect.
18 AEA Energy & Environment Restricted – Commercial Review and update of UK abatement costs curves for AEA/ED43333/Issue 3 the industrial, domestic and non-domestic sectors
In total, it is estimated that the industrial coverage of CCAs based on 2004 returns data is 79 MtCO2, of which 54 MtCO2 are direct emissions (end-use fossil fuel based) and approximately 24.6 MtCO2 from indirect (electricity use). The total EU ETS coverage, on the basis of 2003 site useage (used in developing the Phase II National Allocation Plan) is in total of 42.5 MtCO2 (the sum of sector and CHP component emissions) with a further 26 MtCO2 covered by process emissions. There are an additional 6.9 MtCO2 pa involved in cement and lime processes with a further 18.1 MtCO2 from steelmaking. ‘EUETS only’ coverage amounts to 10.3 MtCO2, with the majority in the chemicals sector.
The CRC coverage of industry is estimated at 7.95 MtCO2 compared with a total national estimate of 19 14.87 MtCO2 . This sectoral analysis is based on a CRC threshold of 3,000 MWh. The latest position on CRC is that the threshold will be 6000 MWh; this makes only a relatively small difference to the overall emissions captured, reducing it to 14.2 MtCO2.
19 The additional emissions coverage is from the commercial and services sectors, which contribute a further 6.9 MtCO2
AEA Energy & Environment 19 Review and update of UK abatement costs curves for Restricted – Commercial the industrial, domestic and non-domestic sectors AEA/ED43333/Issue 3
Table 3.7: Coverage of trading instruments by manufacturing sector at 2003/04
CCAs total CCA CCA EU ETS CHP EU ETS Overlap EU ETS CRC returns data direct indirect coverage component process (CCA/EU Sector only coverage at 2004 element 2004 at 2003 2003 emissions ETS) (ktCO2) (ktCO2) (ktCO2) 2004 (ktCO2) (ktCO2) (ktCO2) (ktCO2) estimate Bricks 1,212 1,017 195 983 - 580 - 983 - Cement+lime 4,956 4,357 598 4,510 - 6,918 - 4,510 - Ceramics 737 493 244 - - - - 0 Chemical 15,042 7,813 7,229 9,760 5,416 - 8,621 6,55520 657 Construction ------45 Electrical Eng 466 82 385 13 - - - 13 298 Food and Drink 10,829 6,091 4,738 3,627 2,291 - - 5,918 280 Glass 1,955 1,203 752 1,273 23 428 - 1,203 80 Mechanical Eng 2,778 1,287.50 1,490 - - - - - 775 Aluminium 5,441 3,195 2,247 2,750 - - 2,750 121 Other NFM 668 365 303 - - - - - Non Met Minerals 1,158 728 430 484 - 79 484 - Other Industries 589 368 221 729 492 - 853 368 1745 Paper and board 5,206 3,467 1,739 3,553 2,756 - - 3,467 526 Printing 595 199 397 ------Rubber+plastics 1,964 1,747 217 - - - - - 1140 Steel 23,559 20,652 2,907 3,074 - 18,150 - 3,074 1384 Textiles 525 375 150 - - - - - Vehicle Eng+Aerospace 1,124 749 375 665 187 - 852 - 698 Water ------202
78,803 54,188 24,615 31,420 11,167 26,155 10,326 29,324 7,950
20 Based on a NERA/FES assessment (Jan 04) – private communication.
20 AEA Energy & Environment Restricted – Commercial Review and update of UK abatement costs curves for AEA/ED43333/Issue 3 the industrial, domestic and non-domestic sectors
The CRC coverage of the manufacturing sector is 8 MtCO2. The estimate excludes emissions that are already covered by CCAs (and hence are not included in the CRC) and reflects an electricity consumption threshold of 3,000 MWh per year.
The following table shows the ENUSIM sectors base-year emissions (both total and indirect), which allows from the previous table of emissions coverage, an estimate of the percentage policy coverage to be calculated for CCAs, EU ETS and CRCs as shown in the later table. In a few cases including non-ferrous metals, non-metallic minerals and brick-making, the estimated coverage (based on the fuel emissions calculation made here) exceeds 100% when compared with ENUSIM base year figures. This is particularly the case for non-ferrous metals and produces an even larger ratio when using national energy statistics. A probable explanation here is the assumption that the split between direct and indirect emission within the policy coverage areas is assumed to be the same as the total sector national split in our calculations. Where the estimates suggest greater than 100% coverage, we suggest using a 100% coverage figure, which sets the emissions covered at the maximum level consistent with ENUSIM base year figures.
Table 3.8: ENUSIM base year emissions, ktCO2 (total direct plus indirect and direct only)
Base Actual total Sector Direct emissions Year emissions Steel 2003 25,977 23,764 Paper 2002 8,927 5,293 NFMetal 2006 4,244 1,563 Glass 2002 2,114 1,467 Other Industries 2006 4,289 2,535 Non Met Minerals 2006 1,391 829 Bricks 2002 1,297 1,083 Electric Eng 2006 2,283 695 Cement (+Lime) 2005 5,737 5,003 Ceramics 2002 1,283 805 Chemical 2002 25,865 18,609 Mechanical Eng 2006 8,800 4,153 Plastics 2006 11,676 2,638 Food and Drink 2006 14,904 9,546 Textiles 2006 2,694 1,807 Vehicle Eng 2002 2,319 1,312
Lime 2005 506 461 Aerospace 2002 538 298 Construction 2006 636 636 Water 2006 tbc tbc Other Vehicles 2006 272 118
AEA Energy & Environment 21 Review and update of UK abatement costs curves for Restricted – Commercial the industrial, domestic and non-domestic sectors AEA/ED43333/Issue 3
Table 3.9: Policies with trading and their percentage coverage based on ENUSIM base year energy consumption
Base CCA EU ETS CRC Sector Year coverage coverage coverage Steel 2003 91% 13% 5% Paper 2002 61% 66% 6% NFMetal 2006 149% 129% 3% Glass 2002 86% 66% 4% Other Industries 2006 13% 32% 40% Non Met Minerals 2006 106% 60% 0% Bricks 2002 107% 115% 0% Electric Eng 2006 22% 2% 14% Cement (+Lime) 2005 79% 77% 0% Ceramics 2002 64% 0% 0% Chemical 2002 73% 56% 3% Mechanical Eng 2006 31% 0% 9% Plastics 2006 17% 0% 10% Food and Drink 2006 70% 37% 2% Textiles 2006 20% 0% 0% Vehicle 2002 41% 43% 25% Eng+Aerospace Water (BERR data) 2006 0% 0% 7% Construction 2006 0% 0% 7%
3.2.5 Hidden and Missing Costs
In the previous wide ranging review of UK cost curves, Enviros (2006) aimed to estimate and incorporate additional costs directly into the cost of the measures themselves. The overall results of Enviros’ literature review and in-house estimates differentiate between high energy intensity industry and low energy intensity.
The approach suggested by Enviros has been implemented in the output spreadsheet. Technologies are classified into the categories behavioural, engineering and other and high energy intensive and low energy intensive industry. The specified cost adjustments are made, as well as the risk factor for delivery. New specific costs are then recalculated. The user may specify the parameters to be applied.
The resulting hidden and missing costs are shown in Appendix 8.
3.2.6 Technological progress and availability of new technologies in the MACCs
Two main approaches can be used to estimate the dynamics and replenishing of the abatement measures. The object being to ensure an accurate picture of the cost curves well into the future by adequately accounting for emerging future technologies. The two approaches are:
. Approach by NERA/Enviros (2006) which assumes that the pool of future abatement potential remains constant over time (at least under the existing ‘ with policy’ baseline projections) via the interaction of the BaU uptake of measures and a top-down estimate of improvement in energy intensity of new measures of 0.3% per year.
22 AEA Energy & Environment Restricted – Commercial Review and update of UK abatement costs curves for AEA/ED43333/Issue 3 the industrial, domestic and non-domestic sectors
. A previous study by AEA and the Carbon Consortium defined the emerging technologies explicitly (AEA 2005).
In this latter study the new technologies were not integrated into the industrial ENUSIM model, but added via a spreadsheet approach. However, this approach is not very flexible and in this project we have implemented some of the technologies directly in ENUSIM. In the previous study, the technologies were not well defined and the following approach based on expert judgement has been followed.
The four technologies for energy end use in industry are:
Advanced Plant Design The technology is the next-step plant design, optimised for energy and environmental benefits. The new design package looks to combine unit operations, and takes an integrated approach to thermal integration/waste heat recovery. Process control is an integral part of the design process. Intelligent agent-based optimisers Intelligent agents are new computing technigues that could substantially increase the productivity and flexibility of manufacturing More energy efficient separation processes Diverse set of new separation techniques Second-stage waste heat recovery Extraction of remaining heat, principally from ‘dirty’ but still reasonably high-temperature processes
Advanced plant design as envisaged is an innovative and high capital cost technology. It has therefore been assumed that it would most likely be implemented in sectors first where there is significant growth projected.
These include: . Glass . Electronics . Aerospace
Concerning intelligent agent based optimisers, there are already a number of technologies included in ENUSIM on control of processes and the definition from the EEIR report is not sufficiently robust to distinguish it from the existing technologies. It has not therefore been included at this stage.
Energy efficient separation process have been implemented for drying and evaporation in chemicals, food and drink (dairies, confectionery and sugar) and ceramics.
Second stage waste heat recovery had only a small potential in 2020 in the original report. Given the slow progress for this technology since the report was published it was assumed that implementation would be pushed back from 2020 and would not therefore be significant in the period to 2022.
3.3 Realistic potential scenario
The marginal abatement curves give the remaining potential for technologies without any consideration of the timescales for delivery. The realistic potential scenario we have used tries to take these timescales into account. As a secondary consideration in this work, we consider it important that realistic potential estimates are considered again in more detail so as to provide a more robust basis for target setting.
Significant potential is available in replacing plant by new more efficient plant. However, plant replacement requires significant capital and usually involves disruption to production. It has therefore
AEA Energy & Environment 23 Review and update of UK abatement costs curves for Restricted – Commercial the industrial, domestic and non-domestic sectors AEA/ED43333/Issue 3 been assumed that this will only occur when plant reaches the end of its life. The proportion of new plants that can be implemented in a particular year is (MACC year – 2003)/lifetime of the plant21.
Hence, looking forward from the 2008 MACC (for 5 years up to the next available MACC in 2012) if an existing plant type has an average lifetime of 20 years then approximately 25% of the stock will reach the end of its life over this period22. When looking at the MACCs in subsequent years (2012, 2017 and 2022) a longer period of time will have elapsed, relative to 2008, and so a higher proportion of plant will have reached the end of its life. For example, any plant type with a lifetime of ≤19 years will have seen the complete stock of this plant reach the end of its life by 2022, and so all abatement potential from more efficient plant is realistically accessible by this point.
For retrofit engineering technologies there is an argument that they would only be implemented when the plant is shut down. However, most sectors have a regular shut down for maintenance (less than 5 year intervals). For the iron and steel plants this period is assumed to occur every ten years and for glass plants every seven years23. Taking this from the base year though, by 2012 all plants will have shut down at least once and so practically there will have been an opportunity to implement retrofit.
Supply side constraints may affect a limited number of the technologies but an analysis of this would be complex and beyond the scope of this project
For behavioural measures there is no reason to suppose that if the incentives were strong enough there would be any restriction on their implementation. The likely effectiveness is already taken into account in the presumed percentage savings.
Capital restrictions would also affect the rate of implementation. Using information on the technologies it would be possible to estimate implied capital spend but not within the timescales or resources for the project.
3.4 Emissions gaps Gaps in the coverage of sector emissions in the MAC curves
Emissions gaps occur for two main reasons: . Not all end-use sectors are considered in the MACC models; . There are other sources of CO2 emissions that are not fuel based (process emissions).
The following table summarises the known gaps.
21 Up to a maximum of 100%. 22 Assuming the current ages of plant are distributed evenly across all of the stock of this plant type. 23 Source: personal communication from industry expert Ernst Worrell.
24 AEA Energy & Environment Restricted – Commercial Review and update of UK abatement costs curves for AEA/ED43333/Issue 3 the industrial, domestic and non-domestic sectors
Table 3.10 Gaps in coverage of MAC curves
Areas Estimated Abatement options not covered emissions (ktCO2)
Additional industrial sectors (direct emissions only) Coal mining 15 Unclear Other oil and gas 2,500 tbc Munitions 100 General energy efficiency measures Mineral wool 450 General energy efficiency measures Carbon black 170 Unclear Gypsum 400 General energy efficiency measures Refineries and 24 18,500 offshore Estimated sub-total 22,000 (Indirect emissions may well add a further 2,000 ktCO2) Other sectors (total emissions) Agriculture 3,200 General energy efficiency measures Miscellaneous classification in 4,800 Unclear (theatres, sports halls, recycling, other) DUKES Estimated sub-total 8,000
Process emissions25 Cement and lime 6,900 Option for more cement blending Glass manufacture 430 Increased recycling Brick making 580 Raw material changes (difficult), or product substitution Iron and steel 18,000 Unclear Other ceramics 80 None apparent other than product design or substitution Estimated sub-total 26,000
As a comparison the ENUSIM total (including indirect) industrial emissions is 130,000 ktCO2 p.a. and public and commercial sector (2006 DUKES) is 65,000 ktCO2 p.a.
24 Data from BERR publication: Updating EnergyProjections May 2007 25 Estimates made from NAP II coverage data
AEA Energy & Environment 25 Review and update of UK abatement costs curves for Restricted – Commercial the industrial, domestic and non-domestic sectors AEA/ED43333/Issue 3
4 Review and update of building sector MACCs
The update of the building sector MACCs has primarily been undertaken by the Building Research Establishment (BRE) under a separate contract with CCC. Under our contract, we have reviewed the resulting cost curves, and associated assumptions,26 and intergated work from other cost curves (as described in Section 5).
In the main, the building sector cost curves are derived from models developed by BRE, namely the BREDEM (BRE Domestic Energy Model) and N-DEEM (National Non-Domestic Energy and Emissions Model) models. The updates to these models by BRE have been restricted to ensuring that the baseline assumptions for these cost curves in the budget years are consistent with the policy situation pre-EWP 2007. This has primarily meant revisiting assumptions on penetration rates, mainly modelled as S-curves.27
Most other assumptions on technologies listed, costs, and energy savings are consistent with earlier work published by BRE (BRE 2005a, BRE 2005b, BRE 2007).
4.1 Generic issues for domestic / non-domestic sectors
There are a range of issues that are generic to both building sectors that are important to highlight:
Interaction of measures
Two types of MACCs have been provided by BRE – one that provides an understanding of carbon saving potential for each measure in isolation, and a variant that provides potential for each measure after all other measures have been implemented, known as measures combined.
A commonly cited example of an interaction is that between between different types of measures which act on the same end use – e.g. simultaneous application of fabric insulation, more efficient heating equipment and more effective heating controls. The introduction of insulation has an impact on the potential of a condensing boiler. As described by BRE in their documentation, these interactions can only be dealt with effectively by carrying out detailed modelling studies as there will be non- linearities in the way in which the system interacts with the building arising from the timing of the service demand and the layout of the building.
The measures in isolation case only takes into account interactions of abatement measures with measures taken up in the baseline. For example, the implementation of condensing boilers in a given year will take account of the levels of insulation already taken up in the stock, and will account for these interactions. What this case does not provide an understanding of is the impact of the interactions between different measures that could be implemented in a given year (which if course are not included in the baseline).
The measures combined variant provides a lower bound abatement potential, due to the accounting for all possible interactions. In other words, the baseline housing stock (based on the characterisation of an average house) has most possible measures incorporated. Therefore, implementation of additional measures will capture any combination of measures in place, and the resulting impact of interactions on abatement potential.
Neither case are correct in reality, with the abatement potential in a given year being somewhere within the range provided by these two cases. Determining a realistic reduction potential in a given year requires that we understand what measures (in addition to the reference case) have been implemented. This can be done through developing scenarios of additional policies, to illustrate the
26 BRE have provided technical documentation of their update work - BRE MAC Curves - Technical Documentationt1.4.doc 27 S-curves provide a projected uptake of a given measure, based on current updatke levels and a saturation level in the future.
26 AEA Energy & Environment Restricted – Commercial Review and update of UK abatement costs curves for AEA/ED43333/Issue 3 the industrial, domestic and non-domestic sectors implementation level of specific types of measures. Interactions based on this mix of measures can then be modelled in the respective building sector models. The CCC recognises the limitations of MACCs in this respect, and the need for intelligent use of the information provided, and is undertaking additional analysis to inform understanding of realistic potentials.
Conventional vs. additional measures
Conventional technologies, as described by BRE, refers to standard, readily available and common technologies such as insulation and replacing equipment, whilst additional technologies include heat pumps which involve fuel switching and onsite generation of renewable energy such as solar hot water heating.
In theory, interactions associated with additional measures in the baseline are considered; however, in reality, the uptake in the baseline of such measures is very small, and therefore any impact of interactions is negligible.
No account for the interaction between additional technology measures and conventional technology measures in the combined measures case has been made due to these interactions being complex and requiring additional scenario modelling to resolve. It is probable that for the time horizon considered for this study, the uptake of these technologies above what is assumed in the baseline will be limited, and therefore, not accounting for interactions of measures may not be too problematic.
BRE make the following point in their technical documentation:
The uptake of such technologies is not expected to be significant within the timescale of the first three carbon budgets (see the uptake curves which have been determined for PV, wind generators, and solar water heating, presented in the domestic sector technical note), so the effect that they will have on the savings potentials from other measures will be small enough to be safely neglected. If such alternative technologies were to be taken up in significant numbers this would require the development of an alternative scenario which defines the trajectories of these technologies in terms of uptake rates, and then re-running the detailed models to calculate the costs and benefits arising.
In addition, the BRE data only cover existing additional measures; additional measures not in the current mix of options (e.g. fuel cell powered micro-CHP) could also lead to significant reductions in energy use in future years but possibly not in the time horizon of this study.
The potentials for these additional technologies can be shown in the MAC curves that present the savings from individual measures in isolation (i.e. where the intention is just to rank measures in terms of their overall potential and cost-effectiveness), but they cannot be included in the MAC curves that consider the combined effect of all measures.
Building stock
The BRE cost curves apply to the existing building stock, and therefore do not take account of the potential for measures in new build. In addition, the analysis is based on a ‘static’ building stock, meaning the savings that are still available in the current building stock are also available in future years (in the absence of the implementation of measures). This therefore does not take account of demolition rates or changes to existing stock (e.g. building extensions etc).
BRE make the following comments on the prospects for savings from new build:
Although the new build rate for both domestic and non-domestic buildings is relatively low, currently around 1-2% pa, if this continues they will still account for around 20% of the building stock by 2022. Tighter Building Regulations mean that the scope for cost effective improvement to the building envelope will be very small, and as the Regulations also cover the main building services, the scope for improvement here will also be limited - although perhaps less so for air conditioning where there is still significant technical headroom for improving plant efficiency.
AEA Energy & Environment 27 Review and update of UK abatement costs curves for Restricted – Commercial the industrial, domestic and non-domestic sectors AEA/ED43333/Issue 3
The majority of the potential for carbon savings in new building is likely to lay with lighting, appliances and (particularly for the more complex non-domestic buildings) improved energy management. In addition avoiding air conditioning could achieve significant carbon reductions in non-domestic buildings. Air conditioning typically accounts for around half the carbon emissions from a building, so options which restrict the use of air conditioning could make a significant impact.
4.2 Domestic Sector MACC
Energy savings from the domestic sector are estimated using the BREDEM model, and statistics on the UK building stock. BREDEM characterises a typical UK home – based on house type, size, heating systems and controls, and heat loss (measured by U-values) – to provide an understanding of what measures could be implemented to reduce energy loss / increase efficiency, and reduce carbon through fuel switching. The UK potential can be determined by scaling up, based on housing stock data, and the fuel consumption profile for the sector.
Measures are characterised by the amount of energy savings, and therefore carbon, (or carbon only in the case of fuel switching) they are likely to achieve in a given year, and the cost of the measure. The measure is said to be cost-effective where annualised investment costs (plus additional annual O&M and fuel costs) are lower than the cost of the energy saved in a single year.
As with most cost curves, the other key characteristics of measures include the remaining potential for emission savings in a specific year, in addition to the savings that are expected to be achieved in the reference case.
A key issue worth highlighting, and which has been assessed in recent research, concerns the potential of insulation measures. A report by Sanders and Phillipson (2006), and referenced in BRE (2007), assesses the difference between theoretical and actual energy savings achieved by cavity wall insulation (CWI) measures. The characterisation of this difference is known as the reduction factor. In addition, the potential savings from insulation measures are also known to be reduced due to the comfort factor issue (see next point). Both factors combined result in approximately 50% of savings achieved relative to those estimated on a theoretical basis. These numbers are still thought to be quite uncertain, which is important given that these are high potential saving measures, which are also cost- effective.
For specific measures, primarily insulation (or fabric-based) measures, a comfort factor is assumed. This is the proportion of the potential energy savings of a given measure that is taken as additional comfort and therefore does not lead to actual energy savings e.g. people adjust their thermostat up. A comfort factor of 15% has been applied where relevant. Note that this is included in the reduction factor of 50%.
Another important issue that is accounted for in the BRE analysis is the Heat Replacement Effect (HRE). When appliances are inefficient most of the energy they waste is released as heat. This additional heat may mean that the building heating system does not have to be used as if this waste heat was not available. This extra energy used in generating heat must therefore be netted-off from the energy savings attributed to efficient appliances. HRE factors will vary between products depending on the extent to which they are correlated with the heating system, making interactions complex.28 This effect is accounted for in both the measure in isolation and measures combined cases.
28 This is explained in two MTP Briefing Notes: BNXS05 http://www.mtprog.com/ApprovedBriefingNotes/PDF/MTP_BNXS05_2007May23.pdf and BNXS24 http://www.mtprog.com/ApprovedBriefingNotes/PDF/MTP_BNXS24_2007May23.pdf
28 AEA Energy & Environment Restricted – Commercial Review and update of UK abatement costs curves for AEA/ED43333/Issue 3 the industrial, domestic and non-domestic sectors
The main characteristics of the domestic sector model are listed in Table 4.1.
Table 4.1 Key characteristics – domestic building MACC
Issue Key assumption Comment Source model BREDEM-12 BRE Domestic Energy Model Assessment of technical potential, Purpose of MACC and cost-effective potential Based on BERR assumptions, but Fuel prices assumed to remain constant in future Note for this work set centrally by CCC. years 2050 estimate assumes 45% of electricity from carbon- Emission factors – Based on BERR assumptions free sources; rest of mix is the same as in 2020. Note for electricity this work, factors will be set centrally by CCC Emission factors – Based on Standard Assessment Note for this work, factors will be set centrally by CCC fossil fuels Procedure Guidelines Model includes estimates for 2010, Model is currently being developed in order to produce Time horizon 2020 and 2050 MACCs for budget periods (2012, 2017 and 2022) Base year 2001 (BRE 2005a); 2005 (BRE 2007) Inclusion of measures in UEP 29 Baseline measures (pre-EWP 2007) Sector None Average house assumed disaggregation Spatial Disaggregation being developed for DAs as part of UK only disaggregation parallel work being undertaken by BRE No account of behavioural measures; solar heating / PV / Coverage of micro wind included; (biomass boilers / GSHP in 2050, Comprehensive coverage measures therefore not in this study; other microgeneration (incl. CHP) excluded Potential of Based on Domestic Energy Fact File
measures (BRE) Fitted to uptake from past trend, and continuation of this Penetration rates S-curves assumed trend. Uncertainty where past trend not applicable. For appliances, rates from MTP used. Hidden and missing Not included Work by Enviros (2006) not subsequently included costs Other costs Policy costs not included Most analysis do not include the social cost of carbon Measure of cost- Net Annual Cost basis (although limited apparent impact on cost-effectiveness effectiveness for most measures) Low / high estimate help capture the uncertainty of Investment costs Low / high cost estimates provided. estimates. Future costs held at 2001 levels, except PV Investment period Lifetime of measure In line with Treasury guidance. This study will undertake Discount rates 3.5% across all technologies sensitivity analysis around discount rates determined by CCC Two approaches, which provides the range of most and least optimistic estimates of cost-effectiveness. One Interaction / overlaps Two approaches – in isolation and assumes no other measures (and therefore ignore between measures after all other measures implemented interactions) whilst the other assumes measure implementation after all others taken (least optimistic). No account taken of early retirement, Timing of measures and resulting increased annualised costs
Beyond the updates that BRE have implemented, consideration of how the above cost curve could be further developed across the key issues that this project is addressing (listed in Section 1.1) is made in Section 4.4.
4.3 Non-domestic sectors (commercial / public sector)
The non-domestic modelling is broadly consistent in the main assumptions with those outlined for the domestic modelling. However, the approach to estimating abatement potential across the buildings stock is different, primarily due to the heterogeneous nature of buildings across different non-domestic sectors.
AEA Energy & Environment 29 Review and update of UK abatement costs curves for Restricted – Commercial the industrial, domestic and non-domestic sectors AEA/ED43333/Issue 3
Information collected on different energy efficiency measures was used to assess potential savings from the builidng stock, based on a sample of c.700 premises which had been subject to an energy audit. This information on savings across different sectors was then scaled up to the UK, based on UK building stock data.
In a similar approach to the domestic sector, the potential of different measures has been assessed individually, and then in combination. The analysis of the combination of measures is required to take account of: . The interactions between heating and fabric measures e.g. condensing boilers vs. insulation, . Potential double counting from overlaps, where two measures affect the same energy use e.g. a condensing boiler and a thermostat. In this case when considering measure combinations, the measure which achieves most energy savings is implemented first.
So called additional measures are not included in the analysis of combined measures. These include PV technology, solar thermal, ground pumps and CHP.
For the cost curve provided by BRE for use under this contract, aggregate numbers have been provided at a non-domestic sector level. No subsector detail is provided, and energy / carbon savings and costs are provided as aggregate values for each measure. This reduced level of detail appears to be a function of providing a flexible excel based tool that can be effectively integrated into a UK wide MACC.
The main characteristics of the non-domestic sector model are listed in Table 4.2.
