Wake Up Reforming EU ETS - FTI Consulting · 6.2 Impact of the combinations of options on ETS...

2 February 2017

Wake up!: Reforming the EU

Emission Trading Scheme

WAKE UP!: REFORMING THE EU EMISSION TRADING SCHEME

Copyright © FTI Consulting, Inc., 2017 i

Contents

Contents ........................................................................................................................................................... i

List of figures ................................................................................................................................................ iii

List of tables .................................................................................................................................................. iv

Abbreviations and technical units .............................................................................................................. v

Preface ........................................................................................................................................................... vi

Executive summary ................................................................................................................................. 1 1.

1.1 Context and objectives of the study ........................................................................................... 1

1.2 The European Commission’s proposed reform is not ambitious enough to address the EU ETS structural issues .................................................................................................................. 1

1.3 Fixing the EU ETS: Six alternative options for reform have been modelled and several combinations of reforms could address the ETS issues ........................................................... 2

Context and objectives of the study .................................................................................................... 5 2.

2.1 Context ........................................................................................................................................ 5

2.2 Objectives of the study ............................................................................................................... 7

2.3 Modelling framework .................................................................................................................. 7

2.4 Outline of the report .................................................................................................................... 8

Diagnosis: Key issues with the current EC proposal for EU ETS reform ....................................... 9 3.

3.1 EC proposals for amending the EU ETS ..................................................................................... 9

3.2 The current emissions trajectory is not in line with the objective of limiting global warming to +1.5°C .......................................................................................................................................... 9

3.3 EU ETS carbon price is too low to provide adequate short-term economic signal for carbon abatement .................................................................................................................................. 11

3.4 EU ETS carbon price is too low to provide adequate long-term economic signal for carbon abatement ................................................................................................................................. 12

3.5 ETS market is a residual market for abatement with vulnerability to market and policy shocks ....................................................................................................................................... 14

3.6 The reform needs to be more ambitious if it is to tackle the different ETS issues and rebuild the credibility of the ETS ........................................................................................................... 17

Fixing the EU ETS: Possible approaches for reform ........................................................................ 19 4.

4.1 Higher LRF and Rebasing .......................................................................................................... 19

4.2 Voluntary allowance cancellation ............................................................................................ 20

4.3 Adjustment for overlapping policies ........................................................................................ 20

4.4 Price corridor ............................................................................................................................ 21

4.5 Stronger Market Stability Reserve ........................................................................................... 21

4.6 Modelling approach for the different options ........................................................................... 22

Assessment of the different options for reform of the ETS.......................................................... 23 5.


Copyright © FTI Consulting, Inc., 2017 ii

5.1 Preliminary assessment of options .......................................................................................... 23

5.2 Only Higher LRF and Rebasing options lead to emissions in line with the EU long-term climate targets .......................................................................................................................... 24

5.3 Rebasing, Price corridor and Stronger MSR options support short-term carbon prices for decarbonisation ........................................................................................................................ 25

5.4 All options but Voluntary cancellation affect carbon intensive plants margins and life extension decisions ...................................................................................................................26

5.5 Auction Revenue ....................................................................................................................... 27

Assessment of the combinations of options for reform of the ETS ............................................ 28 6.

6.1 Impact of the combinations of options on short and long-term emission levels ....................29

6.2 Impact of the combinations of options on ETS prices in the short and long term ................. 30

6.3 All options prevent coal plant life extensions and lock-in of emissions in the long term ....... 31

6.4 Auction Revenue ....................................................................................................................... 31

Conclusion and policy recommendations ......................................................................................... 33 7.

Appendix A – Detailed results ............................................................................................................ 34 8.

8.1 Comparison of options.............................................................................................................. 34

8.2 Comparison of combination of options ....................................................................................36

Appendix B – Modelling framework .................................................................................................. 38 9.

9.1 ETS carbon model .....................................................................................................................38

9.2 Supply ........................................................................................................................................39

9.3 Emissions ..................................................................................................................................39

9.4 Banking ..................................................................................................................................... 40

9.5 Market equilibrium .................................................................................................................... 42

9.6 Power market dispatch ............................................................................................................. 42

9.7 Key assumptions .......................................................................................................................44


Copyright © FTI Consulting, Inc., 2017 iii

List of figures

Figure 2-1: EU ETS carbon model ........................................................................................................................... 8 Figure 3-1: EU ETS emissions, 2015 – 2040, Baseline ........................................................................................ 10 Figure 3-2:EU ETS carbon price (real 2015), 2015 – 2040, Baseline ................................................................. 11 Figure 3-3: Coal and lignite capacity outlook, 2015 – 2040, Baseline ................................................................13 Figure 3-4: European utilisation factor, 2015 – 2040, Baseline ......................................................................... 14 Figure 3-5: Allowance cap reduction over 2021 - 2030 .......................................................................................15 Figure 3-6: Surplus and MSR, 2015 – 2030, Baseline ......................................................................................... 17 Figure 5-1: EU ETS emissions, 2015 – 2040, Options ......................................................................................... 24 Figure 5-2: EU ETS carbon price (real 2015), 2015 – 2040, Options ................................................................. 25 Figure 5-3: Net revenue from auctions, 2015 – 2040, Options .......................................................................... 27 Figure 6-1: Combinations of reform options ........................................................................................................ 28 Figure 6-2: EU ETS emissions, 2015 – 2040, Combinations of options ............................................................ 29 Figure 6-3: EU ETS carbon price, 2015 – 2040, Combinations of options ........................................................ 30 Figure 6-4: Net revenue from auctions, 2015 – 2040, Combinations of options ..............................................31 Figure 9-1: EU ETS carbon model ......................................................................................................................... 38 Figure 9-2: Marginal abatement costs curve, Industrial sectors. ....................................................................... 40 Figure 9-3: Power market dispatch model ........................................................................................................... 43 Figure 9-4: Power market model geographic scope ........................................................................................... 44


Copyright © FTI Consulting, Inc., 2017 iv

List of tables

Table 1-1: Main issues of the European Commission’s proposals for ETS reform .............................................. 2 Table 1-2: Assessment of ETS reform options, compared to the European Commission’s proposals ............ 3 Table 2-1: Assessment criteria ................................................................................................................................ 7 Table 3-1: Impact assessment of RES-E and EE policies by 2030 ......................................................................15 Table 3-2: Main issues of the European Commission’s proposals for ETS reform .......................................... 18 Table 4-1: Modelling approach .............................................................................................................................. 22 Table 5-1: Assessment of ETS reform options, compared to Baseline ............................................................. 23 Table 5-2: Life extension decisions, Options ....................................................................................................... 26 Table 6-1: Life extension decisions, Combinations of options .............................................................................31 Table 8-1: Detailed indicators, Options ................................................................................................................. 34 Table 8-2: Detailed indicators on power sector, Options .................................................................................... 35 Table 8-3: Detailed indicators, Combinations of options .................................................................................... 36 Table 8-4: Detailed indicators on power sector, Combinations of options ....................................................... 37 Table 9-1: Core assumptions, Baseline scenario ................................................................................................. 45 Table 9-2: Core assumptions, Power market model ........................................................................................... 46


Copyright © FTI Consulting, Inc., 2017 v

Abbreviations and technical units

CCGT Combined-cycle gas turbine CCS Carbon Capture and Storage CHP Combined heat plants EC European Commission ETS Emissions Trading Scheme GDP Gross Domestic Product GW Gigawatt IEA International Energy Agency LRF Linear Reduction Factor MSR Market Stability Reserve Mt Million tonnes MW Megawatts MWh Megawatt hours NPV Net Present value PPE French « Programmation Pluriannuelle de l’Energie » RES Renewable Energy Sources t Metric Tonnes TSO Transmission System Operator TWh Terawatt hours UK United Kingdom WEO14 IEA World Energy Outlook 2014 WEO15 IEA World Energy Outlook 2015


Copyright © FTI Consulting, Inc., 2017 vi

Preface

This report follows a multi-client study which started in September 2016 and concluded with a public launch event on 8 November 2016. The multi-client study provided a comprehensive assessment of the reform options currently discussed for the EU’s Emission Trading Scheme for carbon (“EU ETS”). Building on such results, this independent report presents our analysis of the current reform proposals from the European Commission (“EC”), maps the other possible reform options, assesses them, and further analyses combinations of options. FTI-CL Energy would like to thank the seven sponsor companies (CEZ, EDF, EDP, ENEL, ENGIE, Fortum, Iberdrola) of the multi-client study for their support as well as stimulation during the numerous discussions throughout the assignment, Editorial of this report closed on 30 January 2017. The views and analysis presented in this report are those of the FTI-CL Energy authors and not the views of FTI Consulting, Inc. or its management, its subsidiaries, its affiliates, or its other professionals. The right of Fabien Roques, Emmanuel Grand, Guillaume Duquesne and Yves Le Thieis to be identified as the authors of this work have been asserted in accordance with the Copyright, Designs and Patents Act 1988. CONTACTS Fabien Roques Compass Lexecon Senior Vice President froques@ compasslexecon.com

Emmanuel Grand FTI Consulting Senior Director emmanuel.grand@ fticonsulting.com

Guillaume Duquesne Compass Lexecon Senior Economist gduquesne@ compasslexecon.com

Yves Le Thieis Compass Lexecon Economist ylethieis@ compasslexecon.com

22 Place de la Madeleine 4th Floor Paris, 75008 France phone +33 1 53 05 36 15 fax +33 1 53 05 36 16 Published by FTI Consulting LLP Copyright © FTI Consulting, Inc., 2017 All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without the prior permission of the publishers. Requests for permission should be directed to [email protected] or mailed to: Compass Lexecon, 22 Place de la Madeleine, 4th Floor, Paris, 75008.


Copyright © FTI Consulting, Inc., 2017 vii

Notice: The authors and the publisher of this work have used sources believed to be reliable in their efforts to provide information that is complete and generally in accord with the standards accepted at the time of publication. Neither the authors nor the publisher nor any party involved in the preparation or publication of this work warrants that the information contained herein is in every respect accurate or complete and do not assume responsibility for any errors or omissions or for the results obtained from use of such information. The authors and the publisher expressly disclaim any express or implied warranty, including any implied warranty of merchantability or fitness for a specific purpose, or that the use of the information contained in this work is free from intellectual property infringement. This work and all information are supplied "AS IS”. Readers are encouraged to confirm the information contained herein with other sources. The information provided herein is not intended to replace professional advice. The authors and the publisher make no representations or warranties with respect to any action or failure to act by any person following the information offered or provided within or through this work. The authors and the publisher will not be liable for any direct, indirect, consequential, special, exemplary, or other damages arising therefrom. Statements or opinions expressed in the work are those of the respective author. The views expressed in this work do not necessarily represent the views of the publisher, its management or employees, and the publisher is not responsible for, and disclaims any and all liability for the content of statements written by authors.


Copyright © FTI Consulting, Inc., 2017 1

Executive summary 1.

1.1 CONTEXT AND OBJECTIVES OF THE STUDY

The EU established an Emissions Trading Scheme (“ETS”) in 2005 as the cornerstone of its climate change policy. Yet a series of economic and political factors have led to a structural imbalance of supply and demand and depressed carbon prices. This risks increasing the costs of mitigating climate change and potentially missing emission targets. The mere existence of the ETS is threatened as another decade of low prices would likely undermine its credibility and lead to further implementation of overlapping national policies and/or regulations or financial support for some specific technologies. The ongoing amendment of the Emissions Trading Directive represents a unique window of opportunity to rebuild confidence in the ETS. Accordingly, 11 European energy companies rang an alarm bell in June 2016, stressing the urgency to restore the credibility of the ETS and the EU decarbonisation strategy. In that context, we have been mandated by 7 sponsor companies (CEZ, EDF, EDP, ENEL, ENGIE, Fortum, Iberdrola) to conduct a study to provide fact-based evidence by modelling the impact of different approaches for reforming the ETS, using our models of the ETS and the European power sector. Building on this work, we have produced further independent results that are summarised hereafter.