30 AEA Energy & Environment Restricted – Commercial Review and update of UK abatement costs curves for AEA/ED43333/Issue 3 the industrial, domestic and non-domestic sectors
Table 4.2 Key characteristics – non-domestic building MACC
Issue Key assumption Comment Source model N-DEEM National Non-Domestic Energy and Emissions Model Assessment of technical potential, Purpose of MACC and cost-effective potential Based on BERR assumptions, but Fuel prices assumed to remain constant in future Note for this work set centrally by CCC. years Based on BERR assumptions, 2050 estimate assumes 45% of electricity from carbon- Emission factors – except 2002 which is based on BRE free sources; rest of mix is the same as in 2020. Note for electricity estimates this work, factors will be set centrally by CCC Emission factors – Note for this work, factors will be set centrally by CCC fossil fuels Model includes estimates for 2010, Model is currently being developed in order to produce Time horizon 2020 and 2050 MACCs for budget periods (2012, 2017 and 2022) Base year 2002 Inclusion of measures in UEP 29 Baseline measures (pre-EWP 2007) Commercial, communication & transport, education, Sector 10 sectors. Based on energy audit of government, health, hotels & catering, retail, sports & disaggregation c. 700 premises, scaled up to UK leisure, warehouses, other Spatial Disaggregation being developed for DAs as part of UK only disaggregation parallel work being undertaken by BRE Coverage of Comprehensive coverage measures Potential of
measures Fitted to uptake from past trend, and continuation of this Penetration rates S-curves assumed trend. Hidden and missing Not included Work by Enviros (2006) not included subsequently costs Other costs Policy costs not included Most analysis do not include the social cost of carbon Measure of cost- 1 Net Annual Cost basis (although limited apparent impact on cost-effectiveness effectiveness for most measures) For replacement measure, additional cost assumed Wholesale price used (excl. VAT / (relative to alternative), but no additional labour cost. Add Investment costs delivery) on measures include full capital and labour costs. Future costs held at current levels Based on manufacturer’s data – in absence, default of 10 Investment period Lifetime of measure years for mechanical items and 5 years for electrical items In line with Treasury guidance. This study will undertake Discount rates 3.5% across all technologies sensitivity analysis around discount rates determined by CCC Two approaches, which provides the range of most and least optimistic estimates of cost-effectiveness. One Interaction / overlaps Two approaches – in isolation and assumes no other measures (and therefore ignore between measures after all other measures implemented interactions) whilst the other assumes measure implementation after all others taken. This is the least optimistic view. No account taken of early retirement, Timing of measures and resulting increased annualised costs
1 A measure may be cost-effective in one sub-sector but not in another; hence data are often presented for each measure in two components
Again, consideration of how the above cost curve could be developed across the key issues that this project is addressing (listed in Section 1.1) is made in Section 4.4.
AEA Energy & Environment 31 Review and update of UK abatement costs curves for Restricted – Commercial the industrial, domestic and non-domestic sectors AEA/ED43333/Issue 3
4.4 Update of building sector MACCs
BRE supplied updated cost curves for use in this project in the middle of February 2008. The updates proposed in this section were on the basis of the flexible cost curve spreadsheets provided by BRE, supporting documentation, discussions with the team at BRE, previous BRE reports and other related analysis, specifically the Enviros (2006) analysis for the EEIR.
Key developments to the MACCs in this study included:
. Linking assumptions on emission factors and fuel prices to the CCC Control Panel, where central assumptions are controlled. . Reviewing additional measures, with the inclusion of additional renewable heat measures, and product appliance categories. . Inclusion of hidden and missing costs, primarily based on work undertaken by Enviros (2006). . Consideration of the timing of measures, and impact of measure cost . Behavioural measures
This section describes the above changes in detail, and how the key issues considered in this project have been addressed within the updated MACCs. The review of some issues has not necessarily led to changes because 1) they are covered adequately in the current approach or 2) they are not in the scope of work.
All of the MACCs supplied ensure appropriate disaggregation of the cost curve data, on the following basis:
. Temporal – MACCs correspond to CCC budget periods . Spatial – MACCs are disaggregated by Devolved Administration. . Sectoral – Non-domestic MACC has been disaggregated between traded and non-traded sub-sectors of the commercial sector, taking account of the Carbon Reduction Commitment (CRC).
4.4.1 Consistency and selection of underlying assumptions
Core underlying assumptions used in the analysis for this study – discount rates, fuel prices, emission factors – need to be be approached consistently across all MACCs. These are therefore fixed in a central control panel provided by the CCC to ensure intergation, which all of the different MACCs have been linked to. Decisions about these, in particular discount rates, have been subject to wider discussion with CCC in relation to all of the MACC work currently being undertaken.
Fuel prices. These are currently consistent with BERR assumptions but are likely to be subject to changes as a result of revisions at the end of March.
Emission factors. These are consistent with those published by BERR as part of their energy projections publication. The carbon intensity of electricity will be determined by the marginal generation technology modelled in the power sector module.
Discount rate. This is another important exogenous input assumption included in the MACCs. In the Enviros (2006) review of carbon abatement cost curves for the Energy Innovation Review, a key recommendation was that models should use discount rates that reflect the costs of capital for investment in measures (>10%), not a social discount rate (3.5%). The issue of discount rate level is clearly critical in this analysis, as higher rates lead to lower cost-effective potential (due to higher costs of capital).
This issue has already been discussed earlier in this report (under Section 2), and it is likely that CCC will want to consider both social and private discount rates, and have recently made recommendations in line with this. In the Enviros (2006) analysis, 8% and 15% rates were used for domestic and non-
32 AEA Energy & Environment Restricted – Commercial Review and update of UK abatement costs curves for AEA/ED43333/Issue 3 the industrial, domestic and non-domestic sectors domestic sectors respectively, although it is not clear why these specific values were used (as opposed to 10% and 20% for example).
The use of a given discount rate depends significantly on the purpose of the analysis being undertaken (see Section 2), and what the discount rate is trying to represent.
Concerning what a discount rate is trying to represent, it is often increased to reflect non-financial risks (e.g. to model barriers to uptake) in addition to investment risks. For instance previous analysis of energy efficiency measures in the non-domestic sector has used discount rates as high as 25% (BRE 2002). This is often refered to a hurdle rate (for which there is a signfiicant amount of literature e.g. Sanstad et al. (1995)). Given that we are considering hidden / missing costs separately, to avoid double counting it is important that these discount rates reflect costs of capital only.
The CCC have suggested the use of a 3.5% discount rate for the domestic sector. Private rates for the non-domestic sector for use in sensitivity analysis to reflect cost of borrowing at 8%.
4.4.2 Hidden and Missing Costs
It has often been observed that energy efficiency measures, despite being estimated as cost-effective, are not taken up for a number of reasons – consumer behaviour, costs of capital or not accounting for full costs.
The final reason is captured by the issue of hidden and missing costs. These are costs that are missing i.e. they are incurred by the project investor but not recognised in appraisal of the costs, and costs that are hidden i.e. these costs are not obvious or easily quanitified but which are important in estimating true costs of a measure.
In their earlier study, Enviros (2006) concluded that hidden and missing costs were unlikely to be included in the cost data currently included in the BRE models. This has been confirmed by the BRE team. In so far as delivering a cost curve that accurately reflects the true costs of measures, we believe that this is a key aspect of the revised methodological approach.
Enviros (2006) proposed two approaches for the assessment of hidden and missing costs. The first, a top-down approach using a payback criterion, whilst the second is a bottom-up approach, which aims to estimate and incorporate these additional costs directly into the model (detail of these approaches can be found in Section 3.2.5). The bottom-up approach has been used for this project, in the main using hidden and missing cost estimates from the aforementioned Enviros study.
The following types of hidden and missing costs were identified in the Enviros study (2006); values used are shown in Appendix 8.
Table 4.3 Types of hidden and missing costs
Type of cost Description Cost category Project appreciation Time required to investigate EE measure Missing Project appraisal Once identified measure, time to assess measure more Missing comprehensively Project commissioning Time to commission project, and manage installation Missing process Production disruption Installation may cause disruption to economic activity Missing Additional engineering Additional work required due to the installation of measure Missing Perceived delivery risk Risk of measure not performing to perceived required Hidden standard Ongoing management / Time required for managing measure once implemented Hidden supervision time
* The missing cost categories are sequential in the process of project development and implementation
Two categories of missing cost that potentially do not feature in Table 4.3 are the time required for decision making (either after project appreciation and / or project appraisal) and time required for securing funding (prior to commissioning). Although these could be considered as independent steps in project development and implementation, these may already be included in the above categories,
AEA Energy & Environment 33 Review and update of UK abatement costs curves for Restricted – Commercial the industrial, domestic and non-domestic sectors AEA/ED43333/Issue 3 although it is not clear. We have decided not to include either of these missing costs to avoid potential double-counting. As is the case with any of the missing cost categories, the time required will depend on the project in question.
The hidden cost of perceived delivery risk relates to the implemented measure not leading to the energy savings estimated prior to implementation. This is a cost, as the decision making process would have been founded on the estimated potential energy savings. Although this cost has currently been included in the MACC, there is a concern that for insulation based measures, there could be potential for double counting given that adjustments have already been made to account for theoretical versus actual savings. The spreadsheet has the flexibility to amend any of these additional costs. It is important to note that there are potentially other risks including poor installation, and shorter than predicted lifetime, again which could have affected investment decisions. In the absence of information, additional costs have not been assumed beyond those considered by Enviros (2006).
The value used in the Enviros (2006) analysis for the domestic sector was £26 per hour, based on DfT appraisal metholdogy (as recommended in the HMT Green Book). This is equivalent to the value of business time for car drivers - equivalent to the average hourly wage for business car users. The non- domestic hourly rate assumed was £70 per hour, representing the opportunity cost of a manager’s time.
In measuring the value of time of individuals the HM Treasury Green Book supports the use of the Department for Transport’s Values of Time and Operating Costs TAG unit (3.5.6)29. The guidance outlines values of working time per person for the employer and values for non-working time per person for traveller’s own time. This is based on the latest research conducted by the Institute of Transport Studies in 2003 based on willingness to pay. In this project the value of non-working time has been applied for the domestic sector as this represents the time when individuals will consider and undertake energy efficiency improvements.30
A value of £3.68 per hour is therefore attributed to individual’s non-working time31. This is based on 2002 prices and at resource cost (excludes indirect taxes). A caveat to this value is that an individual’s value of non-working time will vary significantly and depend on such factors as the income of the individual, the value of the purpose and its urgency. A value of £3.99, adjusted to 2006 prices32 has been applied to all non-working time.
These costs can of course be flexed in the MACC spreadsheet, and therefore represent default values. A range of values could be considered included those that represented time off work which could be closer to the Enviros value. In summary, it is probably a range of different values depending on the person / project in question.
Using an average wage to determine the value of non-working time is often used as an alternative approach for valuing an individual’s time. The World Bank uses a default value of 30% of household income per hour33. However, the value of individual’s times is dependent on individual’s indifference curve and budget constraint, rather than an average wage that is based on the business activities of the employer. This approach has therefore not been applied in favour of the DfT research.
In terms of the value of time used for the commercial sector, it would seem reasonable to use the Enviros (2006) value; it is equivalent to stating that a manager could be charged out for £70 per hour and therefore that is the opportunity cost or money that could of otherwise be earned. Opportunity cost generally refers to the marginal costs and therefore it could be argued that in fact it is just the cost associated with that individual (wage, NI, pension) rather than the income (consultancy based approach). The approach of using the Enviros approach could be justified if the role was created for the purpose of the additional requirements rather than the using existing manager’s time or where charge out was unknown.
29 http://www.webtag.org.uk/webdocuments/3_Expert/5_Economy_Objective/3.5.6.htm 30 It could be argued that implementation of a given project could be implemented in work time; therefore an opportunity cost may be better reflected by a value of work time for certain additional cost elements. 31 Measured as the opportunity cost against individuals’ other time, for example leisure trips. This would be £4.46 at market rather than resource cost. 32 WebTag value of time growth rates, http://www.webtag.org.uk/webdocuments/3_Expert/5_Economy_Objective/3.5.6.htm 33 http://www.worldbank.org/transport/publicat/td-ot5.htm
34 AEA Energy & Environment Restricted – Commercial Review and update of UK abatement costs curves for AEA/ED43333/Issue 3 the industrial, domestic and non-domestic sectors
Senstivity analysis has been undertaken to explore the impact on cost-effectiveness of including low and high estimates of hidden and missing costs. This is presented in Section 6.
4.4.3 Coverage of measures
Measures in the domestic cost curve can be split into the following:
. Fabric measures, aimed at reducing the heat loss from the building structure e.g. insulation . Heating measures, aimed at improving water and space heating efficiency e.g. replacement with condensing boilers . More efficient electrical appliances e.g. EE lighting . Additional measures, focusing on renewable heat and microgeneration - PV generation, solar water heating, micro and mini wind turbines
There is a good coverage of measures in the existing building stock. However, a number of measures are missing, and these have been subsequently added:
. Selected renewable heat measures . District heating . Electronics and ICT products . Behavioural measures
Note that the above measures have been added by AEA subsequent to receiving the cost curves from BRE, and therefore do not take account of interactions with other measures in the baseline under the measures in isolation case. This would have to be done in the detailed modelling assessment of energy savings, and is a limitation of this cost curve development post-modelling. They cannot legimately be considered in the combined measures case as again interaxctions with other potential measures are not taken into account.
Renewable heat measures added include biomass heating, heat pump technologies, and district heating. The cost curve already includes heating from solar thermal. Description of the assumptions are provided in the section on intergration (5), primarily sourced from the OCC heat cost curve work / Ernst and Young (2007) analysis but with additional cost updates based on recent Element Energy (2008) analysis.
The inclusion of micro-CHP has been considered in this report. Based on this review, both Stirling engine and fuel cell types have been included in the set of domestic sector measures. Current evidence suggests that carbon savings will be limited and the costs will be high. This analysis is described in detail in Section 5.
Heating using electric appliances could be viewed as a potential abatement measure, where electricity provided off the grid is decarbonised. This is probably a longer term option that relies of significant re- structuring of the electricity generation sector. We have therefore not added this as an option.
Enviros (2006) noted the absence of behavioural measures in the domestic sector as a potential weakness. It is our understanding that while the inclusion of such measures would ensure completeness, very few data are available on which to estimate abatement potential; what data there are suggest limited potential relative to other measures. This issue is further complicated by probable variation in accounting for behaviour across different household types e.g. owner-occupier or not, socio-economic status. The BRE modelling does not consider different types of household; therefore such variations would be difficult to properly account for. Behavioural measures have been considered in this update work, based on a study being undertaken in parallel by AEA for Defra on household behavioural measures to reduce environmental impacts. This is discussed in greater detail in Section 5.5.
Measures in the non-domestic cost curve can be split into the following:
. Fabric measures . Heating measures . Lighting measures
AEA Energy & Environment 35 Review and update of UK abatement costs curves for Restricted – Commercial the industrial, domestic and non-domestic sectors AEA/ED43333/Issue 3
. More efficient electrical appliances, particularly appliances associated with commercial offices . Additional measures, including PV generation, solar water heating, CHP and heat pumps
Coverage of measures is comprehensive, with incorporation of behavioural measures, and inclusion where relevant of measures from the review of measures under the EEIR (AEA 2005). Additional measures to be considered include wind generation and biomass heating.
4.4.4 Quality of the data used in MACCs
Issues of modelling approach were highlighted in the previous review by Enviros (2006). For the domestic sector, the following issues were highlighted:
. There is no differentiation between existing housing stock types, with an average UK house assumed. Different measures may be more applicable to different housing stock types, which may be characterised by different thermal properties. Differentiation is also important for the implementation of hidden and missing costs.
. There is no distinction between household type e.g. owner-occupier, rented etc. This could be particularly important when considering additional hidden costs of specific measures, and barriers to implementation.
Both of these issues are still important. The lack of differentiation between housing type is a function of the modelling approach used, whereby implementation of measures is considered at the micro-level for an average house and then scaled up. The modelling becomes much more time consuming and complex if a number of different housing types are assumed.
Concerning the data assumptions used, these appear robust, with updated modelling of the reduction potential and costs of different measures, and the rate of uptake in the reference case. BRE acknowledge that limited updating of additional measures and electrical appliances has been undertaken, particularly given that these areas are covered by the integration task (see Section 5).
Revised assumptions concerning renewable heat and product-based measures are described in Section 5.
For the non-domestic sector, the following issues were highlighted by Enviros (2006):
. Base data to inform measure implementation and abatement potential is based on an energy audit data of around 700 commercial premises. An ‘average’ property approach is not taken due to the varying nature of the building stock. These data were considered to be quite old, and with significant changes in commercial buidlings in recent years, the recommendation was that they should be updated. No updates have been undertaken subsequently (presumably to an absence of better data and lack of funding for newer audits).
. Penetration rates are based on much poorer information than used in the domestic sector modelling. No updates have been undertaken primarily due to the lack of data for this sector.
It is clear that there are some concerns about the building stock data, and the evidence basis for the uptake of measures. However, the lack of new data allow for significant improvements, and the current analysis consitutes best available knowledge. Coverage of measures and associated potential and cost data are comprehensive.
4.4.5 Timing of measures
The cost curves being developed under this contract for the CCC indicates realistic and technical potential in a given year, subject to assumptions made about uptake of measures in the reference case. What they do not do is provide any insights into additional uptake of measures based on specific policies, only the potential abatement at a given marginal cost.
However, it is likely that consideration will be given to different policy targets in due course, and what this means in terms of abatement potential in a given year. It is possible that a given policy could lead
36 AEA Energy & Environment Restricted – Commercial Review and update of UK abatement costs curves for AEA/ED43333/Issue 3 the industrial, domestic and non-domestic sectors to the increased uptake of replacement measures in a specific year, resulting in the early retirement of equipment being replaced. This early retirement means that the replacement measure incurs additional costs, due to the cost implications of any early scrapping of equipment (e.g. sunken capital costs over a shorter lifetime) to meet required emission reductions.
An analysis by Enviros (2006) for the EEIR introduced this dimension into the commercial and domestic cost curves. Their analysis considered the effects of a policy that deliberately substituted a technology prematurely (before the end of its working life). Their calculations aimed at delivering reductions in specific target years and ‘costed’ the obligation of retirement of plant that was not a ‘sunk cost’ (varying time constraints). This clearly has the effect of raising technology costs for replacement measures as the policy delivery requirement tightens.
The following table lists the services sector technologies Enviros identified as particular important in the early retirement issue and also shows the incremental cost step (£/tCO2) for each year of early retirement. The base year margin costs are for 2006 as presented in the original EEIR study data.
Table 4.4: The marginal and incremental costs (for early retirement) of technologies considered as potentially important for early replacement (Enviros)
Cost step for 2006 cost each year of Measure £/tCO2 early retirement (£/tCO2) Lights – 16mm Fluorescent Tubes Replace 38mm -825.7 - Lights - 16 mm Fluorescent Tubes Replace 26mm 277.9 +1.3 Motor - 2 Pole Motor - EFF1 replace 2 Pole 792.4 - Motor - 4 Pole Motor - EFF1 replace 4 Pole 949.8 +32.5 Catering - A Class Freezer -241.3 +4.9 Catering - A Class Fridge-Freezer 66.8 +8.0 Lights - Compact Fluorescent Lamps without ECG & Tungsten -291.3 +0.4 Heating - Condensing Boiler -87.6 +42.0 OffEq Flat Screen Monitors 5,758.8 +11.7 Lights – HF Ballast 626.2 +1.1 Lights - Metal Halide Floods 118.3 +8.4 OffEq Flat Screen Monitors pc Only 1,225.7 +1.8 Windows - Double Glazing Air Filled 1,805.5 +0.8
It is evident that in the majority of situations identified, the technology cost is well above the carbon price level likely to be of interest in near-term carbon budget setting, whilst in other cases, the negative costs (highly cost effective measures) are associated with small incremental cost increases through early retirement. Only in the case of condensing boilers are early retirement options likely to present marginal cost values in the range of interest.
This is demonstrated further in the following table, which considers the latest BRE MAC curve cost data. In this case we also show the implied marginal cost assuming a premature retirement of plant by 5 years (the maximum considered by Enviros) using the incremental cost step data for the original Enviros work34.
Condensing boiler replacing existing boilers is the only technology that is significant in spanning a carbon price in the range £0 to +103 /tCO2 for early retirement.
34 Whilst the original analysis for the EEIR study is not available in a complete form, we have been able to obtain some of the Enviros analysis and have cross-checked their incremental cost increases (by additional year of early retirements) with our own calculations. These are consistent with the values used by Enviros.
AEA Energy & Environment 37 Review and update of UK abatement costs curves for Restricted – Commercial the industrial, domestic and non-domestic sectors AEA/ED43333/Issue 3
Table 4.5: The marginal and incremental cost, plus cost at a 5 year early retirement of services sector technologies considered as potentially important for early replacement
Incremental Latest cost for each £/tCO2 by 5 yr
£/tCO2 year of early early retirement retirement Lights - 16 mm Fluorescent Tubes Replace 26mm -186 +1.3 -179 Motor - 4 Pole Motor - EFF1 replace 4 Pole -171 +32.5 -8 Catering - A Class Freezer -211 +4.9 -186 Catering - A Class Fridge-Freezer -103 +8.0 -63 Lights - Compact Fluorescent Lamps -223 +0.4 -221 Heating - Condensing Boiler -107 +42.0 103 OffEq Flat Screen Monitors -197 +11.7 -139 Lights - HF Ballast -40 +1.1 -34 Lights - Metal Halide Floods -121 +8.4 -79 OffEq Flat Screen Monitors pc Only -213 +1.8 -204 Windows - Double Glazing Air Filled -143 +0.8 -139
In our current approach, we follow the line of analysis as outlined in the following figure, where we assume a 15 year boiler lifetime; and thus by 2022 there are very few conventional boilers left to replace35. In the interim in each five-year period one third of the remaining potential identified in the BRE (total potential data) will available for retirement at the end of ‘useful life’. This will continue until 2022.
Figure 4.1: Retirement profile of existing conventional boilers (2007- 2022)
However, additional replacement of existing conventional boilers could occur; at an additional cost as shown in the following figure, where we replace the current BRE MACC step (for the replacement of all remaining conventional boilers by condensing boilers in the domestic sector) with 6 steps - 5 for earlier retirement and 1 for the savings at actual 'end of life' in the five year period.
35 Building Regulations introduced in 2005/6 essentially eliminated the installation of conventional boilers in households.
38 AEA Energy & Environment Restricted – Commercial Review and update of UK abatement costs curves for AEA/ED43333/Issue 3 the industrial, domestic and non-domestic sectors
Figure 4.2: The cost and contribution of the early replacement of existing domestic boilers with condensing boilers at 2008
100 Retire 5 years earlier cost step and contribution 50 4 years earlier
3 years earlier etc 0 Retire 1 year earlier cost 0 5 10 15 20 step and contribution Early retirement options -50 Savings possible BRE MACC step £/tCO2 2008 to 2012 at actual end of life -100 Current BRE M ACC savings possible at 2008 -150
5.9 M t C O2 -200 CUM MtCO2
Comparing the replacement cost and contribution ‘steps’ with the BRE total MACC ‘step’ evidently provides a lower overall savings contribution because the BRE total potential data has been replaced by the more realistic time dependent potential, which also include the early retirement options. This analysis assumes that the maximum level of early retirement will be up to 5 years, which effectively doubles the allowed replacement level of conventional boilers in the five year period between MAC curves. This cut-off seems reasonable since anything higher i.e. an early retirement of greater than 5 years, would incur costs in excess of £100/tCO2. In other words, the above only assumes boilers coming to the end of their useful life in the next 5 years are replaced.
The next figure shows the replacement MAC curves (2008,2012,2017,2022), substituting the current contribution in the BRE domestic sector MACCs, to take account of: the restrictions on actual end of life replacement (from the ‘age-population’ spread) and the option to retire existing boilers early.
What we are seeing is an opportunity to gain ‘earlier’ savings - from up to 4 years early replacement of a depleting population of conventional boilers- at a cost of less than £25/ tCO2. As we move forward in time this opportunity declines as conventional boilers are replaced at their normal end of life.
AEA Energy & Environment 39 Review and update of UK abatement costs curves for Restricted – Commercial the industrial, domestic and non-domestic sectors AEA/ED43333/Issue 3
Figure 4.3: Revised MACC ‘steps’ for A- rated condensing boilers replacing a declining population of existing conventional boilers (due to BAU uptake of condensing boilers)
100
50
0 0 2 4 6 8 10 12 14 2008 2012 -50
2017 £/tCO2 2022 -100
-150
-200 CUM MtCO2
These revisions can be seen in the revised BRE MACC data for the domestic sector; data are available for the commercial sector but have not yet been linked in. Note that no other technologies have been introduced into these early retirement revisions, because they will not materially affect the MACC shape in the carbon price range £0 to 100 /tCO2. However, an adjustment for the realistic ‘time dependent’ technical potential, rather than the total remaining total potential has been made.
4.4.6 Limited account of interactions between measures
This issue was addressed at the beginning of Section 4.1. In this section, we consider it again in more detail. Note that all building sector MACCs avoid double accounting due to overlaps; this section therefore focuses only on interactions.
The issue here is that the introduction of a measure in a building (condensing boiler) which already has a measure implemented (insulation) could results in lower efficiency gains for the new measure (condensing boiler) due to interaction. As stated in BRE (2007), the savings from insulation would be lower if the savings potential for heating efficiency was met first and the saving from increasing heating efficiency will be lower if the savings potential for insulation was met first.
Not accounting for interactions properly could result in an overestimate of cost-effective potential. In setting budgets, the CCC will want to assess cost-effective potential in relation to a given carbon price. This will involve looking across the cost curve and estimating total savings based for a given marginal cost. The cumulative savings from the package of measures will need to exclude interactions that would otherwise result in overestimates of abatement potential.
In the reference case, interactions between measures (and the resulting impact on abatement potential) in each budget year are taken into account in the modelling – as the implementation of energy efficiency measures is known. The means that the starting point from which further potential abatement opportunities can be considered adequately reflects interactions.