1.2 THE EUROPEAN COMMISSION’S PROPOSED REFORM IS NOT AMBITIOUS ENOUGH TO ADDRESS THE EU ETS STRUCTURAL ISSUES

In July 2015, the European Commission (“EC”) issued a proposal for amending the EU ETS, with the following major measures:

� An increase in the speed of decline of the annual emissions cap (the “Linear Reduction Factor”) from -1.74%/year to -2.20%/year;

� An enhanced carbon leakage framework to preserve the competitiveness of the European industry.

Furthermore, at the end of 2015, the EC approved legislation on the Market Stability Reserve (“MSR”), which could annually park 12% of the surplus allowances accumulated in the previous years. Our modelling shows that the ETS reform should be more ambitious than the EC proposal since this would not lead to a decarbonisation path that is (1) underpinned by efficient short and long term price signals, (2) in line with the objectives of the Paris Agreement, (3) robust to overlapping policy interventions, (4) supportive of investment in low-carbon technologies, and (5) avoidant of lock-in of conventional fossil fuel technologies. Table 1-1 summarises our assessment.



Table 1-1: Main issues of the European Commission’s proposals for ETS reform

Issues Impact

Short term signal ineffective for affecting operational decisions

ETS carbon price will remain too low to provide an adequate short-term economic signal for carbon abatement via operation of existing plants:

� Carbon prices will only reach levels that support switching away from coal to gas after 2030 (switching range increases from 15-50€/t in 2020 to 30-80€/t in 2030).

� Coal and lignite plants will keep operating at comparable high capacity factors to today (about 45%-68% in 2016) until well beyond 2030 and gas plants’ capacity factor will also increase from 30% to 38% in 2040.

Long term signal too weak to drive efficient investment

EU ETS carbon price too low to provide an adequate long-term economic signal for decarbonisation:

� Price too low to drive investment in clean technologies, continuing the need for targeted support for specific technologies: estimates of the efficient carbon price for decarbonisation range from 20-70€/t in 2020 and 40-110€/t in 2030 – these levels are not achieved.

� Low carbon price would maintain significant fossil fuel technologies in the mix: about 360 GW of fossil fuel plants would still be in operation in 2040.

� Carbon prices below 20€/tonne by 2020 and 25€/tonne by 2025 would drive the lock-in of emissions via re-investment in 187 GW of fossil technologies over 2025-2040 (52 GW of coal and lignite power plants lifetime expansions, and 137 GW of new gas capacity).

Current emissions trajectory not in line with the goal of limiting global warming to 2°C, and a fortiori, with the ambition of limiting it to 1.5°C:

� ETS emissions 85 Mt above (+11%) a 2040 emission target which would correspond to 90% reduction in 2050 vs. 2005 emissions.

Overlap with energy and climate policies and vulnerability to shocks

ETS is a residual market for abatement with vulnerability to market and policy shocks, such as impact of overlapping policies which administratively reduce demand for allowances:

� Other low-carbon policies, mainly related to energy efficiency, are to reduce demand by 148 Mt in 2030 (11% of cap).

� National policies such as regulatory driven coal/lignite phase-out plans could further reduce demand.

Source: FTI-CL Energy

1.3 FIXING THE EU ETS: SIX ALTERNATIVE OPTIONS FOR REFORM HAVE BEEN MODELLED AND SEVERAL COMBINATIONS OF REFORMS COULD ADDRESS THE ETS ISSUES

We have modelled the impact of six (6) main alternative options for the reform of the ETS:

� Higher LRF - Setting the linear reduction factor (“LRF”) at a higher rate (e.g. 2.6% in our modelling) from 2021 would result in a faster decreasing amount of EU-wide carbon emission allowances.

� Rebasing - Setting the cap for 2021 based on recent historic emissions rather than continuation of trajectory set on 2008-2012 emissions would lower the cap. For this option, the LRF is set to the EC proposed level, i.e. 2.2% from 2021.

� Voluntary allowance cancellation - Member states could be allowed to voluntarily cancel surplus allowances that result from the success of national policies, such as ambitious national “green” plans that are aimed at more stringent carbon compliance terms, in parallel to other decarbonisation measures. This option is modelled by assuming that a “Green Club” of countries continuously purchases allowances. The Club removes a maximum of 1,489 Mt CO2 over 2021 – 2030, as we assume a budget constraint of €745 million/year (0.007% of participant GDP, same as the current Swedish approach).

� Adjustment for overlapping policies - Another option would be to allow the adjustment of the emissions cap, taking into account EU and complementary national energy and climate policies. Forward-looking, the yearly allowance supply could be automatically adjusted based on a pre-agreed methodology for



assessing the impact of various new policies, such as energy efficiency, coal phase-out and renewables support. This option is modelled by reducing the emission allowances cap by 810Mt over 2021-2030, which would neutralise the impact of the more ambitious 2030 energy efficiency and renewable targets.

� Price corridor - A price corridor could be introduced into the EU ETS between which the actual price of carbon permits can oscillate. This option is modelled by setting a floor and/or a cap for ETS prices over 2021 – 2040. The corridor could be implemented through unlimited purchase or sales of allowances by the EC at the floor or cap price, respectively. The choice of the floor and cap is paramount in letting the ETS market play within a sufficiently wide range of possible prices and reinforce its credibility and resilience to shocks. We model a price corridor at 20-50 €/t starting in 2020 and growing at 5%/year.

� Stronger MSR - The MSR outtake rate is currently set at 12%, but it could be modified to reabsorb current allowance surplus faster. This option is modelled by removing 24% of total surplus from the market as soon as surplus exceeds 833 Mt (from 2019). 100 Mt are to be released each year if the oversupply is below 400 Mt.

Each option has its own purpose and would face different implementation challenges – which are not the focus of our study. The six (6) alternative options all contribute in different ways to addressing the current ETS issues: some have positive effects in the long-term (Reduction of cap, Adjustment for overlapping policies), in the short-term (Reduction of cap, Price corridor, Stronger MSR) or in improving the ETS robustness (Price corridor, Stronger MSR). We find, however, that no single option can address all the issues of the ETS (see Table 1-2). Table 1-2: Assessment of ETS reform options, compared to the European Commission’s proposals

Options Short term Long term

Higher LRF ✘ Limited short-term impact, due to market players’ limited foresight and gradual impact of reform

✔ Growing impact as the market is drying up post-2020. LRF ≥2.6% required to have a material impact.

Rebasing ✔ Strong short-term impact as rebasing in 2021 leads market to rebalance and to significant banking

✘ Moderate impact in the long run as most effect is reduction of initial surplus

Voluntary allowance cancellation

✘

Limited impact due to budget constraint: increase of carbon price limits the amount of credits that are removed from the market under a set budget. In order to make it effective, Member States would have to invest growing amounts.

Adjustment for overlapping policies

✔ Positive impact in short/medium term as cap is gradually reduced

~ Impact depends on implementation of adjustment

Price corridor ✔

Price floor supports short-term carbon prices, stimulates investment in clean technologies and prevents lock-in of fossil plants

~ Impact depends on values of cap and floor; in any case strengthening credibility of ETS and its robustness to potential future shocks

Stronger MSR ✔

Positive short-term impact as stronger MSR rebalances market faster; higher carbon prices support investment in clean technologies and prevents lock-in of fossil plants

✘ Limited impact as stronger MSR does not alter supply and demand balance in the long term when surplus has been cleared


As no single option can address all the issues of the ETS, a range of combinations of options for the ETS reform have been modelled in order to address both short-term and long-term issues.1

1 (i) Higher LRF + Price corridor; (ii) Higher LRF + Adjustment for overlapping policies ; (iii) Higher LRF + Stronger MSR ; (iv)

Rebasing + Price corridor; (iv) Rebasing + Adjustment for overlapping policies ; (vi) Rebasing + Stronger MSR.



Our modelling demonstrates that these combinations of options can efficiently address both short-term and long-term issues of the ETS, and improve its robustness and credibility. In addition, the implementation of one of the suggested combination of options could bring up to 132 billion of euros in additional auction revenue to Member States, which could be partly used to ensure proper compensation for carbon leakage risk – an essential pre-requisite of any ambitious ETS reform. In summary, we recommend an ambitious yet realistic ETS reform based on a combination of options anchored into the following underlying principles:

� A European market based approach in order to drive decarbonisation at least cost, support investment in clean technologies, and avoid uncoordinated national interventions;

� A comprehensive answer to both short-term and long-term ETS issues, addressing the overlap of policies and improving its resilience to shocks, thereby strengthening the robustness and credibility of the ETS;

� Appropriate compensation mechanisms in place for sectors at risk of carbon leakage.



Context and objectives of the study 2.

2.1 CONTEXT

The EU established a pioneering CO2 ETS in 2003 as the cornerstone of its climate change strategy.2 The EU ETS is based on a “cap-and-trade” principle, in which an overall cap on greenhouse gas emissions is set and companies are allowed to trade emissions allowances. The allowance cap was set to be consistent with long-term goals of reducing greenhouse gas emissions by at least 80% by 2050 (as compared to emissions levels in 1990). Yet, a series of economic and political factors have led to a significant surplus of allowances in the ETS, requiring urgent and decisive action. First, shocks and structural issues have led to an imbalance of supply and demand. The high cumulated surplus of allowances resulted from a combination of the high imports of international credits and the reduction in industrial demand for permits during the recession that followed the economic crisis of 2008. The surplus amounted to around 2 billion allowances at the start of Phase 3 and further increased to more than 2.1 billion in 2013. In 2015, it was reduced to around 1.78 billion as a consequence of back-loading. Without this, the surplus would have been almost 40% higher at the end of 2015. Second, the depressed and uncertain carbon price signals do not support the industry efficiently investing in low-carbon technologies. The price of carbon emissions remained low throughout Phase 1 (2005 – 2007) and Phase 2 (2008 – 2012) of the EU ETS. In the beginning of Phase 3 (2013 – 2020), the price for emitting one tonne of carbon dioxide reached €2.81.3 Such carbon prices levels are insufficient for incentivising investment into energy efficiency and alternative energy sources, which are crucial for greater emissions reductions. This increases the costs of mitigating climate change, potentially missing targets for the EU decarbonisation. Third, the mere existence of the ETS is threatened as a few more years of low prices would likely undermine its credibility and lead to the implementation of overlapping policies. The International Emissions Trading Association estimates that the impact of the Energy Efficiency Directive and the Renewable Energy Directive alone would lead to an additional reduction in demand for ETS allowances of more than 700 million tonnes of CO2 by 2020, which would further increase the oversupply of allowances.4 The combined impact of all other existing policies may result in the continuation of the vulnerability of the EU ETS system. With this in mind, the EC has been considering reforms to the system since 2012. A report from the EC, “The state of the European carbon market in 2012”,5 presented the need for adjustments and suggested short-term back loading, along with a six (6) categories of structural options intended to reform the EU ETS:

� Back-loading: As a short-term measure the EC postponed the auctioning of 900 million allowances until 2019-2020. This “back-loading” of auction volumes does not reduce the overall number of allowances to be auctioned during phase 3, as it only changes the distribution of auctions over the period. The auction volume is reduced by 400 million allowances in 2014; 300 million in 2015 and 200 million in 2016. The back-loading would rebalance supply and demand in the short term and reduce price volatility without any significant impacts on competitiveness.

2 European Commission, 2003, “Directive 2003/87/EC of the European Parliament and of the Council”.

3 The Guardian, 2013, “EU carbon price crashes to a record low”.

4 International Emissions Trading Association, 2015, “Overlapping Policies with the EU ETS”.

5 European Commission, 2012, “The state of the European carbon market in 2012”.



� Increasing the EU reduction target to 30% (vs. 20%) in 2020: The EC analysed the implications of a volume of a retirement of allowances that would align the EU ETS cap up to 2020 with an overall target of 30% compared to 1990 and the EU's agreed long-term objective of 80-95% by 2050 compared to 1990. Such a volume would be equal to around 1.4 billion allowances. The Commission analysed the associated implications at Member State level. This option would not only require changes to the quantity of allowances in the EU ETS but also affect the targets adopted under the Effort Sharing Decision.

� Retiring a number of allowances in phase 3: The results sought of this option are to reduce the number of allowances issued in phase 3 by permanently retiring a number of allowances from the amount foreseen to be auctioned. Specifically, the measure can be effective in addressing the overall supply-demand imbalance over phase 3. It would implicitly increase the numerical reduction target for 2020 and thus (partially) restore the ambition level of the 2008 climate-energy package, but it would not directly affect the framework after 2020. It would reduce the surplus of allowances in phase 3 and, depending on the amount retired, ensure that the ETS contributes to renewables and energy efficiency objectives.

� Early revision of the annual linear reduction factor (to be increased from 1.74%): This structural measure could not only address the imbalance and (partially) restore the ambition level up to 2020, but would also impact the ambition level after 2020. As such the linear factor could be set at levels in-line with an overall EU target of 30% GHG reductions compared to 1990. The previous linear factor leads to a just over 70% reduction in the ETS cap by 2050, which is not consistent with the EU's agreed long-term objective of 80-95% reduction by 2050 compared to 1990, as the Commission has pointed out in the 2050 Low-carbon Roadmap.6

� Extension of the scope of the EU ETS to other sectors, less exposed to economic cycles: Whereas the emissions in the EU ETS decreased in 2009 by more than 11%, in the sectors outside the EU ETS this reduction was only around 4%. This difference may be partially explained by differing impacts of the economic crisis on individual sectors. The coverage of the EU ETS could therefore be expanded to other energy related CO2 emissions in sectors currently outside the EU ETS by for instance including fuel consumption in other sectors.

� Limit access to international credits: Following the exceptional macro-economic developments and the fact that emissions have been substantially lower than the cap, the quantity limit of international credits in the period 2008 to 2020 has turned out to be rather generous and is a major driver for the build-up of the surplus. Without international credits, the surplus in the EU ETS by 2020 would potentially be only around a quarter (25%) of the presently expected surplus.

� Discretionary price management mechanisms (such as a carbon price floor or price management reserve): To reduce volatility and prevent price drops due to temporary mismatch between supply and demand, a carbon price floor would create more certainty about the minimum price, giving a better signal for investors.

To date, the EC has implemented one amendment; in 2014, EU ETS regulation was amended to implement back-loading of 900 million allowances in 2014 – 2016. Despite the proposed reforms, stakeholders have questioned whether sufficient measures are in place to meet the EU’s long-term energy goals for 2050, or if the EC has retained too much focus on short-term goals for 2030. Subsequent to a period of public consultation, the EC proposed further reforms to the EU ETS in July 2015, by increasing the speed of decline of the annual emissions cap from 1.74% p.a. to 2.20% p.a. and by enhancing carbon leakage framework to preserve the competitiveness of the European industry. Furthermore, at the end of 2015, the EC introduced the MSR, which could annually park 12% of the surplus allowances accumulated in the previous years.

6 European Commission, 2011, “A Roadmap for moving to a competitive low-carbon economy in 2050”.



2.2 OBJECTIVES OF THE STUDY

There is a unique window of opportunity (which will be closed in a few months) to align the ETS with the increased ambition of the Paris climate agreement, as amendments to the ETS directive are currently being discussed in the Council and Parliament. In this context, the study aims at contributing to the public debate on potential means to strengthen the short- and long-term carbon price signal within the EU ETS so as to decarbonise the European economy at the least cost. We conduct a fact-based study to assess the impact of the ETS reform tabled by the EC as well as six (6) alternative options, and their effect on the power sector. Using a multi-criteria analysis, the study discusses the pros and cons of alternative approaches for ETS reform and provides policy recommendations. The scope of our work is specifically focused on supply and demand balance and price, as well as the power sector; other ETS issues such as auctioning, compensation for energy-intensive industries, and the proposed fund for innovation are disregarded.

2.3 MODELLING FRAMEWORK

We assess the impact of the ETS reform tabled by the EC as well as six (6) alternative options along both well-known short-term and long-term ETS issues (cf. Table 2-1 below). Table 2-1: Assessment criteria

Quantity Short term Long term

ETS Market

Emissions In line with short-term EU targets? In line with long-term EU targets?

Prices Provide efficient signal for carbon abatement via coal-gas switching?

Drive investment in clean technologies?

ETS Robustness Policies overlap Robustness to market shocks and complementary policies?


Accordingly, our modelling takes specific features of the EU ETS into account to properly assess the likely effect of the different options of reform on emissions, carbon prices, and investments decisions in the power sectors in both the short term and the long run. We pay particular attention to modelling in a realistic way the behaviour of market players in order to capture the potential impact of the different reforms. The EU ETS Market is an intertemporal market in which agents follow different banking behaviours:

� Arbitrage, which consists of buying allowances and simultaneously selling forward, future or option contracts, so as to avoid exposure to carbon price risk;

� Hedging, which consists of holding allowances to meet future carbon credit needs as an input to the production process, and thus avoiding the exposure to carbon price risk; and

� Speculation, which consists of taking an open position in carbon allowances and carrying the carbon price risk in expectation that the carbon price will evolve favourably.

EU ETS markets are thus linked across time in a dynamic way through banking. Future (expected) reforms would thus have an immediate impact on the market while current reforms would have an impact on banking behaviours and thus on future market conditions. This specific feature of the EU ETS has two mains implication for our study:

� First, there is a need to model properly banking behaviour (speculation and hedging) of participants; and

� Second, when assessing options for reform, we should keep in mind the intertemporal dimensions of the EU ETS market.



With this respect, we develop an extended version of the Zephyr ETS model7 supported by our in-house power sector model8 (cf. Figure 2-1). The model calculates the EU ETS carbon price and emissions from the power and industrial sectors based on a detailed representation of ETS market supply and demand fundamentals. It aims at both replicating the imperfect inter-temporal compliance behaviour observed ex post – due to short-sightedness and imperfect response from market participants to uncertainty and complexity9 – and observes how they may interact with exogenous scenarios such as implementation of ETS reforms. Our EU ETS model is calibrated on the recent EC Reference Scenario 2016 while our detailed power sector model is based on the latest announcements from TSOs, regulators and market participants. The supply and demand inputs into the power sector model incorporate a combination of TSO reference scenario outlooks, expected projects for the future, and national plans. Commodity price assumptions rely on forwards until 2020, after which they converge to WEO 2015 New Policy by 2040.

Figure 2-1: EU ETS carbon model Source: FTI-CL Energy

2.4 OUTLINE OF THE REPORT

The paper is structured as follows. We first identify key issues with the current proposal of EC for EU ETS reform. We show that the EC’s proposed reform is not ambitious enough to address the EU ETS structural issues. Then, we assess the potential of others approaches. Six alternative options for reform are modelled. We show that several combinations of reforms may address the ETS issues. Finally, we conclude and present our policy recommendations.

7 The EU ETS modelling approach is inspired from the ZEPHYR model developed by Raphaël Trotignon & Boris Solier (Paris

Dauphine University, Chaire Economie du Climat). http://www.chaireeconomieduclimat.org

8 A detailed description of our modelling framework can be found in 9.

9 https://ec.europa.eu/clima/policies/ets/index_en.htm



Diagnosis: Key issues with the current EC 3.proposal for EU ETS reform

3.1 EC PROPOSALS FOR AMENDING THE EU ETS

In July 2015, the EC issued a proposal for amending the EU ETS,10 with the following major amendments that are relevant to our study:

� A targeted emission reduction of 43% in 2030 versus 2005 levels. A centrepiece of the 2030 policy framework is the binding target to reduce overall EU greenhouse gas emissions by at least 40% domestically below 1990 levels by 2030. To achieve this target cost-effectively, the sectors covered by the EU emissions trading system (EU ETS) will have to reduce their emissions by 43% as compared to 2005 while non-ETS sectors will have to reduce their emissions by 30% as compared to 2005.

� An increase of the annual linear reduction factor, which is the percentage by which the number of allowances is reduced annually, from 1.74% to 2.2%. This corresponds to an additional reduction of around 556 million tonnes of carbon dioxide in the period 2021-2030 compared to the current annual decline of 1.74%.

� The EC has proposed establishing a MSR in 2019, which will allow the EC to assess the cumulative under- or oversupply each year, injecting or withdrawing allowances as needed if certain standards are met. 12% of oversupply in previous years is to be withdrawn, if the oversupply was above 833 million. 100 million are to be released if the oversupply was below 400 million. Release and withdrawals will be smoothed over a 12-month period, and 900 million allowances back-loaded from 2014 – 2016 will be transferred to the MSR, instead of being put back on the market.

� The EC has proposed enhancing competitiveness and strengthening the carbon leakage framework. Under the current framework, nearly all industrial sectors receive 100% free allocation based on historical production levels and best-in-class technology benchmarks, as nearly all industrial sectors are considered to be at risk of carbon leakage. With Phase 4 revisions, free allocation will be more focussed on the highest carbon leakage risk industries (about 50 sectors), and benchmarks will be more regularly updated. Furthermore, the Innovation Fund will be increased from 300 million to 400 million allowances, and a new Modernisation Fund of 310 million allowances will be created to support energy transition investments in the 10 lower-income Member States. The New Entrants Reserve will be decreased from 480 million to 250 million allowances.

Our modelling shows that the ETS reform should be more ambitious than the EC proposal since this would not lead to a decarbonisation path that is underpinned by efficient short- and long-term price signals, in line with the objectives of the Paris Agreement, robust to overlapping policy interventions, supportive of investment in low-carbon technologies, and avoidant of the lock-in of conventional fossil fuel technologies.

3.2 THE CURRENT EMISSIONS TRAJECTORY IS NOT IN LINE WITH THE OBJECTIVE OF LIMITING GLOBAL WARMING TO +1.5°C

The “2050 low-carbon roadmap”11 has resulted in a linear reduction factor increase to 1.74% in order to meet the EU goal of limiting global climate change to a temperature increase of +2°C by 2050. This linear reduction factor has been further increased to 2.2% within the “EU Climate and Energy framework”. 12

10 https://ec.europa.eu/transparency/regdoc/rep/1/2015/EN/1-2015-337-EN-F1-1.PDF

11 https://ec.europa.eu/clima/policies/strategies/2050/index_en.htm

12 https://ec.europa.eu/clima/policies/strategies/2030/documentation_en.htm



Despite this increase, the EC has recognised that this linear reduction factor would still be insufficient for achieving 90% emissions reduction by 2050.13 As the EU now agreed to pursue efforts to stay below 1.5°C in accordance with the Paris Agreement, the intermediate targets should be revised. Specifically, assuming that setting the linear reduction factor to 2.4% would be sufficient to comply with the 90% reduction target by 2050 – what is stated by the EC – the ETS targets should be revised downwards. In 2030, the target should be fixed at 1,289 Mt as compared to 1,333 Mt for an objective of 80% reduction by 2050. In 2040, the target should be fixed at 761 Mt as compared to 849 Mt for an objective of 80% reduction by 2050. Accordingly, Figure 3-1 plots the trajectory of emissions reductions from 2015 to 2040 based on the current system and on the proposed EC reform as well as EC targets for both 80% and 90% reduction by 2050.

Figure 3-1: EU ETS emissions, 2015 – 2040, Baseline Sources: EC (2014), “Impact assessment 2014 - A policy framework for climate and energy in the period from 2020 up to 2030“; FTI-CL Energy Notes: (i) EU ETS targets calculated based on the verified emissions for ETS sectors as of 2005, and EU emissions reduction targets expressed in % 2005 emissions reduction. (ii) Business-as-Usual stand for a LRF of 1.74% with MSR active from 2019.

As shown in Figure 3-1, Baseline meets EU ETS targets of 80% reduction by 2050. However, the current emissions trajectory is not in line with the target of 90% reduction by 2050, and a fortiori, the objective of

13 European Commission, 2014, “Impact Assessment 2014 “: « In order to set the cap equal to this level [90% emissions

reduction by 2050], the LRF in the ETS would need to further increase to -2.4% until 2050 ».



limiting global warming to +1.5°C as suggested by the Paris Agreement. In 2040, ETS emissions are 85 Mt CO2-eq. above the emissions reduction target in line with 90% reduction by 2050 – corresponding to 11% of the target.