The difficulty is then to reflect (in the flexible output spreadsheets) the interactions between measures that may considered in addition to those in the reference case. This is needed because the CCC will be considering an optimal package of measures based on a given marginal cost. Consideration of how interactions are modelled for that optimal package of measures is needed.
40 AEA Energy & Environment Restricted – Commercial Review and update of UK abatement costs curves for AEA/ED43333/Issue 3 the industrial, domestic and non-domestic sectors
This is difficult to do; BRE have suggested that this can only be done using a scenario approach, where the implementation of different in a given year above the reference case is known. We agree that this appears to be the only solution to adequately reflect the correct level of interaction.
The CCC intend to use cost curves to determine what measures are cost effective or what measures might be available at a given marginal price of carbon e.g. £30 per tCO2. On the basis of a least cost optimal mix, this choice will be made on a cost-effectiveness basis. This approach would only be sensible where the realistic potential was known, as you would only want to consider a level of implementation that could realistically be achieved in a given year.
The order of implementation of measures would be:
. Introduction of the most cost-effective measure (1) . Introduction of the next most cost-effective measure (2), reflecting any interactions based on (1) already being in place . Introduction of the next most cost-effective measure (3), reflecting any interactions based on (1) and (2) already being in place . And so on
Account would need to be taken of the potential for changing ordering of measures (within a sensitivity analysis e.g. due to addition of hidden / missing costs).
BRE have provided two MACCs identifying technical potential by measure; one that provides an indication of potential for each measure without accounting for interactions beyond the baseline36 – providing an upper bound – and one that assumes all possible interactions (including the heat replacement effect), providing a lower bound estimate of potential of all measures yet to be implemented. This latter case does not include so-called additional measures.
Results for the domestic sector suggest that interactions between measures do not result in significant differences from the case where no interactions are accounted for (13% reduction in CE potential in 2012; 8% in 2022). In the non-domestic sector, the impact appears relatively more significant (36% reduction in CE potential in 2012; 32% in 2022).
It is worth highlighting that in both the non-domestic and domestic MACCs, most of the cost-effective potential comes from fabric, heating and appliance-based measures, whilst non cost-effective measures are primarily from so-called additional measures (e.g. renewables).
In the case of the domestic MACC, BRE and CCC did discuss the possibility of applying adjustment factors to the potential of a given measure e.g. insulation on the basis of increasing implementation of another measure e.g. more efficient boilers. However, the conclusion was reached that simplistic assumptions would not add value to the analysis because, at least between two interacting measures, the impacts were not large (and were very simplistic for a complex problem).
It will be for CCC to determine where the potential between the two cases (measures in isolation and combined measures) lies, and this is likely to require additional scenario based modelling.
4.4.7 Penetration rates of measures
The technical potential uptake, and the penetration rates of measures determine the remaining abatement potential in a given year. This is what has been provided in the BRE MACCs. In addition, the CCC want to understand what might be realistically achieved in a given year; the remaining technical potential does not provide such insights, rather what is remaining to be implemented from that year onwards – in effect, it is time independent.
Two issues are considered here; firstly, the uptake of measures in the reference case and secondly, realistic potentials based on understanding of measures implemented in the reference case and remaining technical potential.
36 The heat replacement effect is taken account of in the domestic and non-domestic MACCs.
AEA Energy & Environment 41 Review and update of UK abatement costs curves for Restricted – Commercial the industrial, domestic and non-domestic sectors AEA/ED43333/Issue 3
Reference case uptake
The domestic model uses the most up to date information available on the penetration of domestic sector energy efficiency technologies, mostly from Defra’s Market Transformation Programme and the Domestic Energy Factfile. This information comes from regular surveys of retailers and provides good quality information on the take up of energy saving measures.
Historical and current data are fitted to S-curves to indicate the future uptake of specific measures over time, and when the potential is at or near saturation. The shape of the S-curve is informed by time series information, supply chain issues, and policy development – and is measure specific.
It has already been noted that the available information on which to base non-domestic S-curves is limited. Additional work, however, has been undertaken as part of this work to update the domestic sector uptake rates.
BRE have recently revisited assumptions for cavity wall and loft insulation to ensure consistency with UEP 29 assumptions e.g. no policy proposal beyond CERT (ending in 2011). Prior to these updates, BRE modelled uptake as follows:
The trend is based on a known starting point (i.e. from recent ownership data) with the effect of policies (mainly EEC figures and CERT Illustrative Mix estimates) added on to establish an S-curve that can be projected forwards to 2022. It is assumed that such projection is entirely appropriate given the Government announcement that some form of supplier obligation, with a level of ambition at least as great as that of CERT, will be in place until at least 2020.
Essentially this assumed that a policy proposal would continue the uptake trend established under CERT (which in reality is what is happening – see section 7.1 on the Supplier Obligation). However, the budget setting will be relative to a reference emissions projection which is based on policies in UEP 29 and therefore does not include continued policy effort beyond CERT or policies in the EWP 07 (many of which are still subject to consultation). Changes made by BRE now show much more potential for loft insulation and CWI now in 2022.
Concerning other uptake rates, significant revisions have been made by BRE (relative to earlier studies) to assumptions on solid wall insulation. Estimates of growth from 2001 to 2005 are low, implying an S-curve that is likely to saturate at perhaps only 10% of the 6.7 million solid wall dwellings (in England) which could be treated. About 8% are treated by 2022 according to the S-curve.
Assumptions around the uptake of solid wall insulation are uncertain. This issue is important, given that solid wall insulation in the MACC offers a large cost-effective potential in 2022. Understanding the realistic potential is therefore particularly important. Our view is that even with additional policy action, a lack of real market solution to the significant uptake of this measure means it is a difficult measure with which to realise significant reductions in CO2. This is demonstrated by the limited growth in implementation in recent years.
The uptake of more efficient products is in line with modelling under the Market Transformation Programme (MTP). None of the penetration rates have been revised. MTP are currently in the process of revising all of their projections concerning uptake under the reference case and policy scenarios.
The uptake of microgeneration and renewable heat measures is uncertain out to 2020, particularly due to the current low uptake of such measures. We do not believe that such measures will significantly penetrate the domestic housing stock in the next 10 years under a view of realistic potential, although clearly they will have an important role to play in the context of new build and the zero carbon homes proposal.
Penetration rates for the non-domestic sector are expressed as a percentage of the potential estimated in 2002 that is remaining. Therefore, due to the way this modelling has been developed, it is not clear what this means in terms of building stock – or with regard to penetration across different commercial and public sub-sectors (as sector detail has not been provided). It is our understanding that these penetration rates have not changed significantly since previous analysis (BRE 2005b) due to the lack of data to inform them.
42 AEA Energy & Environment Restricted – Commercial Review and update of UK abatement costs curves for AEA/ED43333/Issue 3 the industrial, domestic and non-domestic sectors
‘Realistic’ uptake of measures
Another important issue, albeit secondary objective, that this project has attempted to address is that of realistic potential in a given year. The MACCs provided by BRE give an indication of technical potential; that is all remaining potential for this measure at a given point in time, accounting for the business-as-usual uptake. However, this potential is time-independent in the sense that this does not provide an indication of what might be achievable by this point in time; it is simply all potential remaining, which could actually take another 20 years to fully implement. Hence the need for the realistic potential factors to identify potential take-up by a given year.
Some attempt has been made to develop very approximate factors for the domestic sector that reduce technical potentials to what might be realistic in a given year, 2022. Of course, what is considered ‘realistic’ is subjective, and requires more careful scenario-based analysis, including consideration of supply chain constraints, population behaviour, types of mechanisms to encourage uptake etc. In addition, BRE would be best placed to inform these estimates, in particular for the non-domestic sector which remains largely undeveloped in respect to development of realistic potentials.
Realistic estimates are compared to technical potentials for the domestic sector in Table 4.6 below. In 2022, realistic potential is estimated at 24.1 MtCO2 (almost 85% is cost effective) – or 6.6 MtC (excluding Zero Carbon Homes policy). 43% of the technical potential from fabric, heating and appliance-based measures is realised (assuming that only 15% of SWI potential would be achieved). This value has been compared to previous policy scenario analysis (BRE 2005a), which looked at potential abatement out to 2050, based on different scenarios. The value of 6.6 MtC compares to a policy scenario (5 MtC abated by 2020) versus an efficiency (stronger policy) case (8 MtC abated by 2020).
This at least provides some confidence that realistic potentials, at least at the aggregate level, are similar to what other analyses have estimated as achievable. At the individual measure level, there is a need for further investigation (and input by BRE) to reduce the underlying uncertainties.
Table 4.6 Proportion of technical potential ‘achievable’ in 2022for domestic sector measures Category of measure TECH (2022) REAL (2022) FABRIC 27.6 9.2 HEATING 3.2 2.3 APPLIANCES 5.9 4.3 D HEATING 3.0 1.0 MICROGEN 57.7 7.3 TOTAL 97.4 24.0
4.4.8 New build
As indicated earlier in the report, new build is not covered in the BRE models; the potential for additional carbon savings is likely to be limited in terms of fabric and heating measures due to increasingly tighter building regulations and the introduction of zero carbon homes and buildings in both sectors (CLG 2006). Additional savings could potentially be achieved by reducing the requirements for air conditioning and other energy using products and improving efficiency of those products.
In section 7, a range of measures announced as part of the EWP 07, or subsequently, have been considered to understand what abatement potential estimated in the MACC might be achieved and at what cost by policy already in place. For the domestic sector, zero-carbon homes is one such policy. The potential that this policy might achieve in terms of CO2 abatement and costs have been assessed in this section (based on an analysis by Cyril Sweet (2007)); such values could be subsequently used in the MACC to reflect technical potential (and associated costs), although have not been included at present in the flexible spreadsheet.
AEA Energy & Environment 43 Review and update of UK abatement costs curves for Restricted – Commercial the industrial, domestic and non-domestic sectors AEA/ED43333/Issue 3
For the non-domestic sector, we are unaware of any analyses that have considered the cost and potential of CO2 reductions in future new build. This will no doubt be considered in the near future in light of the new announcement for zero carbon buildings by 2019.
44 AEA Energy & Environment Restricted – Commercial Review and update of UK abatement costs curves for AEA/ED43333/Issue 3 the industrial, domestic and non-domestic sectors
5 Integration of cross-sectoral MACCs
An important objective of the end-use energy MACC project was to consider other cost curve work being undertaken by other Government departments, and consider the options for integration into the industrial and building sector MACCs.
The focus of this section is on the integration of the following:
. Heat cost curve developed by the Office of Climate Change (OCC) . CHP estimates, based on analysis by AEA’s CHP team / Carbon Consortium . Product policy cost curve, developed by under the Market Transformation Programme by Defra / AEA . Behavioural measures in the domestic sector, based on work undertaken by AEA for Defra . Micro-CHP, based on work being undertaken for Defra by AEA / Carbon Consortium
In the first instance, the above work was reviewed to help inform possibilities for intergation with the industrial and building sector MACCs. Towards the end of the study, estimates for microgeneration from recent analysis (Element Energy 2008) were also considered for incorporation into the MACCs.
5.1 Review of cross-sectoral cost curves
5.1.1 OCC Heat cost curve
A wide ranging project has been undertaken across several Government departments, led by the Office of Climate Change, to assess options for decarbonising heat.37 The areas covered by the project include renewable heat, industrial CHP, and district heating (DH). As part of this work, a MACC has been developed to identify cost-effective abatement potential.
The OCC cost curve represents the costs of and abatement potential of different heat-based measures in 2020, with most cost-effective potential found in industrial CHP and to a much lesser extent, the use of biomass in the residential sector (as shown in Figure 5.1 below).
37 OCC Heat project, http://www.occ.gov.uk/activities/heat.htm
AEA Energy & Environment 45 Review and update of UK abatement costs curves for Restricted – Commercial the industrial, domestic and non-domestic sectors AEA/ED43333/Issue 3
Figure 5.1. Marginal abatement cost curve for heat – realistic potential version
Marginal Abatement Cost curve for Heat
Residential biomass relative to electricity
300 Solar thermal relative to electric heating
Conversion of electric heating to district heating 250 Industrial CHP
Conversion of existing district heating to CHP 200
Commercial biomass relative to oil
) ) 2 Surplus heat from industry 150 District heating in dense residential areas
100 Industrial biomass relative to oil District heating in new build housing
50 AD CHP
Industrial biomass relative to gas
0 Commercial biomass relative to electricity
Abatement Cost (£/tCO Industrial biomass relative to electricity -50 Commercial biomass relative to gas
Residential biomass relative to gas -100 Residential biomass relative to oil
Solar thermal relative to oil -150
EfW CHP
12
2.4
1.6
18.1
23.6
16.1 20.3 24.8
19.3
12.8 20.9 Solar thermal relative to gas
Abatement potential 2020 (MtCO2) Heat pumps relative to oil
As stated in the documentation, unlike the EWP MAC curve the Heat MAC curve is not a policy appraisal tool. It is an indicator of what abatement potential in 2020 different technologies represent in a market context. This consideration of the market context means that the cost curve was initially developed to reflect a realistic potential (although estimates of technical potential have also been provided). This means that consideration has been given to what might be realistically achievable for a given carbon budget in view of current or emerging policy, supply chain constraints, infrastructure development etc. Care has been taken to ensure that the different sections of the cost curve do not overlap so as to avoid double counting.
The cost-effective potential under the realistic case is 16.4 MtCO2 compared to 48.6 MtCO2 under the technical potential case, as shown below in Figure 5.2. The difference between the two curves illustrates the importance of considering a more realistic view of potentials in future years, and also the range of possibilities.
46 AEA Energy & Environment Restricted – Commercial Review and update of UK abatement costs curves for AEA/ED43333/Issue 3 the industrial, domestic and non-domestic sectors
Figure 5.2 Heat MACC (2020) – full and realistic potential comparisons
300
250
200
150
100
£2006 /tCO2 50
0 0 10 20 30 40 50 60 70 80 90 100 -50 MtCO2 abated -100 Full potential Haircut potential -150
The heat cost curve illustrates that an important contribution to cost effective abatement can be realised through CHP. The contribution of industrial CHP is based upon a single example technology for costs, while the potential is derived as the difference between estimates in the AEA potential study (2007) and the assessments by BERR of likely BAU take-up.
Biomass also offer a significant potential for abatement potential – a realistic potential of 4.33 MtCO2 in 2020, or technical potential of 12.7 MtCO2. The cost-effectiveness of biomass depends on the reference technology assumed; against electricity-based heating it is cost effective at £-101/tCO2. Marginal abatement costs against oil are higher, at £60/tCO2.
The characteristics of the heat cost curve modelling are listed in Table 5.1, using the review framework adopted across all cost curves reviewed in this study. Our understanding of this work has been informed by a meeting with the OCC analyst responsible for the heat MAC curve work, plus supporting documentation38 and data received.
38 Internal documentation labelled OCC Heat Project Marginal Abatement Curve received from OCC
AEA Energy & Environment 47 Review and update of UK abatement costs curves for Restricted – Commercial the industrial, domestic and non-domestic sectors AEA/ED43333/Issue 3
Table 5.1 Key characteristics – OCC heat cost curve
Issue Key assumption Comment Source model OCC Heat cost curve Purpose of MACC Potential in view of current trends / policy A full technical potential has also been provided Fuel prices BERR 2020 estimates Emission factors - Defra, GHG Conversion factors for Company
electricity reporting Emission factors – Defra, GHG Conversion factors for Company
fossil fuels reporting Time horizon 2020 Base year Baseline measures UEP 29 (pre-EWP 2007) No sub-sectoral disaggregation of the industry Sector disaggregation To some extent sector Consideration of split by DA / traded sectors will Spatial disaggregation No be made by AEA where these data are integrated Coverage of No breakdown of CHP for industry by sub-sector Comprehensive measures or CHP type Renewable heat estimates from Defra / BERR Potential of measures Range of sources (2007), CHP / part of district heating from AEA (2007) Penetration rates As per the sources on abatement potentials Hidden and missing No indication these were included costs Other costs None Measure of cost- 25-year Net Present Value (NPV) of the heat
effectiveness technology divided by lifetime CO2 savings Investment costs Investment period Lifetime of measure A public sector discount rate of 3.5% is used, Costs of capital between 5-12% depending on Discount rates in addition to a separate annualised cost of measure capital If any measures are included from this cost curve Interaction / overlaps Overlaps between measures identified and in the domestic / non-domestic sectors some between measures removed consideration of interactions may be required No account taken of early retirement, and Timing of measures resulting increased annualised costs
5.1.2 Defra product policy cost curve
A product policy cost curve has been developed by Defra (supported by AEA), primarily using data from the Market Transformation Programme (MTP) evidence base39 but also other sources. The objective of the cost curve was to help understand the potential costs of CO2 reductions (and cost- effectiveness) associated with product policy, as estimated for the EWP 2007. It estimates the potential for carbon savings and associated costs in 2020 based on the current policies in place, and assuming MTP drives penetration of appliances in the market place as projected. There is therefore a reference projection, and a policy scenario, known as P1. A cost curve reflecting technical potential has not been developed, although estimates of potential carbon reductions have been, using a scenario called Earliest Best Practice (EBP). Additional detail on the different scenarios is provided below:
The following are extracts from the consultation documents “Sustainable products – Improving the efficiency of energy-using products” which were published in December 2007, specifically the domestic lighting paper (available from http://www.mtprog.com/Whitepaper.aspx#Downloads).
The Reference projection takes into account underlying trends in markets and technologies and the estimated or implicit impacts of historical and current policy measures. It does not, as yet, take account of the impact of policies announced in the Energy White Paper, which are still being developed and are not targeted at specific products (for example, CERT and successor schemes). The intention is to revise these projections once it becomes clearer how these new policy measures will affect domestic lighting.
The Earliest Best Practice (EBP) projection shows what would happen if all new UK sales were based on the most resource efficient options, taking into account design and production cycles, but not taking account of price or other market barriers. There is no cost information for this case; in addition, no ‘whole product policy case’ has been developed, so no account is taken of overlaps.
39 MTP What IF? Tool, http://whatif.mtprog.com/
48 AEA Energy & Environment Restricted – Commercial Review and update of UK abatement costs curves for AEA/ED43333/Issue 3 the industrial, domestic and non-domestic sectors
The P1 projection sets a target level of ambition that the Government is proposing could be delivered at a reasonable cost, taking into account such things as current UK and global performance benchmarks, economies of scale and the capacity of the supply chain to take coherent action to deliver more energy efficient products.
The product policy cost curve has been developed using two approaches; 1) by assessing each product group individually, and 2) for product policy as a whole. The product policy assessment (known as whole product policy in this report) is probably a more realistic reflection of cost-effective potential as it has involved provision of additional information on product overlap, adjusting figures against BERR’s figures for energy use, and allowing for savings made by policies related to products owned by other parts of Defra and other Government departments (e.g. CERT, Building regulations, Enhanced Capital Allowances).
Key characteristics of the product policy cost curve are set out in Table 5.2. Our understanding of this work has been informed by discussion with AEA colleagues from the MTP team, documentation from the Defra team40, and a meeting to discuss the work.
Table 5.2 Key characteristics – MTP product policy cost (PP) curve
Issue Key assumption Comment Source model Energy using products cost curve Purpose of MACC Policy assessment Data sourced from BERR. Long-run marginal resource Market prices assumed / Sensitivity Fuel prices cost (LRMC) nets-off the cost of infrastructure from using LRMC energy prices Emission factors – Consistent with Defra data (no changes
electricity in future years) Emission factors – fossil Consistent with Defra data fuels Time horizon Appraisal period – 2007-2020 Base year 2007 Some measures announced in the EWP 2007 (UEP 30) have been included in the P1 scenario. Some have Baseline measures Consistent with BERR UEP29 not due to lack of detail on proposed policy measure. None have been included in the base case. By product group – and by end use No split by product group for the assessment of whole Sector disaggregation sector product policy case Spatial disaggregation UK only Comprehensive coverage of product Coverage of measures groups Another scenario – Early Best Practice (EBP) – MTP outlook on future market plus Potential of measures provides full potential but overlaps between product current and envisaged policy groups not accounted for Penetration rates MTP modelling Based on What-If market penetration model Hidden and missing Not included costs These are key to this analysis; increasing product Other costs Policy costs included1 penetration in the market place requires significant resource Measure of cost- NPV (using 13 year appraisal period) effectiveness ICT (non-domestic and domestic sector), and domestic cooking (gas) and heating (oil and gas) assumed to Investment costs Additional cost of replacement measure have zero capital costs, assuming no link between efficiency and price Investment period Lifetime of measure Discount rates 3.5% Interaction / overlaps Consideration of overlaps in whole Overlaps not removed in individual PP cost curves between measures product policy No account taken of early retirement, Timing of measures and resulting increased annualised costs
1 Policy costs are estimated for influencing, regulatory and intervention policies. For influencing and regulatory policies, the government cost is administration, and the industry cost is for compliance testing. For intervention policies, two elements are identified: an administration cost to government for each product area, and the intervention itself, which generally yields a cost to one party and a benefit to another.
40 Internal documentation labelled Energy using Products: Analysis of Cost-effectiveness received from Defra’s SCP unit
AEA Energy & Environment 49 Review and update of UK abatement costs curves for Restricted – Commercial the industrial, domestic and non-domestic sectors AEA/ED43333/Issue 3
The cost curve analysis suggests that the potential for cost-effective savings is very high. In 2020, under the whole product policy case (having accounted for overlaps etc), 14 MtCO2 could be saved at an average abatement cost of -£80/tCO2, in effect a negative cost or saving. This analysis reflects a scenario of 50% of (net-of-tax) retail prices which is likely to reflect LRM(R)C. 100% (net-of-tax) retail price was used for domestic heating (oil).
The range of abatement costs (by product policy group) are shown in Figure 5.3 below.
Figure 5.3 Costs per tonne CO2 abated by product group,Defra product policy cost curve Product Areas 50
0
-50 Non-domestic Light Domestic Light Domestic ICT Domestic Electronics £/tonne Non-domestic ICT -100 Domestic Heating (Oil) Non-domestic Fridge Non-domestic Motors Domestic Heating (Gas) Domestic Cooking (Gas) Domestic Fridge -150 Street & Traffic Lighting Non-domestic AC Domestic Cooking (Electricity) Domestic Wet
-200
Source: Defra internal documentation
5.1.3 Micro-CHP
The main focus for micro-CHP (micro Combined Heat and Power) systems, are in a domestic or small business premises to provide heat and electricity, which can either be used on site or exported to the electricity grid.
These are typically small installations, and are normally defined as mini-generators at the lower end of the conventional CHP technology range (up to 75kWe), or as particularly small installations (~1kWe) in domestic and small commercial premises.
A current study of domestic micro-CHP (Stirling engine applications)41 using latest Carbon Trust trial data, gives a significantly worse view of the potential for carbon savings compared with earlier assessments; with in many cases no carbon being saved. The following figure shows a histogram of the distribution of the lifetime carbon savings; and compares the savings values with the original study savings. Less than half of trial households save carbon.
41 The Impacts of Distributed Generation on the Wider UK Energy System (April 2008); an update study by AEA and the Carbon Consortium for Defra. This latest study is an update on a previous 2007 report and is based on new trial data.
50 AEA Energy & Environment Restricted – Commercial Review and update of UK abatement costs curves for AEA/ED43333/Issue 3 the industrial, domestic and non-domestic sectors
Figure 5.4: Lifetime carbon saving distribution for individual household (tCO2)
9 Losses Savings 8
7
6 Orginal
5 analysis savings; 4 3.29tCO2
3
2
1
0 -10 -9 -8 -7 -6 -5 -4 -3 -2 -1 0 1 2 3 4 5 6 7 8 9 10
Carbon savings (tCO2)
For those households which do save carbon, costs are high. The following table summarises the net costs to a typical sample of these user, showing that in every case, the NPV is positive and that the technology costs (in £/tCO2) is high at a minimum of +£270/tCO2 saved, but typically much higher.
Table 5.3: Estimated lifetime CO2 savings and cost factors for those CT trial households showing positive CO2 savings
Lifetime Lifetime NPV Lifetime tCO2 (£, 3.5% DCR) £/tCO2 saving Site A +2.85 +1,605 +560 Site B +0.28 +1,843 +6,620 Site C +0.37 +1,893 +5,140 Site D +0.46 +1,796 +3,870 Site E +5.29 +1,436 +270
The micro-CHP studies to date, have only trialled Stirling Engine systems and have not included fuel- cell based systems. The operation of as yet ‘unspecified’ fuel cells in the above domestic situations, are more likely to save carbon, because of the inherent operating efficiency improvements, but are unlikely to offer operating costs that provide cost effective investments. This is supported by a recent example run with the AEA micro-CHP model, where an example fuel cell operating in a typical household, gave a lifetime carbon saving of +10.5 tCO2 but at a cost of +£280/tCO2 saved.
Characteristics of MICRO-CHP technologies in modelling work
Stirling engine: Electricity output 1.2kWe, heat to power ratio 10:1, electricity generation efficiency 7.8%, total efficiency 72.8% - heat out 12 kW Fuel cell: Electricity output 1.0kWe, heat to power ratio 5.5:1, electricity generation efficiency 12.18%, total efficiency 79.2% - heat out 5.5 kW
Data suggests that there are 1.6 million boilers replaced in the UK each year and we assume that only a very small fraction of these by 2012-2017 could be micro-CHP, mostly because of the difficulties in getting the industry off the ground, but also because Stirling engines would only be suitable for large homes. A ‘realistic potential’ would be a few tens of thousands, offering annual savings of 4-5 ktCO2 by 2017. By 2022, fuel cells could be commercially available offering higher potential savings of 15 ktCO2 pa at a marginally lower cost.
For the commercial sector, micro/mini-CHP (33 kWe up to 75 kWe engines) does save carbon. However, again total system net costs are all positive under baseline conditions; despite central
AEA Energy & Environment 51 Review and update of UK abatement costs curves for Restricted – Commercial the industrial, domestic and non-domestic sectors AEA/ED43333/Issue 3 generation cost savings. Results show that for all of the service sector cases studied, the overall cost effectiveness (to 2022) range from between +(plus) £6/tCO2 for hotels/restaurants (medium to large), up to +£70/tCO2 for the education sector. Central generation savings (both cost and carbon) tend to balance the CHP losses; which for the user range between +£16 to +150/tCO2 saved. We include the commercial micro/mini-CHP as a potential option covered in the commercial/services sector MACCs.