3.3 EU ETS CARBON PRICE IS TOO LOW TO PROVIDE ADEQUATE SHORT-TERM ECONOMIC SIGNAL FOR CARBON ABATEMENT

The EU ETS carbon price level is too low to drive investment in clean technologies (RES, nuclear, etc.) and avoid (re-)investments in fossil fuels technologies. Figure 3-2 below plots CO2 prices in our Baseline scenario. We also report (i) a range of CO2 breakeven price benchmarks, which are prices at which the industry is indifferent between coal-produced energy and gas-produced energy;14 as well as (ii) the social cost of carbon which is the marginal global damage costs of carbon emissions for Europe. The latter is estimated as the net present value (hereafter “NPV”) of climate change impacts in the long-term of one additional ton of carbon emitted today.

Figure 3-2: EU ETS carbon price (real 2015), 2015 – 2040, Baseline Source: FTI-CL Energy Notes: (i) The range for the “social cost of carbon” is based on various academic and government benchmarks, in particular Knopf (2013), “The EMF28 Study on Scenarios for Transforming the European Energy System”. (ii) Given the range of efficiencies of existing plants, the fuel switch would be triggered between a range of CO2 prices.

As shown in Figure 3-2, the current proposals support an inefficient decarbonisation path. Over period 2015 – 2040, Baseline CO2 prices fail to meet the lowest estimate of the cost of environmental damages caused by carbon emissions – ranging from about 20-70€/t in 2020, and 40-110€/t in 2030.

14 Given the range of efficiencies of existing plants, the fuel switch would be triggered between a range of CO2 prices.



Similarly, ETS carbon price levels are too low to provide a reliable short-term economic signal for switching to low-carbon technology in the power sector. Our modelling shows that the proposed ETS reform would lead to carbon price levels that will only reach levels that support switching away from coal and lignite to gas after 2030. Accordingly, energy players are expected to invest systematically to extend the lifetime of their coal and lignite power plants, as the NPV of such extensions would largely be positive. These observations are robust to a range of coal and gas price combinations based on external sources. Precisely, across a range of commodity price scenarios, carbon prices in Baseline are still too low to trigger the coal-gas fuel switching point.15

3.4 EU ETS CARBON PRICE IS TOO LOW TO PROVIDE ADEQUATE LONG-TERM ECONOMIC SIGNAL FOR CARBON ABATEMENT

The ETS baseline scenario leads to a significant long-term lock-in of fossil generation capacity. The modelling of the expected remaining operational life of coal and lignite plants highlights two dynamics:

� Overall installed coal and lignite capacity decrease: Current regulations and the ETS baseline scenario would lead to a slow decrease of coal and lignite capacity in Europe. Around 75GW (43%) of the coal and lignite capacity would close by 2030, and a remaining 100GW are already compliant with the latest regulation and not subject to national phase-out plans. These plants would be subject to the ETS price. A further 40GW would close from 2025 to 2040, leaving about 67GW that would still be operational in 2040.

� Economic extension decisions: Carbon prices below 20€/tonne by 2020 and 25€/tonne by 2025 would drive lock-in of emissions via (re)investment in 187 GW of fossil technologies over 2025-2040 (52 GW of coal and lignite power plants lifetime expansions and 137 GW of gas new capacity). With regards to high carbon technology generation, EU ETS carbon prices are too low to incentivise early closure of coal and lignite plants, as most plants compliant with emissions standards would likely be extended. Specifically, while the short-term margins are close to zero due to a large overcapacity across Europe, with the steady tightening of the EU power market, margins from the energy market would increase in the medium term. The ETS baseline and current regulations would thus lead to a positive NPV for coal and lignite plants to be extended to 50 years. Over period 2025 – 2040, about 52 GW of coal and lignite capacity will expand their life time.

Figure 3-3 shows coal and lignite capacity outlook in Baseline.

15 We have benchmarked our fossil fuel price assumptions against a range of external scenarios: (i) The International Energy

Agency World Energy Outlook 2015 (WEO) “Current policies”, “New Policies”, and “450” scenarios (ii) The European Commission 2016 scenario.



Figure 3-3: Coal and lignite capacity outlook, 2015 – 2040, Baseline Sources: Platts, national registers, LCP dataset, Transitional National Plan and operators announcements; FTI-CL Energy Note: We use plant-specific information on all coal & lignite plants. In case of no data, assumption of a standard lifetime of 50 years coherent with Germany G7 Coal analysis (September 2015).

Overall low carbon price would maintain significant carbon emitting technologies in the mix: about 360 GW of fossil fuel plants would still be in operation in 2040 (67 GW of coal and 293 GW of gas). New investments in high-carbon generation like coal and lignite are strong indications of an inefficient path to decarbonisation. The lock-in of high carbon technologies through renewed investments would not support a progressive decarbonisation but rather a wall of low-carbon investments at period-end (2050), unlikely to minimise the total costs for the sector. It worth noting that remaining high carbon technology plants will keep operating at high capacity factors from today until well beyond 2030, as shown in Figure 3-4.



Figure 3-4: European utilisation factor, 2015 – 2040, Baseline Source: FTI-CL Energy

In a low CO2 price scenario (below the CO2 fuel-switch level), coal and lignite plants will keep operating at high capacity factors, while CCGT will keep a low capacity factor. Specifically, coal and lignite plants will keep operating at comparable high capacity factors to today (about 45%-68% in 2016) until well beyond 2030 and gas plants’ capacity factor will also increase from 30% to 38% in 2040. In other words, coal and lignite plants will remain baseload plants, while gas plants will remain mid-merit to peaking plants. This will put significant pressure on existing CCGT plants.

3.5 ETS MARKET IS A RESIDUAL MARKET FOR ABATEMENT WITH VULNERABILITY TO MARKET AND POLICY SHOCKS

Other European and national policies and measures driving CO2 emission reductions (RES, efficiency and emission standards) should be neutralised so that ETS does not compound uncertainty. EU policies in the power sector reduce the demand for carbon allowances, threatening the ETS balance and strength of the price signal After the 2014 EC Impact Assessment, RES-E and EE targets have been further increased to achieve 40% GHG reduction by 2030. To neutralise the impact of this overlapping policy, the allowance cap should be modified to reflect the reduction of the demand for ETS credits. Table 3-1 sets out an impact assessment of these policies by 2030.



Table 3-1: Impact assessment of RES-E and EE policies by 2030

RES-E 2030 Target EE 2030 Target

Overall Power

Old reference

GHG 40 scenario 26.5% 47.3% 25.1%

Old reference

EU 2030 Targets 27% 48.2% 30%

Delta New vs. Old

Generation equivalent 24 TWh in 2030 187 TWh in 2030

Delta New vs. Old

CO2 emission eq. 17 Mt in 2030 131 Mt in 2030

Sources: EC (2007), “Reference scenario to 2050”; FTI-CL Energy Note: Calculations are based on the 2007 baseline final consumption in 2030 of 3,809 TWh and a marginal emission intensity of 0.7t/MWh.

The increase of the RES-E 2030 target from 26.5% to 27% – or an additional 24 TWh of renewable generation in EU final consumption – would reduce demand for allowances by 17 Mt CO2-eq. in 2030. Similarly, the increase of the EE 2030 target from 25.1% to 30% – or a reduction of the EU demand of 187 TWh in 2030 – would reduce emissions by 131 Mt in 2030. For the sake of completeness, Figure 3-5 below quantifies the impact that RES-E and EE policies would have on CO2 emissions reductions from 2021 to 2030.

Figure 3-5: Allowance cap reduction over 2021 - 2030 Sources: EC (2007), “Reference scenario to 2050”; FTI-CL Energy Note: Calculations are based on the 2007 baseline final consumption in 2030 of 3,809 TWh and a marginal emission intensity of 0.7t/MWh.



Figure 3-5 shows that emission allowances cap should be reduced by 810Mt over 2021-2030 to neutralise the impact of more ambitious RES-E and EE 2030 targets. The inability of the EU ETS to provide credible signals to decarbonise the power sector has led Member States to step in Several member states have already implemented national policies in the power sector, further threatening the credibility of the EU ETS market and achieving mandated emission reductions at a higher cost than ETS-driven abatement. In 2015, the UK government introduced a plan to phase out coal by 2025.16 With 11 coal plants still in operation, the forced closure of these plants could reduce significantly CO2 emissions, leading to CO2 emissions reduction of 207 Mt over period 2015 – 2040. Alternative coal national phase-down policies will lead to further 341 Mt reduction over the same period:

� As part of the “Climate Program 2020”,17 the German power plant fleet is required to contribute an extra 22 Mt in CO2 savings, including the gradual take out of 2.7 GW of old German lignite plants from the market by 2020.

� The Netherlands enacted a coal phase-out plan in two phases.18 First, the oldest coal plants from the 1980’s were closed in parallel to the commissioning of new coal plants. However, the second phase envisages closing the remaining coal plants (including the new 2.4 GW) by 2030.

� As part of the PPE to 2023,19 France plans to shut down the 3GW remaining coal plants by 2023.

The MSR will quickly grow to several billion, introducing additional instability to the ETS market Under current proposed reforms, the MSR may grow too quickly, introducing additional instability to the ETS market. As shown in Figure 3-6, while the ETS surplus will only slowly be reabsorbed, the MSR will quickly grow to several billion allowances – unless some credits are cancelled.

16 UK Government to phase out unabated coal generation by 2025, 9 Nov 16

https://www.gov.uk/government/consultations/coal-generation-in-great-britain-the-pathway-to-a-low-carbon-future

17 Climate reserve introduction as part of the German Energy transition strategy https://www.bmwi-energiewende.de/EWD/Redaktion/EN/Newsletter/2015/07/Meldung/topthema-herausforderung-energiewende.html

18 Dutch lawmakers back coal phase out, Nov 2015. http://news.vattenfall.com/en/article/coal-be-phased-out-netherlands

19 Décret n° 2016-1442 du 27 octobre 2016 relatif à la programmation pluriannuelle de l'énergie.



Figure 3-6: Surplus and MSR, 2015 – 2030, Baseline Source: FTI-CL Energy

The allowance surplus could increase up to ~2,300 Mt CO2-eq by 2020. Investors anticipating MSR from 2019 would take speculative positions to profit from the anticipated ETS price increase as the market gets tighter. From 2019, the surplus would start declining to reach 860 Mt CO2-eq in 2040. Several effects are at work. First, investors would start unlocking their speculative positions as soon as the MSR is implemented. Second, the MSR would start absorbing credits from the ETS market, which in turn would lead sectors to decrease their emissions, reducing their hedging needs. In 2040, the total number of allowances in the reserve could amount to ~3,960 Mt CO2-eq.

3.6 THE REFORM NEEDS TO BE MORE AMBITIOUS IF IT IS TO TACKLE THE DIFFERENT ETS ISSUES AND REBUILD THE CREDIBILITY OF THE ETS

As discussed above, our modelling shows that the ETS reform should be more ambitious than the EC proposal since this would not lead to a decarbonisation path that is underpinned by efficient short- and long-term price signals, in line with the objectives of the Paris Agreement, robust to overlapping policy interventions, supportive of investment in low-carbon technologies, and avoiding lock-in of conventional fossil fuel technologies. Table 3-2 provides a synthetic assessment of the issues identified for ETS reform – classifying them as short-term or long-term issues.



Table 3-2: Main issues of the European Commission’s proposals for ETS reform

Short term Long term

Emissions ✔ Emissions below target – largely driven by complementary policies

✘ Not in line with the goal of limiting global warming to 2°C, and a fortiori, with the ambition of limiting it to 1.5°C

Prices ✘ Too low to provide efficient signal for carbon abatement via coal-gas switching, and driving lock-in of fossil plants

✘ Too low to drive investment in clean technologies leading to continuation of need for targeted support for specific technologies

Policies overlap ✘ Vulnerability to market shocks and overlap with complementary policies. Overlapping low-carbon policies achieve mandated abatement at a high cost and displace ETS-driven efficient abatement.


To sum up, the current EC proposal for reforms suffer from three main shortcomings:

� Short-term signal ineffective to affect operational decisions – ETS carbon price will remain too low to provide adequate short-term economic signal for carbon abatement via operation of existing plants. Carbon prices will only reach levels that support switching away from coal to gas after 2030. While, coal and lignite plants will keep operating at comparable high capacity factors to today (about 45%-68% in 2016) until well beyond 2030 and gas plants’ capacity factor will also increase from 30% to 38% in 2040.

� Long-term signal too weak to drive efficient investment – EU ETS carbon price too low to provide adequate long-term economic signal for decarbonisation. Prices are too low to drive investment in clean technologies leading to continuation of need for targeted support for specific technologies. To the contrary, they would drive the lock-in of emissions via re-investment in 187 GW of fossil technologies over 2025-2040. Carbon price would therefore maintain significant fossil fuel technologies in the mix. Mechanically, current emissions trajectory not in line with the goal of limiting global warming to 2°C, and a fortiori, with the ambition of limiting it to 1.5°C.

� Overlap with energy and climate policies and vulnerability to shocks – ETS is a residual market for abatement with vulnerability to market and policy shocks, such as the impact of overlapping policies which administratively reduce demand for allowances. Other EU low-carbon policies, as well as national policies such as regulatory driven coal/lignite phase-out plans, are to reduce demand.

In the next section we consider alternative strengthening measures.



Fixing the EU ETS: Possible approaches for 4.reform

We model the impact of six (6) alternative options to the EC Reform, namely:

� Higher LRF – To set the LRF at a higher rate (e.g. 2.6% in our modelling) from 2021.

� Rebasing – To set the cap for 2021 based on recent historic emissions rather than continuation of trajectory set on 2008-2012 emissions.

� Voluntary allowance cancellation – To allow Member States to voluntarily cancel surplus carbon permits.

� Adjustment for overlapping policies – To allow the adjustment of the emissions cap, taking into account EU and complementary national energy and climate policies.

� Price corridor – To introduce a price corridor into the EU ETS between which the actual price of carbon permits can oscillate.

� Stronger MSR – To modify the MSR outtake rate to reabsorb current emission credit surplus faster.

In the following, we discuss each of those options in more detail.

4.1 HIGHER LRF AND REBASING

The EC has suggested changing the LRF from 1.74% to 2.2% from 2021, but this may not be sufficient for reaching the 2050 decarbonisation objective. Setting the LRF at a higher rate would be an easily implementable policy amendment which would result in a faster decreasing amount of EU-wide carbon emission allowances – allowing a faster decrease of EU-wide carbon allowances supply and reducing the cumulative oversupply. Various proposals have been submitted to the European Parliament, ranging from 1.74% to 2.8% LRFs, with the rationale that this increased level will allow the EU to reach 80% - 95% emissions reduction by 2050, as per the European Commission’s current 2050 low-carbon economy vision.20 Furthermore, following the Paris Agreement, stakeholders have revised upwards proposals. Most notably, a report on the behalf of the French government proposed a linear reduction factor of 2.4% to 2.6% per year from 202021 to be in line with the objective of limiting global warming to +1.5°C. Similarly, Eurelectric has proposed an increase of the LRF beyond 2.2%,22 while Carbon Market Watch notes that a minimum of 2.4% is required to be consistent with the low-carbon path.23 Sandbag, a UK-based not-for-profit think tank, also believes that ETS needs a 2.4% linear reduction factor in order to reach 90% emission reduction by 2050 (as compared to 1990 levels).24 We considered two variants:

20 European Commission, 2016, “Stakeholder feedback on the ETS revision legislative proposal – Documentation, Revision

for phase 4 (2021 – 2030)”; European Commission, 2011, “2050 low-carbon economy”.

21 Canfin, Pascal, Alain Grandjean, and Gérard Mestrallet, 2016, « Rapport de la Mission : Proposition pour des prix du carbone alignés avec l’Accord de Paris ».

22 Eurelectric, 2016, “Reform of the EU ETS: A Eurelectric Statement”, May 2016.

23 Carbon Market Watch, 2015, “Four magic potions to turn the EU ETS into an effective climate mitigation tool”.

24 Sandbag Blog, 2016, “A wake-up call for European emissions targets”.



� Higher LRF - Setting the LRF at a higher rate (e.g. 2.6% in our modelling) from 2021 would result in a faster decreasing amount of EU-wide carbon emission allowances.

� Rebasing - Setting the cap for 2021 based on recent historic emissions rather than continuation of trajectory set on 2008-2012 emissions would lower the cap. For this option, the LRF is set to the EC proposed level, i.e. 2.2% from 2021.

Those options may result in higher predictability for market participants, compared to other options, given its simplicity and link to long-term commitments. However, it does not allow for any flexibility in adjusting to external changes, such as economic or policy shocks.

4.2 VOLUNTARY ALLOWANCE CANCELLATION

Member states could be allowed to voluntarily cancel surplus carbon permits that result from the success of national policies, such as ambitious national “green” plans that are aimed at more stringent carbon compliance terms, in parallel to other decarbonisation measures. Alternatively, the EU can implement a wider version of the Swedish Government’s cancellation program,25 such as a policy which would have specific subsets of countries buying and cancelling a pre-agreed amount of allowances. This is not an unusual policy, as a number of non-profit organisations currently also engage in this practice. In our modelling we consider that a “Green Club” of countries continuously purchase allowances. The Club removes a maximum of 1,489 Mt CO2 over 2021 – 2030, as we assume a budget constraint of €745 million/year (0.007% of participant GDP – same as the current Swedish approach). A voluntary allowance cancellation would signal strong commitment from the member states in question, which would have an immediate impact. However, it may result in free riding from non-participation member states; the cancellation would only be paid for by a subset of countries, which would be politically difficult to “sell” to all member states.

4.3 ADJUSTMENT FOR OVERLAPPING POLICIES

Another option would be to allow the adjustment of the emissions cap, taking into account EU and complementary national energy and climate policies. Several suggestions for dynamic cap adjustment have been submitted to the European Parliament. For instance:

� The retirement of the share of auctioned volumes attributed to retired power plants would increase predictability in the carbon market;26 and

� Creating or transforming another mechanism into a “dynamic symmetric mechanism” which is able to temporarily release or withdraw allowances from the market when production changes by more than 10% would allow greater sensitivity to production and shocks brought on by a swing in business cycles.27

25 Government Offices of Sweden, 2016, “Real emission reductions and more pressure on the EU due to new Swedish EU

ETS policy”.

26 An amendment proposed to the European Parliament’s Committee on Industry, Research and Energy (ITRE) suggested this adjustment in order to increase predictability in the carbon market.

27 An amendment proposed to ITRE suggested this mechanism because ex-post dynamic adjustments more sensitive to smaller changes in production are needed to prevent over-compensation as well as under-compensation of free allowances.



In our modelling, we have only considered28 backward-looking corrections. The yearly allowance supply is automatically adjusted based on a pre-defined methodology for assessing the impact of various overlapping policies, such as energy efficiency, coal phase-out and renewables support. This option is modelled by reducing the emission allowances cap by 810Mt over 2021-2030, which would neutralise the impact of the more ambitious 2030 energy efficiency and renewable targets. It is worth noting that it would be difficult in practice to agree upon an uncontroversial methodology for measuring the impact of national policies.

4.4 PRICE CORRIDOR

A price corridor could be introduced to the EU ETS, as in North America’s emission trading scheme;29 this would be a price range in between which the actual price of carbon permits is allowed to oscillate. A price corridor may be accomplished either through an auction reserve price or via a modification of the MSR, which would absorb or release allowances based on their price rather than on supply-based rules. The price corridor could also be supported by a reserve, where allowances below the price floor could be held by the reserve until the price ceiling is reached. A price corridor can provide price certainty and a clear long-term carbon price signal by partially insulating the price from economic and policy shocks. Specifically, the price floor of the corridor will ensure minimum efforts in abatement through a guaranteed allowance value, while the price ceiling will enhance the mechanism credibility by capping the costs of compliance. However, optimal levels for a price floor and ceiling will be difficult to determine, as the EC would risk over- or undershooting quantity targets; a price corridor that is too narrow, for instance, could defeat the principle of market price discovery and reduce economic efficiency (as with other tax-like measures). The existence of a price corridor could also make a globally unified emissions trading scheme less feasible. This option is modelled by setting a floor and/or a cap for ETS prices over 2021 – 2040. The corridor could be implemented through unlimited purchase or sales of ETS credits by the EC at the floor or cap price, respectively. The choice of the floor and cap is paramount for letting the ETS market play within a sufficiently wide range of possible prices and reinforce its credibility and resilience to shocks. We model a price corridor at 20-50 €/t starting in 2020 and growing at 5%/year.

4.5 STRONGER MARKET STABILITY RESERVE

Starting 2019, the MSR will remove or add carbon allowances from the EU ETS based on quantity thresholds; if the allowances surplus exceeds a certain threshold, a number of allowances are removed based on an outtake rate. The outtake rate is currently set at 12%, but it could be modified to reabsorb current emission credit surplus more quickly. This amendment will have the benefit of reducing uncertainty for market participants even in the case of market shock, as supply will be determined by a pre-set adjustment rule. However, a 12% outtake rate is too

28 Modelling this option involved (i) evaluating increases in RES-E and EE policy targets; (ii) converting generation equivalents

(TWh) that would have otherwise been produced by thermal production; and, (iii) estimating CO2 emission equivalent that would otherwise have been produced by thermal generation and removing from the ETS market.

29 California and Quebec have price floors that increase 5% + inflation annually, and RGGI in northeast US increases its price floor 2.5% each year. Outside of North America, New Zealand sets a ceiling price at 50 NZD (30€), and several cities in China set price floors and/or ceilings (Hubei, Shanghai, and Shenzhen set price floors respectively at 20 yuan (3€), 46 yuan (6€), and 35 yuan (5€).



low to act as a countercyclical mechanism; at least 5 years would be required to bring oversupply to the target amount (400 – 833 Mt). A number of commentaries30 submitted in response to the EC’s proposed reforms have discussed alternatives to a 12% outtake rate. For example, Sandbag has suggested introducing a supply adjustment equal to 33% of the difference between the surplus and the upper end of the band, to avoid over- or under-adjusting supply. This option is modelled by removing 24% of total surplus from the market as soon as surplus exceeds 833 Mt (from 2019). 100 Mt are to be released each year if the oversupply is below 400 Mt. Despite the number of variations within this option, a quantity-based MSR outtake rate may still be insufficient for offsetting oversupply. The time delay of the instrument makes it potentially slow to stabilise prices, which could ultimately result in countercyclical carbon prices. Throughout the rest of the document, we conduct an in-deep assessment of each of all those proposed reform.

4.6 MODELLING APPROACH FOR THE DIFFERENT OPTIONS

Our framework allows us to model options for reform in a simple way. As shown in Table 4-1 below, options for reform would affect supply and/or demand – which would in turn alter banking behaviours. Table 4-1: Modelling approach

Options Supply Demand

Higher LRF Annual supply of credit is decreased consistently with EU cap reduction

Unchanged

Rebasing Annual supply of credit is decreased consistently with rebasing

Unchanged

Voluntary allowance Cancellation

Unchanged A third party is assumed to buy allowances (subject to budget constraints) which triggers prices – and implicitly reduces demand


Annual supply of credit is decreased consistently with neutralisation policies

Unchanged

Price corridor Unchanged Price cap is converted into unlimited demand for credits when prices are below the floor.

Stronger MSR Reduction of auctioned volumes Unchanged


30 European Commission, 2016, “Stakeholder feedback on the ETS revision legislative proposal – Documentation, Revision

for phase 4 (2021 – 2030)”.



Assessment of the different options for reform 5.of the ETS

In the following, we have assessed the results of each option, in relation to the criteria set out when examining the key issues with the current reform.

5.1 PRELIMINARY ASSESSMENT OF OPTIONS

Table 5-1 provides a preliminary assessment of options. Table 5-1: Assessment of ETS reform options, compared to Baseline

Options Short term Long term

Higher LRF ✘ Limited short-term impact, due to market players’ limited foresight and gradual impact of reform

✔ Growing impact as the market is drying up post-2020. LRF ≥2.6% required to have a material impact.