Since completing the analysis, new information has been provided through a major new report on microgeneration by Element Energy (2008). This report reflects a more optimistic view of the role of micro-CHP than has been determined in this report, based on the assessment by the Carbon Trust. It is worth noting the key uncertainties in the performance of these technologies in future years, particularly emerging technologies, such as the fuel cell version.
5.1.4 Behavioural measures
Defra’s Sustainable Consumption and Production team contracted AEA under a separate study to assess the carbon abatement potential and costs of a range of different behavioural measures. This project explores the potential uptake of such measures, taking account of the current and future behaviour of different socio-economic population groups. This project is not only considering behavioural measures associated with energy use but also transport, waste and food, and therefore is broad in scope. At the time this report was finalised, this work had not yet been published.
5.2 Proposal for integration: Heat technologies
There are three technology types covered by the OCC heat curve, which we have considered for integration – CHP, district heating and renewable heat. It is important to note that a robust integration of the above technologies in the buildings sectors would take account of interactions between different measures. For the purposes of this work, the interactions relating to so-called additional measures have been ignored. In the time horizon being considered, we do not think that this is problematic, due to the low projected uptake of such measures.
5.2.1 CHP
The OCC heat curve includes industrial CHP; however, only as a single technology ‘step’ and assuming the costs associated with a single ‘typical’ industrial CHP application.
Past estimates of the MACC for industrial CHP (carried out in connection with evaluating EU ETS sectors) have been carried out at a more detailed sub-sector level. The chemicals industry, paper and vehicle manufacture were considered in the earlier study associated with updating ENUSIM, as they were included in the EU ETS through combustion plant.
There were two aspects to the opportunities for abatement in these sectors: the first is associated with the overall reduction in demand for process heat, and the consequent effect on direct fossil fuel emission from heat raising plant (steam boilers), which is dealt with in ENUSIM; the second is the scope for increasing CHP. For CHP we developed a methodology for deriving MAC curves and applied it to chemicals, paper and board, and vehicle manufacture.
The methodology developed to create sector specific CHP supply curves was based in part on the outputs from the Cambridge Econometrics 2003 CHP projections study42, however using additional sector disaggregation, and a more detailed analysis of technology costs and their future take-up distribution.
For the current study we have extended the sector coverage to all key industrial sectors (relevant for CHP technology) and have reviewed and updated the previous study. We have used the outputs from the AEA modelling work in connection with the EC Cogeneration Directive43, which is based upon a detailed bottom-up model of potential CHP application areas:
42 Modelling Good Quality CHP Capacity to 2010: Revised Projections’ November 2003 43 CHP potential in the UK, AEA report, March 2007.
52 AEA Energy & Environment Restricted – Commercial Review and update of UK abatement costs curves for AEA/ED43333/Issue 3 the industrial, domestic and non-domestic sectors
Chemicals; Engineering; Food and drink; Paper and board; Other
The model provides estimates of total technical potential going forward and it is therefore necessary to deduct the BERR ‘with existing policies’ CHP projections44 from these potentials, as shown in the following tables of capacity and carbon savings projections, to produce the estimate of the remaining technical potential.
Table 5.4: A comparison of the all technically possible level of CHP by year (AEA model) compared with the estimated BAU projections (BERR)
2008 From AEA model BAU comparison
MW e ktCO2 MW e ktCO2
Chemicals 1,956 -6,108 1,013 -1,275 Food 1,158 -3,474 524 -663 Engineering45 3,056 -7,480 26 -15 Paper 431 -1,595 537 -821 Other 413 -1,034 46 -84
Commercial 2,374 -6,140 187 -166
2012 From AEA model BAU comparison
MW e ktCO2 MW e ktCO2
Chemicals 3,146 -9,186 1,050 -1,321 Food 1,197 -3,671 609 -770 Engineering 3,916 -9,608 45 -26 Paper 440 -1,681 601 -918 Other 446 -1,128 69 -126
Commercial 2,574 -6,717 338 -277
2017 From AEA model BAU comparison
MW e ktCO2 MW e ktCO2
Chemicals 3,775 -10,797 1,077 -1,356 Food 1,168 -3,743 728 -922 Engineering 4,402 -10,881 85 -49 Paper 412 -1,669 660 -1,008 Other 415 -1,093 137 -249
Commercial 2,874 -7,483 368 -301
44 The BERR projections have been obtained from the BERR modelling team and are consistent with the CHP projections used for the White Paper analysis. These have not changed substantially from earlier UEP projections. 45 The AEA model projects a very high potential for the engineering sector, which has now been reviewed and reduced by AEA experts in their assessment of the ‘possible potential’ by year for the current MAC curves.
AEA Energy & Environment 53 Review and update of UK abatement costs curves for Restricted – Commercial the industrial, domestic and non-domestic sectors AEA/ED43333/Issue 3
2022 From AEA model BAU comparison
MW e ktCO2 MW e ktCO2
Chemicals 4,480 -12,596 1,105 -1,391 Food 1,083 -3,705 872 -1,103 Engineering 4,372 -10,928 160 -92 Paper 375 -1,637 725 -1,107 Other 321 -946 270 -492
Commercial 3,208 -8,337 400 -327
The total technical potential remaining, and the percentage of that remaining potential which is technically achievable by year is shown in the following table and figure. The ‘achievable’ potential has been estimated by the AEA CHP QA team; it also includes an assessment of the commercial/services sector potentials.
A part of the extension of this study to consult industry over some of the sector MAC curves will include a further review of the achieveable potential for CHP. The CHP MACC model allows the achievable remaining potential to be changed using the same ‘percentage’ format as in the following table.
Table 5.5: Percentage of remaining potential technically achievable by year – latest AEA CHP QA team assessment
2008 2012 2017 2022 Chemicals 3.3% 10% 25% 45% Food 3.3% 10% 25% 40% Engineering 5.0% 15% 30% 40% Paper 10.0% 30% 60% 75% Other 5.0% 15% 25% 50%
Commercial 8.3% 25% 40% 50%
54 AEA Energy & Environment Restricted – Commercial Review and update of UK abatement costs curves for AEA/ED43333/Issue 3 the industrial, domestic and non-domestic sectors
Figure 5.5: Estimate of the total technical potential (installed capacity, MWe) of CHP compared with the estimated BAU projections (BERR), the potential remaining and the assessment of that technically achievable by year
Chemicals Chemicals remaining and achievable by year
6,000 6,000
5,000 5,000
4,000 4,000 Total potential Potential remaining 3,000 Potential remaining 3,000 Technically achievable BAU 2,000 2,000 by year
1,000 1,000
0 0 2006 2012 2017 2022 2006 2012 2017 2022
Food Food remaining and achievable by year
2,500 2,500
2,000 2,000
1,500 Total potential 1,500 Potential remaining Potential remaining Technically achievable 1,000 BAU 1,000 by year
500 500
0 0 2006 2012 2017 2022 2006 2012 2017 2022
Engineering Engineeringremaining and achievable by year
1,200 1,200
1,000 1,000
800 800 Total potential Potential remaining 600 Potential remaining 600 Technically achievable BAU 400 400 by year
200 200
0 0 2006 2012 2017 2022 2006 2012 2017 2022
AEA Energy & Environment 55 Review and update of UK abatement costs curves for Restricted – Commercial the industrial, domestic and non-domestic sectors AEA/ED43333/Issue 3
Continued
Paper Paper remaining and achievable by year
1,200 1,200
1,000 1,000
800 800 Total potential Potential remaining 600 Potential remaining 600 Technically achievable BAU 400 400 by year
200 200
0 0 2006 2012 2017 2022 2006 2012 2017 2022
Other Other remaining and achievable by year
700 700
600 600
500 500 Potential remaining 400 Total potential 400 Potential remaining 300 300 Technically achievable BAU by year 200 200
100 100
0 0 2006 2012 2017 2022 2006 2012 2017 2022
Commercial Commer remaining and achievable by year
4,000 4,000 3,500 3,500 3,000 3,000
2,500 Total potential 2,500 Potential remaining 2,000 Potential remaining 2,000 Technically achievable 1,500 BAU 1,500 by year 1,000 1,000 500 500 0 0 2006 2012 2017 2022 2006 2012 2017 2022
For the assessment of CHP costs we have adopted the AEA cost modelling procedure, which creates the individual CHP application costs and benefits to derive marginal cost data (£/tCO2 saved). For this we require, by sector and potential application:
. The appropriate technology types and typical sizes (capacities); . Operating performance data relevant to the sector and technology type – capacity, heat and electricity output, efficiency (and hence, fuel input); . Capital and installation costs and operating costs by type and sector;
The technologies considered include to following (prime mover) types46: . Gas engine; . Simple cycle gas turbine (GT); . Combined cycle gas turbine (CCGT); . Pass out condensing steam turbine; . Back pressure steam turbine.
The next set of figures show the combined manufacturing CHP MAC curves - for the technically achievable potential by year. This potential moves from 700 ktCO2 pa at 2008 to almost 10 MtCO2 by 2022, with a projected cost effective contribution of almost 7.5 MtCO2 by 2022.
46 The latter two are not considered as potential technologies for the future, but are included in the current population of CHP plants.
56 AEA Energy & Environment Restricted – Commercial Review and update of UK abatement costs curves for AEA/ED43333/Issue 3 the industrial, domestic and non-domestic sectors
Figure 5.6: Manufacturing CHP MAC curves - for the ‘technically achievable’ potential by year
150
100
50
2008 £/tCO2 0 0 100 200 300 400 500 600 700 800
-50
-100 CUM savings (ktCO2)
150
100
50
2012 £/tCO2 0 0 500 1,000 1,500 2,000 2,500
-50
-100 CUM savings (ktCO2)
150
100
50
0 2017
£/tCO2 0 1,000 2,000 3,000 4,000 5,000 6,000
-50
-100
-150 CUM savings (ktCO2)
150
100
50
0 2022
£/tCO2 0 2,000 4,000 6,000 8,000 10,000
-50
-100
-150 CUM savings (ktCO2)
AEA Energy & Environment 57 Review and update of UK abatement costs curves for Restricted – Commercial the industrial, domestic and non-domestic sectors AEA/ED43333/Issue 3
The CHP analysis has been provided separately from the buildings and industry sector cost curve spreadsheets.
5.2.2 District heating
Five district heating measures are included in the heat cost curve:
. City centre district heating schemes . Conversion of electric heated high rise to CHP . Conversion of existing district heating to CHP . District heating in high heat density residential areas . District heating in new build
Currently, district heating is not considered in the BRE sourced building sector MACCs. For this analysis, we have considered the integration of all of the above options with the exception of district heating in new build (as the building sector analysis only covers the existing building stock).
The following data have been included in the MACC for each district heating option:
Table 5.6District heating option data for build sector MACCs (in 2022) - 3.5% DR assumed
MACC Option £NAC/tCO2 £NPV/tCO2 Realistic Technical potential (Mt potential (Mt CO2) CO2) High density Domestic 2 -22 -14 0.11 11.75 urban schemes City centre Domestic 1 -86 -56 0.01 0.98 schemes Fossil district Domestic 3 -17 0.58 1.16 heating to CHP Electric high rise Domestic 3 -49 0.42 0.83 to CHP Non- City centre 1 -86 -56 0.04 2.94 domestic schemes
1 Assumes 25% of potential for domestic and 75% for non-domestic; assumes 20 schemes for realistic potential, and over 1500 under technical potential 2 Assumes 35 schemes for realistic potential, and over 3800 under technical potential
None of the assumptions have been changed from those used in the OCC heat cost curve. However, there are some areas of significant uncertainty, particularly concerning the technical potential, which is less than robust.
5.2.3 Renewable heat
Renewable heat options in both MACCs include solar thermal. The non-domestic sector also includes ground source heat pumps. We propose the inclusion of biomass heating, and heat pumps specifically in the domestic sector.
The estimates of abatement potential and cost originally used in the building sector MACCs were based on estimates from the OCC heat cost curve. New information from the recent Element Energy (2008) report has provided additional data on cost estimates, and is discussed in the next section on microgeneration.
58 AEA Energy & Environment Restricted – Commercial Review and update of UK abatement costs curves for AEA/ED43333/Issue 3 the industrial, domestic and non-domestic sectors
5.3 Proposal for integration: microgeneration
Estimates for the microgeneration technologies, including both renewable heat and electricity generating technologies, were first sourced from the OCC heat cost curve and earlier studies on microgeneration (EST 2007). In recent months, a new report by Element Energy (2008) has led to a revision of some of the initial assumptions made for domestic sector measures, primarily concerning costs.
In respect of electricity generation technologies, the MACCs currently provide good coverage. Only cost assumptions have been revised for the following technologies - photovoltaics , mini wind (5 kW) and micro-wind (1 kW). Energy savings by technology have not been revised against what was already in the MACC.
For renewable heat technologies, biomass and ground source heat pumps have been added, with technical characteristics from the OCC heat cost curve, and cost data from Element Energy’s recent report (2008). Solar thermal was already included; however, the cost estimates have been updated.
Micro-CHP is not currently included in the MACC. Further information on this technology is provided in Section 5.1.3 above.
Cost of technologies The capital cost assumptions are shown in Table 5.7 below.
Table 5.7 Microgeneration investment costs (£ unit installed – 2005 basis) in domestic sector
Measure 2012 2017 2022 PV 8108 6627 5732 Solar Water Heating 3602 3370 3225 Micro wind turbines (1kW) 3742 3245 3090 Mini wind turbines (5 kW) 16805 15901 15619 Residential Biomass (off gas grid) 7029 6122 5264 Heat pumps (off gas grid) 9816 8556 7820
Previous estimates from EST / Element Energy analysis (EST 2007) suggested greater reductions in the level of costs; however, new information from manufacturers and understanding of global technology development has resulted in more conservative estimates of progress.
For all of the above technologies, the assumption about technology performance remains the same. There is also no account for wind speed variation, another critical factor affecting cost-effectiveness, with an assumption of consistent performance across all buildings.
Technology potential Determining maximum technical potential is difficult for emerging microgeneration technologies. For the domestic sector, the following assumptions have been made by BRE (2007):
PV - Assuming all homes could have PV installed somewhere (even those without a roof) and based on the small number of known installations before 2005, the remaining potential in that year was estimated to be 24.992 million homes.
Mini-wind - It is estimated that about 1 million homes might be suitable for mini wind turbines and that none were installed in 2005. There is very little information on which to make this estimate, but clearly mini turbines (which are fairly substantial in size) will not be suitable for a majority of homes.
Micro-wind - Assuming 90% of homes with a roof could be fitted with a micro wind turbine and that there were zero installed in 2005, leaves an estimated potential of 17.1 million. However, a significant number of these would be considered unsuitable for a wind turbine based on current siting advice, so the potential would be rather lower in practice (in this work it has been set to 4 million).
AEA Energy & Environment 59 Review and update of UK abatement costs curves for Restricted – Commercial the industrial, domestic and non-domestic sectors AEA/ED43333/Issue 3
All of the above assumptions remain for the analysis. One potential improvement, particularly if this MACC was to be extended beyond 2022 could be to split wind potentials to reflect poorer / better sites, or urban / rural sites. The associated assumptions to calculate cost-effectiveness would also differ.
Solar thermal - in the BRE domestic sector MACC, it was assumed that over 800,000 existing homes will have solar thermal installed by 2022. This leaves an estimated 18.1 million dwelling potential in 2022, or 6.3 MtCO2 based on an estimation of all dwellings with roofs.
Biomass and heat pumps – assumed penetration of 800,000 homes for biomass and 100,000 homes for heat pumps by 2022 under the baseline. Total potential assumed is all off-gas grid properties of which there are 4.4 million.
5.4 Proposal for integration: Product policy MACC
There are a number of issues that make integration of the product policy (PP) cost curve into the building sector MACCs difficult.
1. Two approaches have been used for this MACC; one that accounts for overlaps but cannot be easily disaggregated by product group (whole product policy) and one that considers the potential and cost-effectiveness of individual products – with very limited account of overlap. 2. In this project one can only consider further the whole product policy case. As mentioned above, apportioning carbon savings by product group is difficult particularly due to the calibration required to match BERR data, and the apportioning of savings from policies to specific product groups. Cost per tonne values (based on the NPV calculation) can, however, be identified by product group. 3. The whole product policy case only provides what might be considered a realistic potential based on what policy might achieve. Technical potentials were not calculated as part of the product policy cost curve project. 4. The product policy is based on sales information, which presumably covers both existing and new buildings. This again makes comparison problematic.
There are a number of issues that we have considered in the process of integrating product policy data into the build sector MACCs. These are listed in Table 5.8.
Table 5.8 Integration issues for product policy (PP) cost curve
Issue Coverage of product For the non-domestic sector, the BRE analysis present much more detailed area measures. It does not cover the following PP product groups explicitly – motors and street lighting. Motors are likely to be covered in specific measures in the BRE analysis rather than as a generic group.
For the domestic sector, two PP groups, ICT and electronics, are not comprehensively covered in the BRE data – although integrated digital TVs and reduced standby consumption are included. These two product groups are estimated to account for up to 61%47 of overall savings from products in 2020 (excluding heating) so are potentially important.
Comparison of In 2022, the remaining technical potential for the products in the BRE analysis is 2.73 potential savings MtCO2. The PP domestic realistic potential for similar appliances is 6.85 MtCO2 (2.65 without ICT / electronics).
Whilst the PP analysis is based on realistic potential across old and new build, we consider that there may be cost-effective opportunities in electronics and ICT not covered in the BRE analysis.
In the Defra analysis, for the non-domestic sector, the realistic potential for product groups (not heating-based) is 6.44 MtCO2 in 2020. It is lower in the BRE curve when
47 Note that this estimate is highly uncertain for reasons discussed concerning splitting out of potentials by product group.
60 AEA Energy & Environment Restricted – Commercial Review and update of UK abatement costs curves for AEA/ED43333/Issue 3 the industrial, domestic and non-domestic sectors
account is taken of all of the interactions in the sector. The BRE work is very comprehensive on appliances, and relies on the data used in the MTP.
Cost data Assumption in the BRE analysis of the domestic sector is that the additional costs of assumptions more efficient appliances is small e.g. additional £20 for cold appliances, £10 for wet appliances.
MTP suggest that there is little evidence that price and efficiency are linked with brand and capacity more important factors. The PP analysis put additional cost for the above categories at an additional £200 for cold appliances, £150 for wet appliances (including VAT). However, they stress that these figures are very uncertain.
We have used the same assumptions as used in the BRE MACCs. Modelling of heat The heat replacement factor reflects the fact that reductions in energy use by replacement effect appliances needs to be compensated by additional heating, as appliances actually provide some of the household heating.
The HRE factor is the proportion of total CO2 savings (net of increased energy requirements for heating) as a proportion of total electricity savings. Across all product groups in the BRE analysis this broadly the same, and is around 66% in 2022 (although this will be a function of electricity carbon intensity).
Whilst there is more variation between product groups in the PP analysis, the above values used by BRE broadly look correct.
Penetration rates The penetration rates assumed by BRE are based on information from MTP. We are therefore assured that the two analysis will be broadly consistent with regards to assumed baseline uptake.
In general, we think that using the BRE data is the more suitable data – as this is in a consistent format with other measures, is part of the the BRE building model analysis, and suited to the needs of the MACC analysis. In addition, much of the information informing the BRE analysis is sourced from MTP, particularly in respect to uptake of measures.
We have used the product policy cost curve analysis to make refinements to the BRE analysis in the following ways:
. Added electronic and ICT product groups to the domestic sector MACC . Used the uptakes rates to inform the realistic potential
Incorporation of the electronic and ICT product groups in domestic sector model Electronic and ICT product groups have been incorporated into the domestic model. Products within these groups are shown below.
ICT products Electronic products Non-thermal printers and MFD Power supplies Thermal printers and MFD Set-top boxes Computers TVs Monitors Video
To avoid double counting, digital TVs have been removed as a product category from the BRE cost curve.
The information for these new product groups is on the basis of sales data, and therefore does not link to the exisiting house stock in the same way as other measures in the cost curve. It is on the basis of all sales (and the existing stock) so would cover new build in the domestic sector.
Development of realistic reduction potential estimates The information provided by BRE on remaining potential is based on the technical potential having taken account of uptake in the baseline.
AEA Energy & Environment 61 Review and update of UK abatement costs curves for Restricted – Commercial the industrial, domestic and non-domestic sectors AEA/ED43333/Issue 3
Data used in the product policy cost curve from the MTP programme provides an understanding of the carbon emission savings achievable in a given year, relative to the reference scenario, for two scenarios – Policy (P1) and Early Best Practice (EBP). We have assumed that P1 (as used in the product policy analysis) represent realistic potential in a given year, whilst the EBP scenario provides an indication of technical potential (at least on the basis of annual product sales).
The difference between EBP and P1 scenarios provides an indication of the difference between realistic and technical potential – and has been used to inform what the realistic potential might be across the products in the MACC.
5.5 Proposal for integration: Behavioural measures
The non-domestic sector cost curve covers behavioural measures quite comprehensively, where a behavioural measure refers to energy saving behaviour rather than technology choices. However, the domestic sector does not have any representation of behavioural measures.
We have therefore incorporated information on behavioural measures from ongoing work by AEA for Defra, which is assessing the impact of behavioural-based measures on policy scenarios, accounting for different behaviour of population groups in society. Due to the uncertainty surrounding the uptake rates of these measures, there is an option to include or not to include them in the cost curve.
Three energy-related measures have been considered for inclusion: . Reduce household heating by 1 C . Switch lights off when not in the room . Use 30 C setting on washing machine
The measure to install a 'smart' meter (energy usage monitor) has not been considered due to the potential overlap with the above three measures i.e. a smart meter helps realise the above behaviour.
The uptake of these measures in the reference case is based on work to understand the response of different socio-economic groups to policies promoting pro-environmental behaviour. The reference case is based on UEP30 basis, although a lower scenario (based on UEP29) has also been included. Two additional policy scenarios – Concerted Effort and Far Reaching – illustrate what might be possible by 2020, depending on resourcing of policy. We have used the Concerted Effort scenario to help inform the realistic potential.
Estimates for the following measures have been checked and agreed with EST.
Reduce household heating Estimate of 1340 kWh saving is based on the 1degC/10% energy reduction assumption. This was suggested for use by EST.
Reduce lighting ECI (1997) have estimated that the elimination of 'unintended on-time' (i.e. switching off the lights on leaving a room) could save 10% of lighting electricity consumption annually. A number of issues were not included in this estimate, for example cleaning bulbs / shades and not using higher wattage bulbs than necessary.
Domestic electricity consumption for lighting is estimated at 700 kWh/yr per household in 2007 (based on MTP data – see Appendix E6, Sustainable Products 2006: Policy Analysis & Projections). Taking account of the Heat Replacement Effect i.e. the additional heating requirements due to savings in lighting, an annual saving of 11.5 kWh/yr has been estimated.
Use lower washing temperature ‘Use 30oC setting on washing machine’ action, is based on current estimate of energy savings on information provided by the EST website: “washing clothes at 30oC instead of a higher temperature can use around 40% less electricity. This is estimated to save 80 kWh/yr per annum, on the basis that o all washes are at 30 C. 40% saving is again based on MTP data on washing machine energy use.
62 AEA Energy & Environment Restricted – Commercial Review and update of UK abatement costs curves for AEA/ED43333/Issue 3 the industrial, domestic and non-domestic sectors
These measures all have barriers to take up – and the requirement for resources into policies is apparent. Whilst this cost curve analysis does not incorporate policy costs, it need to be recognised that such costs are likely to be necessary to ensure that such measures are effectively implemented. In addition, cost associated with reduction in consumer welfare are not accounted for, which could result for example from lower comfort levels.
AEA Energy & Environment 63 Review and update of UK abatement costs curves for Restricted – Commercial the industrial, domestic and non-domestic sectors AEA/ED43333/Issue 3
6 MACC outputs
This section presents and discusses the MACCs developed for each sector. The MACCs provide an understanding of the remaining abatement potential (both cost-effective and not) at a given point in time based on a set of assumptions about future fuel prices, discount rates etc. This view of technical potential does not provide a clear picture of whether this could be achieved or realised at a given point in time (the realistic potential). CCC plans to build on this work to develop a better understanding of the realistic potential and to explore the effects of using different data and methodological assumptions. 6.1 Industry
The figures below show the industry cost curves for the years 2008, 2012, 2017, and 2022. For illustrative purposes the graphs have been created with the following assumptions: BAU scenario, fuel prices reflecting 100% CCL levels, hidden and missing costs are assumed as in the Enviros report, no realistic potential adjustment has been applied, emissions factors are as given by the CCC in the control panel.
The four curves represent the:
(Blue) The original ENUSIM run (with a fixed 15% discount rate applied to the run) without hidden and missing costs. (Pink) The user’s selected variables in the aggregated output sheet (in this case a 3.5% discount rate has been selected) and no hidden and missing costs applied. (Red and Green) The pink curve as above but with low and high, hidden and missing costs applied, respectively.
Any change to the output data will be reflected in all of the curves except the blue one, which shows the results from the original ENUSIM run.
2008
200 Original w ith 15% DR (+ user selected EFs) 150 User selected variables (exc. H&M costs) User selected variables + Low H&M costs 100 User selected variables + High H&M costs
50
0
£/tCO2 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 -50
-100
-150
-200 MtCO2
64 AEA Energy & Environment Restricted – Commercial Review and update of UK abatement costs curves for AEA/ED43333/Issue 3 the industrial, domestic and non-domestic sectors
2012
200 Original w ith 15% DR (+ user selected EFs) 150 User selected variables (exc. H&M costs) User selected variables + Low H&M costs 100 User selected variables + High H&M costs
50
0
£/tCO2 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 -50
-100
-150
-200 MtCO2
2017
200 Original w ith 15% DR (+ user selected EFs) 150 User selected variables (exc. H&M costs) User selected variables + Low H&M costs 100 User selected variables + High H&M costs
50
0
£/tCO2 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 -50
-100
-150
-200 MtCO2
AEA Energy & Environment 65 Review and update of UK abatement costs curves for Restricted – Commercial the industrial, domestic and non-domestic sectors AEA/ED43333/Issue 3
2022
200 Original w ith 15% DR (+ user selected EFs) 150 User selected variables (exc. H&M costs) User selected variables + Low H&M costs 100 User selected variables + High H&M costs
50
0
£/tCO2 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 -50
-100
-150
-200 MtCO2
The MAC curves reflect changes to the ENUSIM model which was revised after a brief consultation (reported in section 9). The sectors consulted were the food and drinks, engineering and chemicals. Other sectors in the model were recently updated following a previous round of consultation for the beginning of the UK ETS. The tight deadline for the consultation meant that there was limited input from stakeholders despite their willingness to engage in the process. A detailed revision of all the devices and technologies was not possible, nonetheless, several modifications to the model were made based on the feedback. These include a revised steam emission factor in all sectors and a update on several measures regarding penetration rates and emissions savings potentials. As a result, the new emissions savings from the revised model are slightly lower in all sectors.