Rebasing ✔ Strong short-term impact as rebasing in 2021 leads market to rebalance and to significant banking

✘ Moderate impact in the long run as the strongest effect is reduction of initial surplus

Voluntary allowance cancellation

✘

Limited impact due to budget constraint: increase of carbon price limits the amount of credits that are removed from the market under a set budget. In order to make it effective, Member States would have to invest growing amounts.


✔ Positive impact in short/medium term as cap is gradually reduced

~ Impact depends on implementation of adjustment

Price corridor ✔ Price floor supports short-term carbon prices, stimulates investment in clean technologies and prevents lock-in of fossil plants

~ Impact depends on values of cap and floor; in any case strengthening credibility of ETS and its robustness to potential future shocks

Stronger MSR ✔

Positive short-term impact as stronger MSR rebalances market faster; higher carbon prices support investment in clean technologies and prevents lock-in of fossil plants

✘

Limited impact as stronger MSR does not alter supply and demand balance in the long term when surplus has been cleared


The six (6) alternative options modelled all would contribute in different ways to address the current ETS issues. Some would have positive effects in the long-term (Higher LRF and Rebasing), in the short-term (Price corridor, Stronger MSR), or in improving the ETS robustness (Price corridor, Stronger MSR, Adjustment for overlapping policies). We find that no single option can address all the issues of the ETS.



5.2 ONLY HIGHER LRF AND REBASING OPTIONS LEAD TO EMISSIONS IN LINE WITH THE EU LONG-TERM CLIMATE TARGETS

Figure 5-1 below forecasted emissions reductions with the implementation of each option.

Figure 5-1: EU ETS emissions, 2015 – 2040, Options Source: FTI-CL Energy

Out of all the option scenarios, only the Higher LRF option would result in emissions reductions that would fulfil the EU’s long-term targets for 2040 (761 Mt CO2-equivalent). As long as the LRF is greater than 2.4%, the emissions reduction will be greater than the EU ETS target level, and at a LRF of 2.6%, emissions will be at 738 Mt CO2-equivalent in 2040. It is worth noting that the Rebasing option would come close to meeting long-term EU ETS goals, but it falls short at 786 MT CO2-equivalent in 2040. Of the remaining scenarios, Voluntary allowances cancellation option has very little impact as compared to the Baseline scenario; projected emissions level is 847 Mt CO2-equivalent in 2040 as compared to 846 Mt CO2-equivalent in Baseline. This policy does not impact market players’ behaviour, as cancellations are not predictable or credible, and the increase in ETS prices limits the amount of credits that countries can effectively remove. Market players can anticipate prices increases with Price corridor and Stronger MSR options, which would result in a strong short-run impact. As the price floor ceases to become binding and the speed at which credits are removed from the ETS decreases, the long-run impact will also decrease, and emissions levels will be 846 and 842 Mt CO2-equivalent respectively in 2040. With the Adjustment for overlapping policies, the impact will initially be moderate, due to the limited foresight of agents and a small static impact in the first years of implementation. Subsequently, this option scenario will have a stronger medium-term impact as the supply of credits decreases strongly post-2020, and emissions will reduce from 1,127 Mt CO2-equivalent in 2030 to 834 Mt CO2-equivalent in 2040.



5.3 REBASING, PRICE CORRIDOR AND STRONGER MSR OPTIONS SUPPORT SHORT-TERM CARBON PRICES FOR DECARBONISATION

Figure 5-2 below shows the projected EU ETS carbon prices in the selected option scenarios.

Figure 5-2: EU ETS carbon price (real 2015), 2015 – 2040, Options Source: FTI-CL Energy

Higher LRF and Adjustment for overlapping policies options do not foster coal-gas switching and investment in clean technologies in the short run, due to market participants’ limited foresight and gradual impact of reform, while Rebasing, Price corridor and Stronger MSR options support short-term carbon prices, favouring coal-gas switching and investment in clean technologies. According to our analysis, only Rebasing and Higher LRF options would allow carbon prices to exceed the lowest coal-gas switching price in the benchmark range. With option Higher LRF, EU ETS carbon prices would initially remain low, as agents would not perceive the scale of the reforms, but carbon prices would reach the breakeven price between 2025 – 2030 as the reform has a growing impact. EU ETS prices would increase from 34.2 €/t in 2030 to 92.1 €/t in 2040. Rebasing would result in prices ranging from 34.9 €/t in 2030 to 75.1 €/t in 2040. The remaining options would have a limited impact on the carbon price, with 2040 carbon prices of 58.4, 58.4, 56.8, and 56.4 €/t in option Voluntary allowances cancellation, Adjustment for overlapping policies, Price corridor and Stronger MSR, respectively. Given the budget constraint, the scale of the Voluntary allowances cancellation option is not sufficient for maintaining a continuous impact on EU ETS prices. For Adjustment for overlapping policies, Price corridor and Stronger MSR options, market participants will buy additional credits in anticipation of higher ETS prices in the future, which would cause the prices to go up in the short term. However, this impact would not be sustained in the long term.



5.4 ALL OPTIONS BUT VOLUNTARY CANCELLATION AFFECT CARBON INTENSIVE PLANTS MARGINS AND LIFE EXTENSION DECISIONS

Table 5-2 below estimates the impact of each option on installed capacity, relying on forecasts of the NPV of coal plants. Table 5-2: Life extension decisions, Options

Options

(Re)investment in fossil technologies over 2025-2040

(as compared to Baseline EC Reform) Positive impacts of options vs. current package

Higher LRF -12GW Reduction of cap and Rebasing affects plant life extension decisions and thereby reduce long-term lock-in effects.

Rebasing -46GW

Voluntary allowance Cancellation

Unchanged Voluntary cancellation would not trigger coal-gas switching in either the short or long terms, and does not affect plant life extension decisions.


-12GW Adjustment for overlapping policies accelerates the reduction of carbon intensive capacity in the medium-term.

Price corridor -26GW Price corridor accelerates the reduction of carbon intensive capacity in the medium-term. Its impact on long-term lock-in depends on the parameters.

Stronger MSR -46GW Stronger MSR accelerates the reduction of carbon intensive capacity in the medium and long terms.

Source: FTI-CL Energy Note: Fossil technologies include both coal and gas power plants.

Without a significant impact on the NPV in the short run, there will be no incentive for the closure of coal plants, leading to plant life extensions and the lock-in of emissions in the long term. All options except for Voluntary allowance cancellation would have a sufficiently strong short-term impact to change projected installed capacity in the long term. Each of these options would result in additional closures of 12GW to 46GW over period 2025-40. A faster reduction of coal and lignite capacity avoids the lock-in of emission capacity.



5.5 AUCTION REVENUE

Figure 5-3 sets out the forecasted total net revenue from allowance auctions from all option scenarios.

Figure 5-3: Net revenue from auctions, 2015 – 2040, Options Source: FTI-CL Energy Note: Net revenue is defined as the net present value of cash flows over 2015 – 2040. Cash flows are discounted using a -0.7% real interest rate.

Two opposite forces affect the total net revenue from auctions: the volume and the price effects. The price effect dominates; as the market gets tighter, the EU ETS price will increase, which in turn increases the revenue from auctioned allowances. The volume effect acts against the price effect, causing a negative impact on revenues as the number of auctioned allowances decreases. All option result in net revenue from auctions that exceeds that of the baseline. The implementation of one of the suggested options could bring up to 86 billion of euros in additional auction revenue to Member States that could be in part used to ensure a proper compensation for carbon leakage risk, an essential pre-requisite of any ambitious ETS reform.



Assessment of the combinations of options for 6.reform of the ETS

As no single option can address all the issues of the ETS, a range of combinations of options for the ETS reform have been modelled in order to address both short-term and long-term issues. In the following, we considered six (6) combination of options design to address both short term and long term issues (cf. Figure 6-1).

� Reduction of emissions cap + Price corridor;

� Reduction of emissions cap + Adjustment for overlapping policies ;

� Reduction of emissions cap + Stronger MSR ;

� Rebasing of emissions cap + Price corridor;

� Rebasing of emissions cap + Adjustment for overlapping policies ;

� Rebasing of emissions cap + Stronger MSR.

Figure 6-1: Combinations of reform options Source: FTI-CL Energy

We have modelled combinations of options where one option has a strong short-term impact and the other option has a strong long-term impact, which delivers the “best of both worlds”:

� The Adjustment for overlapping policies, Price corridor and Stronger MSR options have limited impact in the long term as those options do not materially alter supply and demand balance in the long run as compared to Baseline. However, they lead to a strong short/medium-term price signal for market participants.



� The Higher LRF option has growing impact as the market is drying up post-2020. The Rebasing option has a more moderate impact.

6.1 IMPACT OF THE COMBINATIONS OF OPTIONS ON SHORT AND LONG-TERM EMISSION LEVELS

All combinations of options would help reducing the emission levels in the short and long terms, in line with EU targets of 90% reduction by 2050.

Figure 6-2: EU ETS emissions, 2015 – 2040, Combinations of options Source: FTI-CL Energy



6.2 IMPACT OF THE COMBINATIONS OF OPTIONS ON ETS PRICES IN THE SHORT AND LONG TERM

All combinations of options modelled in Figure 6-3 would support ETS prices in the short and long terms.

Figure 6-3: EU ETS carbon price, 2015 – 2040, Combinations of options Source: FTI-CL Energy

In Figure 6-3 above, the EU ETS price reaches the social cost of carbon in 2025 or 2030 depending on combination. From 2020 onwards, the EU ETS price is close to the CO2 coal-gas switching price. As discussed above, we have chosen a combination of options with strong short/medium-term abatement signals and options with strong static effects in the long run:

� The Adjustment for overlapping policies, Price corridor and Stronger MSR options provide strong short- and medium-term abatement signals. Market players anticipate higher prices in the future due to the quick reduction of surplus, implementation of the corridor, etc.; they buy additional credits in the current market for future use, and this leads to a significant price increase.

� The Higher LRF and Rebasing options have strong static/intertemporal effects in the long run, as the supply of credits decreases strongly post-2020.



6.3 ALL OPTIONS PREVENT COAL PLANT LIFE EXTENSIONS AND LOCK-IN OF EMISSIONS IN THE LONG TERM

All combinations prevent coal plant life extensions in the short term and lock-in of emissions in the long term. Table 6-1 estimates the impact of each combination of options on installed capacity, relying on forecasts of the NPV of coal plants. Table 6-1: Life extension decisions, Combinations of options

Options (Re)investment in fossil technologies over 2025-2040

(as compared to Baseline EC Reform)

Higher LRF [2.6%] + Cap & floor [20-50€/t] -52 GW

Higher LRF [2.6%] + Stronger MSR [24%] -25 GW

Higher LRF [2.6%] + Adjustment for overlapping policies -29 GW

Rebasing [2021] + Cap & floor [20-50€/t] -52 GW

Rebasing [2021] + Stronger MSR [24%] -47 GW

Rebasing [2021] + Adjustment for overlapping policies -46 GW

Source: FTI-CL Energy Note: Fossil technologies include both coal and gas power plants.

With a significant impact on the NPV in the short run, there will be incentive for the closure of coal plants, leading to plant life extensions and the lock-in of emissions in the long term. All combination of options would have a sufficiently strong short-term impact to change projected installed capacity in the long term. Each of these combination options would result in additional closures of 25GW to 52GW over period 2025-40. A faster reduction of coal and lignite capacity avoids the lock-in of emission capacity.

6.4 AUCTION REVENUE

Figure 6-4 sets out the forecasted total net revenue from allowance auctions from all combination of options scenarios.

Figure 6-4: Net revenue from auctions, 2015 – 2040, Combinations of options Source: FTI-CL Energy

All combination of options result in net revenue from auctions that exceeds that of the baseline. The implementation of one of the suggested options could bring up to 132 billion of euros in additional auction



revenue to Member States that could be in part used to ensure a proper compensation for carbon leakage risk, an essential pre-requisite of any ambitious ETS reform.