6.2 Buildings (domestic)
The output of the building sector MACC has been assessed and the main findings of this assessment are discussed in this section. The two MACCs (measures in isolation and combined measures) for the three key years 2012, 2017, 2022 were analysed, and considering the following:
. The cost effective measures and the associated carbon saved . The cost effective measures given a marginal abatement cost of £30 (NPV MtCO2/yr) . Discount rates
The findings below exclude the cost curve behavioural measures because these are still to be finalised.
6.2.1 Cost-effective saving potential
Table 6.1 below shows the cost-effective (CE) potential for different cases in different budget years, both with and without renewable energy measures. These figures do not provide any indication of the realistic potential, as the estimates of what policy could achieve are not complete.
66 AEA Energy & Environment Restricted – Commercial Review and update of UK abatement costs curves for AEA/ED43333/Issue 3 the industrial, domestic and non-domestic sectors
Table 6.1 Mt CO2 saving in budget year – measures in isolation and combined measures cases
Year Group of Measures in isolation Combined measures measures Total Total – CE* Total Total – CE* 2012 All measures 113.2 51.1 105.8 45.1 2012 Excl. Ren 52.0 48.1 45.3 42.1 2017 All measures 104.9 43.2 99.2 38.3 2017 Excl. Ren 44.3 40.2 38.9 35.3 2022 All measures 97.4 35.6 93.9 33.0 2022 Excl. Ren 36.8 32.6 33.9 30.1
* Cost-effective assessment using NPV measure; ‘Excl. Ren’ indicate totals without so called ‘additional measures’ many of which are renewables.
An obvious conclusion from the above data is that renewable measures could contribute to significant savings in the longer term (due to their high potential); however, the potential savings are generally not cost-effective. This changes to an extent by 2022, as assumed learning rate has an effect.
Most cost-effective potential comes from fabric and heating-based measures, and electricity using appliances. In particular, solid wall insulation has a significant contribution to make (up to 30% of the CE total) although, as mentioned earlier in this report, there are a range of problems in implementing this measure e.g. limited success in finding market solutions. Potential savings in products such as ICT and electronics is also significant, as is the potential saving from condensing boilers. Cavity wall and loft insulation also offer significant cost-effective potential (in the absence of an assumption about post-CERT policy implementation).
The cost curve, showing technical potential in 2022, is shown below in Figure 6.6.
Figure 6.1 Domestic sector building cost curve in 2022 (using 3.5% DR) - measures in isolation case
1000
53 800
600
400
50 51 200
£/tCO2 48 49
47 0 41 39 40 Total 97.4 MtCO2 32 34 21 26 141720 -200 22 8 9
6 5 -400 2 3
-600 0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100
MtCO2
AEA Energy & Environment 67 Review and update of UK abatement costs curves for Restricted – Commercial the industrial, domestic and non-domestic sectors AEA/ED43333/Issue 3
1 Reduced standby consumption 19 Loft insulation 0 - 270mm 37 A-rated condensing boiler 3 2 ICT products 20 Pre76 cavity wall insulation 38 Loft insulation 150 - 270mm 3 Electronic products 21 Glazing - old double to future double 39 Fossil DH to CHP 4 Integrated digital TVs 22 DIY floor insulation (susp. timber floors) 40 Glazing - new double to future double 5 A+ rated wet appliances 23 A-rated condensing boiler 2 41 Residential biomass (off gas grid) 6 A++ rated cold appliances 24 Room thermostat to control heating 42 Urban retrofit DH 7 A rated ovens 25 Loft insulation 25 - 270mm 43 A-rated condensing boiler 4 8 A-rated condensing boiler 0 26 76-83 cavity wall insulation 44 Paper type solid wall insulation 9 City centre DH 27 Loft insulation 50 - 270mm 45 A-rated condensing boiler 5 10 Efficient lighting 28 Loft insulation 75 - 270mm 46 Modestly insulated cyl to high performance 11 Induction hobs 29 Thermostatic radiator valves 47 mini wind turbines 12 A-rated condensing boiler 1 30 Post '83 cavity wall insulation 48 Photovoltaic generation 13 Glazing - old double to new double 31 Improve airtightness 49 Heat pumps (off gas grid) 14 Glazing - single to new 32 Installed floor insulation (susp.TFs) 50 Micro-CHP Fuel cell 15 Insulated doors 33 Electric high rise to CHP 51 Solar water heating 16 Insulate primary pipework 34 Solid wall insulation 52 Hot water cylinder 'stat 17 Uninsulated cylinder to high performance35 Loft insulation 100 - 270mm 53 micro wind turbines 18 Glazing - single to future double 36 Loft insulation 125 - 270mm
6.2.2 Cost-effectiveness relative to a £30/tCO2 marginal cost
Our analysis has also considered the level of cost effective measures relative to a £30/tCO2 level across the different years.
Table 6.2 Mt CO2 saving in budget year – measures in isolation and combined measures cases
Year Group of measures Measures in isolation Total < £30 /tCO2 % increase from CE* 2012 All measures 53.1 -4% 2012 Excl. Ren 50.1 -4% 2017 All measures 45.9 -6% 2017 Excl. Ren 42.9 -7% 2022 All measures 59.0 -66% 2022 Excl. Ren 56.0 -72%
* Cost-effective assessment using NPV measure; ‘Excl. Ren’ indicate totals without so called ‘additional measures’ many of which are renewables.
The potential based on the £30 /tCO2 level increases as would be expected, particularly driven by renewable measures in 2022, with the reduction in costs associated with biomass. Most other measures (fabric, heating, products) are cost-effective in the main anyway, so the increases are much smaller. Including the high H&M cost estimates, this results in the cost-effective potential halving; this is using a non-working value of time for specific H&M categories. For the low case, the reduction is significantly lower – between 2-10% reductions depending on the year in question. The exception is for the year 2017 where, in the interactions case, solid wall insulation becomes marginally non-cost- effective and hence the CE potential reduces.
6.2.3 The Impact of increasing the discount rate
Some sensitivity analysis has been undertaken to assess the impact of increasing the discount rate to 7.5%. Table 6.5 below shows the cost-effective (CE) potential for different cases in different budget years, both with and without additional measures.
68 AEA Energy & Environment Restricted – Commercial Review and update of UK abatement costs curves for AEA/ED43333/Issue 3 the industrial, domestic and non-domestic sectors
Table 6.5 MtCO2 saving in budget year - measures in isolation and combined measures cases
Year Group of Measures in isolation Combined measures measures Total Total – CE* Total Total – CE* 2012 All measures 113.2 50.8 105.8 43.7 2012 Excl. Ren 52.0 47.8 45.3 40.8 2017 All measures 104.9 42.9 99.2 37.5 2017 Excl. Ren 44.3 39.9 38.9 34.5 2022 All measures 97.0 35.3 93.9 32.6 2022 Excl. Ren 36.8 32.3 33.9 29.6
* Cost-effective assessment using NPV measure; ‘Excl. Ren’ indicate totals without so called ‘additional measures’ many of which are renewables.
Increasing the discount rate to 7.5% has a very limited impact on the cost effective potential, with small percentage reductions. This discount rate leads to many measures only just being cost-effective, such as solid wall insulation and the thicker loft insulation types. A move to 8.5% discount rates sees many of these measures becoming non-cost-effective, with significant reductions in cost-effective potential e.g. in 2022, CE potential reduces by 40%. Using a 7.5% discount rate therefore results in many measures only being marginally cost-effective.
6.3 Buildings (non-domestic)
The abatement potential seen in the non-domestic MACC for the three budget years is provided in Table 6.4 below. These estimates are based on total potential, given that estimates of realistic potential have not been adequately developed for this sector, and require better understanding.
Table 6.3 Abatement potential for non-domestic sector, MtCO2 (using 3.5% discount rate)
2012 2017 2022 Total Total CE Total Total CE Total Total CE Measures in isolation 48.15 22.88 45.87 20.02 44.98 20.03 Measures in isolation - 18.69 19.86 18.43 renewables only Measures in isolation – 29.46 22.88 26.00 20.02 26.54 20.03 non-renewables only Combined measures – 19.57 14.94 17.35 13.23 17.62 13.21 non-renewables only
The cost curve shown below illustrates the above numbers for 2022, and the changes in the potential for CO2 savings.
AEA Energy & Environment 69 Review and update of UK abatement costs curves for Restricted – Commercial the industrial, domestic and non-domestic sectors AEA/ED43333/Issue 3
Figure 6.2 Commercial / public sector building cost curve in 2022 (using 3.5% DR) - measures in isolation case
900
700 53 54
500 50
300 £/tCO2 49
46 100
-100 32 24 25 14 7 11 20 2 4 6 -300 0 5 10 15 20 25 30 35 40 45 50
MtCO2
1 Lights- Most EE Replacement Tungsten Total 31 OffEq - Most EE Monitor Total 2 Heating - More efficient air conditioning Total 32 presence detector Total 3 Compressed air Total 33 Motor - 4 Pole Motor - EFF1 replace 4 Pole Total 4 Heating - Optimising Start Times Total 34 Heating - TRVs Fully Installed Total 5 Lights - Sunrise-Sunset Timers Total 35 Heating - most EE boiler Total 6 Heating - Programmable Thermostats High Total 36 Most EE pitched roof insulation Total 7 Lights - Basic Timer Total 37 Most ee double glazing Total 8 Lights - Light Detectors Total 38 Most EE flat Roof insulation Total 9 stairwell timer Total 39 Most EE external wall insulation Total 10 Most EE fridge Total 40 Lights - HF Ballast Total 11 Heating - Reducing Room Temperature Total 41 Most ee cavity wall insulation Total 12 Most EE freezer Total 42 Biomass 13 Photocopiers - Energy Management Total 43 Lights- Most EE Replacement Tungsten Total 14 Monitors - Energy Management Total 44 Most EE fridge-freezer Total 15 Printers - Energy Management Total 45 Lights - Metal Halide Floods Total 16 OffEq Fax Machine switch off Total 46 Lights - IRC Tungsten-Halogen - Spots Total 17 Vending Machines Energy management Total 47 Variable Speed Drives Total 18 OffEq - Most EE Monitor pc only Total 48 Heating - most EE boiler Total 19 Computers - Energy Management Total 49 Most EE fridge-freezer Total 20 Lights - Turn off Lights for an extra hour Total 50 Most EE freezer Total 21 OffEq - Most EE Monitor Total 51 Most EE fridge Total 22 presence detector Total 52 Most ee cavity wall insulation Total 23 Motor - 4 Pole Motor - EFF1 replace 4 Pole Total 53 Lights- Most EE Replacement 26mm Total 24 Heating - TRVs Fully Installed Total 54 Most EE pitched roof insulation Total 25 Heating - most EE boiler Total 55 Most EE flat Roof insulation Total 26 Most EE pitched roof insulation Total 56 Lights - HF Ballast Total 27 Most ee double glazing Total 57 Lights- Most EE Replacement 26mm Total 28 Most EE flat Roof insulation Total 58 Most EE external wall insulation Total 29 Most EE external wall insulation Total 30 Lights - HF Ballast Total
There is large carbon abatement potential in the all measures in isolation case, shown above, however, less than half is cost effective. The main non-cost effective measures are lighting measures (for example replacement with the most energy efficient), and also specific wall and roof insulation measures. Heating measures tend to be one of the most cost effective measures, and have high abatement potential.
Where all possible interactions are taken into account there is a fairly large reduction in the abatement potential. So-called additional measures are not included in the analysis as the interactions were not modelled.
70 AEA Energy & Environment Restricted – Commercial Review and update of UK abatement costs curves for AEA/ED43333/Issue 3 the industrial, domestic and non-domestic sectors
7 Comparison of abatement potential with the impact of policies
This section outlines the potential impact of new policies outside of the MACC baseline, with estimates of the abatement potential of these measures. The following policies have been considered:
. Supplier Obligation . Zero Carbon Homes . More energy efficient products . CRC . EU ETS . Continuation of CCAs
7.1 Supplier Obligation
The 2007 Energy White Paper reaffirmed the Government’s commitment to maintain some form of obligation on household energy suppliers until at least 2020, with an ambition level at least equal to that under the Carbon Emissions Reductions Target (CERT). Defra (2007) launched a Call for Evidence in summer 2007 to consult on the possible post-CERT mechanism.
In the above document the following saving of 13.2 MtCO2 were assumed on an annual basis by 2020 (in MtCO2):
. Cavity wall and loft Insulation – 3.7 . Lights / appliances – 2.9 . Behavioural change / reducing waste – 3.7 . Fuel switching – 1.8 . Microgeneration – 1.1
The above estimates pre-judge the types of options that energy suppliers will be considering.
Cavity wall and loft insulation In 2022, the MACC estimates that between 3.5 – 3.7 MtCO2 is remaining. The proposed Supplier Obligation (SO) would mean that most of that insulation potential would be taken up. This remaining potential is consistent with the value outlined above of 3.7 MtC.
Lights / appliances 6.7 MtCO2 potential remains in the 2022 MACC; based on the SO estimates, this would leave just over half of the potential from lights and appliances remaining.
Reducing waste The domestic MACC does not cover behavioural measures extensively. Of the three generic measures included, the potential in 2022 is still being investigated.
Fuel switching This is not explicitly covered by the MACC.
Microgeneration The domestic MACC indicates that there is a significant potential for abatement from microgeneration type measures (>75 MtCO2). The estimates of realistic potential in 2022 reduce this figure to 5.6 MtCO2. For comparison, the Supplier Obligation is expected to lead to about 1.1 MtCO2 of savings from microgeneration.
Overall, the SO is estimated to make annual savings of about 13 MtCO2; this compares to over 24 MtCO2 that is potentially available under an estimate of realistic potential.
AEA Energy & Environment 71 Review and update of UK abatement costs curves for Restricted – Commercial the industrial, domestic and non-domestic sectors AEA/ED43333/Issue 3
7.2 Zero carbon homes
The Government have set out an ambition in 2006 to make all new build ‘zero carbon’ by 2016. Communites and Local Government published Building a Greener Future: policy statement in July 2007. This policy statement confirmed the government's intention for all new homes to be zero carbon by 2016 with a progressive tightening of the energy efficiency building regulations - by 25% in 2010 and by 44% in 2013 - up to the zero carbon target in 2016.
Cyril Sweett (2007) have undertaken an analysis for Government to assess the likely carbon savings from meeting the zero-carbon target, and the cost of abatement. An optimisation model was used to assess the different measures that would need to be considered to meet the interim objectives relating to improvements in Part L1a 2006, and the zero carbon target in 2016. Different options were considered including:
1. All carbon measure savings had to be onsite 2. Allowing offsite generation after 2016 (use of offsite renewable energy generation to offset the carbon emissions from proposed developments) 3. Allowing offsite generation after 2016 but with mandatory energy efficiency backstops (presence of mandatory minimum levels of energy efficiency)
Annual emission reductions that would be achieved in 2020 would be just over 2.6 MtCO2. These savings are small in comparison to potential in the existing building stock; however, in the longer term the savings would be much more significant, based on one-third of the stock being built between now and 2050. If new build was included in the MACC, this would approximately represent the technical potential in 2022.
Abatement costs associated with the reductions in 2020 range from between £210/tCO2 (all carbon savings onsite), to £134/tCO2 (allow offsite generation but mandatory energy efficiency) down to £60/tCO2 where offsite renewable generation can be used.
For option 1, the main technologies of use in 2016 are biomass CHP (65% in overall cost terms), biomass heating (10%), and PV (25%). This is based on cost optimization assessment to meet the zero carbon homes target. For option 2, most energy needs are met exclusively by ROCs while for option 3, ROCs are important but thermal efficiency requirements mean that a significant proportion of costs are spent on this.
A zero-carbon homes policy proposal which is fully implemented in effect will mean that the full technical potential is likely to be achieved.
7.3 More energy efficient products
The realistic potential determined in the MACC for products and appliances is taken from recent work by Defra to assess what product policy could achieve by 2020. This was a more robust analysis of the cost-effectiveness of product-based measures following the EWP 07.
In 2022, the realistic potential for the domestic sector of appliances (excl. heating) was 4.8 MtCO2 versus a technical potential of 6.6 MtCO2. This realistic potential appears to be a reasonable estimate of what product policy could achieve in the domestic sector.
7.4 Carbon Reduction Commitment (CRC)
CRCs are estimated to deliver 0.76 MtCO2 pa in manufacturing industry by 2015 and 1.66 MtCO2 pa 48 by 2020. The following table shows the marginal delivery of CO2 from the adjusted MACC curves for
48 The MACC curves used in the delivery analysis have been developed as appropriate for the policy coverage and using delivered fuel price with the relevant level of CCL. Baseline assumptions from the original Enusim run are used (i.e. 15% Discount Rate and no Hidden & Missing Costs) with 100% CCL .
72 AEA Energy & Environment Restricted – Commercial Review and update of UK abatement costs curves for AEA/ED43333/Issue 3 the industrial, domestic and non-domestic sectors manufacturing at a zero carbon value and at £25/tCO2. At 2017 and 2022 the carbon delivery is typically less than 0.48 MtCO2, which suggests that the CRC delivery levels quoted are going to be expensive49.
Table 7.1 Carbon delivery at a zero and £25/tCO2 value of carbon using MACC curves for manufacturing industry covered by the CRC
CRC – manufacturing 2008 2012 2017 2022 Baseline - cost-effective at £0/tCO2 0.80 0.63 0.47 0.46 Baseline - cost-effective at £25/tCO2 0.80 0.63 0.47 0.46 Baseline (realistic potential) - cost-effective at £0/tCO2 0.66 0.53 0.41 0.42 Baseline (realistic potential) - cost-effective at £25/tCO2 0.66 0.53 0.41 0.42
These CRC sector coverage figures are based on the NERA/Enviros work for the CRC assessments, which were reproduced in the Regulatory Impact Assessment for CRCs. Meeting these commitments would not just be about building measures as smaller industries also have opportunities in many of the technologies listed in the industry MAC curves. Indeed the largest CRC coverage is precisely with those industries such as engineering, small food and drink, plastics which are low energy intensity. Measures such as improved motors and drives, better furnaces and improved compressed air systems are relevant here and likely to be cost-competitive with buildings measures.
From the non-domestic (commercial and public sector) buildings sector, delivery at the above prices is shown below. These figures exclude buildings measures in manufacturing industry, which are included in the industry MACC and in the figures given in Table 7.1.
Table 7.2 Carbon delivery at a zero and £25/tCO2 value of carbon using MACC curves for commercial and public sector buildings covered by the CRC
CRC – non-domestic buildings 2008 2012 2017 2022 Baseline - cost-effective at £0/tCO2 - 7.2 6.4 6.3 Baseline - cost-effective at £25/tCO2 - 7.2 6.4 6.3 * Based on combined combined measures case (3.5% discount rate)
7.5 Climate Change Agreements (CCAs)
CCAs have already over-delivered on the 10.6 MtCO2 policy delivery 2000-2010, which is included in the baseline for the industry MACC. The following table indicates that a further 3.8 MtCO2 (realistic potential) appears possible by 2012. Again no account has been taken of future structural shifts within manufacturing, which could also contribute to further emissions reductions.
Table 7.3 Carbon delivery at a zero and £25/tCO2 value of carbon using MACC curves for CCA coverage 50
CCAs 2008 2012 2017 2022 Cost-effective at £0/tCO2 5.08 4.76 4.4 3.87 Cost-effective at £25/tCO2 5.11 4.81 4.46 3.9 Baseline (realistic potential) - cost-effective at £0/tCO2 3.65 3.75 3.89 3.45 Baseline (realistic potential) - cost-effective at £25/tCO2 3.69 3.81 3.95 3.48
49 This analysis does not take account of future structural shifts within manufacturing, which could also contribute to reducing emissions. 50 Baseline assumptions from original Enusim run are used (i.e. 15% Discount Rate and no Hidden & Missing Costs) with 20% CCL.
AEA Energy & Environment 73 Review and update of UK abatement costs curves for Restricted – Commercial the industrial, domestic and non-domestic sectors AEA/ED43333/Issue 3
7.6 EU Emissions Trading Scheme (EU ETS)
For Phase II of the EU ETS to 2012 no additional delivery is required above 'with existing policy'. The following table shows that after 2012 an additional 1.2 MtCO2 could be possible.
Table 7.4 Carbon delivery at a zero and £25/tCO2 value of carbon using MACC curves for EU ETS coverage51
CCAs 2008 2012 2017 2022 Cost-effective at £0/tCO2 1.64 1.38 1.24 1.26 Cost-effective at £25/tCO2 1.66 1.4 1.26 1.27 Baseline (realistic potential) - cost-effective at £0/tCO2 1.2 1.18 1.17 1.2 Baseline (realistic potential) - cost-effective at £25/tCO2 1.21 1.20 1.19 1.21
The combined contributions from the above policy areas are not directly additive as policy overlaps occur in manufacturing between CCAs and the EU ETS (see previous section 3.2.4).
7.7 CHP measures
Both EU ETS and CCAs cover the types of manufacturing sites (and emissions) relevant for CHP. Most manufacturing sites in CRCs are likely to be unfavouravble for CHP applications; but not so in service sectors.
The manufacturing CHP MAC curves, with realistic technical potential and no hidden or missing costs, indicate the following cost effective potentials by year.
Table 7.5 Carbon delivery at £25/tCO2 value of carbon using MACC curves for manufacturing CHP
Manufacturing CHP 2008 2012 2017 2022 Cost-effective at £25/tCO2 0.53 1.71 4.22 7.61
The CHP delivery figures indicated are large; and are comparable, and by 2022, become significantly larger than, the potential CCA delivery for the manufacturing end-use areas (from ENUSIM). This result requires further industrial consultation, particularly regarding the ‘realistic technical potential’ we have currently assumed for 2017 and 2022.
From the above analysis, it is clear that there are a number of different policies introduced or reviewed post EWP 07 that target significant emissions reductions over the next 10 years. Significant additional potential remains, as reflected in the MACCs, but which is less cost-effective, that could be considered for future policy action.
51 Baseline assumptions from original Enusim run are used (i.e. 15% Discount Rate and no Hidden & Missing Costs) with 20% CCL. The EU ETS MACC is based on direct emissions only.
74 AEA Energy & Environment Restricted – Commercial Review and update of UK abatement costs curves for AEA/ED43333/Issue 3 the industrial, domestic and non-domestic sectors
8 References
AEA (2007), Analysis of the UK potential for Combined Heat and Power, On behalf of Defra, October 2007, http://www.defra.gov.uk/environment/climatechange/uk/energy/chp/pdf/potential-report.pdf
AEA (2005), Assessment of Emerging Innovative Energy Efficient Technologies as part of the Energy Efficiency Innovation Review, Prepared for by AEA (previously known as FES) on behalf of Defra, June 2005, http://www.defra.gov.uk/environment/climatechange/uk/energy/research/pdf/fes-report.pdf
AEA52 / Carbon Consortium (2005), Industrial sector carbon dioxide, A report for the Department for Environment, Food and Rural Affairs (Defra), July 2005
BERR (2007), Updated energy and carbon emission projections, Energy White Paper, Department for Business, Enterprise & Regulatory Reform, May 2007, URN 07/947. Also known as UEP 30.
BRE (2007), Delivering cost-effective carbon saving measures to existing homes, Prepared by John Henderson on behalf of Defra, Client report 239-552, Building Research Establishment Ltd, October 2007
BRE (2005a), Reducing carbon emissions from the UK housing stock, Prepared by L Shorrock, J Hendersonand J Utley on behalf of Defra, Building Research Establishment Ltd
BRE (2005b), Reducing carbon emissions from commercial and public sector buildings in the UK, Prepared by Christine Pout and Fiona Mackenzie on behalf of Defra, Client report 211-104, Building Research Establishment Ltd, 16th December 2005
BRE (2002), Carbon dioxide emissions from non-domestic buildings: 2000 and beyond, Prepared by C Pout, F Mackenzie and R Bettle on behalf of Defra, Building Research Establishment Ltd, 2nd March 2002, ISBN 1860815456
Carbon Consortium and Oxford Economics (2006), Research on Output Growth Rates and Carbon Dioxide Emissions of the Industrial Sectors of EU-ETS, On behalf of BERR, February 2006, http://www.berr.gov.uk/files/file26365.pdf
CLG (2006), Building A Greener Future: Towards Zero Carbon Development, Department for Communities and Local Government, December 2006
Cyril Sweet (2007), Research to Assess the Costs and Benefits of the Government’s Proposals to Reduce the Carbon Footprint of New Housing Development, Draft Final Report – Contract Number RAE 3/15/9, On behalf of Communities and Local Government, October 2007.
Defra / BERR (2007), Renewable heat initial business case, Analysis by Ernst and Young, URN 07/1468, September 2007, http://www.berr.gov.uk/files/file41432.pdf
Defra (2007), The Household Energy Supplier Obligation from 2011: A Call for Evidence, Department for Environment, Food and Rural Affairs, June 2007
ECI (1997), DECADE 2MtC report, Environmental Change Institute
Element Energy (2008), The growth potential for Microgeneration in England, Wales and Scotland, A report funded by The Ashden Trust, BERR, British Gas Service Ltd, Ceres Power plc, The Energy Saving Trust, E.On UK plc, Micropower Council, National Energy Action, Renewable Energy Foundation, North West Development Agency, East Midlands Development Agency, South West of England Development Agency, South East Development Agency and London Development Agency, June 2008.