Conclusion and policy recommendations 7.

A number of economic and political factors have led to a significant surplus of allowances and low price levels in the ETS. The ETS requires urgent and decisive reform if it is to remain as the cornerstone of EU’s decarbonisation strategy and to align the ETS with the ambition of the Paris Climate Agreement. The current depressed and uncertain carbon price signal does not support investment in clean technologies and efficient emissions reductions. Decarbonisation is instead driven by national regulations and financial support for some specific technologies, in a vicious circle that risks further undermining the credibility of the ETS and ultimately threatening the mere existence of the ETS. The ETS reform should be more ambitious than the EC proposal since this would not lead to a decarbonisation path that is underpinned by efficient short- and long-term price signals in line with the objectives of the Paris Agreement, robust to overlapping policy interventions, supportive of investment in low-carbon technologies, and avoids lock-in of conventional fossil fuel technologies. We assessed the impact of the ETS reform tabled by the EC as well as 6 alternative options. They all contribute in different ways to address the current ETS issues: some have positive effects in the short or long term, or in improving the ETS robustness to potential shocks or overlapping policies. A combination of several options is therefore required for a comprehensive ETS reform. Our modelling demonstrates that combinations of options can efficiently address both short-term and long-term issues of the ETS, and improve its robustness and credibility. In addition, the implementation of one of the suggested combination of options could bring up to 132 B€ in additional auction revenue to Member States that could be, in part, used to ensure a proper compensation for carbon leakage risk – an essential pre-requisite of any ambitious ETS reform. In summary, we recommend an ambitious yet realistic ETS reform based on a combination of options anchored into the following underlying principles: (i) a European market-based approach in order to drive decarbonisation at least cost, support investment in clean technologies, and avoid uncoordinated national interventions; (ii) a comprehensive answer to both short-term and long-term ETS issues, addressing the overlap of policies and improving its resilience to shocks, thereby strengthening the robustness and credibility of the ETS; and (iii) appropriate compensation mechanisms in place for sectors at risk of carbon leakage.



Appendix A – Detailed results 8.

8.1 COMPARISON OF OPTIONS

Table 8-1: Detailed indicators, Options

Baseline- EC Reform

Higher LRF Rebasing Voluntary

cancellation Adjustment Price corridor Stronger MSR

CO2 emissions

(Mt CO2-eq)

(% reduction compared to baseline)

2020 1,636 1,619

(1%)

1,606

(2%)

1,636

(0%)

1,609

(2%)

1,524

(7%)

1,530

(6%)

2030 1,241 1,172

(6%)

1,173

(5%)

1,216

(2%)

1,127

(9%)

1,200

(3%)

1,109

(11%)

2040 846 738

(13%)

786

(7%)

847

(0%)

834

(1%)

846

(0%)

842

(0%)

Emissions price (€/tCO2)

2020 4.7 5.9 9.4 4.7 6.5 20.0 10.3

2030 30.4 34.2 34.9 31.7 37.0 31.0 33.5

2040 59.1 92.1 75.1 58.4 58.4 56.8 56.4

Auction revenues

(bn€)

2021-2030 227 244 256 235 248 247 254

2031-2040 381 446 415 381 378 369 382




Table 8-2: Detailed indicators on power sector, Options

Baseline- EC Reform



Power sector CO2 emissions

(Mt CO2-eq)


2020 1,029 1,020

(1%)

1,002

(3%)

1,029

(0%)

1,020

(1%)

986

(4%)

995

(3%)

2030 885 832

(6%)

845

(5%)

874

(1%)

826

(7%)

879

(1%)

825

(7%)

2040 675 579

(15%)

639

(5%)

678

(0%)

576

(15%)

686

(-2%)

689

(-2%)

Installed fossil generation capacity (GW)

2020 415 415 415 415 415 415 415

2030 390 390 378 390 390 378 378

2040 367 356 354 367 354 367 354

Net Present Value of plant extension (€/kW)

2021-2030 n/a n/a n/a n/a n/a n/a n/a

2031-2040 47 17 -8 36 8 -15 -30




8.2 COMPARISON OF COMBINATION OF OPTIONS

Table 8-3: Detailed indicators, Combinations of options

Baseline- EC Reform



CO2 emissions

(Mt CO2-eq)


2020 1,636 1,514

(7%)

1,512

(8%)

1,592

(3%)

1,478

(10%)

1,449

(11%)

1,533

(6%)

2030 1,241 1,161

(6%)

1,048

(16%)

1,058

(15%)

1,145

(8%)

1,032

(17%)

1,035

(17%)

2040 846 738

(13%)

737

(13%)

730

(14%)

774

(9%)

778

(8%)

764

(10%)

Emissions price (€/tCO2)

2020 4.7 20.0 11,2 7.6 20.0 14.2 10.9

2030 30.4 33.5 41.5 41.8 34.4 41.9 42.8

2040 59.1 88.9 86.2 90.7 73.8 68.7 74.6

Auction revenues

(bn€)

2021-2030 227 254 263 260 256 271 269

2031-2040 381 432 435 464 408 410 429




Table 8-4: Detailed indicators on power sector, Combinations of options

Baseline- EC Reform



Power sector CO2 emissions

(Mt CO2-eq)


2020 1,029 980

(5%)

986

(4%)

1,013

(2%)

963

(6%)

955

(7%)

989

(4%)

2030 885 857

(3%)

789

(11%)

787

(11%)

850

(4%)

786

(11%)

780

(12%)

2040 675 610

(10%)

616

(9%)

607

(10%)

642

(5%)

653

(3%)

640

(5%)

Installed fossil generation capacity (GW)

2020 415 415 415 415 415 415 415

2030 390 378 372 384 378 372 372

2040 367 354 351 351 353 351 351

Net Present Value of plant extension (€/kW)

2021-2030 n/a n/a n/a n/a n/a n/a n/a

2031-2040 47 -22 -48 -15 -33 -77 -46




Appendix B – Modelling framework 9.

Our impact assessment is based our ETS model, supported by a plant-by-plant EU power market dispatch model.

9.1 ETS CARBON MODEL

Our EU ETS model can be viewed as an extended version of the Zephyr model.31 It calculates the EU ETS carbon price and emissions from the power and industrial sectors, based on a detailed representation of ETS market supply and demand fundamentals. It aims at both replicating the imperfect inter-temporal compliance behaviour observed ex post – due to short-sightedness and imperfect response from market participants to uncertainty and complexity32 – and observing how they may interact with exogenous scenarios such as implementation of ETS reforms. The model is solved on an annual basis over the time period covering 2015 to 2040. The model takes allocation, annual marginal abatement cost curves and macro-economic scenarios as exogenous inputs. It then computes the reaction of participants to those inputs in a simulated compliance framework, where each can buy or sell allowances depending on its needs and its anticipations. The model solves annual supply-demand equilibriums iteratively and the outputs are the resulting carbon price and emissions trajectories. Figure 9-1 below describes the general framework of the model, showing how the different blocks of the model are connected with one another.

Figure 9-1: EU ETS carbon model Source: FTI-CL Energy

31 The EU ETS modelling approach is inspired from the ZEPHYR model developed by Raphaël Trotignon & Boris Solier,

http://www.chaireeconomieduclimat.org.

32 https://ec.europa.eu/clima/policies/ets/index_en.htm



9.2 SUPPLY

The main features of the credits supply side foreseen in the Directive33 are properly taken into account in our modelling. Annual supply is thus made of the EU allowances cap,34 international credits (about 600Mt over Phase III) and the release of allowances that have been banked in the past. Each year, allowances are attributed to different industrial sectors in proportion to realised emissions. The share of free allowance is set in accordance with the Directive.35 Remaining allowances are auctioned. The total amount of allowances effectively supplied in auctions is thus made of the EU auction allowance cap, international credits as well as release banking (these points will be further developed). Impacts of both the MSR and the back-loading are taken into account by removing the proper number of credits from the auction volume. In such a framework, the overall supply of credits (over Phases 3 and beyond) is the result of regulation. However, the annual supply of credits highly relies on banking behaviours of participants.

9.3 EMISSIONS

Participants trading in the model are economic sectors of the 31 countries included in the EU ETS as a whole. The actual emissions of sectors covered by the EU ETS respond to carbon price according to marginal abatement costs curves (hereafter “MACC”). The MACC of the power sector are derived from our power dispatch model while the MACC of industrial sectors are implicitly derived from the EC 2016 Reference scenario to 2050.36 Following Trotignon et al.,37 the model considers annual baseline emissions (��) for each industrial sector �, i.e. counterfactual emissions as they would be absent the EU ETS. They are derived from an initial exogenous value �� and the following iterative relation:

��,� = ��,� � − ��,� �� × �1 + ��

The parameter � represents the share of emission reductions that are deemed to be long-term in the sense that they result from investments in low-carbon equipment rather than mere fuel-switching or reduced output. The parameter � represents the annual baseline emissions change rate which is a function of both the EU GDP growth and the elasticity of emissions to production. The annual marginal abatement cost curve (��) is exogenously given. Specifically, we assume that, for each year �, for a price of carbon ��:

33 DIRECTIVE 2003/87/EC OF THE EUROPEAN PARLIAMENT AND OF THE COUNCIL of 13 October 2003 “establishing a

scheme for greenhouse gas emission allowance trading within the Community and amending Council Directive 96/61/EC”.

34 We distinguish allowance cap for fixed installation and for aviation. For aviation, we take account of the “stop the clock” https://ec.europa.eu/clima/policies/ets/cap_en.

35 Airlines (intra-EU): 85%; Metals/Cement: 100% (significant risk of carbon leakage); Power generators: 0%; others sectors: from 80% in 2013 to 30% in 2020, 30% after 2020.

36 Alternatively, we have use MACC reported in Landis, F. (2015): Final Report on Marginal Abatement Cost Curves for the Evaluation of the Market Stability Reserve. ZEW Working Paper Documentation Nr. 15-01.

37 Trotignon, Raphaël, et al. European carbon market: lessons on the impact of a market stability reserve using the Zephyr model. No. 1511. 2015.



�,� = � × �1 − �� ! "#$% Where � describes the share of baseline emissions that can be abated at an infinite price, & and ! are the two parameters that conjointly determine the shape of the abatement curve. For the sake of simplicity, those parameters are assumed to be common to all industrial sectors. Annual marginal abatement cost curves are calibrated on the EC 2016 Reference Scenario to 2050. The EC Reference Scenario gives emissions level associated to different carbon prices for each years of the period 2016 - 2050. It thus implicitly defines a demand curve for CO2 credits. Figure 9-2 below shows the MACC we implicitly derived from this EC dataset.

Figure 9-2: Marginal abatement costs curve, Industrial sectors Source: EC (2016), « Reference scenario to 2050 »

Note: α = 65%, β =1.4, τ = 32.

In a given year, abatement is modelled as cost-efficient in the sense that there is a perfect recognition from participants of the opportunity cost of emission reductions, i.e. marginal abatement costs are equalised across sectors to the market price. This allows us to simply express the demand of permits by industrial sector �, denoted �,�, to reflect that it is equal to verified emissions, as a function of both baseline emissions and the carbon price:

�,� = ��,� − �,� 9.4 BANKING

The literature makes clear that the EU ETS Market is an intertemporal market in which agents follow different banking behaviours:

� Arbitrage, which consists of buying allowances and simultaneously selling forward, future or option contracts, so as to avoid exposure to carbon price risk. This requires the capital to buy and keep the allowances. Early studies on the EU ETS identified the possible arbitrage opportunities in the market.38 In recent years, front-year contracts for 2011 were traded at about 3-5% discount below 2012 contracts, and contracts for 2012 at 7% premium below 2013 contracts.39

38 Milunovich, George, and Roselyne Joyeux. "Testing market efficiency and price discovery in European carbon markets."

(2007).

39 Neuhoff, Karsten, et al. "Banking of Surplus Emissions Allowances: Does the Volume Matter?." (2012).



� Hedging, which consists of holding allowances to meet future carbon credit needs as an input to the production process, and thus avoiding the exposure to carbon price risk. Interviews with power, industry and finance actors confirmed discounts on future prices are applied in the order of 5% per year.40 Studies found that power generators generally hold CO2 allowances in the EU ETS to hedge for future power sales.