52 AEA published under old business unit name Future Energy Solutions (FES)
AEA Energy & Environment 75 Review and update of UK abatement costs curves for Restricted – Commercial the industrial, domestic and non-domestic sectors AEA/ED43333/Issue 3
Enviros (2006), Review and development of carbon dioxide abatement curves for available technologies as part of the Energy Efficiency Innovation Review, Final report by Enviros Consulting Ltd, January 2006, http://www.defra.gov.uk/environment/climatechange/uk/energy/research/pdf/enviros-report.pdf
EST (2007), Generating the Future: An analysis of policy interventions to achieve widespread microgeneration penetration, TBC
EST (2005), Potential for Microgeneration, Undertaken by the Energy Saving Trust (EST) on behalf of the DTI, in conjunction with Element Energy Limited, E-Connect and Cambridge University Faculty of Economics, 14th November 2005, http://www.berr.gov.uk/files/file27558.pdf
HM Treasury (2007),The Green Book - Appraisal and Evaluation in Central Government, Treasury Guidance, The Stationary Office, http://www.hm- treasury.gov.uk/economic_data_and_tools/greenbook/data_greenbook_index.cfm
NERA/Enviros (2006), Energy Efficiency and Trading Part II: Options for the Implementation of a New Mandatory UK Emissions Trading Scheme, On behalf of Defra, 28 April 2006, http://www.defra.gov.uk/environment/climatechange/uk/business/crc/pdf/nera-enviros-report- 060428.pdf
Oxford Economics (2007), Report on modelling the macroeconomic impacts of achieving the UK’s carbon emission reduction goal, Report for BERR, May 2007
Sanders C and Phillipson M (2006), Review of Differences between Measured and Theoretical Energy Savings for Insulation Measures, On behalf of EST and funded by Defra, Glasgow Caledonian University, December 2006
Sanstad A, Blumstein C, Stoft S (1995), How high are option values in energy-efficiency investments? Energy Policy 1995, Vol. 23, No. 9, pp. 730-743
76 AEA Energy & Environment Restricted – Commercial Review and update of UK abatement costs curves for AEA/ED43333/Issue 3 the industrial, domestic and non-domestic sectors
AEA Energy & Environment 77
Appendices
Appendix 1: Sectoral growth rate projections Appendix 2: Fuel price projections Appendix 3: Relating the MACC annualised costs to the cost of targets using the DCF (NPV) calculation Appendix 4: An overview of ENUSIM Appendix 5: Description of ENUSIM Appendix 6: ENUSIM file structure and process steps Appendix 7: ENUSIM based bottom-up modelling approach – methodology linking MAC curves to target or policy costs Appendix 8: Hidden and missing costs Appendix 9: Industry consultation for the update of ENUSIM: results and conclusions
AEA Energy & Environment
Appendix 1
Projected sectoral growth rate data
AEA Energy & Environment
Table A1.1: BERR UEP projections at SIC 2 –digit consistent with EWP 07
UK Manufacturing Sector Data/Forecasts - Budget 07_EWP From BERRProportional revision1.00 manufacturing2.00 0.94 1.00 1.78 1.00 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 Proportional revision1.00 GDP1.22 1.00 0.92 1.10 0.80 0.80 0.80 0.80 0.80 0.80 0.80 0.70 0.70 0.70 0.70
SIC(92) Description 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 codes (2002=100) 15-37 Manufacturing Output 100.0 100.1 101.8 100.7 102.2 104.1 106.2 108.3 109.5 110.7 112.0 113.2 114.5 115.8 117.1 118.4 119.7 121.1 122.5 Of which (@ 2-digit disaggregation):
15,16 Food,Beverages & Tobacco 100.0 99.9 101.6 102.1 100.7 101.3 102.1 103.8 104.7 105.4 106.2 106.9 107.7 108.4 109.2 109.9 110.6 111.3 111.9 17,18,19 Textiles, Leather & Clothing 100.0 98.1 87.5 85.6 77.2 73.4 70.5 65.6 63.1 60.7 58.3 56.1 53.9 51.8 49.7 47.7 45.8 44.0 42.3 20 Wood & Wood Products 100.0 100.8 103.4 98.3 98.0 98.7 99.6 99.8 99.8 99.8 99.7 99.7 99.6 99.5 99.4 99.3 99.1 98.9 98.7 21,22 Pulp, Paper, Printing & Publishing 100.0 98.5 97.5 92.5 92.0 93.8 95.5 97.1 97.8 98.4 99.0 99.6 100.2 100.8 101.3 101.9 102.5 103.0 103.5 Of which: 21 Pulp, Paper and Paper products 100.0 100.2 101.4 99.2 97.2 99.2 101.0 102.9 103.6 104.2 104.7 105.3 105.9 106.4 107.0 107.5 107.9 108.4 108.8 22 Printing & Publishing 100.0 98.1 96.6 90.9 90.8 92.5 94.3 95.7 96.4 97.0 97.6 98.3 98.9 99.5 100.1 100.7 101.2 101.7 102.2 23 Coke, Petroleum & Nuclear Fuels 24 Chemicals & Man-made Fibres 100.0 100.9 104.2 105.2 108.8 111.3 114.7 120.4 124.1 127.8 131.7 135.6 139.6 143.7 148.0 152.3 156.7 161.1 165.7 25 Rubber & Plastics 100.0 100.8 99.3 95.5 97.6 98.8 100.6 102.7 103.9 105.0 106.2 107.3 108.5 109.6 110.8 112.0 113.1 114.2 115.3 26 Other Non-metallic Minerals - corrected for glass 100.0 105.8 111.8 111.8 115.2 116.5 118.3 121.0 122.3 123.5 124.6 125.9 127.2 128.5 129.7 130.9 132.2 133.4 134.6 27 Basic Metals 100.0 100.3 103.2 100.8 109.5 114.3 116.6 118.2 118.5 118.8 119.0 119.2 119.3 119.4 119.5 119.6 119.7 119.7 119.7 28 to 35 Engineering & vehicles 100.0 100.3 104.1 103.4 107.0 110.3 113.5 116.0 117.6 119.2 120.9 122.5 124.2 125.8 127.5 129.2 131.0 132.8 134.6 Of which: 28 Fabricated Metal Products 100.0 97.0 100.3 100.9 103.3 104.9 106.3 108.0 108.6 109.2 109.8 110.2 110.6 111.0 111.5 111.9 112.2 112.5 112.8 29 Machinery & Equipment not elsewhere classified 100.0 101.7 107.4 110.6 111.8 113.8 115.6 116.7 117.1 117.3 117.5 117.6 117.6 117.6 117.6 117.6 117.5 117.4 117.3 30 Computers and Office Equipment 100.0 94.5 71.7 69.3 72.4 74.2 76.4 79.9 81.9 83.9 86.1 88.3 90.5 92.8 95.0 97.3 99.7 102.1 104.6 31 Electrical Machinery & Equipment n.e.c. 100.0 94.1 96.7 91.4 90.9 93.3 95.9 98.1 98.9 99.5 100.0 100.6 101.2 101.7 102.2 102.7 103.2 103.7 104.2 32 Electronics & Telecommunications Equipment 100.0 87.6 96.8 86.9 93.8 99.3 105.9 115.5 121.4 127.1 133.1 139.4 145.9 152.8 159.9 167.3 175.1 183.2 191.6 33 Medical, Precision & Optical Instruments 100.0 106.6 111.2 110.6 116.1 119.2 122.8 127.6 129.2 130.6 131.9 133.3 134.7 136.0 137.3 138.4 139.7 141.0 142.2 34 Motor Vehicles & Parts 100.0 103.9 105.3 103.7 102.5 104.6 108.2 105.5 107.1 110.1 113.1 116.0 119.0 122.0 125.2 128.5 131.9 135.3 138.8 35 Other Transport Equipment 100.0 107.5 117.7 118.4 133.6 143.0 149.7 156.9 159.9 162.7 165.6 168.5 171.5 174.5 177.3 180.2 183.1 186.2 189.3 36,37 Manufacturing n.e.c. (incl. Furniture & Recycling) 100.0 100.3 100.1 96.8 95.0 95.2 95.6 95.8 95.9 95.9 96.0 96.0 96.1 96.1 96.1 96.1 96.0 95.9 95.8
45 Construction 100.0 105.2 108.7 109.9 112.3 115.7 118.5 121.1 123.1 124.8 126.5 128.2 130.0 131.8 133.6 135.2 136.8 138.4 140.1
COI DTI 'Other industry' category 20, 25, 36, 37, 45, 13, 14, 41 100.0 103.9 106.3 106.5 109.3 111.8 114.2 117.4 119.3 121.0 122.6 124.3 126.0 127.7 129.5 131.3 133.1 134.9 136.8
AEA Energy & Environment
Table A1.2: ENUSIM projections at SIC 3,4 –digit consistent with EWP 07
Detailed Manufacturing Sector Output SIC(92) Description 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 codes (2002=100) @ 4-digit SIC disaggregation where appropriate
15 Food & Beverages 100.0 100.0 102.1 103.0 102.3 103.0 104.0 106.0 107.0 107.9 108.8 109.7 110.6 111.5 112.3 113.2 114.0 114.8 115.6 Of which:
155 Dairy Products 100.0 99.9 99.2 98.2 97.7 98.9 100.0 101.9 102.8 103.7 104.6 1583 Sugar 100.0 103.9 99.6 101.6 96.2 96.7 97.6 98.6 99.0 99.3 99.6 1512 Poultry 100.0 102.2 102.8 105.6 107.6 110.4 113.6 119.8 123.3 126.9 130.5 159 Beverages 100.0 102.8 104.9 103.8 103.8 104.1 105.1 107.0 107.9 108.8 109.6 Of which: 1591/2 Distilled spirits 100.0 109.5 107.7 112.8 105.1 101.6 102.2 103.1 103.6 104.1 104.6 1593 Wine 100.0 91.2 104.9 96.5 88.7 88.8 89.1 89.4 89.2 88.8 88.2 1596 Beer 100.0 99.7 106.5 101.9 101.0 101.3 101.9 103.2 103.8 104.3 104.8 1597 Malt 100.0 100.4 95.5 85.5 100.8 108.5 116.7 125.1 129.8 134.6 139.5
16 Tobacco 100.0 97.7 92.8 85.6 74.8 73.0 71.3 68.4 66.9 65.3 63.8 62.3 60.9 59.6 58.2 56.8 55.5 54.2 52.9
20 Wood & Wood Products 100.0 100.8 103.4 98.3 98.0 98.7 99.6 99.8 99.8 99.8 99.7 99.7 99.6 99.5 99.4 99.3 99.1 98.9 98.7 Of which: 202 Woodboard 100.0 97.4 105.2 89.4 91.2 95.3 99.7 102.3 103.3 104.2 105.1 105.9 106.8 107.7 108.5 109.4 110.2 111.0 111.7
21 Pulp, Paper & Paper Products 100.0 100.2 101.4 99.2 97.2 99.2 101.0 102.9 103.6 104.2 104.7 105.3 105.9 106.4 107.0 107.5 107.9 108.4 108.8 Of which: 211 Pulp & Paper 100.0 98.6 95.8 99.4 97.0 98.6 100.7 104.8 106.2 107.6 108.9 110.3 111.7 113.0 114.4 115.7 117.0 118.4 119.7 212 Paper Products 100.0 100.8 103.6 99.2 97.3 99.5 101.0 102.2 102.5 102.9 103.1 103.4 103.6 103.9 104.1 104.3 104.4 104.5 104.6
24 Chemicals & Man-made Fibres 100.0 100.9 104.2 105.2 108.8 111.3 114.7 120.4 124.1 127.8 131.7 135.6 139.6 143.7 148.0 152.3 156.7 161.1 165.7 Of which: 2414 Petrochemicals 100.0 103.5 111.9 116.8 114.1 111.3 111.6 118.9 120.7 122.4 124.0 125.8 127.6 129.4 131.2 133.1 135.0 136.9 138.8
25 Rubber & Plastics 100.0 100.8 99.3 95.5 97.6 98.8 100.6 102.7 103.9 105.0 106.2 107.3 108.5 109.6 110.8 112.0 113.1 114.2 115.3 Of which: 2511 Tyres 100.0 93.1 90.0 86.3 79.1 75.9 74.9 72.8 71.4 70.0 68.6 67.2 65.9 64.5 63.2 61.9 60.6 59.3 58.1
Table A1.2: ENUSIM projections at SIC 3,4 –digit consistent with EWP 07 - continued
SIC(92) Description 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020
26 Other Non-metallic Minerals - corrected for glass 100.0 105.8 111.8 111.8 115.2 116.5 118.3 121.0 122.3 123.5 124.6 125.9 127.2 128.5 129.7 130.9 132.2 133.4 134.6 Of which:
261 Glass & Glass Products 100.0 99.3 100.0 96.2 134.2 137.1 140.2 145.8 148.8 151.7 154.4 Of which: 2611,2612 Shaping & Processing of Flat Glass 100.0 93.4 113.4 108.9 116.2 119.7 123.5 130.3 134.3 138.2 142.1 2613 Hollow Glass 100.0 102.4 105.7 108.6 136.7 137.5 138.7 140.5 140.9 141.1 141.1 2614 Glass Fibres 100.0 99.7 99.8 96.6 99.5 100.0 100.6 101.2 101.2 101.0 100.7 2615 Other Glass, Including Technical Glassware 100.0 108.0 108.0 92.5 124.0 127.3 131.2 137.6 141.0 144.2 147.3
262 Refractory & Non-refractory Ceramics (non-construction)100.0 96.3 97.8 100.5 99.0 98.5 98.2 97.4 96.6 95.7 94.7 Of which: 2626 Refractory ceramic products 100.0 100.2 96.7 98.2 98.2 97.6 97.3 96.3 95.4 94.4 93.4 92.5 91.7 90.9 90.0 89.2 88.4 87.6 86.8 262 less Other ceramics 2626 100.0 94.8 98.2 101.4 99.3 98.8 98.6 97.8 97.1 96.2 95.3 94.5 93.8 93.1 92.4 91.7 91.0 90.3 89.6 263 Ceramic Tiles & Flags 100.0 111.7 130.8 119.6 124.4 129.1 134.9 140.4 142.9 145.9 149.0 264 Brick, Tiles & Construction Products in Baked Clay 100.0 93.5 99.4 112.8 94.0 96.7 99.2 103.4 105.6 107.6 109.5 265 Cement, Lime & Plaster 100.0 155.3 128.5 144.7 146.0 149.2 152.7 155.5 157.0 158.4 159.6 266 Articles of Concrete, Plaster & Cement 100.0 103.3 116.2 109.4 111.9 111.6 112.3 113.6 114.1 114.4 114.6 Of which: 2661 Concrete Products (for construction purposes) - - - 2662 Plaster Products ('Gypsum')) 100.0 107.2 122.1 125.9 143.4 152.1 161.3 169.1 173.7 178.1 182.6 187.4 192.4 197.4 202.5 207.7 213.0 218.4 223.8 2663-2666 Ready-mixed Concrete, Mortars, Fibre Cement etc. - - -
268 Other Non-Metallic Mineral Products 100.0 108.9 103.1 105.4 106.4 106.9 107.4 107.6 107.3 106.5 105.5
27 Basic Metals 100.0 100.3 103.2 100.8 109.5 114.3 116.6 118.2 118.5 118.8 119.0 119.2 119.3 119.4 119.5 119.6 119.7 119.7 119.7 Of which: 271-273 Iron & steel 100.0 123.3 125.7 120.5 145.4 156.4 161.1 165.4 166.1 166.7 167.3 167.7 168.1 168.5 168.8 169.1 169.3 169.5 169.7 274 Basic Precious & Non-Ferrous Metals 100.0 87.2 94.9 97.0 98.2 99.9 101.1 101.6 102.0 102.3 102.5 102.8 103.0 103.2 103.4 103.6 103.7 103.8 103.9 Of which: 2742 Aluminium Production 100.0 93.5 97.8 92.4 93.7 96.1 97.4 98.3 99.0 99.4 99.9 100.4 100.8 101.2 101.5 102.0 102.4 102.7 103.0
AEA Energy & Environment
Table A1.2: ENUSIM projections at SIC 3,4 –digit consistent with EWP 07 – continued
SIC(92) Description 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020
29 Machinery & Equipment not elsewhere classified 100.0 101.7 107.4 110.6 111.8 113.8 115.6 116.7 117.1 117.3 117.5 117.6 117.6 117.6 117.6 117.6 117.5 117.4 117.3 Of which: 261 Munitions 100.0 129.0 140.9 150.3 147.0 145.4 143.8 140.8 139.2 137.5 135.9
32 Electronics & Telecommunications Equipment 100.0 87.6 96.8 86.9 93.8 99.3 105.9 115.5 121.4 127.1 133.1 139.4 145.9 152.8 159.9 167.3 175.1 183.2 191.6 Of which: 321 Electronic components (Semiconductors) 100.0 105.2 112.1 103.7 110.7 115.9 123.5 135.6 143.1 150.5 158.3 166.6 175.2 184.2 193.5 203.3 213.6 224.3 235.6 322 Radio/TV transmission & telecommunication 100.0 60.1 61.8 55.2 59.4 63.3 66.8 71.2 73.8 76.1 78.5 80.9 83.4 86.0 88.6 91.3 94.1 96.9 99.7 323 TV, radio, sound & video equipment 100.0 107.1 132.3 114.7 125.9 134.6 145.0 159.4 168.3 177.2 186.6 196.4 206.7 217.6 228.8 240.6 253.1 266.1 279.7
35 Other Transport Equipment 100.0 107.5 117.7 118.4 133.6 143.0 149.7 156.9 159.9 162.7 165.6 168.5 171.5 174.5 177.3 180.2 183.1 186.2 189.3 Of which: 353 Aircraft & Spacecraft 100.0 108.6 118.9 123.0 141.4 152.2 160.5 169.4 173.3 177.2 181.1 185.0 189.0 193.1 197.0 200.9 205.0 209.2 213.4
Table A1.3: Oxford Economics physical output projections revised to be consistent with EWP 07
UK Manufacturing Sector Data/Forecasts - Budget 07_EWP From BERR Proportional revision manufacturing1.00 2.00 0.94 1.00 1.78 1.00 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 Proportional revision GDP1.00 1.22 1.00 0.92 1.10 0.80 0.80 0.80 0.80 0.80 0.80 0.80 0.70 0.70 0.70 0.70
SIC(92) Description 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 codes '000s tonnes 21 Pulp, Paper & Paper Products Of which: 211 Paper only (excludes pulp) 6,217 6,226 6,240 6033.0 5887.2 6041.8 6262.7 6570.5 6658.5 6743.6 6828.4 6914.2 7000.3 7086.4 7169.7 7256.3 7337.7 7420.5 7502.9
26 Other Non-metallic Minerals Of which: 2614 & 2682 Glass fibre (part) & Other non-metallic minerals nec (part)
265 Cement, Lime & Plaster Of which: 2651 Cement - BCA 2651 Cement - OEF 11,088 11,220 11,400 11,563 12,239 12,536 12,792 13,143 13,379 13,577 13,777 13,924 14,070 14,214 14,350 14,480 14,618 14,755 14,891 2652 Lime - BLA 2652 Lime - OEF 1,702 1,819 1,940 1898.4 2176.5 2279.8 2335.1 2403.6 2428.8 2451.3 2474.5 2497.6 2519.7 2541.1 2561.9 2582.1 2601.0 2619.0 2636.1
27 Basic Metals Of which: 271 Crude steel 11,524 13,127 13,763 13,246 16,386 17,586 18,139 18,680 18,748 18,800 18,862 18,918 18,966 19,006 19,039 19,065 19,080 19,087 19,085 Of which: Oxygen converter (Blast Furnace) 8,955 10,630 10,667 10,550 12,354 12,751 13,094 13,304 13,350 13,387 13,430 13,470 13,505 13,533 13,556 13,575 13,586 13,590 13,589 Electric Arc 2,569 2,497 3,096 2696.0 4032.2 4891.5 5108.9 5460.2 5483.1 5498.6 5516.9 5533.3 5547.3 5559.0 5568.6 5576.2 5580.7 5582.6 5582.0
2742 Primary & secondary aluminium 548 550 546 541.7 538.9 552.5 559.1 566.1 567.8 569.9 571.9 573.9 575.8 577.6 579.4 581.0 582.5 583.9 585.1 Of which: (excl. all rolled, extruded and cast products and alumina) Primary 342 343 360 368.2 369.2 379.1 383.5 389.4 391.5 394.1 396.7 399.3 401.9 404.3 406.7 409.0 411.3 413.4 415.5 Secondary 205 207 186 173.5 169.7 173.3 175.6 176.7 176.4 175.8 175.3 174.7 174.1 173.5 172.8 172.1 171.4 170.6 169.8
AEA Energy & Environment
Appendix 2
Fuel price projections
AEA Energy & Environment
Table A2.1: Fuel price projection (with 20% CCL) consistent with EWP projection at ENUSIM disaggregation
Fuel Period1 Period2 Period3 Period4 Period5 Period6 Period7 Period8 Period9 Period10Period11Period12Period13Period14Period15 20% CCL 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 COAL (L) 1.57 1.54 1.50 1.46 1.42 1.43 1.44 1.44 1.45 1.46 1.47 1.48 1.49 1.50 1.51 COAL (M) 2.41 2.35 2.28 2.22 2.15 2.16 2.18 2.19 2.20 2.22 2.23 2.24 2.26 2.27 2.29 COAL (S) 2.99 2.89 2.79 2.68 2.58 2.59 2.61 2.63 2.64 2.66 2.68 2.69 2.71 2.73 2.74 COKE 6.84 6.65 6.46 6.28 6.09 6.13 6.17 6.21 6.25 6.29 6.34 6.38 6.42 6.46 6.50 ELEC (L) 14.56 14.27 13.98 13.69 13.40 13.43 13.46 13.48 13.51 13.54 13.57 13.59 13.62 13.65 13.67 ELEC (M) 17.29 16.92 16.56 16.20 15.83 15.86 15.90 15.93 15.96 15.99 16.02 16.05 16.09 16.12 16.15 ELEC (S) 19.58 19.30 19.01 18.72 18.44 18.48 18.51 18.55 18.59 18.62 18.66 18.70 18.74 18.77 18.81 GAS (FIRM-L) 4.96 4.83 4.70 4.57 4.45 4.37 4.28 4.20 4.12 4.04 4.07 4.10 4.13 4.16 4.19 GAS (FIRM-M) 5.87 5.69 5.51 5.33 5.15 5.06 4.97 4.87 4.78 4.69 4.72 4.76 4.79 4.82 4.86 GAS (FIRM-S) 6.49 6.27 6.06 5.84 5.62 5.52 5.42 5.31 5.21 5.11 5.15 5.18 5.22 5.26 5.30 GAS (INT-L) 4.96 4.83 4.70 4.57 4.45 4.37 4.28 4.20 4.12 4.04 4.07 4.10 4.13 4.16 4.19 GAS (INT-M) 5.87 5.69 5.51 5.33 5.15 5.06 4.97 4.87 4.78 4.69 4.72 4.76 4.79 4.82 4.86 GAS (INT-S) 6.49 6.27 6.06 5.84 5.62 5.52 5.42 5.31 5.21 5.11 5.15 5.18 5.22 5.26 5.30 OIL - HFO 5.99 5.75 5.52 5.28 5.04 4.91 4.79 4.67 4.55 4.43 4.47 4.52 4.56 4.60 4.65 OIL - LFO 8.57 8.28 7.98 7.68 7.38 7.20 7.03 6.85 6.67 6.49 6.55 6.62 6.68 6.75 6.81 OIL - LPG 8.49 8.20 7.90 7.61 7.32 7.14 6.96 6.79 6.61 6.44 6.50 6.56 6.63 6.69 6.76 OTHGASES 2.13 2.45 2.45 2.51 2.58 2.65 2.71 2.73 2.75 2.77 2.79 2.86 2.93 3.01 3.08 STEAM (L) 8.02 7.77 7.53 7.29 7.05 6.92 6.79 6.66 6.54 6.41 6.45 6.50 6.55 6.59 6.64 STEAM (M) 8.02 7.77 7.53 7.29 7.05 6.92 6.79 6.66 6.54 6.41 6.45 6.50 6.55 6.59 6.64 STEAM (S) 8.02 7.77 7.53 7.29 7.05 6.92 6.79 6.66 6.54 6.41 6.45 6.50 6.55 6.59 6.64
AEA Energy & Environment
Table A2.2: Fuel price projection (with full CCL) at ENUSIM disaggregation
Fuel Period1 Period2 Period3 Period4 Period5 Period6 Period7 Period8 Period9 Period10Period11Period12Period13Period14Period15 full CCL 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 COAL (L) 1.91 1.88 1.84 1.80 1.76 1.77 1.78 1.79 1.80 1.81 1.82 1.83 1.84 1.84 1.85 COAL (M) 2.75 2.70 2.63 2.56 2.49 2.51 2.52 2.54 2.55 2.56 2.58 2.59 2.60 2.62 2.63 COAL (S) 3.33 3.24 3.13 3.03 2.92 2.94 2.96 2.97 2.99 3.01 3.02 3.04 3.06 3.07 3.09 COKE 6.84 6.65 6.46 6.28 6.09 6.13 6.17 6.21 6.25 6.29 6.34 6.38 6.42 6.46 6.50 ELEC (L) 15.51 15.22 14.93 14.64 14.35 14.38 14.41 14.43 14.46 14.49 14.51 14.54 14.57 14.60 14.62 ELEC (M) 18.24 17.87 17.51 17.15 16.78 16.81 16.84 16.88 16.91 16.94 16.97 17.00 17.04 17.07 17.10 ELEC (S) 20.54 20.25 19.96 19.67 19.39 19.43 19.46 19.50 19.54 19.57 19.61 19.65 19.69 19.72 19.76 GAS (FIRM-L) 5.29 5.16 5.03 4.90 4.78 4.70 4.62 4.54 4.46 4.38 4.40 4.43 4.46 4.49 4.52 GAS (FIRM-M) 6.21 6.02 5.84 5.67 5.49 5.39 5.30 5.21 5.11 5.02 5.05 5.09 5.12 5.16 5.19 GAS (FIRM-S) 6.83 6.60 6.39 6.17 5.95 5.85 5.75 5.65 5.54 5.44 5.48 5.52 5.55 5.59 5.63 GAS (INT-L) 5.29 5.16 5.03 4.90 4.78 4.70 4.62 4.54 4.46 4.38 4.40 4.43 4.46 4.49 4.52 GAS (INT-M) 6.21 6.02 5.84 5.67 5.49 5.39 5.30 5.21 5.11 5.02 5.05 5.09 5.12 5.16 5.19 GAS (INT-S) 6.83 6.60 6.39 6.17 5.95 5.85 5.75 5.65 5.54 5.44 5.48 5.52 5.55 5.59 5.63 OIL - HFO 5.99 5.75 5.52 5.28 5.04 4.91 4.79 4.67 4.55 4.43 4.47 4.52 4.56 4.60 4.65 OIL - LFO 8.57 8.28 7.98 7.68 7.38 7.20 7.03 6.85 6.67 6.49 6.55 6.62 6.68 6.75 6.81 OIL - LPG 8.65 8.35 8.06 7.76 7.47 7.30 7.12 6.94 6.77 6.59 6.66 6.72 6.78 6.85 6.91 OTHGASES 2.13 2.45 2.45 2.51 2.58 2.65 2.71 2.73 2.75 2.77 2.79 2.86 2.93 3.01 3.08 STEAM (L) 8.48 8.24 7.99 7.75 7.51 7.38 7.25 7.12 7.00 6.87 6.92 6.96 7.01 7.05 7.10 STEAM (M) 8.48 8.24 7.99 7.75 7.51 7.38 7.25 7.12 7.00 6.87 6.92 6.96 7.01 7.05 7.10 STEAM (S) 8.48 8.24 7.99 7.75 7.51 7.38 7.25 7.12 7.00 6.87 6.92 6.96 7.01 7.05 7.10
AEA Energy & Environment
Appendix 3
Relating the MACC annualised costs to the cost of targets using the DCF (NPV) calculation
It is known that the relationship between the marginal costs (£/tCO2 saved) from the MAC curve calculations and the £/tCO2 figure derived from the policy analysis approach of dividing the total NPV of cost and savings by the total cumulative carbon savings differ. This section of the report clarifies why this is the case, and shows how they can be made to correspond. The consequence of this is that the annualised cost results from MACCs could be used directly in calculating the costs of targets or policies by introducing the marginal costs directly into a DCF calculation sheet; without needing to identify individual cost components such as implied capital costs, energy savings, and other cost benefits.