� Speculation, which consists of taking an open position in carbon allowances and carrying the carbon price risk in expectation that the carbon price will evolve favourably. Studies based on trading volume and open interest data found that there is a high degree of speculative behaviour at various points throughout each EU ETS Phase – particularly Phase II.41 Experiences from other commodities suggest that speculative buyers generally expect returns in the range of 10-15% per annum.42

EU ETS markets are thus linked across time in a dynamic way through banking. Specifically, future (expected) shocks – such as a reform removing allowances – would have an immediate impact on the market while current reform will have an impact on banking behaviours and thus on market conditions in the future. This specific feature of the EU ETS has two mains implication for our study. First, there is a need to properly model banking behaviour (speculation and hedging) of participants. Second, when assessing options for reform, we should consider keeping in mind the intertemporal dimensions of the EU ETS market. Specifically, the reform will reduce the aggregate stock of emissions available in the market at a given point in time (static impact) when the reform is implemented. The magnitude of the instantaneous impact depends on the strength of the reform. Then, given the expected static impact of the measure on the aggregate stock of emissions, market participants will adjust their banking behaviour in the current market. This adjustment is likely to smooth the overall impact of the reform across time lead to differentiating short- and long-term impacts. It is worth noticing that in order for the reform to affect banking behaviours, the reform should be predictable and credible.43 Our modelling of banking of surplus emissions allowances encompasses both hedging and speculative behaviours observed in the EU ETS market. In our framework, banking behaviours are driven by anticipated market and individual emission reduction efforts. The degree of participants’ farsightedness is measured by the length, in year, of the anticipation period, which may vary from one participant to another. Market behaviours of hedgers – or equivalently each sector � covered by the ETS – depend on the current position (*,�) and of hedging needs (Δ,�) over an anticipation period of three (3) years:44 ,,�- = − .��*,�� × /*,�/ + 1012,34�56 × |8,�| Where,

40 Eurelectric (2009). “EU ETS Phase 3 Auctioning – Timing and Futures versus Spot”.

41 Lucia, Julio J., Maria Mansanet-Bataller, and Ángel Pardo. "Speculative and hedging activities in the European carbon market." Energy Policy 82 (2015): 342-351.

42 Bessembinder, H. (1992). Systematic risk, hedging pressure, and risk premiums in futures markets. Review of Financial Studies, 5(4), 637-667; Wang, C. (2001). Investor sentiment and return predictability in agricultural futures markets. Journal of Futures Markets, 21(10), 929-952.

43 Ahrens (2002) outlines the key principles of effective reform programmes as accountability, credibility, participation, predictability, and transparency. He notes that “without commitments that are credible…neither effective policy reform nor sustained development can be expected”. Wara and Victor (2008) argue that the cost of a cap-and-trade system has to be predictable so that industry can plan investment.

44 Each year t, hedgers anticipate two quantitative aspects for year: (i) the future amount of free allocation they should receive; and (ii) their future annual emissions over the anticipation period.



� The current position (*,�) is defined as the difference between (i) the surplus – i.e. the stock of banked allowances that participants hold on their accounts at the beginning of each year, plus freely allocated allowances received for the current year – and (ii) participants surrendered allowances/verified emissions over the year. The sign of (*,�) indicates whether one specific sector is buyer or seller of credits for a given year.

� Hedging needs (Δ,�) are measured by the difference between expected future emissions over the anticipation period and amount of allowances received for free. In accordance with the literature, the anticipation period is set to three (3) years. Following Trotignon et al., we opted for the simplest form for the anticipation of future emissions that is based on the emission level of the current year and increased by anticipated growth of emissions, which could reasonably represent imperfect information on the participants’ side.

� To replicate observed banking behaviours and integrate other considerations involved in banking decisions, the parameter scales the intensity of hedging (6), i.e. the proportion of anticipated hedging needs that are banked.

Market behaviours of speculators would only look at the future tightness of the market within a period of five (5) years. Each year, they unlock part of their speculative positions – corresponding to a constant fraction of their portfolio. The demand/supply from speculators in year � is thus given by the following formula: ,�9 = 1013,:4�56 × /8�,;/ <=> ?�9 = @�9/B9 Where,

� @�9 represents the surplus – i.e. the stock of banked allowances that participants hold on their accounts at the beginning of each year.

� The anticipated tightness of the market (Δ�,;) is measured by the difference between expected supply (free + auctioned) and demand. In other words, speculators will bank credits when they perceive a difference between anticipated emissions and overall amount of allowances in future markets. The anticipation period is set to five (5) years.

9.5 MARKET EQUILIBRIUM

The model conducts a forward-looking optimisation. The annual computation of the market equilibrium is realised via a loop on an incrementally increasing carbon price ��. Specifically, the model computes compliance behaviours of participants, calculates supply and demand as defined, and checks whether there is market equilibrium at 0 price. If not, the price is gradually incremented, which is set to stop as soon as it reaches the first or minimal value of �� that allows a supply-demand equilibrium to form on the market. Each year, abatement is thus made at the least cost and banking levels, obtained at our estimate of market clearing levels.

9.6 POWER MARKET DISPATCH

Our dispatch model covers the CO2 EU ETS power markets. It is calibrated to reproduce historical power market prices and generation level. Our power market dispatch model uses a detailed thermal plants’ list covering the EU ETS power markets. The model captures the granular impact of existing capacities phase-down national plans as well as retirements. Our power market model is also designed to model renewable generation. Hourly wind and solar profiles are derived from our in-house methodology which converts wind speed and solar irradiation into power outputs. Hydro generation is derived from a hydro thermal co-optimisation algorithm embedded at the heart of Plexos.



The model is solved on an annual basis over the time period covering 2015 to 2040. The model takes demand, fuels prices, installed capacity, regulatory framework and macro-economic scenarios as exogenous inputs. The model then optimises market participants’ operational and investment/retirement decisions, based on their expected costs and revenues. The outputs are the resulting wholesale power prices capacity prices, emissions, fuel consumption, etc. Figure 9-3 below describes the general framework of the model, showing how the different blocks of the model are connected with one another.

Figure 9-3: Power market dispatch model Source: FTI-CL Energy

The central block of the power market model is the European hourly dispatch model. It is implemented in the commercial modelling platform Plexos® Integrated Energy Model. This modelling platform is most commonly used in the European electricity industry by utilities, regulators and transmission system operators. The dispatch model uses a detailed bottom-up methodology. The supply from flexible thermal power plants is modelled individually to meet the demand net of the supply of must-run renewable generators. The dispatch is determined to minimise the costs of generation in the EU-28 countries while satisfying the unit commitment constraints of generators as well as the flow constraints over the European transmission network. The model uses the zonal transmission network representation that matches with the price zones currently implemented in Europe and the commercial transmission boundaries. In each price zone, the model calculates the price as the marginal value of energy delivered in that zone based on the simulated bids of flexible generators. The transmission network covered by FTI-CL dispatch model is shown in Figure 9-4.



Figure 9-4: Power market model geographic scope Source: FTI-CL Energy

9.7 KEY ASSUMPTIONS

Our baseline scenario is based on a robust set of assumptions. EU ETS Model FTI-CL Baseline scenario is based on the recent EC Reference Scenario 2016, but differs on some key parameters. First, we assume that the linear reduction factor will increase to 2.2% after 2020, and international credit of 600 Mt CO2-equivalent for Phase III. The allowance cap for the aviation sector is assumed to be at historical levels (“stop-the-clock”). Our model has derived marginal abatement cost curves from the EC scenario for industrial sectors. The marginal abatement cost curve of the power sector has been modelled in detail by our power dispatch model. Second, our model factors in the inter-temporality and anticipations from the different market participants actually observed in the ETS market. We assumed myopic agents with foresight of 3 – 5 years.

Table 9-1 compares milestone assumptions of our baseline scenario with the recent EC Reference Scenario 2016.45

45 EC (2016), “Reference Scenario to 2050”,

https://ec.europa.eu/energy/sites/ener/files/documents/20160712_Summary_Ref_scenario_MAIN_RESULTS%20(2)-web.pdf



Table 9-1: Core assumptions, Baseline scenario

Framework EU Reference Scenario 2016 Our modelling

Supply of allowances

EU ETS Cap • Fixed installations: 1.74% p.a. from 2013

• Aviation: 95% of avg. 2004-06)

• Fixed installations: 1,74% p.a. until 2020, 2,20% p.a. after

• Aviation: historical level

International credit • Included As-Is • 600 Mt CO2-eq for Phase III

Free/ auctioning • Allocation rules for the different sectors foreseen in the legislation

Back-loading • 400 Mt (2014), 300 Mt (2015), and 200Mt (2016)

MSR • Active from 2019

Emissions

Sector

• All sectors covered by the EU ETS

• Do not take account of “stop-the-clock” for extra-EU aviation

• All sectors covered by the EU ETS

• “Stop-the-clock” for extra-EU aviation

Region • All 31 countries participating in the EU ETS

GDP growth • 1.2% p.a. over period 2015 – 2020

• 1.5% p.a. from 2021

MACC • Primes model

• Industrial sectors: Derived from EC scenario

• Power: FTI-CL Energy model

Banking

Anticipation • Perfect foresight (until 2050) • Myopic agents (3-5 years)


Power dispatch Model FTI-CL’s detailed power sector model is based on the latest announcements from TSOs, regulators and market participants. The supply and demand inputs into the power sector model incorporate a combination of TSO reference scenario outlooks, expected projects for the future, and national plans. Commodity price assumptions rely on forwards until 2020, after which they converge to WEO 2015 New Policy by 2040. While power demand, RES and Nuclear capacity outlook are exogenous, thermal power plants’ operators’ investment, extension or retirement decisions are endogenous. For those which are not under national plans, decisions are dynamically modelled through an iterative process and net present value calculations. To model the current power market conditions as well as long-term commodity equilibrium, commodity price assumptions are derived from a two-step approach.

� Short-term commodity prices are based on forwards (period 2015- 2020). They represent the market expectation for the next coming years; and

� Long-term commodity prices converge to the WEO 2015 New Policy by 2040. The WEO 2015 New Policy scenario assumes that current and planned energy policies are carried forward. The resulting commodity price outlook therefore includes the relative growth reduction of global coal and gas demand.

Table 9-2 below presents the core assumptions of FTI-CL Energy power market model.



Table 9-2: Core assumptions, Power market model

Items Sources

Demand

Power demand

• Latest TSOs reference scenario outlooks

• ENTSOE MAF 2016 Expected progress scenario

• Median long-term Vision 2 & 3 of ENTSOE TYNDP 2016

Supply

RES capacity

• Latest RES National plans

• Further increase post-2025 due to continuation of support mechanisms and increasing RES competitiveness

Nuclear capacity • Latest National plans on phase-down or phase-out

• Latest announcement on plant life extension and new projects

Thermal capacity

• Latest announcements from operators and National plans on phase-out or conversion to biomass

• Latest announcement on refurbishment and new projects in the short-term

• Dynamically optimised in the longer term based on NPV of anticipated costs and revenues (post 2025-30)

Commodity Price Assumptions

Gas price • Forwards until 2020

• Converge to WEO 2015 New Policy by 2040

Coal price • Forwards until 2020

• Converge to WEO 2015 New Policy by 2040



Copyright © FTI Consulting, Inc.,

CONTACTS

Fabien Roques Compass Lexecon Senior Vice President froques@ compasslexecon.com

Emmanuel Grand FTI Consulting Senior Director emmanuel.grand@ fticonsulting.com

Guillaume Duquesne Compass Lexecon Senior Economist gduquesne@ compasslexecon.com

Yves Le Thieis Compass Lexecon Economist ylethieis@ compasslexecon.com

Date post:	11-May-2018
Category:	Documents
Upload:	lamque
View:	214 times
Download:	1 times

Wake Up Reforming EU ETS - FTI Consulting · 6.2 Impact of the combinations of options on ETS...

Documents