There is however advantage in understanding the cost structures implied in achieving different targets and this probably outweighs the additional effort required to export more detailed cost data from ENUSIM and the BRE models.
The technology analysis approach used in developing MAC curves is based on annualised cost calculations, where the £ per tonne of carbon saved is the annualised ‘current’ net cost divided by the annual carbon saving. Policy or target ‘costing DCF calculations’ are based on a £ per tonne of carbon saving calculated by taking the total net present value (NPV) of cost and savings and dividing by the cumulative carbon saving over the lifetime of policy or target.
Comparing the output of the two calculation methods gives the following differences for an example: £100 capital cost, with a £40 per annum saving, at a discount rate to 15 per cent over 12 years.
The annualised cost per tonne of carbon saved:
Annualised capital cost £18.4553 Savings per annum £40 pa Carbon saving per annum 5 t CO2 pa
Annualised cost per tonne = minus (-) £4.30/tCO2 from (£18.45 minus £40)/5 tCO2
The NPV based cost per tonne of carbon saved:
Total capital cost £100 NPV of savings (12 years at 15%) £216.82 Cumulative carbon savings 60 tCO2
Total cost per tonne = minus (-) £1.947/tCO2 from (£100 minus £216.82)/60 tCO2
The annualised cost per tonne calculation is based on a ‘current value’ annualised cost. If the net annualised saving is discounted over the 12 year period at 15 per cent DCR, and this value is divided by the total cumulative CO2 saving; this also produces a minus (-) £1.947/ tCO2 value.
Cost of per tonne = minus (-) £1.947/ tCO2 from (the NPV of - £21.6 pa over 12 years/ 60 tCO2).
The difference between the annualised cost calculation used in the MAC curve and the calculation based on total NPV divided by cumulative CO2 is simply resolved by the discounting of the future annualised costs 54.
53 From the Excel function PMT(15%, 12 yrs, - £100 ) 54 This approach is used in the technology cost evaluations carried out by OCC for the MAC Curve for heat. In this case however, the annualised capital cost is produced using discounting at a commercial rate (15%), whilst the NPV calculation discounts the annualised total net cost at 3.5%.
This works obviously for a single technology analysis. However, in assessing the cost of future targets, the situation is more complex because of the build-up of the penetration of technologies. It is therefore not so apparent how the marginal annualised £/tCO2 derived from the MAC curve relates to the cost of delivering a particular target trajectory.
The following figure demonstrates the effects. This shows the implied costs of delivering a target, showing the cumulative investment cost, cumulative energy savings and the other savings. It also shows the cumulative net annualised costs derived by the difference between the annualised capital cost and the value of energy and other benefits. All values shown are discounted at 3.5 per cent over the lifetime of the investments (example 20 years) that go to make up the target delivery.
120 0.60
100
80 0.40 Total CUM investment 60 + Annualised capital cost 40 0.20 Energy cost saving
20 + Other cost saving £m pa£m 0 0.00- MtCO2 Net annualised 2006 2008 2010 2012 2014 2016 2018 2020 2022 2024 -20 cost pa Carbon saved -40 -0.20
-60
-80 -0.40
All investment ends in 2025; however, the persistence of the benefits continued to deliver for their total lifetime of 20 years (for later investments out to 2045).
The results of this analysis are shown in the following figure, where a comparison is made between the total £/tCO2 savings methodologies - comparing the annualised cost from MACC calculations and the total NPV based estimates.
The annualised cost derived is minus (-) £118/tCO2, whilst the total NPV based estimate (the green arrow) is minus (-) £138/tCO2. The alternative calculation, which takes the NPV of the total net annualised cost (from the MACC calculations) divided by the total cumulative carbon saving, arrives at the same minus (-) £138/tCO2 value.
AEA Energy & Environment
In essence, if the net annualised cost (which is a ‘current’ value) is discounted over the target period, and then divided by the cumulative carbon savings, the resulting value is the same as the standard total NPV target cost evaluation method.
The consequence of this is that the annualised costs resulting from MACCs could be used directly in calculating the costs of targets or policies by introducing the marginal costs directly into a DCF calculation sheet; without needing to know individual cost components such as implied capital costs, energy savings, and other cost benefits.
It is therefore possible to go directly from the marginal cost data in the MAC curve to a target cost evaluation. This is generally simple if a linear progression of targets is required; however, this can be more complex in those cases where non-linear progress is to be evaluated. Also, it does not reveal the various elements of the cost analysis such as total capital investments, total energy cost savings and other cost savings.
It is also less conceptually clear how the calculation is being carried out.
To make the link between the ENUSIM MAC curve data and the total NPV target cost analysis does, nevertheless, require that both the total capital investment costs and other costs be exported from the ENUSIM model.
There is however, advantage in understanding the cost structures implied in achieving different targets and this probably outweighs the additional effort required to export more detailed cost data from ENUSIM (and the BRE models).
We intend that this will become part of the model updating exercise to be carried out in this project.
Appendix 4
An Overview of ENUSIM
The ENUSIM model simulates the uptake of energy efficiency technologies, given assumptions on the base year energy use, the technical detail of the technologies, projections of fuel prices and projections of growth rates. A more detailed description of ENUSIM is attached at Appendix 5.
Taking Account of Existing Technology Stock
ENUSIM simulates the uptake of cost-effective technologies by assuming that the penetration of each technology follows an ‘S-curve’. As a technology enters the market its initial penetration rate is slow. The penetration rate increases as the market develops and slows as it approaches saturation. The ‘S- curve’ for each technology is specified by the final penetration and the time taken to reach maximum penetration. The user can also specify the point at which the technology can start along this curve, and as long as the technology remains cost-effective, the penetration of the technology will proceed smoothly along the remainder of the curve. For example, in Figure 8.1, Technology 1 has been available for some time and is already fitted to 24% of the devices55 in the base year, whereas Technology 2 is newly available and is not fitted to any device.
As can be seen from Figure 8.1, the level at which the uptake of the technology saturates (i.e. the maximum technology penetration) can differ by technology. In the illustrated example, Technology 1 saturates at 80% and Technology 2 at 50%. The maximum penetration levels and penetration rates can also be adjusted for different scenarios. ENUSIM is set up to handle three scenarios: Business-as-usual (BAU), which assumes that technologies will continue to be taken up at current rates, and that the final saturation point is determined by behavioural preferences in the market, not necessarily determined by the cost-effectiveness of the technology option All-cost-effective (ACE), assumes that technologies are taken up wherever it is cost-effective to do so, without regard for constraints on capital, management time and other barriers to uptake All-technically-possible (ATP), which assumes that the technology is applied wherever it is technically possible to do so without any regard for cost-effectiveness
Figure 8.1 Example S-curves for two technologies
100%
80%
60% Tech 1 Tech 2 40%
20% % Technology Penetration 0% 0 10 20 30 40 Years
55 Device is the term used in ENUSIM to refer to the ultimate end user of energy for example kilns or motors and drives
AEA Energy & Environment
The shape and slope of the curve is used to represent barriers to the uptake of a technology e.g. for a technology which can only be fitted when production stops and equipment is overhauled, the rate of uptake will be slower than for one where the equipment can be brought easily off-line.
Calculation of Cost Curves by ENUSIM
ENUSIM calculates cost curves by taking the difference between the penetration of technologies in any given year under the BAU scenario, and the maximum technology penetration defined for that technology under the ACE scenario. This gap represents the remaining cost-effective potential for that technology. The vertical arrow in the upper part of Figure 8.2 Relationship between technology penetration and the cost curves Figure 8.2 shows this gap at 2005, in the cost curve this gap is represented as a step in the curve (horizontal arrow in the lower part of Figure 8.2). The calculation of this remaining potential does not take into account the amount of time over which the investments could be made cost-effectively. In other words, it should not be assumed that the potential saving implied by the cost-curves can be implemented immediately and still be cost-effective. The ENUSIM model takes this gap and calculates the energy or CO2 saving that would be associated with such an additional uptake of the technology. This energy saving is the basis for the cost curve data.
Figure 8.2 Relationship between technology penetration and the cost curves
90% 80% ACE Max 70% Penetration 60% 50% BAU 40% 30% 20%
% Technology Penetration 10% 0% The gap between BAU and 2000ACE maximum2005 penetration2010 2015 2020 2025 is the basis for the cost curve energy savings Years
Cost curve - % energy saved 2005
6 4 2 0 -20% 5% 10% 15% 20% -4
-6 Pounds per GJ per Pounds -8 -10
Each ‘step’ on the cost curve represents the remaining cost-effective potential for a technology and their sum gives the cumulative potential energy saving for the sector. In calculating the potential for each technology, reduced energy use in that device resulting from the application of more cost effective technologies is taken into account. The value of the potential energy saving from each technology is calculated to provide net cost per unit of energy saved.
ENUSIM also calculates CO2 cost curves by converting the potential energy savings from each technology into potential CO2 savings using standard emission factors. One limitation of the model as it stands is that the emissions factors are assumed to remain constant throughout the run, which particularly for emissions from electricity use is an over-simplification as the mix of electricity generation changes with time. This problem can be overcome by re-running the model each time a cost curve for a different year is required with the relevant emissions factors for that year, but as the focus in this study is direct emissions (from fuel use on site) this was not done.
In general, because the stock of any given technology increases over time, the remaining cost- effective potential for that technology reduces over time. This effect is balanced by the addition of technologies to the curve, which are only expected to become available at a specified time within the model run, since the cost curve for a given year does not include technologies that are not yet available in that year.
Interpretation of the cost curve
The cost curve shows the theoretical potential for savings relative to a given business as usual baseline. However, there are a number of factors that need to be taken into account in interpreting the cost curve: The potential is shown at a given year, but this does not give an indication of the timescales over which that potential can be realised. Technologies with a slow penetration under business as usual will be shown with a large potential at a given year, but the same barriers would still apply and that potential could only be realised over long timescales. No account is taken of any limitations on the availability of capital, even very cost effective technologies may not be implemented because the return on capital may be greater elsewhere or because the focus is on improvement in other core business areas. As far as possible, the costs included for technologies are those that would be considered by companies is determining project viability. However, there are often ‘hidden’ costs involved in these projects that are not well defined or quantified and these are not included in the calculation.
AEA Energy & Environment
Appendix 5
Description of ENUSIM
ENUSIM, the industrial Energy End-Use Simulation Model is a technology based, "bottom up", industrial energy end-use simulation model which models the uptake or retrofit of energy saving and/or fuel switching technologies in industry considering both economic and behavioural factors which affect investment in new technology.
The ENUSIM model comprises two main components:
A Microsoft Access Database system, which holds data on device and technologies and projections on economic activity and fuel prices for 19 industrial sectors. ENUSIM’s simulation engine, which simulates the evolution of energy saving processes and calculates the uptake of energy efficiency measures considering both economic and behavioural factors.
Main features of the ENUSIM model
The ENUSIM model is a ‘bottom-up’ technology based model for simulating the uptake, or retrofit of energy saving and/or fuel saving technologies in industry. The main methodology used by ENUSIM is to calculate future energy consumption E for each industrial sub-sector as the product:
E= UED * SEC where UED is a factor expressing output of a production process/sector (indexed to the base year); and SEC is the average energy required to produce a unit of output (indexed to the base year). The UED is entirely determined off model and its sole influence on energy projection is expressed in the above identity; hence, there are no feedback effects between output and other variables in the model. The SEC is largely determined within the model by the take up of energy-saving technologies within industry.56 The take up of a technology is jointly determined by a net present value evaluation (using information on the cost of alternative technologies and fuel prices) and an index of industry penetration of the technology (represented by a logistics or S-curve). The span and slope of S curve is determined by the cost-effectiveness of the technology (as determined in the net present value evaluation), constrained by a set of imposed 'behavioural' assumptions.
There are three behavioural types in ENUSIM: BAU – Business as Usual; ACE – All Cost Effective and ATP – All Technically Possible. The BAU assumption is intended to characterise a continuation of recent trends, with industry continuing to take up energy-efficient technologies and energy- management procedures in the way it has done in the past. However, there is presumption of no step change in the implementation of energy-saving technologies. The mechanisms (or variables) used in ENUSIM to represent this behaviour are twofold:
The speed with which a technology is taken-up ( ie 'the time taken to reach the 50% level'), and; The extent to which a technology is taken-up (ie 'the maximum penetration of a technology').
These variables are set at very low levels for the 'business as usual' simulation. Generally, for the ACE and ATP simulations, they are set at a much higher level or at the maximum level.
The ACE simulation shows what happen if each sector adopts all available cost-effective management and technical, energy-efficiency measures. In common with all such bottom-up approaches, this
56 The user specifies the initial values for the SEC
scenario places no limits on the overall available management time or capital needed for implementing all the possible measures. It is therefore inherently optimistic. The simulation results are broad indicators and not predictions for the future and are only valid in the context of the key assumptions on which it is based.
Projections are normally performed over a period of between twenty or thirty years, although the user can set this to be any value. Projections are performed year by year.
Sector and Sub-sector Disaggregation
The ENUSIM model disaggregates the industrial sector into a number of sectors, sub-sectors, devices and technologies where the user is able to determine the degree of disaggregation of each sector to suit the particular requirements of the model.
Sectors and sub-sectors are used by ENUSIM to identify where in industry devices are used. Energy saving technologies are then applied or retrofitted to these devices in various combinations in order to save energy.
The way in which ENUSIM disaggregates each sector is illustrated in Figure 8.3. Using the ‘Bricks’ sector as an example, ENUSIM requires that this sector is split into one or more sub-sectors. In this example you will see that Bricks is split into two sub-sectors: 'Flettons’ and 'Non-Flettons’. (Other levels of disaggregation shown in Figure 8.3 are discussed in later sections)
Figure 8.3 Structural Disaggregation of an Industrial Sector
Bricks Sector
Flettons Non-Flettons Subsector
Hoff Kilns Mix and Convey Device
Current Alternate Age Category
HEMs Soft Starters VSDs Technology
AEA Energy & Environment
Output Projection
Output projections are based on projections of tonnes of product, gross output or value added; it is not necessary to have a common measure of output, but one measure will allow comparability of the SEC value across sector and device. These projections are input at the ‘sub-sector’ level.
Devices
ENUSIM further requires that each sub-sector contain at least one Device. Within ENUSIM, devices represent the basic unit around which all calculations relating to the uptake of energy efficient technology options is made. The correct definition of devices within each sub-sector is therefore vital to ensure that the sector is accurately modelled.
In Figure 8.3 above, two devices are defined in the Flettons sub-sector – Hoffmann Kilns and Mixing and Conveying. ENUSIM allows you to define whether the device is a ‘Current’ device or an ‘Alternate’ device. Devices existing in the base year are classified as ‘Current’ (this is the default within ENUSIM). ‘Alternate’ devices are used by ENUSIM to represent an alternative method of producing the same output as the current device. They may not necessarily be available at the start year of the simulation run but at some future user-specified time when they would gradually (and linearly) replace the existing ‘Current’ device over the remaining periods of the model.
Device Throughput
Although output projections are input into ENUSIM at the sub-sector level, as explained above, the demands are actually applied in ENUSIM at the device level and therefore some mechanism is required to translate sub-sector demands down to the device level. In general, the throughput of most devices in a sub-sector will be exactly the same as the sub-sector throughput, but there will be occasions where the device throughput may be different.
Output projections are made for each ‘Device’ in the database. There can be no competition between ‘Devices’ since they all perform separate functions. Consider the example shown in Figure 8.4 below for a sub-sector with an annual throughput of 100 units per year. This sub-sector contains four devices, one for space heating and three related to furnaces. We can use a single device (Device 1 – Space Heating) to describe space heating because we know that space heating is delivered throughout this sub-sector by similar boilers etc. and that the SEC for each of these boilers is similar. We also know that we can apply the same technology options for improving energy efficiency to each of these boilers. If space heating were to be delivered by different boiler types with significantly different SECs and/or each boiler type fitted a different set of energy efficiency technology options then we would have to consider defining more than one device to accurately model space heating (see furnaces below). As it is though, we can define a single device in ENUSIM, and define the throughput of this device as the sub-sector throughput (i.e. 100 units) and calculate the device SEC accordingly to give the correct fuel consumption across the sub-sector.
Figure 8.4 Relationship between Sub-sector Demand and Device Throughput.
Device 1 100 Space Heater
Subsector Throughput Device 2 60 Device 4 30 100 units Furnace A Recyling
Device 3 40 Furnace B
We have also defined two furnaces in this sub-sector, Device 2 (Furnace A) and Device 3 (Furnace B). The reason for defining two furnaces maybe because they use different fuels and/or they have different SECs and/or they may want to retrofit different technology options. Clearly we cannot define both furnaces as having the same throughput as the sub-sector , otherwise the fuel use in the sub- sector will probably be incorrect and so we have to define the percentage of the sub-sector throughput which relates to each furnace type. In this example we say that Furnace A accounts for 60% of the sub-sector throughput, and Furnace B accounts for the remaining 40% of throughput. As a further complication, the example also contains one more device, Device 4 (Recycler), which models the fact that it is possible to recycle 50% of the output from Type A furnaces through a Recycler. As the throughput of Furnace A is 60% of the sub-sector throughput then the throughput of the Recycler will be 30% of the sub-sector throughput.
The Economic Workings of ENUSIM
The ENUSIM engine simulates how each device takes up cost-effective energy efficiency and fuel switching measures for each year of the model time horizon. One of the main features of the simulation engine is the way in which the engine uses behavioural routines to model the take up of these technologies based on the cost-effectiveness of the technology compared to competing technology options.
Technology Combinations
ENUSIM allows up to 7 technology options to be fitted to a device in almost any combination.
The number of technology options for each device can be increased if required, but the time taken to perform an ENUSIM simulation run will increase significantly as the number of technology options increases.
ENUSIM will only fit a technology to a device if it is cost-effective to do so and the mechanism by which it decides if it is cost-effective is critical to the way in which the ENUSIM model works. One of the main features of the ENUSIM model is the way in which these technology options are combined together to represent the different stages of technology uptake.
To illustrate the way in which technology options are combined together consider the example below in Table 1 for a device with four possible technology retrofit options available to the device to improve energy efficiency. (We have used four technology options here simply to keep the example manageable. This example can easily be extended using the same methodology for up to seven or more technology options).
They are ordered according to the annual saving which each technology option can bring to the device. The annual saving calculation is performed at run time and is defined as the annual saving for the user specified normal output of the device. This annual saving is calculated as:
Annual saving = (fuel saving from energy saving + fuel saving from fuel switching) – annualised capital costs – fixed costs – variable costs where fuel saving calculations are scenario specific and are based on period 1 fuel costs, and annualised capital costs are calculated using the user specified discount rates calculated over the user specified investment period.
Technology Energy Capital Fixed Variable Invest Discount Annual Option Saving Cost Cost Cost Period Rate Saving (£k) (£/t) (£/t) (years) Tech 1 10% 1 0.04 0.004 5 25% £1.605k Tech 2 8% 1 0.03 0.003 5 25% £1.445k Tech 3 6% 1 0.02 0.002 5 25% £1.284k Tech 4 4% 1 0.01 0.001 5 25% £1.124k Table 8.1: Technology options available for ‘Device 1’.
AEA Energy & Environment
(Based on a fuel price in period 1 = 1.91 GJ/t, normal output of the device as 100 ktonnes/year and a device SEC of 1 GJ/t)
Once ENUSIM has calculated the annual saving for each technology option, it orders them by the annual saving each technology makes, so that the technology providing the greatest annual saving appears at the top of the list. For ENUSIM this is the most cost-effective technology.
ENUSIM then generates a number of technology combinations in a tree-like structure shown above to represent all the possible technology combinations allowed within the ENUSIM model. The individual technology options are first retrofitted to the base device, starting with the most cost-effective technology, Tech 1, and finishing with the least cost-effective option, Tech 4. The Base + Tech 1 combination is then able to retrofit (in order) Tech 2, Tech 3 and Tech 4, while the Base + Tech 2 combination is able to retrofit (again in order) Tech 3 and Tech 4. The Base + Tech 3 combination can only retrofit the Tech 4 technology option, and the Base + Tech 4 combination has no more technology options to retrofit. This process of adding technology options to existing technology combinations continues until each branch of the combination tree has exhausted all possible technology options.
If it is not technically possible to retrofit two technologies to a device, the user may exclude this combination.
Modelling Behaviour in ENUSIM
As mentioned earlier, the model runs on a year by year basis, carrying forward information from one year to the next. Each fuel price may vary from period to period. As they vary they will impact upon the annual saving for some technology options and thus the payback period. In ENUSIM, this in turn impacts on the extent of final penetration and the speed of penetration.
In general, as fuel prices increase we would expect: The annual savings due to energy saving/fuel switching to increase Payback times to decrease Final extent of penetration to increase Speed of penetration to increase
Similarly as fuel prices decrease we would expect:
The annual savings due to energy saving/fuel switching to decrease Payback times to increase Final extent of penetration to decrease Speed of penetration to decrease
To understand how these relationships are defined within ENUSIM it is first important to understand a little about how ENUSIM models behaviour.
Behavioural sub-routines are considered to be essential for an accurate simulation of the way in which industry reacts when there are cost effective technologies available. The reasons that all cost- effective measures are not taken up immediately are well understood and reported (lack of capital, lack of information, other business objectives, etc.). It is known that cost-effective technologies are taken up with a characteristic “S-curve” shape. There are three distinct stages of the S-curve:-
when a technology first becomes cost-effective, industry will only install it when they know about it (i.e. it has been marketed) and when they are convinced that it is cost-effective and practical (i.e. there are case studies in other factories proving the cost-effectiveness of the technology). Both of these processes take time, and so the initial part of the S-curve shows low take-up rates (less than 4% after 0.2 of total time) but at an increasing rate with time;
once marketing and case studies have been completed, a rapid take-up technologies occurs, illustrated in the central part of the S-curve (technology penetration is 90% of the maximum after 71% of the time);
the final part of the S-curve shows a slowing down of take-up rates, due to the fact that the market for the device is becoming saturated due to the installations which occurred in previous
years. The total penetration of the technology approaches a maximum asymptotically (there will always be certain customers who are unaware of the opportunity and/or who remain unconvinced of the economic benefits and/or who have no capital to invest, etc.).
In ENUSIM the shape of the S-curve is contained in the table NORMDIST. The S-curve is defined by 601 points (i_values) and is normalised to one illustrated in Figure 5.Figure 8.1
Figure 8.5: ENUSIM S-curve
ENUSIM S-Curve
1 0.9 0.8
0.7 0.6 0.5
s_value 0.4 0.3
0.2 0.1 0 0 100 200 300 400 500 600 i_value
The shape of the S-curve for a given technology in a given industry sub-sector will vary from one technology to another. ENUSIM models these variations by parametising the Standardised S-curve as follows:-
maximum penetration of a given technology (fraction); time to maximum penetration time (years), i.e. time taken to get to the point maximum penetration has occurred;
The user can also specify the point at which the technology can start along this curve, and as long as the technology remains cost-effective, the penetration of the technology will proceed smoothly along the remainder of the curve.
AEA Energy & Environment
Appendix 6
ENUSIM File Structure and Process Steps
File structure
The main Enusim database file is held in the root directory (ENUSIM 2002 new)
The other Enusim files are organised into 3 main folders, the access folder, the COSTCURV folder and the output folder as shown below.
access folder
The Access folder is the where the input data needed to run the model are stored. The Access folder contains 3 subfolders called Enusim input, Excel and Mod02.
The Enusim input folder contains two excel spreadsheets which in turn contain information on fuel prices and demand projections for all sectors. Any changes to the fuel prices should be saved on the excel spreadsheet called “Energy Price Relationships_CENTRAL price scenario.xls”. The spreadsheet is linked to fuel prices spreadsheets relevant for each sectors and therefore the changes will propagate to all the sectors.
The Excel folder contains 22 subfolders with input data (fuel prices and demand projcetions) for all the sectors run by Enusim. These are not hard-linked directly to the model (but are linked to the sheets in the Enusim input folder as mentioned above). The model contains a function to import new price/demand data based on the sheets in the Excel folder. The energy subfolder is empty as in this project only energy used is considered and not energy produced.
The mod02 folder contains individual databases for each sector necessary to run Enusim. When the main Enusim database file (in the ENUSIM 2002 new folder) is opened and a sector is selected the main database becomes linked to the specific sector database in the mod02 folder.
There are two ways to update the sector database files:
1. From Enusim. Open Enusim, select the sector to update go to Edit select the table to update (i.e. fuel prices), copy-paste the link to the relevant excel table (from the sector specific sub- folder in the Excel folder) with the updated data where prompted.
2. Directly open the .dat files and make the necessary changes, for example copy-paste a new set of fuel prices in the Fuel Scenario Table.
All changes to the numbers made by the front window of Enusim (fuel prices, technology data, etc) will be automatically saved in the correspondent sector .dat file and the previous data will be lost. Direct changes to the dat files are also permanent as the information is overwritten. It is advisable to have backup copies of the dat files before making changes.
As illustrated in the figure there are two sub-folders in the mod02 folder. These contain identical copies of the sector database with the exception of the fuel prices (i.e. the 100% CCL folder contains files with fuel prices reflecting 100% CCL levels). This is simply to save time – as the alternative is to import updated fuel prices into a single set of sector database files each time you want to change them.
COSTCURV folder
AEA Energy & Environment
The COSTCURV folder contains output Excel sheets from Enusim. All cost curves output data will be stored in the respective sector subfolder and will be overwritten after each run. For this reason it is advisable to save the data soon after the end of the run in a separate folder to avoid data losses. For this purpose the separate Cost curve results folder has been created.
Output folder
The output folder is where the output data from the Normal Enusim Run (i.e. projections as opposed to cost curve data) is created and stored. For the purpose of this project the output folder is empty (whilst part of the cost curve function does require Enusim to undertake a Normal BaU projection run, the data is not output as part of this).
ENUSIM Process Steps
Step 0: Before running Enusim for the first time two changes need to be made to ensure that the model looks at the appropriate directory locations – dependent on where the whole of the model is saved. Open the main Enusim database, close the main form and:
1) go to the “Tables” window (on the left of the screen)and open the System table. Paste the path to the folder where the mod02.dat files are stored (chose 100% CCL or 20% CL) in the last column in this table. Make sure that the end of the path has a backward slash \.
2) go to the “Modules” window and open Global Constants,
Change the line Global Const DB_pathname As String = "XXXX” such that XXXX points to the location on your computer. All output files will be passed to the relevant subfolders (e.g. COSTCURV) at this location. .
Step 1: Run ENUSIM. Select any sector from the main menu and wait for it to load. From the top menu bar select “Run Enusim” -> “Cost Curve”. Enter the required discount rate. Click “Select Sectors”, tick those required and “Save Selection”. Finally click “Run Selected Sectors”57. The time for a run varies between a few minutes for the smaller sectors to over half an hour for the most fragmented sectors (e.g. chemicals and food&drink).
Step 2: Make sure the COSTCURV folders contain the excel spreadsheet relevant to your run.
Step 3: Import cost curve data into supporting workbook (named “CCC workbook to aggregate Enusim outputs.xls”) by clicking on the button “Import Enusim Data” on the Config tab. CELL D1 will need to be changed to point to the relevant COSTCURV folder path. N.B. make sure the main output spreadsheet (named “CCC Industry MACC output sheet - X% CCL.xls –vX- X”) is not open before importing otherwise the macro in “CCC workbook to aggregate Enusim outputs.xls” will not work.
Step 4: Copy “B” tab from “CCC workbook to aggregate Enusim outputs.xls” into “B” tab in the main output workbook named “CCC Industry MACC output sheet - X% CCL.xls –vX-X”..
Step 5: To change variables and obtain different sets of cost curves, go to the “Variable” tab and make the changes. All the changes will affect the ‘output for CCC tab’. Refer to spreadsheet description for more information.
57 N.B. clicking “Run Enusim” from this menu only runs the sector selected from the main menu. In addition, from the top menu bar (after a sector has been loaded), selecting “Run Enusim” -> “Run All Sectors” peforms a normal run across all sectors (i.e. a projections run as opposed to a cost-curve run).
AEA Energy & Environment
Appendix 7
ENUSIM based bottom-up modelling approach – methodology linking MAC curves to target or policy costs
This note describes how ENUSIM output can be used to generate targets or policy cost information, assuming a required delivery of CO2 reduction, and an assumed coverage of policies. The approach was used in two studies during 2005 and 2006:
1. A Carbon Trust (CT) study of the modelling of business sector Climate Change Programme options. The aim was to gain an understanding of the relative effectiveness and the impact of existing CCP policies, and to suggest how the policy ‘architecture’ should evolve over time to meet any existing policy gaps and to meet post-2010 obligations.
2. An extension of the CT analysis for Defra, for the UK Innovation Review, to cost revised policy options.
The Defra ENUSIM model is used as the abatement MAC curve generator in the ‘bottom-up’ targets or policy cost model. Outputs from the cost curves are used in a ‘DCF-based’ costing spreadsheet for each industrial and services sub-sector. For the Innovation Review, a DCF of 3.5% was used.
The ENUSIM Model
ENUSIM is a technology penetration simulation model, in which a view is taken about the likely BAU penetration58 of energy saving technologies. This is represented by a series of ‘S’ curves for the different technologies. The ‘with policy’ penetration59 level will, in general, be higher.
120%
100%
80% B A U level
60% 'with policy' penetration level 40%
20% Technology penetration level 0% 0 200 400 600 800 Tim e
The ENUSIM emissions MAC curves for each year show the total remaining technical potential for each individual measure from the position reached by BAU delivery of technologies to that date.
Changes in the abatement curve from 2005 to 2010 and to 2020, includes BAU uptake of existing technologies, sector growth factors, and the introduction of new technologies at future dates.
58 This BAU penetration is assuming ‘with existing policies’ 59 The ‘with policy’ penetration assumes new policies.
0.5
0.4
0.3
0.2 2005 CSC 2010 CSC
0.1 2020 CSC Cost (£k/tCO2) 0 0 5,000 10,000 15,000 20,000 25,000 -0.1
-0.2 Cum CO2 (ktCO2)
ENUSIM MACCs are based on ‘with existing policy’ delivery (BAU). We need to compare the BERR UEP ‘with policy’ projections against the BAU baseline.
Example: This is shown below for a sector (chemicals) Climate Change Agreement policy (CCAs).
130
120
110 BAU intensity
100 CCA targets
Index 90 Grow th factors 80 Additional policy (2000=100) delivery at 2010 70 over BAU projection
60
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
The BAU MACCs need to be modified to take account of ‘additionality’ from revised targets or policies. We achieve this by assuming that ‘all cost effective’ (CE) technologies will increase their penetration, proportionately, to meet the ‘additionality’ requirement level. This has the effect of removing more of the existing CE technologies from the MACC technical potential.
0.3 Additional policy delivery at 2010 over BAU 0.2 projection
BAU 2010 Revised CSC 0.1
ACE BAU £k/tCO2 Revised ACE
0 0 5000 10000 15000 20000
-0.1 CUM CO2 (ktCO2)
AEA Energy & Environment
ENUSIM assumes that if a technology is cost effective it will be taken up, but according to different penetration rates depending upon the types of measure (level of interest, strategic importance etc.). In accounting for the additional level of technology penetration to meet a target delivery, it is assumed that all CE technologies will have increased take up, by the same ratio, to meet the ‘additionality’ requirement.
The use of MAC curves to establish target or policy delivery costs, 2005-2010 and 2010-202060
Between 2005 and 2010 we match the absolute CO2 delivery requirement from targets, with the adjusted MACC associated with 2005. The level of delivery along the x-axis of the MACC (to identify marginal cost) is known as the ‘efficiency change effect’, calculated as the difference between the growth adjusted 2005 absolute emission level (at 2010), and the actual absolute policy delivery level at 2010, taken from UEP (which takes account of growth)61.
The change in absolute emissions level required by the target at 2010, relative to 2005, is made up of two components: a growth effect and an efficiency (intensity) effect. The growth effect is defined as the emission level at 2010 without efficiency improvements = emission at 2005 x growth index. The efficiency effect = absolute emission requirement at 2010 – (minus) the growth effect.
0.1 Absolute policy delivery at 2010 minus grow th effect on 2005 emissions Revised CSC 0
£k/tCO2 0 5000 10000 15000 Marginal cost of abatement
-0.1 CUM CO2 (ktCO2)
Between 2010 and 2020, the calculation is made in the same manner, using the adjusted 2010 MACC and the intensity change effect 2010-2020.
Note: For the 2005-2010 delivery analysis, we still needed to adjust the 2005 MACC to allow for policy delivery between 2000 (start of the CCP) and 2005.
As more demanding policies force technology delivery above the ‘net cost = zero’ level (above the level of ‘all cost effective’), we need to alter the revised MACC to also include increased uptake of some non-CE measures. We adopt a similar procedure as described previously. However in this case, for the 2010 curve, we assume that all technologies up to the target delivery level required in the previous period, 2005-2010, increase their penetration. In the 2010 curve, this attenuates their future contribution, again all by the same ratio, to meet the target ‘additionality’ required.
60 The example date periods used here were used in the original Innovation Review study. In the current target definition work the target dates will be 2007-2012, 2012-2017 and 2017-2022. 61 The efficiency effect can also include the effects of the restructuring of a sector i.e. plant closures and new openings, rationalisation etc. Using only the bottom-up technology analysis implicit in ENUSIM ignores these effects.
0.3
0.2 Additional policy delivery at 2010 over BAU Policy 0.1 projection delivery BAU 2010 £k/tCO2 marginal Revised CSC cost 0 0 5000 10000 15000 20000
-0.1 CUM CO2 (ktCO2)
If additional policy delivery demands the take-up of technologies, which are above the net cost = zero level, we assume that the penetration of these, and all of the cost effective measures, increases by the same proportion. The marginal cost of the policy delivery in 2005-2010, defines the upper limit of net costs for those technologies which have their uptake increased for the 2010 curve.
The delivery and price calculation is as described previously. However, with fuel price changes, an appropriate MACC is used and the BAU, against which policy delivery is assessed, also changes 62. This is shown in the figure as the additional abatement step required, through baseline adjustment. The MACC is that appropriate to the fuel prices in the scenario being modelled.
0.3
Additional abatement 0.2 requirement Absolutel policy through delivery at 2010 baseline minus grow th adjustment ef f ect
0.1 Revised CSC £k/tC O2
0 Marginal 0 5000 10000 15000cost of 20000 abatement
-0.1
ENUSIM allows for the introduction of new technologies not available at the start of the ENUSIM base year; a negligible number are introduced before 2005 and a few at 2015 and 2020. These technologies are difficult to separate out explicitly, but are implicit in the 2005, 2010 and 2020 MACCs described previously. Brick making in particular, demonstrates the effect, with a growth in the MACC potential from 2005 to 2010 due to new technologies becoming available.
62 The ‘with existing policy’ projection varies with changes in the fuel price projections according to BERR UEP modelling.
AEA Energy & Environment
0.2
0.15
0.1
0.05 Bricks 2005 0 Bricks 2010 £k/tCO2 0 50 100 150 200 250 -0.05 Bricks 2020
-0.1
-0.15
-0.2 CUM CO2 (ktCO2)
Aggregating across all sectors, the trend using the (then at 2005) current ENUSIM technical data, is for a reducing ‘all technical’ potential with time, as shown below. For some sectors, particularly minerals and metal, the new technology effect is more evident than in others such as food and engineering.
2005 1 2010 2020
0 5,000 10,000 15,000 20,000 25,000 All technical potential (ktCO2)
Fuel price effects on the MACCs
The MACCs shown next are for total industry at 2020 and identify the differences between a ’high CCL, high EUETS price’ (HH) scenario and the ‘low low’ (LL) equivalent, for a 20% CCL level and for the corresponding 100% CCL level. These are compared with the no CCL, no EUETS case. The effect of a move from a 20% to 100% CCL is the most significant 63. At 2005 and 2010, curves are not sensitive to HH, LL differences.
Changes in fuel prices in ENUSIM will affect the cost effecitiveness and penetration rate of the measures. However it is possible to change fuel prices without re-running ENUSIM in the aggregate spreadsheet. This option will allow to see the effects on the curves of with and without taxes scenarios but the results will be crude as the changes to the fuel prices on the spreadsheet will be only reflected on the costs effectiveness of the measures and not their penetration rates.
63 This would be equivalent to sites and sectors with CCA agreements and those without.
1
0.8
0.6 HH LL 0.4 100%HH
100%LL £k/tCO2 0.2 noCCL,no EUETS
0 0 5000 10000 15000 -0.2 CUM CO2 (ktCO2)
The assessment of total target or policy costs
This part of the analysis uses a DCF cost sheet, which reads data directly from the MACC data sheets 64. Example information read directly from the MACC sheets (for 2005-2010 and 2010-2020) is shown below.
to 2010 2010-2020 Prog. life 16 years years Addit saving pa 0.012 MtC pa 2010 0.023 MtC pa 2020 Life of saving 26 years years Total investment 17.0 £m by 2010. 24.1 £m by 2020. Energy cost saving 3.9 £m by 2010. 4.7 £m by 2020. Other cost saving 3.4 £m by 2010. 4.7 £m by 2020.
The policies are analysed to 2020 from 2005 (16 years). The life of the savings are taken over a 26- year period; out to 2030. A 3.5% DCR is used; and a linear delivery of benefits and cost growth is assumed over the periods 2005-2010 and 2010-2020.65
64 For Carbon Trust and Innovation Review work a policy analysis spreadsheet was developed, which contained all of the cost supply curve information by sector and scenario. 65 The additional savings per annum at 2010 and 2020 is the value described previously as the absolute policy delivery (from UEP) minus the growth effect. It is therefore not the policy ‘additionality’ over BAU, which can only be provided at a more aggregated sector level at present. The additional savings value is an efficiency improvement value, which needs to be compared with the BAU efficiency improvements to derive policy additionality. Total investment, energy cost savings and other cost saving are read automatically from the MACC data. Policy administration costs are not included in the sector level analysis; they are added as a total aggregated cost.
AEA Energy & Environment
Targets and policy costing outputs
The policy analysis sheet identifies costs to individual sub-sectors and as such, differentiates the cost burdens from targets or policies. The next table is taken from the output sheets as an example of the data currently available 66.
Lifetime net cost benefit -154.0 £M Net cost benefit to 2010 -0.6 £M
Policy cost effectiveness (lifetime savings) na £/tCO2 Overall cost effectiveness (lifetime) -67 £/tCO2 Overall cost effectiveness (to 2010) -5 £/tCO2
Barrier cost estimate 14 £/tCO2
Marginal cost - next measure at 2010 -78 £/tCO2 Remaining cost effective potential at 2010 0.029 MtC pa Remaining technical potential at 2010 0.040 MtC pa
66 The lifetime benefit is currently taken up to 2030. The barrier cost estimate shown is a ‘notional value’; it was removed for the final analysis and report.
AEA Energy & Environment
Appendix 8
Hidden and missing costs
This appendix provides the hidden and missing costs used in the MACCs, sourced from Enviros (2006).
Industry
High Energy Low Energy Intensive Industry Intensive Industry Units
Hidden Cost Split of Measure Low High Low High Behavioural 3 5 6 10 Hrs Engineering Measures 3 5 6 10 Hrs Project Non-engineering identification measures 3 5 6 10 Hrs No cost 0 0 0 0 Hrs Project Low cost 20 40 10 30 Hrs appraisal High cost 50 100 40 80 Hrs No Cost 1 5 1 5 Of capital costs Project Low Cost 3 6 3 6 Of capital costs commissioning High cost 1 5 1 5 Of capital costs Behavioural 0% 0% 0% 0% Of capital costs Production Engineering Measures 0% 5% 0% 5% Of capital costs Disruption Other Measures 0% 5% 0% 5% Of capital costs Behavioural 5% 10% 0% 5% Of capital costs Additional Engineering Measures 5% 10% 0% 5% Of capital costs Engineering Other Measures 5% 10% 0% 5% Of capital costs Behavioural -10% -20% -10% -20% Of energy savings Engineering Measures -5% -10% -5% -15% Of energy savings Risk of Delivery Other Measures -10% 20% 10% 20% Of energy savings Behavioural 6 12 6 12 hrs/yr Ongoing Engineering Measures 6 12 6 12 hrs/yr Management Other Measures 12 60 12 60 hrs/yr
Domestic sector
Owner Social/Local Occupied Rented Authority Units Hidden Cost Split of Measure Low High Low High Low High Behavioural 0.5 2 0.5 2 0.5 0.5 Hrs Engineering Measures 0.5 2 0.5 2 0.5 0.5 Hrs Project Non-engineering identification measures 0.5 2 0.5 2 0.5 0.5 Hrs Project No cost 0 0 0 0 0 0 Hrs appraisal Low cost 1 3 1 3 1 2 Hrs
High cost 3 6 3 6 2 5 Hrs No Cost 0 0 0 0 0 0 Hrs Project Low Cost 1 2 2 4 1 2 Hrs commissioning High cost 5 10 10 15 5 10 Hrs Of capital Behavioural 0% 0% 0% 0% 0% 0% costs Of capital Engineering Measures 0% 0% 0% 0% 0% 0% costs Production Non-engineering Of capital Disruption measures 0% 0% 0% 0% 0% 0% costs Of capital Behavioural 0% 0% 0% 0% 0% 0% costs Of capital Engineering Measures 5% 10% 5% 10% 5% 10% costs Additional Non-engineering Of capital Engineering measures 5% 10% 5% 10% 5% 10% costs Of energy Behavioural 5% 10% 5% 10% 5% 10% savings Of energy Engineering Measures 10% 20% 20% 40% 15% 30% savings Non-engineering Of energy Risk of Delivery measures 15% 30% 25% 50% 20% 40% savings Behavioural 0.5 1 0.5 1 0.5 1 hrs/yr Engineering Measures 0 0 0 0 0 0 hrs/yr Ongoing Non-engineering Management measures 0 0 0 0 0 0 hrs/yr
Non-domestic sector
Private Sector Public Sector Units
Hidden Cost Split of Measure Low High Low High Behavioural 0.5 1 0.5 2 Hrs Project Engineering Measures 0.5 2 0.5 2 Hrs identification Non-engineering measures 0.5 2 0.5 2 Hrs No cost 0.5 1 2.5 5 Hrs Project Low cost 2 5 5 10 Hrs appraisal High cost 2 5 5 10 Hrs No Cost 0.5 1 0.5 1 Hrs Project Low Cost 1 5 1 5 Hrs commissioning High cost 1 5 1 5 Hrs Behavioural 0% 0% 0% 0% Of capital costs Production Engineering Measures 1% 3% 1% 3% Of capital costs Disruption Non-engineering measures 1% 3% 1% 3% Of capital costs Behavioural 0% 0% 0% 0% Of capital costs Additional Engineering Measures 1% 5% 1% 5% Of capital costs Engineering Non-engineering measures 1% 5% 1% 5% Of capital costs Behavioural 10% 20% 10% 20% Of energy savings Engineering Measures 5% 10% 5% 10% Of energy savings Risk of Delivery Non-engineering measures 10% 20% 10% 20% Of energy savings Behavioural 6 12 6 12 hrs/yr Ongoing Engineering Measures 3 3 3 6 hrs/yr Management Non-engineering measures 0 0 0 5 hrs/yr
AEA Energy & Environment
Appendix 9
Industry consultation for the update of
ENUSIM: results and conclusions
Introduction
The consultation was carried out over a short period so concentrated only on the main inputs. Consultees were provided with the following information:
o Cost effective potential savings of emissions for the whole sectors for the years 2012, 2017 and 2022. o Cost of technologies o Current penetration rates and theoretical potential of penetration of technologies
The consultation was limited to only three industrial sectors: o Chemicals o Food and Drinks o Engineering (vehicles manufacture, electrical engineering and mechanical engineering)
The following stakeholders were invited to consult on the ENUSIM outputs: o Chemicals: Chemical Industry Association (CIA), Nick Sturgeon o F&D: Food and Drink Federation (FDF), Stephen Reeson o Engineering: Society of Motor Manufacturers and Traders Limited (SMMT) and the Engineering Employers Federation (EEF) for general engineering.
Consultation
Food and Drinks
The consultation with the FDF has resulted in the following conclusions:
o The total emissions for the sector in the base year output by Enusim are higher than the FDF estimates. The FDF do not represent the whole sector as classified by BERR so this gives a slight difference but the bigger difference is due to the emission factor used for steam in Enusim. The emission factor is too high in the food and drink sector. o Given the short timescales of the consultation no specific feedback was given for the technologies. o Over the period 1999 – 2006 FDF Climate Change Agreement data for Milestone 3 shows energy efficiency improved by 13.3%, which is higher than the model’s10%, and therefore effectively the sector should have no cost effective saving potential left. However it was recognised that the CCA improvement is due to a number of factors such as throughput/product mix changes and use of low carbon technologies not covered by Enusim (e.g. renewables, biomass etc.) although a large part has been delivered by energy efficiency measures as covered by the model.
o The FDF state that their view is that the cost effective potential is somewhere between 0-10%, probably about 5%. The ENUSIM estimate of 10% is at the top end of the range, which would be expected as many of the penetration rates for the technologies are set at 0.
Chemicals
o In a similar way to FDF, the emissions in the base year were too high due to the steam emission factor. o There were some specific, and critical, comments about the data on technologies and subsectors. However, because of the short timescales of the consultation no concrete suggestions to improve the data were made. This is also because the structure of the chemicals sector is complex and the categories in ENUSIM necessarily broad. o Under the CCAs, energy efficiency improvement of 35% over 1990-2006 have been made and it is argued that no more very cost effective technologies will be available. This is unlikely to be true universally across the industry but the potential is almost certainly more limited than ENUSIM implies. However, as mentioned previously there is no concrete information to update ENUSIM input. o CHP potential data was also supplied to the CIA; however, no feedback has been received. o The CIA asks that more resources be put into ensuring the data are correct.
Engineering
A meeting organised by the SMMT with BMW Mini Plant in Oxford resulted in the update of several technologies in terms of penetration rates, savings and costs. The short timescales meant SMMT only had limited opportunities to discuss the model with members, however prior to the meeting they provided a summary note for discussion, which made the overall points:
o The industry has made significant progress in delivering both relative and absolute emissions reductions, evident in their CCA and EU ETS performance. o Members had delivered many of the measures listed and their experience indicates smaller potential savings than those in the model.
Discussion of the issues at the BMW plant resulted in the following changes: o A number of the technologies were removed o The energy balance was adjusted to allow for a greater proportion in the paint shop, based on the information provided by BMW and further advice from SMMT. o The savings and costs were revised for some of the technologies, based on reports by BMW and experience by SMMT members. Penetration rates were adjusted to reflect many plants had already invested in some areas negating some of the potential savings.
The overall saving potential was considered on the high end although within the magnitude expected. The modification of the individual parameters within ENUSIM is provided below. Data supplied to the EEF and a follow up telephone discussion did not produce sufficient comment for extensive updates to be made. EEF have pointed out that significantly more effort than they currently have available would be needed to rigorously assess the current ENUSIM data. Because of this we have not been able to make any major update to the mechanical or electrical engineering sectors in ENUSIM, other than for some of the electricity savings technologies, as described below.
Enusim updates
In practice the discussion with the stakeholders resulted in the modification of a few parameters within Enusim. These were:
AEA Energy & Environment
Food and drink: steam emission factor Chemicals: steam emission factor and the parameters in the table
Sector Device Technology Parameter changed Ammonia Process Heating New plant Capital cost Control of electrode Chloro-alkali Electrolysis gap Energy saving Chloro-alkali Pumping HEMs Current penetration Chloro-alkali Electrolysis Energy management Current penetration General Organics Compressors HEMs Current penetration
Engineering: steam emission factor and the parameters in the table
Sector Device Technology Parameter changed Energy saving + current Land Transport Compressed Air Control sy penetration Land Transport Compressed Air New plant Energy saving Land Transport Compressed Air O&M Improv Energy saving Land Transport Forming/welding/machining New plant Energy saving Process Land Transport Forming/welding/machining improvement Energy saving Controls & Land Transport Furnaces/Heat Treat insulation Energy saving Energy saving + current Land Transport Furnaces/Heat Treat Heat recovery penetration + cost Land Transport Furnaces/Heat Treat House and maint Energy saving Land Transport Furnaces/Heat Treat new plant Energy saving Land Transport IT New plant Energy saving Land Transport Lighting Controls Energy saving Convert to high effy Land Transport Lighting luminaire Current penetration Land Transport Other aux/pumping VSDs Current penetration Land Transport Painting/curing House and maint Energy saving Energy saving + current Land Transport Space Heating Boiler Controls penetration Controls & BMS & Energy saving + current Land Transport Space Heating insulation penetration Land Transport Space Heating House and Maint Energy saving Land Transport Space Heating New boiler plant Energy saving Other Transport Compressed Air O&M Improv Energy saving Convert to high effy Other Transport Lighting luminaire Current penetration Other Transport Plating/pickling House and maint Energy saving Other Transport Space Heating Central to decentral Current penetration Controls & BMS & Other Transport Space Heating insulation Energy saving Other Transport Space Heating House and Maint Energy saving Other Transport Space Heating New plant Energy saving
In addition values for Variable Speed Drives (VSD) and High Efficiency Motors (HEM) for the mechanical and electrical engineering sectors were also updated to reflect the changes highlighted in the vehicle engineering sector.
Summary
. The tight deadline for the consultation meant that there was limited input from stakeholders despite their willingness to engage in the process. A detailed revision of all the devices and technologies was not possible.. . The emissions savings from the revised model are slightly lower in all sectors. . All stakeholders stated a willingness to help if there was scope for further consultation.
The Gemini Building Fermi Avenue Harwell International Business Centre Didcot Oxfordshire OX11 0QR
Tel: 0845 345 3302 Fax: 0870 190 6318
E-mail: [email protected] www.aea-energy-and-environment.co.uk
AEA Energy & Environment